Professional Documents
Culture Documents
Practitioner's
Guide to
Psychoactive
Drugs
FOURTH EDITION
The
Practitioner's
Guide to
Psychoactive
Drugs
FOURTH EDITION
Edited by
and
The information in this book is based on the experience s and research of the editors and authors.
This material is for informational purpose s only and should not be construe d as prescribing
information for individual patients. The editors, authors, and publisher assum e no responsibility for
any treatmen t undertake n by the practitioner with individual patients. Companies , interventions,
and products are mentioned without bias to increase your knowledge only.
ISBN 978-1-4613-7695-8
© 1997 Springer Science+Busines s Media New York
Originall y published by Plenum Publishing Corporation in 1997, 1991, 1983, 1977
Softcover reprint of the hardcover 4th edition 1997
Al l rights reserve
d
No part of this book may be reproduced , stored in a retrieval system, or transmitted in any form or
by any means , electronic, mechanical
, photocopying , microfilming, recording, or otherwise,
without written permission from the Publisher
To Sara and Rebecca Gelenberg
-AJG
Lee S. Cohen, M.D. Director, Perinatal and Reproductive Psychiatry Clinical Re-
search Program, Clinical Psychopharmacology Unit, Massachusetts General Hos-
pital, Boston, Massachusetts 02114; Associate Professor of Psychiatry, Harvard
Medical School, Boston, Massachusetts 02115
Thomas G. Gutheil, M.D. Program in Psychiatry and the Law, Massachusetts Men-
tal Health Center, Boston, Massachusetts 02115
Vicki L. Heller, M.D. Associate in Obstetrics and Gynecology, Brigham and Wom-
en's Hospital, Boston, Massachusetts 02115
vii
viii CONTRIBUTORS
David B. Herzog, M.D. Director of the Eating Disorders Unit, Massachusetts Gen-
eral Hospital, Boston, Massachusetts, 02114; Associate Professor of Psychiatry,
Harvard Medical School, Boston, Massachusetts 02115
Roger D. Weiss, M.D. Alcohol and Drug Abuse Program, McLean Hospital, Bel-
mont, Massachusetts 02178; Department of Psychiatry, Harvard Medical School,
Boston, Massachusetts 02115
CONTRIBUTORS ix
Once again, in their new edition of a classic American handbook of clinical psycho-
pharmacology, Drs. Gelenberg and Bassuk and their colleagues have produced a master-
work of sound clinical guidelines for the use of medicines as an increasingly central
component of contemporary psychiatric practice. They have recruited an outstanding
panel of coauthors, but have managed to maintain a high level of consistency of quality
and style throughout the many chapters on specific classes of psychiatric illnesses and
their corresponding treatments. The book continues to encapsulate the clinically rele-
vant essentials of the pharmacology of each major class of psychotropic agents in a
way that clinicians can easily grasp. Moreover, it provides sound and carefully consid-
ered specific guidelines to diagnosis, drug selection, and dosing and patient assessment,
with particularly rich presentations on the recognition and management of adverse
effects. Inclusion of chapters on pediatric and geriatric populations and on topics
pertaining to substance abuse disorders and medicolegal aspects of the field are unusual
in books of this kind, and so are particularly valuable. Students and trainees will
appreciate the handbook's well-organized and readable style, and practicing clinicans
should be satisified with its balanced consideration of older, standard treatments as well
as the latest medicines and trends in clinical psychopharmacology, with glimpses into
future developments. The fresh and up-to-date quality of this edition is particularly
attractive in this era of progress and appearance of new treatments that is unprece-
dented since the early decades of our field. Sensitivity to the art of the practice of
clinical psychopharmacology and specific consideration of cost-effectiveness are also
very welcome in a modem textbook. Finally, the format and presentation of the book
continue to be attractive, easily handled, and especially appealing to busy trainees and
clinicians. This is a book to be used by students, trainees, and clinicians from many
disciplines, not only by psychiatrists and mental health professionals.
Since the third edition of this guide was published in 1991, clinical psychopharmacol-
ogy has made a leap forward that rivals the rapid advances of the 1950s. In the
intervening years, the role of serotonin-selective reuptake inhibitors (SSRIs) has ex-
panded, and their efficacy has been affirmed in the treatment of depression, obsessive-
compulsive disorder and related conditions, and panic and other anxiety disorders, and
perhaps in the adjunctive treatment of substance abuse. The greater tolerability of these
agents and their relative safety in overdose have benefited hundreds of thousands of
people who suffer from these conditions. Other antidepressants with unique phar-
macologic and clinical profiles have been introduced during this time, and their place
in psychiatric pharmacotherapy is currently being investigated.
As we write this Preface, scientists are studying the pharmacologic treatment of
psychoses. We are learning more about the potential and the limitations of the landmark
drug clozapine, and several new and novel antipsychotic agents are being tested and
probably will be released soon. At the same time, gradual progress is being made in the
use of drugs to treat addictive disorders.
Biological research is likely to inform clinical practice in greater and more mean-
ingful ways over the next few years. We may soon understand the genetic origins of
some psychiatric diseases, such as bipolar disorder. Studies of the P450 liver enzymes
have given us greater insight into interindividual differences in the pharmacokinetic
handling of medications and also the potential for drug-drug and drug-food interac-
tions. The tryptophan depletion paradigm has held up in multiple replication studies
following the innovative research of Delgado and others (1994), and it has been
extended from depression studies to studies of other conditions as a way to explore the
mechanism of antidepressant agents. Functional and structural imaging technologies
promise to reveal even more about how psychotropic drugs work.
In parallel with the unprecedented discoveries and imminent breakthroughs in
psychiatric pharmacotherapy, a virtual revolution in healthcare is occurring in the
United States. Practices have changed and continue to be modified at a dizzying pace,
but the ultimate shape and structure of our healthcare system, including that portion
devoted to the treatment of mental and emotional disorders, remains unclear. Although
cost drove the equation in the early 1990s, we hope quality will have a growing
influence in the closing of this millennium. We anticipate that practitioners conversant
with both pharmacologic and psychosocial approaches to behavioral conditions will
function prominently in the new system.
The editors appreciate the important practical role The Practitioner's Guide has
come to play for many clinicians. We have kept that in mind in planning and executing
this edition. We hope it will continue to assist practitioners in their daily work and
ultimately benefit many patients.
xiii
xiv PREFACE
Dr. Gelenberg appreciates the ongoing support of Ms. Josie Riley, who keeps his
daily schedule rational, and Ms. Heather Hopkins, whose writing and research assis-
tance with this and countless other projects has been invaluable. The University of
Arizona Department of Psychiatry and College of Medicine have provided a supportive
and intellectually stimulating milieu. Finally, he wants to acknowledge and thank his
wonderful daughters, whose love has provided vital sustenance through the most
difficult times. Dr. Gelenberg is particularly honored that his first teacher in psycho-
pharmacology, an internationally renowned and award-winning expert in the field,
Professor Ross Baldessarini, has agreed to write the Foreword to this book.
Dr. Bassuk wants to thank Jayne Samuda for her good-humored support and
assistance in completing this manuscript. Special thanks go to her two wonderful
children, Sarah and Danny, for being part of her life.
REFERENCE
Delgado P. L., Price, L. H., Miller, H. L., Salomon, R. M., Aghajanian, G. K., Heninger, G. R., Charney, D. S.
1994, Serotonin and the neurobiology of depression. Effects of tryptophan depletion in drug-free
depressed patients. Arch Gen Psychiatry 51 :865-874.
Contents
2. Depression
Alan 1. Gelenberg, M.D., and Pedro L. Delgado, M.D.
I. Introduction ............................................. . 19
II. Causal Models ........................................... . 20
A. Early Models and Historical Perspective 20
xv
xvi CONTENTS
3. Bipolar Disorder
Heather S. Hopkins and Alan 1. Gelenberg, M.D.
I. Introduction .............................................. 99
A. Clinical Presentation .................................... 100
B. Course of the Illness .................................... 101
II. General Measures for Treating Bipolar Disorder ................. 102
A. Electroconvulsive Therapy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102
B. Milieu Therapy. . . . . . . . . . . . . .. . . . . . . . . .. . . . . . . . . . . . . . . . . 103
C. Group Therapy. . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . 103
D. Psychotherapy and the Therapeutic Relationship .............. 104
III. Phannacotherapy .......................................... 104
A. Lithium ............................................... 105
1. Phannacokinetics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106
2. Adverse Reactions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107
a. Endocrine ....................................... 108
i. Thyroid ..................................... 108
ii. Diabetes Mellitus ............................. 108
b. Renal .......................................... 109
i. Decreased Concentration ....................... 109
ii. Structural Changes and Glomerular
Filtration Alterations .......................... 109
iii. Renal Function ............................... 109
iv. Other Kidney Effects .......................... 110
c. Hematological ................................... 110
d. Cardiovascular ................................... 110
e. Cutaneous . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111
f. Gastrointestinal .................................. 111
g. Central Nervous System and Neuromuscular . . . . . . . . . . . 111
h. Ocular. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113
i. Weight Gain. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113
j. Other Adverse Reactions. . . . . . . . . . . . . . . . . . . . . . . . . . . 113
3. Drug Interactions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113
4. Lithium Toxicity .................................... 115
5. Pregnancy ......................................... 117
6. Preparations and Dosage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117
7. Other Possible Indications for Lithium .................. 118
B. Valproate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118
1. Phannacokinetics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119
CONTENTS xix
4. Psychoses
Alan J. Gelenberg, M.D., and Samuel Keith, M.D.
I. Introduction .............................................. 153
II. Other Treatments .......................................... 154
A. Nonbiological .......................................... 154
B. Nondrug, Biological . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155
1. Electroconvulsive Therapy ............................ 156
2. Psychosurgery ...................................... 156
III. Antipsychotic Drugs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 156
A. Introduction and Terms .................................. 156
B. Effects on Behavior and the Nervous System ................ 157
C. Mechanism of Action. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 159
D. Classes and Chemistry. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162
E. General Principles of Use ................................ 166
F. Pharmacokinetics ....................................... 168
G. Adverse Effects and Toxicity. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 170
1. Neurological ....................................... 170
a. Extrapyramidal Syndromes ......................... 170
i. Acute Dystonic Reactions ...................... 170
ii. Akathisia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172
iii. Parkinson's Syndrome ......................... 173
iv. Tardive Dyskinesia and Tardive Dystonia. . . . . . . . . . 176
b. Sedation ........................................ 185
c. Seizures ........................................ 185
d. Neuroleptic Malignant Syndrome .................... 186
2. Anticholinergic ..................................... 187
a. Peripheral ....................................... 187
b. Central ......................................... 188
c. Serious Toxicity .................................. 188
3. Cardiovascular and Respiratory ........................ 188
a. Hypotension ..................................... 188
b. Cardiac ......................................... 189
4. Ocular ............................................ 190
5. Cutaneous ......................................... 191
6. Hormonal, Sexual, and Hypothalamic Reactions .......... 191
7. Hepatic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 192
8. Hematological ...................................... 193
9. Pregnancy and Lactation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 194
10. Withdrawal Reactions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 195
11. Overdose .......................................... 195
H. Drug Interactions and Combinations . . . . . . . . . . . . . . . . . . . . . . . . 196
I. Laboratory Tests and Monitoring .......................... 198
J. Clinical Uses of the Antipsychotic Drugs. . . . . . . . . . . . . . . . . . . . 199
1. Acute Treatment .................................... 199
CONTENTS xxi
5. Anxiety
Eric M. Reiman, M.D.
I. Introduction .............................................. 213
II. Current Concepts/Etiological Theories ......................... 213
A. Symptoms ............................................. 213
B. Substrates ............................................. 214
1. Neuroanatomical Substrates ........................... 214
2. Neurochemical Substrates ..................... ~ . . . . . . . 215
3. Psychological Substrates. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 216
C. Integrating Biological and Psychosocial Contributions
to the Problem ......................................... 217
D. Differential Diagnosis ................................... 218
1. The Anxious Response to Everyday Stressors ............ 219
2. Anxiety Disorders ................................... 219
a. The Physiological Effect of a Nonpsychiatric
Medical Condition ................................ 219
b. The Physiological Effect of a Substance or
Its Discontinuation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 220
c. Panic Disorder ................................... 220
d. Agoraphobia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 221
e. Social Phobia .................................... 221
f. Specific Phobia .................................. 222
g. Obsessive-Compulsive Disorder .................... 223
h. The Response to a Catastrophic Stressor .............. 225
i. Generalized Anxiety Disorder . . . . . . . . . . . . . . . . . . . . . . . 226
3. Other Psychiatric Disorders ........................... 227
III. Nonpharmacological Treatment Measures ...................... 227
A. Supportive Therapy ..................................... 227
B. Cognitive and Behavioral Therapies ........................ 228
C. Neurosurgical Treatments ................................ 229
IV. Pharmacotherapy .......................................... 229
A. Benzodiazepines .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 229
1. Introduction ........................................ 229
2. Mechanism of Action ................................ 231
3. Chemistry ......................................... 231
xxii CONTENTS
6. Insomnia
Jeffrey B. Weilburg, M.D., and Heather S. Hopkins
8. Geriatric Psychopharmacology
Carl Salzman, M.D.
9. Pediatric Psychopharmacology
Janet Wozniak, M.D., Joseph Biederman, M.D., Thomas Spencer, M.D.,
and Timothy Wilens, M.D.
I. HISTORICAL PERSPECTIVE
The use of psychotropic drugs-chemicals that affect the mind-probably dates from
the dawn of humankind. Early people turned to plant products such as alcohol, opiates,
and hallucinogens to relieve pain, quell anxiety, promote sleep, or facilitate religious
experiences. Although many of these uses fell outside the realm of medical healing,
many others were among the physicians' more potent and commonly employed treat-
ment options.
Modem psychopharmacology did not begin until the middle of the 20th century.
Before then, there were few chemical compounds used to treat psychiatric symptoms
and many potential pitfalls. In the l840s, clinicians prescribed bromides as sedatives.
In the latter part of the 19th century, other sedative agents were synthesized-includ-
ing paraldehyde and chloral hydrate. During this same period, Sigmund Freud sug-
gested that cocaine was a useful psychoactive drug, and Emil Kraepelin started the first
laboratory for testing drugs in humans. At the beginning of the 20th century, re-
searchers introduced the barbiturates.
In 1940, lithium chloride was given to cardiac patients as a salt substitute, but
many developed severe toxic reactions. Cade gave lithium to animals to determine if it
increased the solubility of uric acid. Noting its sedative properties, he then gave the salt
to several manic and agitated patients. l However, because of continued concerns about
toxic reactions, the drug was not approved for use in the United States until 1970.
The modem revolution in psychopharmacology began in 1949, when Charpentier
synthesized chlorpromazine (Thorazine@ and others), and by 1952, several individuals
had noted its beneficial effects on mania, paranoia, and other psychoses. 2 They observed
Alan J. Gelenberg, M.D. • Department of Psychiatry, University of Arizona Health Sciences Center,
Tucson, Arizona 85724.
The Practitioner's Guide to Psychoactive Drugs (Fourth Edition), Gelenberg and Bassuk, eds.
Plenum Publishing Corporation, New York, 1998
3
4 ALAN J. GELENBERG, M.D.
may soon be determined, as investigators construct new and more complex models of
how psychotropic drugs work in the brain, and as clinical researchers examine the
interface of pharmacotherapy and psychotherapy and the applications of both in the
long-term treatment of chronic disorders.
rapport with a patient. This increases the quantity and reliability of information from
the patient, leading to a diagnosis, and solidifies the relationship on which compliance
with the medication regimen will be built. The clinician's "third ear" will help detect
inconsistencies, disparities, and missing pieces of data that might change diagnostic
impressions or highlight problems in following a medication regimen. A sophisticated
practitioner will recognize the nuances among different personalities and be sensitive
to their interactions with medication-responsive psychiatric syndromes.
At the same time, the science that grounds the clinician's practice continues to
expand and demands attention to a growing body of knowledge. The up-to-date practi-
tioner must recognize the increasing number of psychiatric syndromes identified as
responsive to medication and those drugs currently indicated for such disorders. He or
she must know which features of an illness are likely to respond to drug treatment and
be aware of adjuncts or second- or third-line treatments in case the first fails to produce
expected improvement. The pharmacokinetics of drugs and their potential interactions
are becoming more important, and clinicians have to maintain familiarity with medical
conditions and substance abuse problems (and their treatments) and how these might
interface with the presentation of psychiatric disorders and their treatment.
A growing scientific literature defines and assesses the roles of various forms of
rigorously described psychotherapies in the treatment of phenomenologically defined
psychiatric syndromes. Interpersonal and cognitive-behavioral therapies are the best
studied. Both were initially derived from psychoanalytic constructs and clinical obser-
vations. Unfortunately, more formal psychoanalytic treatments have not been systemat-
ically studied as much, although many experienced psychiatrists value the theories and
techniques.
The current literature suggests that for some patients and certain syndromes (as
will be described in detail in the subsequent chapters), medication combined with
minimal psychotherapy is sufficient for symptomatic relief. Increasingly, though, expe-
rienced clinicians recognize that long-term compliance with treatment requires a re-
spectful, interactive relationship, often with significant others as well as the patient, and
abundant opportunities for exchange of information that allow the patient to be an
informed consumer. For some conditions, psychotherapy is as effective as phar-
macotherapy, particularly for those at the milder end of symptomatic spectra. To date,
no research has established superior efficacy for one type of psychotherapy over
another in any defined disorder. What the effective components of psychotherapy are-
such as warmth, support, insight, behavior change, or altered cognitive schemata-
remain to be understood. At a minimum, practitioners should attempt to identify
psychological and interpersonal misunderstandings and dysfunctional patterns of
thought and behavior that increase stress and lower resistance, leading to symptomatic
exacerbation of long-term psychiatric disorders.
In summary, many patients with psychiatric disorders can be treated with psycho-
therapy alone, but the decision not to prescribe medication should be based on up-to-
date knowledge about the patient's diagnosis and condition. Moreover, prolonged
psychotherapy alone should not be used in disorders known to be responsive to medica-
tion without considering drug therapy. More severe disorders known to respond to
medication should be considered for medication therapy, often along with psycho-
therapy, from early clinical contact. Of course, patient preference must be taken into
account, but the clinician is obligated to provide up-to-date information about treat-
ment options for the condition. The clinician is similarly obligated to remain abreast of
the latest knowledge about types of therapy known to be helpful for specific psychiatric
disorders. For many psychiatric patients, particularly those with serious and chronic
mental illnesses, some form of psychotherapy in tandem with medication treatment is
worthwhile, whether provided by the same clinician or two different ones.
A. Patient Evaluation
Evaluating a patient for psychoactive medication is inherently similar to any other
medical assessment. The clinician should begin with a comprehensive history, includ-
ing a careful review of medical history, other medications, and recent medical changes
and-very important-a review of systems. Observations of family members or other
10 ALAN J. GELENBERG, M.D.
observers of the patient's behavior can be highly valuable, particularly for a psychiatric
diagnosis. A family history of medical, neurological, and psychiatric disorders often
will affect diagnostic considerations and treatment decisions. Knowledge of a patient's
past treatment successes and failures will be important, not only from a biological
perspective but also to determine a patient's comfort with various kinds of interperson-
al interactions. The patient should undergo a physical examination, unless one has been
done recently, and various laboratory tests, as determined by the clinician's differential
diagnosis. Although there are no laboratory tests that specifically rule in a psychiatric
diagnosis, many-such as the assessment of endocrine function or brain imaging-can
rule out other diagnoses that could produce emotional and behavior changes. As in all
medical diagnoses, hypotheses tend to be framed in the course of history taking, and
these should be worked and reworked as more information becomes available.
(continued)
12 ALAN J. GELENBERG, M.D.
TABLE 1. (Continued)
diltiazem, Erythromycin,
ethinylestradiol, felodipine,
imipramine, lidocaine,
midazolam, nefazodone,
nifedipine,
O-desmethylvenlafaxine,
propafenone, quinidine,
sertraline, tamoxifen,
terfenadine, b triazolam,
venlafaxine, verapamil,
vinblastine
2C Amitriptyline, citalopram, Auoxetine, fluvoxamine, 18% of Japanese, 19% of
clomipramine, sertraline African Americans, 8% of
desmethyldiazepam, Africans, and 3-5%
diazepam, hexobarbital, Caucasians are poor
imipramine, metabolizers
mephobarbital,
moclobemide, omeprazole,
phenytoin,b proguanil,
propranolol,
S-mephenytoin,
tolbutamide, warfarinb
"The information in this table is gleaned from both in vitro and in vivo studies and case reports, which are not always consistent
with each other. The interactions of drugs with the P450 isoenzymes are currently the subject of rigorous clinical investigation.
In particular. there is controversy over which serotonin-selective reuptake inhibitors (SSRIs) inhibit which isoenzymes at what
doses and to what degree.
bHas a narrow therapeutic index. Monitor plasma levels closely and consider decreasing dose when coadministered with one of
the drugs that inhibits this enzyme.
<Contraindicated with the nonsedating antihistamines terfenadine and astemizole.
dContraindicated with cisapride.
For a drug to be effective, a patient must take it as prescribed and promptly report its
effects, both positive and negative, to the clinician. Education-about illness and
medicine-is a crucial factor toward establishing an interactive relationship. Another
important factor is the practitioner's respect for the patient and capacity to build trust,
individualized to the patient's temperament and sophistication. Beyond these essential
elements, compliance may be additionally enhanced by simplifying the medication
regimen. For most psychotropic agents, once-daily dosing is possible and easier to
remember. Some medications need to be given in divided doses, however, to limit the
risk of adverse effects. When multiple daily doses or combinations of medications
become part of a treatment regimen, compliance can be enhanced by the use of
medication boxes, calendar notations, or other mnemonic devices.
1. INTRODUCTION 13
brand when prescriptions are refilled and to monitor closely the appearance of adverse
reactions and the reemergence or worsening of clinical symptoms. Because of the
possible differences in effectiveness of the preparations administered, the practitioner
may also have to adjust the dosage.
In addition to being easier for the patient, once-a-day dosing can also be more cost-
effective, because it allows the use of higher-strength tablets or capsules. A lOO-mg
tablet, for example, tends to cost little more than a 50-mg tablet. Therefore, a single daily
dose of 100 mg can be more cost-effective than two 50-mg doses. If a patient is
hospitalized, less frequent dosing also means less staff time preparing and administering
the compound. If the dose must be divided, a lOO-mg tablet can be split in half to keep
expenses down.
A therapy's cost-effectiveness cannot always be measured strictly by the cost of the
medication doses. If side effects are unacceptable to a patient, for example, and if he or
she therefore stops taking the medication, it will obviously be ineffective. In such cases,
the cost of the illness is usually higher than the price of the medicine. If a medication
produces side effects such as sedation or ataxia, which lead to the risk of automobile
accidents or falls, the costs of possible injuries must be factored into the overall cost of
the medication. The savvy practitioner, therefore, becomes familiar not only with the
price of drugs but also with their relative value to a patient, based on treatment
effectiveness and adverse events.
Medically ill patients often are older and therefore have an increased risk of adverse
reactions and a decreased tolerance to medication. The elderly often are taking medica-
tions that may interact with psychoactive agents. The medical condition itself may
represent a relative contraindication to drug treatment or may obligate the clinician to
alter the dosage or methods of administration significantly. Because of these limitations,
the clinician should:
1. monitor possible drug interactions, adverse and toxic reactions, and special
metabolic problems;
2. assess the role of the medical condition in the etiology of the psychological
symptoms; and
3. administer the lowest effective dose.
v. CONCLUSION
REFERENCES
1. Cade J. F. J.: Lithium salts in the treatment of psychotic excitement. Med J Aust 2:349-352, 1949.
2. Laborit H., Huguenard P., Alluaume R.: Un nouveau stabilisateur vegetatif, Ie 4560 RP. Presse Med 60:206-
208,1952.
3. Lehmann H. E., Hunrahan G. E.: Chlorpromazine, a new inhibiting agent for psychomotor excitement and
manic states. Arch Neurol Psychiatry 71 :227 -237, 1954.
4. Sternbach L. H., Randall L. 0., Gustafson, S. R.: 1,4-Benzodiazepines: Chlordiazepoxide and related
compounds, in Gordon M. (ed): Psychopharmacological Agents. New York, Academic Press, 1964, vol 1,
pp 137-224.
5. Kuhn R.: The treatment of depressive states with G22355 (imipramine hydrochloride). Am J Psychiatry
115:459-464,1958.
6. Kramer P.O.: Listening to Prozac. New York, Viking, 1993.
7. Diagnostic and Statistical Manual of Mental Disorders, 4th ed. Washington, D.C., American Psychiatric
Association, 1994.
II
Major Psychiatric Disorders
2
Depression
I. INTRODUCTION
Major depression is a common, serious, and costly disorder. More than just feeling sad,
a depressed person may be plagued by self-doubt, guilt, and recriminations and lose
interest in formerly pleasurable activities (including eating and sex). Sleep, weight, and
other bodily functions may be altered, and the patient might lose touch with reality,
often attempting or actually succeeding in committing suicide. A recent survey found
the lifetime prevalence of major depression to be extremely variable among different
geographic locales, ranging from 1.5% in Taiwan to 19% in Beirut.) In this survey, the
rate in the United States was 5.2%. The Epidemiologic Catchment Area study in five
U.S. communities in the early 1980s reported a I-month prevalence of 1.6% for men
and 2.9% for women. 2 More recently, the National Comorbidity Survey found a 30-day
prevalence of 4.9% and a lifetime prevalence of 17.1 % for major depressive episode. 3
Unfortunately, in a survey it is impossible to distinguish a true difference reflecting
culture or genetics from artifact and diagnostic variables. Less in question is the
female-to-male ratio, which was approximately 2: 1 in all countries surveyed. For at
least half of those who experience a major depressive episode, similar episodes will
recur in the future, at times frequently and to a debilitating degree, and in some taking
on a chronic course without periods of remission. Fortunately, the increasing availabili-
ty of modem therapeutics, both biological and nonbiological, allows most people who
suffer from depression the chance to be effectively treated.
Alan J. Gelenberg, M.D., and Pedro L. Delgado, M.D. • Department of Psychiatry, University of
Arizona Health Sciences Center, Tucson, Arizona 85724.
The Practitioner's Guide to Psychoactive Drugs (Fourth Edition), Gelenberg and Bassuk, eds.
Plenum Publishing Corporation, New York, 1998
19
20 ALAN J. GELENBERG, M.D., AND PEDRO L. DELGADO, M.D.
B. Contemporary Models
1. Psychological
In his now classic paper "Mourning and Melancholia," Freud emphasized intra-
psychic conflict and life experience in the etiology of depression. 4 Abraham postulated
intense oral fixation in depressed individuals and differentiated grief from morbid
depression. 5 Moreover, he viewed depression as aggression turned toward the self.
Freud embraced these ideas but elaborated the concept of object loss. 4 He contended
that the real or fantasied loss of an ambivalently loved person precipitates an ego
regression with introjection and intense anger toward the lost object. This process
results in self-deprecating, guilty feelings. Bibring viewed depression as a discrete ego
state resulting from an individual's inability to achieve his ego idea1. 6 He believed that
aggression was not the driving force behind depression; depleted self-esteem was the
primary issue and secondarily led to hostility.
Other researchers stressed the role of deprivation during critical developmental
stages. The Harlows, Spitz, Ainsworth, Bowlby, the Robertsons, and Mahler empha-
sized early attachments and the process of separation-individuation.7. 8 Early depriva-
tion resulting from the loss of or separation from a major attachment (i.e., mothering
figure) may have permanent ramifications if it occurs during a critical period and is not
replaced by a meaningful relationship. Spitz observed that infants separated from their
mothers during the first year of life developed various symptoms that progressed from
apprehension and crying to withdrawal, motor slowing, despair, and, finally, detach-
ment and even retarded physical and emotional development. This "anaclitic depres-
sion" was also described by Robertson and Bowlby in older children and by Harlow in
primates. Clinical studies have since corroborated the importance of developmental
deficits in the etiology of depression.
From the perspective of self-psychology, patients who have a poor sense of self
and a lack of soothing self-objects frequently feel overburdened, helpless, worthless,
2. DEPRESSION 21
and depressed. 9 In clinical practice, those with severe narcissistic disorders and bor-
derline states often develop adult "anaclitic" conditions marked by feelings of empti-
ness, loneliness, and lowered mood, even in the presence of others. 10
From foundations in Ellis's rational emotive therapy and ego psychology, Beck et
al. envision depression as primarily a disorder of thinking. I I Like the psychoanalysts,
Beck postulates that distorted views of the self, the world, and the future result from
early learning. This negative set consists of disordered cognitive "schemata" and
cognitive "errors" that are stimulated by events in the patient's life and cause the
patient to feel hopeless and helpless in the face of life's difficulties. Beck, in a way,
bridges the psychodynamic and behavioral models.
Seligman's concept of "learned helplessness"12 is based on the hypothesis that
people get depressed when they feel a loss of control over positively reinforcing
experiences. Individuals exposed to unavoidable aversive stimuli may respond with
passivity and helplessness to similar "stresses" in the future. In addition, this reaction
can become a generalized personality trait; the individual believes that all adaptive
efforts will eventually fail. Lewinsohn et al., 13 like Seligman, view the lack of reinforc-
ing stimuli as central to the development of depression. However, they state that the
important deficit is the low number of positive reinforcements "contingent" on the
efforts of the individual. Moreover, according to this model, patients exposed to pun-
ishing or nonreinforcing responses from their environment may stop behaving in ways
that will yield rewarding responses.
Some authors have focused on social and existential themes. They emphasize not
only the importance of relationships but also the broader social context. Problems with
role status, individual purpose and meaning, and cultural myths may contribute to a
person's sense of alienation and despair. 14
2. Biological
Biological hypotheses of the etiology of depression have evolved over the past 40
years, paralleling the growth of the neurosciences and the development of effective
medication treatments. Theoretical formulations followed the fortuitous discovery of
agents that alleviated severe depression.
Initial assumptions underlying early neurobiological models of depression were
greatly influenced by the discovery of the dopamine deficiencies in the central nervous
system (eNS) of patients suffering from Parkinson's disease l5 and by the remarkable
therapeutic effects of the dopamine precursor L-dopa. 16 These groundbreaking discov-
eries were reported two years after dopamine was first identified in the basal ganglia. 17
It was in this historical context that the therapeutic and pharmacological effects of
antidepressant drugs were discovered and the first neurobiological theories of depres-
sion established.
The catecholamine deficiency hypothesis of depression was based on the observa-
tion that many antidepressant drugs increased synaptic concentrations of norepineph-
rine, whereas the catecholamine-depleting drug reserpine seemed to cause depres-
sionlike symptoms. 18, 19 This hypothesis proposed that depression was caused by a
deficiency of norepinephrine, and mania by an excess. The indoleamine hypothesis
22 ALAN J. GELENBERG, M.D., AND PEDRO L. DELGADO, M.D.
stated that a deficit of brain serotonin was responsible for depression, whereas drugs
that increased synaptic serotonin, such as monoamine oxidase (MAO) inhibitors or
serotonin precursors such as 5-hydroxytryptophan (5-HTP) and L-tryptophan, relieved
depression. 2o- 22
The deficiency hypotheses have been only partially supported. 23 Deficiencies of
norepinephrine or serotonin or their metabolites in cerebrospinal fluid (CSF), blood, or
urine have not been consistently demonstrated in depressed patients despite intensive
efforts to do SO.24 A subgroup of depressed patients with a history of impulsivity or
suicide do appear to have decreased CSF levels of the primary metabolite of serotonin,
5-hydroxyindoleacetic acid (5-HIAA). However, decreased CSF 5-HIAA is also found
in subjects with other diagnoses. 25 .26
As new data emerged, it became clear that alterations in CNS neurotransmitter
system function could occur by changes in presynaptic and postsynaptic receptor
sensitivity without the amount of the neurotransmitter itself altering. Thus, the defi-
ciency hypotheses were modified, and the "receptor sensitivity hypothesis" of anti-
depressant action was proposed. This hypothesis stated that the delayed therapeutic
effects of antidepressant treatment were related to time-dependent alterations in cate-
cholamine and indoleamine receptor sensitivity and implied that the pathophysiology
of depression may be more related to abnormal regulation of receptor sensitivity than to
deficiencies of a neurotransmitter. 24 The "dysregulation hypothesis," subsequently pro-
posed by Siever and Davis,27 stated that in affective disorders, regulatory or homeosta-
tic mechanisms controlling neurotransmitter function were dysregulated and effective
pharmacologic agents would restore normal regulation to these systems. 27
The dysregulation and receptor sensitivity hypotheses went beyond neuro-
transmitter deficiencies, proposing that functional deficits in neurotransmission could
occur with normal neurotransmitter content. These hypotheses specifically focused on
the monoamine systems and, until recently, tended to conceptualize depression as a
dysfunction in a single neurotransmitter system. These theories assumed that the neuro-
biological basis of depression involved the opposite neurochemical state to that caused
by antidepressant drugs and, therefore, antidepressant treatments worked by reversing
an actual or functional monoamine deficit state.
These models are limited because they are based on simplistic models of the
nervous system, focusing on a single neurotransmitter system. They are also limited by
the assumption that the neurobiological cause of depression is a mirror deficit of the
pharmacologic effects of antidepressants. This is comparable to assuming that the
cause of rheumatoid arthritis is a corticosteroid deficiency.
3. Integrative
More recent models of the neurobiological basis of depression have attempted to
integrate data from a variety of sources, including clinical observations, epidemiologi-
cal studies, functional brain imaging studies, human neurobiological challenge studies,
and animal models. These new models have tried to account for the diversity of clinical
presentation and course of illness in the context of what is known about the brain
systems that modulate affective and emotional states.
2. DEPRESSION 23
A consensus has now developed that major depression is not a single disorder or
disease but a syndrome. 28 This is based on the notable variation in the age of onset,
course of illness, and symptom profile of major depression. 29 A variety of structured
diagnostic instruments and classification nomenclatures have been developed to aid in
standardizing diagnosis. The concept of "major depression" has emerged from this
process as the umbrella diagnosis under which this broad spectrum of symptoms have
been grouped. As described in the fourth edition of the Diagnostic and Statistical
Manual of Mental Disorders (DSM-IV), the diagnosis of major depression is made
when a mixture of symptoms are present for most of the day, nearly every day during a
2-week period. 3o These symptoms include depressed mood as well as significant
changes in sleep, appetite, mental concentration, energy, ability to experience pleasure,
self-esteem, sense of hope, guilt, and desire to live.
Extremes of this diagnostic grouping are seen daily in clinical practice and can
range from a patient who describes a depressed mood distinctly different from sadness
and is experiencing insomnia, loss of appetite and weight, decreased libido, energy,
interest, and concentration, and suicidal ideas to another patient who describes a
slightly depressed mood with prominent irritability, loss of energy and mental concen-
tration, decreased libido, moderate generalized anxiety, and inappropriate excessive
worry but normal appetite, weight, and sleep. Yet another patient may describe daily
crying spells and intense sadness and severe inappropriate guilt, along with increased
appetite, weight, and sleep (sleeping 12-14 hours/day).
While the etiology of major depression may be heterogeneous, the brain systems
involved in determining specific emotions, cognitions, and somatic processes are likely
to be similar across patients. There is considerable evidence that specific brain net-
works and the neuroanatomical circuits that join them underlie emotional experience,
cognition, and somatic function. 31 While our current understanding of the role and
function of these circuits and networks remains primitive, a rough outline is beginning
to emerge. 32 As most antidepressant drugs are comparably effective across patients-
in spite of the sometimes dramatic differences in individual symptoms between pa-
tients with major depression-these drugs probably cause widespread effects in many
brain regions.
Akiskal and McKinney have described depression as a psychobiological "final
common pathway of the various interlocking processes at chemical, experiential, and
behavioral levels."33 They suggest that many factors-including biological vul-
nerability, developmental deficits, and physiological and psychosocial stressors-can
contribute in varying degrees to functional impairment of the diencephalic centers that
help to maintain mood, motor activity, appetite, sleep, and libido.
If these hypotheses are correct, then depressed patients suffering from similar
symptoms most likely have similar dysfunction of the same networks and circuits. The
dysfunction's etiology may differ between patients, but the net functional impact is
similar. An analogous situation would be a cardiac arrhythmia leading to an increase in
heart rate. In some situations this could be caused by an electrolyte imbalance, and in
others it might arise from infarction of a critical region of the heart. The same symptom
arises from different pathological processes. These types of integrative models begin to
account for the diversity of clinical presentations and course and offer explanations of
24 ALAN J. GELENBERG, M.D., AND PEDRO L. DELGADO, M.D.
how neurotransmitter systems, brain circuits, and the life experiences that affect the
development of these systems could interact to produce vulnerabilities to depression
and how life stress or neurobiological dysfunction could lead to a similar clinical
syndrome.
A. Clinical Presentations
Whether or not to treat a patient with an antidepressant medication is a common
and important question for mental health clinicians and primary-care doctors. Some-
day, biological tests may give us clues or even crisp answers to this question. For now,
however, psychiatric treatments are based on verbal reports and behavioral observa-
tions, coupled with knowledge gained from clinical research.
Several theoretical and practical issues remain extant and troubling to our field.
The very term "antidepressant" is unsatisfying. When a class of agents can treat
anuresis and chronic pain, panic and phobias, obsessions and compulsions, and other
symptoms, does this mean that these other conditions are somehow linked to mood
pathology, that they share a common brain center or pathway, or simply that the
pharmacologic action of the drugs is salutary for a heterogeneous group of conditions?
And what of the increasingly common observations of comorbidity? Patients with
depression frequently have coexisting anxiety disorders, substance abuse conditions,
obsessive-compulsive disorder, eating disorders, and more. Does one condition render
them vulnerable to others? Is a vulnerable brain more vulnerable to other conditions
(coincidentally or causally)? Or are the observations of frequent comorbidity simply
artifacts of our observational, rating, and diagnostic techniques?
If you are sitting with a patient who claims or appears to be depressed, what
should lead you to recommend or prescribe an antidepressant? The first question is
whether we are observing normal sadness or grief in reaction to a real, perceived, or
feared loss. Such losses can include not only a person (through separation or death), but
a pet, job, dream, wish, or plan. In psychologically healthy individuals, time, thought,
alternatives, and caring others can promote healing and restoration in most circum-
stances. Clinicians tend to judge the appropriateness of grief or "reactive" depression
by its duration and magnitude in response to the loss: greater losses are expected to
produce psychological responses of comparatively greater magnitude and duration.
Because normal grief responses can vary considerably among people from different
cultural backgrounds, the further your own cultural background diverges from your
patient's, the more circumspect you should be in drawing conclusions about the mean-
ing of the loss and the appropriateness of the reaction. Others more familiar with the
patient's culture should be consulted. The process of grieving is an iterative, dynamic,
progressive, and variable one. Most important is that the individual not become "stuck"
in the process: unable to move, to utilize the resources of others, or to experience the
cleansing relief of the grieving process and the waves of alternating sadness and
distraction that characterize this normal human experience. When symptoms and dys-
2. DEPRESSION 25
function persist, are severe, or threaten life, medical or psychological intervention may
become necessary. In some cases, the stress of grief can lead to the morbid mood
reaction of depression.
Mild, nonpsychotic depression has been shown to benefit from carefully defined
and studied forms of psychotherapy, such as interpersonal and cognitive-behavioral
techniques. Psychotherapy may also be useful when the patient engages in behavior
patterns that lead to loss and bereavement. Some patients, for example, habitually put
themselves psychologically "in harm's way"-for example, by pursuing or selecting
inappropriate mates or work situations. In these circumstances, successful psycho-
therapy might clarify misperceptions and lead to behavioral changes that can prevent
repetition of these patterns and recurrence of depression.
When symptoms of depression are severe and persistent, or when psychosocial
intervention has failed to yield notable relief and change over a matter of months,
antidepressant drugs usually are indicated. Particularly with the newer generation of
antidepressants, the risk-benefit ratio almost always favors at least a few therapeutic
trials of these agents.
There are several circumstances in which long-standing behavior patterns affect the
likelihood of response to antidepressant pharmacotherapy. One is a coexisting borderline
personality. Here the work of Soloff and others has shown that, although patients with
borderline personality disorder often qualify for diagnosis of major depression, their
response to antidepressant drugs tends to unsatisfactory (see Chapter 13).
On the other hand, when a patient has a long-standing depression, clinicians,
friends and family, and even the patient may come to believe that this is just "who they
are," in other words, that depression is part of the individual's character. A series of
recent studies, however, has indicated that chronic depression-whether dysthymia,
"double depression" (major depression superimposed on dysthymia), or chronic major
depression-is as responsive to antidepressant pharmacotherapy as depressions of
shorter duration. 34 Thus, chronicity should not in and of itself dampen clinicians' ardor
for prescribing. If, however, the patient has had a reasonable number of adequate
antidepressant trials and has failed to respond, the clinician may want to inquire again
about life circumstances, unrecognized stressors, behavior patterns, and occult medical
conditions that may suggest alternate approaches to the problem.
What can we learn from the pattern of depressive symptoms? The existence of a
DSM-IV diagnosis of depression does not automatically indicate the prescription of an
antidepressant medication. As noted above, mild depression may respond to psychoso-
cial intervention. Conversely, as noted above, depression that falls below the formal
symptomatic threshold of major depression should not automatically preclude the
prescription of an antidepressant drug-particularly if the symptoms of depression are
chronic. Cluster typing of depression symptoms may give some guidance. Depression
with melancholic features may be more likely to respond to antidepressants in general
or electroconvulsive therapy. Atypical depression has been shown to respond better to
monoamine oxidase inhibitor (MAOI) antidepressants than to tricyclics. 35
The concomitant presence of psychotic features roughly halves the estimated
response rate to an antidepressant drug (when used alone), whether a heterocyclic agent
or an MAOI. Certain types of personality features, such as hypochondriasis and histri-
26 ALAN J. GELENBERG, M.D., AND PEDRO L. DELGADO, M.D.
onic traits, bode poorly for antidepressant response. Prior resistance to biological
therapy in multiple previous depressive episodes may also be a poor prognostic sign.36
As long as psychiatry remains empirical and phenomenologically based, we will
be treating syndromes that can reflect diverse etiologies. Undoubtedly, this accounts
for the variable responses to psychotropic medications in different settings and studies.
The syndrome of depression is no different. It can be caused or mimicked by a variety
of metabolic, endocrine, toxic, pharmacologic, and neurologic conditions. It makes no
sense for clinicians to memorize all such possible "organic" causes of depression, and it
would be frivolous to conduct laboratory testing on a routine basis to rule them out.
Rather, by conducting a review of systems and medical history, an astute clinician with
sensitive medical "antennae" can detect and become suspicious of symptoms, signs, or
medical history that could indicate further testing and broaden the differential diag-
nosis.
B. Classifications
Historically, depressions have been subtyped according to dichotomous categories
(e.g., endogenous-reactive and neurotic-psychotic). Recently, this view has been re-
fined and broadened to include primary-secondary and unipolar-bipolar. In general,
endogenous depressions correspond to the melancholic subtype of major depressive
illness. The abandonment of the endogenous-reactive dichotomy has been important,
since it implied that some depressions do not have exclusively environmental causes
and require pharmacotherapy whereas others result from life stresses and should be
"worked through" psychotherapeutically. This clinical stance would preclude the use of
pharmacotherapy for patients with "reactive depressions" and the use of psychotherapy
for "endogenous" depressions. In fact, the decision to medicate with antidepressants
relies not so much on presumed causality but on the presence of particular symptom
clusters. The neurotic-psychotic classification of depressions is equally misleading,
since the definitions of these terms are variable and do not correlate well with drug
response.
More recently, diagnosticians have reclassified depressions. The primary-second-
ary dichotomy avoids some of the problems of the older continua. In primary depres-
sions, there is no previous history of psychiatric illness except for episodes of depres-
sion or mania; secondary depressions occur in patients with a previously documented
psychiatric disorder (other than an affective episode) or a medical illness. Investigators
originally thought that primary depressions might preferentially respond to antidepres-
sants. However, secondary depressions generally have similar symptoms and responses
to medication. Clinically, patients with secondary depressions have a somewhat greater
incidence of somatic and psychotic symptoms, phobias, anger, and suicidal behaviors.
Winokur, 115 in particular, believes that secondary depressions represent an acute pro-
cess superimposed on a chronic set of problems.
The unipolar-bipolar dichotomy represents a further subclassification of primary
mood disorders. Bipolar disorders must have a history of both depression and mania. In
contrast, a unipolar depressive disorder must include a history of depressions only,
although the number and frequency of depressions require further definition. By study-
2. DEPRESSION 27
ing families, Winokur and Coryell have further subdivided primary depression into
pure depressive disease, depressive spectrum disease, and sporadic depressive illness
(nonfamilial).28 Pure depressive disease occurs equally in men and women, but the
subgroup in which it occurs most often is older males who have one or more first-
degree relatives with depressions. Depressive spectrum disease, in contrast, occurs
most frequently in women under 40 years of age. Their female relatives have an
increased frequency of depression, and first-degree relatives have a high frequency of
alcoholism or antisocial characteristics. Males with depressive spectrum disease usu-
ally have alcoholic or antisocial symptoms.
Sporadic depressive disorders comprise the largest group of unipolar depressions
(about 40%). Although unipolar illnesses can be divided into relatively homogeneous
groups, the clinical ramifications of these findings await clarification. In addition,
researchers must further correlate epidemiologic, familial, and clinical data with bio-
chemical findings.
The DSM-IV is the currently adopted diagnostic framework for psychiatric disor-
ders. However, because DSM-IV must ensure a high degree of reliability and validity,
some more atypical presentations of depression are excluded. Despite limitations, it
defines a group of major depressive disorders (see Table 1) that are generally:
C. Biological Diagnosis
Since antidepressant drugs were first found to be effective in treating depression,
there has been hope that a biological test could be discovered to identify patients who
were more likely to improve on a specific drug. As new antidepressant drugs and
combination and augmentation strategies have been discovered, this hope has further
increased. Without doubt, treating depression with drugs will change remarkably if and
when we have biological tests similar to those in other fields of medicine: these will
enable us to subdivide an illness by etiology and select an appropriate treatment or
combination of treatments. Indeed, as Klerman has quipped, this is psychiatry's "search
for the holy grail." It stands to reason that there may be biological subtypes of depres-
sion and that these will prove differentially responsive to pharmacologic therapies.
When practical markers become available, they will represent a great leap forward,
shortening the lag time before remission and making a series of empirical trials of
antidepressant drugs a historical curiosity.
To date, unfortunately, we have no such biological testing available to practi-
tioners,36 but it has not been for lack of effort from investigators, attention in the
scientific literature, or research funding. Many hopes have risen and been subsequently
dashed. Candidates for biological markers have included urinary metabolites of central
monoamine neurotransmitters [e.g., 3-methoxy-4-hydroxyphenylglycol (MHPG)], pro-
vocative challenge tests (e.g., amphetamines), neuroendocrine markers [e.g., the Dexa-
methasone Suppression Test (DST) and the thyrotropin-releasing hormone (TRH) stim-
ulation test], the prolactin response to intravenous tryptophan, and the growth hormone
response to clonidine administration. Neurophysiological markers, such as the latency
before the first rapid eye movement (REM) period in the sleep cycle, also are being
actively studied. Although some markers have shown promise in predicting lack of
placebo response (e.g., failure to suppress cortisol appropriately during DST testing),
none of the markers studied to date nor the clinical subtypes identified have been
shown to have sufficient sensitivity or specificity to be useful as a marker of specific
medication response or nonresponse.
Attention is currently being focused on a variety of markers for specific neuro-
transmitter systems using positron-emission tomography (PET) technology. Radio-
labeled markers for specific components of the serotonergic, noradrenergic, and dopa-
minergic systems as well as markers of the functional state of these systems are
actively being investigated. Also being investigated are clinical neurotransmitter deple-
tion strategies. Identifying the reasons why some patients respond to an antidepressant
while others do not and why some patients respond to certain antidepressants and not
others continues to be just beyond our grasp. As our knowledge of basic neuroscience
increases and as new technologies are developed, we continue to hope that these
findings will lead to tests or procedures that can be successfully used to improve the
effectiveness of our treatments.
B. Psychotherapy
Depression is a set of symptoms with complex biopsychosocial determinants. For
the most severe depressions, biological factors rather than environmental or psycho-
logical factors seem to determine the course of the illness. These syndromes require
pharmacotherapy and may be only slightly responsive to specific psychotherapeutic
interventions. However, for depressions of moderate severity, various types of short-
term psychotherapy can play an important role in treatment.
Weissman et aI., in the Boston-New Haven Collaborative Studies, found that both
antidepressant medication and structured, short-term interpersonal psychotherapy
worked better than a mock treatment for patients with nonpsychotic major depressive
30 ALAN J. GELENBERG, M.D., AND PEDRO L. DELGADO, M.D.
illness. 38 Both drug therapy and combined drug and psychotherapy provided equal
symptom relief and prevention of relapse. Psychotherapy alone offered somewhat less
symptom relief but more improvement in social adjustment.
As part of the research effort, the investigators developed a method of therapy that
has been specified in detail, tested in adequately controlled studies, and applied consis-
tently in concert with pharmacotherapy.39.40 The technique, including 12-16 once-
a-week sessions focusing on current interpersonal relationships, is comprised of
several structured phases. Initially, the clinician uses a semistructured evaluation inter-
view to clarify symptom history and to review typical depressive symptoms and the
natural course of depressive illness with the patient. The goal of the assessment is to
define interpersonal problems of four distinct types: grief reactions, interpersonal con-
flicts, role transitions, and interpersonal deficits.
The core treatment sessions focus on present, not past, relationships and behav-
iors. During the process, therapists actively help patients understand their range of
feelings, delineate past model relationships that were positive experiences, and select
and refine potentially helpful interpersonal strategies and actions. The clinician uses
both directive and interpretive techniques. However, the focus of intervention is on
here-and-now interpersonal relationships, not historical, intrapsychic, or ego-structural
problems. Perhaps most important, interpersonal psychotherapy has demonstrated an
important synergy between biological and psychosocial interventions for moderately
severe depressions. In concert with drugs, this psychotherapy seems to decrease symp-
toms better than drugs alone during the acute illness, increase social functioning during
recovery, and produce better treatment outcomes overall. 41
Cognitive therapy, developed by Beck and Rush, represents another important
contribution to the treatment of depressive illness. Instead of an interpersonal relation-
ship focus, cognitive therapy confronts the distorted cognition of depressed patients. I I
The therapy aims to alter patients' punitive, undermining negative thoughts, images,
and underlying assumptions about themselves and others. Cognitions are those
thoughts and images elicited in specific situations. In depressed individuals, cognitions
include several kinds of logical errors, such as overpersonalization, ignoring positive
parts of situations, and overgeneralizing or exaggerating negative aspects of situations.
Beck et a1. 11 also emphasize that patients' hopelessness and helplessness stem, at
least in part, from long-standing negative views of themselves, the world, and the
future (cognitive triad) and that depressed individuals habitually perpetuate those
views, called schemata. These chronic underlying assumptions, derived from early
developmental experience, color the way cognitions are used day to day. In the case of
depressed patients, undermining schemata and negative cognitions fuel deep, self-
denigrating perspectives.
Therapy to correct these misattributions is generally short-term, consisting of one
or two sessions per week for 14 weeks. Initially, treatment focuses on helping the
patient specify and record cognitions and schemata that are self-deprecating. Subse-
quently, the therapist outlines possible new or revised cognitions that are more self-
empowering and more aligned with objective reality. Later in therapy, the clinician
helps the patient identify the schemata, or hidden assumptions, that underlie depressive
cognitions.
2. DEPRESSION 31
The final part of therapy focuses on generating new cognitions and schemata and
applying them to actual and anticipated life events. Patients may complete this phase of
care during the last session of the initial 14-week therapy or may require additional
once- or twice-a-month sessions extending over several months. The most important
goals of cognitive therapy include helping the patient recognize self-denigrating cogni-
tive structures and develop practical alternative viewpoints and actions. Increasingly
studied and applied, cognitive therapy appears to be effective for moderately severe
depression and may have a synergistic effect when used together with antidepressants.
Therapists who are more behaviorally oriented use similar techniques to treat
depression. 41 In addition to identifying and confronting cognitive structures, behavioral
therapists employ formal cognitive restructuring, skills training (including assertive-
ness training), and environment manipulation.
In general, evidence indicates not only that structured psychotherapies, including
various forms of short-term dynamic psychotherapy and crisis intervention, help many
depressions but also that their effectiveness might be equal to that of pharmacotherapy
for some milder depressions. However, we must be careful in drawing definitive
conclusions pending further investigation. Studies often have employed antidepres-
sants and specific psychotherapeutic methods inadequately. No investigations compar-
ing psychotherapy and pharmacotherapy have addressed the problem of "fit" between a
patient and a particular drug or a specific model of therapy. In addition, the accumulat-
ing evidence in psychotherapy research indicates that many factors other than the type
of therapy significantly promote positive outcome, including a patient's positive atti-
tude toward therapy and the therapist, a therapist's positive feeling about the patient,
and the patient's motivation.
Which depressed patients should receive psychotherapy? Every patient should be
involved in a therapeutic relationship. Individuals taking drugs should receive emotion-
al support and encouragement and the opportunity to address their feelings about the
depressive episode. Moreover, for selected individuals with significant character pa-
thology (e.g., borderline states, narcissistic personality disorders, masochistic or de-
pressive characters) who develop mild to moderately severe depressions in response to
chronic disappointment and low self-esteem, long-term insight-oriented psychotherapy
might be the treatment of choice. These patients, however, frequently develop more
severe and autonomous depressive symptoms accompanied by vegetative signs; such
symptoms may improve with medication. In addition, some therapies may be equal to
or better than medication for individuals who become depressed in response to severe
stress, life crises, or real or fantasied losses or from a severely negative view of the self,
world, or others.
C. Electroconvulsive Therapy
1. Introduction
Although its mechanism of action remains unknown, electroconvulsive therapy
(ECT) effectively treats most patients with severe depression. In fact, 60,000 to
100,000 individuals receive this treatment each year, most without significant mor-
32 ALAN J. GELENBERG, M.D., AND PEDRO L. DELGADO, M.D.
2. Mechanism of Action
3. Indications
Numerous controlled studies demonstrate that ECT works better than placebo and
at least as well as heterocyclic antidepressants in the treatment of severe depression. In
addition, ECT may be more effective than heterocyclics alone for treating psychotic
depressions, particularly when they are accompanied by agitation. ECT also alleviates
the symptoms of extreme agitation and catatonia in patients with schizophrenia who
respond poorly to medication, making these patients more amenable to other treatment.
ECT offers significant advantages over medication in certain clinical situations.
The strongest indications include:
1. Previous good response to ECT without significant untoward effects.
2. Strong family history of positive response to ECT.
3. Depression characterized by psychotic features, particularly somatic or ni-
hilistic delusions.
4. Relative or absolute contraindication to medication, such as in patients with
significant cardiovascular disease or other medical conditions that preclude
an adequate trial of drugs, including pregnancy or history of neuroleptic
malignant syndrome.
S. Need for immediate containment of symptoms (e.g., severely suicidal pa-
tients, severe mania, extremely agitated or assaultive patients, and individuals
with severe retarded depression) that impair self-care and feeding.
6. Poor response to trials of medication.
7. Severe, persistent behavior problems or mildly disorganized thinking in
schizophrenic patients.
ECT should be strongly considered for psychotically depressed patients and
those who have not responded to medication. In this situation, the clinician may start
or continue an antidepressant drug, since ECT and heterocyclics may work syner-
gistically. However, the anesthesia procedure occasionally may produce additional risk
of cardiotoxicity from the combination of ECT with antidepressants and anti psycho-
2. DEPRESSION 33
Several medical conditions may limit the use of ECT. Increased risk generally
stems from the anesthesia, the induced seizure, or the transient increase in CNS pres-
sure and blood flow. Relative contraindications include recent myocardial infarction,
active pulmonary inflammation, CNS tumors, increased intracranial pressure,
and recent carbon monoxide poisoning. (ECT occasionally has resulted in brain
damage in these cases.)
The risks that attend preparatory anesthesia include significant respiratory depres-
sion, allergy to premedications, and cardiac failure. The main adverse effects of ECT
are confusion and short-term memory loss (which may last five times as long in elderly
patients). Both risks are significantly reduced with unilateral treatments. 47
5. Technique
The clinician should inform patients about what they will experience during and
after the procedure: they will be asked to lie down, a sedative will be administered, and
they may wake up with some discomfort (and rarely with a sense of being paralyzed).
In addition, each ind\vidual should know the possible risks (particularly memory loss)
and treatment alternatives. The practitioner should obtain informed consent (from a
relative or guardian, if necessary) and should carefully document discussions and
procedures.
The procedure begins with the administration of an anticholinergic agent
(scopolamine or atropine), a short-acting anesthetic agent (usually a briefly acting
barbiturate), and a muscle relaxant [succinylcholine (Anectine@ and others)]. The
barbiturate usually takes about 1 min to work. During anesthesia, because of muscle
paralysis, assisted ventilation with oxygen is required. A physician then applies a brief
electrical stimulation. The electrical impulse is titrated to produce a generalized CNS
34 ALAN J. GELENBERG, M.D., AND PEDRO L. DELGADO, M.D.
seizure of about 40- to 60-sec duration. The strong muscle relaxant protects the patient
from physical injury but paralyzes the respiratory and limb muscles. Therefore, to
monitor the seizure, the clinician can tighten a blood pressure cuff on one arm, which
will prevent the succinylcholine from affecting that limb and allow the seizure activity
to be observed. Alternatively, the clinician may use an electroencephalogram (EEG)
tracing to observe the spike- and-wave pattern. After the seizure, recovery usually takes
about 10-20 min.
The number and frequency of treatments vary with the clinical needs of the
patient. Therapy should be administered until the patient experiences complete recov-
ery. The typical regimen nsually consists of every-other-day or twice-weekly treat-
ments for a total of 5-10 sessions.
Clinical research and enhanced ECT technology have improved our understanding
and technique in recent years. Newer ECT devices make it easier to monitor seizure
duration electroencephalographically, and noninvasive fingertip cuffs allow anesthe-
siologists to monitor arterial blood gases throughout the treatment. For all patients and
particularly for those receiving unilateral ECT, it is important to verify that a seizure
has occurred. Moreover, when the unilateral method is used, a suprathreshold stimulus
provides better efficacy than a lower-energy stimulus that barely triggers a convulsion.
Some practitioners locate the threshold stimulus needed to trigger a seizure on a
patient's first treatment and then subsequently administer stimulation at 100% greater
energy for bilateral electrode placement and 150-200% greater energy for unilateral
electrode placement. 48 An alternative approach is to monitor the seizure duration,
which optimally should be between 40 and 120 seconds. The energy of stimulation is
then adjusted upward or downward to keep the seizure duration within that time range.
Increasingly, bilateral stimulation is being recognized as more effective and more rapid
for most patients (particularly the elderly), with unilateral treatments reserved for those
particularly sensitive to the cognitive side effects of ECT or who have already shown
deterioration in cognition and memory.
v. PHARMACOTHERAPY
Several years later, research with structural analogues of the newly discovered
phenothiazines gave rise to imipramine (Tofranil®). Although its two-dimensional
configuration closely resembles that of its phenothiazine parents, minor changes in the
central "bridging" ring in this tricyclic compound, which gave the class its name,
resulted in alterations in its three-dimensional configuration. This, in tum, resulted in
major pharmacologic and clinical differences from the phenothiazines.
The remainder of the 1950s and the following two decades saw no comparable
breakthrough in antidepressant pharmacotherapy. Rather, a parade of "me-too" anti-
depressants entered the American market, driven largely by the financial needs of their
manufacturers. Discovered through the use of preclinical paradigms that virtually
ensured similarity to the prototypes, these agents possessed only minor to moderate
differences in side-effect profiles.
As the years went by, other countries began to introduce substantially different
drugs for treating depression. Then, in 1979, trimipramine (Surmontil®) was marketed
in the United States. There was some suggestion that it possessed a unique neurochemi-
cal and clinical profile, but the manufacturer has failed to pursue these possibilities
with definitive research; the best "hard" evidence suggests that trimiprarnine is just a
seventh "me-too" tricyclic. The years 1980, 1981, and 1982 saw the sequential intro-
duction of three somewhat different antidepressants: amoxapine (Asendin®),
maprotiline (Ludiomil® and others), and trazodone (Desyrel® and others). Although
each was introduced with considerable advertising ballyhoo, subsequent research and
clinical impressions suggest that these agents offer less than initially met our hopeful
eyes. By now, amoxapine and maprotiline are infrequently prescribed; trazodone is
used largely for its sedating properties.
The subsequent introductions into the U.S. market of nomifensine (no longer
available) and then bupropion (Wellbutrin®) were fraught with even greater complica-
tions. After less than a year, nomifensine was withdrawn worldwide because of its
association with a high rate of immunologic abnormalities, most notably hemolytic
anemia, which is potentially fatal. Although this experience highlighted potential pit-
falls in introducing new drugs, many clinicians have grieved the total withdrawal of
this agent, because some patients appeared uniquely responsive to it. It is possible that
neurochemical features distinguishing nomifensine from older compounds accounted
for these cases of distinct efficacy: most commonly mentioned is its enhancement of
central dopaminergic activity.
Bupropion spent a short time on the American market before it was withdrawn
because of its association with a high incidence of epileptic seizures. In 1989 it was
reintroduced, but with warnings about its effects on the seizure threshold. Since then, it
has demonstrated substantial efficacy "in the field," the seizure incidence appears
manageable, and its lack of sexual side effects has been a distinguishing feature.
Fluoxetine (Prozac® and others) was introduced into the United States at the
beginning of 1988 and rapidly enjoyed remarkable commercial and clinical success. In
addition to its apparently comparable efficacy to older antidepressants, two factors
were largely responsible for fluoxetine's popUlarity: fluoxetine has fewer side effects
than tricyclics, and it has an apparently simpler dosing schedule. Both factors made it
easier for primary-care physicians to prescribe this new agent.
As every educated layperson knows by now, fluoxetine began what is no less than
36 ALAN J. GELENBERG, M.D., AND PEDRO L. DELGADO, M.D.
explain its rare association with orthostatic hypotension and its relatively low incidence
of anticholinergic side effects, respectively. Published reports do not show greater
efficacy for mirtazapine over previously available antidepressants.
Clomipramine (Anafranil®), introduced in the United States in 1990 after many
years of availability abroad, is a potent serotonin reuptake inhibitor (SRI) but a non-
selective one. It was the first drug labeled for the treatment of obsessive-compulsive
disorder, and although SSRIs also appear efficacious for this condition, there are some
suggestions that clomipramine may be superior. On the "downside," its side effects-
including gastrointestinal upset, sexual dysfunction, and seizures-tend to be more
problematic. Because of diverse reactions and toxicity, the drug is not labeled for the
treatment of depression and is seldom used to treat depression in the United States,
although it probably is at least as efficacious against depression as other tricyclics.
Selegiline (Eldepryl®), a selective inhibitor of the B type of MAO, is available
now in the United States and labeled for the treatment of Parkinson's disease. Although
some have reported apparent efficacy of selegiline in the treatment of depression, this
has never been conclusively demonstrated. Moreover, at doses that may be efficacious
in depression, selegiline loses the specificity that makes it relatively immune from the
"cheese effect," by which exogenous tyramine can cause hypertensive reactions. Pa-
tients taking selegiline at any dose are at risk to develop the hypermetabolic reaction
("serotonin syndrome") if they concomitantly take a drug like meperidine (Demerol®
and others) or an SSRI. Serious and sometimes fatal reactions have also occurred when
patients have taken selegiline in combination with a tricyclic antidepressant.
Several reversible inhibitors of type A MAO (RIMAs) are currently in use abroad
and are being tested in the United States. These agents should have less risk of the
"cheese effect" than MAOIs and greater tolerance. However, there have been reports of
excitatory effects with the RIMA moclobemide and adverse interactions with cimetidine
(Tagamet®) and meperidine. RIMAs should elevate brain levels of the catecholamines
norepinephrine and dopamine, which could explain stimulantlike properties. They
should also raise brain serotonin levels, and the "serotonin syndrome" has been associ-
ated with moclobemide, especially in combination with the SRI clomipramine.
From the time the third edition of The Practitioner's Guide went to press until the
time of this writing, the new generation of antidepressants have substantially changed
the prescribing patterns of American physicians. Unless new (and unexpected) toxicity
emerges, or suggested problems of tolerance are confirmed and serious, the new
generation of antidepressants are likely to be employed more and more as first-line
antidepressant agents. Without doubt, they give us more options, and most patients can
take them with a greater degree of comfort, leading primary-care doctors in particular
to be more willing to prescribe them, and patients more willing to take them regularly.
Beyond the availability of more and different drugs, which has led to an increased
number of options for patients refractory to initial trials, practitioners who treat depres-
sion will only get to "the promised land" when we will have tests like the culture
sensitivity available in the treatment of infectious disease-tests that can tailor treat-
ment to the underlying problem. Only then will we be able to select a drug to treat a
patient's presumed neurochemical defect, thus sparing people unnecessary drug expo-
sure and prolonged empirical trials. For now, however, the greatest challenge lies in
38 ALAN J. GELENBERG, M.D., AND PEDRO L. DELGADO, M.D.
convincing the public and health care givers to recognize depression, appreciate its
profound human and economic costs, and learn about the high degree of efficacy of
currently available antidepressant drugs.
A. Heterocyclic Antidepressants
1. Chemistry
The chemical structures of the heterocyclic antidepressants currently available in
the United States are shown in Figs. 1-3. The chemical structure of the tricyclic
antidepressants (Fig. 1) differs from that of the phenothiazines only in the substitution of
a carbon-carbon double bond for the sulfur atom in the bridge between the two benzene
rings. This minor difference, however, changes the three-dimensional conformation of
the molecules, which in tum alters their pharmacological and clinical properties.
With two methyl groups on the terminal nitrogen atom of the side chain, im-
ipramine, amitriptyline (Elavil® and others), trimipramine, doxepin (Sinequan® and
others), and clomipramine are called tertiary amines. The demethylation of imipramine
and amitriptyline results in, respectively, desipramine and nortriptyline (Pamelor® and
others), which also are active antidepressants. With a hydrogen substituted for one of
the methyl groups on the terminal nitrogen, desipramine and nortriptyline, along with
protriptyline (Vivactil®), are known as secondary-amine tricyclics. By and large, the
secondary-amine tricyclics have weaker anticholinergic properties and other neuro-
chemical differences that affect their clinical profile and possibly their blood level
relationship as well. Much as the addition of a chlorine atom to the phenothiazine
promazine (Sparine®), resulting in chlorpromazine (Thorazine® and others), produced
an agent with distinctive pharmacologic and clinical properties, so the addition of a
chlorine atom to imipramine, producing clomipramine, has altered its pharmacologic
and clinical properties. This has resulted in an agent that is more highly serotonergic
and possibly has dopamine-blocking properties, and one that appears to be effective in
alleviating obsessions and compulsions.
Amoxapine also has a three-ring structure, but it is a demethylated metabolite of
the antipsychotic loxapine, which it resembles pharmacologically and clinically; both
are termed dibenzoxazepines. Maprotiline has a fourth ring and is, therefore, termed a
tetracyclic, but it has few pharmacologic or clinical differences from the tricyclics.
Trazodone is a phenylpiperazine derivative of triazolopyridine-chemically, phar-
macologically, and clinically distinct from the tricyclic antidepressants.
As noted above, there are four SSRI antidepressants (Fig. 3). Fluoxetine's chemi-
cal structure makes it a straight-chain phenylpropylamine. Sertraline is a
naphthylamine derivative, paroxetine is a phenylpiperidine, and fluvoxamine is one in
a series of 2-aminoethyl oximes of aralkyl ketones. Although each SSRI is structurally
unique, they all have an electron-dense phenyl ring four to five carbons over from a
nitrogen. It may be this structural similarity that enables the inhibition of serotonin
reuptake.
Different from other antidepressant drugs is bupropion. This drug is structurally
related to phenylethylamine, which probably accounts for some of its stimulant proper-
....
~ CHCH2CH2N(CH3)2 • HCI
0c:J0 I
oyo CHCH2CH2NHCH3 • HCI
CH2CH2CH2N(CH3)2 • HCI
HaC\
C2HS
N) )=~
o-OCH.cH2~N""CH2CH2CH2"N~N<O(· HCI
otb CI
CH 3
N/
~-CH.cH.cH.rQ<O(
H -HCI
-\ HCI CI ;o45 3
CH30
6
Fluoxetine Hydrochloride
F-Q-CNH
Paroxetine
Sertraline Hydrochlooride
2. Mechanism of Action
All effective antidepressant medications and ECT have potent pharmacologic
effects on monoamine neurotransmitter systems. While some drugs inhibit the reuptake
of monoamines, other such as mianserin* and nefazodone primarily block neuro-
transmitter receptors. Potent serotonin reuptake (fluoxetine), norepinephrine reuptake
(desipramine), and dopamine reuptake (nomifensine*) inhibitors effectively treat de-
pression. Current thinking is that these effects on monoamine systems underlie the
initiation of antidepressant action. Through diverse pharmacologic properties, this
group of drugs increases net transmission through either serotonergic, noradrenergic, or
dopaminergic systems. This is the case even for drugs that act as receptor antagonists
(such as nefazodone). Because stimulation of serotonin 2A receptors inhibits the effects
of serotonin on the serotonin lA receptor, when nefazodone blocks the serotonin 2A
receptor, it actually potentiates the effects of serotonin on the serotonin lA receptor.
Recent evidence from clinical studies suggests that while increased synaptic levels
of monoamines is not temporally correlated with therapeutic response, it is an essential
aspect of the mechanism of action of some medications. Selective depletion of seroto-
nin49 ,5o or norepinephrine and dopamine 51 causes a rapid relapse of depression in
depressed patients who have had therapeutic responses to and are being maintained on
antidepressant drugs. The relapse is qualitatively similar to the depression prior to
treatment, it correlates with the time course of monoamine depletion, and the patients
improve rapidly as monoamine levels return to baseline levels. Depleting 5-HT causes
a rapid return of depression in depressed patients who have responded to fluoxetine,
fluvoxamine, MAOIs, and imipramine but not those who have responded to desi-
pramine, nortriptyline, or bupropion. 50,52 Depleting norepinephrine and dopamine
causes a rapid return of depression in those depressed patients who have improved on
desipramine but not in those who have improved on fluoxetine. 51 These studies under-
score the importance of monoamine reuptake inhibition and suggest that more than one
mechanism of action exists.
Monoaminergic systems in the brain modulate many behaviors, cognitions, and
somatic functions. By altering the monoaminergic systems, antidepressant drugs poten-
tly affect the core brain circuits involved in modulating stress responses. Antidepres-
sant drugs could increase neurotransmission in normal monoaminergic systems, lead-
ing to a beneficial impact on dysfunctional brain networks and restoring more
appropriate overall function. This may account for the fact that antidepressants are
effective in the acute treatment of milder mood disorders such as dysthymia, as well as
in generalized anxiety disorder, panic disorder, social phobia, obsessive-compulsive
disorder, bulimia and anorexia nervosa, posttraumatic stress disorder, and possibly
alcohol abuse. Perhaps antidepressant drugs "nonspecific ally" restore function irre-
spective of the underlying pathology, analogously to corticosteroids reducing inflam-
mation regardless of etiology.
We probably will not find a single mechanism for all antidepressants. In addition
to the existence of depressive SUbtypes, each distinct syndrome may result from a
complex interaction of presynaptic, neurohumoral, and postsynaptic receptor properties
of specific neural pathways. As Maas postulated more than 20 years ago, antidepres-
sants may act in various complicated ways to restabilize an unbalanced system. 53
3. Adverse Reactions
The clinician should know a drug's adverse reactions and a patient's suscep-
tibilities. Some of these effects cause minor discomfort, but others may interfere with
drug trials or produce serious morbidity. Of the common reactions, cardiac, anti-
cholinergic, and CNS effects produce the most clinical morbidity. Untoward effects
occur more often when cyclics are coadministered with other drugs or when a patient
has a coexisting physical disorder.
Overall, tricyclic antidepressants resemble one another in their adverse-effect
profile. In general, secondary-amine tricyclics are less anticholinergic and sedating and
produce less weight gain and postural hypotension. Maprotiline's and amoxapine's
adverse effects are comparable to those of the tricyclics. However, amoxapine also
resembles the neuroleptics in producing both extrapyramidal reactions and neuroen-
docrine changes.
Trazodone presents a somewhat distinct clinical profile, certainly different from
that of the tricyclic antidepressants. Highly sedating, trazodone produces postural
hypotension and some gastric irritation. It also is associated with occasional cases of
priapism and can exacerbate preexisting myocardial irritability. It is unlikely, however,
to produce anticholinergic effects and appears very safe in overdose. Chemically re-
2. DEPRESSION 43
lated to trazodone, nefazodone differs somewhat in its side-effect profile. The most
common adverse effects with nefazodone are headache, dry mouth, nausea, drowsi-
ness, dizziness, constipation, lack of energy, insomnia, diarrhea, and light-headedness.
Nefazodone does not appear to affect sexual function adversely nor to cause priapism.
It may cause postural hypotension. Initial activation is a problem for some patients, and
cognitive dulling can be bothersome, especially in the elderly.
The SSRIs cause nausea, headache, nervousness, and insomnia more commonly
than the tricyclics but generally do not cause acute weight gain, are less likely to cause
anticholinergic effects or orthostatic hypotension, and are much safer in overdosage.
With prolonged use, some patients do gain weight on SSRI treatment. Agitation and
increased anxiety may occur in the first weeks of treatment and subside later. All SSRIs
appear to cause a high incidence of anorgasmia and other sexual dysfunction. Akathisia
can be troublesome and may be more common with fluoxetine. Many antidepressants
are metabolized by the liver's P450 enzyme systems and can alter the pharmacokinetics
of other drugs (see Table 1 in Chapter 1 and Section V.A.3.k). SSRIs, in particular, can
raise blood levels of many other pharmaceuticals, thus magnifying their possible side
effects.
Bupropion may cause agitation and, rarely, seizures but causes fewer anti-
cholinergic effects than the tricyclic antidepressants, little if any orthostatic hypoten-
sion, and a relatively low incidence of sexual dysfunction. Venlafaxine generally is
similar to the SSRIs in its adverse effects, but it has been associated with a sustained
increase in diastolic blood pressure. Nausea may limit rapid dose increases.
Transient somnolence occurs in more than 50% of patients taking mirtazapine,
and increased appetite and weight gain are also common problems. Bone marrow
suppression has occurred in a few patients. There are unsubstantiated suggestions that,
unlike other antidepressants, mirtazapine has a low incidence of adverse sexual side
effects. However, this has not been systematically studied.
b. Hematological
i. Leukocytic Effects and Purpura. Heterocyclic antidepressants only rarely pro-
duce alterations in white cell count. Clinicians have reported leukocytosis, leukopenia,
Loeffler's syndrome, eosinophilia, thrombocytopenia, and purpura. These conditions
should be medically evaluated, although they seldom cause significant morbidity. The
patient can usually continue the medication at the same dosage.
ii. Agranulocytosis and Other Hemolytic Effects. Both the heterocyclic anti-
depressants and the phenothiazines can cause agranulocytosis, but the incidence is
much lower with antidepressants than with antipsychotics. Death from infection sel-
dom occurs if the drug is discontinued immediately and the patient placed immediately
in reverse isolation. White blood cell numbers usually return to normal within several
weeks of discontinuing the drug. Most reported cases have occurred with imipramine
(although this may be an artifact, since it is the drug most often studied).
Agranulocytosis seems to be an allergic response of sudden onset that usually
appears 40-70 days after initiation of the medication. Patients have a greater risk with
advancing age and concomitant physical illnesses. The clinical syndrome is charac-
terized by a low white count (composed almost completely of lymphocytes), normal
red count, and infection (usually involving the oropharynx) accompanied by fatigue
and malaise.
Routine blood studies do not help with early detection, since the syndrome devel-
ops very rapidly. Instead, the clinician should evaluate and treat any sign of infection,
especially of the pharynx. He or she should discontinue the heterocyclic and never
administer it to that patient again. The practitioner might try an antidepressant with a
different structure and follow the patient closely for adverse hematological effects.
It is unclear whether antidepressants ever cause aplastic anemia. Nomifensine was
withdrawn from the market worldwide because of its association with hemolytic ane-
mia. The tetracyclic mianserin, widely used in England, has been associated with a
higher incidence of bone marrow suppression than tricyclic antidepressants.
atrioventricular conduction system, between the bundle of His and the ventricles, and is
clinically inconsequential for a majority of patients. Patients at risk include those with
second-degree heart block or with right or left bundle branch block, who may develop a
brady arrhythmia when treated with a tricyclic antidepressant. If the physician wishes to
prescribe a tricyclic for such a patient, the initial dose should be low, and dosage should
be increased very gradually. A cardiologist should follow the patient and should con-
duct Holter monitoring before and after each dose increment. Newer antidepressants
and MAO inhibitors do not slow cardiac conduction.
iv. Antiarrhythmic!Arrhythmogenic Activity. In overdose, with toxic blood levels,
tricyclic antidepressants have long been associated with malignant ventricular ar-
rhythmias, often leading to cardiac arrest. Within a broad range of therapeutic concen-
trations, however, tricyclic antidepressants possess a quinidinelike (type IA) antiar-
rhythmic activity. Thus, in patients with frequent atrial or ventricular premature
contractions, a tricyclic antidepressant will usually result in diminished extra beats. The
physician should be wary, of course, when coprescribing a tricyclic antidepressant with
another type IA antiarrhythmic, such as quinidine, procainamide (Pronestyl® and oth-
ers), or disopyramide (Norpace® and others), lest additive toxicity occur.
Evidence from a recent cardiac arrhythmia suppression trial (CAST), which found
increased mortality among cardiac patients on long-term treatment with type I antiar-
rhythmics, has raised new concerns about the use of tricyclic antidepressants in cardiac
patients. 57 Although these agents do, in fact, suppress extra cardiac beats, when cardiac
ischemia and anoxia occur, patients taking type I antiarrhythmics appear to die at a
significantly greater frequency than those taking placebo. This has led Glassman and
Roose, who have conducted research on the cardiac effects of tricyclics for many years,
to caution about the long-term use of tricyclic drugs in patients vulnerable to suffering
myocardial ischemia. 57 Newer antidepressant agents do not appear to carry this risk,
but, as Glassman and Roose have noted, their efficacy in depression of greater severity
may not be comparable to that of the tricyclics in some patients. ECT may be consid-
ered for some patients in this category.
There is no evidence that nontriCYclic antidepressants possess antiarrhythmic
capacities. Trazodone may trigger ventricular tachycardia in patients with preexisting
myocardial irritability. Because most drugs introduced to the market have been studied
only in patients with excellent health status, it may be some time before possible risks
of arrhythmogenicity are fully identified in newer antidepressant drugs. 58 ,59 However,
the SSRIs and bupropion have been used now for a long time in a lot of patients,
including patients with various cardiac conditions, without evidence of either antiar-
rhythmic or arrhythmogenic effects.
Guidelines for Clinical Use
1. Take a good medical (including cardiac) history, as well as a review of
symptoms.
2. Discuss possible cardiovascular reactions with the patient.
3. In young adult patients with no apparent cardiac problems, the choice of
drug need not take into account cardiovascular side effects. If these occur,
however, they should affect choice of subsequent treatments.
48 ALAN J. GELENBERG, M.D., AND PEDRO L. DELGADO, M.D.
the other drugs, sometimes to toxic levels. Interactions may also occur with
paroxetine but seem to be less of a problem with sertraline, at least at the
lower end of its dosage range (see Chapter 1, Table 1).
10. When treating depression in a patient with significant cardiovascular dis-
ease, who is likely to be taking one or more cardiotonic medications, work
closely with the patient's cardiologist. Consider consultation with a psychia-
trist well versed in psychopharmacology.
g. Neurological
i. Sedation/Stimulation. Tricyclic antidepressants tend to be sedating; tertiary
amines are even more so than their secondary-amine metabolites. Doxepin and amitrip-
tyline are particularly sedating, as is the nontricyclic trazodone. While SSRIs and other
newer antidepressants cause stimulation more often than sedation, many patients do
experience the latter, sometimes to a troublesome degree. Among newer antidepres-
sants, sedation may be problematic with nefazodone and mirtazapine. In some cases,
sedation becomes more of a problem after a patient has been taking an antidepressant
for weeks or several months.
If sedation is a problem during the first few hours following ingestion of an
antidepressant, advise the patient to take it at or soon before bedtime. Morning-after
sedation (hangover) may be a problem initially, but tolerance usually develops to the
sedative side effects of any drug. If reassurance and the use of bedtime dosing do not
alleviate sedation sufficiently, try lowering the dose. Since it is unclear which patients
will experience significant sedation from an antidepressant (except for the tertiary-
amine tricyclics, which are sedating for just about everyone), consider having the
patient take the first dose or two of an antidepressant on a weekend or holiday, when he
or she does not have to go to work and can experiment with the drug's side effects. Be
certain also that the patient has a phone number to reach a knowledgeable clinician
about early and possibly idiosyncratic reactions to the medicine. (Easy access to
information, reassurance, and guidance can enhance compliance and, therefore, effec-
tiveness of medications and decrease liability exposure.)
When a patient has been taking an antidepressant for a number of weeks or a few
months and then experiences a gradual increase in sedation and decrease in energy, a
downward adjustment in dose may be called for. Particularly with a drug with a long
half-life, such as fiuoxetine, steady-state levels (which typically take more than a
month to achieve) may be associated with the side effects, which signal the need to
lower the dose for longer-term treatment.
Fluoxetine (and to a lesser degree other SSRIs), bupropion, venlafaxine, and
nefazodone stimulate many patients, particularly initially. In such cases, ask the patient
to take more of the daily dose earlier in the day. Begin with the lowest possible dose
and raise the daily dose slowly. If the patient has difficulty falling asleep, the last dose
should be in the late afternoon. As with sedation, tolerance tends to develop to the side
effect of unwanted stimulation and difficulty falling asleep. Consider a dosage reduc-
tion if necessary, with very gradual dosage increments. In some cases, consider adding
a benzodiazepine or trazodone at bedtime for a short period of time to facilitate sleep.
A benzodiazepine during the daytime may help diminish excessive feelings of agitation
50 ALAN J. GELENBERG, M.D., AND PEDRO L. DELGADO, M.D.
Amoxapine is a neuroleptic drug and, as such, has been associated with the full
range of neuroleptic extrapyramidal reactions (see Chapter 4). These include acute
dystonias and dyskinesias, akathisia, Parkinson's syndrome, and late-onset and with-
drawal dyskinesias and dystonias.
There have been suggestions that fluoxetine may cause akathisia, although there is
some controversy as to whether the motor restlessness associated with nonsedating
antidepressants may not be a distinct condition. 62 The highly controversial (and highly
public) reports suggesting that fluoxetine might cause or exacerbate thoughts of suicide
(now largely disbelieved among psychiatrists) have been linked in at least some cases
by some authors to akathisia. 63 It is conceivable that some patients develop akathisia
while taking SSRI antidepressants (analogous to the same syndrome caused by neuro-
leptics) and experience feelings of desperation as a result of these feelings of wanting
to "jump out of their skin." There are also anecdotal data suggesting that when com-
bined with a neuroleptic, fluoxetine might enhance the probability of extrapyramidal
reactions. 62
vii. Withdrawal Reactions. Tricyclic antidepressants are associated with a distinct
withdrawal syndrome. As with most psychoactive compounds, the likelihood of a
withdrawal reaction and its severity vary as a function of the dose and the length of
time the patient has taken the drug at the time of withdrawal. It may be mitigated in part
by very gradual withdrawal.
Symptoms typically begin when the dose has been reduced to a particularly low
level or the drug is stopped entirely. Clinical manifestations that have been attributed to
anticholinergic rebound include both peripheral and central effects. Autonomic reac-
tions, the opposite of so-called anticholinergic symptoms, include hypersalivation,
nausea, vomiting, abdominal cramps, and diarrhea. Central manifestations typically
include headaches, vivid dreams, and sleep disturbances. Especially when patients have
taken these drugs chronically, withdrawal symptoms may persist for 2 months or
longer.
Symptoms associated with withdrawal from SSRIs include dizziness, light-head-
edness, headaches, confusion, memory difficulties, decreased energy, weakness, ex-
tremity tingling, chest tightness, tremor, nausea, anorexia, insomnia, nightmares, cold
hands, and sweating. The proposed mechanism is a decrease in serotonin. There is
insufficient data to describe withdrawal syndromes with some of the newer antidepres-
sants, but the likelihood is that they exist. Withdrawal symptoms with venlafaxine may
be particularly severe.
Gradual withdrawal of the antidepressant is the best preventive and management
strategy. At the lower end of the dosage scale, consider dividing tablets in half. An
alternative (or additional) strategy is to end the drug by administering a dose every
other day, then every 3 days, before discontinuing it entirely. If there is no rush to
discontinue the agent, tapering of the drug can be protracted over several months,
partjcularly in patients who have previously taken the antidepressant for years.
When symptoms are particularly uncomfortable, the patient can be returned to a
higher dose of the antidepressant and then the taper made more gradual. Antidotes to
tricyclic withdrawal can include a sedating antihistamine with anticholinergic proper-
ties [e.g., diphenhydramine (Benadryl® and others)], a less sedating centrally active
52 ALAN J. GELENBERG, M.D., AND PEDRO L. DELGADO, M.D.
anticholinergic agent [e.g., benztropine (Cogentin® and others)], or, if peripheral auto-
nomic manifestations predominate, a peripherally active anticholinergic drug [e.g.,
methscopolamine (Pamine® and others)].
viii. Psychiatric Side Effects. It seems that virtually any effective antidepressant
can precipitate mania in some patients. Many such patients have a prior history of
mania and qualify for a bipolar diagnosis. Occasional patients, however, experience
mania only on "provocation" with an antidepressant drug, and some authorities have
suggested that these patients fall clinically along a bipolar spectrum. In DSM-IV,
however, a patient who has experienced an episode of mania or hypomania only while
taking an antidepressant drug does not qualify for a bipolar (or bipolar II) diagnosis.
Instead, the diagnosis is of mania or hypomania "due to" an antidepressant.
The psychotic symptoms of patients with schizophrenia may be made worse by
treatment with an antidepressant drug. When a full-blown depression is present in a
schizophrenic patient, however, it is acceptable to administer an antidepressant drug
along with an antipsychotic agent. Bupropion has been reported to exacerbate psycho-
tic symptoms in patients with schizoaffective disorder; this may be related to its
inhibition of neuronal dopamine uptake. 64
Patients with organic brain syndromes may become increasingly confused when
treated with a strongly anticholinergic drug such as a tricyclic antidepressant. Use of
antidepressants with minimal anticholinergic activity, however, may be considered.
A controversial article published in 1990 suggested that, in some patients, anti-
depressant treatment with fluoxetine could lead to de novo thoughts of suicide. 65 Many
other case series and retrospective analyses have failed to confirm an association
between fluoxetine or other antidepressants and new or exacerbated thoughts of sui-
cide. The prevailing wisdom is that untreated (or undertreated) depression carries a
much greater risk of suicide. Nonetheless, as noted earlier (see Section V.A.3.g.vL),
there may be occasional patients who react to side effects of a drug (such as akathisia)
by becoming deeply distressed and thinking about suicide. Clinicians must be continu-
ously alert to how patients may react to a new factor in their lives, such as a drug.
Common side effects should be predicted, and any patient newly starting on a medica-
tion should be instructed to call immediately and report any distressing reactions. A
knowledgeable clinician should be available 24 hours a day. Some patients are dis-
tressed that they do not feel better immediately and can be reassured. Others are deeply
troubled by one or another side effect of a new medicine and can be comforted by
information about the drug.
Sedation Anticholinergic
High High
Amitriptyline Amitriptyline
Trimipramine Imipramine
Doxepin Trimipramine
Trazodone Doxepin
Moderate Moderate
Imipramine Amoxapine
Amoxapine Nortriptyline
Nortriptyline Maprotiline
Maprotiline Low
Sertraline Desipramine
Low Protriptyline
Desipramine Very low
Protriptyline Trazodone
Fluoxetine Fluoxetine
Bupropion Bupropion
Paroxetine Nefazodone
Fluvoxamine Venlafaxine
Venlafaxine Sertraline
Paroxetine
Fluvoxamine
may cause discomfort, mouth infections, and dental caries (sometimes rampant). In
severe cases of xerostomia, the clinician should intervene by:
1. Encouraging adequate hydration.
2. Lowering the dosage of medication if possible.
3. Discouraging the use of sugar-laden drinks, chewing gum, or candy.
4. Considering switching to a less anticholinergic (i.e., nontricyclic) antidepres-
sant.
5. Providing alternative means of mouth lubrication:
a. "Sugarless" gums (although these actually have sugars that in high doses
can contribute to the formation of caries).
b. Fluoride lozenges.
c. Direct lubricants (for a list of saliva substitutes, see Table 3).
6. Administering bethanechol chloride (Urecholine® and others), a peripherally
active cholinergic agent, 25 mg up to 3 times per day, in patients whose
symptoms do not respond to conservative measures, who would not be able to
complete a drug trial because of the severity of symptoms, and who are not at
increased risk from cholinergic stimulation (e.g., patients with asthma, poten-
tial cardiac compromise, or ulcer disease).
The major cardiovascular anticholinergic effect is tachycardia. Tolerance to this
reaction usually develops quickly. Blurred vision, resulting from failure of accom-
54 ALAN J. GELENBERG, M.D., AND PEDRO L. DELGADO, M.D.
Drug Form
aReprinted from The Medical Letter l13 with permission of the publisher.
bPreservative-free.
"Sodium-free.
j. Sexual Side Effects. Psychoactive drugs frequently interfere with sexual desire
and function, and antidepressant agents are no exception. The mechanism of action
remains obscure, although the interplay of various autonomic effects must bear some
responsibility.
Delayed, incomplete, or altered orgasm is reported occasionally by both men and
women. Decreased libido also is sometimes described by both sexes. In addition, men
may experience difficulties achieving or maintaining a penile erection and/or seminal
ejaculation. Rates of sexual side effects seem to be highest with SSRIs and
clomipramine, followed by MAOIs and tricyclics, and least common with bupropion
and possibly nefazodone. In one study, clomipramine was found to be associated with
sexual side effects in over 90% of patients.66 The incidence of sexual dysfunction with
fluoxetine has been variably reported as 8% to 75%,67-70 although prerelease studies
found an incidence of only 1.9%.71 Reported rates of sexual dysfunction with sertraline
are 15.2% for men and 1.7% for women; for paroxetine, they are 23% for men, 13% for
women. Due to underreporting and based on clinical experience, however, the "true"
incidences are probably much higher.
Strategies for treating adverse sexual effects include (1) waiting for tolerance to
develop, (2) reducing the antidepressant dose, (3) changing to a different antidepressant
(especially bupropion72), and (4) adding a new medication. A weekend without the
antidepressant can restore sexual functioning temporarily for short-half-life agents (not
fluoxetine), but with a risk of increased symptoms of depression.?3
Adjunctive medications found to be helpful include the peripherally active para-
sympathomimetic bethanechol (Urecholine® and others), the serotonin antagonist cy-
proheptadine (Periactin® and others), the a-2 antagonist yohimbine (Yocon® and oth-
ers), amantadine, bupropion, and buspirone. None of these has been systematically
studied in controlled trials, but single cases and small, open series have reported
varying degrees of success. Cyproheptadine has been reported to reverse antidepres-
sant-induced anorgasmia in both sexes and impaired ejaculation in men. 74 - 81 An-
orgasmia from tricyclic antidepressants, one MAOI, and fluoxetine has been suc-
cessfully treated. In the majority of successful cases, patients used cyproheptadine on
an as-needed basis, taking 2-16 mg a few hours before sexual activity. It was also
effective when taken in a daily dose. Cyproheptadine has reversed the effects of the
antidepressant in seven cases. Yohimbine has been reported to ameliorate decreased
libido, erectile dysfunction, and anorgasmia caused by fluoxetine and clomipra-
mine. 67 .82 ,83 Effective doses have ranged from 5.4 to 16.2 mg taken as needed 2-4 hr
before sexual activity or 5.4 mg taken 3 times daily. Bethanechol has reversed the
sexual side effects of anorgasmia, impotence, and impaired ejaculation caused by both
tricyclic antidepressants and MAOIs with doses of 10-40 mg as needed or 30-100 mg
daily.84-88 Amantadine reversed fluoxetine-induced anorgasmia in five patients taking
56 ALAN J. GELENBERG, M.D., AND PEDRO L. DELGADO, M.D.
k. Drug Interactions. Drug interactions fall into two broad categories: phar-
macokinetic and pharmacodynamic. The former occur when a drug affects another
drug's absorption, distribution, biotransformation, or excretion. Pharmacodynamic in-
teractions, by contrast, occur when two drugs have additive (or antagonistic) effects at
a target site. Thus, drugs with anticholinergic properties can be additive with the
anticholinergic properties of tricyclics. Hypotension can be a problem if another hypo-
tensive agent is added to an antidepressant that can lower blood pressure-e.g., tri-
cyclics or nefazodone. Other side effects increased by pharmacodynamic interactions
between drugs include sedation, stimulation, and sexual dysfunction.
Virtually all drugs with serotonergic properties-SSRls, venlafaxine, nefazodone,
and strongly anticholinergic tricyclics such as clomipramine-can produce a poten-
tially fatal hypermetabolic ("serotonin") syndrome when coadministered with (or pre-
scribed in close temporal proximity to) a MAOI. (For more on this syndrome, see
Section Y.B.3.b.)
A number of pharmacokinetic interactions are observed with modem antidepres-
sants. A growing understanding about the liver's P450 drug-metabolizing system (see
Table 1 in Chapter 1) has helped to clarify the nature of many of these. All SSRIs
except fluvoxamine are potent inhibitors of the P4502D6 isoenzyme. (Sertraline is less
potent at the lower end of its dosage range-50 mg daily.) Therefore, when SSRls are
coadministered with tricyclic antidepressants and some anti psychotics and antiar-
rhythmic agents, the levels of these other drugs may rise. The package inserts for
fluoxetine and sertraline caution against administering these SSRls with drugs that are
metabolized by P4502D6 and have a low therapeutic index, such as the tricyclics,
propafenone, and flecainide. Fluvoxamine is a potent inhibitor of the P4501A2 iso-
enzyme and can raise the levels of coadministered theophylline, haloperidol, clozapine,
some other antipsychotics, and tricyclics. Nefazodone inhibits the P4503A4 iso-
enzyme. It can thereby raise blood levels of cisapride and the nonsedating anti-
histamines terfenadine and astemizole and is considered contraindicated with these
three agents, because elevated blood levels can slow cardiac conduction and produce
potentially fatal ventricular arrhythmias. Other drugs whose levels can be raised by
nefazodone through its inhibition of the 3A4 system include the benzodiazepines
alprazolam, triazolam, and midazolam, as well as carbamazepine, quinidine, and lido-
caine. Many clinicians will lower the dose of coadministered drugs in anticipation of
this effect. Although it has not been definitely demonstrated that fluvoxamine inhibits
2. DEPRESSION 57
P4503A4, a substantial interaction has been observed between fluvoxamine and al-
prazolam, a drug that is known to be metabolized by the 3A4 isoenzyme. Therefore,
like nefazodone, fluvoxamine is contraindicated with terfenadine and astemizole and
should be used with caution with cisapride.
Mirtazapine is metabolized by several cytochrome P450 isoenzymes, including
2D6, lA2, and 3A4. Concomitant use of an inhibitor of one of these isoenzymes could
increase mirtazapine levels. Because mirtazapine does not inhibit P450 isoenzymes, it
probably has no effect on other drugs metabolized by them. If mirtazapine is combined
with alcohol or benzodiazepines, it may produce additive effects on sedation and
cognitive and motor impairment. Although there are no data on the combination of
mirtazapine with MAOIs, the manufacturer recommends that mirtazapine not be used
in combination with an MAOI or within 14 days of discontinuing one.
Fluoxetine, sertraline, and paroxetine (but not fluvoxamine and venlafaxine) are
highly bound to plasma proteins and can displace other drugs from their protein
binding sites, thus increasing their blood levels. This may effect the clinical activity of
such diverse agents as warfarin and carbamazepine.
Having available an increased number of antidepressants with different phar-
macologic profiles brings many benefits but also increases the potential for drug
interactions. Knowing the underlying basis for these potential interactions can help
clinicians to rationalize treatment regimens. In general, the addition or deletion of any
drug from a medication regimen should raise interaction as a possibility. Among
available antidepressants, as of now, venlafaxine appears the least likely to cause
interactions with other medicines.
for several days following delivery. These symptoms do not seem to cause long-term
morbidity. Therefore, the pregnant patient with a severe depression accompanying a
serious suicidal risk or an inability to care for herself might receive a trial of a tricyclic
antidepressant. The clinician and patient should carefully weigh the risks and benefits
(see Chapter 10). The effects of MAOI and newer antidepressants on the developing
fetus are less clear; if possible, these drugs should be avoided in a pregnant woman.
ECT is an option to be seriously considered when a pregnant patient develops a
depression of moderate or greater severity.
5. Acute Toxicity
statistically significant association between serum drug levels and the occurrence of
seizures or ventricular arrhythmias but did observe a correlation between QRS duration
and overdose outcome. 91 Patients with QRS durations of <0.1 sec had neither seizures
nor ventricular arrhythmias, whereas those with QRS 2:::0.1 sec had a 50% incidence of
ventricular arrhythmias. By contrast, another study failed to find either blood anti-
depressant concentrations or the EKG helpful in predicting seizures or ventricular
arrhythmias. 92 Still, all patients in this latter study with tricyclic antidepressant serum
levels of 2:::1000 ng/mL had QRS intervals of 2:::0.1 sec, and 2 cases that resulted in
fatalities of 102 consecutive tricyclic overdoses met these criteria.
The authors of the latter study recommend that an antidepressant level of 2::: 1000
ng/mL with a QRS interval of 2:::0.1 sec should be taken as a danger sign but that a
lower blood level or a normal QRS interval should not be a license to relax. Rather,
antidepressant blood level and QRS interval should be considered together with the
duration of time that has elapsed since the drug ingestion, the level of the patient's
consciousness, the presence of seizures, and the patient's vital signs in deciding on the
length of observation and the intensity of management.
Management of serious tricyclic overdoses poses serious difticulties, since
cardiorespiratory collapse, arrhythmias, and coma present both an immediate
and a persistent life threat (see Table 4). Any patient with a suspected tricyclic
antidepressant overdose-no matter how small the amount alleged to have been taken
or how alert the patient seems-should be rushed to an emergency room. Many
fatalities occur before a patient even reaches medical help. In the hospital, an IV line
should be established immediately, and the stomach emptied by gastric aspiration and
lavage. Next, a physician should administer 50-100 mg of activated charcoal, either as
an oral slurry mixed with water or by orogastric tube; it is too thick to administer via a
nasogastric tube. The clinician should avoid mixing it with milk, jelly, or other food-
stuffs, which can interfere with the charcoal's adsorptive capacity. Repeated adminis-
tration of charcoal every 4 hr can help remove additional quantities of the antidepres-
sant, which is recirculated via the enterohepatic system.
Catharsis can further speed elimination of the drug. Some preparations of charcoal
come mixed with sorbitol, which serves this purpose. An alternative cathartic is magne-
sium sulfate, 250 mg/kg per dose as a 20% solution.
An EKG should be performed immediately when the patient arrives at the emer-
gency room. If the patient has altered consciousness, arterial blood gases should be
drawn. Blood should also be drawn to establish antidepressant concentrations (which
may later rise if absorption from the gastrointestinal tract continues), to complete tests
to rule out the presence of other toxins (most overdoses are multiple), and to identify
any coincident medical conditions or complications.
The crucial time for observation in antidepressant overdoses is the first 6 hours. If
a patient has no alterations in consciousness, EKG abnormalities, respiratory depres-
sion, seizures, or hypotension, he or she can be given a final dose of charcoal and be
transferred to a psychiatric unit or given an appointment for psychiatric follow-up. On
the other hand, the presence of any clinical sign of toxicity should prompt immediate
admission to a medical intensive-care unit for at least 48 hours of cardiac monitoring.
Treatment of life-threatening consequences of heterocyclic antidepressant over-
60 ALAN J. GELENBERG, M.D., AND PEDRO L. DELGADO, M.D.
Initial measures
Induce emesis and catharsis (magnesium sulfate, 250 mg/kg in 20% solution)
Implement gastric aspiration and lavage with activated charcoal (Charcodote® powder), 50-100 mg,
every 4 to 6 hr for 24 to 48 hr
Complete physical examination and obtain vital signs
Adequately ventilate; monitor by EKG
Insert an intravenous line with cardiac pacing capacity
Insert pulmonary artery catheter
Obtain blood chemistries including antidepressant serum level
Maintenance procedures
Monitor vital signs, electrolytes, serum heterocyclic levels, and EKG
Maintain fluid and electrolyte balance
Ensure adequate ventilation and skin care if the patient is comatose
Hypertension
Elevate legs
Adequately hydrate
Administer phentolamine (Regitine®), 5 mg (if persistent and severe)
Seizures
Administer diazepam (Valium® and others), 5-10 mg 1M or IV pm (avoid barbiturates)
Central and peripheral anticholinergic syndrome
Slowly inject physostigmine (Antilirium®), 1-2 mg IV every 30 to 60 min pm
Hyperpyrexia
Use ice mattress, ice packs, or cold sponges
Cardiac interventions
Avoid type A antiarrhythmics [procainamide (Pronestyl® and others), quinidine, and disopyramide
(Norpace® and others) I
Use lidocaine (Xylocaine® and others) for ventricular arrhythmia
Use lidocaine or phenytoin (Dilantin® and others) for heart block arrhythmia
Use physostigmine for supraventricular arrhythmia; avoid in cases of cardiovascular or respiratory
compromise
Implement volume expansion for decreased left atrial pressure
Use dopamine (Intropin®) for increased left atrial pressure
Alkalinize the urine by giving IV NaHC0 3 (acid imbalance may predispose to ventricular irritability)
Persistent arrhythmia with heart failure
Use cardiac pacing or cardioversion
doses is complex and incompletely formulated. Because of their high degree of li-
pophilicity and extensive tissue and protein binding, heterocyclic antidepressants are
minimally removed by dialysis. The key medical treatments are symptomatic and hinge
on treating (or preventing) seizures, respiratory arrest, cardiac arrhythmias, and resul-
tant complications, such as renal failure and brain damage.
Seizures are often treated with intravenous diazepam, 5-10 mg repeated as
needed. Seizures may be more common with maprotiline and amoxapine overdose.
Amoxapine-related seizures, in particular, appear difficult to treat. Very high doses of
diazepam might be helpful.
If a patient lapses into stupor or coma, respiration should be monitored continu-
ously, respiratory arrest anticipated, and ventilatory assistance readily accessible. Ap-
2. DEPRESSION 61
b. Anticholinergic Syndromes
i. Description. Many psychoactive drugs have both peripheral and central anti-
cholinergic effects. Anticholinergic effects occur with tricyclics, maprotiline, and
amoxapine, antipsychotics, some hypnotics, antihistamines, and anti parkinson agents.
Combinations of these drugs, such as a psychoactive drug (antidepressant or antip-
sychotic or both) and an antiparkinson agent, may produce additive anticholinergic
effects. Older patients seem particularly sensitive. Acute overdoses of the above-
mentioned agents also frequently cause anticholinergic crises.
The anticholinergic syndrome may present with a mixture or a predominance of
either peripheral or central symptoms. In its florid state, the CNS picture consists of
confusion, delirium with disorientation, agitation, visual and auditory hallucina-
tions, anxiety, motor restlessness, pseudoseizures (myoclonic jerks and choreo-
athetoid movements with EEG seizure activity), and a thought disorder (e.g.,
delusions). The peripheral syndrome may be manifested by decreased bowel
sounds and constipation, urinary retention, anhidrosis, mydriasis, dry mouth,
cycloplegia (decreased accommodation), increased body temperature, motor inco-
ordination, flushing, and tachycardia. When these syndromes are caused by hetero-
cyclic antidepressants, there is also a high risk of life-threatening arrhythmias. Alipha-
tic and piperidine phenothiazines [especially thioridazine (Mellaril® and others)] and
the tricyclics amitriptyline, doxepin, and imipramine are the most anticholinergic. A
combination of the above drugs or of these drugs with other anticholinergic agents
greatly increases the risk of an anticholinergic syndrome.
ii. Anticholinesterase Therapy. Anticholinesterase therapy has proven very effec-
tive for the treatment of anticholinergic syndromes. All anticholinesterases counteract
the peripheral manifestations of the syndrome. However, clinicians should use either
physostigmine or pyridostigmine because they cross the blood-brain barrier and
therefore counteract the central symptoms. Most clinical studies have focused on
physostigmine, but equivalent doses of pyridostigmine also are effective. 94 •95 In gener-
al, the clinician should avoid using physostigmine in patients with unstable vital
signs. In these cases, cardiac arrhythmias from cholinergic stimulation are common.
Cholinergic effects can also produce seizures (particularly with rapid injection of
physostigmine) and respiratory arrest in selected patients. Although practitioners have
tried many different regimens, physostigmine salicylate, 1-2 mg 1M or IV, will
relieve symptoms dramatically. The clinician should infuse the drug very slowly (e.g.,
1 mg over 2 min), monitor the cardiac status, and have means for respiratory
support available. If no improvement occurs within 15-20 min, another dose of 1-
2 mg should be given. Up to 4 mg may be administered over 10-15 min. The body
degrades physostigmine almost completely within 11/2-2 hr. Since the toxic agents may
2. DEPRESSION 63
disappear more slowly, additional 1- to 2-mg doses at 30-min intervals may be neces-
sary, even if the initial treatment is successful.
Although physostigmine treatment provides dramatic and sometimes lifesaving
relief from heterocyclic-induced arrhythmias, it also has risks from cholinergic stimula-
tion. Excessive acetylcholine can result in tearing, salivation, rhinorrhea, sweating,
pallor, bronchial constriction, hypotension, muscle weakness and fasciculations, nausea
and vomiting, abdominal cramps, urinary frequency, and bradycardia. The antidote is
atropine.
SSRIs and other new antidepressants have negligible anticholinergic effects. This
allows clinicians considerable flexibility in avoiding anticholinergic toxicity where it
has been or may become a problem.
6. Pharmacokinetics
Tricyclic antidepressants (except protriptyline) are generally rapidly absorbed
from the gastrointestinal tract, typically peak in the plasma within 2-8 hr, and remain
unbound for about 30 min. Between 80 and 90% of the drug is protein bound, although
individual differences in binding can produce a fourfold variation in the amount of free
drug. Tissue distribution and the first step in metabolism of tertiary cyclics, demethyla-
tion, occur rapidly. Heterocyclics typically have a high volume of distribution because
they are quite soluble and actively bind to various tissues. Patients usually achieve a
steady state for each dosage after 1 to 4 weeks (depending on the drug used). Liver
metabolism, including demethylation, hydroxylation, and glucuronide conjugation,
also accounts for large variations in serum levels. Individual differences in patients'
microsomal enzyme activity produce steady-state plasma concentrations (bound and
unbound) that may vary as much as 40-fold. These characteristics suggest that individ-
ual patients may require widely varying doses to produce clinically effective serum
levels.
During the initial pass through the liver, 30-70% of the drug is degraded ("first-
pass effect"), as only 30-70% of orally administered antidepressant is "biodegrad-
able." Doxepin may produce lower steady-state plasma levels than other tricyclic
antidepressants per milligram ingested. This might reflect lower plasma protein bind-
ing and greater first-pass degradation. Protriptyline, by contrast, might produce higher
plasma levels per milligram, possibly because of comparatively low first-pass metabo-
lism. The first demethylation of tertiary amines results in an active compound, whereas
a second demethylation and hydroxylation at the 2-position of the central ring of
tricyclics produce degradation products (probably therapeutically inactive but possibly
with cardiotoxic effects). Glucuronidation occurs after hydroxylation and makes the
derivative water-soluble. Initially, about 50% of tricyclics are excreted through the bile,
but because of an active enterohepatic circulation, two-thirds of the drugs are eventu-
ally eliminated in the urine. Clearance of the cyclics is relatively slow, but they have a
wide range of half-lives (see Table 5).
Fluoxetine is well absorbed with a small first-pass effect. Food does not affect the
extent of absorption, although the rate may be slightly decreased. Fluoxetine is metabo-
lized by demethylation in the liver by the cytochrome P4502D6 isoenzyme to the active
64 ALAN J. GELENBERG, M.D., AND PEDRO L. DELGADO, M.D.
Amitripty line 16
Amoxapine 30
Bupropion 14
Clomipramine 24
Desipramine 22
Fluoxetine 48
Fluvoxamine 20
Imipramine 16
Maprotiline 47
Nefazodone 3
Nortripty line 24
Paroxetine 24
Protriptyline 126
Sertraline 25
Trazodone 5
Venlafaxine 4
protein bound, nefazodone could increase the free concentrations of other highly pro-
tein-bound drugs if coadministered. After oral or IV administration, urinary excretion
of the parent drug is negligible. Steady-state plasma levels are achieved by the fourth or
fifth day of multiple dosing. At steady state, nefazodone half-life is dependent on dose:
the mean value varies from about 2 hr at twice-daily doses of 50 mg to between 4 and 5
hr at twice-daily doses of 300 mg. Decreased clearance of nefazodone has been ob-
served in the elderly, in hepatic ally impaired individuals, and in subjects who are poor
metabolizers of dextromethorphan. A moderate decrease in nefazodone plasma levels
occurs when the drug is administered with food.
Mirtazapine is rapidly and completely absorbed following oral administration and
extensively metabolized in the liver. Peak plasma levels occur within about 2 hr after
an oral dose. Food has a minimal effect on the rate and extent of absorption. Mir-
tazapine is excreted mainly in urine and has a half-life of 20-40 hr. Steady-state levels
are reached within 5 days. Mirtazapine is approximately 85% bound to plasma pro-
teins.
7. Plasma Levels
To document serum levels of cyclics in "nonresponders" after a few weeks of generally adequate doses
of medication
To determine if a reticent patient is taking medication
To monitor plasma levels in overdose cases
To monitor plasma levels in medically ill patients
To document serum levels within the "therapeutic window" for nortriptyline (i.e., 50-ISO mg/mL)
To determine if a potentially therapeutic dose of imipramine has been reached (Le., >200 mg/mL)
To minimize dosage levels in elderly patients
To determine plasma level in the context of a severe adverse reaction or persistent, bothersome side
effects
to order a plasma level determination (of imipramine and desipramine) along with an
EKG to ensure against unanticipated toxicity. If the blood level is still comparatively
low (say, under 200 ng/mL), the physician is justified in this instance in gradually
elevating the imipramine dose until either the depression remits or the patient is unable
to tolerate side effects. With each successive increment in dose, it would be prudent to
order a repeated blood level and EKG. Some patients may require (and tolerate)
imipramine doses in excess of 500 mg/day.
In treating toxicity, plasma level determinations can help to establish when the
peak occurred and how high the peak was and to follow the egress of the drug. Plasma
tricyclic antidepressant levels in excess of 1000 ng/mL have been correlated with
severe and life-threatening toxicity, but lower levels should not be taken as a guarantee
of safety, since absorption may still be progressing in some cases. There is no clear
correlation between blood levels and toxicity because of interindividual differences in
sensitivity and also the fact that a patient's response to a given blood level when
concentrations are rising may not be identical to what happens at the same blood level
on the way down.
Correlations between plasma concentrations and therapeutic response or toxicity
remain obscure with SSRIs and newer antidepressants.
lism, and liver enzyme activity. The tricyclic compounds amitriptyline, imipramine,
and doxepin have usual dose ranges of 150-300 mg/day. The tetracyclic maprotiline
has comparable potency, but because of its dose-related tendency to provoke seizures,
its daily dose should be capped at 225 mg. Nortriptyline is about twice as potent,
whereas protriptyline is about four times as potent. Doxepin is somewhat less potent
than the other dimethylated tricyclics. Amoxapine and trazodone are about half as
potent as imipramine, with clinically effective doses ranging from 150 to 600 mg/day,
although there is a hint that trazodone is more effective at lower than at higher doses.
When fluoxetine was introduced in 1988, it appeared to have a very simple dose-
response relationship. The prevaling wisdom at the time was that most patients bene-
fited optimally from a daily dose of 20 mg. Since then, the picture has become
complicated. As noted in other chapters (see Chapters 5, 12, and 13), dose-response
relationships may vary according to the condition being treated. Even in a single
condition (at least as homogeneous as we can become in modem psychiatry), such as
depression, the dose-response relationship appears to vary widely among different
patients, presumably due to interindividual differences in pharmacokinetics but also to
possible differences in actual effects in brain.
Most healthy adults of average size will still start treatment with 20 mg of
fluoxetine daily. If side effects are a problem, the dose can be reduced, with the
possibility of gradual increases. For the first few years of its availability, fluoxetine was
available only in 20-mg capsules. To take daily doses below 20 mg, the patient had to
break the capsule open and dissolve the contents in liquid, thereby creating a prepara-
tion that could be dispensed in lower dosages. More recently, fluoxetine has become
available in a liquid preparation and in a lO-mg capsule. Although data gleaned from
large, controlled trials still suggest an optimal daily dose for depressed adults of 20 mg,
with no apparent benefit from increased dosages, clinical observations suggest that
some patients may benefit from daily doses up to and beyond 100 mg. 96 Still other
patients are unusually sensitive to fluoxetine's side effects and might tolerate the drug
better at daily doses below 20 mg, with possible increments as tolerance develops. As
noted earlier, some patients become sedated, fatigued, and apathetic after weeks or a
few months of treatment with fluoxetine. 97 Such individuals may benefit from a reduc-
tion in the daily dose. Given the long half-lives of fluoxetine and norfluoxetine, the
reduction in average daily dose can be achieved (with improved daily cost of the
medication) by the use of 20-mg fluoxetine capsules in staggered daily dosing sched-
ules: e.g., one capsule every other day, one or two capsules in 3 days, etc.
Dose-response relationships for SSRI antidepressants introduced more recently
than fluoxetine are even less clear-cut. With sertraline, many patients appear to reach
optimum therapeutic benefit at a daily dose of 50 mg. Some tolerate the drug better
when started at 25 mg/day, at least for the first few days. Depending on therapeutic
response, many clinicians increase the dosage of sertraline by 50-mg/day increments
every 2 weeks up to a maximum of 200 mg daily. The dose-response curve for
sertraline remains a matter of active debate.
For paroxetine, the recommendation is to begin adult patients at a daily dose of 20
mg and increase the dose every 2 weeks up to a maximum daily dose of 50 mg.
Occasional patients benefit from even higher doses. Older patients should be started at
2. DEPRESSION 69
1. Chemistry
Usual
Initial daily therapeutic
Heterocyclic Trade name Preparations dosage dosage
Amitriptyline Elavilill> and others 10-, 25-, 50-, 75-, 100-, and 25-75 mg 150-300 mg
150-mg tablets; 25- and 20-
mg capsules; vials of 10 ml
with 10 mg/ml
Imipramine Tofranilill> and 10-, 25-, 50-, and lOO-mg 25-75 mg 150-300 mg
others tablets; 75-, and 150-mg
capsules; ampules of 20 ml
with 25 mgl2 ml
Doxepin Sinequanill> and 10-, 25-, 50-, 75-, 100-, and 25-75 mg 150-300 mg
others 100- and 150-mg capsules;
solution of 10 mg/ml
Nortriptyline Pamelor® and 10- and 25-mg capsules; 20-40 mg 75-150 mg
others solution of 10 mg/5 ml
Desipramine Norpramin® and 25-, 50-, 75-, 100-, and 150- 25-75 mg 75-200 mg
others mg tablets; 25- and 50-mg
capsules
Protriptyline Vivactil® 5- and IO-mg tablets 10-20 mg 20-60 mg
Trimipramine Surmontilill> 25- and 50-mg capsules 25-75 mg 75-300 mg
Amoxapine Asendinill> 30-, 100-, and 150-mg tablets 50-ISO mg 150-600 mg
Maprotiline Ludiomilill> and 25- and 50-mg tablets 25-75 mg 75-225 mg
others
Trazodone Desyrelill> and others 25-, 50-, and 1000mg tablets 50-100 mg 150-600 mg
Auoxetine Prozac® 10- and 20-mg capsules; liquid 20mg 20mg
preparation
Bupropion Wellbutrin® 75- and loo-mg tablets 100 mg bj.d. 450 mg
or 75 mg
t.i.d.
Sertraline Zoloft® 50- and loo-mg tablets 25-50 mg 50-250 mg
Paroxetine Paxil® 20- and 30-mg tablets 20mg 50 mg
Auvoxamine Luvox® 50- and loo-mg tablets 50 mg 100-300 mg
Nefazodone Serzoneill> 100-, 150-, 200-, and 250-mg 100 mg bj.d. 200-600 mg
tablets
Venlafaxine Effexor® 25-, 37.5-, 50-, 75-, and 100- 75 mg 75-225 mg
mg tablets
2. Mechanism of Action
Tranylcypromine
phenylalanine, whereas type A (accounting for 20% of CNS MAO but most of the
gastrointestinal MAO) primarily degrades serotonin and norepinephrine. Although the
antidepressant MAOIs that are currently available affect both enzymes, experimental
agents (such as clorgyline) that selectively inhibit type A MAO (MAO-A) seem to be
more effective in treating depression than selective inhibitors of type B MAO [such as
selegiline (Eldepryl®) or some available drugs used for other purposes, including
pargyline (Eutonyl®), procarbazine (Matulane®), or furazolidone (Furoxone®)]. Se-
legiline is available by prescription for the treatment of Parkinson's disease as an
adjunct to levodopa/carbidopa therapy. At its recommended dose, 10 mg/day, it does
not appear liable to the food, beverage, and medication interactions of the nonselective
MAOIs. If it possesses antidepressant activity at all (an uncertain assumption), how-
ever, it is probably at a daily dose higher than that recommended for the treatment of
Parkinson's disease, at which its specificity for MAO-B is lost.
Several reversible inhibitors of MAO-A (RIMAs) are currently in use abroad, but
one such agent, brofaromine, has been removed from testing by its manufacturer, and
the chances of another one, moclobemide, coming to the United States is doubtful at
this point. Ostensibly, RIMAs should provide less risk of the "cheese effect" than
nonselective MAOIs and greater tolerance. The literature on controlled studies with
moclobemide, however, shows a relatively high incidence of excitatory effects such as
insomnia/restlessness, manialike symptoms, irritability/excitation, agitation, and anxi-
ety. There may also be the risk of the "serotonin syndrome," a potentially lethal
hypermetabolic reaction, occurring with RIMAs. A RIMA should elevate brain levels
of the catecholamines norepinephrine and dopamine, which could explain stimulantlike
properties. It should also raise brain serotonin levels, and the serotonin syndrome
sometimes includes symptoms of CNS hyperexcitability. In addition to reporting ad-
verse effects with moclobemide, investigators have also questioned whether its efficacy
equals that of older antidepressants. The role of RIMAs in the treatment of depression
remains in question.
72 ALAN J. GELENBERG, M.D., AND PEDRO L. DELGADO, M.D.
3. Adverse Reactions
TABLE 9. Food and Beverage Products that Can Cause Hypertension with MAOIsa
Cheeses
1. High tyramine content: boursault, Camembert, cheddar, Gruyere, Stilton. For all fermented cheeses,
tyramine content may be highest near rind and fermentation holes.
2. Moderate tyramine content: Gouda, Parmesan
3. Low tyramine content: American, ricotta, cottage cheese, cream cheese
Other foods
I. High tyramine content: aged and processed meats, sausages, lox, pickled herring (but a recent study
found no tyramine in pickled herring excepting brine), bean curd, possibly any overripe produce
2. Moderate tyramine content: salted herring, figs, raisins, broad beans (fava beans), concentrated yeast
extracts (but not Brewer's tablets or flakes), pickles, sauerkraut, coffee, chocolate, cocoa, soy sauce,
sour cream, snails, avocado, banana peels, licorice
3. Questionable tyramine content: liver (fresh liver probably safe), protein extracts (often in soups),
oriental foods. Beware of any gravy, soup, sauce, or dressing of unknown content.
Beverages
I. High tyramine content: some imported beers and ales
2. Moderate tyramine content: most sherry and beer
3. Low tyramine content: champagne, most Italian red wines, Riesling, Sauterne, "hard" liquor (but
reactions have occurred with whiskey; cause unknown)
4. Questionable tyramine content: Chianti, liqueurs, nonalcoholic beers
Additives
1. Questionable effects: cyclamates, monosodium glutamate
aGeneral rule of thumb: Try to eat only fresh foods; avoid overripe or prepared foods; and favor foods that have been
prepared simply.
TABLE 10. Drugs that Can Cause Adverse Interactions with MAOIsa
aFrom Harrison et al l14 with pennission. Copyright 1989, Physicians Postgraduate Press.
occur with a relatively selective MAOI such as selegiline and with RIMAs. Clinical
manifestations include profound hyperthermia, neuromuscular irritability, altered con-
sciousness (through coma), seizures, and possibly death. Offending agents have in-
cluded meperidine (Demerol® and others), SSRIs, and some tricyclic antidepressants,
most notably clomipramine. A nonselective SRI such as venlafaxine has to be treated
as if it too can cause this reaction. Because of its very long half-life, fluoxetine's
manufacturer recommends an interval of 5 weeks between the discontinuation of
fluoxetine and initial treatment with an MAO I. For sertraline and paroxetine, the
manufacturers recommend discontinuing the SSRI for 2 weeks before initiating an
MAOI. Milder syndromes have been associated with the antitussive dextromethorphan
(contained in many over-the-counter "cold" preparations) and tryptophan. The anti-
spasticity agent dantrolene (Dantrium®), useful in malignant hyperthermia of anesthe-
sia and possibly in the neuroleptic malignant syndrome, may be considered an antidote.
d. Cardiac Effects. The MAOIs do not have direct cardiac effects or slow cardi-
ac conduction. Thus, they may be considered alternatives to tricyclic antidepressants
for patients with bundle branch block or second-degree atrioventricular (A-V) block.
f. Edema and Weight Gain. Weight gain, often associated with increased food
intake, is sometimes observed in MAOI-treated patients. This is more common with
phenelzine and isocarboxazid than with the amphetaminelike tranylcypromine. If diet
76 ALAN J. GELENBERG, M.D., AND PEDRO L. DELGADO, M.D.
g. Neuropsychiatric
i. Sleep Disturbances. Frequently patients discontinue MAOI agents because of the
combination of restless nocturnal sleep plus irresistible afternoon somnolence. Possibly
more common in patients with "bipolar spectrum" depressions, these side effects have
been observed with phenelzine and tranylcypromine. There is a suspicion that they may
be less common with isocarboxazid.lOo This syndrome tends to resist attempts at
management, whether with sleeping pills at night or stimulants during the day.
ii. Psychiatric. Occasional patients develop manic episodes during MAOI thera-
py. More common in patients with histories of mania, such pathological euphoria
occasionally develops de novo on provocation with an antidepressant. Some patients
treated with MAOIs also develop confusional states, sometimes while taking the drugs
and occasionally as part of a withdrawal syndrome.
iii. Other Neurological Effects. Exaggerated nocturnal myoclonus in patients tak-
ing MAOIs may respond to a change in medication schedule from bedtime to daytime
or to divided doses or occasionally to clonazepam (Klonopin®), 0.5-2 mg at bedtime.
Paresthesias, often experienced as "pins-and-needles" sensations, are sometimes sec-
ondary to pyridoxine (vitamin B6 ) deficiency induced by MAOIs and may respond to
supplementation with pyridoxine, 50-150 mg daily by mouth.
4. Contraindications
The clinician should administer MAOIs only after a careful medical and psychi-
atric history and review of systems. In various medical situations, however, the clini-
cian should use MAOIs with caution. These include liver disease, advanced renal
disease, pheochromocytoma, cardiovascular disease, hypertension, asthma, or
chronic bronchitis (since pressor agents like epinephrine or theophylline may be
necessary). In addition, MAOIs should seldom be given to patients who take drugs that
can produce dangerous synergisms.
6. Toxicity
Experience with overdoses of MAOI drugs remains limited but is beginning to
expand as their use increases. Mortality has been reported following doses of phe-
nelzine, 375-1500 mg, and tranylcypromine, 170-650 mg. Because clinical signs may
not become manifest for 12 hr or more after ingestion, patients should be intensively
monitored for an extended period in a medical emergency setting.102
Headache, dizziness, and precordial pain may be early symptoms. Signs of hyper-
metabolism predominate and can include hyperthermia, increased heart and respiratory
rates, muscular rigidity, metabolic acidosis, decreased plasma oxygen, and increased
carbon dioxide. Irritability of the nervous system may be manifested by restlessness,
brisk reflexes, and, at times, seizures. Hypertension may occur, but hypotension, some-
times occurring late in the course, appears to be a poorer prognostic sign.
The MAOIs may delay gut motility, so gastric lavage should be performed even
hours after the ingestion. Acidification can speed MAOI elimination and increase the
amount of MAOI eliminated in the urine. In contrast, with tricyclic antidepressants,
hemodialysis can effectively remove much of the MAOI. Similar to the treatment of
heterocyclic overdoses, activated charcoal and magnesium citrate can help reduce
absorption.
78 ALAN J. GELENBERG, M.D., AND PEDRO L. DELGADO, M.D.
7. Pharmacokinetics
Monoamine oxidase inhibitors are rapidly and fully absorbed from the gastroin-
testinal tract. They undergo biotransformation in the liver and are excreted very quickly
through the intestinal tract and, to a lesser extent, via the kidneys. The time to peak
plasma concentration is 2-4 hr after an oral dose for phenelzine and 1-3.5 hr for
tranylcypromine. The onset of action may be as early as 7-10 days with appropriate
dosage in some patients, but it may take up to 4-8 weeks to achieve full therapeutic
effect. It takes at least 10 days for MAO activity to be recovered due to irreversible
binding of MAOIs. The half-life of MAOIs in the body is very short. However, the
traditional MAOIs have long-lasting pharmacologic effects, since they permanently
inactivate enzymes. Therefore, MAO inhibition continues for up to several weeks after
discontinuation of the drug. The body must resynthesize the enzymes before normal
metabolism of body amines resumes, a process taking 1-2 weeks.
Little is know about the metabolic fate of the three MAOI antidepressants, either
in animals or in humans. Some research has focused on acetylation of hydrazines as a
possible predictor of clinical response. At least for phenelzine, the primary degrading
process appears to be oxidation rather than acetylation. Furthermore, the genetic poly-
morphism of the liver enzyme system that determines whether individuals are rapid or
slow acetylators appears to have no clinical usefulness in predicting treatment response
or side effects. Although patients with a degree of platelet MAO inhibition greater than
80% by 2 weeks of therapy are more likely to have a favorable treatment outcome with
phenelzine, ensuring adequate dosage is probably equally valuable and certainly less
expensive. There is no evident correlation between platelet MAO inhibition and clini-
cal response with either tranylcypromine or isocarboxazid.
alsocarboxazid is only available in the United States through special request from the manufacturer.
c. Stimulants
Psycho stimulants, such as the amphetamines, were used before the tricyclics and
MAOIs to treat depressed patients. Often, they did little more than convert psychomo-
tor retardation to anxiety or agitation. Additionally, tolerance developed to their effects,
anorexia and insomnia were unwanted reactions, and these drugs could promote depen-
dency and abuse. Furthermore, there was little scientific validation of their antidepres-
sant efficacy, and for this reason, coupled with increasing government regulation over
the prescription of these substances, they fell into disfavor and disuse among the
medical community.
More recently, however, there has been heightened interest in the use of stimu-
lants, growing largely out of the experience of consultation-liaison psychiatrists. Be-
cause of their very different profile of clinical response and adverse reactions, these
drugs have been found safer and more useful than standard antidepressants, particularly
in the medically ill and the elderly. Low doses of methylphenidate or D-amphetamine,
administered for up to several weeks, can be helpful in turning the ~orner in depressive
symptoms in some such patients. Despite the paucity of scientific validation of these
clinical impressions, many psychiatrists practicing in general hospitals are pleased to
have them available.
Earlier research suggesting that stimulant response could be used to predict anti-
depressant response has not withstood the tests of science and time. In some instances,
clinicians have used stimulants to get a "leg up" on the lag time while waiting for a
standard antidepressant to work or to "jump start" a treatment-resistant patient. Anoth-
er use of stimulants as adjuncts to antidepressant medications is in patients suffering
from drug-induced postural hypotension.
complaints. These may take many fonns, from a request for a physical examination,
complaints of fatigue, weight loss, or insomnia to specific manifestation of somatic
symptoms (often careful questioning reveals that several organ systems are involved).
Along with or instead of physical symptoms, the patient frequently has both emotional
(e.g., lowered mood, anhedonia, negativism) and psychological symptoms [e.g., irri-
tability, loss of interest, ruminations, poor concentration, suicidal preoccupations, and
even pseudodementia (see Chapter 8)]. In addition to defining the specific nature and
history of the mood disorder, the clinician should also evaluate the other important
components of the patient's life that might affect treatment: his or her environmental
supports and important relationships, family history, coping style, and ego strengths.
After completing a careful evaluation, the clinician should decide how to manage
acute symptoms, possible precipitants, and sequelae of the affective episode (e.g.,
stigmatization, lowered self-esteem, alienation from social supports).
2. DEPRESSION 81
a Association does not mean causality. Some agents (e.g., reserpine and steroids) have a clear relationship with depression,
whereas many other drugs do not.
other laboratory tests, such as an EKG, also may be worthwhile in such circumstances.
When a patient has been partially responsive to standard doses of an antidepressant, an
alternate strategy to raising the antidepressant dose or switching antidepressants is to
use an adjunctive drug, such as lithium or triiodothyronine (T 3). When the antidepres-
sant is an SSRI, some clinicians favor the addition of low doses of a tricyclic (see also
Section V.E.I0).
As noted earlier, overdoses with tricyclic and MAOI antidepressants produce a
substantial risk of lethality. Prescribing these drugs, therefore, to a patient with the risk
of suicide demands careful thought and discussion with the patient and friends or
family members. If the risk of suicide is considered substantial, the practitioner will
want to hospitalize the patient or arrange for other circumstances in which the patient
can be monitored in an ongoing fashion for the patient's protection. If the risk of
suicide is not considered to be at that level, and a potentially hazardous agent such as a
tricyclic or MAOI is to be prescribed, the patient and concerned others must be advised
about the lethal potential of the drugs (to avoid an unintended "gesture" that inadver-
tently becomes lethal). Consider limiting the patient's access to potentially lethal
quantities of the drug (for tricyclics, as little as one gram). Along these lines, when
children may have access to these hazardous drugs, the drugs should be kept in safe
places and in childproof containers.
Although many antidepressants have been introduced with the promise of faster
onset of action, these claims have not withstood the test of time. It seems that all
antidepressant drugs currently known-tricyclics, MAOIs, and newer agents-take
some weeks to become maximally effective. The conventional wisdom is that the lag
time is 2-3 weeks, but this is oversimplified.
Prospective studies have found differences in symptom responses between drug
responders and nonresponders as early as the first week of treatment. In other words,
patients who will go on to become responders show the beginnings of improvement in
a wide range of depressive symptoms by the seventh day of antidepressant drug
therapy. Typically, patients may be the last to know (or to acknowledge) that they are
better; a significant other, or the clinician, may be more perceptive. Sleep, by the way,
is not a useful symptom for distinguishing responders from nonresponders; the non-
specific sedating effects of these drugs can improve sleep (or cause excessive drowsi-
ness) independent of treatment response.
Some patients show an early and dramatic improvement in depressive symp-
toms-perhaps a complete remission within the first week. Others show a variable and
stuttering course of symptom improvement. Both patterns may be more likely to be
placebo responses than a true biological antidepressant reaction. The caveat here is that
these patients are more likely to lose their apparent benefit from the antidepressant.
Anticipating this, the clinician can be prepared to counteract a tendency to frustration,
despair, or anger by suggesting the need to reevaluate the drug and its dose and also to
emphasize the role of interpersonal and intrapsychic factors.
While awaiting treatment response, a knowledgeable clinician should be in fre-
84 ALAN J. GELENBERG, M.D., AND PEDRO L. DELGADO, M.D.
quent contact with the patient, and possibly family members, to provide support and
reassurance and to assess suicide potential. Family and patients also should be advised
from the outset about the lag time before the drugs become effective and should be
informed about common side effects. Education about unwanted effects of these drugs
can tum an annoyance into "cement" for the therapeutic alliance: the patient can treat
dry mouth, for example, as evidence that the drug is in his or her system and working; it
also affirms the knowledge of the doctor.
The maximal efficacy of an antidepressant may not plateau for 6 to 8 weeks. Thus,
if a patient's improvement has progressed from 80% to 90% between weeks 4 and 5 of
treatment, the clinician can stay with the same dose for at least another week, awaiting
continued improvement. On the other hand, a plateau in improvement at anything less
than 100%, or a return to status quo ante, should prompt a reassessment of the drug
dose.
The goal of antidepressant drug therapy should be complete remission of symp-
toms. Needless to say, character problems will not usually improve, but sometimes
symptoms attributed to a patient's personality may reflect untreated depression. Still,
for patients with dysthymia who later develop major depression (so-called "double
depression"), antidepressant drugs may reverse the major depression but return the
patient to a stable depressed character type.
The clinician would do best to continue adjusting the dose of the original anti-
depressant drug until either the acute depression remits completely or the patient is
unable to tolerate side effects. If side effects are the limiting variable, and dosage
reduction brings loss of efficacy, then the physician must switch to an alternate anti-
depressant. In this circumstance, one chooses a drug with fewer of the offending side
effects, for example, one that is less anticholinergic, less hypotensive, less likely to
promote weight gain, or less likely to impair sexual functioning.
Failure of an antidepressant to treat symptoms effectively despite adequate dosing
for a sufficient time period is more troublesome. There is no evidence that switching
among tricyclics is apt to produce efficacy when one tricyclic has failed, although "the
jury is still out" about differences among SSRIs. In some patients who have achieved
partial but incomplete remission, lithium, often starting with as little as 300 mg a day,
may provide a needed "jump start." For a patient partially responsive to an SSRI, the
addition of a low-dose tricyclic, such as desipramine, may close the therapeutic gap.
Perhaps this will not be necessary with combined serotonin and norepinephrine reup-
take inhibitors such as venlafaxine. Sometimes two, three, or even more drugs are
being combined in increasingly creative ways by clinicians specializing in hard-to-treat
depression.
When one agent fails to induce remission, switching to an antidepressant with
substantially different pharmacologic characteristics (e.g., tricyclics, MAOIs, SSRIs,
bupropion, venlafaxine, nefazodone) makes rational sense. Before starting an MAOI
after discontinuing a different antidepressant, the patient should be free of the prior
antidepressant for at least a week after a tricyclic, 2 weeks after another MAOI or an
SRI, and 5 weeks to 2 months (depending on duration of treatment) after fluoxetine.
Electroconvulsive therapy always remains an important backup option.
When a patient has proved refractory to several adequate trials of biological
2. DEPRESSION 85
3. Continuation Therapy
Having successfully ameliorated the acute symptoms of depression, the practi-
tioner should now embark on what is termed continuation therapy. Typically lasting
about 6 months, continuation therapy protects against a relapse of the depression.
Presumably, we are "holding the fort" against depressive symptoms while the underly-
ing process runs its course.
For continuation treatment, the clinician should usually continue the same dose
that was effective during the acute phase of therapy. If, however, the patient is plagued
by annoying but persistent side effects, the physician may try to lower the dose very
slowly, although both patient and doctor should be acutely aware that this enhances the
risk of relapse. At times, a switch to a different agent might be considered, but again the
benefits in comfort must be weighted against the risk of recrudescent symptoms.
After about 6 months of successful antidepressant therapy, the physician evaluates
the patient for possible discontinuation of the drug. If depressive symptoms have been
successfully and completely suppressed for at least 4 months, a gradual taper may
be attempted. If, however, there has been a "flickering" of symptoms (whether during a
missed dose, a time of stress, or for no apparent reason), it is most prudent to continue
the continuation therapy for at least another few months. Tapering the medication after
an acute episode of depression should not take place until and unless depression has
been virtually absent for at least 4 months.
To avoid antidepressant withdrawal symptoms (see Section V.A.3.g.vii), tapering
should be very gradual. Withdrawal symptoms after only 6 months of therapy are not
likely to be as intense as those following many years of treatment. Still, if there is no
medical urgency about discontinuing the antidepressant, prolong the taper over a month
or more. Toward the end of the taper, tablets can be broken in half or administered
every other day to make the final phases even more gradual.
mood disorder require maintenance therapy rather than intermittent treatment with
medication.
Whether early or late in the course of a depressive illness, if the physician cannot
wean the patient from antidepressant medication at the end of continuation therapy
because of reemerging symptoms, this suggests that a chronic course is developing. In
such cases, if antidepressant medication has successfully suppressed all or most of the
depressive symptoms, there is no reason not to continue it indefinitely, possibly with
periodic attempts to taper. Obviously, the physician seeks to find the lowest effective
dose of the antidepressant, but because most patients can take these agents chronically
with impunity, there is no pressing need to compromise efficacy in the interest of
lowering the dose. (This situation is different from the concern with neuroleptics that
higher doses increase the risk of tardive dyskinesia during the course of maintenance
treatment.) In fact, recent research suggests that the most effective doses for mainte-
nance antidepressant therapy, at least with tricyclics, are the same as those effective in
the acute treatment of depression. I 05 Therefore, if at all possible, maintain the patient
on maintenance therapy at the same doses that were effective (and presumably toler-
ated) in the acute and continuation phases of treatment. Occasional patients may have
required ECT during an acute episode but will benefit from maintenance antidepressant
drugs.
For a patient with multiple recurring episodes of depression, the question of when
to consider maintenance therapy will hinge on the number of episodes as well as their
frequency and severity. To be sure, the more numerous, frequent, and severe the
depressive episodes have been, the earlier and more vigorously the doctor should
pursue a course of maintenance. The need for such an approach should be discussed
actively with the patient and possibly with significant others. The benefits in terms of
prevention of relapse, with all the attendant morbidity and mortality, are obvious; the
risks are few and usually center around uncomfortable rather than hazardous side
effects.
There are no known risks that increase with long-term exposure to these drugs.
However, as a patient ages, it may be necessary to reassess the dose of the drug and its
interaction with illnesses that may have developed or with medications taken to treat
them.
If an antidepressant is ineffective in the prophylaxis of recurrent depressive epi-
sodes, the practitioner may consider maintenance lithium therapy. This is especially
true if the patient has a history of episodes suggesting hypomania or a family bipolar
history. Most nonbipolar patients, however, will not require lithium, which probably is
less effective than antidepressants in maintenance therapy. There are no good studies
about the optimal plasma range of lithium in depression. We prefer the dose levels most
effective in bipolar patients, 0.8-1.0 mEq/L. Of course, for bipolar patients, lithium is
the maintenance therapy of choice and should be augmented as needed with antidepres-
sant drugs.
than depressive recurrences. When minor depressive symptoms emerge, time, support,
and psychotherapy may suffice. Emergence of a full-blown episode of major depres-
sion, however, can be serious and usually requires prompt and vigorous intervention.
For a major depressive episode in a bipolar patient, an antidepressant may be
added to the maintenance lithium regimen. Despite the lithium, however, there is a
finite risk of triggering a switch into mania, although in some patients this would
happen whether or not an antidepressant was prescribed. To minimize this risk, discon-
tinue the antidepressant drug soon after the patient has achieved a remission from
depression. This also lowers the possibility that prolonged antidepressant therapy can
speed up the rate of a bipolar patient's mood cycles.
There is limited evidence that the MAOI tranylcypromine is particularly effective
in treating the bipolar depressed patient, especially when depression is accompanied by
psychomotor retardation, apathy, and withdrawal.! 06 There is less experience with
some of the newer antidepressants in bipolar patients, although there have been sugges-
tions that bupropion may be beneficial for this population. Bupropion and SSRIs may
cause switches into mania less often than tricyclics.107 ECT is highly effective and
should particularly be considered in emergency cases. (Most psychiatrists discontinue
lithium during the course of ECT to avoid neurotoxicity.)
When the symptoms of major depression are present consistently over a period of
weeks, most psychiatrists will consider antidepressant drug therapy. If the patient
simultaneously has a number of melancholic symptoms, the likelihood of improvement
with an antidepressant drug (or ECT) becomes even better. These predictive factors
tend to be independent of the degree of severity or longevity of the depression. Neither
mildness nor chronicity should preclude consideration of an adequate antidepres-
sant trial, nor do they suggest use of lower doses. The only way to know if a
patient's depressive symptoms will remit on antidepressant drug therapy is to try
it. Obviously, if depressive symptoms yield to psychodynamic interpretations or
changes in living or work arrangements, antidepressant drug therapy may become
moot. However, analogously, if symptoms remit on prescription of an antidepressant
drug, major life changes or long-term psychotherapy may be unnecessary. Until a
vigorous and adequate antidepressant trial has been attempted and failed, it is
impossible to say whether a patient's symptoms are "characterologic" or ''neurot-
ic" rather than drug responsive.
An appropriate role for antidepressants in treating the patient with dysthymia
remains to be elucidated. Indeed, dysthymia most probably reflects a range of disor-
ders, including both those that are responsive to medication and those that are not.
Again, the only way to distinguish the two is empirically: most, if not all, dysthymic
patients deserve at least one vigorous antidepressant trial.
In the early days of modem psychopharmacology, there were hopes and expecta-
tions that antidepressants that were more or less sedating could be tailored to depres-
sive subtypes such as agitated, anxious, or retarded. This has not proved to be the case.
Nevertheless, some patients may appreciate the sedation provided by drugs such as
88 ALAN J. GELENBERG, M.D., AND PEDRO L. DELGADO, M.D.
7. "Atypical" Depression
Mood reactive
Has exhibited at least two of the following:
I. Increased appetite or weight gain while depressed (defined as present if appetite markedly increased
or at least a 4.5-kg weight gain)
2. Oversleeping (at least 10 hr/day)
3. Severe fatigue that creates a sensation of leaden paralysis or extreme heaviness of arms or legs
(defined as present if a marked decrease in energy or if heaviness of limbs is a prominent symptom)
4. Sensitivity to rejection as a trait throughout adulthood (defined as present if rejection results in
depression with functional impairment, e.g., missing work or school)
"Ref. 35.
2. DEPRESSION 89
How the newer antidepressants will compare in efficacy for atypical depression is
unclear, although patients for whom weight gain is a substantial problem may appreci-
ate the weight loss often associated with SSRIs. Electroconvulsive therapy is not
usually thought of in this condition.
13). Individuals for whom depressive behaviors are part of their lifestyle and interper-
sonal coping strategies also may do poorly on antidepressant therapy, although they
should still be given at least several trials of adequate pharmacotherapy. A sensitive
clinician will always inquire about neurotic and character mechanisms that may be
"driving" depressive symptoms. An interview with a significant other can shed new
light on the meaning of an individual's symptoms and on the accuracy of the history.
Unresolved grief is another phenomenon that sometimes appears to be a treatment-
refractory depression.
Many biological factors can similarly render a patient unresponsive to antidepres-
sant medicines. Consider possible adverse interactions with other medications a patient
may be taking: prescribed, over-the-counter, or substances of abuse. A wide range of
medical illnesses-especially endocrinopathies, infections (including AIDS), brain
diseases, or neoplasms-can similarly produce a depression that might fail to respond
to antidepressants. To consider all of these and many other biological possibilities,
return to the medical history and review of systems, obtain additional data from past
records, other treating physicians, or family members, see that an adequate physical
examination has recently been performed, and order any laboratory testing indicated by
the foregoing.
What if nothing new turns up from either a psychosocial or a biological perspec-
tive? At this point, the physician can consider a wide variety of add-ons and medication
combinations, unfortunately few of which have been studied systematically. Probably
best studied and most frequently used today in clinical practice is adjunctive lithium.
Many knowledgeable clinicians add lithium to a first or second antidepressant trial in
an incompletely responsive patient. Some patients appear to respond dramatically and
quickly to minimal doses of lithium, whereas others respond more gradually to some-
what higher doses. Perhaps partial responders are most likely to progress to complete
response, or maybe some patients are hastened to what would ultimately be a complete
recovery anyway. Despite these doubts, the usual augmentation doses are between 300
and 1200 mg daily, maintained for several weeks. One approach is to begin with 300
mg bj.d. and, if there is no response within a week, to increase to 900 mg daily. Some
clinicians prefer to bring patients to a "standard" blood lithium level approaching 1.0
mEq/L. Most physicians try to stop the dose of lithium after several weeks, whether it
is effective (in which case perhaps no longer needed) or ineffective (in which case
something else should be tried).
Adjunctive thyroid hormones have been tried, although our reading of the data
suggests that, in patients chemically euthyroid, this approach is less likely to meet with
success. The usual strategy is to initiate treatment with 10-25 j.1g of liothyronine (T3 ,
Cytomel®), increasing the dose every few days to a maximum of 50-75 j.1g/day. Some
recent studies suggest that thyroxine (T4 ) may be an effective supplement in treatment-
resistant depressed patients. Other adjunctive strategies with even weaker data bases
include the coadministration of reserpine (Serpasil® and others), an antipsychotic drug,
conjugated estrogens in women, or methylphenidate with an SSRI. Tryptophan and
tyrosine, the amino acid precursors of serotonin and norepinephrine, respectively, also
have been used adjunctively with varying reports of success. For patients with seasonal
affective disorder, spending upwards of 2 hr a day in front of high-intensity full-
92 ALAN J. GELENBERG, M.D., AND PEDRO L. DELGADO, M.D.
spectrum lights has been reported to yield dramatic results, although the scientific
validation of phototherapy as an antidepressant treatment remains insufficient. We have
already mentioned the combination of an SSRI and a tricyclic, and some clinicians are
combining bupropion with other agents.
There is no scientific evidence that the combination of a tricyclic and an MAOI
antidepressant provides more efficacy than either drug alone. Still, there is a wealth of
anecdotal reports suggesting that some patients appear to respond better to the combi-
nation than to monotherapy. Many patients have taken this combination with impunity,
although a number of caveats are in order. Obviously, as with any regimen involving an
MAOI, the usual food, beverage, and medication prohibitions apply. Certain medica-
tions are best avoided in this combination, more to prevent the hypermetabolic than the
hypertensive crisis. Imipramine, desipramine, and especially clomipramine are best
avoided as the tricyclic. Tranylcypromine may be chancier than phenelzine on the
MAOI side. It is best to stop previous medications for at least a week and then to begin
a low dose of a tricyclic, e.g., 50 mg of amitriptyline, trimipramine, or doxepin.
Isocarboxazid or phenelzine is then added in low doses, and, if tolerated, the doses of
the tricyclic, alternating with that of the MAOI, are "walked up" every few days
gradually and with cautious monitoring of the patient's symptoms and vital signs.
Patients with difficult-to-treat depressions are typically managed with different
drugs sequentially or often with combination therapies. This means a need for height-
ened awareness of potential interactions. Switching between heterocyclics and MAOIs,
or between two different MAOIs, presents potential hazards. At least a week's washout
should intervene when changing from a tricyclic to an MAOI, and probably 2 weeks
between an MAOI and a tricyclic. The very long half-lives of fluoxetine and its active
metabolite suggest the need for an even longer washout between fluoxetine and an
MAOI, since cases of potentially lethal hypermetabolic syndrome (presumably a seroto-
nin syndrome) have occurred following this switch of medicines. When patients have
been switched from phenelzine to tranylcypromine without sufficient washout, cases of
hypertensive crises and strokes have developed. A pharmacokinetic interaction has been
described between SSRls and tricyclic antidepressants, in which SSRI coadministration
markedly elevates the plasma level of the tricyclic (see Chapter 1, Table 1).
For further discussion and details on the treatment of the treatment-resistant
depression, the reader is referred to Ref. 112.
VI. CONCLUSION
The 1980s saw the introduction of a number of new antidepressant drugs. In the
early 1980s, the only thing new that they seemed to provide was toxicity. Newer drugs,
however, have broadened our range of therapeutic options, both in terms of conditions
that are treatable and different adverse-effect profiles.
Research also has told us more about the appropriate use of old drugs, such as
tricyclics and MAOIs. One of the most important and troubling findings in recent years
is that a majority of depressed patients go untreated or undertreated with antidepressant
2. DEPRESSION 93
medications, making vigorous pharmacotherapy a wat~hword for the tum of the mil-
lennium.
As we look to the future, health care professionals anticipate the advent of biolog-
ical tests that will allow us to select treatments with greater specificity. We also look
forward to new agents greater in efficacy but safer in practice.
Whatever the fruits of neuroscience, there always will be a role for sensitive and
skillful interpersonal interactions in the approach to depressed human beings. The
response of a patient to an antidepressant pharmaceutical is independent of whether the
etiology of that patient's condition is biological, psychosocial, or some combination of
the two. To paraphrase Maimonides, seeing in the patient the suffering human being
often is the greatest challenge for the clinician, yet the greatest blessing we may
bestow.
REFERENCES
I. Weissman M. M., Bland R. C., Canino G. 1., et al: Cross-national epidemiology of major depression
and bipolar disorder. lAMA 276:293-299, 1996.
2. Regier D. A., Myers J. K., Kramer M., et al: The NIMH Epidemiologic Catchment Area program:
Historical context, major objectives, and study population characteristics. Arch Gen Psychiatry
41:934-941, 1984.
3. Blazer D. G., Kessler R. C., McGonagle K. A., Swartz M. S.: The prevalence and distribution of major
depression in a national community sample: The National Comorbidity Survey. Am 1 Psychiatry
151 :979-986, 1994.
4. Freud S.: Mourning and melancholia, in Anonymous: Collected Papers. London, Hogarth Press, 1946,
152-170.
5. Abraham K.: Notes on the Psychoanalytic Investigation and Treatment of Manic-Depressive Insanity
and Allied Conditions (1911). New York, Basic Books, 1960, p. 137.
6. Bibring E.: The mechanism of depression, in Greenacre P. (ed): Affective Disorders. New York,
International Universities Press, 1965, pp. 13-48.
7. Harlow H., Harlow M.: Learning to love. Sci Am 54:244-272, 1966.
8. Mahler M. S.: On sadness and grief in infancy and childhood: Loss and restoration of the symbiotic
love object. Psychoanal Study Child 16:332-351, 1961.
9. Kohut H.: The Analysis of the Self, New York, International Universities Press, 1971.
10. Kernberg 0.: Borderline Conditions and Pathological Narcissism. New York, Jason Aronson, 1975.
II. Beck A. T., Rush A. J., Shaw B., et al: Cognitive Therapy of Depression. New York, Guilford Press,
1979, p. I.
12. Seligman M., Maier S., Geer J.: The alleviation of learned helplessness and the dog. 1 Abnorm Soc
PsychoI73:256-262, 1968.
13. Lewinsohn P., Biglar A., Zeiss A.: Behavioral treatment of depression, in Davidson P. (ed): The
Behavioral Management of Anxiety, Depression and Pain. New York, Brunner-Maze!, 1976, pp. 91-
146.
14. Decker E.: The Revolution in Psychiatry. London, Collier-Macmillan, 1964, p. 108.
15. Ehringer H., Hornykiewicz 0.: Verteilung von noradrenalin and D opamin (3-hydroxytyramin) im
gehirn des menschen und ihr verhalten bei erkrankungen des estrapyramidalen systems. Klin
Wochenschr 38: 1236-1239, 1960.
16. Cotzias G. c., Van Woert M. H., Schiffer L. M.: Aromatic amino acids and modification of parkinson-
ism. N Engl 1 Med 276:374-378, 1967.
17. Carlsson A., Lindquist M., Magnusson T., Woaldeck B.: On the presence of 3-hydroxytyramine in
brain. Science 127:471, 1958.
94 ALAN J. GELENBERG, M.D., AND PEDRO L. DELGADO, M.D.
18. Schildkraut J. 1.: The catecholamine hypothesis of affective disorders: A review of supporting evi-
dence. Am J Psychiatry 122:509-522, 1965.
19. Bunney W. E. 1., Davis J. M.: Norepinephrine in depressive reactions: A review. Arch Gen Psychiatry
13:483-494, 1965.
20. Coppen A: The biochemistry of affective disorders. Br J Psychiatry 113:1237-1264, 1967.
21. Lapin I. P., Oxenkrug G. E: Intensification of the central serotonergic processes as a possible determi-
nant of the thymoleptic effect. Lancet 1:132-136,1969.
22. Murphy D. L., Cambell I., Costa 1. L.: Current status of the indoleamine hypothesis of the affective
disorders, in Lipton M. A, DiMascio A, Killam K. E (eds): Psychopharmacology: A Generation of
Progress. New York, Raven Press, 1978, pp. 1235-1247.
23. Heninger G. R., Charney D. S., Delgado P. L.: Neurobiology of treatments for refractory depression, in
Tasman A, Goldfinger S. M., Kaufman C. A. (eds): Review of Psychiatry. Washington, D.C., American
Psychiatric Press, 1990, pp. 33-58.
24. Charney D. S., Menekes D. B., Heninger G. R.: Receptor sensitivity and the mechanism of action of
antidepressant treatment. Arch Gen Psychiatry 38: 1160-1180, 1981.
25. Asberg M., Schalling D., Traskman-Bendz L., Wagner A.: Psychobiology of suicide, impulsivity, and
related phenomena, in Meltzer H. Y. (ed): Psychopharmacology:The Third Generation of Progress.
New York, Raven Press, 1987, pp. 655-668.
26. Roy A., Virkkunen M., Linnoila M.: Serotonin in suicide, violence, and alcoholism, in Coccaro E. E,
Murphy D. L. (eds): Serotonin in Major Psychiatric Disorders. Washington, D.C., American Psychi-
atric Press, 1990, pp. 187-208.
27. Siever L. 1., Davis K. L.: Overview: Toward a dysregulation hypothesis of depression. Am J Psychiatry
142: 1017-1031, 1985.
28. Winokur G., Coryell w.: Familial subtypes of unipolar depression: A prospective study of familial pure
depressive disease compared to depression spectrum disease. Bioi Psychiatry 32:1012-1018, 1992.
29. Bech P.: Symptoms and assessment of depression, in Paykel E. S. (ed): Handbook of Affective Disor-
ders. New York, Churchill Livingstone, 1992, pp. 3-13.
30. American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders, 4th ed.
Washington, D.C., American Psychiatric Association, 1994.
31. Panksepp 1.: Towards a general psychobiological theory of emotions. Behav Brain Sci 5:407-467,
1982.
32. Halgren E., Marinkovic K.: Neurophysiological networks integrating human emotions, in Gazzaniga
M. S. (ed): The Cognitive Neurosciences. Cambridge, MIT Press, 1995, pp. 1137-1151.
33. Akiskal H. S., McKinney W. I., Jr.: Overview of recent research in depression; integration of ten
conceptual models into a comprehensive clinical frame. Arch Gen Psychiatry 32:285-305, 1975.
34. Keller M. B., Harrison W., Fawcett 1. A., et al: Treatment of chronic depression with sertraline or
imipramine: Preliminary blinded response rates and high rates of undertreatment in the community.
Psychopharmacol Bull 31:205-212, 1995.
35. Quitkin EM., McGrath P. 1., Stewart J. w., et al: Atypical depression, panic attacks, and response to
imipramine and phenelzine. Arch Gen Psychiatry 47:935-941, 1990.
36. Brotman A. w., Falk W. E., Gelenberg A. 1.: Pharmacologic treatment of acute depressive subtypes, in
Meltzer H. (ed): Psychopharmacology: The Third Generation of Progress. New York, Raven Press,
1987, pp. 1031-1040.
37. Weissman M. M.: The psychological treatment of depression. Arch Gen Psychiatry 36:1261-1268,
1979.
38. Weissman M. M., Prusoff B. A., DiMascio A., et al: The efficacy of drugs and psychotherapy in the
treatment of acute depression episodes. Am J Psychiatry 136:555-558, 1979.
39. Herceg-Baron R. L., Prusoff B. A., Weissman M. M., et al: Pharmacotherapy and psychotherapy in
acute depressed patients. A study of attrition patterns in a clinical trial. Compr Psychiatry 20:315-325,
1979.
40. Weissman M. M., Klerman G. L., Prusoff B. A., et al: Depressed outpatients: Results one year after
treatment with drugs and/or interpersonal psychotherapy. Arch Gen Psychiatry 38:51-55, 1981.
2. DEPRESSION 95
41. Kovacs M.: The efficacy of cognitive and behavior therapies for depression. Am J Psychiatry
137:1495-1501, 1980.
42. Frankel F. H.: Current perspectives on ECT: A discussion. Am J Psychiatry 134:1014-1019, 1977.
43. Grahme-Smith D. S., Green A. R., Costain D. W.: Mechanism of the antidepressant action of electro-
convulsive therapy. Lancet 1:254-256, 1978.
44. Blier P., de Montigny C., Chaput Y.: A role for the serotonin system in the mechanism of action of
antidepressant treatments: Preclinical evidence. J Clin Psychiatry 51: 14-20, 1990.
45. Small 1. G., Kellams 1. 1., Milstein v., et al: Complications with electroconvulsive treatment combined
with lithium. Bioi Psychiatry 15:103-112, 1980.
46. Mandel M. R., Madsen J., Miller A. L., et al: Intoxication associated with lithium and ECT. Am J
Psychiatry 137: 1107 -1109, 1980.
47. Squire L. R., Slater P. C.: Bilateral and unilateral ECT: Effects on verbal and nonverbal memory. Am J
Psychiatry 135:1316-1320, 1978.
48. Fink M.: An adequate treatment? Editorial. Convulsive Ther 5:311-313, 1989.
49. Delgado P. L., Charney D. S., Price L. H., Aghajanian G. K., Landis H., Heninger G. R.: Serotonin
function and the mechanism of antidepressant action: Reversal of antidepressant induced remission by
rapid depletion of plasma tryptophan. Arch Gen Psychiatry 47:411-418, 1990.
50. Delgado P. L., Price L. H., Miller H. M., et al: Rapid serotonin depletion as a provocative challenge test
for patients with major depression: Relevance to antidepressant action and the neurobiology of depres-
sion. Psychopharmacol Bull 27:321-330, 1991.
51. Delgado P. L., Miller H. M., Salomon R. M., Licinio 1., Gelenberg A. 1., Charney D. S.: Monoamines
and the mechanism of antidepressant action: Effects of catecholeamine depletion on mood in patients
treated with antidepressants. Psychopharmacol Bull 29:389-396, 1993.
52. Delgado P. L., Gelenberg A. J.: Decision making in the use of antidepressants. in Directions in
Psychiatry, Vol. 14, New York, Hatherleigh, 1994, pp. 1-8.
53. Maas J. W.: Biogenic amines and depression: Biochemical and pharmacological separation of two types
of depression. Arch Gen Psychiatry 32:1357-1367, 1975.
54. Feighner 1. P.: The role of venlafaxine in rational antidepressant therapy. J Clin Psychiatry 55 (Suppl
9A):62-68, 1994.
55. Ellison 1. M., Milofsky J. E., Ely E.: Fluoxetine-induced bradycardia and syncope in two patients. J
Clin Psychiatry 51 :385-386, 1990.
56. Feder R.: Bradycardia and syncope induced by fluoxetine. J Clin Psychiatry 52:139, 1991.
57. Glassman A. H., Roose S. P., Bigger 1. T., Jr.: The safety of tricyclic antidepressants in cardiac patients:
Risk/benefit reconsidered. JAMA 269:2673-2675, 1993.
58. Jefferson J. W: Cardiovascular effects and toxicity of anxiolytics and antidepressants. J Clin Psychiatry
50:368-378, 1989.
59. Glassman A. H., Roose S. P.: Cardiovascular effects of tricyclic antidepressants. Psychiatr Ann
17:340-347, 1987.
60. Davidson J.: Seizures and bupropion: A review. J Clin Psychiatry 50:256-261, 1989.
61. Gelenberg A. J., Jefferson J. W: Lithium tremor. J Clin Psychiatry 56:283-287, 1995.
62. Lipinski J. F., Mallya G., Zimmerman P., Pope H.: F1uoxetine-induced akathisia: Clinical and theoreti-
cal implications. J Clin Psychiatry 50:339-342, 1989.
63. Rothschild A. J., Locke C. A.: Reexposure to fluoxetine after serious suicide attempts by three patients:
The role of akathisia. J Clin Psychiatry 52:491-493, 1991.
64. Preskorn S. H., Othmer S. c.: Evaluation of bupropion hydrochloride: The first of a new class of
atypical antidepressants. Pharmacotherapy 4:20-34, 1984.
65. Teicher M. H., Glod c., Cole 1. 0.: Emergence of intense suicidal preoccupation during fluoxetine
treatment. Am J Psychiatry 147:207-210, 1990.
66. Monteiro W.O., Noshirvani H. E, Marks 1. M., et al: Anorgasmia from clomipramine in obsessive-
compulsive disorder: A controlled trial. Br J Psychiatry 151: 107-112, 1987.
67. Jacobsen F. M.: F1uoxetine-induced sexual dysfunction and an open trial of yohimbine. J Clin Psychia-
try 53:119-122, 1992.
96 ALAN J. GELENBERG, M.D., AND PEDRO L. DELGADO, M.D.
68. Herman J. B., Brotman A. W., Pollack M. H., Falk W. E., Biederman J., Rosenbaum 1. E: F1uoxetine-
induced sexual dysfunction. J c/in Psychiatry 51:25-27, 1990.
69. Musher 1. S.: Anorgasmia with the use of fluoxetine. Am J Psychiatry 147:948, 1990.
70. Patterson W. M.: F1uoxetine-induced sexual dysfunction. J Clin Psychiatry 54:71, 1993.
71. Physicians' Desk Reference, Montvale, New Jersey, Medical Economics, 1994, p. 879.
72. Walker P. w., Cole J. 0., Gardner E. A., et al: Improvement in fluoxetine-associated sexual dysfunction
in patients switched to bupropion. J c/in Psychiatry 54:459-465, 1993.
73. Rothschild A. 1.: Selective serotonin reuptake inhibitor-induced sexual dysfunction: Efficacy of a drug
holiday. Am J Psychiatry 152:1514-1516, 1995.
74. Sovner R: Treatment of tricyclic antidepressant-induced orgasmic inhibition with cyproheptadine. J
Clin Psychopharmacol4:169, 1984.
75. Decastro R M.: Reversal of MAOI-induced anorgasmia with cyproheptadine. Am J Psychiatry
142:783, 1985.
76. Riley A. 1., Riley E. 1.: Cyproheptadine and antidepressant-induced anorgasmia. Br J Psychiatry
142:217 -218, 1986.
77. Kahn D. A.: Possible toxic interaction between cyproheptadine and phenelzine. Am J Psychiatry
144:1242-1243, 1987.
78. Steele T. E., Howell E. E: Cyproheptadine for imipramine-induced anorgasmia. J c/in Psychophar-
macoI6:326-327, 1986.
79. Goldbloom D. S., Kennedy S. H.: Adverse interaction of fluoxetine and cyproheptadine in two patients
with bulimia nervosa. J c/in Psychiatry 52:261-262, 1991.
80. Zajecka J., Fawcett 1., Schaff M., Jeffriess H., Guy C.: The role of serotonin in sexual dysfunction:
F1uoxetine-associated orgasm dysfunction. J Clin Psychiatry 52:66-68, 1991.
81. Cohen A. 1.: F1uoxetine-induced yawning and anorgasmia reversed by cyproheptadine treatment. J c/in
Psychiatry 53: 174, 1992.
82. Price 1., Grunhaus L. J.: Treatment of clomipramine-induced anorgasmia with yohimbine: A case
report. J Clin Psychiatry 51 :32-33, 1990.
83. Hollander E., McCarley A.: Yohimbine treatment of sexual side effects induced by serotonin reuptake
blockers. J c/in Psychiatry 53:207-209, 1992.
84. Gross M. D.: Reversal by bethanechol of sexual dysfunction caused by anticholinergic antidepressants.
Am J Psychiatry 139: 1193-1194, 1982.
85. Sorscher S. M., Dilsaver S. c.: Antidepressant-induced sexual dysfunction in men: Due to cholinergic
blockade? J Clin Psychopharmacol 6:53-55, 1986.
86. Yager J.: Bethanechol chloride can reverse erectile and ejaculatory dysfunction induced by tricyclic
antidepressants and mazindol: Case report. J c/in Psychiatry 47:2\0-211, 1986.
87. Segraves R T.: Reversal by bethanechol of imipramine-induced ejaculatory dysfunction. Am J Psychia-
try 144:1243-1244, 1987.
88. Pollack M. H., Rosenbaum J. E: Management of antidepressant-induced side effects: A practical guide
for the clinician. J c/in Psychiatry 48:3-8, 1987.
89. Balogh S., Hendricks S. E., Kang J.: Treatment of fluoxetine-induced anorgasmia with amantadine. J
c/in Psychiatry 53:212-213, 1992.
90. Labbate L. A., Pollack M. H.: Treatment of flouxetine-induced sexual dysfunction with bupropion: A
case report. Ann Clin Psychiatry 6: 13-15, 1994.
91. Boehnert M. T., Lovejoy E H., Jr.: Value of the QRS duration vs the serum drug level in predicting
seizures and ventricular arrhythmias after an acute overdose of tricyclic antidepressants. N Engl J Med
313:474-479, 1985.
92. Foulke G. E., Albertson R. E.: QRS interval in tricyclic antidepressant overdose: Inaccuracy as a
toxicity indicator in emergency setting. Ann Emerg Med 16:160-163, 1987.
93. Kapur S., Mieczkowski T., Mann 1. J.: Antidepressant medications and the relative risk of suicide
attempt and suicide. JAMA 268:3441-3445, 1992.
94. Crome P., Dawling S., Braithwaite R. A., et al: Effect of activated charcoal on absorption of nortrip-
tyline. Lancet 2:1203-1205, 1977.
2. DEPRESSION 97
95. Granacher R. D., Baldessarini R. J.: Physostigmine: Its use in acute anticholinergic syndrome with
antidepressant and antiparkinson drugs. Arch Gen Psychiatry 32:375-379, 1975.
96. Stoll A. L., Pope H. G., McElroy S. L.: High-dose fluoxetine: Safety and efficacy in 27 cases. J Clin
Psychopharmacol 1\ :225-226, 1991.
97. Beasley C. M., Jr., Sayler M. E., Weiss A M., Potvin J. H.: Auoxetine: Activating and sedating effects
at multiple fixed doses. J Clin Psychopharmacol 12:328-333, 1992.
98. Keck P. E., Jr., Yuckovic A, Pope H. G., Jr., Nierenberg A. A., Gribble G. w., White K.: Acute
cardiovascular response to monoamine oxidase inhibitors: A prospective assessment. J Clin Psycho-
pharmacoI9:203-206, 1989.
99. Gelenberg A. J.: MAOIs: Assorted concerns. Bioi Ther Psychiatry 11:33-36, 1988.
100. Teicher M. H., Cohen B. M., Baldessarini R. J., et al: Severe daytime somnolence in patients treated
with an MAOL Am J Psychiatry 145:1552-1556, 1988.
101. Dilsaver S. C.: Monoamine oxidase inhibitor withdrawal phenomena: Symptoms and pathophysiology.
Acta Psychiatr Scand 78:1-7, 1988.
102. Tollefson G. D.: Monoamine oxidase inhibitors: A review. J Clin Psychiatry 44:280-288, 1983.
103. Kaplan R. E, Feinglass N. G., Webster w., et al: Phenelzine overdose treated with dantrolene sodium.
JAMA 255:642-644, 1986.
104. Breheny EX., Dobb G. J., Clarke G. M.: Phenelzine poisoning, case report. Anesthesia 41:53-56,
1986.
105. Frank E., Kupfer D. J., Perel J. M., et al: Three-year outcomes for maintenance therapies in recurrent
depression. Arch Gen Psychiatry 47:1093-1099, 1990.
106. Himmelhoch J. M., Thase M. E., Mallinger A G., Houck P.: Tranylcypromine versus imipramine in
anergic bipolar depression. Am J Psychiatry 148:910-916, 1991.
107. Sachs G. S., Lafer B., Stoll A. L., et al: A double-blind trial ofbupropion versus desipramine for bipolar
depression. J Clin Psychiatry 55:391-393, 1994.
108. Spiker D. G., Weiss J. C., Dealy R. S., et al: The pharmacological treatment of delusional depression.
Am J Psychiatry 142:430-436, 1985.
109. Anton R. E, Sexauer J. D.: Efficacy of amoxapine in psychotic depression. Am J Psychiatry 140:1344-
1347, 1983.
110. Rothschild A J., Samson J. A, Bessette M. P., Carter-Campbell J. T.: Efficacy of the combination of
fluoxetine and perphenazine in the treatment of psychotic depression. J Clin Psychiatry 54:338-342,
1983.
HI. Siris S. G., Bermanzohn P. C., Gonzalez A., Mason S. E., White C. Y., Shuwall M. A.: The use of
antidepressants for negative symptoms in a subset of schizophrenic patients. Psychopharmacol Bull
27:331-335, 1991.
112. Roose S. P., Glassman A. H.: Treatment Strategies for Refractory Depression. Washington, D.C.,
American Psychiatric Press, Inc., 1990.
113. Anonymous: Treatment of xerostomia. Med Lett 30:74-76, 1988.
114. Harrison W. M., McGrath P. J., Stewart J. W., Quitkin E: MAOIs and hypertensive crises: The role of
OTC drugs. J Clin Psychiatry 50:64-65, 1989.
115. Winokur G.: The concept of secondary depression and its relationship to comorbidity. Psychiatr Clin
North Am, 13:567-583, 1989.
3
Bipolar Disorder
HEATHER S. HOPKINS and
ALAN J. GELENBERG, M.D.
I. INTRODUCTION
Bipolar disorder is one of the most common and treatable of the mood disorders,
affecting over I % of adults in industrialized nations over the course of a lifetime. l The
essential feature of bipolar disorder is the presence of one or more manic episodes;
usually, there is also a history of major depressive episodes. Bipolar II disorder is
characterized by the occurrence of one or more major depressive episodes and at least
one hypomanic episode.
The clinician can use the severity and pattern of a patient's illness as a baseline for
monitoring the effectiveness of pharmacotherapy. Because of the range of clinical
presentations and the wide variation in the course of bipolar illness, the clinician must
follow each individual carefully.
The successful management of bipolar disorder usually requires early, aggressive,
and long-term treatment. Lithium (Eskalith® and others) has had an important impact
on the treatment of bipolar disorder, literally saving the lives of many patients. Without
treatment, mortality is two to three times higher for bipolar patients than it is for the
general population; with lithium treatment, mortality is not significantly higher in
bipolar patients. 2 However, lithium does not work for every patient. An estimated 20-
40% of people with bipolar disorder do not adequately respond to or are unable to
tolerate lithium. Lithium's drawbacks-including side effects, the risk of toxicity, and
limited efficacy in some patients-have encouraged research into other treatments.
Divalproex sodium (Depakote®) is now approved by the u.s. Food and Drug Adminis-
tration (FDA) as a treatment for acute mania, and another anticonvulsant, car-
bamazepine (Tegretol® and others), has shown promise in clinical trials as an adjunct or
alternative to lithium or valproate treatment. Antipsychotic, anxiolytic/hypnotic, and
Heather S. Hopkins and Alan J. Gelenberg • Department of Psychiatry, University of Arizona Health
Sciences Center, Tucson, Arizona 85724.
The Practitioner's Guide to Psychoactive Drugs (Fourth Edition), Gelenberg and Bassuk, eds.
Plenum Publishing Corporation, New York, 1998
99
100 HEATHER S. HOPKINS AND ALAN J. GELENBERG, M.D.
antidepressant medications are used as adjuncts to the mood stabilizers for episodes of
acute mania and acute depression, and nonpharmacological treatments such as electro-
convulsive therapy and psychotherapy can also play important roles in the treatment of
this disorder.
A. Clinical Presentation
Bipolar disorder occurs in about 1.6% of the general population, with an equal
incidence in men and women.) In a cross-national collaborative study, the lifetime rates
of bipolar disorder were found to be more consistent across countries than the rates for
major depression (0.3% in Taiwan to 1.5% in New Zealand).3 Patients usually develop
their first mood episode (most often mania or hypomania) between the ages of 20 and
40 years (average 32), although presentation during the teenage years is not uncom-
mon. Late onset of mania should prompt a neurologic and medical differential diag-
nosis, since mania may be secondary to many other conditions. Gender mayor may not
be a factor in the age of onset of bipolar disorder. 4 - 7 There is some evidence that
female bipolar patients have more depressive episodes than male bipolar patients do,
but again the data are contradictory. 4
In mania, elevated mood is common but not universal. Sometimes, depressive
symptoms break through a facade of emotional lability. Pressure of speech, flight of
ideas, increased motor activity, and decreased sleeping time often accompany the sense
of elation and well-being. These patients are very self-involved but often can sense the
vulnerabilities of others and become inappropriate and intrusive. The manic patient
often gets too involved in pleasurable activities, with painful consequences in family
relationships, social interactions, economic stability, and health. If mania progresses,
the patient's sense of humor may border on anger and irritability. Pressured speech can
deteriorate into fragmented sentences and clang associations, while a sense of well-
being may escalate into grandiosity. In the most extreme form, the patient becomes
psychotic. Auditory hallucinations, delusions, and paranoia are frequent, and incoher-
ence, sensory distortions, agitation, and combativeness often accompany the most
disorganized states. Manic patients frequently require hospitalization to prevent them
from harming themselves or others.
Clinicians have described some cases of prolonged mania or intermittent mania
without depression. Most commonly, however, the illness is bipolar, and the first
hospitalization is usually for mania. Bipolar depression typically is indistinguishable
from other forms of major depression, except that episodes may be more frequent and
shorter in duration. The fourth edition of the Diagnostic and Statistical Manual of
Mental Disorders (DSM-IV) defines the symptom criteria, which include persistently
lowered mood; psychomotor agitation or retardation; feelings of worthlessness or guilt;
suicidal ideation or attempts; and various somatic concerns, such as weight loss,
insomnia, and fatigue. 8
Occasionally, a mood episode in a bipolar patient is not clearly mania or depres-
sion, but a mixture of both. Recent evidence suggests that this mixed picture may be
less treatment-responsive and have a poorer prognosis than simple mania or depres-
3. BIPOLAR DISORDER 101
sion.9 Similarly, rapid cycling, in which a patient has 4 or more episodes in one year,
and "cycling episodes," in which patients move from one mood pole to the other
without an intervening period of euthymia, also have a more guarded prognosis.to,1I
25 can expect to lose approximately 9 years of life, 14 years of major activities such as
work, school, or child rearing, and 12 years of normal health.30 With optimal treatment,
however, the woman will regain 6.5 years of life, 10 years of effective major activity,
and 8.5 years of normal health. Poorly controlled mania can lead to family dysfunction,
economic collapse, auto accidents, and the contraction of sexually transmitted diseases.
Although the trends gleaned from longitudinal studies can provide a general set of
expectations about the course of bipolar disorder, the clinician must assess the nature
and course of each patient's illness individually. An appropriate treatment plan can be
developed only after evaluating the impact and meaning of the illness in the patient's
life, along with objective clinical symptoms.
A. Electroconvulsive Therapy
Electroconvulsive therapy is an effective treatment for mania and for major de-
pression (see Chapter 2). When an acutely manic patient is unresponsive or intolerant
to medication treatment or medication presents a threat to the patient's well-being (e.g.,
during pregnancy), ECT should be seriously considered and may be lifesaving. Al-
though we have no coherent theories about why ECT is effective in acute mania, there
are widespread clinical impressions and evidence from the literature of its safety and
efficacy. Reports in the medical literature over the past 50 years show ECT to be
associated with remission or marked clinical improvement in 470 (80%) of 589 manic
patients: 78% of manic patients in early studies, 85% of patients in six retrospective
studies reported since 1976, and 77% of patients in two prospective studies. 31
Because there is the impression in the field that ECT works faster than phar-
macologic treatment, it is the preferred treatment for patients who are homicidal,
psychotic, or severely suicidal. Unfortunately, there is little experimental evidence to
support this belief. One study found ECT to be superior to lithium during the first 8
weeks of treatment, especially for severely manic patients and those with mixed
states. 32 After 8 weeks, however, there were no significant differences between the
lithium- and ECT-treated patients on ratings of manic and depressive symptoms or in
the incidence of relapses, recurrences, or rehospitalizations.
As in acute mania, ECT is the first choice for bipolar depressed patients who
require a rapid response, cannot tolerate or be exposed to medication, or do not respond
to medication. There is some evidence that ECT has a more rapid therapeutic onset
than antidepressant drugs. 33 - 35 Bilateral ECT may be more efficacious than unilateral
ECT for bipolar patients, and patients who receive a dosage over twice the initial
seizure threshold, with either unilateral and bilateral treatment, may show more rapid
improvement of their depressive symptoms.
3. BIPOLAR DISORDER 103
B. Milieu Therapy
To ensure a manic patient's safety and to contain symptoms, inpatient care may be
necessary. These patients, however, frequently disrupt a ward milieu. The patient's
infectiously elevated mood may be reinforced by staff and other patients, and irri-
tability and hostility can be highly provocative. Structuring daily activities, correcting
cognitive distortions, and limiting external stimulation by staff and patients may reduce
the patient's excitement. Caretakers should remember that milieu therapy and phar-
macotherapy are synergistic. A safe and stable environment promotes medication com-
pliance, and the salutary effect of drugs increases insight, emotional availability, under-
standing, and formation of social attachments.
c. Group Therapy
Fieve has described "lithium groups" designed to discuss issues related to mainte-
nance therapy.36 This approach is not only time efficient for clinicians but also advan-
tageous for patients. Sharing similar concerns about medication usage can be both
supportive and growth-producing and can increase compliance. Studies have shown
that group therapy, in addition to a medication regimen, can decrease rates of hospital-
ization and improve social and occupational functioning. 37 - 39
Families and significant others can also benefit from learning about mood disor-
ders and sharing common approaches to dealing with their loved ones. Support and
practical counseling are valuable, and people close to the patients can learn the prodro-
mal signs of impending relapse, noninflammatory ways of coping with symptoms, how
to predict side effects, the importance of avoiding sleepless periods, and more. Bipolar
patients who have received family therapy in addition to pharmacologic treatment in
several different studies have shown significantly better social functioning and family
interactions and required fewer hospitalizations than those who have received only
standard pharmacotherapy.4O-43
Although group therapy can be helpful during the maintenance phase of treatment
104 HEATHER S. HOPKINS AND ALAN J. GELENBERG, M.D.
for patients with mood disorders, it can be counterproductive during episodes of acute
illness, particularly for hospitalized patients. A depressed patient might become irrita-
ble with forced socialization. A manic patient, in contrast, can become more excited
with interpersonal stimulation and, in tum, may be provocative to others.
Ill. PHARMACOTHERAPY
Pharmacologic agents have markedly decreased the morbidity associated with
bipolar illness, recurrent depressions, and several other cyclic psychiatric disturbances.
Antipsychotic drugs provide an effective means of controlling the behavioral symp-
toms and thought disorder that accompany mania and hypomania. Antidepressant
medications usually reverse depressive episodes. Lithium can diminish the symptoms
of acute mania and, over the long term, often suppresses both manic and depressive
episodes. Valproate is now considered a first-line treatment for acute mania. Car-
bamazepine, clozapine (Clozaril®), and various other agents are being tried as adjuncts
or second-line options for patients who do not respond to lithium satisfactorily.
3. BIPOLAR DISORDER 105
A. Lithium
Naturally occurring salts of the chemical element lithium have been used to treat
various medical conditions since the 1800s. Lithium currently is used to treat acute
hypomanic or manic episodes and recurrent mood disorders. Despite its effective-
ness, its mechanism of action remains unclear. One theory is that lithium alters neuro-
nal function by affecting the distribution and kinetics of other ions that have similar
properties, such as sodium, potassium, magnesium, and calcium. 46 However, abnor-
malities in these systems do not appear to be specific to mood disorders. Lithium's
mechanism of action may be related to its effects on the neurotransmitters serotonin,
dopamine, norepinephrine, or acetylcholine. A third possibility is that lithium stabilizes
mood through its effects on second-messenger and signal-transduction systems 46 - 50 or
by its actions on signal-transducing guanine nucleotide-binding (G) proteins. 51 The G
proteins help regulate neuronal function and maintain the functional balance between
neurotransmitter systems in the central nervous system (eNS) and may be the molecu-
lar sites underlying the pathophysiology of this illness.
In acute hypomanic or manic episodes, lithium is frequently effective within 1-2
weeks but may require several more weeks or even a few months to contain the episode
fully. Sometimes an antipsychotic agent [such as haloperidol (Haldol® and others)]
should be used acutely to manage behavioral excitement and acute psychotic symp-
toms. After the acute symptoms are diminished, lithium maintenance therapy decreases
the number, severity, and frequency of mood episodes. Even with drug treatment,
however, some patients experience various symptoms, intermittent periods of distress,
or unwanted effects. To achieve optimal results, the clinician may need to experiment
with various dosage levels and continue maintenance therapy for a year or longer. In
addition, patients with acute depressive symptoms tend not to respond to lithium alone.
Eliminating significant mood swings is the optimal clinical goal of lithium main-
tenance therapy. However, mood fluctuations, and even the recurrence of a major
episode, do not indicate that lithium treatment has failed. Some clinicians feel that
manic or depressive symptoms that occur within a few months of initiating treatment
represent a relapse of the initial episode and may require medication adjustment.
Patients on a maintenance lithium regimen also may develop mild manic and depres-
sive symptoms that subside spontaneously or respond to interpersonal support or a
transient increase in lithium. Hypomania appears more likely than minor depression to
be a prodrome of a major episode. Optimal dosing of lithium can diminish subsyndro-
mal symptoms. 52
Long-term studies have consistently demonstrated that lithium maintenance sup-
presses mood swings in most bipolar patients. 53 However, individuals must take appro-
priate amounts of medication consistently. Most failures of lithium therapy occur
during the first year of treatment and result from inadequate resolution of the initial
episode, inadequate regulation of dosage, or lack of patient compliance. About 20-
30% of patients discontinue lithium therapy on their own.
1. Pharmacokinetics
TABLE 1. Pharmacokinetics
nate to plasma concentrations, and lithium clearance is relatively constant for each
patient. With a single dose of lithium, 50% is excreted within 5-8 hr. In patients on
maintenance lithium, the half-life is about 24 hr. However, depending on the age and
kidney status of patients, the half-life may vary widely (approximately 18 hr in young,
healthy individuals to 36 hr in older patients).
Effective regulation of lithium depends on the sodium and fluid balance of the
body. When lithium is initially administered, a cation balance among lithium, sodium,
and potassium occurs; sodium and potassium excretion increases. Clinically, the bal-
ance between lithium and sodium is important for the following reasons:
1. Sodium depletion can result in marked lithium retention and possible toxicity.
2. High levels of lithium can lead to increased sodium excretion.
In cases of toxicity, the loss of body sodium obligates more lithium retention by the
kidney, prolonging the toxicity. Decreased fluid intake also can diminish lithium excre-
tion and could lead to intoxication.
2. Adverse Reactions
Lithium causes a wide range of unwanted effects (see Table 2). Many appear at the
onset of therapy, but patients soon become tolerant to them, and they cease to be a
problem. As in all clinical pharmacology, patients' sensitivity to a side effect at a given
Initial
Nausea and vomiting
Diarrhea
Lethargy
Drowsiness
Muscle weakness
Fine tremor
Increased thirst (polydipsia)
Increased urination (polyuria)
Potentially indicative of toxicity
Nausea, vomiting, diarrhea
Drowsiness and mental dullness
Slurred speech
Confusion
Vertigo
Coarse tremor and twitching
Muscle weakness
Choreoathetosis
Persistent but benign
Increased thirst
Increased urination
Fine tremor
Weight gain
Edema
108 HEATHER S. HOPKINS AND ALAN J. GELENBERG, M.D.
serum lithium concentration will vary. For some patients, a concentration necessary for
an effective result will produce unacceptable adverse reactions. For most patients, side
effects are tolerable at therapeutic levels, but the appearance or reappearance of certain
problems-such as tremor, gastrointestinal disturbances, or increased urination-
could herald an increase in blood level to toxic concentrations (see Section III.AA).
a. Endocrine. The long-term use of lithium may cause various endocrine abnor-
malities.
i. Thyroid. Lithium therapy can produce significant changes in thyroid function-
ing. It inhibits several steps in the process of hormone synthesis and degradation,
including iodine uptake by the thyroid gland, iodination of tyrosine, the release of
triiodothyronine (T3) and thyroxine (T4), the peripheral degradation of thyroid hor-
mones, and the stimulating effects of thyroid-stimulating hormone (TSH). Patients can
usually compensate for the initial decrease in thyroid hormones. However, approx-
imately 5% of patients treated with lithium develop signs of hypothyroidism; another
3% develop a diffuse nontender goiter. 56 This latter effect occurs more commonly in
women, between 5 months and 2 years after starting treatment. In addition, about 30%
of patients develop consistently increased levels of TSH indicated by elevated hormone
or abnormal clinical signs and symptoms. 57 Each of these thyroid abnormalities typ-
ically remits when lithium is stopped.
Hypothyroidism can be confused with a depressive disorder. Therefore, the practi-
tioner should be sensitive to clinical manifestations and laboratory findings indicative
of thyroid disease. 58
Since lithium affects thyroid function so profoundly, the clinician should take the
following measures during the various stages of maintenance therapy:
Before Starting Treatment
1. Screen for signs and symptoms of preexisting thyroid disease.
2. Obtain laboratory tests, including TSH, thyroxine (T4) T4I (free thyroxine
index), and T3RU (resin uptake).
During Treatment
1. Obtain a TSH measurement every 6 months during the first year and yearly
thereafter.
2. Follow up an elevated TSH with a complete battery of thyroid tests.
3. Maintain a clinical index of suspicion for thyroid dysregulation.
4. Initiate thyroid replacement therapy if hypothyroidism or elevated TSH oc-
curs (discontinuing lithium is not necessary).
In rare instances, lithium appears to trigger cases of hyperthyroidism. A high
index of suspicion is the best guideline, with appropriate testing of thyroid function to
confirm the clinician's impressions.
ii. Diabetes Mellitus. Lithium has many effects on carbohydrate metabolism,
including altered glucose tolerance, that are incompletely understood. Rarely, patients
may develop diabetes mellitus. 59 In these cases, the clinician must decide whether or
not the risk from diabetes outweighs the morbidity of the individual's affective disor-
3. BIPOLAR DISORDER 109
der. Generally, diabetic patients tolerate lithium well. Sometimes, however, this agent
causes altered sensitivity to insulin, requiring a dosage readjustment.
b. Renal. As noted earlier, the body depends almost entirely on the kidneys to
eliminate lithium. In a patient with no glomerular filtration, who is totally dependent on
dialysis, a dose of lithium will remain in the body from one dialysis until the next.
Patients with diminished renal filtration will have prolonged elimination half-lives of
lithium.
Aside from the kidney being the primary organ of lithium excretion, it is also the
site of some of the most common side effects, polydipsia and polyuria, which affect
roughly 60% of patients taking lithium. Although these effects usually are reversible, in
occasional patients they may persist after lithium has been discontinued, suggesting
some degree of structural impairment. There have also been rare reports of nephrotic
syndrome associated with lithium treatment.
i. Decreased Concentration. Presumably by counteracting the effects of anti-
diuretic hormone (ADH or vasopressin) on the renal tubule, lithium decreases the
kidney's maximal concentrating capacity, analogous to nephrogenic diabetes insipidus.
The obligatory water loss goes up, often to 3 liters or more per day. Patients must not
restrict their water intake, for they will be subject to dehydration with attendant lithium
intoxication and potential renal shutdown. Instead, they must drink as much fluid as it
takes to maintain hydration.
For some patients, once-a-day dosing of lithium can reduce polyuria. 60 For others,
a lower dose of lithium may help, but lower levels can bring reduced protection against
relapse. The use of diuretics (paradoxical though it sounds, with the mechanism poorly
understood) and a diet low in protein and salts are alternative treatment options. 61
(Some diuretics can markedly increase serum lithium levels. If a diuretic is used to
decrease urine volume, serum lithium levels should be closely monitored and the
lithium dose reestablished.)
ii. Structural Changes and Glomerular Filtration Alterations. In 1977 reports
began to appear of kidney structural damage in patients treated with lithium. The
incidence of this finding varied depending on patients studied, and so did the nature of
the lesions and relation to therapy. The changes described were mostly of a tuberoin-
terstitial nature. Although similar changes have been described in patients with bipolar
illness before treatment with lithium, they probably occur at a greater frequency in
patients who have been so treated. Mellerup et al. estimate a 10% incidence of struc-
tural changes in lithium-treated patients. 6o Unquestionably, lithium intoxication can
lead to acute functional and structural alterations.
Fortunately, the most vital kidney function, glomerular filtration, is minimally
impaired by long-term lithium treatment absent toxic levels. Even patients treated for
several decades appear to have negligible decreases in glomerular filtration rates be-
yond the decline expected with age. Only a few cases of renal failure have been
apparently linked to therapeutic lithium. 63
iii. Renal Function. Earlier recommendations for extensive testing of kidney
function-including 24-hr urine collections and concentrating capacity tests-have
proven unnecessary. 58 Instead, pretreatment assessment should include a history and
110 HEATHER S. HOPKINS AND ALAN J. GELENBERG, M.D.
necessary laboratory testing to rule out renal pathology or, if present, to assess its
nature and degree. First, the clinician should complete personal and family medical
histories and review of systems. Laboratory tests should include serum creatinine and
blood urea nitrogen (BUN) determinations along with routine urinalysis.
Once lithium therapy has been initiated, a serum creatinine determination every 6
months should suffice. Beware, however, of progressive rises in serum creatinine, even
within the normal range, as these may herald a decline in glomerular filtration rate.
Similarly, a rise in serum lithium without an increase in dose may reflect decreasing
glomerular filtration. Patients on maintenance therapy should be questioned peri-
odically about changes in urine volume, and their families should be counseled about
the importance of adequate hydration-especially under conditions of heat, increased
physical activity, gastroenteritis, or other situations that may alter fluid and electrolyte
balance.
iv. Other Kidney Effects. Rare cases of nephrotic syndrome have been described
in patients treated with lithium. Although this condition is usually detected by asymp-
tomatic proteinuria on routine urinalysis, occasional patients develop edema. Almost
all cases have been reversible after discontinuing lithium, although at least one patient
required continued steroid therapy.64
Because lithium and sodium are handled similarly in renal reabsorption mecha-
nisms, sodium depletion can lead to increased lithium reabsorption and, consequently,
elevated blood levels. Thus, using salt-lowering diuretics or instituting a restricted-
sodium diet can raise blood lithium levels in a lithium-treated patient. Limited research
on strenuous exercise suggests that, contrary to initial expectations, blood levels of
lithium are more likely to decrease than to increase, presumably because more lithium
than sodium is lost in sweat.
c. Hematological. Mild to moderate elevations of the white blood cell count
(typically 12,000-15,000 cells/mm3 ) occur commonly with both acute and chronic
administration of lithium. 65 Neutrophils are most affected. Lithium leukocytosis is
clinically benign and reversible and can develop at various doses and at any stage of
treatment.
d. Cardiovascular. Lithium has significant effects on the heart. However, ad-
verse reactions in clinical practice are rare, even in patients with known heart disease.
Perhaps because of its effects on potassium balance, lithium causes electrocardiogram
(EKG) T-wave flattening or inversion in 20% or more of individuals. 66 Sometimes with
normal doses, U waves also appear. Since these are common changes, the clinician
should obtain a baseline EKG prior to starting therapy.
Several types of conduction problems have appeared during lithium therapy,
including first-degree atrioventricular (A-V) block, irregular or slowed sinus node
rhythms (particularly in elderly patients), and increased numbers of ventricular prema-
ture contractions. 67 •68 Some individuals have also developed severe congestive heart
failure, cardiomyopathy, and ventricular tachycardia. However, the degree to which
lithium therapy contributed to the development of these disorders was unclear. During
toxic states, some patients have developed persistent ventricular tachycardia, atrial
fibrillation, advanced A-V block, and vascular collapse. 69
3. BIPOLAR DISORDER 111
These symptoms most often remit quickly and usually do not require changing the
dosage.
A fine rapid hand tremor accompanies lithium therapy in 15-70% of patients,
depending on dose. It is an action or postural tremor (made worse by sustained posture
or purposeful movements) and may persist from the beginning of treatment, appear at
any point, or disappear and then recur.70 However, it usually appears when therapy
starts and decreases over time. CNS stimulants (including caffeine), anxiety, muscle
tension, and, occasionally, antiparkinson drugs worsen the tremor, whereas sedative-
hypnotic agents may improve it. Patients with other forms of action tremor (i.e.,
familial, idiopathic, senile) develop lithium-induced tremors more frequently. Rarely,
tremulousness may involve the upper extremities, face, or eyelids. Lowering the dose
of lithium can provide relief. However, if the tremor still persist, propranolol (Inde-
ral. and others), 20-160 mg per day, may help.71 Other remedies that have been
studied include other l3-blockers, primidone, safflower oil or evening primrose oil,
inositol, and potassium, but so far results are inconclusive. 70
Lithium also causes various other troublesome effects, including dysarthria, ver-
tigo, ataxia, tinnitus, nystagmus, autonomic slowing of bladder and bowel function,
visual distortion, muscle irritability (e.g., twitching, fasciculations, facial spasm, in-
creased tendon reflexes, clonus, and choreoathetosis), and even an organic brain syn-
drome. When any such signs appear abruptly in the course of lithium therapy, they may
signal toxicity and should prompt a blood level assay.
Extrapyramidal reactions, although sometimes occurring without any apparent
predisposition, are more likely in the elderly, those taking antipsychotic medication,
and those with toxic blood levels of lithium. Cogwheeling and generalized muscle
rigidity are more common signs.72 Sometimes these reactions appear during the first
months of therapy at therapeutic levels of lithium. However, they more often accom-
pany serum lithium levels greater than 1.5 mEq/L. With levels above 3.0 mEq/L,
patients usually show profound neurotoxic effects that can progress to seizures and
incontinence, stupor, coma, brain damage, and death.
In addition to the adverse reactions already mentioned, rare cases of brain damage
have been reported with usual (i.e., nontoxic) doses of lithium (see Chapter 4). This
neurotoxicity occurs most often in patients who are elderly, schizophrenic, or already
neurologically impaired.
To Minimize CNS Effects
1. Take an appropriate history for schizophrenia, possible CNS disease, tremors,
and antipsychotic medication usage; these will dictate greater caution in
administering lithium.
2. Note and carefully follow any mild, initial neurological signs (i.e., mental
dullness, poor concentration, weakness, lethargy, or tremors).
3. Discontinue lithium and obtain a serum level in patients who develop any
significant neurological symptoms (particularly an organic brain syndrome).
4. Treat hand tremors by reassuring the patient or administering 20-160 mg of
propranolol per day. Failing those, consider lowering the dose.
3. BIPOLAR DISORDER 113
h. Ocular. Lithium therapy rarely causes adverse reactions involving the eyes.
Sometimes tearing, itching, burning, or blurring may occur during the first few weeks
of treatment. With decongestant eyedrops and time, most of these reactions abate.
Occasionally, more significant reactions develop, including exophthalmos and worsen-
ing of cataracts, and eight cases of bilateral papilledema have been reported. 73 - 77
However, the role of lithium in these disorders remains unclear.
i. Weight Gain. Reports about weight gain during lithium therapy vary widely.
Although the etiology of this effect is unclear, many causes have been hypothesized,
including altered carbohydrate and lipid metabolism, increased fluid intake and reten-
tion, improved appetite with the resolution of affective episodes, diminished thyroid
functioning, and increased intake of high-caloric fluids. Since this adverse reaction is
frequent, patients should be warned and instructed about possible dietary restrictions.
For those who gain weight, limiting calories usually results in appropriate weight loss.
Thirst should be quenched with no-calorie beverages.
j. Other Adverse Reactions. These include a persisting metallic taste and pre-
tibial edema.
3. Drug Interactions
Lithium and other drugs, particularly the diuretics, have various synergistic effects
(see Table 3). Thiazide diuretics commonly increase lithium levels by decreasing
clearance. This reaction can occur quickly, resulting in significant increases in blood
levels and potential toxicity. Some other diuretics, notably osmotic drugs and carbonic
anhydrase inhibitors, decrease lithium levels by increasing excretion.
Antipsychotic agents are commonly prescribed with lithium for acutely disor-
ganized patients. This practice is usually safe. However, some individuals have devel-
oped acute neurotoxicity and/or permanent brain damage (described most often with
haloperidol). These reactions occur rarely and may reflect lithium toxicity only or a
neuroleptic malignant syndrome (see Chapter 4). Whether or not a combined toxic
effect occurs is still unclear. See Table 3 for a list of other medications that, when
combined with lithium, can affect lithium levels or increase adverse reactions.
As mentioned earlier, lithium should be discontinued when bipolar patients are
treated with ECT to avoid a possible increase in confusional states.
To Minimize Possible Drug Synergisms
1. Rationalize polypharmacy.78
2. Lower the dose of lithium in patients starting on drugs known to elevate blood
lithium levels.
3. Monitor serum lithium levels more often in patients on osmotic diuretics or a
carbonic anhydrase inhibitor.
4. Administer furosemide (Lasix@ and others) as a diuretic rather than others if it
fits the patient's medical needs. (This drug may not affect blood lithium
levels.)
114 HEATHER S. HOPKINS AND ALAN J. GELENBERG, M.D.
TABLE 3. (Continued)
Adverse reactions:
Antipsychotics: extrapyramidal symptoms, neurotoxicity, somnambulism
Atracurium (Tracium®): prolonged neuromuscular blocking effects
Hydroxyzine (Atarax®, Vistaril®, and others): cardiovascular toxicity
Iodine: antithyroid effects
Methyldopa (Aldomet® and others): hypertension, toxic symptoms at normal blood levels
Pancuronium (Pavulon®): prolonged neuromuscular blocking effects
Succinylcholine (Anectine® and others): prolonged neuromuscular blocking effects
Verapamil (Calan® and others): neurotoxicity and bradycardia
5. Discontinue lithium and obtain a blood level in any patient who develops
signs of an organic brain syndrome while taking antipsychotic medication.
6. Avoid abruptly discontinuing caffeinated beverages (e.g., coffee, tea, colas)
and monitor blood levels more closely in patients who markedly change their
consumption of caffeine.
7. Avoid the use of methyldopa and hydroxyzine with lithium.
8. Warn patients about taking nonsteroidal anti-inflammatory drugs (NSAIDS;
see Table 3 for a list of NSAIDs).
4. Lithium Toxicity
Toxic blood levels of lithium can occur acutely by accidental or intentional over-
dose or chronically through excessive dosing, decreased renal function, drug interac-
tions, diminished salt intake, or other fluid and electrolyte abnormalities. Some patients
will develop toxic symptoms and signs at doses usually considered therapeutic, where-
as other patients will tolerate fairly high plasma concentrations without ill effect. Most
probably, these individual variations reflect differences in brain levels achieved at a
given serum concentration, but they also may reflect differences in end-organ sensi-
tivities.
Patients should not have serum levels above 1.5 mEq/L, and certainly alarm bells
must sound at 2.0 rnEq/L and above. Twelve-hour levels above 3.0 mEq/L are an
emergency and should be dealt with immediately.
Gastrointestinal symptoms may initially appear, followed or accompanied by CNS
depression. This may include somnolence, sluggishness, the various hallmarks of an
organic brain syndrome, dysarthria, seizures, choreoathetoid movements, increased
muscle tone, and increased deep tendon reflexes. Cardiovascular collapse marked by
lowered blood pressure, irregular cardiac rhythm, decreased urine output, and conduc-
tion abnormalities (and EKG changes) can be life-threatening.
Acute intoxication causes significant CNS depression. Patients typically develop
pyramidal tract signs and impaired consciousness or coma. Individuals who chronically
take too much medication develop CNS impairment more gradually. Sluggishness and
drowsiness may progress over a period of days. Often, gastrointestinal symptoms,
116 HEATHER S. HOPKINS AND ALAN J. GELENBERG, M.D.
slurred speech, ataxia, and coarse trernor accornpany these changes. If the initial signs
of chronic intoxication are overlooked, a rnore florid CNS syndrorne rnay develop,
rnost often rnanifested by hyperpyrexia and stupor or corna. It can also include neuro-
logical asymmetries, nystagrnus, stiff neck, and hyperextension of the extrernities.
The possibility of lasting neurological irnpairment as a result of lithiurn intoxica-
tion probably is a product of the rnaxirnal blood concentration and the duration of toxic
levels. Various permanent neurological sequelae are possible, rnost commonly involv-
ing cerebellar and cognitive dysfunctions,?9
Lithiurn toxicity is generally rnanaged by supportive rneasures. If the toxicity
occurs as part of an acute rnedication regirnen or rninor overdose, blood levels never
exceed 3.0 rnEq/L, and kidney function is intact, careful observation usually suffices.
Since lithiurn is excreted rapidly, the syndrorne rnost often abates within a few days.
However, in a large or chronic overdose, very large total-body lithiurn stores rnay
accurnulate. In these cases, the patient often suffers persistent (several days or longer)
life-threatening CNS depression and cardiovascular irnpairment.
1. Rapidly assess (including clinical signs and syrnptorns, serurn lithiurn levels,
electrolytes, and EKG), rnonitor vital signs, and rnake an accurate diagnosis.
2. Discontinue lithiurn.
3. Support vital functions and rnonitor cardiac status.
3. BIPOLAR DISORDER 117
4. Limit absorption
a. If the patient is alert, provide an emetic.
b. If the patient is obtunded, intubate and suction nasogastrically (prolonged
suction may be helpful, since lithium levels in gastric fluid may remain
high for days).
5. Prevent infection in comatose patients by body rotation and pulmonary toilet.
6. In all cases, vigorously hydrate (ideally, 5-6 liters per day); monitor and
balance the electrolytes.
7. In moderately severe cases
a. Implement osmotic diuresis with urea, 20 g IV 2 to 5 times per day, or
mannitol, 50-100 g IV per day.
b. Increase lithium clearance with aminophylline, 0.5 g up to every 6 hr, and
alkalinize the urine with IV sodium lactate.
c. Ensure adequate intake of NaCI to promote excretion of lithium.
8. Implement hemodialysis in the most severe cases. These are characterized by
a. Serum lithium levels between 2.0 and 4.0 mEq/L with severe clinical
signs and symptoms (particularly decreasing urinary output and deepen-
ing CNS depression).
b. Serum lithium levels greater than 4.0 mEq/L. Most patients completely
recover from lithium toxicity; several may die; some develop permanent
neurological damage.
5. Pregnancy
In the United States, lithium comes mostly as a carbonate salt in oral preparations
of 300 mg (8.12 mEq). This dosage and composition usually meet most clinical needs.
However, a few patients exhibit sensitivities to various preparations and can benefit
from alternate forms, such as slow-release tablets or lithium citrate syrup.
118 HEATHER S. HOPKINS AND ALAN J. GELENBERG, M.D.
B. Valproate
Valproic acid, an eight-carbon, branched-chain fatty acid, was synthesized in 1882
and recognized to have antiseizure activity in 1963. Valproate's efficacy in stabilizing
mood disturbances was first reported in 1966, and in 1995, divalproex sodium, an
enteric-coated, stable, equimolar combination of sodium valproate and valproic acid,
was approved by the U.S. Food and Drug Administration (FDA) for treating acutely
manic bipolar patients. 88
Valproate's antimanic mechanism of action has yet to be established but may be
related to its increase of 'Y-aminobutyric acid (GABA) concentrations, by decreasing
either GABA's metabolism or its reuptake in brain tissues. Another theory is that
3. BIPOLAR DISORDER 119
1. Pharmacokinetics
All valproate preparations are rapidly absorbed after oral administration, reaching
peak plasma levels within 2-4 hours of ingestion (see Table 1). Food may delay
absorption but does not affect bioavailability. Valproate is rapidly distributed and
reaches the CNS within minutes of administration. It is highly bound (90-95%) to
plasma proteins at serum concentrations up to 50 j.Lg/mL. As the concentration in-
creases from 50 to 100 j.Lg/mL, the percentage bound decreases to 80-85% and the free
fraction becomes progressively larger, thus increasing the concentration gradient into
the brain and also increasing its metabolism. Valproate's half-life ranges from 5 to 16
hr, depending on whether it is taken alone or with other medications, and it takes 1-4
days to attain steady state. The half-life may be considerably longer in patients with
impaired hepatic function, in the elderly, and in children up to 18 months of age. It may
be much shorter in patients concomitantly taking hepatic enzyme-inducing anticonvul-
sants.
Valproate is primarily metabolized in the liver through a variety of conjugative
and oxidative processes and produces many metabolites, some of which may have
pharmacologic or toxic activity. The increased free serum fraction at higher blood
levels can lead to faster metabolism. Therefore, some patients require a greater dosage
increment to raise a blood level from 100 to 125 j.Lg/mL, for example, than to go from
75 to 100 j.Lg/mL. The rate of metabolism is faster in children and in patients concur-
rently using enzyme-inducing medications, such as phenytoin, phenobarbital, pri-
midone, and carbamazepine. Val pro ate is eliminated by the kidneys, mainly as the
glucuronide conjugate.
2. Adverse Reactions
e. Hepatic. Serum hepatic transaminase elevations are common, dose related, and
usually self-limiting and benign. Occasionally, laboratory test results indicate increases
in serum bilirubin and abnormal changes in other liver function tests. These results may
reflect potentially serious hepatotoxicity. Serious or fatal hepatotoxicity may be pre-
ceded by nonspecific symptoms such as malaise, weakness, lethargy, facial edema,
anorexia, and vomiting. Patients on multiple anticonvulsants, children, those with
congenital metabolic disorders, and those with organic brain disease may be at particu-
lar risk. Fatal hepatotoxicity is extremely rare and usually develops within the first 6
months of valproate therapy. Valproate should be discontinued immediately in the
presence of suspected or apparent significant hepatic dysfunction or if hyperam-
monemia occurs, with or without lethargy or coma.
f. Other. Patients on valproate therapy have also reported irregular menses, sec-
ondary amenorrhea, breast enlargement, galactorrhea, parotid gland swelling, abnormal
thyroid function tests, acute pancreatitis (including rare fatalities), hyponatremia, inap-
propriate ADH secretion, edema of the extremities, and enuresis. Studies of women
taking valproate for epilepsy show an increase in reproductive endocrine disorders,
especially in those who began taking valproate before age 20 years. 4
Drug Interactions
Valproate is metabolized in part by the liver's cytochrome P4502D6 isoenzyme
system, but this is a relatively minor secondary metabolic pathway compared to glu-
curonidation and [3-oxidation. Drugs that inhibit the cytochrome P450 isoenzymes,
therefore, may have little effect on valproate clearance. Drugs that induce hepatic
enzymes, on the other hand, particularly those that elevate levels of glucuronosyl
transferases, may increase the clearance of val pro ate and thus decrease plasma levels.
For example, phenytoin, carbamazepine, and phenobarbital (or primidone) can double
the clearance of valproate, thereby diminishing its efficacy. Serum valproate concentra-
tions should be monitored as a guide to dosage when valproate is prescribed with
carbamazepine or any other enzyme-inducing drug, especially when one of these
medications is added to or withdrawn from an existing regimen.
Although valproate does not induce hepatic metabolism in general, it does appear
to inhibit drug oxidation and may increase serum levels of concomitantly administered
drugs that are oxidatively metabolized, such as phenobarbital, phenytoin, and tricyclic
antidepressants. See Table 5 for a list of other drugs that interact with valproate.
122 HEATHER S. HOPKINS AND ALAN J. GELENBERG, M.D.
4. Overdose
Overdosage with valproate may result in a deep coma. Since valproic acid is
absorbed very rapidly, gastric lavage or emesis will be more effective the sooner after
ingestion it is used. General supportive measures should be applied with particular
attention to maintaining adequate urinary output. Naloxone has been reported to re-
verse the eNS depressant effects of valproate overdosage.
3. BIPOLAR DISORDER 123
5. Pregnancy
Valproate should not be used in pregnant women. It crosses the placenta and has
been reported to cause teratogenic effects, including neural tube defects, in the fetus.
The risk of spina bifida in children born to women who took valproate during the first
trimester of pregnancy is 1-2%, and other congenital anomalies have also been report-
ed. When valproate is taken later in pregnancy, it may affect the development of the
central nervous system. Even though these data are from patients with epilepsy rather
than bipolar disorder, caution dictates the same avoidance. Valproate also should be
avoided in nursing mothers, as it is excreted in breast milk in concentrations of 1-10%
of the maternal serum concentration.
C. Carbamazepine
Carbamazepine is a dibenzazepine derivative structurally related to the tricyclic
antidepressants. Although some double-blind studies have suggested that carbamaz-
epine is effective in treating acute mania, the methods in most of these studies have
been criticized. Large, carefully designed, prospective, controlled trials would be
needed to establish carbamazepine's efficacy and safety, its specific indications, and its
range of side effects. There is also some evidence for carbamazepine's efficacy as a
long-term treatment for preventing recurrences in bipolar disorder, but, again, better
studies are needed to elucidate its role for this indication.
The exact mechanism by which carbamazepine might alleviate mania in bipolar
disorder is unknown. Its antimanic properties could be related to its anticonvulsant or
antineuralgic effects or to its effects on one of the neurotransmitter or second-messen-
ger systems. Carbamazepine increases acetylcholine in the striatum, decreases dopa-
mine turnover, decreases the release of norepinephrine and cerebrospinal nor-
epinephrine, decreases the activity of adenylate and guanylate cyclase, and decreases
GABA turnover; any or all of these actions could be relevant to carbamazepine's
effects in mania.
1. Pharmacokinetics
The pharmacokinetics of carbamazepine are summarized in Table 1. Carbamaz-
epine is almost completely absorbed from the gastrointestinal tract, but its absorption is
slow and variable. Protein binding is moderate: 55-59% in children and 76% in adults.
Peak plasma levels occur within 4-6 hr after ingestion of tablets and within 1.5 hr
following chronic administration of the suspension. Bioavailability is estimated at 85%
but may be less when the drug is taken with meals. Carbamazepine's half-life ranges
124 HEATHER S. HOPKINS AND ALAN J. GELENBERG, M.D.
from 25 to 65 hr after an initial single dose. With chronic dosing, it may decrease to 8-
29 hr (average 12-17) because of autoinduction of metabolism (see below). Its active
metabolite, carbamazepine-10,11-epoxide, has a half-life of 5-8 hr. Carbamazepine is
eliminated primarily through the kidneys.
Like valproate, carbamazepine is metabolized by the hepatic P4502D6 isoenzyme.
It causes an induction of the P450 enzymes, resulting, in some patients, in an increase
in the rate of its own metabolism over time as well as that of other drugs metabolized
by the P450 system. This often results in having to raise the dose after 2-4 months of
treatment. Steady state may be attained within 4-5 days or, in patients in whom
clearance increases as a result of autoinduction, steady state may not be achieved until
3-4 weeks after treatment initiation.
2. Adverse Reactions
Adverse reactions to carbamazepine are listed in Table 6. The most common side
effects of carbamazepine in the first few weeks of treatment are drowsiness, dizziness,
ataxia, diplopia, nausea, blurred vision, and fatigue. These tend to diminish over time
or to respond to a temporary reduction in dose. Less common reactions include gastro-
intestinal upset, hyponatremia, and a variety of skin reactions, some of which are
severe enough to require discontinuation of carbamazepine. About 10% of patients
experience transient leukopenia, but, unless infection develops, carbamazepine may be
continued. More serious hematopoietic reactions, including aplastic anemia and
agranulocytosis, are very rare. Gradual buildup of carbamazepine levels and smaller,
more frequent dosing can help to minimize side effects, particularly in patients con-
comitantly taking other drugs such as lithium. The most severe adverse reactions have
been observed in the hematopoietic system, the skin, and the cardiovascular system.
e. Neurological. Sedation, dizziness, and ataxia may occur at the start of car-
bamazepine therapy unless starting doses are low and the dose is raised slowly. Fatigue,
headache, dizziness, ataxia, and diplopia may appear during maintenance therapy in
some patients. There may be little to no correlation between these problems and dose or
126 HEATHER S. HOPKINS AND ALAN J. GELENBERG, M.D.
blood level. Cognitive slowing, perhaps secondary to sedation, may occur in some
patients. The cognitive impairments produced by carbamazepine are usually less than
those produced by the barbiturates or phenytoin.
Other neurological effects that have been associated with carbamazepine treat-
ment include: confusion, blurred vision, visual hallucinations, oculomotor distur-
bances, nystagmus, speech disturbances, abnormal involuntary movements, peripheral
neuritis and paresthesias, depression with agitation, talkativeness, tinnitus, and hyper-
acusis. There have been reports of associated paralysis and other symptoms of cerebral
arterial insufficiency, but the exact relationship of these reactions to carbamazepine has
not been established.
3. Drug Interactions
Carbamazepine induces the cytochrome P4502D6 system, which may increase the
metabolism, decrease serum concentrations, and reduce the elimination half-lives of
medications metabolized by this system when used concurrently. (See Table 7 for a list
of drugs that interact with carbamazepine.) Valproate may prolong the half-life and
reduce the protein binding of carbamazepine as well as increase the concentration of its
active 1O,1l-epoxide metabolite. Carbamazepine may increase levels of valproate's
hepatotoxic metabolite, 2-propyl-4-pentanoic acid.
The primary step in carbamazepine's metabolism may be via the P4503A4 sys-
tem, so carbamazepine levels may be increased when carbamazepine is used concur-
rently with 3A4 inhibitors, such as danazol (Danocrine® and others), diltiazem (Cardi-
zem® and others), verapamil (Calan® and others), or the influenza virus vaccine.
Carbamazepine toxicity may be delayed for several weeks after initiation of danazol
therapy and 1-2 weeks after the influenza virus vaccination. It is recommended that
nifedipine (Adalat® and others) be used as an alternative to verapamil or diltiazem.
Carbamazepine also increases the metabolism of estrogen-containing oral contra-
ceptives, so women taking these medications may need higher doses to prevent break-
through bleeding and to obtain adequate protection. * Parenteral medroxyprogesterone
or nonhormonal methods of birth control may be considered as alternatives.
Chronic use of a hepatic enzyme-inducing agent such as carbamazepine prior to
anesthesia may increase the metabolism of enflurane (Ethrane®), halothane (Fluo-
thane®), or methoxyflurane (Penthrane®), leading to an increased risk of hepatotoxi-
city. It may also increase the formation of nephrotoxic metabolites of methoxyflurane,
leading to increased risk of nephrotoxicity. In addition, cardiac arrhythmias may occur,
possibly due to sensitization of the myocardium resulting from increased concentra-
tions of norepinephrine.
Carbamazepine has an antidiuretic-hormone-like effect, which may counteract the
opposite effect of lithium (which produces polyuria) when the two drugs are used
together. By itself, carbamazepine has the potential to produce hyponatremia. Concur-
rent use of chlorpropamide (Diabinese®), clofibrate (Atromid-S®), desmopressin,
lypressin, posterior pituitary, thiazide diuretics (when used for their paradoxical anti-
diuretic activity in the treatment of diabetes insipidus), or vasopressin with carbamaze-
pine may potentiate the antidiuretic effect, leading to a lower sodium concentration and
causing adverse effects that include increased seizure activity. Coadministration of
carbamazepine and lithium may increase neurotoxic side effects even at nontoxic blood
concentrations of both drugs.
Carbamazepine can slow atrioventricular cardiac conduction and should be
avoided in patients who have heart block or prolonged QT and administered cautiously
in conjunction with other drugs that also can increase heart block, such as tricyclic
antidepressants and J3-blockers, among others.
The risk of hepatotoxicity with single toxic doses or prolonged use of high doses
*Failure to make this dosage increase in oral contraceptives can lead to an unwanted pregnancy, with the
fetus exposed to a teratogen!
128 HEATHER S. HOPKINS AND ALAN J. GELENBERG, M.D.
4. Overdose
Symptoms of carbamazepine overdose include irregular breathing, respiratory
depression, tachycardia, hypotension or hypertension, shock, conduction disorders,
impairment of consciousness, convulsions (especially in small children), motor rest-
lessness, muscular twitching, tremor, athetoid movements, opisthotonus, ataxia,
drowsiness, dizziness, mydriasis, nystagmus, diadochokinesia, ballism, psychomotor
disturbances, dysmetria, initial hyperreflexia (followed by hyporeflexia), nausea, vom-
iting, anuria or oliguria, and urinary retention. Signs and symptoms begin to appear
after 1-3 hr, with neuromuscular disturbances being the most prominent. When very
high doses of carbamazepine (>60 g) have been ingested, severe cardiac complications
may occur, but cardiovascular problems are generally mild. When an overdose of
carbamazepine is taken in conjunction with alcohol, tricyclic antidepressants, barbitu-
rates, or hydantoins, the signs and symptoms of acute poisoning may be aggravated or
modified.
There is no specific antidote for carbamazepine overdose. The prognosis in cases
of severe poisoning depends on prompt elimination of the drug through induced vomit-
ing' stomach irrigation, or other appropriate measures that diminish absorption. Even
when more than 4 hr have elapsed following ingestion of the drug, the stomach should
be repeatedly irrigated, especially if the patient has consumed alcohol. Activated char-
coal and laxatives can be used to reduce absorption, and forced diuresis may accelerate
elimination. Dialysis is indicated only in severe poisoning associated with renal failure.
Replacement transfusion should be used in severe poisoning in small children.
130 HEATHER S. HOPKINS AND ALAN J. GELENBERG, M.D.
S. Pregnancy
Carbamazepine crosses the placenta and should not be used in pregnant women.
Although adequate and well-controlled studies in humans have not been done, there
have been reports of babies (usually of mothers with epilepsy) prenatally exposed to
carbamazepine having small head circumferences, low birth weights, craniofacial de-
fects, fingernail hypoplasia, developmental delays, and spina bifida. Carbamazepine
also is distributed into breast milk. Concentrations in breast milk and in the plasma of
nursing infants have been reported to reach 60% of the maternal plasma concentration.
Therefore, the possibility exists that carbamazepine may cause adverse effects in a
nursing infant.
D. Antipsychotics
Antipsychotics, particularly haloperidol, can rapidly diminish increased motor
activity and also suppress other symptoms of mania. Lithium, in contrast, has more
specific antimanic effects but may take from several days to several months to control
manic symptoms. In general, lithium is more effective for stabilizing mood and ide-
ation while neuroleptics are superior in controlling hyperactivity. The course of clinical
response to antipsychotics is usually faster than that to lithium. Therefore, based on
research 89 - 92 and clinical experience, anti psychotics are usually used in addition to
lithium (or another antimanic agent) to gain more rapid relief of severe psychotic
symptoms and acute behavioral excitement, typically during an acute manic episode.
Although neuroleptics may have specific antimanic properties in addition to their
sedative and antipsychotic effects, they carry the risk of extrapyramidal symptoms and
may not be as effective over time as other agents. There is also concern about possible
neurotoxicity from the combination of lithium and a neuroleptic. 93 ,94
3. BIPOLAR DISORDER 131
Recent data suggest that the atypical drug clozapine may have particular benefit
for acutely manic and psychotic patients,95 especially those who are treatment resistant
and/or intolerant of other agents. 96 - 98 Clozapine has also been reported to be effective
in a few cases of rapid-cycling bipolar disorder (see Section IV.C.5).99 Although
clozapine is free from most of the extrapyramidal side effects associated with typical
neuroleptics and may not cause tardive movement disorders, it does cause a number of
serious toxic reactions-including agranulocytosis, seizures, and cardiorespiratory
complications (see Chapter 4). Whether clozapine is superior to traditional neuroleptics
in the treatment of bipolar disorder remains to be seen. There is also preliminary
evidence 100. 10 I that another atypical neuroleptic, risperidone, may be beneficial in the
treatment of manic patients, but as several reports lO2 - 104 have found an association
between risperidone and new or increased manic symptoms, the jury is still out as to its
effectiveness in this condition.
Several studies have evaluated the use of depot antipsychotics for maintenance
treatment of bipolar disorder, with mixed results. In general, depot antipsychotics-
including flupenthixol, fluphenazine, haloperidol, pipothiazine, and zuclopenthixol-
reduce manic but not depressive episodes,l°5-107 This seems to be the case whether
depot antipsychotics are used to augment lithium or replace it, although some studies
have found no benefit to depot antipsychotic augmentation,l°8 One study reports a
decrease in relapses and hospitalization for manic, depressive, and mixed episodes
during treatment with depot antipsychotics,109 whereas another found a decrease in
manic episodes but an increase in depressive episodes in patients nonresponsive to or
intolerant of lithium. I 10
Prescription surveys have shown that many bipolar patients are maintained rou-
tinely and for long periods on antipsychotics in addition to lithium or other mood
stabilizers. I I I Reported rates for any antipsychotic use during treatment for bipolar
disorder range from 75% to lOO%, with 40-95% of patients continuing to take an
antipsychotic during maintenance treatment,l12-114 Beyond doubt, antipsychotic drugs
have a role to play in augmenting other antimanic agents during the acute treatment of
excited or psychotic mania, but their efficacy in maintenance treatment has never been
conclusively demonstrated. In addition, the risk of tardive movement disorders with
long-term antipsychotic use is 20-30%; the risk is probably even higher in bipolar
patients. On the other hand, there is a strong feeling in the field that some patients need
antipsychotic drug treatment, along with (or in some cases instead of) mood-stabilizing
agents to remain free of recurrent episodes of mania. To what extent some patients may
be unnecessarily exposed to long-term treatment with these agents is unknown. For the
clinician, the current state of knowledge translates into: Use maintenance antipsychotic
treatment for a bipolar patient if it really helps and the patient gives informed consent.
Document both of these facts.
E. Antidepressants
If mild, a depressive episode in a bipolar patient may be "ridden out" with
continuing lithium or other maintenance therapy and psychological support. More
severe depressions, however, require biological treatment. Early studies found com-
132 HEATHER S. HOPKINS AND ALAN J. GELENBERG, M.D.
F. Other
The anticonvulsant drug lamotrigine (Lamictal®) has not been studied systemat-
ically in patients with bipolar disorder, but clinicians have had some success with this
agent in individual cases and one open trial. 125 Lamotrigine may have antidepressant as
well as mood-stabilizing properties. It appears to prevent antidepressant-induced cy-
cling and to be beneficial for patients with rapid-cycling bipolar disorder or mixed
states. 126 There have been several reports of serious adverse reactions associated with
lamotrigine, including multiple organ failure,127 encephalopathy,128 Stevens-Johnson
syndrome, 129 fatal toxic epidermal necrolysis, 130 and fulminant hepatic failure. 13 I Less
3. BIPOLAR DISORDER 133
serious side effects include dizziness, headaches, diplopia, ataxia, nausea, vomiting,
somnolence, asthenia, and rash. Rash is the side effect that most often causes the drug
to be discontinued and is more likely to develop when initiallamotrigine doses are high
or when lamotrigine is started too rapidly in someone taking valproate. When
lamotrigine is introduced gradually, the risk of rash does not exceed that with car-
bamazepine. In adults, the initial dose of lamotrigine is usually 25 mg b.i.d. with
increases of 25 or 50 mg/day every week or two up to 50-100 mg b.i.d. In patients
taking carbamazepine, larger initial doses and more rapid dose increases are possible,
because carbamazepine-induced enzymes facilitate the metabolism of lamotrigine,
lowering serum lamotrigine levels. Phenobarbital and primidone also lower plasma
lamotrigine levels (by about 40%). As mentioned earlier, valproate increases serum
levels of lamotrigine, and a dosage reduction should be considered if the two medica-
tions are used concomitantly.132 In patients taking valproate, the initial dose of
lamotrigine is often 12.5 mg/day, and the drug is increased by 12.5 or 25 mg/day every
2 weeks.
There have been anecdotal reports of another new anticonvulsant, gabapentin
(Neurontin), being used to treat mania, but no data are yet available on its safety and
efficacy. It seems to have a more benign side-effect profile than lamotrigine, with fewer
reports of serious adverse reactions.
A few, small, double-blind trials have found the calcium channel blocker ve-
rapamil to be equal in efficacy to lithium 133 ,134 and superior to placebo135 in patients
with acute mania. Since there have been reports of neurotoxicity and bradycardia from
the combination of lithium and verapamil, these two medications should be combined
only with great caution. Because verapamil raises carbamazepine serum levels, this
combination should be monitored carefully.
A. Diagnostic Evaluation
Barring contraindications to its use, lithium should be the first agent tried for the
treatment of classic bipolar disorder, although valproate is also an acceptable first-line
alternative for acute mania. Before employing either of these agents for an acute
episode, however, the clinician should have a working diagnosis compatible with this
therapy. As noted earlier, the diagnoses of bipolar, manic, and schizoaffective disorder
clearly justify lithium treatment during the acute phase, whether lithium is used alone
or in conjunction with a neuroleptic. At times, the physician may prescribe lithium on a
trial-and-error basis for another psychiatric indication, usually based on the patient's
insufficient response to another treatment. At other times, the overlapping symptoms of
several possible psychiatric diagnoses-for example, acute mania versus schizo-
phrenia-will lead the physician to try lithium on an empirical basis.
In addition to making a primary diagnosis, the practitioner should become familiar
with factors that help predict outcome. He or she also should assess current stressors,
precipitating events, ego strengths, environmental resources, and psychodynamic, char-
acterologic, and interpersonal issues. Family and friends can be educated and serve as
134 HEATHER S. HOPKINS AND ALAN J. GELENBERG, M.D.
allies. A careful history of the patient's illness over a lifetime is important. So too are
medical and neurologic histories and evidence of current or past substance abuse. The
development of an initial rapport that contributes to a growing therapeutic relationship
occurs simultaneously with this information-gathering process.
If the diagnosis of bipolar disorder is unclear, the clinician might justifiably
hesitate before recommending long-term maintenance lithium therapy. An acute epi-
sode with manic features could also be a brief reactive psychosis, a drug-precipitated
reaction, the beginning of schizophrenic illness, or a manifestation of a medical or
neurologic disorder.136 Especially when this is a first episode in a relatively young
patient without a family history of mood disorders, the practitioner may elect to follow
the patient carefully over time to chart out the course of future episodes, if any.
However, since recurrences cause a worsening offunction between episodes l2 and may
make a patient less likely to respond to medication, 17 intervention needs to be early and
aggressive. Therefore, contacts should be regular; psychosocial, medical, and neuro-
logic data gathering should be open and ongoing; and the physician should be available
between visits, in case intercurrent symptoms arise.
The diagnosis and treatment of bipolar disorder is often complicated by comorbid
substance abuse. While substance use disorders are prevalent among patients with all
psychiatric illnesses, bipolar disorder appears to be the most likely to co-occur with
alcohol or drug abuse.137 In fact, only 39% of bipolar patients do not have comorbid
substance abuse at some time during their lives. 138 Bipolar patients with substance use
disorders tend to have an earlier onset and worse course of illness compared to bipolar
patients who do not abuse drugs or alcohol, and many have a poor response to lithium
treatment. Preliminary data suggest that valproate may be effective in the acute treat-
ment of bipolar episodes complicated by concomitant substance dependence.139
When a bipolar diagnosis is clear-cut, maintenance lithium therapy might be
considered from the very first episode, especially if it is manic and severe-even more
in the presence of a family history of bipolar disorder. Although the course of episodes
in bipolar disorder varies from patient to patient, episodes tend to become more
frequent and severe over time, recurrences are the rule, and it is a lifetime disorder.
Furthermore, each episode carries a risk of mortality and a substantial likelihood of
psychosocial and economic morbidity. Worse, there is some suggestion that the occur-
rence of each episode might actually enhance the risk of future episodes, analogous to
the "kindling" model of epileptic seizures. All of these facts and hypotheses add up to a
strong argument for early maintenance treatment in the course of a bipolar disorder,
once the diagnosis is apparent and other possible causes (e.g., medical, neurologic,
substance abuse) are ruled out. Few clinicians would not consider maintenance treat-
ment after three episodes.
Factors Predicting Lithium Responsiveness 140,141
1. Positive response
a. Family history of mania or depression
b. Prior positive response to lithium
c. Positive response of family member to lithium
d. Mild and uncomplicated illness
3. BIPOLAR DISORDER 135
2. Negative response
a. Rapid cycling (four or more episodes/year)
b. Mixed states or dysphoric mania
c. History of alcohol or drug abuse
d. Noncompliance with previous maintenance treatment
e. Cycle pattern of depression-mania-euthymia
f. Personality disturbance
g. History of poor interepisodic functioning
h. Poor social support system
i. Three or more prior episodes
B. Medical Evaluation
Various medical evaluations should be performed before beginning treatment with
any of the mood-stabilizing drugs used for the treatment of bipolar disorder (see Table
8). Before starting lithium therapy, the clinician should complete a medical history,
including a review of organ systems and a physical examination. The patient should be
screened for pregnancy, thyroid disease, epilepsy, renal disease, cardiovascular disease,
and evidence of brain damage. If the patient has any of these conditions, lithium should
be administered more cautiously. Appropriate laboratory screening prior to lithium
therapy should include a complete blood count; an EKG; urinalysis, BUN, and electro-
lytes; serum creatinine; and serum TSH, T4, T3RU (or other measure of binding
globulin), and free T4 (measured or calculated).
Clinicians should intermittently evaluate thyroid and renal functioning dur-
ing lithium therapy. A consensus about appropriate tests and their timing does not
exist. However, we recommend a serum creatinine determination every 6 months and a
serum TSH determination at 6 months and then annually thereafter. Beware of gradual
creatinine increases, even within the normal range, as these may indicate a gradual
deterioration in glomerular filtration. Similarly, a gradual increment in lithium level
despite a constant dose may reflect a similar renal deterioration. If TSH is elevated, a
full thyroid panel is indicated, and, together with an internist, the psychiatrist may
consider thyroid supplementation. We hasten to emphasize that, despite periodic labo-
ratory screening, the clinical index of suspicion must remain foremost as a means to
detect treatment-emergent medical conditions. Routinely questioning the patient about
physical symptoms, along with psychiatric and other mood changes, may prompt
targeted laboratory tests outside of the schedule we recommend. Any patient-and
particularly those taking maintenance medications-deserves appropriate annual
physical examinations, which, depending on a patient's age, may be supplemented with
an EKG or additional laboratory testing.
Before initiating valproate, the physician should obtain a comprehensive medical
history and perform a physical examination, paying particular attention to suggestions
of liver disease or bleeding abnormalities. Liver and hematologic functions should be
measured before treatment, every 1 to 4 weeks for the first 6 months, and then every 3
to 6 months. Serum ammonia concentrations should be monitored during treatment,
and bleeding time determinations, blood cell counts, and renal function determinations
are recommended prior to treatment and periodically during therapy. Patients should be
screened for pregnancy prior to treatment with valproate, as it is teratogenic.
Before carbamazepine is started, baseline blood and platelet counts, urinalysis,
and hepatic and renal function tests are in order. Although earlier guidelines called for
routine monitoring of some or all of these indices, and some physicians still perform
blood counts once or twice during the first few months of treatment and when plasma
concentrations are drawn, a more general consensus at present is to instruct patients
and family members to contact the doctor immediately if petechiae, pallor, weakness,
fever, or infection occurs. At that time, the physician should order relevant tests. As
with valproate, women of childbearing potential should be tested for pregnancy prior to
treatment with carbarnazepine, another known teratogen.
A first onset of mania after age 40 should prompt a thorough medical and neuro-
logic differential diagnosis. Primary mania seldom appears de novo in midlife or later,
but a wide range of brain insults can precipitate secondary mania, often indistinguish-
able by behavior alone from a primary mood disorder. At a minimum, the physician
should ensure a thorough physical and neurologic examination and routine laboratory
and possibly neuropsychological testing. An electroencephalogram (EEG) and head
computerized tomography (CT) or magnetic resonance imaging (MRI) should be con-
sidered. When a specific and treatable condition is found, it should be addressed
specifically, and the manic syndrome should be treated on a symptomatic basis: often
antipsychotic drugs will be satisfactory. If, however, there is no specific treatment for
the underlying organic condition, a psychiatrist may treat the manic symptoms much as
he or she would primary mania, i.e., with lithium, valproate, neuroleptics, and, if
necessary, other alternatives. Valproate might be superior to lithium for patients whose
mania is secondary to a neurologic condition.
C. Treatment
The therapeutic relationship provides a foundation for all drug therapies. In acute
situations, a caretaker may use his or her position to establish quick rapport and, when
3. BIPOLAR DISORDER 137
necessary, to confront or set limits. His or her initial approach should include various
crisis intervention techniques, such as mobilization of environmental resources. In-
volvement and education of persons close to the patient generally increases medication
compliance. In addition, maintenance therapy requires the formation of a trusting
collaboration between clinician and patient. With many patients, open discussions
about the illness set the tone for the ongoing dialogue. Most individuals have some
trouble accepting their illness and its possible ramifications. Therefore, the clinician
should try to understand how mood episodes affect the patient's life, exploring how
each episode has interfered with the patient's relationships and major roles.
During the early phase of treatment, patients with bipolar illness most often want
to know:
1. Where did I get it?
2. Is it transmitted to my children?
3. Is my problem permanent?
4. What is the treatment?
5. What are the risks of drug therapy?
6. How effective is therapy?
These queries frequently encompass more fundamental, but less obvious, concerns
about control, dependency, personal defectiveness, and "lovableness" in personal rela-
tionships. Therefore, the clinician should try to clarify both the explicit and implicit
concerns and requests before providing answers that might close off continued discus-
sion. No matter what course is taken, however, discussion of the potentially inherited
nature of bipolar illness and its possible course should be coupled closely with explana-
tions about the effectiveness of lithium maintenance and the importance of the thera-
peutic relationship. Establishing an open relationship may take many months. The
clinician should encourage the patient to become an active partner in maintenance
therapy. For example, the patient and family members should be sensitized to emerging
hypomanic and depressive symptoms. The clinician, patient, and family together
should delineate a list of symptoms that are early manifestations of the patient's
affective illness and require immediate clinical contact (such as sleeplessness, spending
extra money, or anorexia). Also, the patient and family should be enlisted to detect
medical conditions that might affect the lithium level and signs and symptoms of
toxicity.
The discussion of these aspects of treatment often stimulates other questions,
including:
1. How long will treatment last?
2. How will we know when to stop lithium?
3. What are the risks of treatment?
4. How will I feel on the medicine?
5. Does lithium interact with other drugs?
Again, these questions should be answered only after determining what they mean to
the patient and how he or she will use the information. In addition, the clinician has a
human and legal responsibility to discuss adverse reactions (see Tables 2, 4, and 6),
potential toxicity, alternative treatments, limitations of maintenance therapy, and long-
138 HEATHER S. HOPKINS AND ALAN J. GELENBERG, M.D.
tenn prognosis. Although medication-free periods may be tried, there is a real risk of
recurrences, and no one seems to "lose" a mood disorder over time; if anything, the
disorder worsens.
achieve the desired blood level but to be observant during the postmanic phase for
signs of toxicity that might accompany a rise of blood lithium level into a toxic range.
After the first week, lithium levels should be determined weekly for a few weeks
and then about once a ,month. Once steady state has been achieved, the frequency of
dosing may be reduced to just once or at most twice daily, even with standard lithium
carbonate preparations. When the manic episode resolves, the clinician must make a
decision about maintenance therapy.
For a patient with acute mania who cannot tolerate lithium or does not respond to
lithium despite adequate levels and appropriate neuroleptic therapy, valproate should
be substituted for lithium. If the patient responds partially to lithium therapy, valproate
may be added to lithium therapy. In each case, carbamazepine should be tried instead of
valproate if valproate also appears to be inefficacious or intolerable to the patient.
When these medications are combined, they may be effective at lower dosages than
those required in monotherapy.
The typical starting dose of valproate for healthy adults is 750 mg daily in divided
doses. (Equivalent oral doses of divalproex sodium and valproic acid capsules deliver
systemically equivalent quantities of the valproate ion.) If helpful in selected cases,
valproate is available as "sprinkles," which can be mixed with food. The dose should
be increased as rapidly as possible to achieve the lowest therapeutic dose that produces
the desired clinical effect or a l2-hr serum valproate concentration between 50 and 125
IJ.g/mL. As serum levels rise, the likelihood of efficacy increases, but so do side effects.
The time of dosing is determined by possible side effects, and, if tolerated, a once-
a-day dosing schedule can be employed. As with lithium, the antimanic response to
valproate typically occurs after 1-2 weeks, although some experts think valproate
works several days faster than lithium. A still faster response may be achieved by oral
loading. Patients given divalproex sodium at a dosage of 20 mg/kg per day from day 1
often attain serum concentrations greater than 50 mg/L by the second day. The maxi-
mum recommended dosage is 60 mg/kg per day. The rapid onset of antimanic and
antipsychotic response to this method may be comparable to that of haloperidol. 146
Geriatric patients taking valproate tend to have elevated free, unbound valproic
acid concentrations and lowered intrinsic clearances. This indicates a decrease in
valproate-metabolizing capacity and a fall in serum albumin. Therefore, these patients
should receive a lower daily dosage, and the serum concentrations should be kept in the
lower therapeutic range.
For carbamazepine used as a monotherapy, the typical starting dose is 200 to 400
mg/day in three to four divided doses, increased to 800-1000 mg/day by the end of the
first week. If clinical improvement is insufficient by the end of the second week, and
the patient has not developed intolerable side effects to the drug, increases to as high as
1600 mg daily may be considered. Although a correlation between blood level and
clinical response has yet to be established, most clinicians are guided by the antiepilep-
sy range: usually 4-15 ng/mL. If carbamazepine is combined with lithium or neurolep-
tics, physicians often prefer to use lower doses and blood levels. If valproate and
carbamazepine are administered simultaneously, blood levels of each should be moni-
tored carefully because of complex interactions between the two agents. When the dose
of carbamazepine is built up rapidly, side effects are more likely.
3. BIPOLAR DISORDER 141
2. Hypomania
Clinicians often can begin lithium treatment in hypomanic patients without hospi-
talization. Lithium is relatively easy to start once the patient has been medically
cleared. Give 300 mg of lithium carbonate 2 or 3 times a day for 1 week and then draw
a serum level. Inform the patient about the blood-drawing procedure and instruct him
or her to allow 12 hr between the last dose and the serum measurement. For each
dosage change, a steady-state level is achieved after about 1 week. After drawing the
first serum level, the clinician can usually approximate a maintenance dose, which can
then be tried and tested with the next blood level. In maintenance therapy, 0.8-1.0
mEq/L may be the most effective range. When side effects are intolerable, reduced
plasma concentrations of lithium may diminish them. However, reduced levels de-
crease protection against relapse and recurrence, particularly for mania. 17 Few, if any,
patients will experience prophylactic benefit from lithium at blood levels below 0.4
mEq/L.
During the first week of therapy, the administration of lithium has few serious
risks in healthy patients because of rapid clearance and low total dosage. Therefore,
telephone contact or short office visits usually suffice unless the patient requires
additional support and reassurance. Gastrointestinal symptoms are common during
initial treatment and during dosage increases. If these symptoms are severe, however,
smaller doses or a different preparation may help. If the dose is changed, another serum
level should be obtained a week later. Once the desired level is reached, monthly
checks are adequate. The clinician should remember that patients might retain more
lithium when they are hypomanic or manic than when they are euthymic. Therefore,
before the number of serum level checks is diminished, the patient's mood should be
stable. For fully stabilized patients, the clinician can check lithium levels as infre-
quently as every 3 months. In any patient taking lithium, and particularly in those
stopping medication, the caretaker should educate the patient and relatives about the
signs of incipient mania. Some patients may want to carry a wallet card or an ID
necklace, bracelet, or anklet indicating their condition and medication.
Lithium can provide dramatic relief for both manic and hypomanic symptoms.
However, a maintenance regimen for the prevention of future episodes may require the
administration of lithium for 12-18 months before the clinician can fully determine its
effectiveness.
Antipsychotics are seldom needed in treating hypomania, but a benzodiazepine at
bedtime often promotes sleep. Although the efficacy of valproate and carbamazepine in
hypomania is incompletely studied, we may assume efficacy for valproate and possibly
carbamazepine, which can be used in a manner similar to what we have described
under mania.
4. Maintenance Therapy
During mania or depression, a patient may not fully understand the illness and
circumstances or have insight and judgment about treatment options. When mainte-
nance therapy is contemplated, in contrast, the patient should be competent to give
informed consent and must be an active partner in treatment. In fact, only the patient
can control many of the vital aspects of therapy. Therefore, the clinician should openly
discuss with the patient the nature of the mood illness and the clinical effects and
pharmacokinetics of lithium or other medications to be employed. He or she should
also inform the patient about possible adverse reactions, normal precautions, and the
importance of monitoring the serum drug levels. For example, a patient taking lithium
should be advised to maintain a stable salt intake and contact the physician when there
are significant changes in salt balance [e.g., after marked sweating, diarrhea from an
illness, institution of a low-sodium diet, or when another medication is coadministered
(particularly diuretics)]. The patient also should be told about common adverse effects:
hand tremor, gastrointestinal upset, polydipsia and polyuria, and weight gain. Patients
should know that changes in caffeine intake may alter lithium levels. In addition,
patients should be instructed to stop lithium and contact the physician if any of these
symptoms persist or get worse or if dizziness, drowsiness, slurred speech, or ataxia
appears. Patients taking other mood stabilizers need analogous information about their
3. BIPOLAR DISORDER 143
5. Rapid Cyclers
Patients with rapid-cycling bipolar disorder-defined as four or more affective
episodes in one year, with or without an intervening period of euthymia-tend to be
less responsive to lithium treatment. lO For many patients, rapid cycling is a phase in
their lifelong illness. Whether rapid cycling is a natural progression of the illness or a
separate disorder has yet to be determined. 122 The onset of rapid cycling has been
associated with antidepressant drugs (especially tricyclic antidepressants) and hypo-
thyroidism. 150,151 Some people also experience ultrarapid cycling, switching between
moods over days or even hours.
Rapid cycling tends to be more prevalent in women, who make up approximately
75% of this population. 4 Premenstrual syndrome (PMS) has been related to major
depressive disorder, and one study found that 15 (60%) of 25 female rapid-cycling
patients had severe PMS symptoms as opposed to 5 (20%) of 25 control subjects. 152 In
144 HEATHER S. HOPKINS AND ALAN J. GELENBERG, M.D.
addition, bipolar patients with severe PMS had more affective episodes in one year than
those with moderate or mild PMS symptoms.
Various therapeutic approaches have been investigated for treating patients with
rapid-cycling bipolar disorder. Although at this time the scientific literature does not
point to one standard protocol, experts rate valproate as the treatment of choice for a
patient with acute mania and rapid cycling.147 Carbamazepine is considered an alter-
nate first-line treatment, whereas lithium, alone or in combination with other drugs
(such as choline 153 ), is considered a second-line treatment. Before beginning any
medication for a rapid-cycling patient, the clinician should obtain a careful history,
searching for possible provocative factors, such as a precipitating effect of antidepres-
sants or the presence of endocrinopathies, other medications, seasonal factors, psycho-
social stressors, and so on. If there is not a specific provocation that can be removed or
remedied, the clinician should attempt to stabilize the patient pharmacologically.
Patients with rapid-cycling bipolar II disorder may require lower doses of valpro-
ate than those currently recommended for bipolar I disorder: 125-500 mg/day, corre-
sponding to a mean serum valproate level of 32.5 j.Lg/mL, was effective in 69% of
patients in one trial. 154 Clozapine and nimodipine may be effective in patients with
rapid-cycling bipolar disorder nonresponsive to conventional treatments. 99 ,155,156
There are also suggestions that sleep deprivation may help alleviate depressive epi-
sodes in rapid-cycling patients, but it also may precipitate mania. 157,158
E. Schizoaffective Disorder
The boundaries between schizophrenia and schizoaffective disorder and between
the latter and bipolar disorder are blurry at best and nonexistent at worst. Most proba-
bly, the schizoaffective label includes patients from both the schizophrenic and mood-
disorder categories, as well as others who might have a distinct condition. Since there
are no valid predictors-clinical or biological-of pharmacological response, the best
strategy is empirical. With lithium as a less hazardous option than neuroleptics for
long-term therapy (no tardive dyskinesia), patients with a suggestion of mood disorder
should be tried on it first. In some cases, the clinician may conclude that combining
lithium with an antipsychotic drug is the maintenance treatment of choice, but because
of the additive risks and adverse effects, the clinical record should clearly justify this
polypharmacy. As clinical experience increases, roles for valproate and other mood
stabilizers and for atypical neuroleptics should become clearer.
3. BIPOLAR DISORDER 145
v. CONCLUSION
Although less common than recurrent major depression, bipolar illness is impres-
sive in its manifestations, more frequently requires hospitalization, and carries consid-
erable hazards in morbidity and mortality, economic and social stability, and risk of
substance abuse. Its response to biological treatment, however, is equally impressive,
and for many patients with this psychiatric condition, lithium is little short of a god-
send. As noted earlier, though, for patients who cannot tolerate or do not benefit fully
from lithium, valproate has been found to be equally effective in treating acute mania.
Perhaps it will tum out to be a good treatment for maintenance treatment also. Car-
bamazepine and clozapine may also be effective for some patients.
All patients with bipolar disorder require a stable relationship with the treating
physician. This is no different from any other chronic medical illness. Many patients
also require some form of psychosocial intervention to enjoy maximum mood stability
over the lifetime course of this condition. Approached sensitively and knowledgeably,
treating patients with bipolar disorder can be both successful and gratifying.
REFERENCES
1. Kessler R C., McGonagle K A., Zhao S., et al: Lifetime and 12-month prevalence of DSM-III-R
psychiatric disorders in the United States. Arch Gen Psychiatry 51:8-19, 1994.
2. Muller-Oerlinghausen B., Ahrens B., Grof E., et al: The effect of long-term lithium treatment on the
mortality of patients with manic-depressive and schizoaffective illness. Acta Psychiatr Scand 86:218-
222, 1992.
3. Weissman M. M., Bland R c., Canino G. 1., et al: Cross-national epidemiology of major depression
and bipolar disorder. JAMA 276:293-299, 1996.
4. Leibenluft E.: Women with bipolar illness: Clinical and research issues. Am J Psychiatry 153:163-173,
1996.
5. Krasa N. R.: Sex differences in age at first affective episode. Am J Psychiatry 152:960, 1995.
6. McMahon E 1., Stine O. C., Chase G. A., Meyers D. A., Simpson S. G., DePaulo J. R., Jr.: Influence of
clinical subtype, sex, and lineality on age at onset of major affective disorder in a family sample. Am J
Psychiatry 151:210-215, 1994.
7. Goodwin E K, Jamison K R: Manic-Depressive Illness. New York, Oxford University Press, 1990.
8. American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders, Washing-
ton, D.C., American Psychiatric Association, 1994.
9. Calabrese J. R, Woyshville M. 1., Kimmel S. E., Rapport D. J.: Mixed states and bipolar rapid cycling
and their treatment with divalproex sodium. Psychiatr Ann 23:70-78, 1993.
10. Dunner D. L., Fieve R R.: Clinical factors in lithium carbonate prophylaxis failure. Arch Gen Psychia-
try 30:229-233, 1974.
11. Keller M. B., Lavori P. W., Coryell W., et al: Differential outcome of pure manic, mixed/cycling, and
pure depressive episodes in patients with bipolar illness. JAMA 255:3138-3142, 1986.
12. Prien R. E, Gelenberg A. J.: Alternatives to lithium for preventive treatment of bipolar disorder. Am J
Psychiatry 146:840-848, 1989.
13. Cohen L. S., Sichel D. A., Robertson L. M., Heckscher E., Rosenbaum J. E: Postpartum prophylaxis
for women with bipolar disorder. Am J Psychiatry 152:1641-1645, 1995.
14. Kendell R E., Chalmers J. C., Platz C.: Epidemiology of puerperal psychoses. Br J Psychiatry
150:662-673, 1987.
15. Winokur G., Clayton P. J., Reich T.: Manic Depressive Disease. SI. Louis, C.V. Mosby, 1969.
16. Post R. M.: Transduction of psychosocial stress into the neurobiology of recurrent affective disorder.
Am J Psychiatry 149:999-1010, 1992.
146 HEATHER S. HOPKINS AND ALAN J. GELENBERG, M.D.
17. Ge1enberg A. 1., Kane 1. M., Keller M. B., et al: Comparison of standard and low serum levels of
lithium for maintenance treatment of bipolar disorder. N Engl J Med 321:1489-1493, 1989.
18. Altshuler L. L.: Bipolar disorder: Are repeated episodes associated with neuroanatomic and cognitive
changes? Bioi Psychiatry 33:563-565, 1993.
19. Friedman M., Culver C., Ferrell R.: On the safety oflong-term treatment with lithium. Am J Psychiatry
134:1123-1126, 1977.
20. Savard R. J., Rey A. C., Post R. M.: Halstead-Reitan category test in bipolar and unipolar affective
disorders: Relationship to age and phase of illness. J Nerv Ment Dis 168:297-304, 1980.
21. Hauser P., Altshuler L. L., Berrettini W., Dauphinais I. D., Gelernter J., Post R. M.: Temporal lobe
measurement in primary affective disorder by magnetic resonance imaging. J Neuropsychol 1:128-
134, 1989.
22. Altshuler L. L., Conrad A., Hauser P., Li X., Guze 8., Post R. M.: Temporal lobe volume in bipolar
disorder: A preliminary report. Arch Gen Psychiatry 48:482-483, 1991.
23. Coffman, J. A., Bornstein R. A., Olson S. C., Schwarzkopf S. B., Nasrallah H.: Cognitive impairment
and cerebral structure by MRI in bipolar disorder. Bioi. Psychiatry.
24. Swayze V. w., II, Andreasen N. C., Alliger R. J., Yuh W. T. C., Ehrhardt J. C.: Subcortical and temporal
structures in affective disorder and schizophrenia: A magnetic resonance imaging study. Bioi Psychia-
try 31:221-240, 1992.
25. Dolan R. J., Calloway P., Mann A. H.: Cerebral ventricular size in depressed subjects. Psychol Med
15:873, 1985.
26. Andreasen N. C., Swayze V. w., II, Aaum M., Alligen R., Cohen G.: Ventricular abnormalities in
affective disorder: Clinical and demographic correlates. Am J Psychiatry 147:893-900, 1990.
27. McInnis M. G., McMahon F. J., Chase G. A., Simpson S. G., Ross C. A., DePaulo J. R. J.: Anticipation
in bipolar affective disorder. Am J Hum Genet 53:385-390, 1993.
28. Clift E.: Data Consortium to search for genes causing manic-depressive illness. Psychiatric Times
1993;10:1-3.
29. Weissman M. M., Leaf P. 1., Livingston B. M., et al: The epidemiology of dysthymia in the community:
Rates, risks, comorbidity and treatment, in 140th Annual Meeting of the American Psychiatric Associa-
tion. Chicago, American Psychiatric Association, 1987.
30. Anonymous: Medical Practice Project: A State-of the-Science Report for the Office of the Assistant
Secretary for the U.S. Department of Health, Education and Welfare, 1979 (Abstract).
31. Mukherjee S., Sackeim H. A., Schnur D. B.: Electroconvulsive therapy of acute manic episodes: A
review of 50 years' experience. Am J Psychiatry 151:169-176, 1994.
32. Small J. G., Klapper M. H., Kellams 1. J., et aI: Electroconvulsive treatment compared with lithium in
the management of manic states. Arch Gen Psychiatry 45:727-732, 1988.
33. Coffey C. E., Winer R. D.: Electroconvulsive therapy: An update. Hosp Community Psychiatry
41:515-521, 1990.
34. Consensus Development Conference: Electroconvulsive therapy. JAMA 254:2103-2108, 1985.
35. Segman R. H., Shapira B., Gorfine M., Lerer B.: Onset and time course of antidepressant action:
Psychopharmacological implications of a controlled trial of electroconvulsive therapy. Psychophar-
macology 119:440-448, 1995.
36. Fieve R. R.: The lithium clinic: A new model for the delivery of psychiatric services. Am J Psychiatry
132:1018-1022, 1975.
37. Shakir S. A., Volkmar F. R., Bacon S., Pfefferbaum A.: Group psychotherapy as an adjunct to lithium
maintenance. Am J Psychiatry 136:455-456, 1979.
38. Kripke D. F., Robinson D.: Ten years with a lithium group. McLean Hosp 10: I-II, 1985.
39. Volkmar F. R., Bacon S., Shakir S. A., et al: Group therapy in the management of manic-depressive
illness. Am J Psychother 35:226-234, 1981.
40. Davenport Y. B., Ebert M. H., Ad1and M. L., Goodwin F. K.: Couples group therapy as an adjunct to
lithium maintenance of the manic patient. Am J Orthopsychiatry 47:495-502, 1977.
41. Glick I. D., Clarkin J. F., Spencer J. H., et aI: A controlled evaluation of inpatient family intervention:
Preliminary results of the six-month follow-up. Arch Gen Psychiatry 42:882-886, 1985.
3. BIPOLAR DISORDER 147
42. Miklowitz D. J., Goldstein M. 1.: Behavioral family treatment for patients with bipolar affective
disorder. Behav Modif 4:457-489, 1990.
43. Miller 1. w., Keitner G. 1., Epstein N. B., Bishop D. S., Ryan C. E.: Families of bipolar patients:
Dysfunction, course of illness, and pilot treatment study. 1991 (Abstract).
44. Scott J.: Psychotherapy for bipolar disorder. Br J Psychiatry 167:581-588, 1995.
45. Cochran S. D.: Preventing medical noncompliance in the outpatient treatment of bipolar affective
disorders. J Consult Clin Psycho I 52:873-878, 1984.
46. Price L. H., Heninger G. R.: Lithium in the treatment of mood disorders. N Engl J Med 331 :591-598,
1994.
47. Manji H. K., Potter W. Z., Lenox R. H.: Signal transduction pathways: Molecular targets for lithium's
actions. Arch Gen Psychiatry 52:531-543, 1995.
48. Pollack S. J., Atack J. R., Knowles M. R., et al: Mechanism of inositol monophosphatase, the putative
target of lithium therapy. Proc Natl Acad Sci U.S.A. 91 :5766-5770, 1994.
49. Berridge M. J., Downes C. P., Hanley M. R.: Neural and developmental actions of lithium: A unifying
hypothesis. Cell 59:411-419, 1989.
50. Baraban J. M.: Toward a crystal-clear view of lithium's site of action. Proc Natl Acad Sci U.S.A.
91:5738-5739, 1994.
51. Manji H. K., Chen G., Shimon H., Hsiao J. K., Potter W. Z., Belmaker R. H.: Guanine nucleotide-
binding proteins in bipolar affective disorder: Effects of long-term lithium treatment. Arch Gen Psychi-
atry 52:135-144, 1995.
52. Keller M. 8., Lavori P. W., Kane 1. M., et al: Subsyndromal symptoms in bipolar disorder: A compari-
son of standard and low serum levels of lithium. Arch Gen Psychiatry 49:371-376, 1992.
53. Fieve R. R.: Overview of therapeutic and prophylactic trials with lithium in psychiatric patients, in
Gershon S., Shopsin B. (eds): Lithium. Its Role in Psychiatric Research and Treatment. New York,
Plenum Press, 1973, p. 336.
54. Van Putten T.: Why do patients with manic-depressive illness stop their lithium? Compr Psychiatry
16:179-183, 1975.
55. Cooper T. B.: Pharmacokinetics of lithium, in Meltzer H. Y. (ed): Psychopharmacology: The Third
Generation of Progress. New York, Raven Press; 1987, pp. 1365-1376.
56. Schou M., Anderson A., Eskajaer-Jensen S., et al: Occurrence of goiter during lithium treatment. Br J
Med 3:710-713,1968.
57. Emerson C. H., Dyson W. L., Utiger R. D.: Serum thyrotropin and thyroxin concentration in patients
receiving lithium carbonate. J Clin Endocrinol Metab 36:338-346, 1973.
58. Jefferson 1. w.: Lithium carbonate-induced hypothyroidism-its many faces. JAMA 242:271-272,
1979.
59. VanderVelde C. D., Gordon M. W.: Manic-depressive illness, diabetes mellitus and lithium carbonate
treatment. Arch Gen Psychiatry 21 :478-485, 1969.
60. Mellerup E. T., Plenge P., Rafaelson O. J.: Renal and other controversial adverse effects of lithium, in
Meltzer H. Y. (ed): Psychopharmacology: The Third Generation of Progress. New York, Raven Press;
1987, pp. 1443-1448.
61. Gelenberg A. J.: Lithium-induced polyuria: Approaches to management. Bioi Ther Psychiatry Newslett
6:17-19, 1983.
62. Gelenberg A. J., Wojcik J. D., Falk W. E., et al: Effects of lithium on the kidney. Acta Psychiatr Scand
75:29-34, 1987.
63. Gitlin M. 1.: Lithium-induced renal insufficiency. J Clin Psychopharmacol 13:276-279, 1993.
64. Gelenberg A. 1.: Lithium and the nephrotic syndrome. Bioi Ther Psychiatry 12:24, 1989.
65. Shopsin B., Gershon S.: Pharmacology-toxicology of the lithium ion, in Gershon S., Shopsin B. (eds):
Lithium: Its Role in Psychiatric Research and Treatment. New York, Plenum Press; 1973, pp. 107-146.
66. Schou M.: Electrocardiographic changes during treatment with lithium and with drugs of the im-
ipramine type. Acta Psychiatr Scand 169:258-259, 1963.
67. Roose S. P., Nurnberger J. L., Dunner D. L., et al: Cardiac sinus node dysfunction during lithium
treatment. Am J Psychiatry 136:804-806, 1979.
148 HEATHER S. HOPKINS AND ALAN J. GELENBERG, M.D.
68. Jaffe C. M.: First-degree atrioventricular block during lithium carbonate treatment. Am 1 Psychiatry
134:88-89, 1977.
69. Worthley L. 1. C.: Lithium toxicity and refractory cardiac arrhythmia treated with intravenous magne-
sium. Anesth Intens Care 2:357-360, 1974.
70. Gelenberg A. 1., Jefferson 1. w.: Lithium tremor. 1 Clin Psychiatry 56:283-287, 1995.
71. Jefferson 1. W., Greist 1. H.: Adverse reactions-neurological tremor, in Jefferson 1. W., Greist 1. H.
(eds): Primer of Lithium Therapy. Baltimore, Williams & Wilkins, 1977, pp. 139-150.
72. Ghandirian A. M., Lehman H. E.: Neurological side effects of lithium: Organic brain syndrome,
seizures, extrapyramidal side effects, and EEG changes. Compr Psychiatry 21:327-335, 1980.
73. Levine S. H., Puchalski c.: Pseudotumor cerebri associated with lithium therapy in two patients. 1 Clin
Psychiatry 51 :251-253, 1990.
74. Alvarez-Cermeno J. c., Fernandez 1. M., O'Neill A., Moral L., Saiz-Ruiz J.: Lithium-induced head-
ache. Headache 29:246-247, 1989.
75. Saul R. E, Hamburger H. A., Selhorst J. B.: Pseudotumor cerebri secondary to lithium carbonate.
lAMA 253:2869-2870, 1985.
76. Pesando P., Nuzzi G., Maraini G.: Bilateral papilloedema in long term therapy with lithium carbonate.
Pharmakopsychiatr Neuro-Psychopharmakol 13:235-239, 1980.
77. Lobo A., Pilek E., Stokes P. E.: Papilledema following therapeutic dosages of lithium carbonate. 1 Nerv
Ment Dis 166:526-529, 1978.
78. Gelenberg A. 1.: Polypharmacy in psychiatry: Balancing caution with daring. CNS Drugs 4:410-411,
1995.
79. Schou M.: Long-lasting neurological sequelae after lithium intoxication. Acta Psychiatr Scand 70:594-
602, 1984.
80. Cohen L. S., Friedman J. M., Jefferson 1. w., Johnson E. M., Weiner M. L.: A reevaluation of risk of in
utero exposure to lithium. lAMA 271:146-150, 1994.
81. Thase M. E., Rush A. J.: Treatment-resistant depression, in Bloom E E., Kupfer D. 1. (eds): Psycho-
pharmacology: The Fourth Generation of Progress. New York, Raven Press; 1995, pp. 1081-1097.
82. Donaldson S. R., Gelenberg A. J., Baldessarini R. 1.: The pharmacologic treatment of schizophrenia: A
progress report. Schizophrenia Bull 9:504-527, 1983.
83. Gjessing L. R.: Lithium citrate loading of a patient with periodic catatonia. Acta Psychiatr Scand
43:372-375, 1967.
84. Kline N. S., Wren J. C., Cooper T. B., et al: Evaluation of lithium therapy in chronic and periodic
alcoholism. Am 1 Med Sci 268:15-22, 1974.
85. Rifkin A., Quitkin E, Carrillo R., et al: Lithium carbonate in emotionally unstable character disorder.
Arch Gen Psychiatry 27:519-522, 1972.
86. Gottfries C. G.: The effect of lithium salts on various kinds of psychiatric disorders. Acta Psychiatr
Scand 203:157-167, 1968.
87. American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders, DSM-Ill-R.
Washington, D.C., American Psychiatric Association, 1987.
88. Bowden C. L., Brugger A. M., Swann A. c., et al: Efficacy of divalproex vs. lithium and placebo in the
treatment of mania. lAMA 271:918-924, 1994.
89. Prien R. E, Caffey E. M., Klett C. J.: A comparison of lithium carbonate and chlorpromazine in the
treatment of acute mania. Arch Gen Psychiatry 26:146-153, 1972.
90. Shopsin B., Gershon S., Thompson H., Collins P.: Psychoactive drugs in mania: A controlled compari-
son of lithium carbonate, chlorpromazine, and haloperidol. Arch Gen Psychiatry 32:34-42, 1975.
91. Post R. M., Jimerson D. c., Bunney W. E., Jr., Goodwin E K.: Dopamine and mania: Behavioral and
biochemical effects of the dopamine receptor blocker pimozide. Psychopharmacology (Berlin) 67 :297-
305, 1980.
92. Cookson J. C., Silverstone T., Wells B.: A double-blind controlled study of pimozide versus chlo-
rpromazine in mania. Psychopharmacol Bull 16:38-41, 1980.
93. Kessel J. B., Verghese C., Simpson G. M.: Neurotoxicity related to lithium and neuroleptic combina-
tions? A retrospective review. 1 Psychiatry Neurosci 17:28-30, 1992.
3. BIPOLAR DISORDER 149
94. Spring G., Frankel M.: New data on lithium and haloperidol incompatibility. Am J Psychiatry 138:818-
821, 1981.
95. Zarate C. A., Tohen M., Baldessarini R. J.: Clozapine in severe mood disorders. J Clin Psychiatry
56:411-417, 1995.
96. Suppes T., McElroy S. L., Gilbert J., Dessain E. c., Cole J. 0.: Clozapine in the treatment of dysphoric
mania. Bioi Psychiatry 32:270-280, 1992.
97. Kimmel S. E., Calabrese J. R., Woyshville M. J., Meltzer H. Y.: Clozapine in treatment-refractory mood
disorders. J Clin Psychiatry 55:91-93, 1994.
98. Calabrese J. R., Kimmel S. E., Woyshville M. 1., et al: Clozapine for treatment-refractory mania. Am J
Psychiatry 153:759-764, 1996.
99. Calabrese J. R., Meltzer H. Y., Markovitz P. J.: Clozapine prophylaxis in rapid cycling bipolar disorder.
J Clin PsychopharmacoI1l:396-397, 1991.
100. Goodnick P. J.: Risperidone treatment of refractory acute mania. J Clin Psychiatry 56:431-432,1995.
101. Tohen M., Zarate C. A., Jr., Centorrino E, Hegarty J., Froeschl M., Zarate S. B.: Risperidone in the
treatment of mania. J Clin Psychiatry 57:249-253, 1996.
102. Dwight M. M., Keck P. E., Jr., Stanton S. P., Strakowski S. M., McElroy S. L.: Antidepressant activity
and mania associated with risperidone treatment of schizoaffective disorder. Lancet 344:554-555,
1994.
103. Koek R. J., Kessler C. C.: Probable induction of mania by risperidone. J Clin Psychiatry 57: 174-175,
1996.
104. Tomlinson W. c.: Risperidone and mania. Am J Psychiatry 153:132-133, 1996.
105. Naylor G., Scott c.: Depot injections for affective disorders. Br J Psychiatry 136:105, 1980.
106. Lowe M., Batchelor D.: Depot neuroleptics and manic depressive psychosis. Int J Clin Psychophar-
macoll:53-62, 1986.
107. White E., Cheung P., Silverstone T.: Depot antipsychotics in bipolar affective disorder.lnt Clin Psycho-
pharmacoI8:1l9-122, 1993.
108. Esparon J., Kalloor J., Naylor G. J., et al: Comparison of the prophylactic action of flupenthixol with
placebo in lithium treated manic depressive patients. BrJ Psychiatry 148:723-725,1986.
109. Littlejohn R., Leslie E, Cookson 1.: Depot antipsychotics in the prophylaxis of bipolar affective
disorder. Br J Psychiatry 165:827-829, 1994.
110. Ahlfors U. G., Baastrup P. C., Dencker S. J., et al: Aupenthixol decanoate in recurrent manic depressive
illness. Acta Psychiatr Scand 64:226-237, 1981.
111. Sachs G. S.: Adjuncts and alternatives to lithium therapy for bipolar affective disorder. J Clin Psychia-
try 50:31-39, 1989.
112. Sernyak M. J., Woods S. W.: Chronic neuroleptic use in manic-depressive illness. [Review]. Psycho-
pharmacol 8uIl29:375-381, 1993.
113. Sernyak M. J., Griffin R. A., Johnson R. M., Pearsall H. R., Wexler B. E., Woods S. W.: Neuroleptic
exposure following inpatient treatment of acute mania with lithium and neuroleptic. Am J Psychiatry
151:133-135, 1994.
114. Vestergaard P., Schou M.: Prospective studies on a lithium cohort. Acta Psychiatr Scand 78:421-426,
1988.
115. Goodwin E K., Murphy D. L., Bunney W. E., Jr.: Lithium carbonate treatment in depression and mania:
A longitudinal double-blind study. Arch Gen Psychiatry 21:486-496, 1969.
116. Goodwin E K., Murphy D. L., Dunner D. L., Bunney W. E., Jr.: Lithium response in unipolar versus
bipolar depression. Am J Psychiatry 129:44-47, 1972.
117. Noyes R., Jr., Dempsey G. M., Blum A., Cavanaugh G. L.: Lithium treatment of depression. Compr
Psychiatry 15:187-193, 1974.
118. Baron M., Gershon E. S., Rudy V., Jonas W. Z., Buchsbaum M.: Lithium carbonate response in
depression: Prediction by unipolar/bipolar illness, average-evoked response, catechol-O-methyl trans-
ferase, and family history. Arch Gen Psychiatry 342:1107-1111, 1975.
119. Mendels J.: Lithium in the treatment of depressive states, in Johnson E N. (ed): Lithium Research and
Therapy. New York, Academic Press; 1975, pp. 43-62.
150 HEATHER S. HOPKINS AND ALAN J. GELENBERG, M.D.
120. Altshuler L. L., Post R. M., Leverich G. S., Mikalauskas K., Rosoff A., Ackerman L.: Antidepressant-
induced mania and cycle acceleration: A controversy revisited. Am J Psychiatry 152: 1130-1138, 1995.
121. Stoll A. L., Mayer P. Y., Kolbrener M., et al: Antidepressant-associated mania: A controlled compari-
son with spontaneous mania. Am J Psychiatry 151:1642-1645, 1994.
122. Bauer M. S., Calabrese J., Dunner D. L., et al: Multisite data reanalysis of the validity of rapid cycling
as a course modifier for bipolar disorder in DSM-IY. Am J Psychiatry 151:506-515, 1994.
123. Sachs G. S., Lafer B., Stoll A. L., et al: A double-blind trial ofbupropion versus desipramine for bipolar
depression. J Clin Psychiatry 55:391-393, 1994.
124. Himmelhoch J. M., Thase M. E., Mallinger A. G., Houck P.: Tranylcypromine versus imipramine in
anergic bipolar depression. Am J Psychiatry 148:910-916, 1991.
125. Bowden C. L.: Role of newer medications for bipolar disorder. J Clin Psychopharmacol 16:48S-55S,
1996.
126. Calabrese J. R, Fatemi S. H., Woyshville M. 1.: Antidepressant effects of lamotrigine in rapid cycling
bipolar disorder. Am J Psychiatry 153: 1236, 1996.
127. Gregersen H., Nielsen J. S., Peterslund N. A.: [Acute porphyria and multiple organ failure during
treatment with lamotrigine). Ugeskr Laeg 158:4091-4092, 1996.
128. Hennessy M. 1., Wiles C. M.: Lamotrigine encephalopathy. Lancet 347:974-975, 1996.
129. Campistol J., Geli M., L1istosella E., Molins 1., L10bert M.: [Stevens-Johnson syndrome after
lamotrigine treatment). Rev NeuroI23:1236-1238, 1995.
130. Sterker M., Berrouschot J., Schneider D.: Fatal course of toxic epidermal necrolysis under treatment
with lamotrigine. Int J Clin Pharmacol Ther 33:595-597, 1995.
131. Makin A. J., Fitt S., Williams R, Duncan J. S.: Fulminant hepatic failure induced by lamotrigine. Br
Med J 311:292, 1995.
132. Anderson G. D., Yau M. K., Gidal B. E., et al: Bidirectional interaction of valproate and lamotrigine in
healthy subjects. Clin Pharmacol Ther 60: 145-156, 1996.
133. Giannini A. 1., Houser W. L., Jr., Loiselle R H., Giannini M. C., Price WA: Antimanic effects of
verapamil. Am J Psychiatry 141:1602-1603, 1984.
134. Garza-Trevino E. S., Overall 1. E., Hollister L. E.: Verapamil versus lithium in acute mania. Am J
Psychiatry 149:121-122, 1992.
135. Dubovsky S. L., Franks R D., Allen S., Murphy J.: Calcium antagonists in mania: A double-blind study
of verapamil. Psychiatry Res 18:309-320, 1986.
136. Popkin M. K.: Patients with multiple sclerosis presenting to psychiatric hospitals. J Clin Psychiatry
56:307-308, 1995.
137. Brady K. T., Sonne S. c.: The relationship between substance abuse and bipolar disorder. J Clin
Psychiatry 56: 19-24, 1995.
138. Regier D. A., Farmer M. E., Rae D. S., et al: Comorbidity of mental disorders with alcohol and other
drug abuse: Results from the Epidemiologic Catchment Area (ECA) study. JAMA 264:2511-2518,
1990.
139. Brady K. T., Sonne S. C., Anton R., Ballenger 1. c.: Valproate in the treatment of acute bipolar affective
episodes complicated by substance abuse: A pilot study. J Clin Psychiatry 56:118-121, 1995.
140. Prien R E, Potter W. Z.: NIMH workshop report on treatment of bipolar disorder. Psychopharmacol
Bull 26:409-427, 1990.
141. Bowden C. L.: Predictors of response to divalproex and lithium. J Clin Psychiatry 56:25-30, 1995.
142. Zetin M., Garber D., De Antonio M., et al: Prediction of lithium dose: A mathematical alternative to the
test-dose method. J Clin Psychiatry 47: 175-178, 1986.
143. Zetin M., Shokooh S., Daghighi-Asi A., et al: Prediction of lithium dose: A follow-up study. Lithium
1:182-183, 1990.
144. Cummings M. A., Haviland M. G., Wareham J. G., et al: A prospective clinical evaluation of an
equation to predict daily lithium dose. J Clin Psychiatry 54:55-58, 1993.
145. Terao T., Oga T., Nozaki S., et al: A further prospective evaluation of an equation to predict daily
lithium dose. J Clin Psychiatry 56: 193-195, 1995.
146. McElroy S. L., Keck P. E., Stanton S. P., Tugrul K. C., Bennett 1. A., Strakowski S. M.: A randomized
3. BIPOLAR DISORDER 151
comparison of divalproex oral loading versus haloperidol in the initial treatment of acute psychotic
mania. J CUn Psychiatry 57:142-146, 1996.
147. Frances A., Docherty J. P., Kahn D. A., Carpenter D.: The expert concensus guideline series: Treatment
of bipolar disorder. J Clin Psychiatry, 57 (Suppl 12A):1-88, 1996.
148. Stoll A. L., Locke C. A., Vuckovic A., Mayer P. V.: Lithium associated cognitive and functional deficits
reduced by a switch to divalproex sodium: A case series. J Clin Psychiatry 57:356-359, 1996.
149. Solomon D. A., Ryan C. E., Keitner G. I., et al: A pilot study of lithium carbonate plus divalproex
sodium for the continuation and maintenance treatment of patients with bipolar I disorder. J Clin
Psychiatry 58:95-99, 1997.
150. Wehr T. S., Goodwin F. K.: Can antidepressants cause mania and worsen the course of affective illness?
Am J Psychiatry 144:1403-1411, 1987.
151. Roy-Byrne P. P., Joffe R. T., Uhde T. w., Post R. M.: Approaches to the evaluation and treatment of
rapid-cycling affective illness. Br J Psychiatry 145:543-550, 1984.
152. Price W. A., DiManio L.: Premenstrual tension syndrome in rapid-cycling bipolar affective disorder. J
Clin Psychiatry 47:415-417, 1986.
153. Stoll A. L., Sachs G. S., Cohen B. M., Lafer B., Christensen 1. D., Renshaw P. F.: Choline in the
treatment of rapid-cycling bipolar disorder: Clinical and neurochemical findings in lithium-treated
patients. Bioi Psychiatry 40:382-388, 1996.
154. Jacobsen F. M.: Low-dose valproate: A new treatment for cyclothymia, mild rapid cycling disorders,
and premenstrual syndrome. J Clin Psychiatry 54:229-234, 1993.
155. Privitera M. R., Lamberti J. S., Maharaj K.: Clozapine in a bipolar depressed patient. Am J Psychiatry
150:986, 1993.
156. Goodnick P. J.: Nimodipine treatment of rapid cycling bipolar disorder. J Clin Psychiatry 56:330, 1995.
157. Benjamin 1., Zohar J.: Sleep deprivation in rapid-cycling bipolar affective disorder: Case report. Eur
NeuropsychopharmacoI2:463-465, 1992.
158. Kasper S., Wehr T. A.: The role of sleep and wakefulness in the genesis of depression and mania.
Encephale 18:45-50, 1992.
4
Psychoses
I. INTRODUCTION
Alan J. Gelenberg, M.D.· Department of Psychiatry, University of Arizona Health Sciences Center,
Tucson, Arizona 85724. Samuel Keith, M.D. Department of Psychiatry, University of New Mexico
School of Medicine, Albuquerque, New Mexico 87131.
The Practitioner's Guide to Psychoactive Drugs (Fourth Edition), Gelenberg and Bassuk, eds.
Plenum Publishing Corporation, New York, 1998
153
154 ALAN J. GELENBERG, M.D., AND SAMUEL KEITH, M.D.
merely anecdotal. However, even when scientific research has not kept pace with
clinical practice, patients continue to require medication; therefore, this chapter dis-
cusses the most up-to-date information on the use of these compounds, blending data
from scientific research with clinical experience.
A. Nonbiological
Psychosocial treatments for schizophrenia have gone through a number of phases
relative to their usefulness in this disorder. This evolution began in the 1960s, when it
was felt unethical not to provide psychotherapy, progressed through a period of high
enthusiasm for pharmacologic treatments during which psychotherapy was considered
useful only if it encouraged compliance, and has reached a point today where specific
therapies are considered based on their specific efficacy. It is safe to say that enthusi-
asm for intensive individual psychotherapy either on the inpatient unit or during fol-
low-up with outpatients has relatively little support from the research literature. Begin-
ning with the Camarillo State Hospital Study, by May et al.,) which examined five
therapeutic modalities-milieu, psychotherapy alone, ECT, psychotherapy with medi-
cation, and medication alone-the major studies have failed to find additional benefit
from psychotherapy, nor have they found that psychotherapy can substitute for medica-
tion. The results of the May study indicated that those patients who were treated with
medication did better than those who were not and that the addition of the psycho-
therapeutic component to medication did not add appreciable benefits. The second
major study, by Stanton, Gunderson, Knapp, et al., 2 was designed to determine if there
were any benefits from intensive psychotherapy over supportive psychotherapy. The
results of this study were twofold: first, the sicker patients dropped out of the more
intensive treatment, but not out of the supportive treatment; and, second, despite this
differential attrition, there were no differences between treatment outcomes on a wide
variety of measures. This does not mean that the person with schizophrenia can be
ignored; it only means that there is little support for using substantial resources on
intensive individual psychotherapy.
The data on group psychotherapy for acutely psychotic inpatients with schizo-
phrenia are even more sobering. 3 There is no evidence to suggest that group therapy is
beneficial, and, from what we know about the sensitivity of people with schizophrenia
to complex, multifocal, intensive, intrusive, critical stimuli, there is little theoretical
support for it either. Therefore, the role of inpatient group meetings should be reconsid-
ered for this population. In the outpatient setting, where the goal changes from remis-
sion of psychosis to the development of interactional skills that will allow patients to
resume some level of productivity, theoretical support and a marginal data base of
empirical evidence finds a role for low-intensity group therapy.
The major change in psychosocial treatments for schizophrenia over the last decade
has been the refinement of treatments that involve the family. Eschewing the historically
4. PSYCHOSES 155
and painfully incorrect ideas of family pathogenesis of schizophrenia, the new forms of
family treatment encourage the family to become a member of the clinical treatment
team. This rationale is based not on what a family does wrong but on what the family is in
a position to do correctly and better than anyone else in the patient's environment. Sixty-
five percent of schizophrenic patients discharged from a hospital return to their families
of origin, and even more in the early stages of the illness. In these days of declining
resources, the family represents a natural support system that would be very difficult to
replicate. However, the family should be a short-term resource and not be expected to
become the new back wards of the community. Further, many families have suffered
under the threefold punishment of having a child with schizophrenia, being accused of
causing the illness, and, despite being excluded from the treatment process, being asked
to pay for it. It is no wonder that families are skeptical over a "new" approach to family
treatment.
The common features of new approaches to family treatment of schizophrenia
include (1) a positive clinical alliance, (2) psychoeducation, (3) teaching general man-
agement skills for the illness, and (4) expanding social networks of the family. The
strategies have differed in location of delivery (home or clinic), in mode (multiple
families, single families; with or without the patient), in timing (in hospital or outpa-
tient), and even in principles (behavioral, educational, or psychodynamic). What has
not differed is the results. Each of the five studies done in the 1980s showed a
remarkable effect of family management programs when compared with other forms of
treatment, including, for example, individual psychotherapy.4,5 The treatment princi-
ples include education of the patient and family, management of stress (particularly
interpersonal stress), general case management, and a special emphasis on early recog-
nition of relapse.
Family management, like all psychosocial approaches, should not be seen as
competitive with pharmacologic treatment but instead as an additive or synergistic
approach to the illness of schizophrenia.
B. Nondrug, Biological
1. Electroconvulsive Therapy
Electroconvulsive therapy (ECT) can playa dramatic, even lifesaving, role in the
treatment of psychotic depression. It also suppresses acute mania. A wealth of clinical
anecdotes testify to similarly impressive results in many cases of catatonic schizo-
phrenia (however, not all nor even most catatonia is schizophrenia, which reduces the
confidence in the anecdotal data6 ).
At the 1985 Consensus Development Conference on ECT sponsored by the Na-
tional Institutes of Health and Mental Health, Dr. Joyce Small reported that ECT can
playa role in treating schizophrenic disorders that are relatively acute and marked by
intense affective symptoms.? By contrast, patients who have been ill for 5 years or
more are unlikely to improve. Antipsychotic drugs appear superior in efficacy to ECT
for the treatment of acute schizophrenia, but the combination of drugs and ECT might
156 ALAN J. GELENBERG, M.D., AND SAMUEL KEITH, M.D.
2. Psychosurgery
The frontal lobotomy, used to treat many schizophrenic patients earlier in the
century, has largely been abandoned because of an unfavorable risk/benefit ratio. More
restrictive psychosurgical procedures, such as cingulotomy, may effectively treat pa-
tients with chronic, severe depressions, obsessive-compulsive disorder, and intractable
pain syndromes, but no substantial evidence supports the use of psychosurgery in the
treatment of psychosis. 8
quilizers." This term developed from the observation that chlorpromazine and reser-
pine produce somnolence and relaxation and the consequent misbelief that their major
action was sedative. However, relatively non sedating antipsychotic compounds just as
effectively combat psychotic symptoms. Moreover, patients often become tolerant to
the sedating effects of antipsychotic drugs but not to the antipsychotic effects them-
selves. Finally, because of the implication that "major tranquilizers" are on a spectrum
with, but more powerful than, sedative-hypnotic and antianxiety agents (sometimes
called "minor tranquilizers"), this term is best avoided. Our concern is that patients
may have these medications prescribed for "tranquilization" and be at risk for long-
term side effects.
Because antipsychotic drugs frequently produce signs of neurological dysfunc-
tion, most notably Parkinson's syndrome and other extrapyramidal reactions, the term
"neuroleptic" was coined. In fact, researchers originally believed that (1) any drug
effective in combating psychosis must produce extrapyramidal effects and (2) in any
given patient, the induction of extrapyramidal signs indicated an optimal therapeutic
dose. However, based on the following evidence, these assumptions are incorrect:
1. The piperidyl phenothiazine thioridazine (Mellaril® and others) produces a
relatively low incidence of short-term extrapyramidal effects yet is as effec-
tive as any other antipsychotic agent.
2. Clozapine (Clozaril®), a dibenzodiazepine compound, is an effective anti-
psychotic agent that produces few, if any, extrapyramidal effects.
3. Many patients show marked clinical improvement in psychotic symptoms
without experiencing parkinsonian signs or related reactions.
Therefore, although all antipsychotic agents on the U.S. market are, in fact, neurolep-
tics, an antipsychotic drug need not be neuroleptic-a fact presenting a challenge for
future research. Furthermore, when treating individual patients, clinicians should
try to avoid the emergence of neurological effects. It is beginning to appear that there
are several dopamine systems in the brain, with the nigrostriatal being responsible for
the extrapyramidal symptoms and those projecting from the ventral tegmental area
being responsible for antipsychotic effects (see Fig. 1). If this is confirmed, it may
mean that it is not necessary for extrapyramidal symptoms to appear at all. Some of the
antipsychotic drugs currently under investigation purport to have this differential ef-
fect. Recent research suggests the intriguing possibility that the first appearance of
slight hypokinesia and rigidity might signal the minimally effective neuroleptic dose
for the treatment of an acutely psychotic patient I 0, II; this remains to be explored more
fully. The introduction of a growing number of "atypical neuroleptics," which are
supposed to be less likely than typical neuroleptics to cause extrapyramidal effects, has
opened new clinical vistas and promises enhanced insights into the nature of these
drugs and their actions.
- - - - - Mesocortical pathway
-"+-t......f - - - - Caudate & putamen
"'fT;I~- Mesolimbic pathway
1 - - - - Median eminence
Tubero-infundibular pathway
Arcuate nucleus of hypothalamus
......--.,;:~-- Nigrostrtatal pathway
- - - Amygdala
1---- Substantia nigra (A9)
- - - Ventral tegmental area (A10)
suppress vital centers in the brainstem: coma, respiratory depression, and cardiovascu-
lar collapse are rare, even at very high doses.
In the electroencephalogram (EEG), phenothiazines and other antipsychotic
chemicals produce slowing and synchronization and a decrease in arousal-induced
changes-effects that are reversed by dopamine agonists. The low-potency agents
(e.g., chlorpromazine) also tend to lower the seizure threshold. Clinically, this effect is
particularly important in patients predisposed to seizures, such as those with epilepsy,
and in individuals undergoing withdrawal from sedative-hypnotic drugs (including
alcohol) (see Chapter 7).
Most antipsychotic agents (with thioridazine as an interesting exception) have
antiemetic effects. They can protect against the nausea and vomiting that usually follow
administration of apomorphine, presumably by blocking the latter's dopamine agonis-
tic effects in the chemoreceptor trigger zone of the medulla.
C. Mechanism of Action
Antipsychotic drugs block dopamine receptors in various pathways within the
brain, which probably accounts for their therapeutic effectiveness as well as for some
of their more prominent unwanted effects. According to widely held theories, antip-
sychotic activity depends on the blockage of postsynaptic receptors in dopamine-
mediated pathways that run from the midbrain to the limbic system (septal nucleus, the
olfactory tubercle, and the amygdala) (see Fig. 1) and to the temporal and frontal lobes
of the cerebral cortex. In fact, the effectiveness of antipsychotic drugs in blocking the
D2 subtype of dopamine receptors correlates with their clinical potency (i.e., usual
daily doses). Presumably, tolerance to the dopamine-blocking action of antipsychotic
drugs does not develop in these mesolimbic and mesocortical pathways, explaining the
impression that tolerance does not develop to their antipsychotic efficacy. [Although
some authors have postulated the development of tolerance to the antipsychotic effec-
tiveness of these drugs ("tardive psychosis"),12 at best this is rare, and most clinicians
and scholars remain to be convinced that it is a valid phenomenon.]
Antipsychotic drugs also block dopamine receptors in the pathways from the
substantia nigra in the midbrain to the head of the caudate nucleus in the basal ganglia
(see Fig. 1). Interruption of communication in this nigrostriatal pathway is thought to
account for Parkinson's syndrome-bradykinesia, rigidity, and tremor. In this neuronal
network, tolerance to the dopamine-blocking action of the drugs does seem to develop.
Chemically blocked dopamine receptors are initially underactive and then become
normally active to overactive, developing what is analogous to denervation supers en-
sitivity. Underactivity of the striatal dopamine receptors presumably results in parkin-
sonian signs; overactivity has been postulated to cause tardive dyskinesia, a syndrome
of abnormal involuntary movements, although this is being called into question, as we
discuss later. Acetylcholine and O-aminobutyric acid (GABA) mediate transmission
within adjacent connecting neuronal systems. Treatment of both Parkinson's syndrome
and tardive dyskinesia may involve drugs purported to act on any or all of these three
neurotransmitters (see Section III.G.1.a).
Researchers believe that a third important dopamine pathway, the tubero-infun-
160 ALAN J. GELENBERG, M.D., AND SAMUEL KEITH, M.D.
dibular system, is affected by antipsychotic drugs. It projects from the arcuate nucleus
of the hypothalamus to the median eminence, where it acts to inhibit (directly or
indirectly) the release of prolactin from the anterior pituitary (see Fig. I). By blocking
dopamine neurotransmission in this system, antipsychotic drugs cause increased pro-
lactin secretion and hyperprolactinemia, producing unwanted effects and possible long-
term toxicity (see Section III.G.6). Although researchers have believed that tolerance
does not develop to the prolactin-elevating effect of antipsychotic drugs, it now appears
that at least partial tolerance may develop over a period of many months or even years.
Some have suggested that the elevation of prolactin in plasma could be used as an
index of the clinical effectiveness of antipsychotic drugs; however, maximum prolactin
elevation occurs at much lower doses than those usually required to treat a psychotic
disorder.
Our recent understanding of dopamine autoreceptors provides a possible clue to
the behavior of the various major dopamine pathways. Autoreceptors are present on the
body or axon of a nerve cell and respond to the cell's own neurotransmitter by decreas-
ing its synthesis and release. Thus, they serve as part of a single-cell feedback inhibi-
tion loop. Interestingly, dopamine autoreceptors do not occur in the mesocortical
pathways but are present in the nigrostriatal and mesolimbic pathways.13 The former
does not exhibit tolerance to antipsychotic drugs, but the latter two do. Similarly, we
think tolerance does not develop to the antipsychotic actions of these drugs but does to
their extrapyramidal and prolactin-raising actions. Thus, it may be the presence or
absence of the dopamine autoreceptors within a system that determines whether toler-
ance to a clinical effect will or will not occur.J3
Our understanding of the mechanisms of action of antipsychotic drugs has ex-
panded in recent years with receptor subtype cloning, new imaging technology, and the
advent of agents with novel and atypical activities. Clozapine, synthesized in the late
1960s and introduced into clinical research in the 1970s (although not approved for
general clinical use in the United States until 1990), showed that it was possible for an
agent to be an effective antipsychotic and have minimal extrapyramidal and prolactin-
elevating effects. This suggested that clozapine might have more selective dopamine-
blocking properties in pathways such as the mesolimbic and mesocortical, with compar-
atively mild effects on nigrostriatal pathways (where Parkinson's syndrome is generated)
and the tubero-infundibular pathway (where prolactin secretion is regulated). Recent
research has shown that both typical antipsychotics and clozapine decrease spontaneous
activity in single dopamine cells in the mesolimbic pathways. Although typical anti-
psychotics similarly suppress spontaneous activity of neurons in the nigrostriatal path,
clozapine does not. 14 An anatomic correlate of this pharmacologic effect has emerged in
the past few years: patients treated with typical antipsychotic drugs have increased
caudate nuclei volumes, but caudate nuclei volume decreases when patients are switched
to clozapine.
Also of interest, clozapine decreases extracellular dopamine concentrations in rat
striatum and nucleus accumbens but increases extracellular dopamine in the medial
prefrontal cortex-a possible explanation for its diminution of negative as well as
positive schizophrenic symptoms. 15 Recent studies also have shown that while all
anti psychotics block the D2 and D3 subtypes of dopamine receptors, clozapine has a
4. PSYCHOSES 161
uniquely high affinity for the D4 receptor, found to be increased in postmortem brains
of schizophrenics. 16 Scientists have shown that drugs that more selectively block DJ
receptors activate exploratory locomotor activity in laboratory animals, while D 2 -
receptor blockers, which include most classical neuroleptics, tend to produce more
sedation. 16 These findings might suggest means to alleviate both negative and positive
symptoms of schizophrenia. Neuroscientists also have paid attention to the fact that
clozapine has high affinity for the 5-HT2 type of serotonin receptor. Some of the newer
agents, such as risperidone (Risperdal®), which have comparatively low tendencies to
produce extrapyramidal reactions at low doses, have similarly high ratios of blocking
ability at the 5-HT2 receptor versus the D2 receptor.
Olanzapine (Zyprexa®) has a relatively high affinity for muscarinic, anti-
cholinergic, 5-HT2, and dopamine D 1, D2, and D4 receptors. Its blockade at a 2 recep-
tors is weaker than that of clozapine or risperidone. Like typical neuroleptics, but
unlike clozapine, it produces a modest rise in prolactin levels. Like olanzapine, sertin-
dole (Serlect®) is a potent blocker of cortical 5-HT2 and dopamine receptors. Other
atypical neuroleptics currently being studied have different properties (see Table 1).
The potent effects of antipsychotic agents on the autonomic nervous system
explain many of their adverse reactions. They block a-noradrenergic receptors, which
probably accounts for their hypotensive action, particularly on a postural basis. At
common clinical doses, low-potency antipsychotic drugs tend to be more potent at
a-adrenergic receptors and to produce a greater drop in orthostatic blood pressure. The
same low-potency agents also cause more sedation, which may reflect a-adrenergic
antagonism at brain receptors. Central effects on noradrenergic systems might be
involved in antipsychotic activity as well.t7
Antipsychotic agents block muscarinic acetylcholine receptors, producing other
autonomic effects (see Section III.G.2). As a group, antipsychotic drugs tend to be
much less potent as anticholinergic agents than the tricyclic antidepressants (see Chap-
ter 2). The most potent antimuscarinic drug among typical neuroleptics is the phe-
nothiazine thioridazine, which approaches the tricyclics in anticholinergic activity.
Clozapine is even more potent in vitro, but in humans it manifests such seemingly
cholinergic effects as increased salivation and enuresis. High-potency antipsychotic
agents are relatively weak blockers of cholinergic receptors.
Typical antipsychotic drugs produce other effects on neurotransmission, although
their significance is unclear. Similarly to tricyclic antidepressants, they block the reup-
take of norepinephrine, but this effect is probably outweighed by their tendency to
antagonize norepinephrine receptors. Other antipsychotic drugs also block serotonin
and histamine receptors; the latter might account for some of their sedative and appe-
tite-increasing tendencies. Other effects involve r3-adrenergic receptors and GABA,
serotonin, and peptide neurotransmitters.
The hypothalamus mediates various antipsychotic drug effects. In addition to
effects on prolactin, antipsychotic drugs inhibit the release of growth hormone (which
might have implications for their use in children). Some of their effects on autonomic
activity may also result from actions within the hypothalamus. Furthermore, they
impair the temperature-regulating mechanisms by making the normally homeothermic
mammalian system poikilothermic (i.e., the body temperature drifts toward that of the
162 ALAN J. GELENBERG, M.D., AND SAMUEL KEITH, M.D.
Sertindole
Amperozide
Clozapine
Savoxepine
environment), which has resulted in cases of hypo- and hyperthermia. In the extreme,
antipsychotic drugs have created what has been termed the neuroleptic malignant
syndrome (see Section III.G.l.d), also thought to involve the hypothalamus. Increased
appetite, yet another hypothalamic effect, often results in weight gain.
Chemical structurel
Nonproprietary name Trade name Approximate potencyb Available as injectable representative agent
Phenothiazines
Aliphatic
Chlorpromazine Thorazine® and others 1:1 Yes
~N~N/CH3
s~ tH3
!lACI
Triflupromazine Vesprin® 4:1 Yes
~N~
~ CH 3
~S....-CH3
Piperidine
Thioridazine Mellaril® and others 1:1 No
Piperazine
Trifluoperazine Stelazine® and others 30:1 Yes
,,&; N ............... N
~ 1
s~ ~N,CH
I 3
CF3
Fluphenazine Prolixin® 50-100:lc Yesd
(continued)
Chemical structure!
Nonproprietary name Trade name Approximate potencyb Available as injectable representative agent
Thioxanthenes
Chlorprothixene Taractan® and others 1:1 Yes
Thiothixene Navane® and others 25:1 Yes
.-!!P t;Yl
"1 C~'CH
I 3 3
S...-N'CH
Butyrophenone
%J cf'b 3
"1
CI 0
Diphenylbutylpiperidine
ifON~F
Pimozide Orap® -' No
9-0
HN3 ~ --
p-o-F
o ~ b
Dibenzodiazepine F
Clozapine Clozaril® 2:1 No
OCH3
O@
I
H
Olanzapine Zyprexa® -' No
Dibenzoxazepine
Loxapine Daxolin®, Loxitane®, 7:1 Yes
and others Nf"'.NCH3
Q@CI
Dihydroindolone
Molindone Moban® 10:1 No
o CH 3
~Jl£-
o V~CH3
H
Benzisoxazole
Risperidone Risperdal® _e No
~CH3 H
,----QCH,CH,()( ~
~ -~F
Imidazolidinone
Sertindole Serlect® _e No
F
cO
I 0
Q
~
0 H N /CH".)lNH
......... CH 2 V
a Adapted from Gelenberg. 87
hMiIligram equivalence to chlorpromazine.
"Oral only. Potency equivalences not established for long-acting injectable forms. Usual dosage for fluphenazine decanoate is 0.5-2.5 ml every 3 to 5 weeks.
d Available both as short-acting fluphenazine Hel injection and as long-acting enanthate and decanoate esters.
eNot established.
166 ALAN J. GELENBERG, M.D., AND SAMUEL KEITH, M.D.
somewhat different cyclic structure results in the piperazine phenothiazines [e.g., tri-
fluoperazine (Stelazine® and others)]. When a piperazine phenothiazine has a terminal
hydroxyl (OH) group, esterifying it with a fatty acid results in a highly fat-soluble
hybrid that diffuses into the body's adipose tissue, releasing the parent phenothiazine
over a period of weeks. Examples are the enanthate and decanoate esters of the
piperazine phenothiazine fluphenazine (Prolixin® and others).
Replacing the nitrogen atom in the central ring with a carbon atom produces a
second group of effective antipsychotic substances, the thioxanthenes. They have either
aliphatic [e.g., chlorprothixene (Taractan®)] or piperazine [e.g., thiothixene (Navane®
and others)] structures and are chemically and pharmacologically similar to the phe-
nothiazines.
The butyrophenones, developed by Janssen, appear structurally quite different
from the phenothiazines but are pharmacologically very similar to the piperazines. The
only butyrophenone labeled for antipsychotic use in the United States is haloperidol
(Haldol® and others), although the anesthetic agent droperidol (Inapsine® and others)
also appears to have antipsychotic properties. Closely related to the butyrophenones are
the diphenylbutylpiperidines, which are undergoing experimental investigation. This
class currently includes only pimozide (Orap®), an extremely potent blocker of the
neurotransmitter dopamine.
The dibenzoxazepine drugs are a fourth antipsychotic group with a three-ring
structure. The only member currently labeled as an antipsychotic agent in the United
States is loxapine (Loxitane® and others). [The demethylation of loxapine has resulted
in amoxapine (Asendin®), a compound with antidepressant properties.) Closely related
to these compounds are clozapine and olanzapine, which are dibenzodiazepines.
The sixth antipsychotic group, the dihydroindolones, are solely represented by
molindone (Moban®). Introduced into clinical use in 1994, risperidone is a novel
antipsychotic agent of a new chemical class, the benzisoxazole derivatives, with a
molecular structure distinctive from all other currently available antipsychotic classes.
Another structurally unique agent is sertindole, an imidazolidinone. Rauwolfia alka-
loids, such as reserpine and tetrabenazine (not available in the United States), are
infrequently used as antipsychotic agents and are primarily of historical interest.
For a tricyclic antipsychotic to be effective, three carbon atoms must lie between
the amino nitrogen and the nitrogen of the center ring. The addition of an electronega-
tive substituent to the benzene ring (e.g., CI, SCH3 , CF3 ) enhances its efficacy, whereas
the piperazine group on the side chain increases its potency. As a rule, molecules of
typical neuroleptics with greater milligram potency produce less sedation and
hypotension but more acute extrapyramidal reactions (see Table 3).
Low-potency drugs
Fewer extrapyramidal reactions (especially thioridazine)
More sedation and postural hypotension
Greater effect on the seizure threshold and electrocardiogram (especially thioridazine)
More likely skin pigmentation and photosensitivity
Occasional cases of cholestatic jaundice
Rare cases of agranulocytosis
High-potency drugs
More frequent extrapyramidal reactions
Less sedation and postural hypotension
Less effect on the seizure threshold, less cardiovascular toxicity
Fewer anticholinergic effects
Occasional cases of neuroleptic malignant syndrome
able to predict that in the not-distant future, the scheduled weekly venopunctures and
white blood cell counts required at the time of this writing may be decreased, and with
increased competition from newer agents, the manufacturer may decrease the actual
price of the medicine.
Rather than considering clozapine a "last-ditch" option for difficult-to-treat pa-
tients, there is an argument for its use as a first-line antipsychotic. True, it has a high
incidence of troublesome and even life-threatening complications. However, chronic
psychosis is a debilitating disease that robs the soul if not physical life itself. With
careful management, the life-threatening complications of clozapine can be kept to a
minimal risk, while the agent itself clearly has superior efficacy to traditional neurolep-
tics. Moreover, its capacity to treat the negative symptoms of schizophrenia, which
cause so much functional debility and suffering, are clearly superior to that of tradition-
al agents, along with much greater normalization of a patient's life.
There is some argument that newer atypical agents, with gentler side-effect pro-
files, should be tried before clozapine. The other atypical agents available in the United
States are risperidone and olanzapine. (Sertindole may be introduced soon.) Ris-
peridone has not been shown to be a superior antipsychotic nor to be as effective as
clozapine in patients who have not responded to traditional agents. In addition, ris-
peridone's pharmacologic profile resembles that of typical antipsychotics in that it
induces hyperprolactinemia, produces extrapyramidal symptoms at higher doses, and
has shown inconsistent effects on the negative symptoms of schizophrenia. As newer
atypical agents are introduced, they must be held to clozapine's standards as we
struggle to create a set of clinical priorities and a rank ordering of medications to try in
patients with schizophrenia and other chronic psychotic disorders.
F. Pharmacokinetics
The rate and completeness of absorption of drugs determine the rapidity and
intensity of onset of their clinical effects. Orally administered antipsychotic drugs are
variably absorbed. When given intramuscularly, the drugs produce higher (2-10 times)
and more reliable blood levels at a more rapid rate. Antipsychotic agents are highly
lipophilic [with the apparent exceptions of thioridazine and its metabolite mesoridazine
(Serentil®)] and bind tightly to protein and body membranes; they are poorly dialyzed.
As with other drugs that cross the blood-brain barrier, they also cross the placental
membrane to enter the fetal circulation and are transported into mammalian milk.
The half-life of a drug determines the time required to achieve a steady-state
concentration in the body (i.e., the point at which tissue concentrations become stable).
Steady state is reached after approximately four half-lives. Drugs with longer half-lives
may be administered less frequently, and they disappear from the body more gradually
when discontinued following chronic treatment. The half-life of an antipsychotic drug
in plasma following a single dose is usually 10-20 hr (see Table 4). However, the
drug's effects typically persist much longer, presumably because the brain half-life is
longer. When these drugs are administered chronically, the brain and body adipose
tissues become saturated; these supplies are then released and excreted very slowly. In
addition, when an antipsychotic drug has many metabolites (e.g., chlorpromazine), they
TABLE 4. Pharmacokinetics of Antipsychotic Agentsa
Phenothiazines
Chlorpromazine 2-4 16-37 (but active metabs 98-99 Many, unclear which of > 160 30-350 ng/mL?
probably much longer) metabs active
Fluphenazine (oral) 12-24 None? 0.5-3.0 ng/mL?
Mesoridazine -6 -25
Perphenazine 8-21 None 0.8-2.4 ng/mL?
Thioridazine 2 (concentrate); 7-42 96-99 Mesoridazine, 1-1.5 ng/mL?
4 (tablets) sulphoridazine
Trifluoperazine 3-6 17-18 Unknown Uncertain (possible demethylated Not established (1-2.3 ng/mL?)
and hydroxylated metabolites)
Butyrophenone
Haloperidol 2-6 12-40 -90 Reduced haloperidol? 5-20ng/mL?
Dibenzoxepine
Loxapine IIMI2PO 3.4 (loxapine) [9 (8-0H -90 8-0H loxapine, 8-OH amoxapine 30-100 ng/mL (pamlt loxapine
loxapine), 30 (8-0H plus 8-0H loxapine)
Amoxapine)]
Dibenzodiazepines
Clozapine 1-4 6-33 >90 Uncertain (demethylated Not established (100-600
metabolites may be active) ng/mL?)
Olanzapine 4-6 20-70 93 None
Ind%ne
Molindone 0.5 1.5 None ?
Thioxanthene
Thiothixene 1-3 a = 3.5;~ = 34 Not studied None identified 2-15ng/mL?
Benzisoxazo/e
Risperidone 3.6 90 9-Hydroxyrisperidone Not established
Imidazolidinone
Sertindole 7-13 24-200 >99 Uncertain (dehydrosenindole and
hydroxylated metabolites may
be active)
are detected in the urine for many months after discontinuation of the drug following
chronic administration.
Attempts to correlate plasma concentrations of antipsychotic drugs with their
clinical efficacy have yielded contradictory results. This unsatisfying situation reflects
problems in research methodology, but also problems in biochemical assays and ques-
tions about active drug metabolites. Although it was once thought that a radioreceptor
assay measuring the level of dopamine-blocking activity in serum would replace direct
chemical measurements of drug concentrations, this is less clear-cut at present. 18 ,19
Available evidence hints at the existence of a curvilinear ("therapeutic window")
relationship for the butyrophenone haloperidol, a drug with no (or possibly one) active
metabolite(s), although the precise upper and lower boundaries on this "window"
remain to be delimited.
Regardless of any given antipsychotic drug's specific half-life, virtually all of
these agents can be administered in a once-daily dose, usually at bedtime. The clinical
effectiveness of this regimen presumably reflects the drugs' prolonged brain effects and
gradual elimination from the body. After chronic dosing, biological effects of these
agents persist for many weeks. After prolonged use of decanoate preparations, clinical
and biological effects can extend beyond 6 months!
As with most psychotropic drugs, lipophilic parent compounds are oxidized to
inactive hydrophilic metabolites largely within microsomal enzymes in the liver. These
metabolites are excreted primarily in the urine and to a lesser extent in bile, Metabo-
lism and excretion of antipsychotic drugs are greatest in healthy young people and less
at either end of the age spectrum.
The blood levels achieved with antipsychotic drugs following a standard dose
vary widely among individuals, which may account for some of the differences in
clinical effects.
ber, or clinician who is unfamiliar with them. They may be uncomfortable. They are
rarely dangerous. However, in rare cases there can be respiratory compromise with the
potential for a fatality. The diagnosis of acute dystonic reaction is usually not difficult if
it is clear that a patient has recently begun taking an antipsychotic drug or has had a
switch in the type or dosage of medication. At times, however, eliciting this informa-
tion may be difficult, especially with patients who are taking prochlorperazine supposi-
tories (and say they are not taking any tranquilizer pills), who do not wish to acknowl-
edge use of antipsychotic drugs, or who have sought antipsychotic drugs for illicit use.
Among the many neuropsychiatric syndromes that must be considered in the differen-
tial diagnosis of acute dystonic reactions are tetanus, seizures, and conversion reac-
tions.
The likelihood of producing acute dystonias is the greatest with the highest-
potency antipsychotic drugs, much less with the low-potency agents, and the least with
thioridazine and clozapine. Young people are at greater risk of developing this syn-
drome than the elderly. We used to think males experienced it more frequently than
females, but this is no longer clear. Patients who have previously experienced acute
dystonia are at greater risk of its recurrence upon rechallenge with a neuroleptic. When
the acute dystonic reactions are frequent or severe in a patient, it may be worthwhile to
assay the serum concentration of calcium, since rare cases of hypocalcemia have been
detected in this way.
The mechanism underlying acute dystonic reactions is unclear. One hypothesis is
that the syndrome reflects an acute increase in dopamine neurotransrnission in the basal
ganglia, which transiently supervenes the blockade of dopamine receptors brought
about by the same drugs. Although dystonic reactions do not occur in naturally occur-
ring Parkinson's disease (paralysis agitans), they are observed in postencephalitic
Parkinson's syndrome, which shares other features with antipsychotic drug-induced
extrapyramidal reactions.
Although the mechanism may be unclear, the treatment of an acute dystonic
reaction is straightforward, readily available, and usually dramatically successful.
Parenteral treatment is preferred for initiating drug therapy in more severe
cases, with intravenous being more rapid than intramuscular. The intravenous
injection of contraactive medication provides both immediate relief and confirmation
of the diagnosis. For this purpose, a number of classes of agents have been employed
with considerable success. An injectable anticholinergic antiparkinson drug such as
benztropine (Cogentin Gt and others), 1 mg, or biperiden (Akineton Gt and others),
2 mg, may be used. Other clinicians prefer to administer an antihistamine drug such
as diphenhydramine (BenadrylGt and others), 50 mg. Some doctors use a ben-
zodiazepine intravenously, such as diazepam (Valium® and others), up to 10 mg ad-
ministered over at least 2 minutes. This drug is relatively safe and does not add
additional anticholinergic effects. As with the injection of any other central depres-
sant compound, equipment for support of the airway should be immediately available
in case an emergency occurs, and care must be taken to avoid accidental intraarterial
injection.
Following immediate relief of acute dystonic signs, the clinician may wish to
begin oral administration of one of these drugs, using the lowest effective dose (for
172 ALAN J. GELENBERG, M.D., AND SAMUEL KEITH, M.D.
range, see Table 5). If this successfully prevents additional reactions, the contraactive
medication can usually be tapered and discontinued within several weeks. Recent data
suggest that coadministering an antiparkinson drug from the beginning of antipsychotic
drug therapy in the hope of avoiding an acute dystonic reaction may offer partial
protection to some patients. 19 The physician may want to consider this approach for
patients at highest risk (i.e., young patients receiving high-potency antipsychotic
agents) and for those in whom a reaction is likely to be clinically disruptive. For other
patients, it makes sense to avoid a drug that may be unnecessary, reserving treatment
until signs of dystonia appear. As atypical antipsychotics are used more often and
typical agents are used at lower doses, concomitant antiparkinson therapy should be
needed less often.
Physicians, nurses, family members, and anyone else who may observe a patient
receiving an antipsychotic medicine should be aware of the possible occurrence of a
dystonic reaction early in therapy. Treatment should be readily available. For most
patients, the risk of dystonic reactions appears to wane with continued antipsychotic
drug therapy.
ii. Akathisia. Akathisia is another extrapyramidal reaction associated with both
antipsychotic drugs and postencephalitic Parkinson's syndrome (rarely with Parkin-
son's disease). Akathisia is a symptom (i.e., subjective) defined as a compulsion to be
in motion. Patients describe an inner restlessness, an intense desire to move about
simply for the sake of moving. Persons who are virtually paralyzed by the akinesia of
postencephalitic Parkinson's disorder have been known to ask other people to move
their limbs, just to relieve this intense compulsion. Patients suffering from akathisia are
often observed to pace aimlessly, fidget, and be markedly restless. At times, akathisia
may cause a worsening of psychosis.20
Akathisia can occur early in the course of drug treatment, or it may not appear for
several months. It, too, appears to be more prevalent with high-potency drugs. The
natural course of akathisia is less clear than that of acute dystonic reaction: at times it
4. PSYCHOSES 173
appears to wane, yet some patients are troubled by it for a long time. Its mechanism is
obscure but may involve a blockade of mesocortical dopamine receptors.
Treatment responses are variable; they include attempting to lower the dose of the
antipsychotic drug, switching to a lower-potency or atypical agent, or adding a contra-
active drug. The same drugs discussed for the treatment of acute dystonic reactions-
anticholinergic antiparkinson agents, antihistamines, and benzodiazepines-also may
be tried orally in cases of akathisia, although results are less universally successful (see
Section III.G.l.a). Whether the dopamine agonist antiparkinson drug amantadine
(Symmetrel® and others) may playa role in the treatment of this disorder is unclear.
Recent research has shown that the l3-adrenergic-blocking drug propranolol (Inde-
ral® and others) can effectively treat both subjective and objective manifestations of
akathisia while producing an acceptable profile of side effects; other l3-blockers may be
safe and effective as well. 19 Another agent that diminishes central noradrenergic activ-
ity, the antihypertensive clonidine (Catapres®), also may be effective but appears to
cause an unacceptable incidence of sedation and hypotension. 2o Akathisia is an impor-
tant syndrome to recognize, as it may severely complicate a patient's response to
antipsychotic drug therapy; perhaps most important, it makes patients extremely un-
happy.
iii. Parkinson's Syndrome. Parkinson's syndrome is characterized by a triad of
signs: tremor, rigidity, and akinesia (or bradykinesia). The tremor (by definition, a
regular and rhythmic oscillation of a body part around a point) is approximately 4-8
cycles per second and is greater at rest than during activity. A parkinsonian tremor is
often observed in the hands; the thumb rubbing against the pad of the index finger may
produce a characteristic "pill-rolling" appearance. The tremor also can involve the
wrists, elbows, head, palate, or virtually any body part. In neuroleptic-induced Parkin-
son's syndrome, the tremor is typically bilateral; unilateral tremors should raise ques-
tions about the etiology. Although tremor is very common and may be one of the earlier
signs in naturally occurring Parkinson's disease, in antipsychotic drug-induced Parkin-
son's syndrome it is less common than rigidity and akinesia, and it may not appear until
relatively late in the drug-related syndrome.
Rigidity is an increase in the normal resting tone of a body part. It is detectable
only by palpation on physical examination. (In other words, a patient does not look
rigid; he or she must feel rigid.) In testing for rigidity, the physician asks the patient to
relax completely and allow body parts to be manipulated without moving them. The
examiner then rotates the head on the neck, moves the major joints in the upper and
lower extremities, and raises each extremity to note the rapidity with which it falls by
gravity. An increased resistance to passive motion and a slow return from a raised
position denote the presence of rigidity. When tremor coexists with rigidity, the rigidity
may take on the feel of a "cogwheel." In extreme forms of the condition, rigidity may
mimic (or actually become) the waxy flexibility with sustained postures characteristic
of catatonia. 21 In antipsychotic drug-induced Parkinson's syndrome, rigidity tends to
be more common than tremor but less common than the third sign of the triad-
akinesia.
Akinesia is a disinclination to move in the absence of paralysis. Literally, akinesia
means the absence of motion, whereas bradykinesia, truer to the clinical reality in most
174 ALAN J. GELENBERG, M.D., AND SAMUEL KEITH, M.D.
neuroleptics. As new atypical agents come to the market, it will be important to assess
whether they are truly like clozapine or whether they are more like thioridazine in their
likelihood of generating extrapyramidal effects.
Pharmacological contraactive therapy for parkinsonism consists of attempts to
counterbalance the decreased dopamine neurotransmission by either blocking acetyl-
choline transmission [anticholinergic antiparkinson drugs include benztropine and tri-
hexyphenidyl (Artane® and others)] or by increasing dopamine neurotransmission
(e.g., by the use of amantadine and others). Other dopamine agonist drugs such as
L-dopa and bromocriptine (Parlodel®) may be useful in other forms of Parkinson's
syndrome but are seldom employed for the drug-induced variety. Table 5 lists available
drugs that are useful for the treatment of drug-induced Parkinson's syndrome. With
constant awareness of the pleomorphic and often subtle manifestations of extrapyrami-
dal reactions, the physician prescribes one of these contraactive drugs as needed for a
patient's comfort and optimal functioning, always seeking the lowest effective dose.
Doctors should periodically attempt to taper and discontinue a concomitant antiparkin-
son drug, although recent evidence has shown that many patients on maintenance
neuroleptic therapy require parallel maintenance with antiparkinson drugs; for most,
this can be carried out with impunity.20
With the exception of amantadine, all of the antiparkinson drugs listed in Table 5
have anticholinergic and/or antihistaminic properties. Among these drugs, trihex-
yphenidyl has a relatively short half-life, whereas benztropine has a relatively long
half-life. Amantadine is a dopamine agonist that has few anticholinergic effects. This
feature may make it preferable in patients particularly sensitive to anticholinergic
reactions. 23 For many patients, the strongly anticholinergic antiparkinson drugs can
impair memory, time perception, and cognition: these central nervous system anti-
cholinergic side effects also might prompt the use of amantadine. 24 Amantadine, 100
mg b.i.d. to 100 mg t.i.d., may be effective in some cases of drug-induced Parkinson's
syndrome, particularly the more severe variety, when anticholinergic antiparkinson
drugs have been ineffectual.25 Amantadine has a relatively long half-life (approx-
imately 24 hr). It is excreted in the urine unchanged and may lead to toxicity (including
psychiatric symptoms) in patients with impaired renal function. It is also particularly
hazardous and potentially lethal in overdose.
Antiparkinson drugs, as well as other chemicals with anticholinergic activity (e.g.,
belladonna-containing compounds), have been used for "recreational" purposes. 26 It
appears that the anticholinergic activity provides a mood-elevating effect for many
people, creating a feeling of euphoria or a "high." Some aficionados of the drug culture
actually use the drugs to create a toxic delirium, which they perceive as pleasurable.
Many schizophrenic patients treated with antiparkinson drugs become more attached to
these drugs than to their antipsychotic agents, possibly because of the relief of extra-
pyramidal symptoms or perhaps because of directly pleasurable effects. Trihexypheni-
dyl has been reported to be more commonly abused than other antiparkinson agents. If
this is true, it could be related to the pharmacology of the drug or simply to its
popularity and, therefore, availability. Anticholinergic toxicity and its treatment are
described in Section III.G.2.
For the last 10 to 15 years, there has been a growing awareness that lower
176 ALAN J. GELENBERG, M.D., AND SAMUEL KEITH, M.D.
patient and interfere with important activities such as eating, talking, and dressing. In
rare instances, tardive dyskinesia can impair breathing and swallowing.
A metaanalysis of 21 studies found that exposure to neuroleptics multiplies the
risk of developing dyskinesia by 2.9 times. 29 It is generally believed that for tardive
dyskinesia to occur, a patient must have been exposed, more or less continuously, to an
antipsychotic drug for at least 3-6 months, although it is probable that a rare, highly
sensitive patient may develop this syndrome after an even briefer period. Presumably,
increasing exposure to antipsychotic drugs increases the risk of developing tardive
dyskinesia, although if a patient has not developed it after some length of time, he or
she is probably less vulnerable.
Patients who are older, female, black, or right-handed have an increased risk of
developing tardive dyskinesia on neuroleptic exposure. Other factors associated with
an increased risk are psychiatric diagnosis of mood disorder, negative schizophrenic
symptoms, a past history of early and severe acute extrapyramidal reactions, and type II
diabetes. Among typical neuroleptics, there is no evidence that anyone drug, chemical
class, or potency grouping is more or less likely to produce tardive dyskinesia. Recent
evidence suggests that clozapine is more likely than traditional neuroleptics to be
associated with a decrease in abnormal involuntary movements in a patient with preex-
isting movements following previous treatment with typical neuroleptic drugS. 30,31 In
addition, clozapine is less likely than typical neuroleptics to be associated with new-
onset cases of tardive dyskinesia in patients with previous histories of exposure to
traditional neuroleptics. 31 Coadministration of anticholinergic agents can exacerbate
existing movements of tardive dyskinesia but does not appear to increase the risk of it
developing. It appears that "drug holidays," i.e., regular, abrupt discontinuation of
antipsychotic drug therapy, do not reduce the risk of tardive dyskinesia. Although it has
been suggested that drug holidays may even increase the risk,27 this is controversial.
Prospective studies suggest a link between cumulative dosage and the development of
tardive dyskinesia, underscoring the wisdom of maintaining patients on the lowest
effective dosage over prolonged periods of time.
Estimates of the prevalence of tardive dyskinesia have varied widely-from 1%
to over 50% of patients currently taking neuroleptic drugs. 32 However, when the most
minimal cases are removed from consideration, an attempt is made at differential
diagnosis, and a careful and reliable screening procedure is used, the prevalence of
tardive dyskinesia in a group of patients maintained on antipsychotic drugs for a
variable period of time (i.e., a typical outpatient schizophrenic population) is likely to
be 20-30%.33 (A comparable group of patients not treated with antipsychotic drugs
may have a prevalence rate of 5%.) A more important question concerns the incidence
of tardive dyskinesia. A review of seven tardive dyskinesia incidence studies by Glazer
et al. in 1993 revealed average rates of new cases per year from approximately 4% to
11 %, with 5-year risks between 17.5% and 42.1 %.34 The prevalence of tardive dyston-
ia may be 1-2%.33
Although the prevalence of tardive dyskinesia is considerable (and actually may
be growing, as the population at risk takes neuroleptics for longer and longer periods),
for most patients the neurological manifestations are comparatively mild. Many pa-
178 ALAN J. GELENBERG, M.D., AND SAMUEL KEITH, M.D.
tients deny even an awareness of the dyskinetic movements, and dysfunction is rare.
The most severely afflicted patients, and those likely to suffer the most, are patients
with tardive dystonia.
The movements of tardive dyskinesia may appear during treatment when a patient
is taking a constant dose of an antipsychotic, or they may initially appear when the
dosage of the drug is lowered or stopped entirely. Conversely, if movements are
present, increasing the dose of an antipsychotic drug can make the movements cease.
This latter phenomenon has been referred to as "masking" of the movements and gives
rise to the concept of "covert dyskinesia" -in other words, dyskinesia observed only
when an antipsychotic drug has been discontinued. 35 On the other hand, when an
antipsychotic drug is discontinued abruptly, a patient may show transient dyskinetic
movements that can disappear in a matter of days or weeks. This phenomenon has been
known as withdrawal dyskinesia, and it might conceivably indicate that a patient is
vulnerable to a persistent dyskinesia if the medication is reinstituted and continued.
Tardive dyskinesia itself is believed to be persistent, although evidence suggests that if
the movements are detected early and medication is discontinued, many patients will
show gradual improvement over time. 36
Anticholinergic drugs frequently exacerbate the movements of tardive dyskinesia
once they are present, and the administration of anticholinergic agents may "unmask"
latent movements. Conversely, discontinuing an anticholinergic drug in a patient with
tardive dyskinesia may improve the movements. Tardive dystonia, by contrast, might
actually improve in some cases on administration of an anticholinergic drug. Anti-
cholinergics do not seem to increase the likelihood that a patient will develop tardive
movements.
The most widely held hypothesis about the mechanism of tardive dyskinesia has
involved the nigrostriatal dopamine pathway. Chronic blockade of dopamine receptors
within the basal ganglia leads first to receptor underactivity, then to overactivity. In
other words, the initial receptor blockade from antipsychotic drugs leads to hypoac-
tivity at the receptor neurons. This state is believed to underlie the genesis of Parkin-
son's syndrome. Prolonged blockade, however, results in a situation analogous to
denervation supersensitivity. In fact, chronic neuroleptic administration to animals has
been shown to cause an actual multiplication in the number of postsynaptic receptor
sites. More recent magnetic resonance imaging (MRI) studies in humans have found
that prolonged exposure to typical neuroleptics increases the size of the caudate (possi-
bly correlating with the small duplication in receptor sites)37-39; switching a patient
from a typical neuroleptic to clozapine or possibly discontinuing a typical neuroleptic
results in a reduction in the increased size of the caudate. 40 A recent computerized
tomographic (CT) study found structural abnormalities in the caudate nucleus and
temporal lobe in patients with tardive dyskinesia-more marked in the left hemi-
sphere. 41 Thus, long-term administration of antipsychotic drugs leads ultimately to an
increase in activity at postsynaptic dopamine receptors. This state of excessive dopa-
minergic activity is believed to result in the abnormal movements of disorders such as
tardive dyskinesia and Huntington's chorea. 42
The "dopamine excess" hypothesis of tardive dyskinesia is not completely satisfy-
ing, however. For example, prolonged exposure to neuroleptic drugs results in dopa-
4. PSYCHOSES 179
mine receptor supersensitivity in virtually all experimental animals, but only some
patients develop tardive dyskinesia. Also, such supersensitivity is less likely to develop
in older animals, but tardive dyskinesia is more likely to develop in older patients.
These and other objections have led many scholars to conjecture that the dopamine
supersensitivity hypothesis of tardive dyskinesia is too simplistic to be universally
applicable. Perhaps such a mechanism is valid in some patients, but alternative neuro-
transmitter systems that may go awry and contribute to the development of tardive
dyskinesia include those involving GABA, acetylcholine, and norepinephrine.
Tardive dyskinesia (and Huntington's chorea) is in many ways the opposite of
Parkinson's syndrome: the former is a disorder of extraneous movements, and the latter
a paucity of movements; the former is thought to reflect dopaminergic overactivity
(and relative cholinergic underactivity), whereas the latter is caused by too little dopa-
mine neurotransmission (and relatively too much cholinergic) (but see above reserva-
tions). Pharmacologic strategies that tend to make one better frequently make the other
worse.
Treatment approaches to tardive dyskinesia are based on our understanding (and
beliefs) about the nature of antipsychotic drugs and tardive dyskinesia. First, of course,
is primary prevention. This means avoiding unnecessary exposure of patients to anti-
psychotic drugs; in particular, avoiding the use of these agents to treat relatively benign
conditions or those that could respond equally well or better to other agents (e.g.,
anxiety, nonpsychotic depression). It also means using the lowest dose for the shortest
period of time when it is necessary to employ antipsychotic drug therapy. Of course,
most schizophrenic patients will require prolonged therapy with these agents, but the
clinician should find the lowest effective maintenance dose. Similarly, in the treatment
of patients with other chronic disorders such as mental retardation and organic brain
syndrome, the use of drugs should be minimized and constantly reevaluated.
Secondary prevention of a disorder means early detection. For tardive dyskinesia,
this entails routine screening and monitoring of patients for the presence of abnormal
movements. A standard neurological examination can be employed for this purpose, or
a clinician may want to use a specific examination procedure and rating scale such as
the Abnormal Involuntary Movement Scale designed by the National Institute of Men-
tal Health (see Fig. 2). It is best to perform an examination and note the results prior to
the initiation of antipsychotic drug therapy (or before too long into the course) and then
to repeat the examination every 6 to 12 months while the patient remains on drug
therapy.
When the presence of abnormal involuntary movements is unequivocal, consider
the diagnosis (see Table 6). Is any other diagnosis likely? Are the movements charac-
teristic of tardive dyskinesia, or might they be tremors or other movements? Are they
chronic mannerisms and stereotypies characteristic of the underlying psychiatric disor-
der, or have these movements only appeared following chronic drug treatment? Family
history should be explored to rule out the presence of hereditary disorders of the central
nervous system such as Huntington's chorea or dystonia. The patient's history also
should be reexplored to establish a reasonable linkage between drug therapy and the
appearance of abnormal movements. Movements that inexorably progress are more
likely caused by a degenerative neurological disorder than by tardive dyskinesia. A
180 ALAN J. GELENBERG, M.D., AND SAMUEL KEITH, M.D.
(Circle one l
FACIAL AND ORAL 1. Muscles of facial expression, a 1 2 3 4
MOVEMENTS e.g., movements of forehead,
eyebrows, periorbital area,
cheeks; include frowning,
blinking, smiling, grimacing.
Figure 2. Abnormal involuntary movement scale (AIMS). Asterisk (*) denotes activated movements. From
Department of Health, Education, and Welfare, Public Health Service, Alcohol, Drug Abuse, and Mental
Health Administration, National Institute of Mental Health.
GLOBAL 8. Severity of abnormal movements. None, normal 0
MOVEMENTS Minimal 1
Mild 2
Moderate 3
Severe 4
EXAMINATION PROCEDURE
Either before or after completing the Examination Procedure, observe the patient unobtrusively, at
rest (e.g., in waiting room).
The chair to be used in this examination should be a hard, firm one without arms.
1. Ask patient whether there is anything in his/her mouth (i.e., gum, candy, etc.) and if there is,
to remove it.
2. Ask patient about the ~ condition of his/her teeth. Ask patient if he/she wears
dentures. Do teeth or dentures bother patient D.QW?
3. Ask patient whether he/she notices any movements in mouth, face, hands, or feet. If yes,
ask to describe and to what extent they ~ bother patient or interfere with his/her
activities.
4. Have patient sit in chair with hands on knees, legs slightly apart, and feet flat on floor. (Look
at entire body for movements while in this position.)
5. Ask patient to sit with hands hanging unsupported. If male, between legs; if female and
wearing a dress, hanging over knees. (Observe hands and other body areas.)
6. Ask patient to open mouth. (Observe tongue at rest within mouth.) Do this twice.
7. Ask patient to protrude tongue. (Observe abnormalities of tongue movement.) Do his twice.
*8. Ask patient to tap thumb, with each finger, as rapidly as possible for 10 to 15 seconds,
separately with right hand, then with left hand. (Observe facial and leg movements.)
9. Flex and extend patient's left and right arms (one at a time). (Note any rigidity.)
10. Ask patient to stand up. (Observe in profile. Observe all body areas again, hips included.)
*11. Ask patient to extend both arms outstretched in front with palms down. (Observe trunk,
legs, and mouth.)
*12. Have patient walk a few paces, tum, and walk back to chair. (Observe hands and gait.) Do
this twice.
182 ALAN J. GELENBERG, M.D., AND SAMUEL KEITH, M.D.
Associated
mental Family
Disorder disturbances history Laboratory tests
aMeige syndrome is a disorder of middle age characterized by progressive oral, lingual, and buccal dystonia together with
blepharospasm. The movements are indistinguishable from those of tardive dyskinesia, but patients with Meige syndrome
need not have had antedating antipsychotic drug exposure, and Meige syndrome is a progressive disorder.
"Be sure to ask patients about the state of their mouth and teeth and whether they have any gum in their mouth when you are
examining them for tardive dyskinesia.
<The abnormal movements of Huntington's disease are primarily chorea with little dystonia or athetosis. The movements are
generalized, producing a ''fidgety'' appearance in the early stages. Although movements of tardive dyskinesia are more
stereotyped and abnormal, the movements of Huntington's disease appear to be normal movements at an increased
frequency. Patients with tardive dyskinesia have dyskinesias other than chorea. In general, patients with Huntington's
disease have more trouble keeping their tongues out of their mouths, whereas tardive dyskinesia patients may have trouble
keeping their tongues in their mouths.
suspected. At times, the psychiatrist may wish to seek consultation from a neurological
colleague.
Helpful
1. History
a. Exposure to antipsychotic drug
b. Nonprogressive disorder
c. Rule out family history of movement disorders
2. Physical examination
a. Movements characteristic of tardive dyskinesia
b. Rule out neurological signs in other systems (e.g., cerebellar, sensory,
pyramidal)
Not helpful
1. What makes movements better (e.g., sleep, relaxation) or worse (e.g., tension,
psychological "gain")
2. Whether the patient can voluntarily inhibit movements
3. Whether movements are worsened or "unmasked" by a decrease in the anti-
psychotic drug dose
4. Laboratory tests (Although laboratory tests are not helpful to rule in tardive
dyskinesia, they can be helpful in ruling out other disorders.)
If the diagnosis of tardive dyskinesia appears firm, what is the course of treat-
ment? Raising the dose of the antipsychotic drug might succeed in partially (or even
completely) suppressing the movements, but the clinician should avoid this strategy if
possible. Available evidence suggests that continuing neuroleptic therapy, although
unlikely to exacerbate the degree of dyskinesia, may enhance the probability that the
dyskinesia will become irreversible. 43 •44 Of course, if continuing an antipsychotic drug
is necessary to control the psychosis or prevent episodic relapses (as is often the case),
then the clinician must recommend this course, and it is up to the patient (or guardian)
to decide whether to consent.
If abnormal movements are detected early, and it is clinically practical to taper
and discontinue the antipsychotic medication, it is more likely that the dyskinesia will
ultimately improve (although it may be exacerbated temporarily). Although this is ideal
from the standpoint of the movement disorder, for most schizophrenic patients the
trade-off in exacerbation of psychosis will not be worth the price. Schizophrenic
psychosis usually is worse than tardive dyskinesia, and most often the clinician, patient,
and family will decide to continue drug therapy, always at the lowest effective dose.
If the movements continue and are troublesome (and irrespective of whether the
patient remains on antipsychotic drug therapy), the clinician will want to alleviate as
much of the patient's discomfort as possible. Unfortunately, there is no standard and
accepted treatment for tardive dyskinesia or dystonia. Therapy with a benzodiazepine
drug such as diazepam or clonazepam (Klonopin®) may help some patients, either by
its sedative or muscle-relaxing properties or possibly via its role in increasing GABA-
ergic tone.
184 ALAN J. GELENBERG, M.D., AND SAMUEL KEITH, M.D.
Preliminary procedure
1. Complete a baseline neurological examination.
2. Inform the patient and family about the risk of tardive dyskinesia.
Primary prevention
1. Avoid unnecessary exposure to antipsychotics.
2. Use lowest dose of antipsychotic for shortest time.
Secondary prevention
1. Routinely screen for dyskinetic movements [routine neurological examina-
tion or rating scale such as Abnormal Involuntary Movement Scale (AIMS)].
a. Administer at the beginning of treatment.
b. Administer every 6 to 12 months.
4. PSYCHOSES 185
h. Sedation. It was initially believed that sedation was important for the effec-
tiveness of antipsychotic drugs, but this no longer appears true. Drugs can have a major
impact on the primary symptoms of psychosis such as hallucinations and thought
disorder without producing somnolence. As mentioned previously, the low-potency
antipsychotic drugs such as chlorpromazine, thioridazine, and chlorprothixene tend to
be more sedating. Clozapine is especially sedating. If around-the-clock sedation is
desired, these drugs can be administered several times daily during initial therapy. If
nighttime sedation is indicated, the drugs may be administered once daily at bedtime.
Tolerance to sedation tends to develop over a matter of days or several weeks.
and/or adding (or increasing) an anticonvulsant. Clozapine lowers the seizure threshold
in a dose-dependent fashion. If a patient is to be treated with very high doses of
clozapine, or if a seizure occurs during clozapine treatment but the clinician and patient
decide that the antipsychotic benefits of clozapine warrant its continuation, coad-
ministration of an anticonvulsant medication may be indicated. In such cases, di-
valproex may be the most logical choice. Knowledgeable clinicians in the United
States tend to avoid coadministering carbamazepine with clozapine for fear of increas-
ing the risk of bone marrow suppression.
2. Anticholinergic
a. Peripheral. Antipsychotic drugs block a SUbtype of cholinergic receptor
known as muscarinic. This type of receptor, which responds to interneuronal release of
acetylcholine, is located on postganglionic neurons of the parasympathetic branch of
the autonomic nervous system (as well as autonomic ganglion cells and certain cortical
and subcortical neurons). Thus, the atropinelike action of antipsychotic drugs antago-
nizes the effects of the parasympathetic nervous system. The most strongly anti-
cholinergic among the traditional antipsychotic drugs is thioridazine, which is almost
as potent as the cyclic antidepressants. High-potency antipsychotic drugs (e.g., flu-
phenazine, haloperidol) are comparatively weak in their atropinelike action.
In the skin, anticholinergic drugs can produce warmth, flushing, and dryness; in
the eye, dilated pupils, difficulty with visual accommodation, and increased intraocular
pressure; in the mouth, dryness; in the lungs, drying of secretions; in the heart, in-
creased rate; in the stomach, decreased acid secretion; in the bowel, diminished mo-
tility with potential constipation; in the urinary tract, smooth muscle slowing, which
can lead to delayed urination; in the penis, delayed (or retrograde) ejaculation.
In general, the elderly are more sensitive to atropinelike effects than are younger
patients. Obviously, certain illnesses are exacerbated by anticholinergic actions of
drugs. An abrupt attack of narrow-angle glaucoma is a rare possibility that can occur in
a predisposed individual. Fortunately, the more common open-angle glaucoma, parti-
cularly when controlled by drugs, is less likely to be made worse by drug administra-
tion. Individuals prone to dental problems may suffer from diminished salivation.
Patients with respiratory problems could be adversely affected by diminished pulmo-
nary secretions. Similarly, those with cardiac disorders may be compromised by tachy-
cardia as well as by more direct cardiotoxic effects, which are discussed below. Indi-
viduals with gastrointestinal disturbances, including those who have recently had
abdominal surgery, may be adversely affected by the diminished bowel motility. Like-
wise, a man with prostatic enlargement could find urination more difficult during
treatment with an anticholinergic drug.
Some degree of tolerance develops to these effects over weeks and months (and
conversely, rebound can occur on drug withdrawal). When peripheral anticholinergic
activity of antipsychotic drugs is a problem, the clinician may wish to lower the dose of
the drug or switch to a drug with less atropinic activity (i.e., a higher-potency agent). At
188 ALAN J. GELENBERG, M.D., AND SAMUEL KEITH, M.D.
times, symptomatic treatment may be offered, such as sugarless gum and lozenges for
dry mouth or a stool softener or mild laxative for constipation. In occasional patients,
the use of a peripherally active parasympathomimetic agent such as bethanechol
(Urecholine® and others) may be indicated.
In its anticholinergic activity, clozapine again is something of an anomaly. Its
binding potency at the muscarinic receptor is higher than that of older antipsychotic
drugs, even thioridazine. However, one of the common side effects of clozapine is
hypersalivation: patients typically awaken with a pillow soaked with saliva. Patients
have also reported experiencing enuresis in association with clozapine treatment. 50
Helpful antidotes were oxybutynin (Ditropan® and others) and desmopressin
(DDAVP®). Risperidone has minimal anticholinergic activity.
common with the low-potency drugs (e.g., chlorpromazine) and may be more of a
problem with the elderly or with patients with preexisting postural hypotension and
vascular instability such as those undergoing a sedative-hypnotic withdrawal reaction.
It is likely that parenteral administration of an antipsychotic drug may provoke more
severe hypotension than oral ingestion of the same drug. Clozapine produces substan-
tial postural hypotension, but risperidone does not.
If postural hypotension does develop, it usually can be managed by keeping the
patient horizontal. The next step, if necessary, is the administration of intravenous
fluids to expand the vascular volume. If these two maneuvers are not effective, then a
pure a-adrenergic pressor agent should be administered. Metaraminol (Aramine® and
others) will usually suffice, although norepinephrine (Levophed® and others) also may
be considered. The use of a mixed a- and l3-stimulating drug such as epinephrine
(Adrenalin® and others) could lead to a paradoxical drop in blood pressure, since the a
receptors are blocked, and the unopposed 13 stimulation can promote further hypoten-
sion. Similarly, the use of a primarily l3-stimulating drug such as isoproterenol (Isup-
rel® and others) should be avoided.
4. Ocular
As already noted, the anticholinergic effects of antipsychotic drugs can affect the
eye. Increased mydriasis can make a patient more light-sensitive. Interference with
visual accommodation may result in complaints of blurred vision. Because of the rare
possibility of precipitating an attack of angle-closure glaucoma, patients should be
asked prior to treatment with any anticholinergic agent whether they have a history of
visual symptoms-such as eye pain, blurring, and halos-that could suggest previous
narrow-angle episodes. Open-angle glaucoma is less likely to be a problem, but the
patient should be managed conjointly with an ophthalmologist.
Prolonged treatment with high doses of low-potency antipsychotic drugs has been
associated with the deposition of pigment in the lens, cornea, conjunctiva, and retina,
often together with skin pigmentation. Except in extreme cases, these are unlikely to
interfere with vision. However, if it is necessary to expose a patient to high doses of a
low-potency agent for long periods of time, periodic ophthalmologic examinations
would be worthwhile, perhaps annually.
Of greater clinical significance is the pigmentary retinopathy that may accompany
treatment with thioridazine or other phenothiazines with piperidine side chains. This
disorder, which can permanently impair vision, is unlikely if doses of thioridazine do
not exceed 800 mg per day, but even this is relative. Thus, thioridazine is the one
antipsychotic agent that should be considered to have an absolute "ceiling dose" -800
mg daily-which must not be exceeded for even brief periods of time. (If a patient
does require high doses of an antipsychotic drug, it is best to change to a different one.)
Although phenothiazines with aliphatic side chains, such as chlorpromazine, are much
less likely to cause ocular complications, there have been rare reports of chlor-
promazine-induced retinopathy, even at low doses. 53
Other effects of chlorpromazine on the eye are minimal. In patients who have been
receiving chlorpromazine at 500 mg/day or more for at least 3 years, however, granular
deposits have been noted on the corneal endothelium and the lens capsule. 54 These
4. PSYCHOSES 191
chlorpromazine effects are probably due to the compound coming out of solution in the
aqueous humor and depositing on surfaces bathed in the aqueous.
5. Cutaneous
7. Hepatic
therapy. In almost all cases, recovery occurs over a matter of weeks and is complete
and without sequelae. In predisposed patients, cholestatic jaundice may lead to chronic
biliary cirrhosis. Subsequent treatment probably should be with a different antipsycho-
tic drug, preferably a high-potency agent. The presence of preexisting liver disease
does not contraindicate the use of antipsychotic drugs. However, when the liver is
impaired, metabolism of antipsychotic drugs may be slowed, and other drugs that use
similar enzyme pathways could be affected by the addition of another liver-metabo-
lized agent. In addition, the central action of antipsychotic drugs may worsen a case of
hepatic encephalopathy.
8. Hematological
11. Overdose
Fortunately, antipsychotic drugs have a high therapeutic index-the ratio of toxic
to therapeutic dose. Ingestion of an antipsychotic agent alone in an overdose attempt
seldom results in death. (However, it should always be remembered that another agent
may have been ingested along with the antipsychotic drug). Possibly more dangerous
than other anti psychotics in overdose are thioridazine and loxapine.
196 ALAN J. GELENBERG, M.D., AND SAMUEL KEITH, M.D.
1. Acute Treatment
In the early placebo-controlled trials of phenothiazines in the treatment of acute
schizophrenic psychosis, about one-fourth of patients who received placebo responded
reasonably well. This means that a minority of psychotic patients who meet behavioral
criteria for this undoubtedly heterogenous condition will improve without antipsycho-
tic therapy. However, there is no clear way to predict who these patients are.
Given the current state of scientific knowledge and clinical experience, there is
wisdom in allowing a day or two to pass before administering antipsychotic drugs to a
patient undergoing a first psychotic episode. In a supportive milieu, away from poten-
tially toxic substances and psychosocial stressors, some patients may show rapid remis-
sion, without medication or with only antianxiety agents. At the same time, staff can
conduct a comprehensive medical, neurologic, and psychiatric diagnosis, psychosocial
history, and screening for substances of abuse. With this approach, unnecessary use of
antipsychotic drugs with their attendant risks can be avoided.
On the other hand, psychotic episodes should not be allowed to continue for more
than a very brief period without trying to alleviate them through administration of
antipsychotics. Not only is each episode painful, disruptive, and potentially dangerous,
the growing evidence suggests that episodes may leave a "neurobiological trace" that
may aggravate the ultimate long-term prognosis and treatment response.
Some physicians used to employ extremely large doses of high-potency anti-
psychotic drugs early in treatment. This approach, known as rapid neuroleptization,
attempted to achieve rapid remission of psychosis and avoidance of hospitalization or
rapid discharge from the hospital. Unfortunately, double-blind studies were unable to
confirm any enhanced efficacy of this form of treatment contrasted with standard
dosage regimens. In other words, treating all psychotic patients with "megadoses" does
not control psychosis better or more rapidly, and it is associated with increased un-
toward effects. Thus, the clinician should treat most patients at the lower end of the
therapeutic range for acute psychosis, e.g., 5-10 mg/day of haloperidol or flu-
phenazine. Additional sedation can be achieved through the adjunctive administration
200 ALAN J. GELENBERG, M.D., AND SAMUEL KEITH, M.D.
of a benzodiazepine [e.g., lorazepam (Ativan® and others)]. Only patients who in the
present or past have been unresponsive to such standard doses administered over a
period of weeks should be considered for higher doses.
For the highly excited or dangerous patient, the combination of a neuroleptic drug
and a benzodiazepine may provide rapid pharmacologic control in the context of a calm
but structured environment. A high-potency drug, such as haloperidol, can be adminis-
tered parenterally and is less likely to produce hypotension or seizures than lower-
potency antipsychotic agents,?7 A dose of 2.5-10 mg can be administered intra-
muscularly every 1-2 hr, alternating with intramuscular injections of lorazepam (the
only benzodiazepine to achieve reliable blood levels when administered through this
route), 1-2 mg, until the patient is quiet. Vital signs should be monitored at least
hourly. An alternative to haloperidol in a combined regimen would be mesoridazine,
25-50 mg, which is more sedating than haloperidol and less likely to produce extra-
pyramidal reactions but causes more hypotension.
Yet a third option for parenteral neuroleptic therapy is droperidol. Available in the
United States only for use in anesthesia, droperidol is a butyrophenone like haloperidol,
but more sedating and possibly faster acting. It has been described as very effective in
emergency management of acutely excited and violent patients, but because no oral
preparation is currently available, the patient must later be switched to another product.
Early in the treatment course, administer antipsychotic drugs in divided doses,
perhaps three or four times per day (more frequently for patients who are severely
excited). Once satisfactory levels have been achieved, and the patient has become
tolerant to unwanted effects (probably within several weeks), a once-daily dosage
usually will suffice. If a patient can be maintained out of the hospital, a once-daily
dosage, even from the beginning, may enhance compliance. Particularly when a more
sedating agent is employed, a bedtime dose may be optimal, as it can improve sleep,
and the peak anticholinergic and hypotensive effects will occur while the patient is
asleep.
What if a patient does not appear to respond to an antipsychotic drug even after 6
to 8 weeks? First, the clinician must wonder whether the patient is actually receiving
the agent. An outpatient may not have filled a prescription or may be taking the drug
irregularly, if at all. An inpatient might be "cheeking" tablets or capsules. Questions to
be asked at this juncture include: Is anyone observing the patient taking the medication,
including an inspection of the patient's mouth? Does the patient have any noticeable
"side effects" from the drug, such as extrapyramidal reactions? If compliance is a
particular problem, consider changing the dosage regimen, administering a liquid prep-
aration, or using a parenteral agent. Always ask patients how they feel about a drug.
They might reveal covert effects such as sexual dysfunction or akathisia, which may be
improved by switching to a different agent, or one may learn of a patient's wish to
remain "crazy" or a patient's feeling that taking the drug is an indication that he or she
is mentally ill.
If compliance is not a problem, yet the patient has not improved, he or she may
require a higher dose of medication. Increase the dosage until improvement occurs or
limiting adverse effects develop. Although U.S. Food and Drug Administration (FDA)-
approved labeling (contained in the Physicians' Desk Reference and in the package
4. PSYCHOSES 201
insert) provides useful guidelines about maximal dosage, the actual upper limit may
vary from patient to patient. The one exception to this rule is thioridazine, which must
not be administered in dosages exceeding 800 mg per day-even for a brief period of
time-because of the danger of pigmentary retinopathy with potential loss of vision.
Some patients metabolize antipsychotic drugs very rapidly in the gut and in the
first pass through the liver. A physician might suspect that this is the case in a patient
who is taking relatively large doses orally yet experiences neither clinical benefit nor
adverse effects. These patients may respond to higher oral doses or to a parenteral
preparation.
Although all antipsychotic drugs are equally effective, some patients appear to do
better with one agent than another. Thus, failure to improve with one drug (after the
above factors have been considered) is a reason to change to a different agent. Clini-
cians should be wary, however, of a tendency to be "too quick on the prescription
trigger," switching agents every few days in a vain attempt to "do something." Al-
though side effects (including sedation) occur early, antipsychotic effects may take 3-6
weeks. Only after a drug has been given an adequate trial (at least 300-600 mg of
chlorpromazine-equivalent dose per day for 6-8 weeks) should the clinician switch
to a different agent. When it is time to change, it makes most sense to choose a drug in
a different chemical class.
One other consideration in the apparently refractory patient is psychosocial fac-
tors, which may be "fueling" the psychosis. Antipsychotic drugs raise the threshold at
which an individual predisposed to psychosis actually becomes psychotic in response
to interpersonal and intrapsychic stresses. A chronic schizophrenic, for example, might
remain quite stable and relatively intact (albeit still awkward, isolated, and apathetic)
when in a low-stress environment, even at a very low dose of antipsychotic. However,
given a precipitant (an argument with a friend, family member, or mental health worker
or being in a program that is too intense or overinvolved), the patient may decompen-
sate into a state of disorganized thinking and be unable to distinguish reality from
fantasy.
It takes more medication to "cap" a psychosis in the face of ongoing stresses and
threats than in more placid circumstances. In fact, if the pressure is sufficient, psycho-
pharmacology alone will not bring the psychosis under control. Hence, the clinician
must consider psychological and environmental factors when treating an acutely de-
compensated patient. If patients are in a stable environment and are unimproved
despite adequate chemotherapy, change the environment; e.g., arrange for hospital
admission. If patients are in the hospital, inquire about ongoing contacts both inside
and outside the hospital. Occasionally, friends and family will need to be counseled on
how to reduce the patient's stress, other patients may need to be controlled from
making sexual advances, or a naive staff member might be getting provocatively close.
In such circumstances, decreasing the intensity of the patient's conflict may allow a
hitherto ineffective drug to become strikingly efficacious.
And how effective is effective for an antipsychotic drug? Traditional drugs are
best at suppressing the most florid symptoms of psychosis such as hallucinations,
delusions, disorganized thinking, and excited, assaultive, and belligerent behavior. If
the patient has a primary disorder of mood, reconstitution should involve what Bleuler
202 ALAN J. GELENBERG, M.D., AND SAMUEL KEITH, M.D.
called restitutio ad integrum-in other words, a return to an integrated and normal state
of personality and functioning. For a schizophrenic patient, however, recovery from
psychosis is likely to involve return to a somewhat strange, awkward, and isolated
existence.
Since the third edition of The Practitioner's Guide went to press several years ago,
the introduction of atypical neuroleptics has substantially altered the pharmacotherapy
of psychotic disorders. Those of us who conducted research on clozapine in the 1970s
were impressed that it was a highly efficacious antipsychotic with minimal (if any)
acute extrapyramidal reactions. What has emerged with subsequent study and clinical
experience, however, is that it is probably a superior antipsychotic, with particular
efficacy in patients previously unresponsive to traditional neuroleptics and against the
cluster of symptoms commonly termed "negative"-e.g., apathy, flattened affect, so-
cial awkwardness, and decreased motivation. However, this interpretation of the litera-
ture is under debate. Some clinicians would say that clozapine does not have direct
impact on negative symptoms, but, as is true with most antipsychotics, the negative
symptoms get better in conjunction with the positive symptoms. 78 ,79
Because of its range of serious and potentially lethal effects, clozapine is typically
reserved for patients unresponsive to or intolerant of traditional neuroleptics. However,
given the magnitude and unique nature of its efficacy against the spectrum of schizo-
phrenic and other psychotic symptoms (it may be more efficacious in psychotic mood
disorders 80,81), and the malignant (and often lethal) nature of chronic psychotic disor-
ders, one might consider whether, despite its risks, clozapine should be used as a first-
line agent for some patients (Table 8). In any case, its use should not be withheld for an
extended time, unless a patient's clinical response (in terms of both safety and efficacy)
is highly satisfactory on a traditional agent.
At the time of this writing, clozapine is expensive. The cost is partly due to the
manufacturer's pricing of the medicine itself and partly due to the currently required
weekly venopunctures and white blood cell (WBC) assays. The former is likely to
decrease with the introduction of new atypical neuroleptics, while the frequency of
blood count monitoring will also probably decrease within the next few years. Even
with the current high annual cost, however, clozapine affords many patients sufficient
clinical and functional benefits so that they are able to be treated in less costly set-
tings-thus providing an "offset" that often exceeds the cost of the medicine and
laboratory testing. 82 The benefits of clozapine treatment appear to accrue through a
second year of maintenance therapy and perhaps even beyond. This may be related to
ongoing neurobiological alterations brought about by the drug itself, or perhaps what-
ever normalizing effect it has on neuronal networks allows rehabilitative therapies and
ongoing social interactions to have a normal and salutary effect on the patient's think-
ing, feelings, and behavior.
Due to clozapine's cost, clinicians often wonder when to discontinue treatment for
lack of efficacy. The best available data suggest that some benefit should be evident
after 3 months of treatment, and if the patient is not clearly better after 6 months of
taking clozapine, the doctor can reasonably taper and stop it.
A proper role for risperidone, the second atypical neuroleptic introduced in the
United States, is less clear-cut at this time. There is no evidence that it is superior in
4. PSYCHOSES 203
Clinical indications
If a patient shows satisfactory response over 3-4 months of acute treatment for any of the indications below,
and if maintenance treatment is indicated, in most cases clozapine will be the maintenance treatment of
choice.
I. Treatment-resistant schizophreniaa
2. Schizophrenia or schizoaffective disorder-when typical neuroleptics have caused severe extra-
pyramidal reactions or tardive movement disorders.
3. Treatment-resistant or -intolerant mood disorders with psychotic features. b
4. Other psychiatric disorders for which an antipsychotic is indicated but in which typical neuroleptics
have been ineffective or poorly tolerated. c
Adverse effects
A. Serious: Agranulocytosis; cardiovascular: EKG changes and arrhythmias, orthostatic hypotension; res-
piratory arrest; seizures
B. Common: Constipation; dizziness; fever; hypertension, sedation, salivation; tachycardia; urinary incon-
tinence; weight gain
Withdrawal
Abrupt withdrawal of clozapine may lead to a more rapid and severe return of psychotic symptoms than
usually observed with traditional antipsychotic agents. Very gradual withdrawal. therefore, is recommended,
except when serious toxicity (e.g., agranulocytosis) occurs.
Overdose
The most serious consequences of clozapine overdose are depressed consciousness, cardiac arrythmias,
respiratory suppression, hypotension, and seizures. There are no specific antiodotes, and management is the
same as for other antipsychotic overdoses.
Cost
Clozapine costs $6,000-$10,000 annually. The high cost is due largely to the requirement of weekly
venipunctures and laboratory tests. As more novel antipsychotics are introduced over the next few years, the
manufacturer is likely to reduce the price of clozapine. If the requirement for WBC counts is liberalized, that
too will reduce the annual costs of treatment.
a As indicated in the text, one may argue about how many "typical" neuroleptics should be tried before c10zapine is
prescribed for a patient. Although c10zapine is unquestionably more toxic and hazardous that standard anti psychotics, its
superior efficacy suggests that in many cases it should be tried sooner rather than later in the course of a lifelong illness.
bEven though c10zapine does not carry FDA-approved labeling for this indication, there is compelling evidence that it is
efficacious and possibly even more efficacious for this indication than it is in the treatment of schizophrenia.
'Outside of FDA labeling.
204 ALAN J. GELENBERG, M.D., AND SAMUEL KEITH, M.D.
efficacy, nor that it has special advantages for patients previously resistant to standard
neuroleptics. At lower doses-8 mg/day or less-it produces a much lower incidence
of acute extrapyramidal reactions than a high-potency neuroleptic such as haloperidol.
As noted earlier, it has produced cases of neuroleptic malignant syndrome (NMS), and
there is currently no basis on which to estimate its relative risk of causing tardive
movement disorders. A common practice at this time among clinicians treating psycho-
tic patients is to try one or two standard neuroleptics and, failing to see optimal
benefits, to then prescribe risperidone, followed by clozapine. There is certainly little
reason to believe that a patient who has failed to respond to clozapine will benefit from
risperidone; however, the reverse is likely to be true. Patients who have benefited from
clozapine and then been switched to risperidone often have not shown comparable
benefits. As mentioned earlier, a rapid discontinuation of clozapine and substitution of
risperidone has produced havoc in many patients, most likely due to a combined
withdrawal effect from clozapine and a lack of sufficient compensation by risperidone.
A gradual discontinuation from clozapine may ease the transition and reduce the risks
of psychotic relapse.
As of this writing, olanzapine has recently been introduced into clinical practice.
Preliminary results have been very favorable. Some clinicians believe it approaches
clozapine in efficacy-particularly in its effects on negative symptoms and in difficult
patients. Early clinical experience suggests that it is well tolerated. Sertindole may soon
be approved for marketing in the United States, but clinical experience beyond phase
III trials does not yet exist.
Other atypical neuroleptics are certain to be introduced over the next few years.
What benefit they will bring in the important safety and efficacy variables discussed
above remains to be seen as additional studies are conducted and clinicians gain
experience in the field with patients typically much more complex than those who
qualify for research protocols. The atypical agents currently being studied have varying
pharmacologic profiles. A common denominator is that virtually all have a compara-
tively low incidence of acute extrapyramidal reactions and uncertain long-term liability
of producing tardive syndromes. The differences among this group tend to be much
greater than differences among the typical neuroleptics, making generalizations impos-
sible.
through the return to home, job, and community. Antipsychotic agents presumably
provide a buffer for the stresses of the transitional period.
If a person resumes a normal life without evidence of recurrent psychosis, gradu-
ally lower the dose of the antipsychotic drug. If dosage tapering proceeds smoothly,
consider discontinuing the agent entirely within about 6 months. However, when a
psychotic illness has shown itself to be chronic, with frequent recurrences (a common
pattern among schizophrenic patients), maintenance therapy may be in order. Data
emerging from recent clinical studies suggest that at least in some patients diagnosed
with schizophrenia, clinicians may need to consider lifelong maintenance antipsychotic
therapy after only two episodes. 83 Factors that might support such a decision include
poor premorbid functioning, a failure to return to baseline, the presence of attenuated
symptoms, and family history of psychosis.
3. Maintenance Therapy
of tardive dyskinesia as well as less troubling but more annoying side effects that might
undermine compliance.
Research has shown that lower antipsychotic drug doses do, on average, lead to
higher rates of relapse, especially with longer periods of follow-up.84 Nevertheless,
relapses in patients on low-dose antipsychotic drug therapy tend to be less severe than
those occurring in patients free of medication: they remit more easily when the drug
dose is raised and are less likely to result in social disruption and the need for hospital-
ization. Patients and families prefer low-dose treatment even with the increased risk for
relapse.
Several groups of investigators have been examining an intermittent or "targeted
dose" strategy for chronic schizophrenic patients. This involves drug treatment only
during an acute episode, with drug discontinuation afterwards. During maintenance
therapy, patients are carefully monitored, and family members know how to reach a
clinician promptly to report prodromal symptoms. This approach relies heavily on
therapeutic rapport, and it also assumes that the clinician can identify prodromal
symptoms, which may include such nonspecific evidence of distress as anxiety and
insomnia. Unfortunately, a much higher rate of psychotic breakdowns with intermittent
than with continuous treatment makes this technique too risky for most patients. A
practical approach is to use low but continuous doses with intermittent increases when
symptoms increase. Patients can be maintained for prolonged periods on very low
doses of antipsychotic drugs, but any evidence of a "roughening" of their clinical
course should prompt a rapid raising of the dose until symptoms subside. Thus, a
patient receiving 0.25 mL of fluphenazine decanoate every 3-4 weeks might receive a
supplement of 1-5 mg of fluphenazine hydrochloride daily by mouth or an additional
0.25 mL of decanoate until acute symptoms have subsided.
The use of "drug holidays" -i.e., periodic, planned withdrawal of neuroleptic
therapy-has fallen out of favor. Initially designed as a means to lower a patient's
exposure to these agents, drug holidays have not appeared to lower the risk of tardive
dyskinesia (and may, in fact, increase it). By contrast, when patients are taking these
medicines chronically, a 5-out-of-7-day schedule (i.e., once-daily dosing Monday
through Friday, with the weekend "off") does not represent a neuropharmacologic
"holiday," since the brain will hardly notice the falloff in levels (because of saturation
of fat and tissue stores). This approach may make sense for patients in day treatment
programs, who can take the medication once each weekday in front of a nurse.
As of this writing, there are three long-acting antipsychotic drugs available by
prescription in the United States: the enanthate and decanoate esters of the phe-
nothiazine fluphenazine (Prolixin Enanthate® and Prolixin Decanoate®) and the de-
canoate ester of the butyrophenone haloperidol (Haldol Decanoate®). Fluphenazine
decanoate probably has a longer duration of action than its enanthate sister, which is
infrequently used.
Fluphenazine decanoate is usually administered intramuscularly (it may also be
given subcutaneously) in doses of 12.5 mg (0.5 mL) through 75 mg (3 mL) every 3-4
weeks. Although some patients have received injections greater than 125 mg (5 mL) at
a time (requiring the use of several injections), there is a growing recognition that
patients can be maintained on the lower end of the dosage scale, with some receiving as
4. PSYCHOSES 207
little as 1.25 mg (0.05 mL) at each dose. Additionally, the dosing interval can be
progressively increased so that patients need only receive injections every 1-2 months,
based on their clinical responses. Symptomatic exacerbation is best treated through the
use of oral or 1M fluphenazine hydrochloride (short-acting), although the maintenance
dose also may be increased if need be. Since a decanoate cannot be removed once it is
injected, it is best to err on the lower side and supplement with shorter-acting prepara-
tions. Unfortunately, there is no established conversion factor to equate a dose of oral
fluphenazine hydrochloride to a dose of the decanoate, but each can be titrated within
the usual therapeutic ranges.
The Use of Fluphenazine Decanoate
1. Initially administer 0.25 mL of fluphenazine decanoate.
2. Supplement with oral or 1M fluphenazine hydrochloride as needed.
3. Give second dose of 0.25 mL of fluphenazine decanoate after 4 weeks.
4. Begin to taper oral fluphenazine.
5. If you cannot stop oral fluphenazine without reemergence of psychotic symp-
toms, increase decanoate dose to 0.5 mL at next 4-week administration.
6. Once patient is stable on fluphenazine decanoate alone, see if a 5- or 6-week
interval suffices. (Some patients may need decanoate only every 2 months or
less often!)
7. If patient begins to become symptomatic toward the end of a dosing interval,
shorten the interval. (Rare patients may need fluphenazine decanoate as often
as every 2 weeks.)
8. If a patient begins to become symptomatic around a time of stress, supple-
ment decanoate with oral or 1M fluphenazine hydrochloride as needed or
increase the decanoate dose for one to two dosing intervals until the stress and
symptoms pass.
Haloperidol decanoate is recommended for once-monthly administration, but
there are probably patients who can take it, too, at less frequent intervals. Concentra-
tions of haloperidol in the decanoate preparations are 50 mg/mL or 100 mg/mL. For
patients already stabilized on oral haloperidol, the recommended conversion formula is
to mUltiply the daily dose by 10-15 and then to administer that dose (in milligrams) of
haloperidol decanoate in a monthly injection. Again, the best approach is to "start low
and go slow," particularly when using a long-acting preparation.
Arguably the most dramatic effects observed with clozapine have been in the
long-term treatment of patients with chronic psychotic disorders. Many patients have
become less symptomatic and more functional on clozapine treatment than they had
been on standard neuroleptics, with benefit increasing steadily over the first several
years. Because of clozapine's unique side-effect profile, specific approaches to clinical
management are required for patients maintained on the drug. 85 As mentioned earlier,
patients who have recovered from clozapine-induced agranulocytosis should never
again be exposed to this drug. Those with seizures can, however, be reexposed, possi-
bly in conjunction with an anticonvulsant (usually divalproex), and certainly at a lower
initial dose. Sedation can be managed with a lower dose or a bedtime biasing of the
dosage schedule, and, as with other psychotropic medications, symptomatic hypoten-
208 ALAN J. GELENBERG, M.D., AND SAMUEL KEITH, M.D.
sion may be treated with support stockings, increased dietary sodium, or the miner-
alocorticoid fludrocortisone (Florinef® and others). Hyperthermia is common with
clozapine, but usually benign and transient; infection and NMS must be ruled out.
Occasional patients who develop nausea and, less commonly, vomiting after weeks or
months of clozapine therapy have been treated successfully with metocloprarnide (Reg-
lan® and others).
IV. CONCLUSION
REFERENCES
I. May P. R. A.: Treatment of Schizophrenia; A Comparative Study of Five Treatment Methods. New York,
Science House, 1968.
2. Stanton A. H., Gunderson J. G., Knapp P. H., et al: Effects of psychotherapy in schizophrenia: I. Design
and implementation of a controlled study. Schizophrenia Bull 10:520-563, 1984.
3. Schooler N. R., Keith S. J.: The clinical research base for the treatment of schizophrenia. Psychophar-
macol Bull 29:431-446, 1993.
4. Leff J., Kuipers L., Berkowitz R., Eberlein-Vries R., Sturgeon D.: A controlled trial of social interven-
tion in the families of schizophrenic patients. Br J Psychiatry 141: 121-134, 1982.
5. Kottgen c., Sonnichsen I., Mollenhauer K., et al: The family relations of young schizophrenic patients:
Results of the Hamburg Camberwell Family Interview Study I. Int J Fam Psychiatry 5:61-70, 1984.
6. Gelenberg A J.: The catatonic syndrome. Lancet 1:1339-1341, 1976.
7. Gelenberg A 1.: ECT: Controversies and consensus. Mass Gen Hosp Bioi Ther Psychiatry Newsletter
8:29-32, 1985.
8. Bartlett 1., Bridges P., Kelly D.: Contemporary indications for psychosurgery. Br J Psychiatry 138:507-
511, 1981.
9. Bassuk E. L.: The homelessness problem. Sci Am 251 :40-45, 1984.
10. McEvoy 1. P.: The neuroleptic threshold as a marker of minimum effective neuroleptic dose. Compr
Psychiatry 27:327-335, 1986.
II. McEvoy 1. P., Stiller R. L., Farr R.: Plasma haloperidol levels drawn at neuroleptic threshold doses: A
pilot study. J Clin Psychopharmacol 6: 133-138, 1986.
12. Chouinard G., Jones B. D.: Neuroleptic-induced supersensitivity psychosis: Clinical and pharmacologic
characteristics. Am J Psychiatry 137:16-31, 1980.
13. Ross C. A: Buspirone in the treatment of tardive dyskinesia. Med Hypotheses 22:321-328, 1987.
14. Gardner E. L., Chen J., Paredes W.: Clozapine produces potent antidopaminergic effects anatomically
specific to the mesolimbic system. J Clin Psychiatry 55: 15-22, 1994.
15. Moghaddam B.: Preferential activation of cortical dopamine neurotransmission by clozapine: Functional
significance. J Clin Psychiatry 55:27-29, 1994.
16. Targeting schizophrenia: Pierre Sokoloff on dopamine receptors. Science Watch 5:3-4, 1994.
17. Cohen B. M., Lipinski J. E: In vivo potencies of antipsychotic drugs in blocking alpha-I noradrenergic
and dopamine D-2 receptors: Implications for drug mechanisms of action. Life Sci 39:2571-2580, 1987.
18. Ko G. N., Korpi E. R., Linnoila M.: On the clinical relevance and methods of quantification of plasma
concentrations of neuroleptics. J Clin Psychopharmacol 5:253-262, 1985.
19. Keepers G. A, Clappison V. J., Casey D. E.: Initial anticholinergic prophylaxis for neuroleptic-induced
extrapyramidal syndromes. Arch Gen Psychiatry 40: 1113-1117, 1983.
20. Van Putten T., Mutalipassi L. R., Malkin M. 0.: Phenothiazine-induced decompensation. Arch Gen
Psychiatry 30:102-106, 1974.
21. Ge1enberg A. J., Mandel M. R.: Catatonic reactions to high potency neuroleptic drugs. Arch Gen
Psychiatry 34:947-950, 1977.
210 ALAN J. GELENBERG, M.D., AND SAMUEL KEITH, M.D.
22. Rifkin A .• Quitkin E, Klein D. E: Akinesia: A poorly recognized drug-induced extrapyramidal behavior
disorder. Arch Gen Psychiatry 32:642-674, 1975.
23. Wojcik J. D.: Antiparkinson drug use. Mass Gen Hosp Bioi Ther Psychiatry Newsletter 2:5-7, 1979.
24. Gelenberg A. J., Van Putten T., Lavori P., et al: Anticholinergic effects on memory: Benztropine versus
amantadine. J Clin PsychopharmacoI9:180-185, 1989.
25. Gelenberg A. 1.: Amantadine in the treatment of benztropine-refractory extrapyramidal disorders in-
duced by antipsychotic drugs. Curr Ther Res 23:375-380, 1978.
26. Smith J. M.: Abuse of the antiparkinson drugs: A review of the literature. J Clin Psychiatry 41:351-354,
1980.
27. Jeste D. V., Wyatt R. J.: Changing epidemiology of tardive dyskinesia: An overview. Am J Psychiatry
138:297-309, 1981.
28. Burke R. E., Fahn S., Jankovic J., et al: Tardive dystonia: Late-onset and persistent dystonia caused by
antipsychotic drugs. Neurology 32: 1335-1346, 1982.
29. Morgenstern H., Glazer W. H., Niedzwiecki D., et a1: The impact of neuroleptic medication on tardive
dyskinesia: A meta-analysis of published studies. Am J Public Health 77:717-724, 1987.
30. Tamminga C. A., Thaker G. K., Moran M., Kakigi T., Gao X. M.: C10zapine in tardive dyskinesia:
Observations from human and animal model studies. J Clin Psychiatry 55:102-106, 1994.
31. Gerlach 1., Peacock L.: Motor and mental side effects of c1ozapine. J Clin Psychiatry 55: 107-109, 1994.
32. Wojcik 1. D., Gelenberg A. 1., Labrie R. A., Mieske M.: Prevalence of tardive dyskinesia in an outpatient
population. Comp Psychiatry 21 :370-380, 1980.
33. Kane J. M., Woerner M., Borenstein M., et al: Integrating incidence and prevalence of tardive dyskin-
esia. IVth World Congress of Biological Psychiatry, Philadelphia, Sept. 8-13, 1985 (Abstract).
34. Glazer W. M., Morgenstern H., Doucette 1. T.: Predicting the long-term risk of tardive dyskinesia in
outpatients maintained on neuroleptic medications. J Clin Psychiatry 54: 133-139, 1993.
35. Gardos G., Cole J. 0., Tarsy D.: Withdrawal syndromes associated with antipsychotic drugs. Am J
Psychiatry 135:1321-1324, 1978.
36. Quitkin E, Rifkin A., Gochfeld L., et al: Tardive dyskinesia: Are first signs reversible? Am J Psychiatry
134:84-87, 1977.
37. Chakos M. H., Lieberman J. A., Bilder R. M., et al: Increase in caudate nuclei volumes of first-episode
schizophrenic patients taking antipsychotic drugs. Am J Psychiatry 151:1430-1431, 1994.
38. Keshavan M. S., Bagwell W. w., Haas G. L., Sweeney 1. A., Schooler N. R., Pettegrew J. w.: Changes in
caudate volume with neuroleptic treatment. Lancet 344:1434, 1994.
39. Doraiswamy P. M., Tupler L. A., Krishnan K. R. R.: Neuroleptic treatment and caudate plasticity. Lancet
345:734-735, 1995.
40. Chakos M. H., Lieberman 1. A., A1vir J., Bilder R., Ashtari M.: Caudate nuclei volumes in schizophrenic
patients treated with typical antipsychotics or cIozapine. Lancet 345:456-457, 1995.
41. Dalgalarrondo P., Gattaz W. E: Basal ganglia abnormalities in tardive dyskinesia: Possible relationship
with duration of neuroleptic treatment. Eur Arch Psychiatry Clin Neurosci 244:272-277, 1994.
42. Baldessarini R. 1., Tarsy D.: Tardive dyskinesia, in Lipton M. A., DiMascio A., Killam K. E (eds):
Psychopharmacology: A Generation of Progress. New York, Raven Press, 1978, pp. 993-1004.
43. Casey D. E., Povisen U. 1., Meidahl B., et al: Neuroleptic-induced tardive dyskinesia and parkinsonism:
Changes during several years of continuing treatment. Psychopharmacol Bull 8:250-253, 1986.
44. Yagi G., Itoh H.: A ten-year follow-up study of tardive dyskinesia-with special reference to the
influence of neuroleptic administration on the long-term prognosis. Keio J Med 34:211-219, 1985.
45. Caroff S. N.: The neuroleptic malignant syndrome. J Clin Psychiatry 41:79-83, 1980.
46. Dabiri L. M., Pasta D., Darby J. K., Mosbacher D.: Effectiveness of vitamin E for treatment of long-term
tardive dyskinesia. Am J Psychiatry 151 :925-926, 1994.
47. Shalev A., Munitz H.: The neuroleptic malignant syndrome: Agent and host interaction. Acta Psychiatr
Scand 73:337-347, 1986.
48. Levinson D. E, Simpson G. M.: Neuroleptic-induced extrapyramidal symptoms with fever: Hetero-
geneity of the "neuroleptic malignant syndrome". Arch Gen Psychiatry 43:839-848, 1986.
49. Gelenberg A. 1., Bellinghausen B., Wojcik J. D., Falk W. E., Farhadi A. M.: Patients with NMS histories:
What happens when they are rehospitalized? J Clin Psychiatry 50:178-180, 1989.
4. PSYCHOSES 211
50. Frankenburg E R., Kando J. c., Centorrino E, Gilbert J. M.: Bladder dysfunction associated with
cIozapine therapy. J Clin Psychiatry 57:39-40, 1996.
51. Baldessarini R. J., Gelenberg A J.: Using physostigmine safely. Am J Psychiatry 136: 1608-1609, 1979.
52. Wilt 1. L., Minnema A M., Johnson R. E, Rosenblum A. M.: Torsade de pointes associated with the use
of intravenous haloperidol. Ann Intern Med 119:391-394, 1993.
53. Mitchell A C., Brown K. W: Chlorpromazine-induced retinopathy. Br J Psychiatry 166:822-823, 1995.
54. Potts A M.: Toxic responses of the eye, in Klaassen C. D., Amdur M. 0., Doull J. (eds): Casarett and
Doull's Toxicology; The Basic Science of Poisons, 5th ed.. New York, McGraw-Hili, 1996, pp. 583-
615.
55. Frankenburg E R., Stormberg D., Gerson S. L.: Unsuccessful reexposure to cIozapine. J Clin Psycho-
pharmacol 14:428-429, 1994.
56. Lieberman 1. A, Johns C. A., Kane 1. M., et al: Clozapine-induced agranulocytosis: Non-cross-reactivity
with other psychotropic drugs. J Clin Psychiatry 49:271-277, 1988.
57. Safferman A Z., Lieberman J. A., Alvir J. M. J., Howard A.: Rechallenge in cIozapine-induced
agranulocytosis. Lancet 339: 1296-1297, 1992.
58. Alvir J. M. J., Lieberman 1. A: Agranulocytosis: Incidence and risk factors. J Clin Psychiatry 55:137-
138,1994.
59. Gerson S. L.: G-CSF and the management of cIozapine-induced agranulocytosis. J Clin Psychiatry
55:139-142, 1994.
60. Hummer M., Spemer-Unterweger B., Kemmler G., Falk M., Kurz M., Oberbauer H., Aeischhacker
W. W: Does eosinophilia predict cIozapine-induced neutropenia? Psychopharmacology 124:201-214,
1996. 1992 (Abstract).
61. Banov M. D., Tohen M., Friedberg 1.: High risk of eosinophilia in women treated with cIozapine. J Clin
Psychiatry 54:466-469, 1993.
62. Zarrabi M. H., Zucker S., Miller E, et al: Immunologic and coagulation disorders in chlorpromazine-
treated patients. Ann Intern Med 91:194-199, 1979.
63. Zucker S., Zarrabi H. M., Schuback W H., et al: Chlorpromazine-induced immunopathy: Progressive
increase in serum IgM. Medicine 69:92-100, 1990.
64. Gelenberg A. 1.: Psychotropic drugs during pregnancy and the perinatal period. Mass Gen Hosp Bioi
Ther Psychiatry Newsletter 2:41-42, 1979.
65. Rivera-Calimlim L., Castaneda L., Lasagna L.: Effect of management on plasma chlorpromazine in
psychiatric patients. Clin Pharmacol Ther 14:978-986, 1973.
66. Dysken M. W., Javaid J. I., Chang S. S., et al: Auphenazine pharmacokinetics and therapeutic response.
Psychopharmacology 73:205-210, 1981.
67. Cohen W J., Cohen N. H.: Lithium carbonate, haloperidol, and irreversible brain damage. JAMA
230:1283-1287, 1974.
68. Spring G. K.: Neurotoxicity with combined use of lithium and thioridazine. J Clin Psychiatry 40:135-
138, 1979.
69. Blake L. M., Marks R. C., Luchins D. J.: Reversible neurologic symptoms with cIozapine and lithium. J
Clin Psychopharmacol 12:297-299, 1992.
70. Gelenberg A. 1.: Is it safe to co-prescribe lithium with a neuroleptic? Mass Gen Hosp Bioi Ther
Psychiatry Newsletter 2:13, 1979.
71. Rivera-Calimlim L., Kerzner B., Karch F. E.: Effect of lithium on plasma chlorpromazine level. Clin
Pharmacol Ther 23:451-455, 1978.
72. Finley P., Warner D.: Potential impact of valproic acid therapy on cIozapine disposition. Bioi Psychiatry
36:487-488, 1994.
73. Tiihonen J., Vartiainen H., Hakola P.: Carbarnazepine-induced changes in plasma levels of neuroleptics.
Pharmacopsychiatry 28:26-28, 1995.
74. Longo L. P., Salzman c.: Valproic acid effects on serum concentrations of c10zapine and norcIozapine.
Am J Psychiatry 152:650, 1995.
75. Ge1enberg A J.: Coffee, tea, and antipsychotic drugs revisited. Mass Gen Hosp Bioi Ther Psychiatry
Newsletter 4:42-43, 1981.
76. Gelenberg A. 1.: Psychiatric emergencies: The psychotic patient. Drug Ther 25-36, 1981.
212 ALAN J. GELENBERG, M.D., AND SAMUEL KEITH, M.D.
77. Granacher R. P., Jr.: Management of the acutely psychotic patient. Symposium on Antipsychotic Drug
Therapy: Current Concepts and Future Trends, Key Biscayne, Florida, March 28, 1986 (Abstract).
78. Meltzer H. Y.: Clozapine: Is another view valid? Am J Psychiatry 152:821-825, 1995.
79. Carpenter W. T. J., Conley R. R., Buchanan R. w., Breier A., Tamminga C. A.: Patient response and
resource management: Another view of clozapine treatment of schizophrenia. Am J Psychiatry 152:827-
832, 1995.
80. McElroy S. L., Dessain E. C., Pope H. G., et al: Clozapine in the treatment of psychotic mood disorders,
schizoaffective disorder, and schizophrenia. J Clin Psychiatry 52:411-414, 1991.
81. Suppes T., McElroy S. L., Gilbert J., Dessain E. c., Cole J. 0.: Clozapine in the treatment of dysphoric
mania. Bioi Psychiatry 32:270-280, 1992.
82. Meltzer H. Y., Cola P. A.: The pharmacoeconomics of clozapine: A review. J Clin Psychiatry 55:161-
165, 1994.
83. Kane 1. M.: Schizophrenia. N Engl J Med 334:34-41, 1996.
84. Marder S. R., Putten T., Mintz J., et al: Low- and conventional-dose maintenance therapy with flu-
phenazine decanoate. Arch Gen Psychiatry 44:518-521, 1987.
85. Lieberman J. A., Kane J. M., Johns C. A.: Clozapine: Guidelines for clinical management. J Clin
Psychiatry 50:329-338, 1989.
86. Marder S. R.: Atypical drugs. Psychiatric limes 2, 1994.
87. Gelenberg A. J.: Treating the outpatient schizophrenic. Postgrad Med 64:48-56, 1978.
88. Taylor W. J., Caviness M. H. D.: A Textbook/or the Clinical Application o/Therapeutic Drug Monitor-
ing, Irving, Texas, Abbott Laboratories, 1986, p. 467.
5
Anxiety
ERIC M. REIMAN, M.D.
I. INTRODUCTION
A. Symptoms
Anxiety consists of cognitive, physiological, and behavioral symptoms. Cognitive
symptoms include feelings of apprehension or fear, a sense of uncertainty, helpless-
ness, or danger, and frightening or worrisome thoughts. Physiological symptoms in-
clude respiratory symptoms such as shortness of breath, a smothering sensation, hyper-
ventilation, and numb or tingling paresthesias; cardiac symptoms such as chest
tightness, sensations of heart pounding or racing, and increased heart rate; gastrointesti-
nal symptoms such as nausea, abdominal distress, diarrhea, a lump in the throat, and
swallowing problems; neurological symptoms such as light-headedness and tremulous-
ness; neuromuscular symptoms such as tension and headaches; and other autonomic
sensations such as hot flashes or chills, sweating, dry mouth, and urinary urgency.
Behavioral symptoms include avoidance of, escaping from, or freezing in the anxiety-
provoking situation, restlessness, fidgeting, nail-biting, and impairments in concentra-
tion and performance.
Eric M. Reiman, M.D. • Department of Psychiatry, University of Arizona; Samaritan PET Center, Good
Samaritan Regional Medical Center, Phoenix, Arizona 85006.
The Practitioner's Guide to Psychoactive Drugs (Fourth Edition), Gelenberg and Bassuk, eds.
Plenum Publishing Corporation, New York, 1998
213
214 ERIC M. REIMAN, M.D.
B. Substrates
1. Neuroanatomical Substrates
The brain regions most commonly implicated in the production of anxiety and fear
include limbic areas (the amygdala, hippocampal formation, and septum), which are
anatomically well situated to attach emotional significance to different kinds of sensory
stimuli; paralimbic areas (orbitofrontal cortex, insular cortex, the temporal poles, the
parahippocampal gyrus, and the cingulate complex), which are the main inputs and
outputs of the limbic areas; the hypothalamus, which receives information from limbic
areas and participates in autonomic and endocrine expressions of anxiety; and certain
brainstem nuclei (such as the locus coeruleus, raphe nuclei, barabrachial nuclei, and
medullary chemoreceptors).
Investigating the effects of selective brain lesions in laboratory rats, LeDoux and
his colleagues l have begun to characterize the neuroanatomical structures and path-
ways that are involved in the dissectable components of anxiety. They demonstrated
that the amygdala participates in the evaluation procedure which labels simple sensory
stimuli (such as a pure tone or flashing light) with emotional significance, that a
projection from the amygdala to the hypothalamus participates in certain autonomic
expressions of anxiety, and that a projection from the amygdala to or through the
midbrain participates in certain behavioral expressions of anxiety. Interestingly, lesions
of the relevant primary sensory areas (e.g., lesions of auditory cortex in rats condi-
tioned to fear a tone) had no effects on the animal's fearful response to the sensory
stimulus, perhaps accounting for the observation that individuals with a phobic disor-
der have difficulty appealing to conscious, rational thought to overcome their fear.
Comparing the effects of antianxiety medications and selective brain lesions on a
laboratory rat's hypervigilant response to novel or unpleasant stimuli, Gray 2 postulated
that the septohippocampal system, its incoming noradrenergic projections from the
locus coeruleus, and its serotonergic projections from the raphe nuclei are involved in
forms of anxiety other than conditioned fear. He postulated that antianxiety medica-
5. ANXIETY 215
tions like the benzodiazepines and barbiturates exert their therapeutic effect by inhibi-
ting the noradrenergic and serotonergic projections to the septohippocampal system.
Using depth electrodes in the presurgical evaluation of awake patients with intrac-
table epilepsy, neurosurgeons found that seizure discharge and electrical stimulation of
the amygdala, hippocampal formation, parahippocampal gyrus, and temporal poles are
most commonly associated with the subjective experience of fear.
Positron-emission tomography (PET) studies find that patients with obsessive-
compulsive disorder have abnormally increased activity in the vicinity of the caudate
and orbitofrontal cortex, that these structures, the thalamus, and the anterior cingulate
gyrus are activated by the elicitation of obsessive-compulsive anxiety, and that antiob-
sessional medication and nonmedication treatments both correct the caudate abnor-
mality. These and other studies led investigators to postulate that the caudate is failing
to filter out adventitious, worrisome thoughts in these patients, that this abnormality
leads to an abnormally active circuit involving the caudate, thalamus, and orbitofrontal
cortex, and that medication, psychological, and neurosurgical treatments exert their
therapeutic effect by interrupting this self-perpetuating circuit.
Other PET studies suggest that patients with panic disorder have a functional
abnormality in the vicinity of the right parahippocampal gyrus, that other paralimbic
regions (such as insular cortex) are activated during experimentally generated panic
attacks (and normal anticipatory anxiety), and that anti panic medication does not
correct the parahippocampal abnormality. According to one theory, the regional brain
abnormality is involved in the genetic predisposition to panic attacks; the abnormal
region responds to a normally innocuous triggering event (such as a normal increase in
locus coeruleus activity) by sending a message to paralimbic alarm centers which are
involved in both pathological and normal forms of anxiety; antipanic medications
interfere with the initiation of panic through its effect on the triggering event (e.g., by
reducing noradrenergic activity); and cognitive-behavioral therapy works downstream,
resetting the alarm centers, and interfering with the elaboration of panic.
2. Neurochemical Substrates
Most serotonergic neurons (i.e., those that synthesize, store, and release serotonin)
arise in the raphe nuclei in the midbrain and project to widespread regions of the brain.
Some studies relate anxiety and some of its disorders to increases in serotonergic
activity, others to decreases. This system may have an especially important role in the
treatment of obsessive-compulsive disorder.
GAB A, the major inhibitory neurotransmitter, is located in intemeurons through-
out the central nervous system. Antianxiety medications, such as the benzodiazepines
and barbiturates, appear to reduce anxiety by facilitating GABA neurotransmission.
When GAB A binds to the postsynaptic GABA A receptor, a chloride channel opens,
causing chloride ions to enter, hyperpolarize, and, thus, inhibit the postsynaptic neuron.
Benzodiazepines, barbiturates, and ethanol all change the conformation of the GABA A
receptor, increasing its affinity for GABA. Benzodiazepines do so by binding to and
activating a postsynaptic benzodiazepine receptor, which exists in a supramolecular
receptor complex with the GABA A receptor and a chloride channel; barbiturates do so
by binding to a separate postsynaptic receptor; and eth.anol appears to do so by altering
the membrane around the GABA A receptor. While GABA inhibits neurons throughout
the brain, some researchers postulate that GABA exerts its antianxiety effects by
inhibiting noradrenergic neurons arising in the locus coeruleus, serotonergic neurons
arising in the raphe nuclei, or both.
Additional chemical processes are likely to have a role in anxiety and its disor-
ders. For instance, investigators recently postulated that noradrenergic, opiate, cor-
ticotropin-releasing factor, and other neurochemical systems are involved in the patho-
physiology and potential treatment of posttraumatic stress disorder.
While researchers continue to uncover the chemical processes that are involved in
the development and treatment of anxiety and its disorders, several caveats are in order.
First, medications act on multiple neurotransmitter systems and neuroreceptor sub-
types, many of which remain to be char~cterized. Second, additional neurotransmitter,
neuropeptide, neuroreceptor, intraneuronal, and endocrine processes are likely to have
a role in different aspects of anxiety, anxiety disorders, and their treatment. Third,
researchers are just beginning to characterize the interactions among different neuro-
transmitter systems. Finally, it is difficult, if not impossible, to make valid inferences
about the pathophysiology of psychiatric disorders on the basis of a medication's
therapeutic effects, since pharmacological agents could work by compensating for
rather than correcting the underlying abnormality.
3. Psychological Substrates
blame is neither warranted nor helpful. At the same time, we need to work together to
address the problem in the most constructive way possible. There are medication and
nonmedication treatments that have been established to block panic attacks. Once the
panic attacks are blocked, I will encourage you to confront and learn to overcome those
situations which you have come to fear and avoid. Let's review the options and decide
the best way to proceed at this time .... "
D. Differential Diagnosis
The differential diagnosis of anxiety is summarized in Table 1 and discussed
below.
5. ANXIETY 219
2. Anxiety Disorders
Still, it is important to consider the possibility that the patient's anxiety symptoms
are the direct physiological consequence of a medical condition. Thus, it is helpful to
consider a temporal association between the onset, exacerbation, or remission of non-
psychiatric medical and anxiety symptoms, especially in patients with atypical features
of an anxiety disorder, such as atypical age of onset, symptoms, or clinical course.
Laboratory tests are considered on the basis of the patients' psychiatric and medical
histories, review of symptoms, and relevant physical findings. For instance, findings of
weight loss, heat intolerance, and ocular, dermatological, or hair changes should lead
the practitioner to consider the possibility that the patient has hyperthyroidism.
(the most common fear), a fear of other kinds of public performances (such as playing a
musical instrument), or fears of eating, drinking, writing, or using rest rooms in the
presence of others. Individuals with social phobia, generalized type, have a persistent,
excessive fear of most performance and social situations; in addition to several of the
performance fears noted above, they fear social situations such as speaking in small
groups, dating, calling strangers on the telephone, returning goods to a store, and
attending parties.
It is also important to make the distinction between social phobia and
agoraphobia. In contrast to social phobia, panic disorder with agoraphobia is charac-
terized by at least some panic attacks that occur at unexpected times; panic disorder
with agoraphobia and agoraphobia without a history of panic disorder are not always
limited to those situations involving possible scrutiny by others (e.g., driving, eleva-
tors).
In the feared situation, there often appears to be a vicious cycle of anticipatory
fear of failure or embarrassment, somatic anxiety symptoms, perceived less of control,
actual or perceived performance problems, embarrassment, anticipatory fear, and so
forth. The fear of public speaking is often exacerbated by a fear of heart racing,
trembling hands, and a trembling voice; the fear of eating or drinking in public is often
related to a fear of hand trembling (as the eating utensil reaches the mouth) or choking
on one's food (perhaps related to anxiety-related esophageal motility problems); the
fear of writing checks in public is characteristically related to a fear of hand trembling;
and the fear of using public rest rooms is often related to a fear of urinary hesitancy.
Social phobia appears to affect up to 13% of the population, depending on the
number of social phobic stressors investigated and the threshold used to characterize
clinically significant distress and disability, and is twice as common in females as in
males. The majority of individuals with this disorder have a fear of more than one kind
of performance or social situation. Social phobia typically begins in the mid-teenage
years, lasts for many years, and waxes and wanes in accordance with demands related
to the individual's social, academic, and occupational advancement, such as those
involving dating, oral reports and oral examinations in school, and work-related oral
presentations. The generalized type of social phobia is often characterized by a child-
hood history of social inhibition and shyness and enduring problems related to poor
self-esteem, hypersensitivity to rejection, and difficulty with assertiveness. Indeed,
some individuals with the generalized type of social phobia also satisfy criteria for
avoidant personality disorder. Family, twin, and developmental studies support the
possibility of a genetic contribution to the problem.
(DSM-IV) distinguishes several types of specific phobia by their natural history, psy-
chophysiological characteristics, and familial transmission. Specific phobia, animal
type, is characterized by an excessive fear of certain animals or insects and is more
common in an individual's first-degree relatives. Specific phobia, natural environment
type, is characterized by fear of environmental situations, such as storms, water, and
heights. Specific phobia, blood-injection-injury type, is characterized by a fear of
seeing blood, injuries, injections, or other invasive medical procedures. In contrast to
the other phobias, it is usually characterized by a strong vasovagal response, with
decreased heart rate, decreased blood pressure, and occasional syncope. Specific pho-
bia, situational type, is characterized by a fear of specific situations, such as driving,
tunnels, bridges, elevators, enclosed places, and flying and is more common in an
individual's first-degree relatives. In contrast to agoraphobia, this and other types of
specific phobia are not characterized by a history of unexpected panic attacks; the focus
of fear is the situation itself rather than a panic attack or its consequences. Specific
phobia, other type, is characterized by a fear of other stimuli, such as costumed
characters or loud noises (which sometimes occurs in children), choking, vomiting, or
acquiring an illness. Individuals with one phobia appear more likely to have additional
phobias within the same SUbtype.
28% had rituals involving ordering, symmetry, and precision; and 18% had hoarding
rituals. About half of these patients had multiple compulsions.
OCD is not the same thing as obsessive-compulsive personality disorder, a condi-
tion which is characterized by a disabling preoccupation with details and rules, dis-
abling perfectionism, excessive devotion to work and productivity, inflexible insistence
on certain ethical values, an inability to discard worthless objects, a reluctance to
delegate tasks to others, a miserly spending style, and stubborn insistence that others
conform to one's way of doing things. In one study, only 14% of OCD patients had
obsessive-compulsive personality features.
Less than one generation ago, OCD was considered an extremely rare disorder,
affecting only 1 in 2000 individuals. Individuals with OCD were unlikely to volunteer
these embarrassing, seemingly "crazy," presumably untreatable symptoms. Clinicians
were unlikely to ask about obsessions or compulsions. Since then, studies have consis-
tently demonstrated that OCD is a common disorder, affecting 2.5% of the population.
If the disorder is recognized and properly treated, symptoms can be significantly
reduced. Thus, it is important to suspect the possibility of OCD. Consider using a
checklist (such as the Yale-Brown Obsessive Compulsive Scale checklist) to identify
symptoms that patients have difficulty volunteering.
Although OCD is equally common in males and females, it tends to develop in the
early teens in males and the early twenties in females. About 10% of cases begin prior
to the age of 10, and about 15% begin after the age of 35. In most cases, OCD is a
chronic, waxing and waning, mildly to moderately disabling condition; in about 5% of
cases, it is episodic; and in about 15% of cases, it is a progressively severe, disabling
disorder. About 10% of individuals have OCD with poor insight, in which the individu-
al does not recognize that the obsessions or compulsions are excessive or unreasonable.
This form of the disorder appears to be more progressive, more disabling, and less
responsive to antiobsessive treatments.
Patients with OCD often have additional problems, such as depressed mood,
major depressive disorder, social phobia, eating disorders, alcohol and sedative-hyp-
notic abuse, panic disorder, Tourette's disorder, and personality disorders. OCD pa-
tients can spend much of their day engaged in the battle between anxiety-eliciting
obsessions and anxiety-relieving but disabling compulsions. Others can avoid an in-
creasing number of situations (e.g., some individuals can avoid handshaking, rest
rooms, and all sorts of situations that are perceived to be associated with possible
contamination). Like those with other obsessions, individuals with intrusive, distress-
ing aggressive or sexual urges do not appear to have an increased risk of suicide
attempts, antisocial behaviors, or acting out on their obsessions.
The differential diagnosis of OCD includes basal ganglia disorders, which are
sometimes associated with obsessive-compulsive symptoms; hypochondriasis, which
is associated with recurrent, distressing thoughts about physical symptoms and ill-
nesses; and "compulsive" eating, gambling, sexual activity, and substance use and
other behaviors which are intrinsically pleasurable but which lead to deleterious conse-
quences. (The role of serotonin reuptake inhibitors in the treatment of these disorders is
now being investigated.) In those individuals who have OCD with poor insight, it is
sometimes difficult to distinguish the disorder from schizophrenia.
S. ANXIETY 225
lasts from 1 to 3 months, and is followed by virtually complete recovery. The others
have chronic PTSD, which by definition lasts for at least 3 months but often lasts
much longer. Most individuals with PTSD develop symptoms within the first 3 months
following the traumatic event; those who first develop symptoms at least 6 months
(sometimes years) later have PTSD with delayed onset.
The onset, severity, and duration of PTSD symptoms appear to be related to the
severity and type of catastrophic stressor. For instance, individuals appear to be more
disabled by the PTSD if the stressor is of human design (such as childhood sexual or
physical abuse, a terrorist attack, or torture). Some studies suggest that the onset and
duration of PTSD are adversely affected by childhood onset, limited social supports,
preexisting psychiatric disorders, and certain personality features. Individuals with
PTSD have an increased risk of substance-related disorders, major depression, and
other anxiety disorders, problems which could well affect the choice of medication
treatment. Some investigators suggest that PTSD is less responsive to treatment when it
is chronic or associated with delayed onset.
those with a later age at onset. Individuals with an age at onset greater than 20 appear
more likely to develop the disorder more rapidly, following a stressful event. Further
studies are needed to confirm the distinction between early-onset and late-onset GAD
and their relationships to different treatments.
Before treating a patient for GAD, it is important to look for, diagnose, and treat
comorbid disorders (such as panic disorder, the treatment of which is discussed in
Section V.C).
A. Supportive Therapy
Supportive therapy is an important part of the management plan. It consists of a
patient, kind, and nonjudgmental attitude and an earnest effort to understand the indi-
vidual and his or her problem. It often involves helping individuals consider the
problem in more constructive ways, providing information about the problem and
treatment alternatives, decreasing misplaced feelings of blame, promoting the individu-
als' coping strengths, and helping them reach out to appropriate family members,
friends, or social agencies for support. Supportive therapy can help individuals handle a
stressful situation, increase their understanding of the problem, increase their self-
confidence, self-esteem, and sense of control, decrease their sense of social isolation,
and help them to consider constructive coping alternatives.
228 ERIC M. REIMAN, M.D.
Patients (and in, some instances, couples) can benefit from participation in a
homogeneous support group that includes other individuals with the same problem.
Support groups can share new information about the disorder and its treatments,
decrease feelings of shame or guilt, help individuals confront phobic situations, and
help them cope with everyday stressors. Support groups have been used in the treat-
ment of panic disorder with agoraphobia, social phobia, obsessive-compulsive disor-
der, and posttraumatic stress disorder.
C. Neurosurgical Treatments
When prefrontal leukotomy was introduced by Moniz in 1936 to treat patients
with severe obsessive-compulsive disorder and schizophrenia, it was associated with
significant reductions in anxiety and agitation. Unfortunately, it was also associated
with impoverished expressions of emotion, decreased creativity, an impairment in
goal-directed behavior, and a high incidence of seizures, causing understandable ethical
concerns and eventually limiting its use. With the advent of stereotactic techniques
approximately 20 years ago, neurosurgeons began to study the effects of much more
discrete lesions in these patients. In patients with severe forms of OCD that are
unresponsive to the conventional treatments, these treatments may be helpful.
Used most extensively at Massachusetts General Hospital, bilateral anterior cin-
gulotomy appears to be associated with significant improvement in about half of
patients with intractable OeD, reducing the severity of symptoms by an average of
about 50%. It also appears to be relatively safe, causing seizures (which can be
controlled with medication) in 1% of patients and no measurable adverse effects on
personality. It remains unclear how much of the improvement is related to a reduction
in obsessions or compulsions or, as some suggest, in the anxiety associated with these
symptoms.
Used most extensively at the Karolinska Institute in Sweden, bilateral anterior,
internal capsulotomy appears to be associated with significant improvement in about
three-quarters of patients. While personality changes, including emotional blunting and
disinhibition, were reported in an earlier series, subsequent studies found this proce-
dure to be associated only with lethargy for up to a few months and no long-term
adverse effects. In both of the studies in which anterior capsulotomy was compared to
anterior cingulotomy, anterior capsulotomy proved more effective.
Limbic leukotomy involves lesions of both the anterior cingulate and inferomedial
orbitofrontal cortex. It was associated with improvement in about 85% of patients with
intractable OeD and produced only transient side effects (headache, confusion, and
perseverative behaviors) and no long-term side effects.
These studies are limited by the absence of randomized, sham-controlled trials, an
issue that could be addressed with the advent of stereotactically guided radiotherapy. In
the meantime, these treatments remain an option for individuals with severe, intractable
OeD when the conventional treatments have been exhausted.
IV. PHARMACOTHERAPY
A. Benzodiazepines
1. Introduction
The benzodiazepines, barbiturates, propanediols, antihistamines, and other medi-
cation classes are commonly called anxiolytics or antianxiety medications (reflecting
their ability to decrease anxiety symptoms), sedative-hypnotics (reflecting the ability
to induce drowsiness and promote sleep), or minor tranquilizers (in contrast to the
230 ERIC M. REIMAN, M.D.
2. Mechanism of Action
3. Chemistry
Eight benzodiazepines in the United States are approved for the treatment of
anxiety and some of its disorders; another benzodiazepine, clonazepam, is approved for
the treatment of epilepsy but is widely used for the treatment of panic disorder. The
term "benzodiazepine" refers to a structural component that all of the drugs in the class
have in common-a benzene ring fused to a seven-membered diazepine ring.
232 ERIC M. REIMAN, M.D.
5. Pharmacokinetics
In order to determine the choice, dose, and schedule of a particular ben-
zodiazepine, it is helpful to consider its pharmacokinetic properties, including its rate
of onset, distribution half-life, elimination half-life, potency, and alternative routes of
administration (Table 2).
a. Onset. The peak onset and intensity of a benzodiazepine's clinical effects are
directly related to its absorption in the gastrointestinal system. Gastrointestinal absorp-
tion is accelerated by taking the medication on an empty stomach and delayed by
anything that retards gastric emptying, such as food and aluminum-containing antacids.
Orally administered benzodiazepines vary widely in their onset rate (Table 2).
Diazepam is the fastest-acting benzodiazepine available, reaching a peak plasma con-
centration within an hour. In comparison to an equivalent dose of another ben-
zodiazepine, a single dose of diazepam is associated with relatively prompt and intense
sedative effects; it is sometimes associated with reports of elation or dysphoria. Praz-
epam, which is slowly absorbed and slowly transformed into an active metabolite, is
the slowest-acting benzodiazepine available, reaching a peak plasma concentration in
about 6 hr. In comparison to an equivalent dose of another benzodiazepine, a single
dose of prazepam is associated with delayed and attenuated sedative effects.
A relatively fast-acting medication is indicated in those individual with a ben-
zodiazepine-responsive problem who require a prompt sedative effect, such as those
treated for initial insomnia (difficulty falling asleep), extreme situational anxiety, cer-
tain psychiatric and nonpsychiatric medical disorders that are associated with severe
agitation, and rapidly emerging withdrawal symptoms. (Since panic attacks are char-
acteristically brief, even the fastest-acting benzodiazepine is unlikely to be helpful
when used "as needed.") A relatively slow-acting benzodiazepine is indicated in
those individuals with a benzodiazepine-responsive problem who could benefit from
delayed or attenuated sedative effects, such as those who experience transient over-
sedation, elation, or dysphoria on a fast-acting benzodiazepine and those in whom the
clinician does not want to reinforce the habit of taking medication in times of need.
(In individuals with a history of alcohol, sedative-hypnotic, or other substance de-
pendence, it is wise to avoid benzodiazepines altogether, even those with a delayed
onset.) Finally, the rate of onset can affect the timing of benzodiazepine administra-
tion in those with situational anxiety or initial insomnia. For instance, it is helpful to
know that lorazepam exerts its maximal effect approximately 2 hr after it is taken
orally in order to know when to prescribe it in relationship to an upcoming airplane
trip in those with a fear of flying.
TABLE 2. Benzodiazepines Used in the Treatment of Anxiety or Its Disorders
Other routes
Generic name Equivalent Onset!peak Distribution Elimination
(brand name) dose (mg) time time half-life SL 1M IV Dosage forms
Alprazolam (Xanax<R>, Intermediate! Intermediate Short X 0.25-,0.5-, and I-mg scored tablets
others) 1-2 hr (6-27 hr)
Chlordiazepoxide 20 Intermediate! Slow Long (30-200 hI"") 5-, 10-, and 25-mg tablets and capsules
(Librium<R>, others) 0.5-4 hr
Clonazepam 0.5 Intermediate! Intermediate Long (18-50 hr) 0.5-, 1-, and 2-mg scored tablets
(KJonopin<R>, others) 1-2 hr
Clorazepate 15 fastll-2 hr Fast Long (30-200 hI"") 3.75-, 7.5-, and 15-mg scored tablets
(Tranxene<R>, others)
Diazepam (Valium<R>, 10 Very fast! Very fast Long (30-200 hr") X X 2-, 5-, and 100mg scored tablets;
others) 0.5-2 hr ampules, syringes, vials
Halazepam (Paxipam<R» 40 Slowll-3 hr Intermediate Long (30-200 hI"") 20- and 40-mg scored tablets
Lorazepam (Ativan<R>, 2 Intermediate! Intermediate Short (10-20 hr) X X X 0.5-, 1-, and 2-mg tablets; syringes,
others) 1-6 hr vials
Oxazepam (Serax<R>, 15 Slowl2-4 hr Intermediate Short 10-, 15-, and 30-mg capsules; 15-mg
others) (3-21 hr) tablets
Prazepam (Centrax<R» 10 Very slow!6 hr Intermediate Long (30-200 hI"") 5- and IO-mg capsules; IO-mg tablets
6. Adverse Effects
b. Uncommon Effects. Additional side effects are uncommon and only rarely
serious. These include increased appetite and weight, allergic skin rashes, headache,
and dizziness. Benzodiazepines have minimal respiratory depressant effects when used
orally, in anxiolytic or hypnotic doses, and in healthy patients. They have more serious
respiratory depressant effects when administered intravenously, in higher doses, and in
some patients with preexisting problems, such as chronic obstructive pulmonary dis-
ease or sleep apnea. Clinicians should be prepared to provide respiratory support for
5. ANXIETY 239
those who take a benzodiazepine overdose in conjunction with other respiratory de-
pressants and those who receive a benzodiazepine intravenously.
reflect a return of the original symptoms; more severe withdrawal can also be associ-
ated with symptoms like auditory, visual, or tactile hallucinations and generalized
seizures. Withdrawal symptoms typically develop within the first 2 days following the
discontinuation of a benzodiazepine that has a short elimination half-life and within the
first 7 -10 days following that of a benzodiazepine that has a long-elimination half-life;
they usually subside within 2-3 weeks following benzodiazepine discontinuation.
The risk of benzodiazepine withdrawal symptoms appears to be increased with
longer use (at least 8-12 weeks), higher doses, higher-potency drugs, drugs having a
shorter elimination half-life, and more abrupt discontinuation. In order to avoid severe
withdrawal symptoms, the medication should be discontinued slowly; patients should
be informed not to suddenly discontinue their medication.
During the past decade, clinicians have had particular difficulty tapering and
discontinuing alprazolam without the development of rebound or withdrawal symp-
toms. Reasons include its widespread use at relatively high doses (which are indicated
for panic disorder), its high potency, and its short elimination half-life. This problem
can be minimized by restricting the long-term use of alprazolam to patients with panic
disorder (and, even in these patients, considering alternative treatments), informing
patients about the risk of sudden discontinuation, and forewarning them about the self-
limited discomfort that is frequently associated with its gradual discontinuation. Begin
by tapering the medication slowly (e.g., decreasing the daily dose by 0.25-0.5 mg
every 5 days until it is the down to 2 mg and then decreasing the daily dose by 0.125-
0.25 mg every 5 days).
In those patients who have difficulty tolerating a slow taper without rebound or
withdrawal symptoms, consider switching the remaining alprazolam dose with an
equipotent dose of a benzodiazepine that has a long elimination half-life. For instance,
replace every 1 mg of alprazolam with 0.5 mg of clonazepam or 25 mg of chlor-
diazepoxide. The patient should continue to take alprazolam up to the previous dose as
needed/or one week only (sufficient time to saturate peripheral tissues with the long-
elimination-half-life benzodiazepine). Like alprazolam, clonazepam is effective in the
treatment of panic disorder, requires fewer daily doses, is associated with less interdose
anxiety (symptoms which sometimes occur between doses when plasma concentrations
reach their trough), and may be associated with less severe withdrawal symptoms. In
comparison to clonazepam substitution, chlordiazepoxide substitution has not been
well studied, but, as a low-potency benzodiazepine, chlordiazepoxide may be easier to
taper and discontinue with minimal discontinuation symptoms. While clonidine and
carbamazepine have also been used to ameliorate withdrawal symptoms during ben-
zodiazepine discontinuation, their efficacy appears to be limited.
have been attributable to medical problems or medications other than the ben-
zodiazepines themselves). Like other medications, benzodiazepines should be avoided
during pregnancy, especially during the first trimester, unless there is a compelling
reason to administer a benzodiazepine.
Persistent benzodiazepine doses in late pregnancy are occasionally associated
with a withdrawal syndrome in the newborn infant. High doses just prior to or during
labor are occasionally associated with hypotonia ("a floppy infant syndrome"), hypo-
thermia, mild respiratory depression, and a poor sucking response in the newborn
infant. Like many other medications, benzodiazepines are secreted in breast milk; thus,
mothers should avoid breast feeding if they are taking this medication.
7. Drug Interactions
8. Laboratory Tests
Prior to treatment, laboratory tests are sometimes indicated to rule out an anxiety
disorder due to a generalized medical condition (such as hyperthyroidism) and, less
commonly, to characterize hepatic or renal impairments that could be associated with
decreased benzodiazepine elimination. During treatment, benzodiazepine doses are
titrated according to their therapeutic and adverse effects rather than plasma levels. In
patients with a suspected benzodiazepine overdose, a urine drug toxicity screen is
indicated to confirm the presence of benzodiazepines and help identify other ingested
substances.
B. Other Sedative-Hypnotics
Although the antihistamines are less effective and more sedating than the ben-
zodiazepines in the treatment of anxiety, they do not appear to be associated with abuse
or withdrawal symptoms. Although their role in the treatment of anxiety and its
disorders has not been well studied, some clinicians use these medications for the acute
relief of anxiety symptoms in patients who are considered to be at risk for substance
abuse or dependence. The antihistamines appear to exert their sedative, hypnotic,
antiemetic (e.g., anti-motion-sickness), and antipruritic effects through the blockade of
central and peripheral histamine HI receptors. In addition to their sedative effects, the
antihistamines can cause mild anticholinergic effects (including potential cognitive
impairment and toxicity in the elderly); they appear to be associated with a slight
increase in the risk of seizures, occasional agitation, and occasional reports of abnor-
mal involuntary movements; and they can potentiate the effects of other central ner-
vous system depressant, anticholinergic, or antihistaminic substances. Because it is not
possible to rule out their teratogenic effects and it remains to be established whether or
not they are secreted in breast milk, antihistamines should be used sparingly in women
who are pregnant or nursing. When clinicians choose to use an antihistamine in the
treatment of anxiety, they most commonly use hydroxyzine (Atarax®, Vistaril® and
others), 50-100 mg orally up to 4 times per day. It is rapidly absorbed, beginning to act
within 15-30 minutes, reaching peak plasma concentrations within 1 hr, and maintain-
ing its effect for at least 24 hr. It remains unclear how hydroxyzine is metabolized.
Although barbiturates such as phenobarbital appear to have acute anxiolytic
effects, they are not indicated for the treatment of anxiety or its disorders. In compari-
son to benzodiazepines, the barbiturates appear to be less effective and more sedating,
more likely to cause abuse, tolerance, and severe withdrawal symptoms, more likely to
produce pharmacokinetic drug interactions (such as a potentiation of oral anticoagulant
effects), and more likely to cause respiratory depression, cardiac contractility reduc-
tions, and death following an overdose. Moderate intoxication is typically associated
with sedation, slurred speech, ataxia, and nystagmus; severe intoxication can also be
associated with coma, respiratory depression, decreased cardiac contractility, and
death. As little as 2 -3 g of a short-acting barbiturate, such as amobarbital, secobarbi-
tal, or pentobarbital, and as little as 6-10 g of a long-acting barbiturate, such as
phenobarbital, can be lethal. Even lower doses can be lethal if taken in conjunction
5. ANXIETY 243
with alcohol, antihistamines, or other central nervous system depressants. The barbitu-
rates are metabolized by, and induce the oxidizing effects of, hepatic microsomal
enzymes. They are contraindicated in patients with acute intermittent porphyria.
Although two propanediols, meprobamate (Miltown®, Equagesic®, Equanil®,
and others) and tybamate (no longer available in the United States), were at one time
commonly prescribed as anxiolytics, they are not indicated for the treatment of anxiety
or its disorders. In comparison to the benzodiazepines, they appear to be less effective
and are more sedating, more likely to cause abuse, tolerance, and severe withdrawal
symptoms, more likely to produce pharmacokinetic drug interactions, more likely to
cause respiratory depression, cardiac contractility reductions, and death following an
overdose, and contraindicated in patients with acute intermittent porphyria. Signs,
symptoms, and complications of propanediol intoxication are similar to those associ-
ated with barbiturate intoxication. Like the barbiturates, the propanediols are metabo-
lized by, and induce the oxidizing effects of, hepatic microsomal enzymes.
c. Buspirone
Buspirone (BuSpar®) is an azospirodecanedione, an anxiolytic that is chemically
distinct from the benzodiazepines. Since it lacks the sedative, anticonvulsant, and
muscle relaxant effects of benzodiazepines, it has been marketed as an "anxioselective"
agent. In contrast to the benzodiazepines, which facilitate the inhibitory effects of
GABA throughout the brain and, in doing so, decrease noradrenergic, doparninergic,
and serotonergic activity, buspirone does not exert an effect on GABA, actually causes
a modest increase in noradrenergic activity (which may account for its ability to
preserve and perhaps modestly increase attention, memory, and psychomotor perfor-
mance), increases doparninergic activity through its effects on presynaptic dopamine
D2 receptors, and at least acutely decreases serotonergic activity through its agonist
effects on the presynaptic serotonin IA autoreceptor.
Although researchers postulate that buspirone exerts its therapeutic effect through
its actions on the serotonergic system, its actions on this system remain to be fully
characterized. Acutely, it appears to be a full agonist of the presynaptic serotonin lA
receptor, decreasing the activity of serotonergic neurons arising in the raphe nuclei, and
a partial agonist of the postsynaptic serotonin lA receptor, exerting an effect on limbic
and cortical neurons in a manner that depends on synaptic serotonin concentrations. In
the presence of low serotonin concentrations, buspirone would act as a postsynaptic
serotonin lA receptor agonist; in the presence high serotonin concentrations, it would
act as a postsynaptic serotonin lA antagonist. Researchers have not yet accounted for
the temporal discrepancy between the medication's acute effects on the serotonergic
system and its delayed clinical effects. Thus, buspirone's adaptive neurochemical ef-
fects, such as the desensitization of serotonin lA and serotonin 2 receptors, merit
additional study.
In comparison to the benzodiazepines and other sedative-hypnotics, buspirone
has several advantages. It does not cause impairments in arousal, attention, memory,
reaction time, or driving, does not lead to abuse, tolerance, or withdrawal, and does not
potentiate the effects of alcohol or sedative-hypnotics. Indeed, it appears remarkably
244 ERIC M. REIMAN, M.D.
free from sedation, other adverse effects, or toxicity. Relatively infrequent side effects
include a sUbjective sense of dizziness (15%), headache (6%), nausea (5%), nervous-
ness (4%), and insomnia (2%).
Studies suggest that buspirone is as effective as benzodiazepines in the treatment
of generalized anxiety disorder. Findings suggest that buspirone (and antidepressants)
may be particularly effective for subjective symptoms of anxiety such as irritability and
worry, whereas the benzodiazepines may be particularly effective for somatic symp-
toms of anxiety such as muscle tension. However, buspirone takes 2-4 weeks to work,
appears less effective in individuals with a prior history of benzodiazepine exposure
(perhaps related to unrealistic expectations about immediate relief and sedation), and is
associated with a surprisingly high percentage of individuals who express dislike of the
medication and discontinue treatment even when it appears to reduce anxiety symp-
toms.
Based on their anecdotal experiences, it is not uncommon for clinicians to de-
scribe buspirone as a medication that has all of the desirable effects of an anxiolytic-
except for efficacy! Explanations for this disappointment might include a failure to
educate patients about the nature, time course, and selectivity of buspirone's clinical
effects; a failure to appreciate its particular benefit in benzodiazepine-naive patients;
and the limited opportunities for psychiatrists to treat patients who have generalized
anxiety disorder in the absence of comorbid disorders.
Buspirone still may be the medication of choice for the treatment of generalized
anxiety disorder (and some other persistently anxious patients), especially in individu-
als who are elderly or cognitively impaired or have a history of an alcohol, sedative-
hypnotic, or other substance use disorder.
Although buspirone does not lead to withdrawal symptoms by itself, its potentia-
tion of the noradrenergic system could potentiate the rebound increase in noradrenergic
activity and occasionally exacerbate associated withdrawal symptoms that follow ben-
zodiazepine discontinuation. Thus, it is prudent to delay the initiation of bus pirone until
after benzodiazepine "detoxification" is complete.
The role of bus pirone in the treatment of disorders other than generalized anxiety
disorder remains unclear. It does not appear to be effective in the treatment of panic
disorder and, thus, should not be used as a primary treatment for this condition.
Although one controlled study suggested that buspirone was as effective as
clomipramine in the treatment of OCD, this finding is not yet supported by clinical
experience or additional controlled studies. While uncontrolled studies suggested that
buspirone, 20 mg t.i.d., might potentiate the effects of a serotonin reuptake inhibitor in
patients with OCD, controlled studies have failed to support its role as an adjuvant
agent. Case reports and one controlled study suggests that buspirone may be helpful in
the management of patients with dementia, autism, and head injuries who have persis-
tent symptoms of anxiety, restlessness, irritability, or assaultive behaviors. Although
some reports suggest that buspirone (in the range of 20 mg t.i.d.) may be helpful in the
treatment of major depressive disorder, this medication is not recommended for this
disorder until trials of better established treatments have been completed. Case reports
which suggest that buspirone may potentiate antidepressants' therapeutic effects await
confirmation in controlled studies.
5. ANXIETY 245
D. ~-Blockers
sures to circumscribed phobic situations, the duration of action, and thus the number of
daily doses, depends on the f3-blocker's elimination half-life. Based on the drug's
elimination half-life, and its longer duration of f3-blocking activity, propranolol should
be taken 2-4 times each day; atenolol, metoprolol, and the slow-release form of
propranolol should be taken 1 or 2 times each day, and nadolol should be taken once each
day. (Still, it should be recognized that the efficacy of standing doses has not been as well
studied as that of as-needed doses in the treatment of circumscribed social phobia.)
Propranolol and metoprolol are eliminated primarily by hepatic metabolism whereas
atenolol and nadolol are eliminated primarily by renal excretion. The elimination half-
lives of these medications are often increased in the elderly.
In addition to their f3-blocking activity, pindolol, propranolol, and perhaps the
other f3-blockers appear to antagonize the presynaptic serotonin IA receptor. The
serotonin reuptake inhibitors lead to a down-regulation of these receptors and, thus, a
temporally delayed potentiation in serotonergic neurotransmission. In a recent open-
label study, pindolol augmentation led some SSRI-treated patients with major depres-
sive disorder to respond more rapidly and some previously SSRI-refractory patients to
respond at all. Although this study needs to be confirmed and extended to other
disorders, it raises the possibility that pindolol, propranolol, and more selective and
potent serotonin IA receptor antagonists might have an adjunctive role in anxiety
disorders such as OeD.
f3-Blockers are indicated for other medical problems including angina, hyperten-
sion, essential or lithium-induced tremors, migraine headaches, and possibly intermit-
tent explosive disorder. They should be used with caution in patients with cardiac
failure, first-degree or greater heart block, or preexisting bradycardia. In patients with
bronchospastic disorders such as asthma, even the relatively cardioselective f3-blockers
should be used with caution. In patients with insulin-dependent diabetes mellitus,
f3-blockers can mask tachycardia symptoms of hypoglycemia, and the nonselective
f3-blockers can delay recovery of serum glucose to normal levels. f3-Blockers are
sometimes associated with central side effects (less likely with the minimally lipophilic
agents) such as lethargy, fatigue, dysphoria, and depression; exacerbation of asthma
and insulin-induced hypoglycemia (less likely with the f31-selective agents); cardio-
vascular side effects such as hypotension, bradycardia, dizziness, and exacerbation of
heart failure; other problems such as nausea and erectile dysfunction; and, upon discon-
tinuation, an exacerbation in symptoms of preexisting angina and thyrotoxicosis.
tion is greater for patients with panic disorder than for those with other anxiety or mood
disorders.) Patients should be advised of this possible side effect, its usually self-
limited nature, and the clinician's efforts to minimize this effect by starting with
extremely low daily doses and increasing the medication slowly.
Phenelzine and, perhaps, the other monoamine oxidase (MAO) inhibitors appear
to be at least as effective in the treatment of panic disorder as the other medications and
slightly more effective in treating the social phobic/agoraphobic complications of this
disorder. Owing to the risks of a hypertensive crisis, a hypermetabolic crisis, and other
drug interactions, patients must agree to comply with the dietary and medication
restrictions described in Chapter 2. Despite this limitation, the MAO inhibitors are
characteristically safe, effective, and underutilized for this and other disorders for
which they have an established role.
Reversible inhibitors of monoamine oxidase A (RIMAs) like brofaromine and
moclobemide are not yet available in the United States. They lack the pressor effects of
nonselective, irreversible MAO inhibitors and, thus, do not require dietary restrictions.
Studies suggest that brofaromine is as effective as other antidepressants in the treat-
ment of panic disorder. Its side-effect profile resembles that of the serotonin-selective
reuptake inhibitors.
The antidepressants and MAO inhibitors appear to block panic attacks completely
in the majority of patients, substantially reduce the severity and frequency of attacks in
many additional patients, and help patients confront and learn to overcome their
agoraphobic situations.
The serotonin reuptake inhibitors are distinguished from the other antidepressants
by their established efficacy in the treatment of obsessive-compulsive disorder. These
medications differ from the other antidepressants by their potent inhibition of serotonin
reuptake in the presynaptic neuronal terminal (an acute effect which precedes clinical
improvement) and a subsequent potentiation of serotonergic neurotransmission that
involves a presynaptic mechanism (down-regulation of serotonin lA autoreceptors).
These medications include the nonselective serotonin reuptake inhibitor clomipramine
and the serotonin-selective reuptake inhibitors (SSRIs) fluoxetine, fluvoxamine, ser-
traline, and paroxetine. These medications typically reduce OCD symptom severity by
about 50% and help about 60-80% of patients. Although direct comparisons are
currently lacking, a metaanalysis of the published treatment studies suggests that
clomipramine is significantly more effective than the SSRIs, that clomipramine was not
associated with more dropouts despite its association with more side effects, and that
all the SSRIs are comparable in their efficacy. The management of OCD with serotonin
reuptake inhibitors requires a higher dose and longer duration of treatment than are
needed to treat major depression. Even though some studies suggest that lower doses
may be equally effective, a therapeutic trial typically involves the use of higher doses
because some patients respond preferentially to higher doses and most patients experi-
ence incomplete improvement. Indeed, patients on paroxetine appear to require a
minimum of 40 mg each day. Thus, a therapeutic trial typically consists of the highest
tolerable dose (up to the maximum recommended) of a serotonin reuptake inhibitor for
at least 8-12 weeks. Later, the medication can be tapered to determine the lowest
effective dose.
248 ERIC M. REIMAN, M.D.
v. CLINICAL CONSIDERATIONS
A. General Principles
The management of anxiety and its disorders includes the following general steps:
1. Identify the underlying problem. Are the patient's symptoms related to a psy-
chosocial stressor, a nonpsychiatric medical disorder, medications or their discontinua-
tion, an anxiety disorder, another psychiatric disorder, or some combination of these
problems?
2. Consider relevant psychosocial factors. Consider the patient's understanding of
his or her problem and treatment options, the patient's coping strengths and vul-
nerabilities, the impact of the problem on the patient's family and his or her education-
al, work, and social activities, how social circumstances contribute to the problem or
the patient's ability to cope with it, and the available support services.
3. Consider relevant biological factors, such as previous medication trials (and
their dose, duration, adverse effects, and benefits), contraindications to a particular
medication, and factors such as aging which can influence the choice and dose of a
medication.
5. ANXIETY 249
4. Quantify the target symptoms and their severity in order to determine the
treatment options and the extent of improvement following treatment.
5. Provide information about your understanding of the problem, the treatment
options, and their anticipated benefits and risks. Correct misconceptions about the
problem, its treatments, and their implications for the patient. Help the patient make an
informed choice about the best treatment for him or her.
6. If a medication is indicated, provide information about its use, including the
anticipated benefits, its role in the overall management plan, common side effects, less
common but potentially important hazards (such as the effects of a medication on
driving or the developing fetus), potentially important drug interactions (such as the
synergistic effects of sedative-hypnotics and alcohol on driving impairment), the med-
ication dose, schedule, and the time required for it to work, and the anticipated duration
of treatment.
7. Ask the patient if he or she has any questions or concerns. Encourage the patient
to raise questions or concerns at any time in the future.
8. Provide an adequate therapeutic trial. If a medication is indicated, increase it to
the optimal dose and continue it for an adequate duration before it is considered
ineffective. If a nonmedication treatment such as cognitive-behavioral therapy is indi-
cated, it should conform as closely as possible to the established protocol before it is
considered ineffective.
9. Following a therapeutic trial, discontinue treatments that are not effective. In
order to minimize the long -term use of unnecessary medication, it is helpful to quantify
baseline symptoms, monitor these symptoms, use one treatment at a time, pay attention
to other factors (such as information and support) that might influence the patient's
symptoms, and provide continuity of care. (In some instances, an adjunctive treatment
is indicated prior to the medication's discontinuation.)
10. Slowly taper and attempt to discontinue the medication if it is no longer
needed. The duration of treatment depends on the underlying problem. Patients with an
adjustment disorder should be prepared to discontinue a benzodiazepine within a few
days or weeks. Patients with certain anxiety disorder, such as OeD, should be prepared
for the possibility of lifelong treatment (or a nonmedication alternative which might be
associated with a lower relapse rate).
The following sections provide practical information about the treatment of spe-
cific problems. A summary of established treatments is presented in Table 3.
B. Adjustment Disorder
Patients who develop intolerable or disabling anxiety symptoms in response to a
stressful situation (including certain medical problems or surgical interventions) can
benefit from the crisis intervention and supportive approaches described in Section
II.D.I. The practitioner should emphatically listen to the patient, understand what the
problem means to him or her, and help the patient explore the options for understanding
and coping with the situation in the most constructive manner.
Some patients benefit from an adjunctive benzodiazepine to help them tolerate the
anxiety symptoms and cope with the problem in a more effective way. These patients
250 ERIC M. REIMAN, M.D.
should be informed that the medication is only a part of the overall management plan,
that it is designed to help mobilize other coping activities, and that the duration of
treatment will be limited in accordance with the stressful situation.
Among the benzodiazepines, lorazepam is a compound having a relatively short
elimination half-life with several attractive features for the treatment of situational
anxiety: a relatively long duration of action (several hours) following the initial doses
due to its distribution half-life; minimal accumulation or hangover effects (especially in
5. ANXIETY 251
the elderly or those with hepatic impairment) due to its single-step metabolism and
short elimination half-life; and the availability of sublingual, intramuscular and intra-
venous administration routes for patients who cannot consume medication orally or
need a more prompt effect. For situational anxiety, lorazepam is usually initiated at a
dose of 0.5-2 mg and increased by that amount as needed as frequently as every 2 hr; a
therapeutic dose is typically less than 6 mg per day, rarely more than 10 mg per day,
taken as needed or in standing doses, and in the latter case the medication is divided
into 2 or 4 doses each day.
Alternatively, chlordiazepoxide is benzodiazepine having a relatively long elim-
ination half-life with attractive features for the treatment of situational anxiety: an even
longer duration of action following the initial doses (due to its long distribution half-
life) and the option of single daily doses after several days (due to its long elimination
half-life). In contrast to lorazepam, it is more likely to accumulate with repeated doses
(e.g., in elderly patients) but less likely to produce withdrawal symptoms following its
discontinuation in the relatively few patients who are treated for several weeks or
months. For situational anxiety, chlordiazepoxide is usually initiated at a dose of 5-25
mg and increased by that amount as needed as frequently as every 2 hr; a therapeutic
dose is rarely more than 100 mg per day, is taken as needed or in standing doses, and in
the latter case the medication is divided into 1-4 doses each day. If a medication with a
long elimination half-life is used in an elderly or hepatic ally impaired patient, use
lower doses and monitor the patient to minimize accumulation effects.
c. Panic Disorder
Patients should be informed that panic disorder is common (i.e., no one with the
illness suffers alone) and that panic attacks are not lethal (an idea that many patients
find difficult to believe). Information about the biological contributions to the develop-
ment and treatment of panic disorder can help allay misplaced feelings of blame.
Information about the influence of nongenetic biological and psychosocial factors can
help the patient accept the ideas that there are both medication and nonmedication
alternatives to the treatment of panic disorder and that when a medication is used, it just
one part of the overall management plan. Information about the established treatments,
their anticipated benefits, side effects (the initial stimulating effect of an antidepressant,
which usually subsides with continued treatment, and the withdrawal symptoms that
can follow discontinuation of a benzodiazepine), and inconvenience can help patients
make an informed decision about the treatment that is best for them.
The established medication treatments of panic disorder include certain anti-
depressants, MAO inhibitors, the RIMA brofaromine, and the high-potency ben-
zodiazepines. These medications appear to block panic attacks completely in over 50%
of patients and reduce the frequency and severity of attacks in about 90% of patients.
Advantages of treatment with these medicaHons include their well-established efficacy,
their availability, their relative ease of use, and the ability of some of these medications
to treat coexisting disorders. Disadvantages include their side effects and the high
relapse rate following medication discontinuation, thus causing some patients to re-
quire lifelong treatment.
252 ERIC M. REIMAN, M.D.
include the small risk of abuse (which can be minimized by reserving them for individ-
uals without a history of alcohol or another psychoactive substance use disorder), the
need for a very gradual taper (among other things, complicating or prolonging
the switch to another medication if the benzodiazepine proves to be suboptimal), and
the likelihood of some uncomfortable rebound or withdrawal symptoms following even
gradual discontinuation.
Effective treatment consists of standing rather than pm doses. Thus, the aim of
treatment is to prevent panic attacks, not to terminate them; it is helpful to recognize
that panic attacks typically end before a pm benzodiazepine is even absorbed. Effective
treatment also consists of relatively high benzodiazepine doses, considerably higher
than that with which many clinicians are familiar. Thus, for instance, the average dose
used in the cross-national study of alprazolam was 5.6 mg per day. As the dose is
increased, patients typically develop tolerance to the medication's sedating effects but
not to its anti panic effects.
Alprazolam (starting dose typically 0.25 mg q.i.d., effective dose 0.5-2.0 mg
q.i.d.) is the best-studied antipanic medication. However, it has several disadvanta.ges,
including the potential for interdose anxiety (which can be minimized by using rela-
tively high and frequent doses), the requirement of multiple daily doses, and apparently
more severe rebound and withdrawal symptoms than with benzodiazepines having a
longer elimination half-life or lower potency.
Although clonazepam is less well studied than alprazolam, the advantages of this
long-elimination-half-life medication include fewer daily doses, less interdose anxiety,
and, at least theoretically, less severe discontinuation symptoms. Although some stud-
ies have raised the possibility that clonazepam is associated with the emergence of
depressive symptoms, these symptoms are relatively uncommon (occurring in less than
10% of patients) and potentially responsive to a reduction in dose or the addition of an
antidepressant; the relative risk of depressive symptoms on clonazepam and other
benzodiazepines remains to be clarified. About twice as potent as alprazolam, the
initial dose of clonazepam is usually 0.25-0.5 mg b.i.d., and the effective dose is
usually 1-4 mg b.i.d. Like the other benzodiazepines, this medication must be tapered
gradually in order to minimize the risk and severity of rebound and withdrawal symp-
toms.
While lorazepam appears to be effective in the treatment of panic disorder, it is
less well studied than alprazolam and does not appear to offer any pharmacokinetic
advantages. About half as potent as alprazolam, the initial dose of lorazepam is typ-
ically 0.5 mg q.i.d., and the effective dose is usually 1-4 mg q.i.d.
When a medication proves to be effective in the treatment of panic disorder, it can
be continued for 6 months to 2 years (depending in part on its role in helping
agoraphobic patients confront their agoraphobic situations with increased confidence
and their progress in overcoming this complication). Periodically, the medication
should be tapered and, if possible, discontinued in order to determine the lowest
effective dose and whether or not additional medication is required. At the same time,
relapse (which should be distinguished from benzodiazepine rebound and withdrawal
symptoms) appears to occur in 50-90% of patients. Since the medications appear to
retain their antipanic efficacy, are usually well tolerated, and do not appear to be
254 ERIC M. REIMAN, M.D.
associated with progressive or irreversible side effects, patients should be prepared for
the possibility of chronic medication treatment.
The best-established psychological treatment of panic disorder is cognitive-be-
havioral therapy (CBT). Although an ongoing multicenter study comparing CBT, im-
ipramine, placebo, and combination treatment promises to shed more light on the role
of CBT in the treatment of panic disorder, published studies suggest that it may be as
effective as anti panic medications. If preliminary studies are supported, CBT offers
several advantages, including freedom from medication side effects, its ability to help
patients tolerate any residual attacks, and an apparently low relapse rate. Like other
psychological treatments, it has several disadvantages, including fewer well-controlled
studies to support its efficacy, the limited availability of properly trained and experi-
enced therapists, the need for highly motivated patients, its inability to treat comorbid
disorders like depression and OCD without additional interventions, and the possibility
that some patients are simply relabeling their symptoms rather than experiencing
symptomatic relief.
Presently, the best-studied and most widely available form of CBT available is the
highly structured, manual-driven treatment offered by Barlow and his colleagues. 3 It
typically involves 12 weeks and 15 sessions of treatment, numerous therapist-adher-
ence guidelines, and lots of homework assignments. Components include education
about panic disorder, the normally adaptive role of anxiety, and treatment rationale;
identification and correction of cognitive distortions; breathing retraining and muscular
relaxation to promote a sense of control over somatic sensations; experimental and
situational exposure to fear-generating physical sensations; and exposure to panic-
eliciting and agoraphobic situations. Patients are encouraged to become observers
rather than victims of their attacks. They are instructed to keep detailed daily records.
The frequency, duration, and ultimate severity of exposure to fear-eliciting sensations
and situations is thought to be extremely important. Although it is unlikely that all of
these components are crucial for the treatment of panic disorder, it is wise to adhere as
closely as possible to the protocol demonstrated to be effective in controlled clinical
studies.
While this treatment was designed to break the links between physical sensations,
catastrophic cognitions, and fear, its efficacy may be more directly related to other
aspects of the treatment, including decreased feelings of helplessness and demoraliza-
tion and increased levels of expectation and self-control. Thus, CBT was not more
effective than a highly structured, manual-driven "nonprescriptive treatment" consist-
ing of 3 educational sessions and 12 sessions that emphasized reflective listening.
Although less well studied, other forms of CBT and an "applied muscular relaxation"
technique (much more detailed and extensive than progressive muscular relaxation)
may also be effective. Some patients and family members benefit from participation in
support groups and reading materials like those provided by the Anxiety Disorders
Association of America.
D. Agoraphobia
Over 95% of patients with agoraphobia have panic disorder with agoraphobia.
Characterization and treatment of recurrent panic attacks will increase the patient's
5. ANXIETY 255
ability to confront and tolerate agoraphobic situations. Once the panic attacks are
addressed, the best-established treatment for agoraphobia is exposure therapy, which
consists of frequent, enduring, and increasingly intense exposure to the phobic situation
(discussed in Section III.B).
G. Specific Phobia
Although specific phobias are extremely common, few individuals present to
psychiatrists or psychologists for treatment. Some patients might find it easier to avoid
the phobic situation, others might fear a treatment that involves exposure to the situa-
tion, and others might be unaware that an effective treatment is available.
The best-established treatment of a specific phobia is exposure therapy, which, as
previously discussed, involves frequent, sufficiently enduring, and increasingly intense
exposures to the phobic stimulus. Although there have been surprisingly few well-
controlled studies, daily exposures are likely to be associated with relatively rapid
5. ANXIETY 257
H. Obsessive-Compulsive Disorder
Since many patients with obsessive-compulsive disorder (OCD) do not discuss
their problem without explicit encouragement-and since it afflicts about lout of
every 40 individuals-clinicians should routinely ask their patients about obsessions
and compulsions.
Established treatments of OCD include the serotonin reuptake inhibitor and a form
of behavioral therapy known as exposure with response prevention. Advantages of
medication treatment include its established value, its availability, its ease of use, its
ability to treat comorbid disorders like depression and panic disorder, its ability to treat
patients with pure obsessions, its efficiency in the treatment of patient with multiple
obsessions and compulsions, and the possibility that it increases compliance with
behavioral therapy; disadvantages include medication side effects and potential drug
interactions, an extremely high relapse rate following medication discontinuation, and
thus the likely need for lifelong treatment. Advantages of exposure with response
prevention include its possibly superior efficacy in those who complete treatment, its
possibly low relapse rate following treatment discontinuation, and the absence of
medication side effects; disadvantages include the limited availability of properly
trained and experienced therapists, the discomfort associated with exposure to obses-
sion-eliciting situations, the need for patients to have a lot of motivation, the need for
patients to have good insight about the senseless nature of their obsessions, an approx-
imately 20% dropout rate, an apparent lack of efficacy in patients with pure obsessions,
the requirement for extended treatment and treatment-related discomfort in patients
with multiple obsessions and compulsions, and a paucity of studies involving symptom
constellations other than those associated with washing and checking rituals. While
additional studies are needed, many clinicians believe that the optimal treatment con-
sists of the combination of a serotonin reuptake inhibitor and behavioral therapy. If a
combination of treatments is considered, it is wise to institute and monitor the effects of
one treatment at a time.
258 ERIC M. REIMAN, M.D.
Medications with established value in the treatment of OCD include the nonselec-
tive serotonin reuptake inhibitor clomipramine and the SSRls fluoxetine, fluvoxamine,
sertraline, and paroxetine. These medications are helpful in about 60-80% of patients,
typically reduce symptoms by about 50%, and appear to maintain their efficacy with
continued use but are associated with an extremely high relapse rate (about 90%
following their discontinuation). Although the improvement associated with these
medications is usually incomplete, it is usually sufficient to improve the patient's
quality of life. While direct comparisons are still needed, a metaanalysis of multicenter
placebo-controlled studies suggests that clomipramine is significantly more effective
than the SSRls and, despite significantly more side effects, is associated with a slightly
lower dropout rate.
By comparison with many other disorders that are responsive to a serotonin
reuptake inhibitor, OCD appears to require a higher dose and duration of treatment.
Although a few studies question the need for a higher dose, a trial of the highest
tolerable dose up to the maximum recommended is usually warranted, since the pa-
tient's response is typically incomplete. Patients usually benefit from clomipramine,
150-250 mg qHS; fluoxetine, 40-80 mg qAM; fluvoxamine, 200-300 mg qHS; ser-
traline, 150-200 mg qHS; or paroxetine, 40-60 mg qHS. A therapeutic trial consists of
a least 8-12 weeks of optimal treatment, and further improvement is often observed
after this time. Medication is usually continued for at least one year and periodically
reduced to characterize the lowest dose required.
Clinical researchers continue to investigate medication strategies for treating pa-
tients who do not have an adequate response to a serotonin reuptake inhibitor. Although
definitive studies are lacking, patients with the poor-insight type of OCD (who appear
less likely to respond to a serotonin reuptake inhibitor) and those with a coexisting
diagnosis of Tourette's disorder might benefit from the addition of an antipsychotic.
Uncontrolled studies suggest that the antiobsessional effects of a serotonin reup-
take inhibitor might be augmented by the addition of fenfluramine (Pondimin®, an
appetite suppressant with serotonin-releasing and reuptake-inhibiting effects), but con-
trolled studies remain to be reported. While the uncontrolled studies used 40 mg per
day in a single dose, it can be increased up to 120 mg per day in single or divided doses.
(Although studies suggest that high doses of fenfluramine can be neurotoxic in labora-
tory animals, findings do not yet support the neurotoxicity of clinical doses.) Although
a serotonin syndrome has not yet been reported, clinicians should increase the medica-
tion gradually and monitor the patients for adverse effects.
While uncontrolled studies suggested that brofaromine augmentation of a seroto-
nin reuptake inhibitor was beneficial in some patients with OCD, controlled clinical
trials fail to support a therapeutic role of brofaromine or brofaromine augmentation in
the treatment of OCD. While lithium augmentation has established value in the treat-
ment of medication-refractory depression, studies fail to support a therapeutic role of
lithium or lithium augmentation in the treatment of OCD. While an uncontrolled study
suggests that augmentation with the serotonin 1A (and j3-adrenergic receptor) antago-
nist pindolol may be helpful in patients with medication-refractory depression, the
adjunctive effects of this medication and investigational serotonin IA antagonists have
not yet been studied in patients with OCD. Although it is tempting to consider a
5. ANXIETY 259
tion in these symptoms, and placebo was associated with no significant improvement.
In the one controlled study of a benzodiazepine, alprazolam was associated with
improvement in anxiety but was unhelpful for the other symptoms of PTSD. While
case reports suggest that other medications, such as the reversible inhibitors of mono-
amine oxidase A, valproate, carbamazepine, bus pirone, f3-blockers, clonidine, and
lithium may have a role in certain patients with PTSD, these reports need to be
confirmed.
Thus, certain antidepressants (e.g., imipramine, 150-300 mg qHS) and mono-
amine oxidase inhibitors (e.g., phenelzine, 30-90 mg per day) may have an adjunctive
role in PTSD, reducing intrusive experiences, possibly reducing arousal symptoms, and
treating comorbid symptoms of major depressive disorder, panic disorder, or gener-
alized anxiety disorder. A comparison of the randomized clinical trials suggests that the
optimal dose of medication should be tried for at least 8 weeks before it is discon-
tinued.
Although differences in the gender of studied patients, type of stressor, and dura-
tion of symptoms make it difficult to compare the randomized clinical trials of behavior
therapy to each other or to the medication studies, exposure therapy remains the best-
established treatment for PTSD. This treatment involves repeated and prolonged expo-
sure to the imagined traumatic event or real stimuli that are associated with the traumat-
ic event. In one study, rape victims with PTSD repeatedly described the traumatic
event, repeatedly listened to audiotape recordings of their description, and repeatedly
confronted feared and avoided situations that were associated with the traumatic event.
This treatment reduced intrusive and arousal symptoms but had no effect on emotional
numbing or avoidance. While this treatment appears to be associated with significant
and long-lasting benefits, a recent study suggests that it can occasionally cause exacer-
bations in depression, alcohol, and panic disorder, especially in those with coexisting
disorders. Cognitive treatments, such as "stress inoculation training," are designed to
promote a sense of control by identifying and revising cognitive distortions, using
relaxation and controlled breathing exercises, and rehearsing strategies to cope with
anticipated stressors. "Eye movement desensitization" therapy is a recently developed
intervention that combines exposure to the traumatic memory with hand-driven volun-
tary saccades to treat PTSD and, perhaps, other anxiety disorders. While preliminary
reports describe dramatic improvements in intrusive and arousal symptoms following
only one or two sessions, randomized trials are needed to confirm the efficacy and
specificity of this treatment. Physiological explanations invoking the role of processes
involved in rapid eye movement (REM) sleep are speculative and unconvincing.
Individuals with PTSD can also benefit from supportive psychotherapy or support
groups that promote their self-esteem, decrease their sense of social isolation, and
bolster their ability to cope with everyday challenges.
asking "Do you worry excessively about minor matters?" Although an affirmative
response is insufficient to establish the diagnosis of GAD, a negative response is likely
to rule it out. Establish that the worry involved matters other than those directly related
to another psychiatric disorder, such as schoolwork, household chores, being late for an
appointment, automobile repairs, financial matters, the health or safety of family mem-
bers, and financial problems. Although few studies have addressed the long-term
efficacy of GAD interventions, the best-established treatments include the ben-
zodiazepines, buspirone, and imipramine.
While many patients report significant improvement with benzodiazepines, the
benefits appear to be relatively incomplete: less good for subjective symptoms like
worry than somatic symptoms like cardiorespiratory arousal. Although placebo-con-
trolled studies have not been conducted to address the possibility of tolerance to the
medication's anxiolytic effects, uncontrolled studies suggest that the medication re-
mains at least partially effective. The optimal benzodiazepine dose for treating GAD is
not well established; many studies employed a daily dose equivalent to about 25 mg of
diazepam, clinicians frequently use a daily dose equivalent to 5-30 mg of diazepam,
and perhaps some patients require doses equivalent to those used in the treatment of
panic disorder. Since the relapse rate following benzodiazepine discontinuation is high,
patients should be prepared for the possibility of long-term treatment.
Randomized clinical trials indicate that buspirone is as effective as benzodiaze-
pines in the treatment of GAD, especially in those who have never received prior
treatment with a benzodiazepine. Despite the limitations which have been discussed in
section IV.C, buspirone's potential advantages make it a possible first-line medication
treatment for GAD, especially in elderly patients, cognitively impaired patients, and
those with a history of substance dependence. The patient should be informed about the
advantages of this medication over benzodiazepines, the absence of sedative or imme-
diate effects, and the need for standing doses. The starting dose of buspirone is typ-
ically 5 mg t.i.d., the effective dose is usually 5-10 mg ti.d., and the medication is
sometimes increased as high as 20 mg t.i.d.
Although less well studied in patients with GAD, antidepressants are under-
utilized in the treatment of this disorder. Several studies have demonstrated the efficacy
of imipramine (starting dose usually 50 mg qHS in a patient without panic attacks;
usual dose 150-300 mg pHS) in the treatment of GAD. If tricyclic antidepressant side
effects such as dry mouth and increased appetite are tolerable, an antidepressant may be
the treatment of choice for some patients. Preliminary studies suggest that venlafaxine
(Effexor®) is effective in the treatment of GAD; if so, this medication could become a
first-line treatment. The efficacy of SSRIs and other antidepressants with fewer side
effects than imipramine remains to be established.
In contrast to several other anxiety disorders, GAD has demonstrated only a
modest response to behavioral treatments such as relaxation training, perhaps because
these treatments did not specifically target the excessive, poorly controlled anxiety that
always accompanies the disorder. Newer cognitive-behavioral treatment programs that
target excessive worry are now being investigated. A supportive approach can help
patients and family members cope with this chronic and frequently bothersome prob-
lem.
262 ERIC M. REIMAN, M.D.
VI. CONCLUSIONS
Anxiety and its disorders are extremely common, often distressing, and potentially
disabling. When patients present with anxiety symptoms, characterize the underlying
problem, associated problems, previous treatments, and established treatment alterna-
tives; consider the patient's understanding of the problem and alternative treatments,
correct misconceptions, and allay misplaced feelings of blame; help the patient make
an informed choice about the optimal treatment; provide an adequate therapeutic trial
(in the case of established medications, use the appropriate dose, schedule, and duration
of treatment; in the case of an established psychotherapy, adhere as closely as possible
to the reported protocol); monitor the patient's response; discontinue treatments that are
ineffective or no longer necessary; and be available to address the concerns of patients
and their families. When anxiety disorders are recognized and properly managed,
patients can be effectively treated.
REFERENCES
I. LeDoux 1. E. Emotion. In Mountcastle V. B., Plum E, Geiger S. R (eds.): Handbook of Physiology.
Section I: The Nervous System, Baltimore, Maryland, Williams & Wilkins, 1987, pp. 419-460.
2. Gray 1. A. G.: The Neuropsychology of Anxiety: An Enquiry into the Functions of the Septohippocampal
System. New York, Oxford University Press, 1982.
3. Barlow D. H., and Cerny 1. A.: Psychological Treatment of Panic. New York, Guilford Press, 1988.
SELECTED READING
American Psychiatric Association: Benzodiazepine Dependence, Toxicity, and Abuse. Washington, D.C.:
American Psychiatry Association, 1990.
American Psychiatry Association: Diagnostic and Statistical Manual of Mental Disorders, 4th ed. Washing-
ton, D.C., American Psychiatry Association, 1994.
Ballenger 1. C., Burrows G. D., DuPont R L., Jr., Lesser I. M., Noyes R, Jr., Pecknold J. c., Rifkin A.,
Swinson R P.: Alprazolam in panic disorder and agoraphobia: results from a multicenter trail. Arch Gen
Psychiatry 45:413-422.
Barlow D. H.: Anxiety and its Disorders. The Guilford Press, New York, 1988.
Beck A. T., Emery G.: Anxiety Disorders and Phobias. Basic Books, New York, 1985.
Black D. W., Wesner R, Bowers W., Gabel 1.: A comparison of fluvoxamine, cognitive therapy, and placebo
in the treatment of panic disorder. Arch Gen Psychiatry 50: 44-50, 1993.
Brown T. A., Barlow D. H., Liebowitz M. R: The empirical basis of generalized anxiety disorder. Am J
Psychiatry 151:1272-1280,1994.
Charney D. S., Heninger G. R: Noradrenergic function and the mechanism of action of antianxiety treat-
ment. Arch Gen Psychiatry 42:458-481, 1985.
Charney D. S., Deutch A. Y., Krystal 1. H., Southwick S. M., Davis M.: Psychobiologic mechanisms of
posttraumatic stress disorder. Arch Gen Psychiatry 50:294-305, 1993.
Davidson J., Kudler H., Smith R, et al: Treatment of posttraumatic stress disorder with amitriptyline and
placebo. Arch Gen Psychiatry 47:259-266, 1990.
den Boer J. A., van Vliet I. M., Westenberg H. G. M.: Recent advances in the psychopharmacology of social
phobia. Prog Neuropsychopharmacol Biol Psychiatry 18:625-645, 1994.
S. ANXIETY 263
Eison A. S., Temple D. L.: Buspirone: Review of its pharmacology and current perspectives on its mecha-
nism of action. Am J Med 80:1-9, 1986.
Fesler F. A.: Valproate in combat-related posttraumatic stress disorder. J Clin Psychiatry 52:361-364,1991.
Frank 1. B., Kosten T. R., Giller E. L., Dan E.: A randomized clinical trial of phenelzine and imipramine for
posttraumatic stress disorder. Am J Psychiatry 145:1289-1291, 1988.
Goodman W. 1., McDougle C. 1., Barr L. C., Aronson S. c., Price L. H.: Biological approaches to treatment-
resistant obsessive compulsive disorder. J Clin Psychiatry 54:16-26, 1993.
Gorman J. M., Liebowitz M. R., Fyer A. 1., Stein J.: A neuroanatomical hypothesis for panic disorder. Am J
Psychiatry 146:148-161, 1990.
Grady T. A., Pigott T. A., L'Heureux F., Hill J. L., Bernstein S. E., Murphy D. L.: Double-blind study of
adjuvant buspirone for fluoxetine-treated patients with obsessive-compulsive disorder. Am J Psychiatry
150:819-821, 1993.
Greenblatt D. 1., Divoll M., Abernethy D. R., et al: Benzodiazepine kinetics: Implications for therapeutic
pharmacogeriatrics. Drug Metab Rev 14:251-292, 1983.
Greist J. H., Jefferson 1. w., Kobak K. A., Katzelnick D. J., Serlin R. C.: Efficacy and tolerability of serotonin
transport inhibitors in obsessive-compulsive disorder. Arch Gen Psychiatry 52:53-60, 1995.
Herman 1. B., Rosenbaum 1. F., Brotman A. W.: The alprazolam to clonazepam switch for the treatment of
panic disorder. J Clin Psychopharmacol7:175, 1987.
Hollander E., DeCaria C., Schneier F. R., Schneier H. A., Liebowitz M. R., Klein D. F.: Fenfluramine
augmentation of serotonin reuptake blockade antiobsessional treatment. J Clin Psychiatry 51: 119-123,
1990.
Insel T. R., Ninan P. T., Aloi J., et al: A benzodiazepine receptor mediated model of anxiety: Studies in
nonhuman primates and clinical implications. Arch Gen Psychiatry 41:741-750, 1984.
Jenike M. A., Baer L., Minichiello W. E.: Obsessive-Compulsive Disorders: Theory and Management. Year
Book Medical Publishers, Chicago, 1990.
Jenike M. A., Baer L., Buttolph L.: Buspirone augmentation of fluoxetine in patients with obsessive
compulsive disorder. J Clin Psychiatry 52:13-14, 1991.
Kahn R. J., McNair D. M., Lipman R. S., et al: Imipramine in chlordiazepoxide in depressive and anxiety
disorders. II. Efficacy in anxious outpatients. Arch Gen Psychiatry 43:79-85, 1986.
Klein E., Colin V., Stolk 1., Lenox R. H.: Alprazolam withdrawal in patients with panic disorder and
generalized anxiety disorder: Vulnerability and effect of carbamazepine. Am J Psychiatry 151: 1760-
1766, 1994.
Laegrid L., Olegard R., Wahlstrom 1., et al: Abnormalities in children exposed to benzodiazepines in utero.
Lancet 1:108, 1987.
Lennox R. H., Shipley 1. E., Peyser J. M., et al: Double-blind comparison of alprazolam and imipramine in
the inpatient treatment of major depressive illness. Psychopharmacol Bull 20:79, 1984.
Liebowitz M. R., Gorman 1. M., Fyer A. 1., Klein D. F.: Social phobia. Arch Gen Psychiatry 42:729-736,
1985.
Liebowitz M. R., Gorman J. M., Fyer A. 1., et al: Pharmacotherapy of social phobia: An interim report of a
placebo-controlled comparison of phenelzine and atenolol. J Clin Psychiatry 49:252-257, 1988.
Lipman R. S., Covi L., Rickels K., et al: Imipramine and chlordiazepoxide in depressive and anxiety
disorders. I. Efficacy in depressed outpatients. Arch Gen Psychiatry 43:68, 1986.
Marshall J. R.: Practical approaches to the treatment of social phobia. J Clin Psychiatry 55:367-374, 1994.
McDougle C. J., Goodman W. K., Leckman J. F., Holzer J. C., Barr L. C., McCance-Katz E., Heninger G. R.,
Price L. H.: Limited therapeutic effect of addition of buspirone in fluvoxamine-refractory obsessive-
compulsive disorder Am J Psychiatry 150:647-649, 1993.
Noyes R., Jr., DuPont R. L., Jr., Pecknold 1. C., Rifkin A., Rubin R. T., Swinson R. P., Ballenger J. C.,
Burrows G. D.: Alprazolam in panic disorder and agoraphobia: Results from a multicenter trial. Arch
Gen Psychiatry 45:423-428, 1988.
Noyes R., Jr., Garvey M. 1., Cook B., Suelzer M.: Controlled discontinuation of benzodiazepine treatment for
patients with panic disorder. Am J Psychiatry 148:517-523, 1991.
Otto M. w., Pollack M. H., Sacs G. S., Reiter S. R., Meltzer-Brody S., Rosenbaum J. F.: Discontinuation of
264 ERIC M. REIMAN, M.D.
benzodiazepine treatment: Efficacy of cognitive-behavioral therapy for patients with panic disorder.
Am J Psychiatry 150:1485-1490, 1993.
Pollack M. H., Tesar G. E., Rosenbaum J. F., Spier S. A.: Clonazepam in the treatment of panic disorder and
agoraphobia: A one-year follow-up. J Clin Psychopharmacol 6:302, 1986.
Reist C., Kauffmann C. D., Haier R. J., et a1: A controlled trial of desipramine in 18 men with posttraumatic
stress disorder. Am J Psychiatry 146:513-516, 1989.
Rickels K. R., Schweizer E., Csanalosi I., Case W. C., Chung H.: Long-term treatment of anxiety and risk of
withdrawal. Arch Gen Psychiatry 45:444-450, 1988.
Rickels K., Case W. G., Schweizer E., Garcia-Espana F., Fridman R.: Long-term benzodiazepine users 3
years after participation in a discontinuation program. Am J Psychiatry 148:757-761, 1991.
Rickels K., Schweizer E., Weiss S., Zvodnick S.: Maintenance drug treatment for panic disorder. Arch Gen
Psychiatry 50:61-68, 1993.
Rickels K., Downing R., Schweizer E., Hassman H.: A placebo-controlled comparison of imipramine,
trazodone, and diazepam. Arch Gen Psychiatry 50:884-885, 1993.
Rosenberg L., Mitchell A. A., Parsells J. L., et al: Lack of relation of oral clefts to diazepam use during
pregnancy. N Engl J Med 309:1282, 1983.
Roy-Byrne P. P., Dager S. R., Cowley D. S., et al: Relapse and rebound following discontinuation of
benzodiazepine treatment of panic attacks: Alprazolam versus diazepam. Am J Psychiatry 146:860,
1989.
Salzman C.: The APA Task Force report on benzodiazepine dependence, toxicity, and abuse. Am J Psychiatry
148:2, 1991.
Salzman C., Shader R. I., Greenblatt D. J., Harmatz J. S.: Long vs. short half-life benzodiazepines in the
elderly. Arch Gen Psychiatry 40:293, 1983.
Schneier F. R., Saoud J. B., Campeas R., Fallon B. A., Hollander E., Coplan 1., Liebowitz M. R.: Buspirone
in social phobia. J Clin Psychopharm 13:251-256, 1993.
Schneier F. R., Heckelman L. R., Garfinkel R., Campeas R., Fallon B. A., Gitow A., Street L., Del Bene D.,
Liebowitz M. R.: Functional impairment in social phobia. J Clin Psychiatry 55,322-331, 1994.
Schweizer E., Rickels K.: Failure of buspirone to manage benzodiazepine withdrawal. Am J Psychiatry
143:1590-1592, 1986.
Schweizer E., Rickels K., Weiss S., Zavodnick S.: Maintenance drug treatment of panic disorder. Arch Gen
Psychiatry 50:51-60, 1993.
Shader R. I., Greenblatt D. J.: Clinical implications of benzodiazepine pharmacokinetics. Am J Psychiatry
134:652-656, 1977.
Shapiro F.: Efficacy of the eye movement desensitization procedure in the treatment of traumatic memories. J
Traumatic Stress 2:199-223, 1989.
Shear M. K., Pilkonis P. A., Cloitre M., Leon A. C.: Cognitive behavioral treatment compared with non-
prescriptive treatment of panic disorder. Arch Gen Psychiatry 51:395-401, 1994.
Solomon S. D., Gerrity E. T., Muff A. M.: Efficacy of treatments for posttraumatic stress disorder. JAMA
268:633-638, 1992.
Southwick S. M., Krystal 1. H., Morgan C. A., Johnson D., Nagy L. M., Nicolaou A., Heninger G. R.:
Abnormal noradrenergic function in posttraumatic stress disorder. Arch Gen Psychiatry 50:266-274,
1993.
Sternbach H.: F1uoxetine treatment of social phobia. J Clin Psychopharm 10:230, 1990.
van Laar M. W., Volkerts E. R., van Willigenburg A. P. P.: Therapeutic effects and effects on actual driving
performance of chronically administered buspirone and diazepam in anxious outpatients. J Clin Psycho-
pharm 12:86-95, 1991.
van Vliet I. M., den Boer J. A., Westenberg H. G. M.: Psychopharmacological treatment of social phobia:
Clinical and biochemical effects of brofaromine, a selective MAO-A inhibitor. Eur Neuropsychophar-
macoI2:21-29, 1992.
Versiani M., Mundim F. D., Nardi A. E., Liebowitz M. R.: Tranylcypromine in social phobia. J Clin
PsychopharmacoI8:279-283, 1988.
Zorumski C. F., Isenberg K. E.: Insights into the structure and function of GABA-benzodiazepine receptors:
Ion channel and psychiatry. Am J Psychiatry 148:162-173, 1991.
6
Insomnia
JEFFREY B. WEILBURG, M.D., and
HEATHER S. HOPKINS
I. INTRODUCTION
Insomnia is defined in the fourth edition of the Diagnostic and Statistical Manual of
Mental Disorders (DSM-IV) as a complaint of difficulty initiating or maintaining
sleep or of nonrestorative sleep that persists for at least 1 month and is associated
with clinically significant distress or functional impairment.) Most patients with in-
somnia have prolonged sleep latency (time in bed to time asleep) and sleep that is
shallow or fragmented by multiple arousals. However, no single disturbance of sleep
physiology is found in all insomniacs. Some patients who complain of insomnia
appear normal on objective measurements of sleep. Even patients whose complaints
are matched by abnormal sleep physiology may have variations in objective sleep
measures. 2
Insomnia is a common problem. Between 15 and 20% of the population complain
of insomnia, and 10% report using "sleeping pills." Twenty percent of all adult general
medical outpatients and 35% of psychiatric outpatients complain of insomnia. Higher
rates are found among women, patients from lower socioeconomic groups, and the
elderly. The elderly receive 35-40% of sedative-hypnotic prescriptions written each
year in the United States, even though they account for only 12% of the population. 2 ,3
The estimated incidence of hypnotic drug use in nursing homes and residential facili-
ties for the elderly is 28-36%.4
Insomnia is also a significant public health problem. Approximately 1.7% of the
U.S. population receive a prescription for "sleeping pills" each year, though this rate
declined through the 1980s.5 Estimates based on survey data suggest that, in 1990, $1.1
265
266 JEFFREY B. WEILBURG, M.D., AND HEATHER S. HOPKINS
billion was spent on medications to promote sleep, with less than half of this spent on
prescription hypnotics. 6 •7 Patients with untreated insomnia have significant discomfort
and dysfunction8 and may have increased mortality in part because of heightened risk
of motor vehicle accidents. In 1990, the combined cost of medical care and self-
treatment of insomnia in the United States reached an estimated $10.9 billion.
A. Normal Sleep
Sleep physiology is observed using polysomnography, which involves continu-
ous, all-night monitoring of the electroencephalogram (EEG), eye movement, muscle
activity, respiration, and electrocardiogram (EKG). Blood oxygen levels, penile tumes-
cence, and video observation may sometimes be included in the evaluation.
The polysomnogram (PSG) indicates that sleep can be divided into two basic
types: rapid eye movement (REM) and non-REM (NREM) sleep. During REM sleep,
the EEG becomes activated, pulse and respiration increase and become irregular,
muscle tone decreases, and the eyes move in rapid bursts. Dreaming is associated with
REM sleep. NREM sleep is subdivided into stages 1-4. Stage 1 is a drowsy, transition-
al state that occupies about 5% of the time healthy adults spend asleep. Stage 2
accounts for about 50% of the time spent asleep. About 10-20% of sleep time is spent
in stages 3 and 4 (slow-wave or delta sleep), which are the deepest levels of sleep.
REM and NREM sleep alternate throughout the night. The pattern of cycling is
called the sleep architecture. The normal sleeper begins with stage 1 and then slips into
stage 2, where sleep formally begins. After a period of deep sleep (stages 3 and 4),
stage 2 returns. The first REM period follows, appearing 90-120 minutes after sleep
onset. NREM and REM sleep then alternate approximately every 90 minutes through-
out the night. Slow-wave sleep is more prominent in the early cycles, and REM sleep is
more prominent toward morning. Sleep architecture changes with age. Children and
adolescents have large amounts of slow-wave sleep, but in adults, wakefulness and
stage 1 sleep increase, while stage 3 and stage 4 sleep decrease.
B. Sleep Disorders
Some patients with chronic primary insomnia and most patients who have insom-
nia related to medical conditions or substance abuse (see discussion of diagnosis
below) are found to have a prolonged (30 min) sleep latency, frequent unexplained
arousals, and shallow, fragmented sleep. The smooth reciprocal cycling of sleep stages
characteristic of normal sleep may be disrupted in these patients. They often complain
of mild to moderate impairment of memory and psychomotor performance and have
elevated scores on mood and anxiety rating scales. Patients with insomnia secondary to
Axis I disorders tend to show the sleep changes characteristic of the underlying disor-
der, e.g., shortened REM latency in depressed patients. However, a significant propor-
tion of patients with insomnia demonstrate normal or nearly normal sleep on the PSG.
Thus, there is no single pathophysiological basis of or marker for insomnia. Rather, it
6. INSOMNIA 267
results from a variety of physiological and psychological factors and is, therefore, best
regarded as a symptom, rather than a disease.
Psychological factors-such as anxiety, anger, and frustration-often contribute
to the development and persistence of insomnia. Perceptual factors-such as aberra-
tions in the capacity to judge level of alertness and to estimate the passage of time-
may playa role in patients with chronic primary insomnia. In addition, environmental
factors-such as changing schedules in shift workers or time changes experienced by
jet travelers-may produce discoordination between internal pacemakers and ambient
day/night conditions and thus playa role in producing insomnia.
Various primary disorders of sleep may present as insomnia; these are outlined
below. If a primary sleep disorder is suspected, polysomnography is indicated. A
complete description of these and other primary and secondary sleep disorders can be
found in Erman's monograph. 9
1. Sleep Apnea
Sleep apnea is a syndrome of repetitive episodes of diminished or absent respira-
tion during sleep, leading to decreased blood oxygen saturation. The two basic forms of
sleep apnea are central and obstructive. Both types may present as insomnia. With
central apnea, the diaphragm is not stimulated to contract; thus, no respiration occurs
until the patient arouses. In obstructive sleep apnea, the airway is occluded by excess
tissue, craniofacial abnormalities (such as a small airway), poorly toned pharyngeal
muscles, or a combination of these conditions, which prevents air flow despite contin-
ued ventilatory effort. Loud snorts, snores, and partial arousal are noted on the PSG.
Many patients have mixed apnea, which includes central and obstructive elements.
Most patients with sleep apnea suffer from excessive sleepiness, but a few com-
plain of insomnia. Since hypnotic medications and other sedatives may raise the
arousal threshold and thereby delay the return of respiration, the use of such agents in
apneics can be dangerous. When respiration ceases, blood oxygen saturation falls, CO 2
rises, pH falls, and blood pressure rises. Serious cardiorespiratory problems may occur
if apnea is untreated. In severe cases, untreated sleep apnea may be fatal. 9 Screening
questions for sleep apnea should be included in the routine insomnia workup, and the
diagnosis pursued by PSG, so that hypnotics are not used inappropriately.
2. Nocturnal Myoclonus
Nocturnal myoclonus is a stereotypic repeated twitching of the leg and other
muscles during sleep. Patients may be unaware of the movements, but bed partners or
family members will often notice. The myoclonic activity may partially arouse the
patient and thus can produce fragmented, nonrestorative sleep. Patients generally com-
plain of excessive sleepiness but, on some occasions, may complain of insomnia.
Nocturnal myoclonus and other sleep-related movement problems are often cate-
gorized as periodic leg movements of sleep (PLMS). PLMS may be precipitated by
psychotropics such as cyclic antidepressants, monoamine oxidase inhibitors (MAOIs),
and antipsychotic agents. 9
268 JEFFREY B. WEILBURG, M.D., AND HEATHER S. HOPKINS
3. Phase Shifts
Some patients have a sleep-wake cycle that is not synchronized with the day-
night, light-dark environment. In the case of phase delay, patients will fall asleep
several hours after a socially desired bedtime (e.g., 4 A.M.) and will, if undisturbed,
awake refreshed at noon. If such patients have to get up for work or school at 7 A.M.,
however, they will feel tired and may complain of insomnia. Behavioral treatment,
such as phase advance or delay, is the effective treatment for phase-shift problems. It is
important to have phase shifts ruled out before initiating drug treatment, since hypnot-
ics may be ineffective or even contraindicated. 10,1 I
4. Psychophysiological Insomnia
Although controversial, there are some patients who appear overaroused but do
not meet criteria for psychiatric disorders. Such patients learn to associate bedtime with
anxiety and develop insomnia by classical conditioning. In these individuals, behav-
ioral management (e.g., deconditioning, relaxation, biofeedback) may be judiciously
combined with benzodiazepines. 12,13
As the name suggests, some patients have objective (on PSG) and subjective
insomnia without an identifiable cause. Some patients with childhood-onset insomnia
may suffer from some idiopathic problem with sleep.
The clinician should attempt to determine the etiology of insomnia before treat-
ment is initiated. Physiological, psychological, and environmental factors should be
considered when the differential diagnosis is constructed. Although this task may be
difficult, it is not an academic exercise. Sedative-hypnotics can be ineffective and in
some cases dangerous if the problem producing the insomnia is not taken into account.
For example, the patient who complains of "insomnia" because of underlying sleep
apnea may develop significant respiratory and secondary cardiovascular problems if
sedatives are given without appropriate caution. The patient with sleep difficulty sec-
ondary to the use of or rebound from alcohol or hypnotics may be at risk of overdosage
or exacerbation of the substance abuse if sedatives are prescribed before the primary
problem is addressed. 14
The classification schema for sleep disorders found in DSM-IV lists different
subtypes of insomnia than those found in the International Classification of Sleep
Disorders: Diagnostic and Coding Manual. 15 Some authors further divide insomnia
into transient and chronic types. Although some controversy about classification exists,
most investigators agree that a significant proportion of patients with insomnia have
underlying psychiatric (including substance use and abuse) disorders. The patho-
6. INSOMNIA 269
1. Psychiatric disorders
a. Mood disorders (including dysthymic disorder)
b. Character disorders
c. Anxiety disorders
d. Others: psychosis, adjustment disorders, etc.
2. Drugs
a. Alcohol (dependence, rebound, or withdrawal phenomenon)
b. Sedatives (dependence, rebound, or withdrawal phenomenon): barbiturates, benzodiazepines, others
c. Stimulants (dependence, rebound, or withdrawal phenomenon): amphetamines, methylphenidate;
pemoline; cocaine; caffeine and stimulant xanthines (coffee, tea, cola, chocolate)
d. Antidepressants: phenelzine, tranylcypromine, protriptyline, desipramine, imipramine, amoxapine,
fluoxetine
e. Antiasthmatics, decongestants
f. Tobacco (dependence, rebound, or withdrawal phenomenon)
3. Medical/surgical problems
a. Pain of any source
b. Respiratory: chronic obstructive pulmonary disease
c. Renal: uremia, urinary tract infection
d. Endocrine: hyper- and hypothyroidism; exogenous steroids
e. Cardiovascular: nocturnal angina, orthopnea
f. Delirium: dementia, infection, medication toxicity (including sedative-hypnotics or over-the-
counter sleep remedies)
4. Primary sleep disorder
a. Sleep apnea
b. Narcolepsy
c. Periodic movements during sleep: nocturnal myoclonus, periodic leg movements of sleep (PLMS)
d. Circadian rhythm disruption: phase shift
5. Others
a. Idiopathic primary insomnia
b. Childhood-onset insomnia
c. Psychophysiological or conditioned insomnia
d. Irregular schedule, inadequate sleep time
e. Unusual polysomnographic pattern
physiological processes that can present as insomnia, abstracted from these sources, are
listed in Table 1.
The insomnia workup should involve specific and detailed interviews with the
patient and, if possible, bed partners and family members. Patients can keep a sleep log,
with entries made each morning detailing time in bed; estimated time of falling asleep;
number, duration, and precipitating factors of nocturnal awakenings; time of waking
and leaving bed; and quality of cognitive and mood state upon retiring and arising. Any
important events or stressors in the preceding day and all medications or drugs con-
sumed (including alcohol, cigarettes, and caffeinated beverages) should be listed each
morning. Such a log, kept over a 2-week period or longer, may reveal an association
between environmental, emotional, substance, or schedule-related factors and sleep
270 JEFFREY B. WEILBURG, M.D., AND HEATHER S. HOPKINS
be used only for sleeping or lovemaking. Patients should get out of bed into a chair for
reading or other quiet activities if they are unable to sleep after 10 minutes. Moderate
exercise early in the day, a regular sleep-wake schedule (including weekends), and a
regular bedtime preparation ritual that allows time to "wind down" are useful. A small
snack of milk or cheese before bed may help; stimulants and daytime naps should be
avoided. Hypnotics may be used adjunctively.
Patients with drug and alcohol abuse (including abuse of hypnotics) must be
detoxified and then helped to remain abstinent. Dry alcoholics may show objective
evidence of disrupted sleep for up to I year after the initiation 6f sobriety. Supportive
psychotherapy, sometimes combined with antidepressants, lithium (Eskalith® and oth-
ers), or antihistamines, may be used to improve insomnia in these cases. Drugs that
cross-react with the drug of abuse, such as benzodiazepines with alcohol, should be
avoided.
Some patients are very sensitive to the effects of caffeine and must avoid all
coffee, tea (including decaffeinated types), colas, chocolate, or stimulants (deconges-
tants, diet pills, etc.) in order to manage their insomnia. Likewise, cigarette smoking
can produce insomnia, possibly related to the stimulant and withdrawal effects of
nicotine, so some patients may have to stop smoking to treat their insomnia suc-
cessfully.ls
Shift workers or those with circadian rhythm disruptions may be helped by sched-
ule manipulation, including incremental realignment of the sleep cycle with the 24-hr
day. Bright light, antidepressants, lithium, and hypnotics may also be used adjunctively
in some patients. I I
Insomnia secondary to a mood disorder may respond well to antidepressants and
lithium (see Chapter 3). Adjusting the dosage schedule or switching to a more sedating
agent may help those who are having difficulty falling asleep because of the stimulat-
ing effects of an antidepressant. If a patient is experiencing insomnia because of a panic
or anxiety disorder, optimal management of the anxiety disorder itself is the best
strategy for improving sleep.
IV. PHARMACOTHERAPY
A. Benzodiazepines
Benzodiazepines tend to decrease sleep latency and diminish the number of noc-
turnal arousals. They also usually increase the amount of stage I and stage 2 sleep, at
the expense of delta sleep (stages 3 and 4), although the total amount of EEG delta
activity probably is not decreased because delta is redistributed throughout other
stages. The effects of benzodiazepines on REM sleep are variable, depending on the
individual, the underlying illness, and the dose and type of drug. 19
The benzodiazepines marketed in the United States primarily as hypnotics are
estazolam (Prosom®), flurazepam (Dalmane®), quazepam (Dormalin®), temazepam
(Restoril and others), and triazolam (Halcion). These medications are indicated for the
short-term treatment of insomnia characterized by difficulty in falling asleep, frequent
nocturnal awakenings, and/or early morning aWakenings. Benzodiazepines that are
marketed as anxiolytics but are sometimes useful in managing insomnia include di-
azepam (Valium®), clonazepam (Klonopin®), and oxazepam (Serax®).20.21 Lorazepam
(Ativan®) is indicated for the treatment of insomnia caused by anxiety or transient
situational stress.
1. Mechanism of Action
See Chapter 5.
2. Pharmacokinetics
Benzodiazepines vary considerably on many pharmacokinetic parameters (see
Table 2). In general, though, they are well absorbed following oral administration and
reach peak plasma levels within 1-4 hr. Food can delay the rate, but not the extent, of
absorption. All are highly protein-bound. With long-half-life benzodiazepines-di-
azepam, flurazepam, and quazepam-repeated dosing leads to accumulation of the
parent compound and active metabolites. For these drugs, steady-state serum concen-
trations are usually reached 5-14 days after treatment is started. When treatment is
discontinued, the drug is eliminated slowly. Active metabolites in the blood can pro-
long the medication's effects for several days to weeks.
When taken orally, diazepam is the most rapidly absorbed benzodiazepine. Ab-
sorption is also usually rapid and complete when diazepam is injected into the deltoid
muscle but can be slow and erratic when given intramuscularly elsewhere. Diazepam is
metabolized by oxidation to active and inactive metabolites.
Flurazepam also is metabolized by oxidation to active and inactive metabolites.
The two inactive metabolites, hydroxethylflurazepam and flurazepam aldehyde, along
with low levels of flurazepam appear in the serum shortly after ingestion. These three
compounds probably induce sleep. They are eliminated over 8-12 hours and do not
accumulate with multiple dosing. Flurazepam's major active component, N-desalkyl-
flurazepam, appears gradually, is slowly eliminated, and may maintain sleep.
Maximum quazepam plasma concentrations occur at approximately 1.5 hr follow-
ing oral ingestion. Quazepam is metabolized by oxidation into two active metabolites,
2-oxoquazepam and N-desalkyl-2-oxoquazepam (which is identical to N-desalkylflur-
6. INSOMNIA 273
TABLE 2. Benzodiazepinesa
a All benzodiazepines can be used as sedative-hypnotics, even though they are not all labeled for this indication.
4. Adverse Effects
The type and duration of adverse effects associated with a benzodiazepine hypnot-
ic are influenced by the elimination half-lives of the drug and any active metabolites.
274 JEFFREY B. WEILBURG, M.D., AND HEATHER S. HOPKINS
As mentioned above, when half-lives are long, the drug or its metabolites may accumu-
late during nightly administration and be associated with impairment of cognitive
and/or motor performance during waking hours. These problems may be amplified
with repeated use during long-term treatment. In contrast, if the half-lives of the parent
compound and active metabolites are short, drug and metabolites will be cleared before
the next dose is ingested, and carryover effects related to excessive sedation or central
nervous system (CNS) depression should be minimal or absent. However, all hypnotics
produce similar adverse effects at higher doses in most patients, so dose is an important
consideration.
a. Next-Day Sedation. Agents and active metabolites with long elimination half-
lives, such as flurazepam and N-desalkylflurazepam, may produce next-day sedation.
Some patients complain of hangover from these agents, whereas others-often those
with increased anxiety levels-may experience this sedation as beneficial. Dose-de-
pendent daytime sedation over several days is not an uncommon reaction to quazepam,
but tolerance to this effect may develop.
b. Rebound Insomnia and Anxiety. Rebound insomnia and rebound anxiety are
problems sometimes caused by benzodiazepine hypnotics, especially agents that have
short or intermediate durations of action, such as triazolam, lorazepam, or tem-
azepam. 22 ,23 In rebound insomnia, the patient experiences subjective and objective
disruption of sleep, greater than the baseline difficulty, during nights when the hypnotic
is not used. This can occur in susceptible individuals even if the drug is used for only I
or 2 nights. Rebound anxiety is an increase in anxiety over baseline during the later
parts of the night or the day following the use of the drug. These conditions can be
managed by tapering a short-acting agent (cut the dose by 50% for 2 days and 50%
again for 2 more days)24 or switching to long-acting agents such as clonazepam,
flurazepam, or quazepam. Patient education is also helpful in avoiding escalating drug
dosages and drug dependence.
removal from the British market. However, studies comparing triazolam's side effects
with those of other hypnotics do not indicate statistically significant differences in
adverse-effect rates. 28 ,29
f. Elderly Patients. The presence of active compounds in the system during the
day following benzodiazepine use may predispose elderly patients (and others whose
nervous systems are sensitive to this kind of adverse effect) to slowed reaction times,
memory disturbance, and unsteadiness of gait. These effects can impair social and
occupational function and lead to falls, which may result in hip fractures. In one study,
current benzodiazepine use increased the risk of a fall leading to a femur fracture by
about 60%.32 The risk appears to be associated less with the pharmacokinetics of the
specific benzodiazepine and more with the total dose and the rate of dosage increase.
However, elderly patients may be at increased risk of falls and disorientation if they
awaken 2-4 hr after ingesting an agent with a short duration of action.
5. Drug Interactions
When benzodiazepines are combined with other medications that depress the
CNS, depressant effects may be increased, and the dosage of one of the medications
should be reduced. Similarly, when a benzodiazepine is combined with an opioid
analgesic, the dosage of the opioid analgesic should be decreased by at least one-third
and given in small increments. CNS depressant effects are also increased when ben-
zodiazepines are combined with alcohol.
Through the induction of hepatic microsomal enzyme activity, carbamazepine can
increase the metabolism of benzodiazepines biotransformed by the liver's P450 iso-
enzyme system, especially clonazepam. This leads to decreased serum concentrations
and reduced elimination half-lives of the benzodiazepines. The hepatic metabolism of
diazepam, estazolam, flurazepam, quazepam, and triazolam may be inhibited by the
concurrent use of cimetidine (Tagamet®), oral estrogen-containing contraceptives, di-
sulfiram, or erythromycin,19-21 resulting in delayed clearance and increased plasma
276 JEFFREY B. WEILBURG, M.D., AND HEATHER S. HOPKINS
concentrations. It may be necessary to reduce the dosage of one of the medications. The
hepatic metabolism of benzodiazepines such as lorazepam, oxazepam, and temazepam
is probably not affected, possibly because cimetidine, oral estrogen-containing contra-
ceptives, disulfiram, and erythromycin do not affect glucuronide conjugation.
Probenecid, however, can impair the glucuronide conjugation of lorazepam, oxazepam,
and temazepam, leading to increased effects and possibly excessive sedation.
In addition to erythromycin, the antibiotics troleandomycin (TAO®) and clari-
thromycin (Biaxin®) and the antidepressant nefazodone (Serzone®) may also increase
plasma levels of triazolam, probably by inhibiting the P4503A4hepatic isoenzyme.
The antifungal agents ketoconazole (Nizoral®) and itraconazole (Sporanox®) also in-
hibit the P4503A4 system and can cause dangerous elevations in triazolam concentra-
tions when coadministered.
Fluoxetine (Prozac®), a serotonin-selective reuptake inhibitor (SSRI), impairs the
clearance of diazepam, and thereby increases diazepam concentrations, but simul-
taneously decreases the formation of the active metabolite desmethyldiazepam, result-
ing in no greater psychomotor impairment from the combination of diazepam with
fluoxetine than from diazepam alone. Fluvoxamine (Luvox®), another SSRI, markedly
elevates levels of diazepam.
The combination of clozapine and a benzodiazepine has led to collapse in a few
cases. Some clinicians recommend that benzodiazepines be discontinued at least a
week before beginning clozapine treatment. If clozapine is to be used concomitantly
with a benzodiazepine, the dose of clozapine should be titrated upward very gradually.
A few patients have developed delirium during the concomitant administration of
clozapine and 10razepam. 33
There are advantages and disadvantages to using the benzodiazepines diazepam or
lorazepam with fentanyl derivatives prior to or during surgery. This combination can
decrease the dose of fentanyl derivative required to induce anesthesia and the time to
loss of consciousness after induction. It can also decrease the chances that a patient will
remember surgery after an operation. However, concurrent use of these medications
increases the risks of severe hypotension and respiratory depression and delays recov-
ery time, especially with intravenous administration of the benzodiazepine.
Benzodiazepines may decrease the therapeutic effects of L-dopa. In theory, the
concurrent use of zidovudine (Retrovir®) with benzodiazepines could competitively
inhibit hepatic glucuronidation and decrease the clearance of zidovudine, thereby in-
creasing its toxicity.
Acute ethanol ingestion may impair triazolam clearance and elimination (as well
as potentiating its psychomotor effects). Concurrent use of isoniazid (Dinacrin® and
others) with triazolam and diazepam can inhibit their elimination, resulting in increased
plasma concentrations.
The concurrent use of antacids could delay but not decrease the absorption of
diazepam. Omeprazole (Prilosec®) can prolong the elimination of diazepam. Concur-
rent use of rifampin (Rifadin® and others) may enhance the elimination of diazepam,
resulting in decreased plasma concentrations. Dosage adjustments may be necessary.
The use of lorazepam with scopolamine (Isopto®) may increase the incidence of
sedation, hallucination, and irrational behavior.
6. INSOMNIA 277
See Chapter 5.
7. Toxicity
See Chapter 5.
Estazolam is available as I-mg and 2-mg scored tablets. The usual adult dose for
sedation is 1 mg, although some people may need 2 mg. The dose for elderly patients is
also 1 mg, but in small or debilitated older patients, the initial dose should be 0.5 mg.
The usual adult daily dose of flurazepam is 15 or 30 mg. Some reports suggest that
15 mg may be as effective as 30 mg, so the lower dose should be tried first. In
debilitated or elderly patients, the initial dose should be 15 mg, increased to 30 mg as
needed and tolerated. Flurazepam is increasingly effective on the second or third night
of consecutive use, and for one or two nights after medication is discontinued, both
sleep latency and total wake time may still be reduced. Flurazepam is available as 15-
mg and 30-mg capsules.
For temazepam, the usual adult dose is 15 mg/day, but 7.5 mg may be sufficient to
improve sleep latency in transient insomnia. Debilitated patients should be started on
7.5 mg, with the dose increased as needed and tolerated. Temazepam is available in
7.5-mg, 15-mg, and 30-mg capsules. It is also available in 15-mg and 30-mg tablets.
It is best to use as low a dose of triazolam as possible, beginning with 0.125
mg/day in adults and not exceeding 0.5 mg; 0.25 mg is considered a usual dose. Some
patients may require 0.5 mg, but this should only be used if they have not responded to
lower doses, as the risk of side effects increases with increased dose. Triazolam is
available in 0.I25-mg tablets and 0.25-mg scored tablets.
Sleep laboratory and clinical studies indicate that a daily dose of 15 mg of
quazepam appears to be effective during short- and long-term (up to 4 weeks) adminis-
tration. Quazepam is available in 7.5-mg and I5-mg tablets. The recommended dosage
is 15 mg at bedtime for healthy adults, reduced to 7.5 mg as needed, and 7.5 mg at
bedtime for the elderly or debilitated. It is recommended that the dose be decreased in
debilitated patients after I or 2 nights of treatment.
278 JEFFREY B. WEILBURG, M.D., AND HEATHER S. HOPKINS
B. Zolpidem
Zolpidem, an imidazopyridine that is thought to have three-dimensional structural
homologies with benzodiazepines, is classified by the U.S. Food and Drug Administra-
tion (FDA) as a schedule IV controlled substance and indicated for use as a hypnotic.
In contrast to most of the benzodiazepines, which nonselectively bind to and activate
all three omega receptor subtypes, zolpidem binds the omega-I receptor preferentially.
This selective binding of zolpidem on the omega-I receptor is not absolute, but it may
explain the relative absence of muscle relaxant and anticonvulsant effects in animal
studies as well as the preservation of deep sleep (stages 3 and 4) in human studies of
zolpidem at hypnotic doses. Some studies suggest that low (but therapeutic) doses of
zolpidem may alter sleep architecture less than other agents.
Low doses of zolpidem (S or 7.S mg) decrease sleep latency and increase total
sleep time. 34 .35 Higher doses (e.g., 20 mg) may disrupt sleep architecture by decreasing
the percent of REM sleep.34 Preliminary studies suggest that, as with temazepam and
triazolam, zolpidem in low to moderate doses may not adversely affect nocturnal
respiration and may improve sleep in patients with mild to moderate COPD who
complain of insomnia. 3o
1. Pharmacokinetics
Zolpidem has a short half-life, no active metabolites, and a short duration of
action. It is rapidly and completely absorbed from the gastrointestinal tract and reaches
peak serum concentrations in less than 2 hr. Taking zolpidem with food may decrease
6. INSOMNIA 279
the completeness of absorption and delay the onset of action. Its elimination half-life is
1.5-4.5 hr. Because it is so quickly eliminated, it will not accumulate when used in
appropriate doses and probably does not impair next-day function. Old age and liver
impainnent slow elimination; the dose should be decreased in elderly patients and
those with hepatic dysfunction. No dosage adjustment is necessary in patients with
compromised renal function, although, as a general precaution, these patients should be
closely monitored.
2. Adverse Effects
Zolpidem's side effects are usually dose-related and include daytime drowsiness,
dizziness, headache, nausea, vomiting, and diarrhea. Approximately 4% of 1701 pa-
tients who received zolpidem at all doses (1.25-90 mg) in U.S. premarketing clinical
trials discontinued treatment because of an adverse clinical event.
A few patients have experienced zolpidem-related cognitive distortions and hallu-
cinations. 36- 38 Two cases of zolpidem-related sleepwalking suggest that somnam-
bulism may occur with imidazopyridines, as it does with the benzodiazepines. 39 .40
Increased wakefulness during the last third of the night, as measured by polysom-
nography, has not been observed in clinical trials.
Zolpidem impairs short-tenn and long-tenn memory, psychomotor perfonnance,
and balance at 1.5 hr after intake, but not at 6 hr. This disturbance is similar in
magnitude and course to that produced by triazolam in equivalent doses. 36
3. Withdrawal
Although not observed in studies of patients with insomnia who received doses of
zolpidem up to 20 mg at bedtime, withdrawal symptoms were reported in two patients
who raised their zolpidem dose up to 50 and 100 mg/day, because they had developed
tolerance to its hypnotic effects.41 Withdrawal effects experienced during the daytime
included anxiety, tremor, sweating, nausea, swallowing difficulties, abdominal pains,
tachycardia, and tachypnea. Treatment with benzodiazepines resolved these symptoms.
4. Drug Interactions
The systemic evaluation of zolpidem in combination with other drugs that affect
the eNS has been limited, and thus careful consideration should be given to the
pharmacology of any eNS-active drug to be used with zolpidem. Any drug that
depresses the eNS could potentially enhance the eNS-depressant effects of zolpidem.
Several case reports indicate a worsening of psychosis in some chronically psy-
chotic patients maintained on benzodiazepines when zolpidem is introduced. Zol-
pidem's sedative effects may be increased when imipramine, chlorpromazine, or alco-
hol are used concomitantly.
5. Overdose
In European postmarketing reports of overdose with zolpidem alone, impainnent
of consciousness has ranged from somnolence to light coma, with one case each of
280 JEFFREY B. WEILBURG, M.D., AND HEATHER S. HOPKINS
6. Pregnancy
Although studies to assess the effects of zolpidem on human reproduction and
development have not been conducted, teratology studies in rats show dose-related
maternal lethargy and ataxia and a dose-related trend to incomplete ossification of fetal
skull bones, which indicates a delay in maturation and is often seen in rats treated with
sedative-hypnotic drugs. In rabbits, dose-related maternal sedation and decreased
weight gain occurred at all doses of zolpidem tested. At the high dose of 16 mg
base/kg, there was an increase in postimplantation fetal loss and underossification of
sternebrae in viable fetuses. These fetal findings in rabbits are often secondary to
reductions in maternal weight gain. There were no frank teratogenic effects in either set
of studies.
Because animal reproduction studies are not always predictive of human response,
this drug should be used during pregnancy only if clearly needed. Children born of
mothers taking sedative-hypnotic drugs may be at some risk for withdrawal symptoms
from the drug during the postnatal period. In addition, neonatal flaccidity has been
reported in infants born of mothers who received sedative-hypnotic drugs during
pregnancy.
Studies in lactating mothers indicate that between 0.004 and 0.019% of the total
administered dose is excreted into milk, but the effect of zolpidem on the infant is
unknown. The use of zolpidem in nursing mothers is not recommended.
C. Antihistamines
Sedating antihistamines such as diphenhydramine (Nytol® and others) and hy-
droxyzine (Atarax® and others) are sometimes used to promote sleep in those who are
sensitive to or unable to use benzodiazepines, such as alcoholics, drug abusers, the
elderly, or medically ill hospitalized patients. Doxylamine (Unisom® and others) and
pyrilamine (Robitussin Night Relief® and others) are the active agents in over-the-
counter hypnotic antihistamines.
There are conflicting data on the hypnotic efficacy of these drugs, with some
studies indicating that they produce no significant change in the objective parameters
of sleep. Antihistamines may produce paradoxical excitement, delirium, psychosis, and
dystonic reactions. Elderly persons may be especially sensitive to these adverse effects.
Antihistamines are therefore not the first choice for treatment of insomnia and must be
used with caution. 42
D. Barbiturates
The short-acting and intermediate-acting barbiturates such as pentobarbital (Nem-
butal®) and secobarbital (Seconal®) are still marketed as hypnotics. Barbiturates de-
crease sleep latency, shorten slow-wave sleep, and noticeably decrease the length of
REM periods. With short-acting barbiturates, these effects tend to wear off later in the
night. Tolerance develops to the hypnotic effects of barbiturates over a matter of days,
and rebound phenomena occur following cessation of treatment. Tolerance, abuse,
dependence, and adverse effects (excessive next-day sedation at higher doses) occur
much more often with barbiturates than with benzodiazepines. Further, the lethality of a
barbiturate overdose is far greater than that of a benzodiazepine overdose. In general,
patients using these medications should be assisted in switching to less toxic agents
such as the benzodiazepines. Barbiturates may be used cautiously in some patients
when other agents have failed.
E. Chloral Hydrate
Chloral hydrate (Noctec® and others) decreases sleep latency and the number of
awakenings and may slightly decrease slow-wave sleep. It has relatively little effect on
REM sleep because some degree of tolerance develops. In fact, if used for more than 2
weeks, chloral hydrate's effectiveness diminishes. Therapeutic doses of chloral hydrate
have little effect on respiration and blood pressure, but toxic doses may impair both.
Chloral hydrate has an irritating and unpleasant taste. Local irritation of the
gastrointestinal tract can cause heartburn, nausea, and vomiting. Gastric irritation in-
creases with the dose but may be minimized by administering the drug with milk or an
antacid. The relatively short half-life of the chloral derivatives means that hangover is
seldom a problem. Allergic reactions are occasionally observed. Because of its poten-
tial to cause direct organ toxicity, chloral hydrate is best avoided in patients with severe
hepatic, renal, or cardiac disease.
Chloral hydrate affects the hepatic drug-metabolizing system and thus may inter-
282 JEFFREY B. WEILBURG, M.D., AND HEATHER S. HOPKINS
fere with some drugs that are metabolized in the liver. One of its metabolites, tri-
chloroacetic acid, displaces acidic drugs from plasma protein-binding sites, thereby
increasing their plasma concentrations. Chloral hydrate also may interact with the
diuretic furosemide (Lasix® and others) to cause a syndrome of sympathetic instability.
Combining chloral hydrate with alcohol increases blood levels of both, adding to the
sedative potentiation.
Most adults will require 1-2 g of chloral hydrate to induce sleep. The toxic dose is
generally around 10 g, but death has occurred with doses of 4 g. Gastric irritation may
limit the amount a patient can take at one time, and vomiting diminishes absorption.
However, gastric necrosis can produce additional complications. In serious intoxica-
tion, support of cardiovascular and respiratory functions is vital. Hemodialysis may
enhance excretion.
Tolerance, physical dependence, and addiction are possible with chloral deriva-
tives. A withdrawal syndrome occurs similar to that seen with other sedatives and
alcohol, involving disrupted sleep and intense nightmares. If a patient has previously
used a chloral derivative and prefers it, then careful administration is acceptable. Some
centers find chloral hydrate useful for alcoholics who complain of insomnia during
inpatient alcohol detoxification. However, there is generally little reason to initiate
treatment with it.
F. Ethchlorvynol
Ethchlorvynol (Placidyl®) has a rapid onset of hypnotic action (15-30 min), and
although its elimination half-life is 10-25 hr, rapid distribution terminates its clinical
effect within a few hours after an acute dose. Chronic dosing can lead to prolonged
effects, but tolerance tends to develop to the sedative effects with regular use. Ethchlor-
vynol resembles the barbiturates in its effect on sleep stages, including suppression of
REM. Rebound occurs following drug withdrawal.
Adverse effects with ethchlorvynol include confusion, hangover, ataxia, nausea
and vomiting, hypotension, facial numbness, amblyopia, and giddiness. Less common
are allergic reactions, blood dyscrasias, and cholestatic jaundice.
Ethchlorvynol interferes with the kinetics of some drugs metabolized by hepatic
isoenzymes. When ethchlorvynol is combined with other sedating drugs or alcohol,
additive sedative effects can occur. Delirium has been reported in patients taking
ethchlorvynol with tricyclic antidepressants. Ethchlorvynol is contraindicated in pa-
tients with acute intermittent porphyria.
Ethchlorvynol is much more toxic than most other hypnotics. The lethal dose
range is considered to be 10-25 g, but death has been reported after a dose as low as
2.5 g. Profound coma can follow an acute overdose. Treatment includes the usual
approaches to diminishing absorption, supporting vital functions, and hastening elim-
ination, possibly using a lipid dialysate with hemodialysis.
There is a high probability that ethchlorvynol use will produce physical dependen-
cy. The abuse potential of this agent is likewise high, as is its lethality in overdose.
Therefore, treatment with this drug should not be initiated, and when patients are
already using it, strenuous efforts should be made to withdraw the drug or to find a
benzodiazepine substitute.
6. INSOMNIA 283
G. Over-the-Counter Drugs
There are two current forms of over-the-counter sleep aids: the pure hypnotics,
which usually contain sedative antihistamine (diphenhydramine or doxylamine), and
the "nighttime analgesics," which also contain a salicylate or acetaminophen.
Salicylates in over-the-counter sleeping aids include salicylamide, aspirin, and
sodium salicylate. The effectiveness of these medications is questionable, although it is
possible that their analgesic activity may promote sleep by alleviating pain in affected
patients. Typical problems with salicylates include gastric irritation and interference
with clotting mechanisms. Intoxication with salicylates can produce tinnitus and acid-
base disturbances. Intoxication with acetaminophen may result in hepatic necrosis.
There is little evidence that any available over-the-counter sleeping pill is more
effective than placebo. Furthermore, these drugs are not innocuous, and some patients
may be particularly sensitive to their adverse effects. However, they are very widely
used, and further study is needed before clear guidelines on their use can be issued.
Melatonin, a hormone secreted by the pineal gland, may be responsible for the
entrainment of human sleep rhythms to light-dark cycles. Available over the counter in
a pill form, melatonin has become popular as a "natural" remedy for insomnia and jet
lag. Preliminary evidence suggests that melatonin may have some benefit for transient
sleep disturbances, but further studies of its safety and efficacy need to be conducted
before its use can be recommended. 43 The purity of the product now available in the
United States, the hypnotic dose, and the adverse effects of taking melatonin are all
unknown.
P.M. and/or time-limited use of benzodiazepine hypnotics may help. Recent evidence
suggests that trazodone, which is very sedating, may be used safely and effectively to
treat the insomnia caused by stimulating antidepressants, such as the SSRIs.
2. Antipsychotics
Antipsychotics can be useful in managing insomnia when agitation, either from
delirium or psychosis, is disrupting sleep. For hospitalized patients, especially those in
intensive-care units, antipsychotics may be used alone or in combination with ben-
zodiazepines. A low-dose antipsychotic, such as haloperidol (Haldol® and others), 0.5
mg, may be better than a benzodiazepine in managing nocturnal agitation in demented
elderly patients. Antipsychotics can induce nocturnal myoclonus, PLMS, somnambul-
ism, and nightmares. Small amounts of haloperidol probably do not worsen respiration
in patients with sleep apnea.
3. Buspirone
Preliminary data show that buspirone (BuSpar®) has no immediate effect on sleep
latency or sleep architecture in patients with generalized anxiety disorder but that a
decrease in nocturnal time awake, with consequent improvement in sleep efficiency,
does appear in these patients after 5 weeks of chronic administration. Buspirone is not a
hypnotic but may be useful in managing insomnia secondary to anxiety.
4. Clonidine
Studies suggest that clonidine (Catapres®) may be helpful in relieving stimulant-
induced insomnia in patients with attention deficit hyperactivity disorder. Doses of 0.1
mg at bedtime will not interfere with daytime blood pressure. 45
with a slower onset of action may be abused less often than those whose effects are felt
more rapidly. The lethality of benzodiazepine and related hypnotics in overdose is
relatively limited, except when these drugs are combined with alcohol or other seda-
tives.
3. Use sedative-hypnotic medications primarily for treating transient insomnia,
i.e., insomnia lasting days to weeks that is related to specific, situational emotional
stress, medical illness, or sleep-wake schedule changes. The long-term use of seda-
tive-hypnotic medications in patients with chronic insomnia remains controversial. In
general, these medications should be used for short periods (i.e., less than 2-3 weeks)
and preferably not on consecutive nights. When long-term use is needed, the patient
should be carefully monitored to avoid dosage escalation and to note the appearance of
new or iatrogenic problems. There should also be periodic reassessments of diagnosis
(with special attention to psychosocial, behavioral, and environmental factors) and
treatment regimen.
4. Use the lowest effective dose. Older patients, who have higher rates of sleep
apnea, greater sensitivity to adverse eNS effects, and somewhat slowed drug elimina-
tion rates and therefore drug accumulation, typically require extremely low doses.
Patients with insomnia secondary to medical disorders, especially those with neuro-
logic, cardiac, pulmonary, or hepatic disease, may be sensitive to respiratory depres-
sion and are more likely to suffer confusional effects from all sedatives. They also may
require lower dosages and careful monitoring.
5. Behavioral techniques and compliance with the rules of good sleep hygiene
should often be tried before medications are used and should be used concomitantly
with medications, especially for those with long-standing insomnia.
Several factors should be considered when developing a clinical algorithm for
choosing a medication for treating insomnia. Patients with trouble falling asleep
(versus staying asleep) should initially be given an agent with a short or intermediate
duration of action and a rapid onset of action, such as triazolam, estazolam, zolpidem,
lorazepam, or diazepam. Doses of triazolam may start with 0.125 mg and should not
exceed 0.5 mg (0.25 mg in the elderly). Triazolam should be tapered to avoid rebound
by cutting the dose in half every 2 days. Zolpidem in doses of 5 mg reliably decreases
sleep latency; higher doses may be employed to improve sleep efficiency. Estazolam
has been shown in some studies to have limited efficacy in reducing sleep latency, but
it may be quite useful for increasing total sleep time, decreasing nocturnal awakenings,
and improving next-day subjective status. Lorazepam may be administered sub-
lingually to shorten the onset of activity.
Patients with difficulty staying asleep (nocturnal awakenings, shortened total
sleep time) or those with daytime anxiety or tendencies toward rebound insomnia
or anxiety may benefit from an agent with an intermediate or long duration of action,
such as estazolam, temazepam, flurazepam, quazepam, or clonazepam. Some patients
find a flurazepam dose of 15 mg as effective as one of 30 mg, so the lower dose should
be used to start. Quazepam may be started at 15 mg for adults and at 7.5 mg in the
elderly. Sedating tricyclic antidepressants are often also quite useful, alone or with
benzodiazepines, in improving sleep continuity disturbances, especially in anxious or
dysthymic patients.
Rarely, if ever, should treatment be initiated with barbiturates, chloral hydrate, or
286 JEFFREY B. WEILBURG, M.D., AND HEATHER S. HOPKINS
ethchlorvynol. If patients with chronic insomnia who are using these agents are seen
because they are still having difficulty sleeping, every effort should be made to switch
the patient to benzodiazepines, antidepressants, and behavioral techniques, alone or in
combination.
VI. CONCLUSION
REFERENCES
I. American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders, 4th ed.
Washington, D.C., American Psychiatric Association, 1994.
2. Weilburg 1. B.: Approach to the patient with insomnia, in Goroll A. H., May L. A., Mully A. G. (eds):
Primary Care Medicine. Philadelphia, JB Lippincott, 1987.
3. Coleman R. M., Roffwarg H. P., Kennedy S. J., et al: Sleep-wake disorders based on a polysom-
nographic diagnosis: A national cooperative study. lAMA 27:997-1003, 1982.
4. Monane M., Glynn R. J., Avorn J.: The impact of sedative-hypnotic use on sleep symptoms in elderly
nursing home residents. Clin Pharmacol Ther 59:83-92, 1996.
5. Mellinger G. D., Balter M. B., Uhlenhuth E. H.: Insomnia and its treatment: Prevalence and correlates.
Arch Gen Psychiatry 42:225-232, 1985.
6. Stoller M. K.: Economic effects of insomnia. CZin Ther 16:873-897, 1994.
7. Walsh J. K., Engelhardt C. L., Hartmann P. G.: The direct economic cost of insomnia, in Nutt D.,
Mendleson W. (eds): Hypnotics and Anxiolytics: Baillieres Clinical Psychiatry. London, Bailliere Tin-
dall, 1995, pp 369-381.
8. Balter M. 8., Uhlenhuth E. H.: The beneficial and adverse effects of hypnotics. 1 Clin Psychiatry 52: 16-
23, 1991.
9. Erman M. K.: Sleep disorders. Psychiatr Clin North Am 10:4, 1987.
10. Weitzman E. D., Czeisler C. A., Coleman R. M., et al: Delayed sleep phase syndrome. Arch Gen
Psychiatry 38:737-746, 1981.
II. Czeisler C. A., Richardson G. S., Coleman R. M., et a1: Chronotherapy: Resetting the circadian clocks of
patients with delayed sleep phase insomnia. Sleep 4:1-21, 1981.
12. Reynolds C. E, Taska L. S., Sewitch D. E.: Persistent psychophysiologic insomnia: Preliminary research
diagnostic criteria and EEG sleep data. Am 1 Psychiatry 141:804-805, 1984.
13. Hauri P.: Treating psychophysiologic insomnia with biofeedback. Arch Gen Psychiatry 38:752-758,
1981.
14. Kramer P. D.: Insomnia: Importance of the differential diagnosis. Psychosomatics 23:129-137, 1982.
15. Thorpy M. J., Diagnostic Classification Steering Committee: International Classification of Sleep Disor-
ders: Diagnostic and Coding Manual. Rochester, Minnesota, American Sleep Disorders Association,
1990.
16. Reynolds C. E, Kupfer D. J., Hoch C. c., et al: Sleep of healthy seniors: A revisit. Sleep 8:20-30, 1985.
17. Reynolds C. E, Kupfer D. 1., Hoch C. C., et al: Sleeping pills for the elderly: Are they ever justified? 1
Clin Psychiatry 46:9-12, 1985.
18. Soldatos C. R., Kales 1. D., Scharf M. B., et al: Cigarette smoking associated with sleep difficulty.
Science 207:551-553, 1980.
19. Abernethy D. R., Greenblatt D. J., Shader R.I.: Benzodiazepine hypnotic metabolism: Drug interactions
and clinical indications. Acta Psychiatr Scand 332:32-38, 1986.
6. INSOMNIA 287
20. Greenblatt D. J., Abernethy D. R, Shader R I.: Benzodiazepine hypnotics: Kinetic and therapeutic
options. Sleep 5:S18-S27, 1982.
21. Greenblatt D. J., Harmatz J. S., Engelhardt N., et al: Pharmacokinetic determinants of dynamic differ-
ences among three benzodiazepine hypnotics. Arch Gen Psychiatry 46:326-332, 1989.
22. Kales A., ScharfM. 8., Kales J. D., et al: Rebound insomnia. lAMA 241:1692-1695,1979.
23. Kales A., Soldatos C. R, Bixler E. 0.: Rebound insomnia and rebound anxiety: A review. Pharmacology
26:121-137,1983.
24. Greenblatt D. J., Harmatz J. S., Zinny M. A., et al: Effect of gradual withdrawal on the rebound sleep
disorder after discontinuation of triazolam. N Engl 1 Med 317:722-728, 1987.
25. Church M. W., Johnson L. C.: Mood and performance of poor sleepers during repeated use of flur-
azepam. Psychopharmacology 61:309-316, 1979.
26. Morris H. H., III., Estes M. L.: Traveler's amnesia: Transient global amnesia secondary to triazolam.
lAMA 258:945-946, 1987.
27. Spinweber C. L., Johnson L. c.: Effects of triazolam on sleep performance, memory and arousal
threshold. Psychopharmacology (Berlin) 76:5-12, 1982.
28. Wysowski D. K., Barash D.: Adverse behavioral reactions attributed to triazolam in the Food and Drug
Administration's Spontaneous Reporting System. Arch Intern Med 151:2003-2008, 1991.
29. Rothschild A. J., Bessette M. P., Carter-CampbeU J., Murray M.: Triazolam and disinhibition. Lancet
341:186, 1993.
30. Steens R. D., Pouliot Z., Millar T w., Kryger M. H., George C. E: Effects of zolpidem and triazolam on sleep
and respiration in mild to moderate chronic obstructive pulmonary disease. Sleep 16:318-326, 1993.
31. Guilleminault C., Clerk A., Labanowski M., Simons J., Stoohs R: Cardiac failure and benzodiazepines.
Sleep 16:524-528, 1993.
32. Herings R M. c., Stricker B. H. C., de Boer A., Bakker A., Sturmans E: Benzodiazepines and the risk of
fa11ing leading to femur fractures: Dosage more important than elimination half-life. Arch Intern Med
155:1801-1807, 1995.
33. Jackson C. W., Markowitz J. S., Brewerton T D.: Delirium associated with clozapine and ben-
zodiazepine combinations. Ann Clin Psychiatry 7:139-141,1995.
34. Merlotti L., Roehrs T., Koshorek G., Zorick E, Lamphere J., Roth T.: The dose effects of zolpidem on
the sleep of healthy normals. 1 Clin PsychopharmacoI9:9-14, 1989.
35. Kryger M. H., Steljes D., Pouliot Z., Neufeld H., Odynski T: Subjective versus objective evaluation of
hypnotic efficacy: Experience with zolpidem. Sleep 14:399-407, 1991.
36. Berlin I., Warot D., Hergueta T., Molinier P., Bagot C., Puech A. J.: Comparison of the effects of
zolpidem and triazolam on memory functions, psychomotor performances, and postural sway in healthy
subjects. 1 Clin Psychopharmacol 13:100-106, 1993.
37. Ansseau M., Pitchot W., Hansenne M., Moreno A. G.: Psychotic reactions to zolpidem. Lancet 339:809,
1992.
38. Iruela L. M., Ibanez-Rojo v., Baca E.: Zolpidem-induced macropsia in anorexic woman. Lancet
342:443-444, 1993.
39. Mendelson W. B.: Sleepwalking associated with zolpidem. 1 Clin Psychopharmacol 14:150, 1994.
40. Sauvanet J. P., Maarek L., Roger M., Renaudin J., Louvel E., Orofiamma 8.: Open long-term trials with
zolpidem in insomnia, in Sauvanet JP, Langer SZ, Morse1li PL (eds): Imidazopyridines in Sleep Disor-
ders. New York, Raven Press; 1988, pp 339-349.
41. CavaUaro R, Regazzetti M. G., CoveUi G., Smeraldi E.: Tolerance and withdrawal with zolpidem.
Lancet 342:374-375, 1993.
42. Roehers T A., Tietz E. I., Zoerck E J., et al: Daytime sleepiness and antihistamines. Sleep 7: 137 -141,
1984.
43. Anonymous: Melatonin. Med Lett 37:111-112,1995.
44. Ware J. c.: Tricyclic antidepressants in the treatment of insomnia. 1 Clin Psychiatry 44:25-28, 1983.
45. Wilens T. E., Biederman J., Spencer T.: Clonidine for sleep disturbances associated with attention deficit
hyperactivity disorder. lAm Acad Child Adolesc Psychiatry 33:424-426, 1994.
46. Consensus conference: Drugs and insomnia: The use of medications to promote sleep. lAMA 251 :2410-
2414, 1984.
III
Psychoactive Substance Abuse
7
Psychoactive Substance
Use Disorders
The self-administration of chemical substances that alter mood or thinking has been a
widespread practice for centuries. When the use of such substances deviates from
societal norms or is deemed harmful to physical or mental health, substance use may
merge into substance abuse and dependence. There are wide cultural variations in
acceptable substance use. According to the fourth edition of the Diagnostic and Statis-
tical Manual of Mental Disorders (DSM-IV), a substance use disorder is characterized
by regular use of psychoactive substances leading to symptoms and maladaptive be-
havioral changes that would be viewed as undesirable in almost all cultures. These
disorders are categorized as either substance dependence or substance abuse.
Substance dependence is described in the DSM-IV as a cluster of cognitive,
behavioral, and physiological symptoms that indicate that a person has impaired con-
trol of substance use and continues this despite adverse consequences. Dependence is
characterized by at least three of the following symptoms occurring at any time in the
same 12-month period: (1) tolerance, as defined either by a need for increased amounts
of the substance to achieve intoxication or the desired effect or by markedly diminished
effect with continued use of the same amount of the substance; (2) withdrawal, as
Shelly F. Greenfield, M.D., M.P.H. • McLean Hospital, Belmont, Massachusetts 02178; Department of
Psychiatry, Harvard Medical School, Boston, Massachusetts 02115. Roger D. Weiss, M.D. • Alco-
hol and Drug Abuse Program, McLean Hospital, Belmont, Massachusetts 02178; Department of Psychiatry,
Harvard Medical School, Boston, Massachusetts 02115. Steven M. Mirin, M.D. • Mclean Hospital,
Belmont, Massachusetts 02178; Department of Psychiatry, Harvard Medical School, Boston, Massachusetts
02115.
The Practitioner's Guide to Psychoactive Drugs (Fourth Edition), Gelenberg and Bassuk, eds.
Plenum Publishing Corporation, New York, 1998
291
292 SHELLY F. GREENFIELD, M.D., M.P.H., ET AL
manifested either by the characteristic withdrawal syndrome for the substance or by the
fact that the substance is taken to relieve or avoid withdrawal symptoms; (3) taking the
substance in larger amounts over a longer period than was intended; (4) a persistent
desire or unsuccessful efforts to cut down or control substance use; (5) spending a great
deal of time in activities necessary to obtain, use, or recover from the substance; (6)
giving up or reducing important social, occupational, or recreational activities because
of substance use; and (7) continued substance use despite knowledge of having a
persistent or recurrent physical or psychological problem that is likely to have been
caused or exacerbated by the substance. The first two characteristics may not apply to
cannabis, hallucinogens, or phencyclidine. With all other substances, when the criteria
include either of the first two characteristics, the diagnosis of substance dependence
with physiological dependence is made. If there is no evidence of either tolerance or
withdrawal, the diagnosis of substance dependence without physiological dependence
is made.
Psychoactive substance abuse is present when a person has never met criteria for
substance dependence but has a maladaptive pattern of psychoactive substance use
leading to clinically significant impairment or distress, as manifested by one or more of
the following characteristics, occurring within a 12-month time period: (1) recurrent
substance use resulting in a failure to fulfill major role obligations at work, school, or
home; (2) recurrent use in situations in which use is physically hazardous; (3) recurrent
substance-related legal problems; or (4) continued use despite knowledge of having
persistent or recurrent social or interpersonal problems caused or exacerbated by use of
the psychoactive substance.
When no criteria for either dependence or abuse have been present for at least one
month, the disorder can be said to be in remission. If the interval of time between
cessation of substance dependence or abuse is less than one year, this is considered to
be early remission; if the period of time is greater than one year, this is considered to be
sustained remission. If during early or sustained remission, no criteria of dependence or
abuse have been present, this is full remission; if at least one criterion of dependence or
abuse has been met, intermittently or continuously during remission, this is partial
remission. Remission can also be further specified as being characterized by the indi-
vidual's being "on agonist or antagonist therapy" (e.g., prescribed methadone or
naltrexone) or "in a controlled environment" (e.g., in residential treatment or in a
halfway house).
From a pharmacological perspective, drug dependence may result from the direct
reinforcing effects of these agents on the central nervous system. The mood-altering
effects of many drugs such as relief of anxiety or production of euphoria are mediated
through their effects on various brain neurotransmitters. Physical dependence implies a
state of altered cellular physiology caused by repetitive drug administration. Repetitive
use of some agents may also result in drug tolerance, a state in which increasingly
larger doses of the drug must be taken to reproduce its original pharmacological effect.
Tolerance may result from an increased rate of drug metabolism or a change in the
sensitivity of target cells to the drug's pharmacological effect. In addition, inducing
tolerance to one member of a particular drug class may be accompanied by develop-
7. PSYCHOACTIVE SUBSTANCE USE DISORDERS 293
ing tolerance to other drugs in the same general class. This phenomenon is called
"cross-tolerance."
In individuals in whom dependence has developed, abrupt withdrawal of that
agent may precipitate a characteristic abstinence syndrome. The wish to avoid absti-
nence symptoms can be a powerful force in maintaining drug-taking behavior. The
experience of withdrawal, however, may cause some individuals to seek medical treat-
ment. The clinical evaluation of such individuals should ideally include a detailed
medical and psychiatric history. The clinician should distinguish emotional problems
preceding drug abuse from those that may be drug induced. Because drug dependence
and withdrawal often are accompanied by symptoms similar to those seen in patients
with other psychiatric disorders (e.g., anxiety, insomnia, depression), a careful devel-
opmental and family history, and perhaps information from significant others, may be
extremely useful. As in other areas of medicine, proper treatment depends on accurate
diagnosis. This chapter reviews each major class of commonly abused drugs with
respect to basic pharmacology, patterns of abuse, and currently held views about
treatment. The use of psychoactive substances during pregnancy and its effects on the
neonate are beyond the scope of this chapter, and the reader is referred to other texts for
additional information on these topics.
A. Classification
We can divide the opioids into the following categories based on their derivation:
1. Natural alkaloids of opium derived from the resin of the opium poppy, includ-
ing opium, morphine, and codeine.
2. Semisynthetic derivatives of morphine, including diacetylmorphine (heroin),
hydromorphone (Dilaudid® and others), and oxycodone (Percodan® and oth-
ers).
3. Purely synthetic opioids, which are not derivatives of morphine, including
meperidine (Demerol® and others), methadone (Dolophine® and others), and
propoxyphene (Darvon® and others).
4. Opioid-containing preparations, such as elixir of terpin hydrate with codeine
and paregoric, having abuse liability.
In the United States, at least 20 drugs are available that have opioid actions, and
other such drugs are available in other countries. The semisynthetic derivatives, like
heroin and oxycodone, and pure synthetics, like meperidine and methadone, are most
commonly abused. A 1990 national survey of high school students by the National
Institute on Drug Abuse found that 1.3% of all high school seniors had used heroin. As
of 1993, the Drug Enforcement Agency noted that heroin available on the streets was
38% pure; in contrast, in 1983 it was only 4% pure. This phenomenon, combined with
the greater availability of heroin, has contributed to a rise in the past 5 years in the rate
294 SHELLY F. GREENFIELD, M.D., M.P.H., ET AI..
B. Pharmacology
The opioid drugs are readily absorbed from the gastrointestinal tract, nasal mu-
cosa, and lung, but parenteral administration more rapidly elevates blood levels and
produces intoxication. For example, following intravenous administration of heroin,
the drug is almost immediately hydrolyzed to morphine by the liver, with peak plasma
morphine levels attained in approximately 30 min. The drug then rapidly leaves the
blood and is concentrated in body tissues. Only small quantities of injected opioids
cross the blood-brain barrier, but plasma levels correlate directly with the level of
intoxication. Morphine is metabolized primarily by conjugation in the liver. Its half-life
(Le., the amount of time necessary for one-half of a given dose to be cleared from the
blood) is approximately 2Y2 hr, and 90% of the total dose administered is excreted in
the urine within the fIrst 24 hr. The remainder is excreted through the biliary tract and
appears in the feces.
The opioids act by selectively binding to stereospecifIc neuronal receptor sites,
which thus far have been identifIed in the brain (e.g., the hypothalamus) and gut.Three
types of opioid receptors (8, /-L, and K) have been identifIed and isolated. These opioid
receptors probably mediate the action of naturally occurring opiate peptides (Le., the
enkephalins and j3-endorphin) as well. Opioid drugs are categorized by their ability to
bind to, and activate, these receptor types. Compounds that bind and activate a receptor
are called agonists; those that bind to but do not activate the receptor are called
antagonists.
C. Acute Effects
1. Central
Acute administration of heroin, especially when taken intravenously, produces a
"rush" lasting 30-60 sec. This is usually followed by a brief period of euphoria
accompanied by a profound sense of tranquility. This state may last for several hours,
depending on the dose administered and the plasma level of morphine attained. It is
characterized by drowsiness ("nodding"), lability of mood, mental clouding, apathy,
and motor retardation. Respiratory depression (Le., diminished volume and slower rate)
secondary to inhibition of the brainstem respiratory center also occurs. Stimulation of
the brainstem chemoreceptor trigger zone for emesis may produce nausea and vomit-
ing. However, with repetitive use, tolerance develops to this effect.
2. Peripheral
In the healthy patient, the cardiovascular effects of opioids like morphine are
minimal. In some patients, however, peripheral vasodilation may contribute to ortho-
static hypotension. These drugs also decrease secretions in the stomach, biliary tract,
and pancreas and inhibit the contractility of smooth muscle. As a result of the latter
7. PSYCHOACTIVE SUBSTANCE USE DISORDERS 295
2. Clinical Manifestations
Patients who overdose on illicitly obtained opioid drugs frequently are alone, and
the untoward effects are immediate. Thus, diagnosis and medical intervention often
occur too late. When these patients present for treatment, they may be in stupor or
coma with constricted pupils and diminished pulse and respiration. Hypothermia and
pulmonary edema (usually noncardiogenic) are seen in severe cases.
The clinician must differentiate opioid overdose from other causes of respiratory
depression and coma. Although pupillary constriction usually is a reliable sign, it also
296 SHELLY F. GREENFIELD, M.D., M.P.H., ET AL.
3. Management
a. General Life-Support Measures. The first task in treating acute opioid poi-
soning is to establish an adequate airway. Aspiration can be prevented either by placing
the patient on his or her side or by using a cuffed endotracheal tube. Gastric lavage
should be performed if the clinician suspects recent oral intake of opioids or other
drugs. If hypoglycemia complicates the clinical picture, establish a reliable intravenous
route and slowly infuse a 50% solution of glucose in water.
c. Other Considerations. Patients who use opioids tend to abuse other drugs as
well, particularly alcohol, cocaine, and antianxiety drugs. For this reason, the clinician
should obtain blood and urine from all patients for toxicological screening. On recov-
ery, the degree of suicidal intent associated with overdose should be assessed; prior to
discharge, patients can be referred to an appropriate treatment facility.
7. PSYCHOACTIVE SUBSTANCE USE DISORDERS 297
F. Opioid Withdrawal
1. Clinical Manifestations
The opioid-dependent patient often comes for treatment when the quality or
availability of heroin or other opioids declines on "the street," making it difficult to
stave off symptoms of withdrawal. When abstinence symptoms occur, their severity
depends on the type of opioid previously used, degree of tolerance, time elapsed since
the last dose, and emotional meaning of the symptoms. The symptoms reflect increased
excitability in organs previously depressed by chronic opioid use.
In the early stages of withdrawal from heroin (Le., 6-12 hr after the last dose), the
patient may yawn and sweat, his or her nose and eyes may run, and he or she may
experience considerable anxiety. Craving for opioids and drug-seeking behavior inten-
sify. As withdrawal progresses, pupils dilate, and the patient may develop gooseflesh,
hot and cold flashes, loss of appetite, muscle cramps, tremor, and insomnia or a restless
tossing sleep. After 18-24 hr have elapsed since the last dose of heroin, nausea,
vomiting, and elevations in blood pressure, pulse, respiratory rate, and temperature
occur. After 24-36 hr, diarrhea and dehydration may develop.
In the case of heroin or morphine, abstinence symptoms generally peak 48-72 hr
after the last dose. By this time, laboratory values reflect the clinical process. Leuko-
cytosis is common, and ketosis and electrolyte imbalance may develop as a result of
dehydration. In untreated cases, most clinical symptoms usually disappear within 7-10
days, although physiological disturbances, such as insomnia and malaise, may be
detected for weeks to months afterward. Withdrawal from methadone is characterized
by slower onset (24-48 hr after the last dose) and more gradual resolution of symptoms
(more than 2 weeks). Indeed, patients often complain of fatigue, weakness"and insom-
nia for several months after stopping the drug. When naloxone is used to displace
methadone, the onset of withdrawal occurs within minutes and a brief (30-60 min)
withdrawal syndrome develops. This subsides again when naloxone is excreted and
methadone that is still present in the body binds to the receptors. Also, although more
observable signs of withdrawal (e.g., pupillary dilation, sweating, yawning) follow the
time courses just specified, subjective symptoms of withdrawal (muscle aches, bone
pain, dysthymia) may begin earlier after cessation of the drug and persist longer.
Finally, in some individuals a protracted abstinence syndrome has been described. This
refers to the persistence of subjective distress and deviations from baseline physiologic
measures (e.g., respiration, temperature) over a period of months. It is postulated that
such a syndrome may contribute to relapse in some individuals.
2. Management
a. Medical Detoxification. The objectives of medically managed detoxification
are to make the syndrome more tolerable to the individual and to initiate the individu-
al's engagement in treatment, rehabilitation, and relapse prevention. The technique
selected may depend on the level of physical discomfort the individual is willing or
able to tolerate, as well as the support system available.
298 SHELLY F. GREENFIELD, M.D., M.P.H., ET AL
I. Observe the patient for objective signs of opioid withdrawal. The patient should exhibit two of
the following four signs: (a) pulse 10 beats/min over baseline, or (if baseline is unknown) greater than
90 beats/ min in the absence of a history of tachycardia; (b) systolic blood pressure greater than 10 mm
Hg over baseline or greater than 160/95 mm Hg in the absence of known hypertension; (c) dilated
pupils; (d) sweating, gooseflesh, rhinorrhea, or lacrimation. When at least two of the four signs are
present, administer methadone, 10 mg PO. Signs of withdrawal should abate, at least temporarily, within
90 min. If they do not, the patient's symptoms may not be caused by opioid withdrawal.
2. The evaluation procedure should be repeated every 2-4 hr, and methadone administered as often as
necessary for 24 hr. Only rarely should a patient receive more than 40 mg of methadone during this
period, unless he or she has been maintained on methadone at higher doses prior to detoxification.
3. The total dose of methadone administered during the first 24 hr (the stabilization dose) should be given
the next day in two divided doses.
4. Beginning on the third day, taper the methadone by 5 mg per day, beginning with the morning dose, until
the patient is completely withdrawn.
5. All methadone should be dispensed in crushed or liquid form and should be consumed under direct
supervision in order to avoid illegal diversion of the drug onto the street.
effects, and/or generalized distrust of the treatment program. More pertinent, however,
is the patients' ambivalence about giving up opioid use and the fact that tolerance
develops rapidly to methadone's euphoriant and analgesic effects. Thus, once the
dosage drops to 15 or 20 mg per day, patients may experience a return of abstinence
symptoms. Patients should be informed at the outset that they cannot expect a symp-
tom-free withdrawal period.
severe hypertension. Clonidine has been shown to have efficacy equal to methadone in
achieving successful short-term detoxification and 6-month abstinence. It has been
seen to have a somewhat greater rate of success in inpatient than in outpatient settings.
The most common side effects of clonidine are hypotension and sedation. In inpatient
settings, it is possible to use higher doses of clonidine to achieve greater blockade of
withdrawal symptoms such as insomnia and still monitor and support hypotensive side
effects. Sedation is also less problematic for the inpatient than the outpatient, who may
be trying to maintain routine activities while in the process of detoxification. The
advantages of clonidine detoxification are several. It is not a narcotic or controlled
substance and thus is usable in a wider range of settings. Second, the induction of
opioid antagonists may be achieved more rapidly following clonidine detoxification.
The disadvantages outlined above may include insomnia and intolerable hypotension
and/or sedation. Furthermore, the use of clonidine for treating opioid withdrawal is not
an approved use of the drug at the time of this writing.
phine does not lead to significant dependence and can be stopped with only moderate
withdrawal symptoms, further work with this drug is required before it can be recom-
mended for widespread use.
are HIV positive, the tuberculin test should include antigen testing because HIV-
positive individuals may give a false-negative PPD.
2. Treatment Approaches
a. General Considerations. Chronic opioid users who enter treatment are often
both physically ill and psychosocially impaired. The clinician should view opioid
dependence as treatable with the constant potential for relapse. There is no known
treatment that is completely effective for all individuals addicted to opioids, but a
number of treatments are effective for different people. Patients often require a wide
range of medical and rehabilitative services. These may include remedial education,
job training and placement, legal assistance, and individual, group, and family therapy,
as well as pharmacological intervention.
In most cases, a multi modal approach will be most helpful, but a treatment
program must be individually designed for each patient. Some of the factors that must
be assessed when planning treatment are the degree and severity of psychological
problems and coexisting psychiatric illness (especially mood disorders). Depending on
the individual's previous educational and vocational status, the patient may need either
rehabilitation or habilitation. These patients must often change their lifestyles to dimin-
ish stimuli that may lead to drug craving, and the clinician must be alert to coexisting
polydrug use that may require additional intervention.
Treatment programs may include a combination of inpatient detoxification and
rehabilitation, outpatient drug-free programs, methadone maintenance, support groups
such as Narcotics Anonymous, therapeutic communities, narcotic antagonists, and
individual, group, and family therapy. Although motivation on the part of the individual
is, of course, important, programs can help maintain motivation by helping the patient
gain positive rewards from a drug-free life, thus causing more substantial losses if a
relapse occurs.
b. Methadone Maintenance
i. Background and Theory. As a treatment modality, methadone maintenance
evolved out of the clinical observation that some individuals feel unable to abstain
totally from opioid drugs. Its theoretical justification derives from the hypothesis that
chronic opioid use induces long-lasting, perhaps permanent, physiological changes at
the cellular level. As a result, opioid addicts may experience "narcotic hunger" for
months or years after withdrawal; the need to satisfy this hunger results in relapse. In
this context, medically supervised maintenance on methadone, in doses high enough to
satisfy narcotic hunger, becomes a logical alternative to psychological and physical
dependence on heroin or other illicitly obtained opioid drugs. The primary goals of
maintenance treatment are to achieve a stable maintenance dose that reduces craving
and illicit use and to help the patient engage in a comprehensive treatment program to
prevent relapse and promote rehabilitation.
ii. Clinical Application. See Section II.E2.hj for a review of the pharmacology of
methadone. Because methadone maintenance leads to the development of physical
dependence on methadone, federal guidelines limit eligibility to patients with opioid
7. PSYCHOACTIVE SUBSTANCE USE DISORDERS 303
dependence for more than one year who also demonstrate physiologic dependence. No
single dose of methadone is optimal for all patients. Methadone maintenance usually
entails administration of 40-120 mg PO per day. Doses exceeding 60 mg per day are
generally considered high-dose methadone maintenance; low-dose maintenance ranges
between 40 and 60 mg per day. Optimal blood levels in a methadone maintenance
patient are between 150 and 600 ng/mL.19 The advantage oflow-dose maintenance is
the fact that it theoretically limits side effects such as drowsiness while providing
sufficient receptor blockade to halt craving. High-dose maintenance theoretically pro-
vides receptor blockade that is more difficult to override with heroin and may, there-
fore, decrease illicit drug use. There is some evidence that higher doses may be
associated with better retention and improved outcome. 30
After determining that the patient is opioid dependent (i.e., by observing objective
signs of withdrawal coupled with a urine specimen positive for the presence of
opioids), the clinician should determine a methadone stabilization dose in the manner
described in Section II.F.2.b.iii. Thereafter, the daily dose is adjusted to a level (usually
greater than 40 mg) at which the patient reports complete blockade of the effects of
subsequently administered opioid drugs (e.g., heroin) and random urine samples indi-
cate that the use of opioids other than methadone has ceased. If the methadone mainte-
nance dose is being calculated on an outpatient basis, it is important that the patient be
observed for 1-2 hr after the initial dose for signs of sedation, and it is preferable to err
on the lower side for the initial dose.
The treatment goals for methadone maintenance include suppression of opioid
withdrawal, absence of sedation or euphoria, and elimination of other opioid use. When
methadone is taken regularly, tolerance develops to its euphoriant effects, and patients
generally function without psychomotor impairment or evidence of intoxication. Some
patients, however, continue to complain of sedation, excessive sweating, constipation,
ankle edema, and decreased interest in sex. The latter is associated with a drug-induced
reduction in plasma testosterone, especially at higher doses.
Patients on methadone maintenance generally come to the clinic daily to receive
the drug and do not take home any doses until considerable movement toward reha-
bilitation is noted. Compulsory clinic attendance makes the patient available for partic-
ipation in individual and/or group counseling, which may be a precondition to receiv-
ing methadone. Treatment programs are variable in their tolerance of concurrent illicit
drug use, level of support (ranging from no ancillary services to a full range of
psychotherapy, counseling, medical services, and vocational and rehabilitation pro-
grams), and duration of program before at least a trial of detoxification from meth-
adone is required. Other requirements of methadone maintenance programs may in-
clude active job seeking and urine specimens to test for illicit drugs.
iii. Detoxification from Methadone. When patients voluntarily or involuntarily
leave methadone maintenance treatment, the clinician can decrease the dosage at a
level ranging from 1 mg a week to 5 mg a day; in this way, withdrawal generally
proceeds gradually. Slow detoxification from methadone maintenance has been associ-
ated with a high dropout rate and low abstinence success rate, probably because the
patient must tolerate withdrawal symptoms for a longer period of time. Many studies
indicate that the use of clonidine in withdrawal from methadone maintenance shortens
304 SHELLY F. GREENFIELD, M.D., M.P.H., ET AL.
the detoxification period and is associated with a higher success rate of abstinence.
Detoxification from methadone maintenance using clonidine can proceed in the same
general manner as described in Section II.F.2.c. Doses used in outpatient detoxification
should generally be lower than inpatient doses, however, because of less frequent
monitoring and the possibility of hypotension.
iv. Current Issues in Methadone Treatment. Methadone maintenance boasts the
highest retention rate of any opioid treatment modality. It is now the primary treatment
modality for approximately 75,000 patients. Of those who begin on methadone mainte-
nance, 50-70% remain in treatment for at least 1 year. The advantages of this treatment
approach have been a demonstrated decrease in illegal drug use and arrest rate, an
improvement in employment and school records, and a reduction in HIV seroconver-
sion. Problems that persist in methadone maintenance programs are (1) the continued
use of illegal drugs, including opioids, cocaine, diazepam (Valium® and others), and
amphetamines, by patients while on methadone maintenance; (2) alcoholism, which
remains a problem for approximately 20% of those on methadone maintenance; (3)
diversion of methadone to illicit markets, especially by those using take-home meth-
adone; and (4) the occasional deaths by overdose, especially in children, teenagers, and
heroin addicts who obtain the methadone illicitly or accidentally.
A review of the numerous studies that have measured the outcome of methadone
maintenance programs has shown that adjunctive psychosocial services (psycho-
therapy, vocational counseling, etc.) increased rates of successful outcome.
v. LAAM Maintenance. L-a-Acetylmethadol (LAAM) is a long-acting preparation
that has recently been approved for use ill opioid maintenance treatment. LAAM can be
prescribed in doses ranging from 10 to 140 mg (with an average dose of 60 mg) and,
because it is longer-acting than methadone, may be dosed Monday, Wednesday, and
Friday. Typically, the Friday dose is 10-15% higher than the Monday or Wednesday
dose. For some individuals, the possibility of fewer clinic visits allows for expanded
integration into work or other rehabilitative activities. Although some studies have
indicated similar results with LAAM-maintained and methadone-maintained patients,
some studies show that higher-dose methadone maintenance treatment is associated
with greater rates of treatment retention.
function tests be monitored at baseline and then each month for 6 months. If serum
aspartate aminotransferase (AST, SGOT) is twice normal at baseline, do not start
naltrexone therapy. Once therapy has begun, if serum aspartate aminotransferase (AST,
SGOT) is elevated to three times baseline, discontinue treatment.
The time of most significant dropout from naltrexone therapy is in the induction
period; this can be minimized by waiting the recommended time between last opioid
use and beginning treatment. Moreover, naltrexone is best used in the setting of strong
psychosocial supports. The most successful outcomes of naltrexone treatment have
been seen in subpopulations that include employed, motivated patients with good
vocational stability and less coexisting drug use. Better outcomes are also more likely
when naltrexone is administered in highly structured multidisciplinary programs that
include family, couple, and individual therapy. Greatest success has been achieved
when there are strong family and social supports and in certain groups such as health
professionals, occasional heroin users, addicts leaving structured settings and pro-
grams, and people from higher socioeconomic groups who do not want to enter metha-
done maintenance treatment.
Clinical experiments with buprenorphine hydrochloride are under way as de-
scribed in Section II.E2.e. Buprenorphine is a mixed mu-receptor agonist and antago-
nist and results in mild euphoria for 3-4 weeks. This may increase retention in the
induction phase but clearly increases the abuse potential. It is 25-50 times more potent
than morphine and equipotent to naloxone as an antagonist. Unlike pure opioid ago-
nists, buprenorphine does not produce greater opioid effects at higher doses. Clinically,
buprenorphine blocks the effects of opioids and satisfies craving. Abrupt withdrawal
produces a mild abstinence syndrome. The duration of action is 24 hours, and it is most
effective when administered subcutaneously or sublingually. Experimentally, a subcu-
taneous dose of 8 mg per day significantly suppresses heroin self-administration in
addicts. Additional clinical trials are needed to assess its long-term efficacy.
A. Introduction
The sedative-hypnotics and benzodiazepines are all included with alcohol in the
category of central nervous system (CNS) depressants despite having widely varying
chemical structures. The term sedative-hypnotic refers to the therapeutic application of
these drugs. Sedative-hypnotics are a diverse group of medications commonly used to
treat anxiety and insomnia. These drugs include the barbiturates and other sedative-
hypnotics such as methaqualone (no longer available in the United States), glu-
tethimide (Doriden®), meprobamate (Miltown® and others), chloral hydrate (Noctec®
and others), paraldehyde, methyprylon (Noludar®), and ethchlorvynol (Placidyl®).
308 SHELLY F. GREENFIELD, M.D., M.P.H., ET AL
C. Classification
1. Benzodiazepines
To a large extent, the benzodiazepines have replaced the barbiturates and other
sedative-hypnotics for the pharmacological treatment of anxiety. In addition, the hyp-
notic effects of these compounds have contributed to their increasing popularity. Taken
for brief periods in therapeutic doses, the benzodiazepines are safe and effective;
however, chronic use may result in the development of tolerance and physical depen-
dence. Benzodiazepines act at one of the subunits of the 'Y-aminobutyric acid (GAB A)
receptors, which has been designated the benzodiazepine receptor. GABA receptors
represent 40% of all synapses in the eNS and are the main inhibitory receptors.
Benzodiazepines act as GAB A agonists, producing the clinical effects of sedation and
anxiety reduction.
The benzodiazepines include the anxiolytic benzodiazepines, which are alpra-
zolam (Xanax®), chlordiazepoxide hydrochloride (Librium® and others), clorazepate
7. PSYCHOACTIVE SUBSTANCE USE DISORDERS 309
2. Barbiturates
The barbiturates are usually classified by their duration of action into ultrashort-,
short-to-intermediate-, and long-acting types. The ultrashort-acting barbiturates, in-
cluding methohexital (Brevital®) and thiopental (Pentothal® and others), are used
primarily as intravenous anesthetics. The short-to-intermediate-acting drugs, including
amobarbital (Amytal® and others), pentobarbital, and secobarbital (Seconal®), are
employed primarily for their sedative and hypnotic properties. The relatively long-
acting barbiturates, including phenobarbital (Tedral® and others), may be employed as
sedative-hypnotics but are most useful in the control of some seizure disorders. They
may also be used to facilitate withdrawal from other sedative-hypnotics.
3. Other Sedative-Hypnotics
overdose, episodic release from body stores may occur, and so the patient's degree of
intoxication may fluctuate widely. Serum concentrations may not correlate well with
the level of consciousness. Dilated pupils may accompany glutethimide poisoning; this
is not the case with other CNS depressants. Overdose with this drug has been impli-
cated in many suicides.
Methyprylon in doses of 200-400 mg will induce sleep. As is the case with other
sedatives, prolonged use of large doses will result in physical dependence. Death has
been reported following the ingestion of as little as 6 g, but individuals vary greatly in
their degree of tolerance.
Ethchlorvynol is a sedative-hypnotic with a rapid onset and short duration of
action. Ingestion of 2-4 g per day over several months can produce physical depen-
dence. As with methyprylon, the potentially lethal dose varies widely. Ingestion of
more than 10 g at once is usually fatal.
Meprobamate, a carbamate derivative, has been widely used as an antianxiety
agent and skeletal muscle relaxant. The benzodiazepines have largely supplanted this
drug both as a therapeutic compound and as a drug of abuse. The usual therapeutic dose
is 400 mg three or four times per day. Chronic administration of slightly higher doses
will produce physical dependence with the possibility of seizures on withdrawal. The
lethal dose varies between 12 and 40 g. Reports of potential teratogenicity make the
ingestion of meprobamate during the first trimester of pregnancy contraindicated.
a long time period and are at particular risk for developing cross-tolerance to, and
dependence on, benzodiazepines.
A third group of abusers of benzodiazepines and sedative-hypnotics are young
people in their teens or early twenties who use illegally obtained medications for
nonmedical purposes to induce euphoria. Frequently, such use occurs in combination
with illicit drugs such as opioids (to enhance euphoria) or cocaine (to counteract the
stimulant effects). These individuals are more likely to administer benzodiazepines or
barbiturates intravenously, increasing the risk of fatal overdose. People in this group
may use these drugs daily or more frequently during sprees of intoxication lasting
several days, during which they may consume massive doses (in excess of 300 mg/day
of diazepam or its equivalent).
It is important to note that many people who are prescribed long-term ben-
zodiazepine treatment for chronic severe anxiety disorders do not meet criteria for
dependence, since the drugs do not interfere with their functioning and they do not
spend excessive time seeking and taking the medications. Significantly, they are con-
tinuing to use the drug for medically prescribed use. However, these individuals may
develop tolerance and physical dependence, since abrupt termination of the drug would
result in characteristic withdrawal symptoms.
2. Clinical Manifestations
Mild to moderate intoxication with CNS depressants closely resembles alcoholic
drunkenness, but patients do not have alcoholic breath. The severity of symptoms
depends on the drug(S) used, the route of administration, and the presence or absence of
other complicating conditions (e.g., head injury). As summarized in Table 2, patients
Drowsiness Ataxia
Slurred speech Hypotonia
Motor incoordination Hyporeflexia
Confusion/agitation Memory impairment
Disorientation Respiratory depression
Nystagmus Inappropriate affect
Tremor Rage reactions
312 SHELLY F. GREENFIELD, M.D., M.P.H., ET AL
usually present with drowsiness, slurred speech, motor incoordination, impaired mem-
ory, confused thinking, and disorientation. Physical examination may reveal both hori-
zontal and vertical nystagmus, tremor, and ataxia. In addition, patients may exhibit
extreme irritability, agitation, inappropriate affect, and paranoia, sometimes accom-
panied by rage reactions and destructive behavior. The latter is most common when
depressant drugs are combined with alcohol.
Severe overdose with CNS depressants is accompanied by signs of cerebrocortical
and medullary depression. Patients may be stuporous or comatose with absent corneal,
gag, and deep tendon reflexes. Plantar stimulation may produce no response or an
extensor response. Patients may also have impaired cardiopulmonary function with
shallow, irregular breathing, hypoxia, respiratory acidosis, and, in the late stages,
paralytic dilatation of the pupils. In terminal cases, patients may develop shock, hypo-
thermia, lung complications (e.g., pulmonary edema, pneumonia), and renal failure.
3. Management
differences in the absorption and metabolism of specific drugs, these conversions can
only be seen as approximations; the actual conversion ratio must be determined empiri-
cally for each patient. Before each dose, the patient is checked for signs of phenobarbi-
tal toxicity such as nystagmus, slurred speech, or ataxia. If nystagmus is present, the
next dose is withheld; if all three signs are present, the next two doses of phenobarbital
are withheld and the daily dose of the next day's phenobarbital is halved. If the patient
is already in acute withdrawal and the risk of seizure is imminent, the first dose of
phenobarbital can be administered intramuscularly. The patient is stabilized on the
starting dose for 1-2 days, and if there are no signs or symptoms of withdrawal or
intoxication, phenobarbital withdrawal begins. Phenobarbital is generally decreased
from the starting dose by 30 mg each day. If intoxication occurs, the next day's dose is
reduced by 50%. If withdrawal signs occur, the next day's dose can be increased by
50%.
The treatment of benzodiazepine or sedative-hypnotic withdrawal is complicated
by the fact that sedative abusers are often inaccurate informants. They may wish to
obtain as much drug from the clinician as possible and/or may have impaired memory
secondary to chronic drug use. For this reason, the pentobarbital tolerance test is a
useful tool in the management of withdrawal states.
ii. The Pentobarbital Tolerance Test. The pentobarbital tolerance test takes advan-
tage of the fact that cross-tolerance exists between the various sedative-hypnotics such
as pentobarbital and phenobarbital. This test is administered in medical settings and
uses the shorter-acting pentobarbital to calculate the appropriate dose of phenobarbital
for stabilization. As illustrated in Table 3, patients showing signs of depressant with-
drawal are given 200 mg of pentobarbital orally (100 mg can be given to certain
patients if the clinician believes that 200 mg may be excessive). One hour later, the
clinician should examine the patient for signs of sedative intoxication such as sedation,
nystagmus, ataxia, and slurred speech. If mildly intoxicated by this dose, the patient has
probably been taking less than the equivalent of 800 mg of pentobarbital a day. The
clinician can stabilize these patients on a dose of 100-200 mg of pentobarbital every 6
hr, depending on the degree of intoxication. If, after an initial test dose (200 mg), no
signs of intoxication appear, the patient's tolerance is probably greater than that in-
duced by prolonged, daily use of 800 mg of pentobarbital. Consequently, additional
increments of 100 mg of pentobarbital may be administered every 2 hr until signs of
intoxication become evident or until a total of 500 mg has been given. The clinician
should then calculate the total dose required to produce intoxication and give this dose
every 6 hr for the next 48 hr. If the patient becomes grossly intoxicated, the next 6-hr
dose may be omitted, and the following day's dose reduced by this amount. Once
stabilization has been achieved, the total daily dose of pentobarbital is reduced by 100
mg each day until withdrawal is completed. Patients should be free of tremulousness,
insomnia, and orthostatic hypotension. If these signs of abstinence recur during the
tapering process, additional 100- to 200-mg doses of pentobarbital may be given.
However, it is usually sufficient to stop the dosage reduction for 1 day and then
cautiously resume tapering the drug.
One disadvantage of pentobarbital-mediated withdrawal is that patients some-
times require intoxicating doses to prevent the development of seizures and perhaps
316 SHELLY F. GREENFIELD, M.D., M.P.H., ET AL
TABLE 3. Detoxification from eNS Depressants Using the Pentobarbital Tolerance Test
delirium. Thus, practical reasons exist for substituting a long-acting barbiturate such as
phenobarbital once stabilization on pentobarbital has been achieved. Use of the long-
er-acting phenobarbital produces a more constant plasma level than can be obtained
with the shorter-acting pentobarbital. Phenobarbital also provides a greater degree of
anticonvulsant activity relative to its sedative effects, so that the patient need not be
intoxicated to avoid the development of seizures. Thirty milligrams of phenobarbital
can be substituted for every 100 mg of pentobarbital. The drug can be given in divided
doses every 8 hr. Once stabilization on phenobarbital has been achieved, the dose is
lowered by 30 mg per day until total withdrawal is achieved. If the patient shows signs
of barbiturate toxicity, one or more doses may be omitted. If signs of withdrawal are
apparent, an additional 60-120 mg of phenobarbital can be given immediately 1M, and
the total daily dose of phenobarbital increased by 25%. Phenobarbital should not be
administered in doses exceeding 500 mg per day.
Treatment with phenobarbital will suppress early symptoms of depressant with-
drawal. Once delirium has developed, however, 24-72 hr of barbiturate treatment may
be required before it clears. Phenytoin (Dilantin® and others) is of questionable effi-
cacy in preventing withdrawal seizures in these patients. The antipsychotic drugs [e.g.,
chlorpromazine (Thorazine® and others)], which themselves lower seizure threshold,
also should be avoided, although some have advocated their use in the treatment of
delirium.
Although some patients have been withdrawn from high-dose benzodiazepines
and sedative-hypnotics as outpatients, this is risky. Thirty milligrams of phenobarbital
are substituted for each hypnotic dose of the substance abused, and the total daily
requirement of phenobarbital is given in three divided doses. The total daily dose is
then reduced by 30 mg per day over 2-3 weeks.
iii. Gradual Dose Reduction. Although the above-described phenobarbital substi-
tution technique has been used successfully in many centers, some questions have
recently emerged regarding the effectiveness of this technique in withdrawing patients
from certain drugs, particularly alprazolam. Thus, many clinicians have tapered their
patients off benzodiazepines such as alprazolam (when the amount taken was clearly
known) by lowering the dose gradually. For example, the Physicians' Desk Reference
recommends that the daily dose be decreased by no more than 0.5 mg every 3 days.
Since even this schedule is very uncomfortable for some long-term, high-dose users,
7. PSYCHOACTIVE SUBSTANCE USE DISORDERS 317
some authors have recommended a gradual switch from alprazolam to clonazepam and
tapering the patient off the latter drug.
In the last two decades, simultaneous abuse of more than one class of psychoac-
tive drug has become commonplace. Heroin addicts and patients on methadone mainte-
nance may also abuse benzodiazepines, sedative-hypnotics, and alcohol. Although
there is no cross-tolerance between opioids and CNS depressants, sedative-hypnotics
and benzodiazepines may partially alleviate the symptoms of opioid withdrawal. How-
ever, neither methadone nor heroin will prevent high-dose benzodiazepine or sedative-
hypnotic withdrawal.
When mixed opioid-sedative dependence is suspected, the clinician should first
determine the degree of sedative-hypnotic or benzodiazepine dependence by using the
pentobarbital tolerance test. Once this is done, the most prudent course is to maintain
the patient on a dose of methadone sufficient to prevent symptoms of opioid with-
drawal while gradually tapering CNS depressants. Some have recommended tapering
both classes of drugs simultaneously, but this approach may complicate the clinical
picture.
A. Introduction
Of all the drugs mentioned in this chapter, alcohol is the most commonly abused.
About 70% of the adult population in this country have consumed alcoholic beverages
in the last year. Within this group, it is estimated that the number of problem drinkers
approaches 10 million, about half of whom are physically dependent on alcohol.
Alcohol use disorders are directly or indirectly implicated in 200,000 deaths annually.
318 SHELLY F. GREENFIELD, M.D., M.P.H., ET AL
In the United States, 13.8% of all adults have had alcohol abuse or dependence in the
course of their lives.
Alcohol abuse and dependence are defined by the same criteria used to define all
psychoactive substance use disorders (see Section I). Among people who meet these
diagnostic criteria, there appear to be several typical patterns of drinking. The first is
characterized by drinking large quantities on a daily basis. The second pattern is so-
called weekend binge drinking, in which people are abstinent during the week but
consume large quantities of alcohol on weekends. Another pattern is characterized by
long periods of abstinence punctuated by alcohol binges that may last for weeks or
months. In addition to these patterns, there are a number of other presentations of
alcohol abuse and dependence. Indeed, the heterogeneity of the disorder is one of its
hallmarks.
B. Pharmacology
Alcohol is rapidly absorbed from the gastrointestinal tract, but the rate of absorp-
tion is modified by food, the volume of liquid ingested, the concentration of alcohol,
and its rate of administration. High concentrations limit intestinal absorption by induc-
ing a reflex pylorospasm. As a result, large quantities of alcohol may remain unab-
sorbed in the stomach before gradually passing into the small intestine. Thus, following
an episode of heavy drinking, an individual may experience fluctuating levels of
intoxication.
Once absorbed, alcohol appears to be uniformly distributed throughout most body
tissues, including the brain, where an equilibrium with the plasma alcohol concentra-
tion is quickly reached. In the liver, alcohol is converted to acetaldehyde through the
action of alcohol dehydrogenase. Acetaldehyde, in tum, is converted into acetyl coen-
zyme A, which is then oxidized through the citric acid cycle or used in other anabolic
reactions. In addition to alcohol dehydrogenase, liver microsomal oxidases also playa
role in alcohol metabolism. The average rate of liver oxidation is 10 mL/hr, with a
daily maximum of approximately 450 mL. A small unoxidized portion of ingested
alcohol is excreted through the kidneys and lungs.
D. Acute Intoxication
1. Simple Type
Although ethyl alcohol affects all organ systems of the body, its most marked and
important effect is on the central nervous system. Physiological signs of intoxication
include slurred speech, ataxia, incoordination, nystagmus, and flushed face. There may
also be maladaptive behavioral changes such as disinhibition, aggressiveness, poor
judgment, impaired attention, poor recent memory, irritability, euphoria, depression,
and emotional lability.
Alcohol-induced inhibition of the cerebral cortex releases lower brain centers
from the inhibitory and integrating control of the cortex. This accounts for the initial
stimulatory effect of alcohol on behavior. In the initial stages of intoxication, the user
may become outgoing, loquacious, and emotionally labile. Increased confidence and
expansiveness often lead to the perception of enhanced verbal and manual perfor-
mance. Careful testing, however, reveals that efficiency in both areas is impaired.
Recent memory, ability to concentrate, and insight are also diminished.
The acute effects of alcohol on the brain are a function of the blood alcohol level,
which itself is determined by dose, speed of absorption, rate of metabolism, and the
efficiency of alcohol excretion. In addition, CNS effects are more pronounced when the
blood alcohol level is rising than when it is falling. In the majority of individuals, a
blood alcohol concentration of 50-100 mg per deciliter (dl) results in mild intoxica-
tion. Concentrations over 100 mg/dL usually impair an individual's ability to operate a
motor vehicle. At 200 mg/dL, most users will be grossly intoxicated, exhibiting ataxia
and slurred speech. Blood alcohol levels over 400 mg/dL are potentially lethal. Alco-
hol usually exerts its lethal effects by leading to respiratory depression or aspiration of
vomitus.
There are a number of potential complications of alcohol intoxication. These
include highway, household, and industrial accidents and violent acts. Fifty percent of
all murders occur when the murderer or victim is intoxicated, and approximately 25%
320 SHELLY F. GREENFIELD, M.D., M.P.H., ET AL
of all suicides occur during a period of intoxication. Falls are common in intoxicated
individuals, leading to potentially serious medical sequelae such as fracture, subdural
hematoma, and other brain trauma.
E. Chronic Intoxication
1. Peripheral and Central Nervous System Effects
Chronic alcohol abuse may lead to substantial nervous system pathology. The
most frequent neurological disorder in alcoholics is polyneuropathy, which results
primarily from an alcohol-induced vitamin deficiency, particularly a deficiency of B
vitamins. A direct toxic effect of alcohol may also be a factor. Most patients are
asymptomatic, with absent ankle jerks as the only sign. In more severe cases, burning
feet, pain, paresthesias, and distal muscle weakness occur. Foot drop and wrist drop are
ultimate complications. Prolonged abstinence, adequate nutrition, and vitamin replace-
ment may gradually reverse this syndrome, but in some the damage is permanent.
CNS effects include Wernicke's encephalopathy, Korsakoff's syndrome, alcohol
dementia and other cognitive impairments, and ataxia. Wernicke's encephalopathy, the
result of an alcohol-induced thiamine (vitamin B l ) deficiency, occurs in some chronic
alcoholics. Nystagmus, bilateral sixth cranial nerve palsies, and paralysis of conjugate
gaze are frequently seen along with ataxia and various mental disturbances. Most
common is a "quiet delirium" characterized by apathy, lassitude, disorientation, and
drowsiness. The symptoms usually have an abrupt onset and may occur singly or in
various combinations.
Approximately 80% of patients with Wernicke's encephalopathy will also develop
Korsakoff's syndrome, or alcohol amnestic disorder, which consists of profound ante-
rograde and retrograde amnesia, impairment of recent memory, sometimes coupled
with confabulation, in an otherwise alert individual. Wernicke's disease and Kor-
sakoff's syndrome are not separate disease entities; rather, the latter is a variably
present component of the former. When both are present, the disease is called the
Wernicke-Korsakoff syndrome. Postmortem brain examination of these patients often
reveals structural lesions in the mammillary bodies, which may be the result of vitamin
deficiencies or the direct toxic effects of alcohol.
7. PSYCHOACTIVE SUBSTANCE USE DISORDERS 321
Over time, heavy alcohol consumption may damage other organ systems. Approx-
imately 15% of heavy alcohol users develop alcoholic cirrhosis and/or pancreatitis.
Liver disease can take the form of alcoholic fatty liver, alcoholic hepatitis, or cirrhosis.
Gastrointestinal disorders include gastritis, pancreatitis, gastric and duodenal ulcers,
and malabsorption syndrome. Many individuals with chronic alcohol use disorders are
anemic, in part because of nutritional (e.g., folic acid) deficiencies that often accom-
pany chronic alcohol abuse. Alcoholic cardiomyopathy also is an important cause of
morbidity and mortality in chronic heavy drinkers. Individuals may develop hyperten-
sion and hypercholesterolemia. Endocrine abnormalities include decreased serum tes-
tosterone, which may lead to testicular atrophy, loss of facial hair, breast enlargement,
decreased libido, and impotence in men. Women may experience late-luteal-phase
dysfunction, amenorrhea, anovulation, and hyperprolactinemia. Finally, individuals
with alcohol use disorders are at increased risk for acquiring HIV infection, usually
through engagement in high-risk behaviors during periods of intoxication.The reader is
referred to standard textbooks of internal medicine for a more complete description of
these clinical entities.
F. Alcohol Withdrawal
1. Minor Abstinence Syndrome
a. Clinical Manifestations. In physically dependent users, the initial stages of
alcohol withdrawal usually begin 4-12 hr after the cessation or reduction of alcohol
use. Symptoms include autonomic hyperactivity including sweating and increased
heart rate (> 100 beats/min), tremulousness ("shakes"), muscle tension, flushing, and a
vague sense of anxiety ("jitters"). This may be followed by nausea, vomiting, anorexia,
and abdominal cramps. The patient may appear restless and agitated, and signs of
322 SHELLY F. GREENFIELD, M.D., M.P.H., ET AL
sy examinations in patients with AWD have failed to elucidate the exact pathophysiolo-
gy of the problem.
b. Management. Patients with AWD are seriously ill and require intensive inpa-
tient care. The major objectives of treatment are reduction of eNS irritability, preven-
tion of exhaustion, and correction of potentially fatal fluid and electrolyte imbalance.
An intensive-care unit is usually preferable for careful monitoring of electrolytes.
Intravenous dextrose and saline should be given to replace fluid losses. Hyperthermia
should be treated aggressively with acetaminophen for temperatures above 101°F and a
cooling blanket for temperatures above 103SF. Parenteral thiamine, 100 mg per day,
should be given. Restraints should be used for severe agitation. For sedation, diazepam,
5-10 mg IV, may be administered every 15-20 min until sedation is achieved. Repeat
doses may be given as needed. The exact dosing should vary with the weight of the
patient, severity of symptoms, and quantity of alcohol consumed prior to withdrawal.
For patients with hepatic insufficiency, oxazepam or lorazepam can be used. Di-
azepam, up to 20 mg q.i.d., oxazepam, 60 mg q.i.d., or lorazepam, 1-2 mg q.i.d., may
be required for sedation. Infections should be treated with appropriate antibiotics.
Following a prolonged drinking bout, some patients who are physically dependent
on alcohol may experience an organic psychosis. This may be characterized by vivid
and persistent auditory or visual hallucinations or delusions that appear during intoxi-
cation or within several days after the last drink. Auditory hallucinations may take the
form of threatening or disturbing voices. Patients may respond by calling the police or
arming themselves. Unlike patients with delirium, however, these individuals are ori-
ented to time, place, and person. In those individuals in whom alcohol-induced psycho-
tic disorder occurs during withdrawal, hallucinations or delusions may last from a few
hours to a week, although about 10% continue for weeks or months. Some patients will
experience a recurrence of hallucinosis following each bout of heavy alcohol use. A
few will develop a chronic form of this disorder, which may be mistaken for schizo-
phrenia. Antipsychotic drugs such as chlorpromazine or trifluoperazine (Stelazine® and
others) are the treatment of choice. On recovery, there is excellent memory for the
psychotic episode.
2. Biological Treatment
ally the patient may develop fulminating (sometimes fatal) hepatitis. A patient with
persistently elevated LFfs should not be given disulfiram until the LFfs return to
normal. Liver function tests should then be monitored every 2 weeks for the first 2
months and then every 3-6 months thereafter. If LFfs become elevated, disulfiram
should be discontinued. Disulfiram is not recommended in patients with moderate to
severe hepatic dysfunction, peripheral neuropathies, pregnancy, renal failure, or cardi-
ac disease.
Disulfiram is probably most useful in older, well-motivated, socially stable pa-
tients whose drinking is precipitated by recurrent psychological stress. The drug should
be taken each morning, when the level of stress and the temptation to drink are usually
lowest. When taking disulfiram, many patients report that situations previously condu-
cive to drinking generate less craving for alcohol than before. On the other hand,
patients with poor impulse control may be at considerable risk of drinking when on
disulfiram.
Disulfiram, by itself, is not an extremely effective treatment for alcoholism. A
recent study indicated that only 23% of patients treated with disulfiram alone remained
alcohol-free for 1 year, compared with 12% of a placebo-treated control group. Taking
disulfiram is a daily decision that is best supported when it is part of a comprehensive
treatment program aimed at helping the patient adjust psychologically and socially to
an alcohol-free existence.
depressed social drinkers and in those with alcohol use disorders but no comorbid
depression. Although alcoholics are frequently anxious, the use of benzodiazepines or
other sedative-hypnotics (e.g., diazepam, clonazepam) other than in the treatment of
withdrawal is problematic, because alcohol-dependent individuals have an increased
risk of developing dependence on other CNS depressants. After four weeks of absti-
nence, if anxiety symptoms have persisted or increased, the individual should have an
evaluation for diagnosis of anxiety disorder. In those patients with co-occurring panic
disorder or agoraphobia, appropriate antidepressant treatments can be prescribed. In
those individuals with more generalized anxiety, a trial of buspirone may be helpful.
3. Aftercare
5. The Psychotherapies
6. Brief Interventions
For less severely affected individuals, brief motivational treatments may be effec-
tive. Typically, these are delivered over 1-3 sessions and include an evaluation of
alcohol-related problems and drinking severity and the provision of education and
recommendations. Such interventions generally have greater efficacy than no interven-
tion and often enhance the effectiveness of treament received at a later date.
Perhaps the most widely useful treatment for alcoholism is Alcoholics Anony-
mous (AA), a worldwide organization that aims to help its members achieve lasting
sobriety. This is essentially a self-help approach in which members share their recovery
experiences and support each other in the struggle to avoid relapse. Unlike the help
offered by most professionals and social agencies, AA groups are almost continuously
available. Experienced group members (sponsors) are asked to care for new members
during periods in which the relapse risk is high. Membership provides an opportunity
to give to others, which enhances self-esteem. It also offers consistent reminders of
7. PSYCHOACTIVE SUBSTANCE USE DISORDERS 329
both the rewards of sobriety and the pitfalls of returning to drinking and its disastrous
consequences. Of those alcoholics who become long-term active members, about 40-
50% remain abstinent for several years, with 60-68% improving to some extent during
their participation.
Alanon, a separate organization for significant others, spouses, or other family
members and friends, offers support and valuable information about dealing with the
alcohol-dependent individual at home. Alateen provides similar help for the children of
alcoholic parents. Adult Children of Alcoholics (ACOA) provides education and sup-
port for those adults who grew up in a household in which a parent was alcohol
dependent. The individual need not be alcohol dependent him/herself to benefit from
this group.
Rational Recovery (RR) is another self-help group that is based on principles of
cognitive therapy. This self-help program does not use the 12-step approach of AA but
instead focuses on patterns of thought and feeling that lead to drinking and helps
individuals to change these patterns. Certain patients will prefer RR to AA, whereas
others will use both self-help approaches effectively.
8. Adjunctive Services
Halfway houses can be useful for alcoholic patients who, for a variety of reasons,
should not live at home. For most people, these houses are transitional facilities
between home and hospital, whereas for others they provide long-term custodial care
and supervision. Since job failure frequently accompanies chronic alcoholism, voca-
tional rehabilitation programs are also a useful adjunct to treatment.
9. Summary
Concern about the rising cost of health care has naturally prompted questions
about the efficacy of various treatments for alcoholism. Some studies have suggested
that less intensive treatment, e.g., outpatient group therapy or brief hospitalization for
detoxification, does not necessarily result in a poorer outcome compared with more
intensive treatment regimens. However, as in other addictive disorders, an individually
tailored approach is clearly the most useful. Reliance on pharmacological intervention
alone or on a single psychologically based treatment modality is usually insufficient to
disrupt a pattern of repetitive relapse. Finally, although the individual and societal costs
of alcoholism are enormous, comparatively little attention has been paid to preventive
approaches, including intensive education, screening and early detection, and brief
interventions. More research is clearly needed in this area.
A. Cocaine
1. Epidemiology
Cocaine use continues to represent a major public health problem in the United
States. Recently, the Substance Abuse and Mental Health Services Administration
330 SHELLY F. GREENFIELD, M.D., M.P.H., ET AL
estimated that 3 million Americans reported having used cocaine within the past
year-a significant decrease from 8.1 million in 1985-but it also reported that half a
million Americans abused cocaine at least weekly, which represents no significant
change since this was first measured in 1985. Thus, although casual use of cocaine has
dropped in the past decade, the number of heavy users has not.
3. Acute Intoxication
Cocaine exerts a direct stimulatory effect on the cerebral cortex, producing excita-
tion, restlessness, euphoria, decreased social inhibitions, increased energy, and feelings
7. PSYCHOACTIVE SUBSTANCE USE DISORDERS 331
of enhanced strength and mental ability. Its sympathomimetic effects include vaso-
constriction, tachycardia, increased blood pressure and temperature, and dilated pupils.
High doses may lead to repetitive stereotyped behavior, bruxism, formication, irri-
tability, restlessness, tremor, panic, emotional lability, and paranoia.
Acute cocaine poisoning is rare among users of low doses but common in high-
doses users and especially among those who inject or smoke cocaine. The symptoms
are similar to those of intoxication but lead more frequently to intense anxiety, para-
noia, and illusions or hallucinations. Elevated blood pressure, tachycardia, ventricular
irritability, hyperthermia, and respiratory depression are symptomatic of severe poison-
ing. Acute heart failure, stroke, and seizures have also been reported.
Initial treatment of acute cocaine poisoning should be symptomatic and support-
ive. Manifestations of sympathetic nervous system hyperactivity such as hypertension,
tachycardia, and tachypnea have been treated in the past with J3-adrenergic blockers
and calcium-channel blockers. However, recent animal data and some clinical experi-
ence suggest that all J3-blockers and dopaminergic antagonists should be avoided in
treating cocaine-induced hypertension. Patients in hypertensive crisis may be given
phentolamine (Regitine®). Intravenous diazepam can be used in patients with repeated
seizures, and ambient cooling is recommended for hyperthermia.
4. Abstinence Symptoms
6. Medical Complications
Medical complications of cocaine use affect many organ systems and create
significant morbidity and mortality. Cardiovascular effects include hypertension, myo-
cardial infarction, arrhythmias, aortic rupture, and angina pectoris. CNS effects include
grand mal seizures, intracranial hemorrhage, ruptured intracranial aneurysms, and other
cerebrovascular accidents and anxiety, paranoia, and psychosis. Because of vaso-
constriction and acute increases in blood pressure, gastrointestinal complications such
as ischemic or gangrenous bowel are occasionally seen. Decreased lung diffusion
capacity, bronchitis, pneumothorax, pneumomediastinum, and pulmonary edema are
especially associated with smoking cocaine; respiratory arrest may be seen as a compli-
cation of cocaine use by any route of administration. Other medical complications
include rhabdomyolysis with subsequent acute renal failure, hepatic damage, and
sometimes death; anosmia and perforation of the nasal septum; HIV infection associ-
ated with intravenous administration or disinhibited sexual activity; viral hepatitis; and
sexual dysfunction.
B. Amphetamines
1. Introduction
Although the chemical structure of amphetamines is dissimilar to that of cocaine,
their actions are similar, though of longer duration and perhaps lesser intensity. Indeed,
research subjects who blindly compared the effects of the two drugs in one study were
unable to tell them apart.
Amphetamine and its related compounds are sympathomimetic agents: their pe-
ripheral effects resemble those produced by stimulation of adrenergic (i.e., sympathet-
ic) nerve endings. They also have central stimulatory effects, which contribute to their
popularity as drugs of abuse. Dextroamphetamine, the d-isomer of amphetamine, is
three to four times as potent as the l-isomer. Other amphetamine derivatives include
methamphetamine (Desoxyn®) (which is also available in injectable and smokable
forms), methylphenidate, phenmetrazine (Preludin®), and diethylpropion (Tenuate®
and others). This section focuses on amphetamines as examples of this class of agents.
2. Pharmacology
Both amphetamine and methamphetamine are well absorbed from the gastroin-
testinal tract and are stored in body tissues, including the CNS. Clinical effects gener-
ally appear within 30 min after oral ingestion. The drugs are metabolized primarily in
the liver via hydroxylation, demethylation, and oxidative deamination. Following oral
administration,· more than 50% is excreted unchanged in the urine, particularly when
urinary acidity is high.
Structurally, the amphetamine molecule strongly resembles the catecholamine
neurotransmitter dopamine. However, amphetamine is more lipid-soluble and more
334 SHELLY F. GREENFIELD, M.D., M.P.H., ET AL
readily crosses the blood-brain barrier. The amphetamines exert effects on dopamine-
containing neurons by facilitating dopamine release and perhaps interfering with the
metabolic breakdown of this neurotransmitter. The paranoid psychosis that sometimes
accompanies chronic use of high doses of amphetamines may result from effects on
dopaminergic neurons.
3. Acute Effects
In therapeutic doses, amphetamines produce wakefulness, mood elevation, a sense
of initiative and confidence, and increased mental alertness and ability to concentrate
on simple tasks. Appetite is diminished, probably through a direct effect on the feeding
center in the lateral hypothalamus. Systolic and diastolic blood pressure are increased,
and heart rate is reflexly slowed. At higher doses, tremulousness, agitation, insomnia,
headache, dizziness, confusion, and dysphoria occur. The electroencephalogram (EEG)
is desynchronized, and both total sleep time and REM sleep are markedly reduced.
5. Patterns of Abuse
Currently, amphetamine and other CNS stimulants such as methylphenidate may
be prescribed for the treatment of narcolepsy, attention-deficit hyperactivity disorder,
and exogenous obesity. Until the late 1960s, these drugs were frequently prescribed for
the treatment of fatigue and depression as well. However, as their abuse potential
became apparent, their prescription was placed under increased regulatory control,
which, in turn, created a substantial illicit market. This has led to widespread drug
substitution, so that much of what is sold on the street as amphetamine is caffeine or
another over-the-counter stimulant.
A recent study by the National Institute on Drug Abuse demonstrated that approx-
imately 1% of the adult population had abused amphetamines or other similarly acting
sympathomimetics within the previous year. There are three common patterns of abuse.
First, some amphetamine abusers use the drug intermittently, ingesting large doses (up
to several grams) in search of euphoria. Members of this group are often young, tend to
abuse other drugs, and are more likely to use methamphetamine either intravenously
7. PSYCHOACTIVE SUBSTANCE USE DISORDERS 335
("crystal") or by smoking ("ice"). The individuals may also concurrently use CNS
depressants or opioids. Intravenous amphetamine produces euphoria more rapidly than
oral administration. With this pattern of use, the drug may be injected many times each
day in doses as high as 1 g every 3-4 hr. Sprees of intoxication may last days or weeks
and are usually followed by abrupt withdrawal ("crashing") and physical exhaustion.
The binge is followed by a period of exhaustion and sleep and subsequent nonuse for
several days or weeks.
Smoking of methamphetamine is a second pattern of abuse. "Ice," as it is called,
was introduced several years ago in Hawaii and northern California, accompanied by a
fear that its use would spread like that of crack. Fortunately, ice use has thus far
remained localized. Abuse patterns of ice are similar to those of crack, but with a
longer duration of action.
A third pattern of amphetamine abuse is chronic daily oral use at either high or
low doses. Continuous long-term use produces diminished euphoriant effects and an
increase in dysphoria because of tolerance. Other symptoms are depression, irritability,
anergy, anhedonia, social isolation, sexual dysfunction, paranoid ideation, attentional
disturbances, and memory problems.
6. Acute Intoxication
7. Amphetamine Withdrawal
A. Introduction
The term hallucinogen means "producer of hallucinations." The hallucinogens are
a group of structurally similar agents that produce perceptual distortion (primarily
visual illusions and hallucinations) and enhance awareness of internal and external
stimuli. They also induce in the user a sense that mundane events are unusually
important, a tendency toward introspection, and profound emotional lability. Some
have referred to this group of substances as "psychedelic" (i.e., "mind expanding"), but
this suggested property may be more illusory than real. The term "psychotomimetic"
also has been applied to these drugs because they produce a state that mimics function-
al psychosis, but the resemblance between hallucinogen intoxication and functional
psychosis is superficial.
Popular in the late 1960s, hallucinogen abuse has declined in recent years. The
drugs have by no means disappeared, however, and the adverse consequences of their
use are still apparent in emergency room settings. For instance, in 1993, hallucinogens
were responsible for 3369 emergency room visits. 9
Throughout the 1980s, the annual prevalence of use of hallucinogens declined. 225
Yet, during the early 1990s, there was some indication of a rise in LSD use. For
example, 5600 doses of LSD were confiscated by the Drug Enforcement Administra-
tion in 1991 compared to only 14 doses in 1990. 235 For the first time since 1976, the
annual survey of high school seniors found that in 1990 and 1991 more seniors had
used LSD than cocaine in the previous 12 months. 228
According to the 1993 National Household Survey on Drug Abuse,9 8.7% of the
total population in the United States reported using hallucinogens at some time in their
lives, 1.2% (or 2,391,000 people) had reported using hallucinogens in the past year, and
0.2% (or 515,000 people) in the past month.
On the street, typical doses of LSD range in size from 70 to 300 micrograms. In
1994, the cost of one street dose was about $5, compared to $50 for 1/2 gram of cocaine
and $70 for 1/4 ounce of marijuana. 230 Because LSD has such a high potency, it can be
applied to paper blotters or to the backs of postage stamps. Additionally, drugs sold on
the street as "hallucinogens" often contain other agents [e.g., amphetamine, phen-
cyclidine (PCP)], so that patients who present with a "bad trip" presumably from
mescaline may, in fact, be suffering from a PCP-induced psychosis.
B. Classification
Pharmacologically, the commonly abused hallucinogenic substances may be
divided into two major groups. The indolealkylamines, including d-lysergic acid
diethylamide (LSD), psilocybin, and dimethyltryptamine (DMT), bear a structural
resemblance to the neurotransmitter 5-hydroxytryptamine (serotonin). The phenyl-
ethylamines, including mescaline, and the phenylisopropylamines, such as 2,5-dimet-
338 SHELLY F. GREENFIELD, M.D., M.P.H., ET AL
With repeated administration, tolerance develops rapidly (i.e., in 2-4 days) to the
behavioral effects of LSD. As a result, even chronic users do not typically take the drug
more often than twice a week, and the vast majority take it less frequently. Consider-
able cross-tolerance exists between LSD and other hallucinogens. No physical depen-
dence has been demonstrated in LSD users, and abstinence symptoms are not observed
after withdrawal. Only one death has been directly linked to an LSD overdose, al-
though fatal accidents and suicides may occur during periods of intoxication.
3. Acute Intoxication
The effects of LSD begin 20-60 min after ingestion, depending on the amount
ingested and the degree of tolerance developed. Sympathomimetic effects include
tachycardia, increased blood pressure and body temperature, and pupillary dilation.
Hyperreflexia, nausea, and muscle weakness also are observed. Peak effects occur
between 2 and 3 hr after ingestion. Visual illusions, wavelike perceptual changes,
macropsia and micropsia, and extreme emotional lability dominate this period. Percep-
tions in one sensory modality may affect or overflow into another (synesthesias), so
that colors are "heard" and sounds "seen." Subjective time is slowed, and a generalized
loss of body and ego boundaries is experienced. Other perceptual changes include
depersonalization, derealization, increased fantasy production, illusions, hallucinations,
and hyperacusis, which occur in a state of full wakefulness and alertness. Hallucina-
tions from these agents are primarily visual and may be distinguished from those of
TABLE 4. Commonly Abused Hallucinogenic Drugs
Psychedelic Peak
Drug Source dose symptoms Duration of action Prominent somatic effects Prominent psychological effects
d-Lysergic acid Synthetic 50 ....g 2-3 hr 8-12 hr; Increased sympathetic Hypervigilance, illusions,
diethylamide (from undulating nervous system emotional lability, loss of
fungi) activity as activity; dilated pupils; body boundaries, time
effect declines increased blood slowing, increased intensity
pressure, pulse, tendor of all sensations
reflexes, temperature,
and blood sugar;
tremor
Psilocybin Mushroom 10 mg 90 min 4-6 hr Like LSD, but milder Like LSD, but less intense,
more visual; more euphoria;
paranoia
Dimethyltryptamine Synthetic 50mg 5-20 min 30-60 min Like LSD, but with more Like LSD, but usually more
(DMT) intense intense, in part because of
sympathomimetic sudden onset. Must be
symptoms smoked or injected; cannot
be taken orally
Mescaline Peyote 200mg 2-3 hr 8-12 hr Nausea, vomiting; Like LSD, but perhaps more
cactus otherwise like LSD; sensory and perceptual
perhaps more intense changes; euphoria prominent
sympathomimetic
effects
Dimethoxymethyl- Synthetic 5 mg 3-5 hr 6-8 hr at doses Minimal effects at low May resemble amphetamine
amphetamine below 5 mg; dose; autonomic combined with LSD, but
(DOM, "STP") 16-24 hr at effects prominent at long-lasting; high incidence
high doses doses above 5 mg of flashbacks, psychosis;
(10-30 mg) chlorpromazine may
aggravate symptoms
340 SHELLY F. GREENFIELD, M.D., M.P.H., ET AL
schizophrenic psychosis, which are usually auditory.235 Behavioral changes may in-
clude impaired judgment, marked anxiety and depression, fear of losing one's mind,
paranoia, and referential thinking. In such a state, the lack of a supportive environment
or companion can be detrimental, as the individual struggles to control his or her
anxiety and prevent ego disintegration. As the effects of the drug begin to wane (i.e., 4-
6 hr after ingestion), the patient experiences intermittent "waves of normalcy." After
8-12 hr, the intoxication syndrome is mostly cleared, although aftereffects, including a
sense of psychic numbness, may last for days.
2. Toxic Delirium
Another untoward effect of hallucinogen use is psychosis that fails to clear after
the drug is metabolized. Schizophreniform reactions have lasted anywhere from weeks
to years following a hallucinogenic experience. 235 This occurs primarily in patients
with underlying mood disorder or schizophrenia, some of whom may use these drugs in
the hope of achieving psychic reintegration or an anxiety-free state. Such patients often
become more disorganized by the drug experience. Treatment may include judicious
use of antipsychotic drugs and, when indicated, mood stabilizers and/or antidepres-
sants. Following resolution of the acute psychotic episode, attention should be directed
toward preventing further drug use.
reassurance that the flashback will pass. In some cases, benzodiazepines may be
helpful.
Phencyclidine (PCP, "angel dust," "hog," "T," "crystal joints," "rocket fuel"), an
arylcyclohexylarnine, was originally developed as a human anesthetic, but its use in
humans was discontinued in 1965 because it induced psychotic and hallucinatory
reactions. It is now legally used only in veterinary medicine as a tranquilizer. Through
the 1980s and early 1990s, PCP abuse has increased in prevalence, especially among
those between the ages of 18 and 34. The number of people reporting known PCP use
increased from 5.6 million in 1985 to 8.4 million in 1993 (4.1 % of the U.S. population),
but more people have taken the drug unknowingly in the form of adulterated marijuana,
heroin, or LSD. Recently, the number of PCP-related emergency room visits decreased
significantly from 12,346 in 1988 to 6527 in 1993. 9 Yet, phencyclidine-related deaths
have increased from 96 in 1981 223 to 206 in 1993, with more than half of those deaths
being males between the ages of 18 and 34. 9 The drug was found in combination with
at least one other drug, often heroin or alcohol, in the majority of PCP-related
deaths.237 Street samples of the drug have been shown to contain nearly 30 analogues
of PCP, many of which are more potent than PCP itself. Because of its behavioral
toxicity, the drug poses a significant health problem, with deaths resulting from drug-
induced violence, suicide, and accidents. The majority of PCP-related deaths were
consequences of external events rather than actual drug overdoses. For instance, of the
104 PCP-related deaths in 1981, 54 were homicides, 29 were accidents, and 14 were
accidental overdoses. 236 Because of the decline in the prevalence of use of phen-
cyclidine and its more recent use concurrently with other drugs, PCP will not be
reviewed here. Readers are referred to other general texts for a review of the signs and
symptoms of PCP intoxication and overdose and its treatment.
7. PSYCHOACTIVE SUBSTANCE USE DISORDERS 343
A. Introduction
The inhalants comprise a heterogeneous group of volatile organic solvents that
have profound toxic effects on the CNS. Some of the commonly inhaled substances and
their major active ingredients include glues and paint thinners (toluene), lighter fluid
(naphtha), aerosols (fluorinated hydrocarbons, nitrous oxide), cleaning solutions (tri-
chloroethylene, carbon tetrachloride), and gasoline (benzene).
B. Patterns of Abuse
Adults at risk for inhalant abuse include those whose work brings them into
contact with these substances, such as shoemakers, cabinetmakers, printers, hair styl-
ists, maintenance workers, painters, and workers in gas stations and dry cleaners.
Children and adolescents are also at risk. Young users tend to be between 7 and 19
years old, but use has been reported in children as young as 4 to 6 years.
The prevalence of inhalant abuse is greater than is commonly appreciated by
health care providers. In 1990, one community survey showed that 2.9% of high school
seniors and 1% of college students had used inhalants within the previous 30 days.241
Inhalants tend to be popular drugs of abuse among boys in their early teens, especially
in poor and rural areas, where more expensive drugs of abuse are less available. 241
Although the prevalence of inhalant abuse decreases with age, many users go on to
abuse other drugs. Most commonly, the volatile gases are inhaled from a handkerchief
or rag that has been soaked with the solvent material. Glue squirted into a paper bag
gives off a vapor that can then be inhaled. Aerosol spray cans are a source of volatile
hydrocarbons and nitrous oxide. 241
c. Acute Intoxication
High lipid solubility and rapid passage across the blood-brain barrier account for
the rapidly intoxicating effects of the inhalants. Drug effects last up to 45 min, depend-
ing on the degree of exposure. CNS impairment usually resolves in minutes to hours.
The clinical picture resembles that of alcohol intoxication and includes euphoria
and giddiness, turning soon to light-headedness, agitation, disinhibition, and drowsi-
ness. With increasing intoxication, ataxia, dizziness, and disorientation can develop.
Irritation of the nasal mucosa, conjunctivitis, and unusual breath odor also suggest the
diagnosis. Several hours after intoxication, individuals may feel lethargic, "hung over,"
and experience headaches. Depending on the solvent used, a variety of toxic effects
also may occur. These include induction of cardiac arrhythmias, hypoxia, seizures, and
death. Inhalants, especially chlorohydrocarbons, paint, and glue, have also produced
glomerulonephritis and distal renal acidosis.
Tolerance to the effects of inhalants develops rather quickly, but no cross-toler-
ance occurs among the various solvents. Although physical dependence has not been
documented, a syndrome resembling delirium tremens has been noted in some chronic
344 SHELLY F. GREENFIELD, M.D., M.P.H., ET AL
users who have abruptly stopped using these agents. Other withdrawal symptoms may
include tremulousness, tachycardia, disorientation, hallucinations, seizures, and agita-
tion if inhalant use is abruptly discontinued after chronic use. Abdominal pain, par-
esthesias, and headaches also can occur during drug-free periods. Strong psychological
dependence often develops in inhalant abusers and accounts for much of the difficulty
in treating these individuals.
Clinical assessment should include careful individual evaluation of adult and child
inhalant abusers and family assessment if the patient is a child or adolescent. Patients
should be assessed for damage to the CNS, liver, kidney, lung, nasal mucosa, heart, and
bone marrow. There are no routine biological assays for most inhalants. However,
tetraethyllead gasoline may be detected by basophilic stippling of the red blood cells
and increased lead levels in a 24-hr urine. Neuropsychological testing may help identi-
fy longer-term effects on attention span, memory, and concentration.
E. Treatment Approaches
Chronic solvent abusers may not respond to the usual drug rehabilitation treatment
modalities because of the combined effects of low socioeconomic status, family prob-
lems, school failure, antisocial behavior, and brain damage. Therapeutic communities
emphasizing work programs, remedial education, and other reality-based therapies may
offer the best chance of success with these patients. Behavior modification programs
might be useful in treating chronic relapsing behavior. When brain injury has become a
prominent symptom of chronic use, response to treatment becomes slower and more
difficult. Urine screening 2-3 times each week for hippuric acid, the major metabolite
of toluene, may be useful. Because of the high co-occurrence of alcohol use disorders
among inhalant abusers, monitoring the patient for alcohol use and abuse may be
helpful.
The term designer drug was first used to describe the results of efforts to chem-
ically alter a controlled substance to produce a potent, psychoactive analogue that was
7. PSYCHOACTIVE SUBSTANCE USE DISORDERS 345
both legal and lucrative. Although several analogues have been successfully produced,
these synthetic drugs are illegal, making up a small portion of the illicit drug market.
The four designer drugs that have received the most attention are MDMA, MPPP, TMF
and other analogues of the opioid fentanyl, and 4-MAM.
MDMA, 3,4-methylenedioxymethamphetamine, also known as "Ecstasy," is a
synthetic analogue of MDA, 3,4-methylenedioxyamphetamine, which was developed
in 1914 as an appetite suppressant. MDMA came into use in the 1970s as an adjunct to
psychotherapy, and in 1983 it became popular as a recreational drug also known as
"XTC," "Adam," and "MDM." MDMA causes euphoria, enhanced sociability, and a
sense of increased personal insight and empathy. It is not associated with hallu-
cinogenesis. It is ingested as a gelatinous capsule or loose powder and has three phases
of action. The drug induces an initial phase of disorientation; effects are felt within 20-
60 min of administration. 246 This is followed by a "rush," which may be accompanied
by spasmodic jerking of the muscles, and finally a feeling of happy sociability. The
effects wear off in 4-6 hr. Some users have reported confusion, anxiety, and de-
pression.
The Federal Drug Enforcement Agency became concerned with the potential for
MDMA abuse and possible dangerous side effects based on ill effects of its analogue
MDA, including selective damage to serotonin nerve terminals. Since that time,
MDMA has been shown to cause serotonergic damage in animals, but its effects on
humans are questionable. A few cases of MDMA intoxication have also been impli-
cated in the onset of panic disorder, which may occur as a result of individual suscep-
tibility. In 1985, MDMA became a Schedule 1 drug, and there was an increase in its
popularity and availability on the black market. On one college campus, the prevalence
of use of MDMA increased from 16% of students in 1986 to 24% in 1990. 243 The
chemically similar MDEA, 3,4-methylenedioxyamphetamine or "Eve," became avail-
able as a nonscheduled drug in 1985 as well.
The addiction liability, abuse potential, and adverse effects of MDMA and MDEA
are controversial. MDMA has been shown to have sympathomimetic effects that have
been associated with cardiac arthythmias and deaths, especially in those with underly-
ing cardiovascular disease. Cardiac arrhythmia associated with MDMA has been asso-
ciated with at least one episode of sudden death in an otherwise healthy young person.
As of 1986, reports of MDMA-associated adverse reactions, treatment admissions, and
police reports were limited. However, increased street demand has led to anticipation of
increasing incidence of adverse reactions associated with adulterated MDMA.
The meperidine analogue I-methyl-4-propionoxy-4-phenylpyridine (MPPP) and
its contaminant MPTP have been shown to cause parkinsonian syndromes in some
of its users. The toxicity of MPPP is mediated by I-methyl-4-phenyl-I,2,3,6-tetra-
hyropyridine (MPTP), which may be a contaminant, especially when MPPP is synthe-
sized under poor temperature and pH control. MPTP causes degeneration in the sub-
stantia nigra when its toxic metabolite (MPP+) interferes with oxidative
phophorylation in this region, causing cell death. One study estimated that 6% of users
exposed to MPTP developed parkinsonian syndromes, indicating a variation in individ-
ual susceptibility.
Derivatives of the opioid fentanyl are of special concern among the controlled
substance analogues as a result of their high potency. Among the several analogue
346 SHELLY F. GREENFIELD, M.D., M.P.H., ET AL
A. Introduction
Marijuana, a plant used for recreational and medicinal purposes for centuries, is
still widely popular today. The term "marijuana" refers to the dried leaves and flowers
of the Indian hemp plant Cannabis sativa, which grows freely in warm climates and
has psychoactive effects when smoked or ingested.
By 1985, the lifetime prevalence of marijuana use in the u.s. had increased to 62
million. Since that time, however, marijuana use has steadily declined. Recent data
indicate that the prevalence of marijuana use dropped from 50.2% in the high school
class of 1987 to 40.7% in the class of 1990. 255 Among college students, the annual
prevalence of use decreased from 51 % in 1980 to 29% in 1990. 255 This decrease in use
may be due to more widespread perception of the risks associated with marijuana: In
1990, 77.8% of high school seniors associated marijuana use with significant risk, as
opposed to 35% in 1977. 255 Interestingly, the content of the major psychoactive sub-
stance in marijuana, a9 -tetrahydrocannabinol (THC) , has increased from 1-3% in
1970 to 5-15% since 1980. This represents a THe content that is more likely to
promote tolerance, dependence, and a withdrawal syndrome when abruptly discon-
tinued. The increased potency of marijuana in the last decade has helped make can-
nabis dependence more common in the United States, in spite of the decrease in
prevalence of overall marijuana use.
second type of dependence is most frequently seen in individuals whose marijuana use
has been detected by routine employee drug screening. These individuals have a
consistent pattern of using marijuana every 24-36 hr for several years. They frequently
do not report any withdrawal symptoms, but relapse following treatment is common.
Chronic daily use of marijuana is frequently associated with a diminution or loss
of the pleasurable effects of the substance, and there may be a corresponding increase
in dysphoric mood. Other symptoms associated with chronic use and dependence are
lethargy, anhedonia, and attentional and memory problems. Long-term daily use of
marijuana has been associated with subsequent abuse of other drugs, especially alcohol
and cocaine.
D. Pharmacology
Although marijuana has both stimulant and sedative properties and will produce
hallucinations when taken in high enough doses, pharmacologically it appears to be in
a class by itself. The plant is extremely complex, containing over 400 identifiable
compounds. The principal psychoactive ingredient, however, appears to be THC,
which is found in highest concentration in the small upper leaves and flowering tops of
the plant. The drug comes in three grades of preparation: bhang, ganja, and hashish (or
charas). Bhang is the cheapest and least potent form; ganja is of a higher quality with a
greater quantity of resin; and hashish, a dried, concentrated resinous exudate of the
flowers, is several times more potent. One form of liquid hashish has been shown to
contain 60% THe. When marijuana is smoked, the degree of absorption varies, but the
average experienced user will absorb approximately half the total dose into his or her
bloodstream. Drug effects are noticeable almost immediately and reach peak intensity
within 30 min. Speed of onset is partly determined by the concentration of THC in the
preparation. After an hour, plasma levels begin to decline, and most of the subjective
effects disappear by 3 hr after the last dose. Oral administration of marijuana is
generally 20-30% as effective as smoking in delivering THC to the bloodstream.
Compared with smoking, the onset of action after ingestion is slower (30-60 min), but
the subjective effects persist for a longer period of time (5-12 hr). The rate of absorp-
tion also is influenced by the food content of the stomach.
Following administration by either route, THC leaves the blood rapidly as a result
348 SHELLY F. GREENFIELD, M.D., M.P.H., ET AL
of both hepatic metabolism and efficient uptake by body tissues. It is stored in fat
depots, where it may remain for 2-3 weeks. Some also remains bound to plasma
proteins. The drug is metabolized to Il-hydroxy-THe, which is excreted via the
gastrointestinal tract and to a lesser extent by the kidneys. Like the parent compound,
THe metabolites also bind tightly to plasma proteins and can be sequestered in fat
depots for long periods of time, with subsequent rerelease into the bloodstream. As a
result, THe can be measured in plasma for up to 6 days after a single episode of
marijuana use; THe metabolites can be detected in urine up to a month after a single
dose. Evidence of the slow rate of elimination of THe has fostered speculation that
repeated use of marijuana may lead to accumulation of THe or its metabolites in body
tissues. The experimental data are unclear at this point.
E. Acute Intoxication
1. Subjective Effects
As with other drugs, the clinical effects following a dose of marijuana depend on
the strength of the preparation, route of administration, individual variables (e.g.,
metabolic rate, prior drug experience, and personal expectations), and the setting in
which the drug is consumed.
Shortly after inhalation and absorption through the lungs, users experience a sense
of well-being or euphoria, accompanied by feelings of friendliness and relaxation.
Intoxicated persons also develop an altered time sense or "temporal disintegration," a
state in which the past, present, and future become fused. Although awareness of their
environment may be heightened, they may have less ability to communicate: their
speech is often disconnected and tangential, and some smokers become remote and
withdrawn. Thought processes are slowed, short-term memory is often impaired, and
users have difficulty concentrating. Some feel that they have achieved special insights;
others find mundane events more humorous or more poignant, and there is accompany-
ing emotional lability. Experienced users of marijuana may manage to appear perfectly
sober in public even while highly intoxicated.
2. Physiological Changes
Two reliable signs of marijuana intoxication are increased heart rate and con-
junctival injection. The rise in heart rate is directly related to the dose of THe.
Decreases in salivation, intraocular pressure, and skin temperature also are found, and,
at high doses, orthostatic hypotension can occur. Bronchodilation occurs acutely, but
chronic marijuana smoking often results in obstructive pulmonary disease similar to
that seen in tobacco smokers. Marijuana only minimally affects the waking surface
EEG; the changes are similar to those seen in the drowsy state. Sleep EEG recordings
show a significant loss of REM sleep and increases in both stage-4 and total sleep time.
2. Toxic Delirium
Following large doses of marijuana, taken either orally or by inhalation, users may
experience psychic disorganization accompanied by feelings of depersonalization and
changes in perception and body image. Some patients are disoriented, with marked
memory impairment. Confusion, derealization, paranoia, violent and bizarre behavior,
visual and auditory hallucinations, and dysphoria are also features of this toxic delir-
ium. Individuals usually remain aware that these effects are drug related. Toxic delir-
ium is more likely after cannabis is ingested, perhaps because of the user's inability to
titrate adequately the dose of THe. In most cases, the process is self-limited and lasts
from a few hours to a few days; it resolves as plasma THC levels decline. Generally, no
pharmacological treatment is indicated, but patients should be carefully observed and
prevented from doing harm to themselves or others.
2. Amotivational Syndrome
A number of authors have described an "amotivational syndrome" that develops
gradually in chronic heavy marijuana users. It is characterized by decreased drive and
ambition, a shortened attention span, poor judgment, a high degree of distractability,
impaired communication skills, and diminished effectiveness in interpersonal situa-
tions. A tendency toward introversion and magical thinking also has been observed.
The individual frequently feels incapacitated and makes few or no plans beyond the
present day. Personal habits deteriorate, and there are a progressive loss of insight and
7. PSYCHOACTIVE SUBSTANCE USE DISORDERS 351
3. Depression
Some evidence has emerged relating depression to marijuana use. In one study,
dysthymic boys were more likely to admit to having used marijuana. An acute depres-
sive reaction has also been known to develop among new users of the drug. These
depressive reactions are generally mild and transient but occasionally do require psy-
chiatric intervention. Symptoms of depression as a result of marijuana use are more
likely to appear in individuals with some underlying degree of depression. Based on
this finding, the effects of the drug may serve to enhance already present depressive
feelings. Research in this area is still ongoing.
4. Medical Problems
SELECTED READING
General
I. American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders, 4th ed.,
revised. Washington, D.C., American Psychiatric Press, 1994.
2. Baldessarini R. 1.: Chemotherapy in Psychiatry. Cambridge, Harvard University Press, 1985.
3. Braude M. C.: Perinatal effects of drugs of abuse: Symposium summary. Fed Proc 46:2446-2453,
1987.
4. Galanter M., Kleber H. D. (eds): The American Psychiatric Press Textbook of Substance Abuse
Treatment. Washington, D.C., American Psychiatric Press, 1994.
5. Lowinson 1. H., Ruiz P., Millman R. B., Langrod 1. G. (eds): Substance Abuse: A Comprehensive
Textbook. Baltimore, Williams and Wilkins, 1992.
6. McLellan A. T., Arndt I. 0., Metzger D. S., et al: The effects of psychosocial services in substance
abuse treatment. lAMA 629:1953-1959, 1993.
7. Meyer R. E. (ed): Psychopathology and Additive Disorders. New York, Guilford Press, 1986.
8. Mirin S. M., Weiss R. D., Michael J.: Psychopathology in substance abusers: Diagnosis and treatment.
Am 1 Drug Alcohol Abuse 14:139-157, 1988.
9. National Household Survey on Drug Abuse: Population Estimates 1993. DHHS Publication No. (SMA)
94-3017, Substance Abuse and Mental Health Services Administration, Drug Abuse Warning Net-
work, Office of Applied Studies, printed 1994.
10. Prochaska J. 0., DiClemente C. C., Norcross 1. C.: In search of how people change: Applications to
addictive behaviors. Am Psychologist 47:1102-1114, 1992.
II. Robins L. N., Regier A. (eds): Psychiatric Disorders in America. New York, The Free Press, 1991.
Opioids
12. Bale R. N., Van Stone W. W., Kuldau 1. M., et al: Therapeutic communities vs methadone maintenance.
A prospective controlled study of narcotic addiction treatment: Design and one-year follow-up. Arch
Gen Psychiatry 37:179-193,1980.
13. Batki S. L., Sorensen 1. L., Faltz B., et al: Psychiatric aspects of treatment of i. v. drug abusers with
AIDS. Hosp Community Psychiatry 39:439-441, 1988.
14. Charney D. S., Riordan C. E., Kleber H. D.: Clonidine and naltrexone: A safe, effective and rapid
treatment of abrupt withdrawal from methadone therapy. Arch Gen Psychiatry 39:1327-1332, 1982.
15. Crabtree B. L.: Review of naltrexone, a long-acting opiate antagonist. Clin Pharm 3:273-280, 1984.
16. Crowley T. J., Wagner J. E., Zerbe G.: Naltrexone-induced dysphoria in former opioid addicts. Am 1
Psychiatry 142:1081-1084, 1985.
17. Cushman P.: The major medical sequelae of opioid addiction. Drug Alcohol Dep 5:239-254, 1980.
18. Davis D. D., Templer D. I.: Neurobehavioral functioning in children exposed to narcotics in utero.
Addict Behav 13:275-283, 1988.
19. Dole V. P.: Implications of methadone maintenance for theories of narcotic addiction. lAMA 260:3025-
3029, 1988.
20. Dole V. P., Nyswander M. E.: Heroin addiction-a metabolic disease. Arch Intern Med 120:19-24,
1967.
21. Edelin K. c., Gurganious L., Golar K.: Methadone maintenance in pregnancy: Consequences to care
and outcome. Obstet GynecoI71:399-404, 1988.
354 SHELLY F. GREENFIELD, M.D., M.P.H., ET AL.
22. Faltz B. G.: Counseling substance abuse clients infected with human immunodeficiency virus. 1
Psychoactive Drugs 20:217-221, 1988.
23. Friedman A. S., Glickman N. w.: Residential program characteristics for completion of treatment by
adolescent drug abusers. 1 Nerv Ment Dis 175:419-424, 1987.
24. Greenstein R A., Arndt I. c., et al: Naltrexone: A clinical perspective. 1 Clin Psychiatry 45:25-28,
1984.
25. Gritz E. R., Shiffman S. M., Jarvik M. E., Mclellan A. T.: Physiological and psychological effects of
methadone in man. Arch Gen Psychiatry 32:237-242, 1975.
26. Haastrup S., Jepsen P. w.: Eleven year follow-up of 300 young opioid addicts. Acta Psychiatr Scand
77:22-26, 1988.
27. Jaffe 1. H.: Drug addiction and drug abuse, in Goodman L. S., Gilman A., Gilman A. G. (eds):
s
Goodman and Gilman The Pharmacological Basis of Therapeutics, 7th ed. New York, Macmillan,
1985, pp 532-581.
28. Jaffe J. H., Martin W. R: Opioid analgesics and antagonists, in Goodman L. S., Gilman A., Gilman
s
A. G. (eds): Goodman and Gilman The Pharmacological Basis of Therapeutics, 7th ed. New York,
Macmillan, 1985, pp 491-531.
29. Jaffe 1. H.: Opiates: Clinical Aspects, in Lowinson J. H., Ruiz P., Millman R B. (eds): Substance
Abuse: A Comprehensive Textbook, 2nd ed. Baltimore, Williams and Wilkins, 1992, pp 186-194.
30. Joe G. W., Simpson D. D., Hubbard R L.: Treatment predictors oftenure in methadone maintenance. 1
Subst Abuse 3:73-84,1991.
31. Johnston L. D., O'Malley P. M., Bachman 1. G., eds.: Drug Use among American High School Seniors,
College Students and Young Adults, 1975-1990. National Institute on Drug Abuse, DHHS Publication
No. (ADM) 91-1813. Washington, D.C., U.S. Government Printing Office, 1991.
32. Jasinski D. R, Johnson R. E., Kuchel T. R.: Clonidine in morphine withdrawal: Differential effects on
signs and symptoms. Arch Gen Psychiatry 42: 1063-1076, 1985.
33. Jasinski D. R., Pevnick 1. S., Griffith J. D.: Human pharmacology and abuse potential of the analgesic
buprenorphine: A potential agent for treating narcotic addiction. Arch Gen Psychiatry 35:501-514,
1978.
34. Kleber H. D.: Naltrexone. 1 Subst Abuse Treat 2:117-122, 1985.
35. Kleber H. D.: Opioids: Detoxification, in Galanter M., Kleber H. D. (eds): Textbook of Substance Abuse
Treatment. Washington, D.C., American Psychiatric Press, 1994, pp 191-208.
36. Kleber H. D.: Treatment of narcotic addicts. Psychiatr Med 3:389-418, 1987.
37. Kleber H. D., Kosten T. R: Naltrexone induction: Psychologic and pharmacologic strategies. 1 Clin
Psychiatry 45:29-38, 1984.
38. Kleber H. D., Riordan C. E., Rounsaville B.: Clonidine in outpatient detoxification from methadone
maintenance. Arch Gen Psychiatry 42:391-394, 1985.
39. Kleber H. D., Siobetz E: Outpatient drug-free treatment, in DuPont R L., Goldstein A., O'Donnell J.
(eds): Handbook on Drug Abuse. Washington, D.C., National Institute on Drug Abuse, 1979, pp 31-38.
40. Kosten T. R, Kleber H. D.: Buprenorphine detoxification from opioid dependence: A pilot study. Life
Sci 42:635-641, 1988.
41. Kosten T. R., Kleber H. D.: Strategies to improve compliance with narcotic antagonists. Am 1 Drug
Alcohol Abuse 11 :249-266, 1984.
42. Kosten T. R, Rounsaville B. J.: Psychopathology in opioid addicts. Psychiatr Clin North Am 9:515-
532, 1986.
43. Kreek M. 1.: Medical safety and side effects of methadone in tolerant individuals. lAMA 223:665-668,
1973.
44. Martin W. R: Naloxone. Ann Intern Med 85:765-768, 1976.
45. Mello N. K., Mendelson 1. H.: Buprenorphine suppresses heroin use by heroin addicts. Science
207:657-659, 1980.
46. Meyer R. E., Mirin S. M.: The Heroin Stimulus: Implications for a Theory of Addiction. New York,
Plenum Press, 1979.
47. Milby 1. B.: Methadone maintenance to abstinence-how many make it? 1 Nerv Ment Dis 176:409-
422, 1988.
7. PSYCHOACTIVE SUBSTANCE USE DISORDERS 355
48. Mirin S. M., Meyer R. E., McNamee H. B.: Psychopathology and mood during heroin use. Arch Gen
Psychiatry 33:1503-1508, 1976.
49. Mirin S. M., Meyer R. E., Mendelson, J. H., et al: Opiate use and sexual function. Am J Psychiatry
137:909-915, 1980.
50. National Research Council Committee on Clinical Evaluation of Narcotic Antagonists: Clinical evalua-
tion of naltrexone treatment of opiate-dependent individuals. Arch Gen Psychiatry 35:335-340, 1978.
51. Newman R. G.: Methadone treatment-defining and evaluating success. N Engl J Med 317:447-450,
1987.
52. O'Brien C. P.: Opioids: Antagonists and Partial Agonists, in Galanter M., Kleber H. D. (eds): Textbook
of Substance Abuse Treatment. Washington, D.C., American Psychiatric Press, 1994, pp 223-239.
53. O'Brien C. P., Testa T., O'Brien T. 1., et al: Conditioned narcotic withdrawal in humans. Science
195: 1000-1002, 1977.
54. Resnick R. B., Schuyten-Resnick E., Washton A. M.: Treatment of opioid dependence with narcotic
antagonists: A review and commentary, in DuPont R. L., Goldstein A., O'Donnell J. (eds): Handbook
on Drug Abuse. Washington, D.C., National Institute on Drug Abuse, 1979, pp 76-104.
55. Rounsaville B. 1., Kosten T., Kleber H.: Success and failure at outpatient detoxification. Evaluating the
process of clonidine and methadone-assisted withdrawal. J Nerv Ment Dis 173: 103-110, 1985.
56. Rounsaville B. J., Weissman M. M., Crits-Cristoph K.: Diagnosis and symptoms of depression in opiate
addicts. Course and relationship to treatment outcome. Arch Gen Psychiatry 39: 151-156, 1982.
57. Schuster C. R.: Intravenous drug use and AIDS prevention. Public Health Rep 103:261-266, 1988.
58. Sells S. B., Simpson D. D.: The case for drug abuse treatment effectiveness, based on the DARP
research program. Br J Addict 75:117-131, 1980.
59. Senay E. C.: Methadone maintenance treatment. Int J Addict 20:803-821, 1985.
60. Simon E. J., Opiates: Neurobiology, in Lowinson 1. H., Ruiz P., Millman R. B. (eds): Substance Abuse:
A Comprehensive Textbook, 2nd ed. Baltimore, Williams and Wilkins, 1992, pp 195-204.
61. Simpson D. D., Savage L. 1.: Drug abuse treatment readmissions and outcomes: Three-year follow-up
of DARP patients. Arch Gen Psychiatry 37:896-901, 1980.
62. Snyder S. H.: Opiate receptors in the brain. N Engl J Med 296:266-278, 1977.
63. Stimmel B., Adamsons K.: Narcotic dependency in pregnancy: Methadone maintenance compared to
use of street drugs. JAMA 235:1121-1124, 1976.
64. Suffet E, Brotman R.: A comprehensive care program for pregnant addicts: Obstetrical, neonatal and
child development outcomes. Int J Addict 19:199-219, 1984.
65. U.S. Department of Health, Education and Welfare, Alcohol, Drug Abuse and Mental Health Adminis-
tration: Management of labor, delivery and immediate post-partum period, in: Drug Dependence in
Pregnancy: Clinical Management of Mother and Child. Washington, D.C., National Institute on Drug
Abuse, Services Research Monograph Series, 1979, Chapter 4.
66. Vaillant G. E.: A 20-year follow-up of New York narcotic addicts. Arch Gen Psychiatry 29:237-241,
1973.
67. Washton A. M., Resnick R. B., Geyer G.: Opiate withdrawal using lofexidine, a clonidine analogue
with fewer side effects. J Clin Psychiatry 44:335-337, 1983.
68. Wikler A.: Dynamics of drug dependence. Arch Gen Psychiatry 28:611-616, 1973.
69. Zelson C., Lee S. J., Casalino M.: Neonatal narcotic addiction: Comparative effects of maternal intake
of heroin and methadone. N Engl J Med 289:1216-1220, 1973.
73. Ciraulo D. A., Sands B. F., Shader R. I.: Abuse liability of benzodiazepines in alcoholics: A critical
review. Am J Psychiatry 145:150\-1506, 1988.
74. Cohen L. S., Rosenbaum J. F.: Clonazepam: New uses and potential problems. J Clin Psychiatry
48(Suppl 10):50-55, 1987.
75. Cronin R. 1., Klingler E. L., Jr., Avashti P. S., et al: The treatment of nonbarbiturate sedative over-
dosage, in Browne P. G. (ed): A Treatment Manual for Acute Drug Abuse Emergencies. Washington,
D.C., National Institute on Drug Abuse, 1974, pp 58-62.
76. Dorpat T. L.: Drug automatism, barbiturate poisoning and suicide behavior. Arch Gen Psychiatry
31:216-220, 1974.
77. Dupont R. L. (ed): Abuse of benzodiazepines: The problems and the solutions. Am J Alcohol Abuse
14(Suppl 1):1-69, 1988.
78. Fleischhacker W. W., Bamas C., Hackenberg B.: Epidemiology of benzodiazepine dependence. Acta
Psychiatr Scand 74(1):80-83, 1986.
79. Gelenberg A. J.: Benzodiazepine withdrawal. Mass Gen Hosp Bioi Ther Psychiatry Newslett 3:9-12,
1980.
80. Geller A: Protracted abstinence, in Miller N. S. (ed.): Comprehensive Handbook of Drug and Alcohol
Addiction, New York, Marcel Dekker, 1991, pp 905-913.
81. Herman J. B., Rosenbaum J. F., Brotman A. W.: The alprazolam to clonazepam switch for the treatment
of panic disorder. J Clin PsychopharmacoI7:175-178, 1987.
82. Juergens S. M., Morse R. M.: Alprazolam dependence in seven patients. Am J Psychiatry 145:625-627,
1988.
83. Mellman T. A, Uhde T. W.: Withdrawal syndrome with gradual tapering of alprazolam. Am J Psychia-
try 143:1464-1466, 1986.
84. Noyes R., Jr., Clancy J., Coryell W. H., et al: A withdrawal syndrome after abrupt discontinuation of
alprazo1am. Am J Psychiatry 142:114-116, 1985.
85. Noyes R., Jr., Perry P. J., Crowe R. R.: Seizures following the withdrawal of alprazolam. J Nerv Ment
Dis 174:50-52, 1986.
86. Pevnick 1. S., Jasinski D. R., Haertzen C. A: Abrupt withdrawal from therapeutically administered
diazepam. Arch Gen Psychiatry 35:995-998, 1978.
87. Piesiur-Streh10w B., Strelow U., Poser W.: Mortality of patients dependent on benzodiazepines. Acta
Psychiatr Scand 73:330-335, 1986.
88. Preskom S. H., Denner L. J.: Benzodiazepines and withdrawal psychosis. Report of three cases. JAMA
237:36-38, 1977.
89. Ray W. A., Blazer D. G., Schaffner W.: Reducing long-term diazepam prescribing in office practice.
JAMA 256:2536-2539, 1986.
90. Setter J. G.: Emergency treatment of acute barbiturate intoxication, in Browne P. G. (ed): A Treatment
Manual for Acute Drug Abuse Emergencies. Washington, D.C., National Institute on Drug Abuse,
1974, pp 49-53.
91. Smith D. E., Wesson D. R.: Phenobarbital technique for treatment of barbiturate dependence. Arch Gen
Psychiatry 24:56-60, 1971.
92. Smith D. E., Wesson D. R.: Benzodiazepines and other sedative-hypnotics, in Galanter M., Kleber
H. D. (eds): Textbook of Substance Abuse Treatment. Washington, D.C., American Psychiatric Press,
1994, pp 179-190.
93. Smith D. E., Wesson D. R., Seymour R. B.: The abuse of barbiturates and other sedative-hypnotics, in
DuPont R. L., Goldstein A, O'Donnell I. (eds): Handbook on Drug Abuse. Washington, D.C., National
Institute on Drug Abuse, 1979, pp 233-240.
94. Swartzburg M., Lieb 1., Schwartz A. H.: Methaqualone withdrawal. Arch Gen Psychiatry 29:46-47,
1973.
95. Tennant F. S., Ir., Pumphrey E. A: Benzodiazepine dependence of several years duration: Clinical
profile and therapeutic benefits. Natllnst Drug Abuse Res Monogr 55:211-216, 1984.
96. Wesson D. R., Smith D. E., Seymour R. B.: Sedative-hypnotics and tricycJics, in Lowinson 1. H., Ruiz P.,
Millman R. B. (eds): Substance Abuse: A Comprehensive Textbook. 2nd ed. Baltimore, Williams and
Wilkins, 1992, pp 271-279.
7. PSYCHOACTIVE SUBSTANCE USE DISORDERS 357
97. National Household Survey on Drug Abuse: Population Estimates 1993. DHHS Publication No.(SMA)
94-3017, Substance Abuse and Mental Health Services Administration, Drug Abuse Warning Net-
work, Office of Applied Studies, printed 1994.
Alcohol
98. Adams R. D., Victor M.: Principles of Neurology. New York, McGraw-Hill, 1985, pp 808-824.
99. Alterman A. I., Hayashida M., O'Brien C. P.: Treatment response and safety of ambulatory medical
detoxification. J Stud Alcohol 49:160-166, 1988.
100. Avins A., Woods W. J., Lindan C. P., et al: HIV Infection and risk behaviors among heterosexuals in
alcohol treatment programs. JAMA 271:515-518, 1994.
101. Bean M.: Alcoholics Anonymous: Chapter I: Principles and methods. Psychiatr Ann 5:7-21, 1975.
102. Blume S. B.: Women and alcohol. A review. JAMA 256:1467-1470, 1986.
103. Brown C. G.: The alcohol withdrawal syndrome. Ann Emerg Med 11:276-280, 1982.
104. Chick 1., Ritson B., Connaughton J., Stewart A., Chick 1.: Advice vs. extended treatment for alcohol-
ism: A controlled study Br J Addict 83: 159-170, 1988.
105. Ciraulo D. A., Barnhill J. G., Greenblatt D. J.: Abuse liability and clinical pharmacokinetics of
alprazolam in alcoholic men. J Clin Psychiatry 49(9):333-337, 1988.
106. Cloninger C. R: Neurogenetic adaptive mechanisms in alcoholism. Science 236:410-416, 1987.
107. Cohen S. (ed): The treatment of alcoholism: Does it work? Drug Abuse Alcoholism Newslett 7:1-3,
1978.
108. Coles C. D., Smith I., Fernhoff P. M.: Neonatal neurobehavioral characteristics as correlates of maternal
alcohol use during gestation. Alcoholism (NY) 9:454-460, 1985.
109. Council on Scientific Affairs: Aversion therapy. JAMA 258:2562-2566, 1987.
110. Dorus W., Ostrow D., Anton R, Cusshman P., et. al.: Lithium treatment of depressed and nondepressed
alcoholics. JAMA 262:1646-1652, 1989.
Ill. Edwards G., Orford 1., Egert S., et al: Alcoholism: A controlled trial of "treatment" and "advice." J
Stud Alcohol 38:1004-1031, 1977.
112. Emrick C. E.: A review of psychologically oriented treatment of alcoholism: II. The relative effective-
ness of different treatment approaches and the effectiveness of treatment versus no treatment. J Stud
Alcohol 36:88-108, 1975.
113. Emrick C. E.: Alcoholics Anonymous: Affilitation processes and effectiveness as treatment. Alcohol-
ism (NY) 11:416-423, 1987.
114. Erhart C. B., Wolf A. W., Emhart C. B., Wolf A. W., Linn P. L.: Alcohol-related birth defects:
Syndromal anomalies, intrauterine growth retardation, and neonatal behavioral assessment. Alcoholism
(NY) 9:447-453, 1985.
115. Ewing 1. A.: Detecting alcoholism: The CAGE questionnaire. JAMA 252:1905-1907, 1984.
116. Frances R. J.: Update on alcohol and drug disorder treatment. J Clin Psychiatry 49(Suppl):13-17,
1988.
117. Fuller R. K., Roth H. P.: Disulfiram for the treatment of alcoholism. Ann Intern Med 90:901-904, 1979.
118. Gallant D: Alcohol, in Galanter M., Kleber H. D. (eds): Textbook of Substance Abuse Treatment.
Washington, D.C., American Psychiatric Press, 1994, pp 67-89.
119. Institute of Medicine: Broadening the Base of Treatment for Alcohol Problems. Washington, D.C.,
National Academy Press, 1990.
120. Holder H. D., Longabaugh R., Miller W. R., Rubonis A. V.: The cost-effectiveness of treatment for
alcoholism: A first approximation. J Stud Alcohol 52:517-540, 1991.
121. Kendler K. S., Heath A. C., Neale M. c., Kessler R c., Eaves L. J.: Alcoholism and major depression
in women. A twin study of the causes of comorbidity. Arch Gen Psychiatry 50:690-698,1993.
122. Kwentus M. D., Major L. E: Disulfiram in the treatment of alcoholism. J Stud Alcohol 40:428-446,
1979.
123. Linnoila M., Mefford I., Nutt D. W.: NIH conference. Alcohol withdrawal and noradrenergic function.
Ann Intern Med 107:875-889, 1987.
358 SHELLY F. GREENFIELD, M.D., M.P.H., ET AL
124. Liskow B. I., Goodwin D. W.: Pharmacological treatment of alcohol intoxication, withdrawal and
dependence: A critical review. J Stud Alcohol 48:356-370, 1987.
125. Liskow B., Mayfield D., Thiele J.: Alcohol and affective disorder: Assessment and treatment. J Clin
Psychiatry 43(4):144-147, 1982.
126. Longabaugh R.: Longitudinal outcome studies. in Rose R. M., Barrett J. (eds): Alcoholism: Origins and
Outcome. New York, Raven Press, 1988, p 81.
127. McKay J. R., Longabauth R., Beattie M. C., Maisto S. A.: The relationship of pretreatment family
functioning to drinking behavior during follow-up by alcoholic patients. Am J Drug Alcohol Abuse
18:445-460, 1992.
128. Mendelson J. H., Babor T. F., Mello N. K., et al: Alcoholism and prevalence of medical and psychiatric
disorders. J Stud Alcohol 47:361-366, 1986.
129. Mendelson J. H., Mello N. K.: Biologic concomitants of alcoholism. N EnglJ Med301:912-92I, 1979.
130. Mendelson J. H., Mello N. K. (eds): The Diagnosis and Treatment of Alcoholism. New York, McGraw-
Hill, 1992.
131. Meyer R. E., Kranzier H. R.: Alcoholism: Clinical implications of recent research. J Clin Psychiatry
49(Suppl):8-12, 1988.
132. Miller N. S.: Motivational interviewing with problem drinkers. Behav Psychother 11:147-172, 1983.
133. Miller W. R., Hester R.: Matching problem drinkers with optimal treatments, in Miller W. R., Heather N.
(eds): Treating Dddictive Behaviors: Process of Change. New York, Plenum, 1986, pp 175-203.
134. Moos R. H., Finney J. W., Cronkite R. C. (eds): Alcoholism Treatment: Context, Process, and Outcome.
New York, Oxford University Press, 1990.
135. Naidoo D. P., Bramdev A., Cooper K.: Wernicke's encephalopathy and alcohol-related disease. Post-
grad Med J 67:978-981, 1991.
136. Naranjo C. A., Kadlec K. E., Sanheuza P., et al.: Fluoxetine differentially alters alcohol intake and other
consummatory behavior in problem drinkers. Clin Pharmacol Ther 47:490-498, 1990.
137. Newsom G., Murray N.: Reversal of dexamethasone suppression test nonsuppression in alcohol abus-
ers. Am J Psychiatry 140:353-354, 1983.
138. Ng S. K., Hauser W. A., Brust J. c.: Alcohol consumption and withdrawal in new-onset seizures. N
Engl J Med 319:666-673, 1988.
139. Nunes E. V., McGrath P. J., Quitkin F. M., et. al.: Imipramine treatment of alcoholism with comorbid
depression. Am J Psychiatry 150:963-965, 1993.
140. O'Parrell T. J., Cutter H. S., Floyd F. J.: Evaluating behavioral marital therapy for male alcoholics:
Effects on marital adjustment and communication from before to after treatment. Behov Ther 16:147-
167, 1985.
141. O'Malley S. S., Jaffe A., Chang G., Schottenfeld R. S., et al: NaItrexone and coping skills therapy for
alcohol dependence: A controlled study. Arch Gen Psychiatry 49:881-887, 1992.
142. Ouelette E. M., Rosett H. L., Rosman N. P., et al: Adverse effects on offspring of maternal alcohol
abuse during pregnancy. N Engl J Med 297:528-530, 1977.
143. Pomerleau 0., Pertschuk M., Stunnett J.: A critical examination of some current assumptions in the
treatment of alcoholism. J Stud Alcohol 37:849-867, 1976.
144. Schuckit M. A.: A clinical review of alcohol, alcoholism and the elderly patient. J Clin Psychiatry
43:396-399, 1982.
145. Sellers E. M., Naranjo C. A., Peachey J. E.: Drug therapy: Drugs to decrease alcohol consumption. N
Engl J Med 305:1255-1262, 1981.
146. Stabenau J. R.: Implications of family history of alcoholism, antisocial personality, and sex differences
in alcohol dependence. Am J Psychiatry 141: 1178-1182, 1984.
147. Staiesey N. L., Fried P. A.: Relationships between moderate maternal alcohol consumption during
pregnancy and infant neurological development. J Stud Alcohol 44:262-270, 1983.
148. Streissguth A. P., Clarren S. K., Jones K. L.: Natural history of the fetal alcohol syndrome: A 10 year
follow-up of eleven patients. Lancet 2:85-91, 1985.
149. Streissguth A. P., Sampson P. D., Barr H. M.: Studying alcohol teratogenesis from the perspective of the
fetal alcohol syndrome: Methodological and statistical issues. Ann NY Acad Sci 488:73-86, 1986.
7. PSYCHOACTIVE SUBSTANCE USE DISORDERS 359
ISO. Thompson W. L., Johnson A. D., Maddrey W. L., et al: Diazepam and paraldehyde for treatment of
severe delirium tremens: A controlled trial. Ann Intern Med 82:175-180, 1975.
151. Vaillant G. E.: The Natural History of Alcoholism: Causes. Patterns. and Paths of Recovery. Cam-
bridge, Mass., Harvard University Press, 1985.
152. Vaillant G. E.: Natural history of male psychological health: VIII. Antecedents of alcoholism and
"orality." Am J Psychiatry 137:181-186, 1980.
153. Volpicelli 1. R., Alterman A. I., Hagashida M., O'Brien C. P.: Naltrexone in the treatment of alcohol
dependnce. Arch Gen Psychiatry 49:876-880, 1992.
154. Walsh D. C.,Hingson R. w., Merrigan D. M., Morelock-Levenson S., et. al: A randomized trial of
tretment options for alcohol-abusing workers. N Engl J Med 325:775-782,1991.
155. Warren K. R., Bast R. 1.: Alcohol-related birth defects: An update. Public Health Rep 103:638-642,
1988.
156. Weissman M. M., Myers 1. K.: Clinical depression in alcoholism. Am J Psychiatry 137:372-374, 1980.
157. Wilson G. T.: Chemical aversion conditioning as a treatment for alcoholism: A reanalysis. Behav Res
Ther 25:503-516, 1987.
158. Wright C., Vafier 1. A., Lake C. R.: Disulfiram-induced fulminating hepatitis: Guidelines for liver panel
monitoring. J Clin Psychiatry 49:430-434, 1988.
Cocaine
159. Aigner T. G., Balster R. L.: Choice behavior in rhesus monkeys: Cocaine versus food. Science
201:534-535, 1978.
160. Ambre J. J., Belknap S. M., Nelson J., et al: Acute tolerance to cocaine in humans. Clin Pharmacol
Ther 44: 1-8, 1988.
161. Batki S. L., Manfredi L. B., Jacob P., III, Jones R. T.: Fluoxetine for cocaine dependence in methadone
maintenance: Quantitative plasma and urine cocaine/benzoylecgonine concentrations. J Clin Psycho-
pharmacol 13:243-250, 1993.
162. Brotman A. w., Witkie S. M., Gelenberg A. J., et al: An open trial of maprotiline for the treatment of
cocaine abuse: A pilot study. J Clin Psychopharmacol 8:125-127, 1988.
163. Byck R., Van Dyke c.: What are the effects of cocaine in man?, in Peterson R. C., Stillman R. C. (eds):
Cocaine: 1977. Washington, D.C., National Institute on Drug Abuse, Research Monograph 13, 1977,
pp 5-16.
164. Chaisson R. E., Bacchetti R., Osmond D., et al: Cocaine use and HIV infection in intravenous drug
users in San Francisco. lAMA 261:561-565, 1989.
165. Chasnoff I., MacGregor S., Chisum G.: Cocaine use during pregnancy: Adverse perinatal outcome, in:
Problems of Drug Dependence. /987. Proceedings of the 49th Annual Scientific Meeting. Committee on
Problems of Drug Dependence. Inc. Washington, D.C., National Institute on Drug Abuse, Research
Monograph 81, 1988, P 265.
166. Anon: Crack. Med Lett 28:69-70, 1986.
167. Cregler L. L., Mark H.: Special report: Medical complications of cocaine abuse. N Engl J Med
315:1495-1500, 1986.
168. Ellinwood E. H., Jr.: Amphetamine psychosis: Individuals, settings and sequences, in Ellinwood E. H.,
Cohen S. (eds): Current Concepts on Amphetamine Abuse. Washington, D.C., National Institute of
Mental Health, 1972, pp 143-158.
169. Ellinwood E. H., Jr.: Amphetamines/anorectics, in DuPont R. L., Goldstein A., O'Donnell 1. (eds):
Handbook on Drug Abuse. Washington, D.C., National Institute on Drug Abuse, 1979, pp 221-231.
170. Finnegan L.: The dilemma of cocaine exposure in the perinatal period, in: Problems of Drug Depen-
dence. 1987. Proceedings of the 49th Annual Scientific Meeting. Committee on Problems of Drug
Dependence. Inc. Washington, D.C., National Institute on Drug Abuse, Research Monograph 81, 1988,
P 379.
171. Fischman M. W., Schuster C. R., Resnekov L., et al: Cardiovascular and subjective effects of intra-
venous cocaine administration in humans. Arch Gen Psychiatry 33:983-989, 1976.
360 SHELLY F. GREENFIELD, M.D., M.P.H., ET AL.
172. Gastfriend D. R., Mendelson J. H., Mello N. K, Teoh S. K: Reif S., Baker S. L.: Long-term effects of
buprenorphine in the outpatient treatment of combined heroin and cocaine dependence. Problems of
Drug Dependence, 1994: Proceedings of the 56th Annual Scientific Meeting, The College on Problems
of Drug Dependence, Inc., Washington, D.C., Rockville, Md., National Institute on Drug Abuse,
Research Monograph 153, 1995, p. 9.
173. Gawin F. H., Ellinwood E. H.: Cocaine and other stimulants: Actions, abuse and treatment. N Engl 1
Med 318:1173-1182, 1988.
174. Gawin F. H., Kleber H. D.: Abstinence symptomatology and psychiatric diagnosis in cocaine abusers:
Clinical observations. Arch Gen Psychiatry 43:107-113, 1986.
175. Gawin E H., Kleber H. D.: Pharmacologic treatments of cocaine abuse. Psychiatr Clin North Am
9:573-583, 1986.
176. Gawin F. H., Kleber H. D., Byck R., et al: Desipramine facilitation of initial cocaine abstinence. Arch
Gen Psychiatry 46:117-121, 1989.
177. Gawin F. H., Morgan C., Kosten T. R, Kleber H. D.: Double-blind evaluation of the effect of acute
amantadine on cocaine craving. Psychopharmacology 97:402-403, 1989.
178. Giannini A. 1., Malone D. A., Giannini M. c.: Treatment of depression in chronic cocaine and
phencyclidine abuse with desipramine. 1 Clin PharmacoI26:211-214, 1986.
179. Gorelick D. A.: Overview of pharmacologic treatment approaches for alcohol and other drug addiction.
Psychiatr Clin North Am 16:141-156, 1993.
180. Goldfrank L. R, Hoffman R S.: The cardiovascular effects of cocaine. Ann Emerg Med 20:165-175,
1991.
181. Grinspoon L., Hedblom P.: The Speed Culture: Amphetamine Use and Abuse in America. Cambridge,
Harvard University Press, 1975.
182. Higgins S. T., Budney A. 1., Bickel W. K, et al: Incentives improve outcome in outpatient behavioral
treatment of cocaine dependence. Arch Gen Psychiatry 51:568-576, 1994.
183. Isner J. M., Estes M., Thompson P. D., et al: Acute cardiac events temporarily related to cocaine abuse.
N Engl 1 Med 315:1438-1443, 1986.
184. Javaid 1. I., Fischman M. w., Schuster C. R., et al: Cocaine plasma concentration: Relation to physi-
ological and subjective effects in humans. Science 202:227-228, 1978.
185. Johnson R E., Jaffe J. H., Fudala P. J.: A controlled trial of buprenorphine treatment for opioid
dependence. lAMA 267:2750-2755, 1992.
186. Kosten T. R, Kleber H. D., Morgan c.: Treatment of cocaine abuse with buprenorphine. Bioi Psychia-
try 26: 170-172, 1989.
187. Kramer J. C.: Introduction to amphetamine abuse, in Ellinwood E. H., Cohen S. (eds): Current
Concepts on Amphetamine Abuse. Washington, D.C., National Institute of Mental Health, 1972, pp
177-184.
188. Kramer 1. c., Fischman V. S., Littlefield D. C.: Amphetamine abuse: Pattern and effects of high doses
taken intravenously. lAMA 201:89-93, 1967.
189. Malcolm R, Hutto, B. R., Phillips 1. D., et al: Pergolide mesylate treatment of cocaine withdrawal. 1
Clin Psychiatry 52:39-40, 1991.
190. Margolin A., Kosten T., Petrakis I., et al: An open pilot study of bupropion and psychotherapy for the
treatment of cocaine abuse in methadone-maintained patients, in Harris L. (ed): Problems of Drug
Dependence 1990, Rockville, Md., National Institute on Drug Abuse, Research Monograph 105, 1991.
191. Mody C. K, Miller B. L., McIntyre H. B., et al: Neurologic complications of cocaine abuse. Neurology
38:1189-1193, 1988.
192. O'Brien C. P., Childress A. R., McLellan T., Ehrman R.: Integrating systematic cue exposure with
standard treatment in recovering drug dependent patients. Addict Behav 15:355-365, 1990.
193. Pollack M. H., Rosenbaum J. E: Fluoxetine treatment of cocaine abuse in heroin addicts. 1 Clin
Psychiatry 52:31-33, 1991.
194. Pollack M. H., Brotman A. W., Rosenbaum 1. F.: Cocaine abuse and treatment. Compr Psychiatry
30:31-44, 1989.
195. Post R M.: Cocaine psychoses: A continuum model. Am 1 Psychiatry 132:225-231, 1975.
7. PSYCHOACTIVE SUBSTANCE USE DISORDERS 361
196. Post R. M., Kotin J., Goodwin F. K.: The effects of cocaine on depressed patients. Am J Psychiatry
131 :511-517, 1974.
197. Resnick R. B., Kestenbaum R. S.: Acute systemic effects of cocaine in man: A controlled study by
intranasal and intravenous routes. Science 195:696-698, 1977.
198. Rosenfeld W., Zabaleta I., Sahdev R., et al: Maternal use of cocaine, methadone, heroin and alcohol:
Comparison of neonatal effects, in: Problems of Drug Dependence, 1987, Proceedings of the 49th
Annual Scientific Meeting, Committee on Problems of Drug Dependence, 1nc. Washington, D.C.,
National Institute on Drug Abuse, Research Monograph 81, 1988, P 264.
198. Roth D., Alarcon F. J., Fernandez J. A.: Acute rhabdomyolysis associated with cocaine intoxication. N
Engl J Med 319:673-677, 1988.
199. Rounsaville B. J., Foley Anton S., Carroll K, Budde D., PrusoffB. A., Gawin F.: Psychiatric diagnoses
of treatment-seeking cocaine abusers. Arch Gen Psychiatry 48:43-51, 1991.
200. Smith D. E., Wesson D. R., Buxton M. E., et al (eds): Amphetamine Use, Misuse and Abuse: Proceed-
ings of the National Conference. Boston, G. K Hall, 1979.
201. National Household Survey on Drug Abuse: Population Estimates 1993. DHHS Publication No. (SMA)
94-3017, Substance Abuse and Mental Health Services Administration, Drug Abuse Waming Net-
work, Office of Applied Studies, printed 1994.
202. Washton A. M.: Nonpharmacologic treatment of cocaine abuse. Psychiatr Clin North Am 9:563-571,
1986.
203. Weiss R. D.: Relapse to cocaine abuse after initiating desipramine treatment. JAMA 260:2545-2546,
1988.
204. Weiss R. D., Gawin F. H.: Protracted elimination of cocaine metabolites in long term high-dose cocaine
abusers. Am J Med 85:879-880, 1988.
205. Weiss R. D., Mirin S. M., Bartel R.: Cocaine. Washington, D.C., American Psychiatric Press, 1993.
206. Weiss R. D., Mirin S. M.: Subtypes of cocaine abusers. Psychiatr Clin North Am 9:491-501, 1986.
207. Weiss R. D., Mirin S. M., Griffin M. L., et al: Psychopathology in cocaine abusers: Changing trends. J
Nerv Ment Dis 176:719-725, 1988.
208. Weiss R. D., Mirin S. M., Griffin M. L., et al: Personality disorders in cocaine dependence. Compr
Psychiatr 34:145-149, 1993.
209. Wolfsohn R., Angrist B.: A pilot trial of levodopa/carbidopa in early cocaine abstinence. J Clin
Psychopharmacol 10:440-442, 1990.
210. Woods, J. R., Plessinger M. A, Clark K. E.: Effect of cocaine on uterine blood flow and fetal
oxygenation. JAMA 257:957-961, 1987.
Hallucinogens
211. Abraham H. D.: Visual phenomenology of the LSD flashback. Arch Gen Psychiatry 40:884-889, 1983.
212. Blacker K H., Reese T. J., Stone G. C., et al: Chronic users of LSD: The "acidheads." Am J Psychiatry
125:341-351, 1968.
213. Bowers M. B., Jr.: Acute psychosis induced by psychotomimetic drug abuse. I. Clinical findings. Arch
Gen Psychiatry 27:437-439, 1972.
214. Bowers M. B., Jr.: Acute psychosis induced by psychotomimetic drug abuse. II. Neurochemical
findings. Arch Gen Psychiatry 27:440-442, 1972.
215. Bowers M. B., Jr., Chipman A, Schwartz A., et al: Dynamics of psychedelic drug abuse: A clinical
study. Arch Gen Psychiatry 16:560-566, 1967.
216. Brawley P., Duffield J. c.: The pharmacology of hallucinogens. Pharmacol Rev 24:31-66, 1972.
217. Cohen S.: Flashbacks. Drug Abuse Alcoholism Newslett 6:1-3, 1977.
218. Faillace L. A., Snyder S. H., Weingartner H.: 2,5-Dimethoxy-l-methylamphetamine: Clinical evalua-
tion of a new hallucinogenic drug. J Nerv Ment Dis 150:119-126, 1970.
219. Freedman D. X.: The use and abuse of LSD. Arch Gen Psychiatry 18:330-347, 1968.
220. Gilmour D. G., Bloom A. D., Lele K. P., et al: Chromosomal aberrations in users of psychoactive drugs.
Arch Gen Psychiatry 24:268-272, 1971.
362 SHELLY F. GREENFIELD, M.D., M.P.H., ET AL
221. Glass G. S., Bowers M. B., Jr.: Chronic psychosis associated with long-term psychotomimetic drug
abuse. Arch Gen Psychiatry 23:97-102, 1970.
222. Grinspoon L., Bakalar 1. B.: Psychedelic Drugs Reconsidered. New York, Basic Books, 1979.
223. Annual Data Report 1981: Data from the Drug Abuse Warning Network. NIDA, Division of Data and
Information Development, May 1982 Data File.
224. Jacobson C. B., Berlin C. M.: Possible reproductive detriment in LSD users. JAMA 222:1367-1373,
1972.
225. Johnston L.: Monitoring the Future: The National High School Senior Survey. Rockville, Md., National
Institute on Drug Abuse, 1990.
226. McGlothlin W. H., Arnold D.O.: LSD revisited. Arch Gen Psychiatry 24:35-49, 1971.
227. McGlothlin W. H., Arnold D.O., Freedman D. X.: Organicity measures following repeated LSD
ingestion. Arch Gen Psychiatry 21:704-709, 1969.
228. Nagy 1.: A Comparison of Drug Use among 8th, 10th, and 12th Graders from NIDA's High School
Senior Survey. Rockville, Md., National Institute on Drug Abuse, 1992.
229. National Household Survey on Drug Abuse: Population Estimates 1993. DHHS Publication No. (SMA)
94-3017, Substance Abuse and Mental Health Services Administration, Drug Abuse Warning Net-
work, Office of Applied Studies, printed 1994.
230. Seligmann J., Mason M., Annin P, et al: The new age of aquarius. Newsweek. February 3, 1992, pp 66-
67.
231. Shick 1. F. E., Smith D. E.: An analysis of the LSD flashback. J Psychedelic Drugs 3:13-19. 1970.
232. Snyder S. H .• Faillace L. A .• Weingartner H.: DOM (STP). a new hallucinogenic drug. and DOET:
Effects in normal subjects. Am J Psychiatry 125:113-120. 1968.
233. Snyder S. H., Weingartner H .• Faillace L. A.: DOET (2.5-dimethoxy-4-ethylamphetamine). a new
psychotropic drug: Effects of varying doses in man. Arch Gen Psychiatry 24:50-55. 1971.
234. Tucker G. 1.. Quinlan D .• Harrow M.: Chronic hallucinogenic drug use and thought disturbance. Arch
Gen Psychiatry 27:443-447, 1972.
235. Ungerleider 1. T.. Pechnick R. N.: Hallucinogens. in Galanter. M .• Kleber. H. D. (eds): Textbook of
Substance Abuse Treatment. Washington. D.C .• American Psychiatric Press, 1994. pp 141-148.
Phencyclidine
236. Kozel N. J., Hudson B.: An epidemiologic assessment of selected drug trends. NIDA Res Monogr 2:5-
19. 1981.
237. Zukin S. R.. Zukin R. S.: Phencyclidine. in Lowinson J. H., Pedro R.• Millman R. B. (eds): Substance
Abuse: A Comprehensive Textbook, 2nd ed. Baltimore, Williams and Wilkins. 1992. pp. 290-302.
Inhalants
238. Cohen S.: Inhalants. in DuPont R. L., Goldstein A., O'Donnell J. (eds): Handbook on Drug Abuse,
Washington. D.C.. National Institute on Drug Abuse. 1979. pp 213-220.
239. Glaser F. B.: Inhalation psychosis and related states. in Browne P. G. (ed): A Treatment Manual for
Acute Drug Abuse Emergencies. Washington, D.C., National Institute on Drug Abuse, 1975. pp 95-
104.
240. Lewis J. D .• Moritz D .• Mellis L. P.: Long-term toluene abuse. Am J Psychiatry 138:368-370. 1981.
241. Sharp C. w.. Rosenberg N. L.: Volatile substances. in Lowinson 1. H .• Ruiz P., Millman R. B. (eds):
Substance Abuse: A Comprehensive Textbook, 2nd ed. Baltimore. Williams and Wilkins. 1992, pp 303-
327.
242. Westermeyer 1.: The psychiatrist and solvent-inhalant abuse: Recognition. assessment. and treatment.
Am J Psychiatry 144:903-907. 1987.
244. Dowing G. P., McDonough E. T., Bost R. 0.: 'Eve' and 'ecstasy': A report of five deaths associated
with the use of MDEA and MDMA. JAMA 257:1615-1617, 1987.
245. Greer G., Strassman R. J.: Information on "ecstasy." Am J Psychiatry 142:1391, 1985.
246. Lowinson J. H., Pedro R., Millman R. B. (eds): Substance Abuse: A Comprehensive Textbook, 2nd ed.
Baltimore, Williiams and Wilkins, 1992, pp. 280-289, 328-333.
247. McCann U. D., Ricaurte G. A.: MDMA ("Ecstasy") and panic disorder: Induction by a single dose.Biol
Psychiatry 32:950-953.
248. Newmeyer 1. A.: Some considerations on the prevalence of MDMA use. J Psychoactive Drugs 18:361-
362, 1986.
249. Pall anti S., Mazzi D.: MDMA (ecstasy) recipitation of panic disorder. Bioi Psychiatry 32:91-95,1992.
Marijuana
250. Benedikt R. A., Cristofaro P., Mendelson 1. H., et al: Effects of acute marijuana smoking in post-
menopausal women. Psychopharmacology 90: 14-17, 1986.
25 I. Bernstein 1. G.: Marijuana-new potential, new problems. Drug Ther 10:38-48, 1980.
252. Block 1. H., Gjerde P. E, Block, J. H.: Personality antecedents of depressive tendencies in 18-year-olds:
A prospective study. J Pers Soc PsychoI60:726-738, 1991.
253. Cohen S.: Cannabis: Impact on motivation. Part I. Drug Abuse Alcoholism Newsleft 9:1-3, 1980.
254. Cohen S.: Cannabis: Impact on motivation. Part II. Drug Abuse Alcoholism Newsleft 10:1-3, 1981.
255. Millman R. B., Beeder A. B.: Cannabis, in Galanter M., Kleber H. D. (eds): Textbook of Substance
Abuse Treatment. Washington, D.C., American Psychiatric Press, 1994, pp 91-110.
256. Harding T., Knight E: Marijuana-modified mania. Arch Gen Psychiatry 29:635-637, 1973.
257. Kupfer D. 1., Detre T., Koral 1., et al: A comment on the "amotivational syndrome" in marijuana
smokers. Am J Psychiatry 130:1319-1322, 1973.
258. Grinspoon L., Bakalar 1. B.: Marihuana, in Lowinson J. H., Pedro R., Millman R. B. (eds): Substance
Abuse: A Comprehensive Textbook. 2nd ed. Baltimore, Williams and Wilkins, 1992, pp 236-246.
259. Mendelson 1. H.: Marijuana, in Meltzer H. Y. (ed): Psychopharmacology: The Third Generation of
Progress. New York, Raven Press, 1987, pp 1565-1571.
260. Mendelson J. H., Mello N. K., Ellingboe 1., et al: Marijuana smoking suppresses luteinizing hormone in
women. J Pharmacol Exp Ther 237:862-886, 1986.
261. Millman R. B., Sbriglio R.: Patterns of use and psychopathology in chronic marijuana users. Psychiatr
c/in North Am 9:533-545, 1986.
262. Peterson R. C. (ed): Marijuana Research Findings: 1980. Washington, D.C., National Institute on Drug
Abuse, Research Monograph 31, 1980.
263. Schnoll S. H., Daghestani A. N.: Treatment of marijuana abuse. Psychiatr Ann 16:249-257, 1986.
264. Tashkin D. P., Shapiro B. 1., Lee Y. E.: Subacute effects of heavy marijuana smoking on pulmonary
function in healthy men. N Engl J Med 294:125-129, 1976.
265. Tennant E S.: The clinical syndrome of marijuana dependence. Psychiatr Ann 16:225-234, 1986.
266. Tennant E S., Jr., Groesbeck C. J.: Psychiatric effects of hashish. Arch Gen Psychiatry 27:133-136,
1972.
267. Wu T., Tashkin D. P., Djahed B., et al: Pulmonary hazards of smoking marijuana as compared with
tobacco. N Engl J Med 318:347-351, 1988.
268. Zuckerman B., Frank D. A., Hingson R., et al: Effects of maternal marijuana and cocaine use on fetal
growth. N Engl J Med 320:762-768, 1989.
IV
Special Topics
8
Geriatric Psychopharmacology
I. INTRODUCTION
Older persons, representing more than 10% of the population in the United States,
take all forms of medication more frequently than younger individuals. Estimates of
psychotropic drug use in the aged range from 7% to 92% in institutional settings and up
to 30% in medical settings. A greater incidence of polypharmacy accompanies in-
creased utilization of drugs. A recent survey in general hospitals showed that older
patients received an average of 5-12 medications per day. These statistics reflect many
factors: a longer life span accompanied by a greater incidence of chronic illnesses;
increased medical sophistication; and drug therapy as a substitute for nonbiological
interventions.
To care for the geriatric patient, the clinician must not only master the usual skills
necessary for effective drug administration but also understand the unique characteris-
tics of older individuals. In this age group, medical disorders commonly mimic or
contribute to psychiatric presentations. Physical illnesses, medical drugs, and signifi-
cant organ system alterations increase the number and severity of adverse reactions
from psychotropic agents. Age-related physiological changes as well as drug-drug
interactions may alter psychotropic drug pharmacokinetics; medications often take
longer to work, last longer in the body, and produce greater clinical effects per milli-
gram dosage. In addition, older individuals frequently have complicated psychosocial
problems that may decrease drug compliance (see Table 1).
Carl Salzman, M.D. • Department of Psychiatry, Harvard Medical School, Boston, Massachusetts
02115.
The Practitioner's Guide to Psychoactive Drugs (Fourth Edition), Gelenberg and Bassuk, eds.
Plenum Publishing Corporation, New York, 1998
367
368 CARL SALZMAN, M.D.
Problem Causes
Forgetting medication Organic, affective, or psychotic disorders may cause the patient to take
too much or too little medication.
Poor compliance Drugs are too expensive.
Adverse effects may be prohibitive.
Arthritic patients cannot open bottles (particularly with safety caps).
Elderly cannot get to the pharmacy (particularly in winter months).
Unusual reactions Unknown polypharmacy; patient taking over-the-counter medications or
other drugs without doctor's knowledge; more than one doctor is
prescribing drugs without knowledge of the other.
Unknown or underestimated use of alcohol, nicotine, or caffeine.
Undiagnosed illness that alters absorption, distribution, metabolism, or
elimination.
Undiagnosed decline in nutritional status, which also may alter drug
metabolism.
Confusion over medicine taking Pills that look alike (shape, color, size) may be interchanged; pills taken
concurrently but on different schedules (e.g., q.i.d. and t.i.d.) may be
confused.
Capricious noncompliance May take either excessive medicines or none at all because of irrational
ideas or struggles over control.
choactive medication. The most frequent and troublesome effects involve the central
nervous system (eNS) (sedation, confusion, and extrapyramidal effects) and heart
(orthostatic hypotension and alterations of heart rate, rhythm, and contractility) (see
Tables 2 and 3). In addition, drug combinations (Table 4), so prevalent in elderly
patients, increase the incidence and severity of both central and peripheral adverse
reactions (see Table 5).
Sedation Most psychotic drugs either produce Increased daytime sedation and
or enhance sedation of other napping; increased confusion and
medication irritability at night
Confusion Anticholinergic properties of cyclic Disorientation, visual hallucina-
antidepressants and antipsychotics tions, agitation, assultiveness,
Lithium loss of memory
Secondary to increased sedation
Orthostatic hypotension Results from cyclic antidepressants Falling when getting out of bed or
and anti psychotics out of a chair
Cardiac toxicity Primarily from cyclic antidepres- Increase in heart rate and in fre-
sants quency of irregular heart beat
rhythms; altered EKG patterns
Extrapyramidal symptoms Antipsychotics Akathisia-restlessness, which is con-
fused with agitation; akinesia-
decreased activity and interest,
which is confused with depression
depressant, diazepam (Valium® and others), or an antipsychotic agent. The eNS de-
pression (oversedation) that results from such drug combinations may lead to a lowered
mood, a sense of helplessness, an attitude of "resignation," and withdrawn or disor-
dered behavior. There may be a decline in self-care and social activities, as though the
older patient had given up and was quietly awaiting death.
Decreased central nervous system arousal level Sedation, apathy, withdrawal, depressed mood, dis-
inhibition, confusion
Peripheral anticholinergic blockade Dry mouth, constipation, atonic bladder, aggravation
of narrow-angle glaucoma and prostatic hyper-
trophy
Central anticholinergic blockade Confusion, disorientation, agitation, assaultiveness,
visual hallucinosis
a-Adrenergic blackade and central pressor Orthostatic hypotension
blockade
Quinidine effect, anticholinergic effect, de- Tachycardia, cardiac arrhythmia, heart block, in-
creased myocardial contractility creased PR interval and widening of QRS com-
plex, decreased inotropic effect, heart failure
Dopaminergic blockade Extrapyramidal symptoms, (7) tardive dyskinesia
TABLE 4. Pharmacokinetic Effects of Drug Interactions
Absorption
Antipsychotics Coffee and tea may cause precipitation in the stomach (although some
authorities now doubt this effect).
Benzodiazepines Aluminum-, magnesium-, or calcium-containing antacids delay hydrolysis;
kaopectate and milk of magnesia decrease absorption.
All drugs Anticholinergic agents [e.g., atropine (Antrocol® and others), scopolamine,
meperidine (Demerol® and others») decrease absorption.
Protein binding
Diazepam (Valium® Cyclic antidepressants increase binding of digoxin (Lanoxin® and others);
and others) thyroid-stimulating hormone (TSH) increases binding.
Excretion
All drugs Propranolol (lnderal® and others) increases serum levels from reduced renal
blood flow.
Cyclic antidepressants Tetracycline (Surmycin® and others) and spectinomycin
(Trobicin®) increase serum levels by reducing clearance; sodium
bicarbonate increases serum levels by increasing tubular reabsorption
secondary to alkalinization of the urine; ascorbic acid and ammonium
chloride decrease serum levels by a decrease in reabsorption secondary to
acidification of the urine.
Lithium (Eskalith® Ibuprofen (Motrin® and others) and indomethacin (Indocin® and others)
and others) (and perhaps all prostaglandin synthetase inhibitors) increase serum levels
by reducing renal clearance; some Na-losing diuretics increase serum
levels by increasing reabsorption; K + -sparing diuretics may decrease
serum levels by decreasing reabsorption; nonsteroidal anti-inflammatory
drugs may decrease increase lithium levels and levels of tricyclic
antidepressant hydroxy-metabolites.
Hepatic metabolism
Decrease metabolism of psychotropic drugs (raise blood level):
Isoniazid (lNH® and others)
Chloramphenicol (Chloromycetin® and others)
Methylphenidate (Ritalin® and others)
Disulfuram (Antabuse®)
Antipsychotics (raises levels of heterocyclics)
Cimetidine (Tagamet®) [affects chlordiazepoxide (Librium® and others), diazepam, prazepam (Centrax®
and others), and clorazepate (Tranxene® and others»)
Propranolol (lnderal® and others)
Norethindrone (Loestrin® and others)
Barbiturates (when toxic doses of heterocyclics are present)
Serotonin-selective reuptake inhibitors (raise levels of tricyclics and antipsychotic drugs; may increase
levels of triazolam and alprazolam benzodiazepines; may also raise levels of ketoconazole, terfenadine,
and astemizole).
Nefazodone may raise levels of ketoconazole, terfenadine, and astemizole.
Increase metabolism of psychotropic drugs (lower blood level):
Alcohol
Barbiturates
Nonbarbiturate sedative-hypnotics [except perhaps flurazepam (Dalmane® and others»)
Caffeine
Smoking
Carbamazepine (Tegretol® and others)
Phenytoin (Dilantin® and others)
Chloral hydrate (Noctec® and others)
Antiparkinson agents
Lithium [affects chlorpromazine (Thorazine® and others); also decreases chlorpromazine levels by delay-
ing gastric emptying, inhibiting absorption, and increasing renal excretion)
TABLE S. Unwanted Clinical Effects of Psychotropic-Psychotropic/Medical
Drug Combinations
2. Confusion
Confusion in the elderly most often results from one of the three etiologies: CNS
depression or oversedation (see above), CNS anticholinergic effects, or toxic effects of
lithium therapy. When significant sedation occurs, older individuals often develop
difficulties with concentration and recall.
Anticholinergic effects can lead to CNS confusion as well as peripheral toxicity.
Central anticholinergic syndromes commonly appear with the use of various anti-
depressants [e.g., amitriptyline, imipramine (Tofranil® and others), trimipramine (Sur-
montil®), doxepin, and nortriptyline (Pamelor® and others)] and some antipsychotics
[e.g., chlorpromazine, thioridazine, and chlorprothixene (Taractan®)]. More often,
however, confusion results from administration of multiple drugs with anticholinergic
properties. For instance, the combination of an antipsychotic, an antidepressant, and an
antiparkinson agent may cause significant toxicity, because the elderly have increased
CNS sensitivity to cholinergic blockade. The clinician should remember that even
medications with few anticholinergic effects can cause CNS impairment when given
alone to sensitive individuals, when given in higher doses, and when administered in
combination with other psychiatric or medical drugs.
The elderly patient with organic brain dysfunction is especially susceptible to
central anticholinergic toxicity. Antipsychotics and cyclic antidepressants may be pre-
scribed with narcotics [e.g., meperidine (Demerol® and others)], atropine-containing
drugs (which are often available as over-the-counter drugs), and antihistamines that can
produce either central and/or peripheral effects. In mild forms, this toxicity may appear
as disorientation, confusion, agitation, and irritability, particularly at night (i.e., the
"sundowner's syndrome"). In more severe cases, it may result in a toxic delirium
consisting of disorientation, hallucinations, other psychotic symptoms, and even as-
saultiveness. This condition can be life-threatening because of self-destructiveness,
combativeness, or poor self-care. Typically, in general hospital or nursing home set-
tings, such patients attempt to climb out of bed, mistake nurses for family members,
have frightening periods of amnesia, or may even hallucinate. A common medical
response is to add more sedating drugs, particularly antipsychotics; this, of course, only
further aggravates the condition or puts the patient to sleep.
Lithium alone commonly causes confusion and disorientation in the elderly. Con-
fusion secondary to neurotoxicity also may result from the combined use of an anti-
psychotic drug and lithium. Other symptoms include extrapyramidal features, seizures,
ataxia, hyperreflexia, and abnormal electroencephalograms (EEGs). If lithium and an
antipsychotic are used together, we recommend that the patient's status be monitored
with EEGs and that the plasma lithium concentrations be kept in the range of 0.2-0.6
meq/L, a plasma level range that is significantly lower than the recommended
therapeutic range for younger adults.
In addition, lithium-type toxicity, including polyuria, polydipsia, decreased free
thyroxine index, and decreased libido, has been noted, (primarily in younger adults)
despite normal serum levels of both drugs. In one case, substitution of carbamazepine
(Tegretol®) for phenytoin resulted in remission of symptoms. The use of anticonvul-
sants for elderly patients is discussed below.
8. GERIATRIC PSYCHOPHARMACOLOGY 373
3. Extrapyramidal
Older patients develop frequent and severe extrapyramidal symptoms. This may
result from altered pharmacokinetics of dopamine-blocking drugs leading to higher
receptor-site levels, decreased dopamine levels in the nigrostriatal pathways of the
elderly due to age-related decreased synthesis of dopamine and increased metabo-
lism, or age-related increased sensitivity of dopamine receptors in the central
nervous system. The most common effects include akathisia, parkinsonism, akinesia,
and the "Pisa syndrome." As in younger patients, akathisia may be misdiagnosed as
agitation; dosage elevations further increase the severity of the symptoms. Akinesia,
consisting of decreased speech and energy and masklike facies, may be mistaken for
depression. The "Pisa syndrome" is a form of dystonia in which the trunk is flexed to
one side. The clinician can treat these extrapyramidal symptoms (EPS) with antiparkin-
son drugs. However, he or she must weigh the added risk of anticholinergic effects (to
which the elderly are more sensitive) against the possible advantages and disadvan-
tages of switching to an antipsychotic drug with lower EPS-inducing properties.
Tardive dyskinesia is much more common in the elderly (especially women) and
in those with neuropathological disorders. No effective treatment exists, although the
symptoms may diminish in severity if medication is stopped shortly after the appear-
ance of symptoms (this occurs less often in older patients; see Chapter 4).
Parkinson's disease itself is more common in old age. Antipsychotic drugs may
worsen a patient's condition by inhibiting his or her response to L-dopa (Larodopa® and
others) (i.e., by blocking dopaminergic receptors in the CNS). When the older patient is
treated with L-dopa and antipsychotics, a change in dosage may be necessary.
Tardive dyskinesia is more likely in the older chronic patient because of the
greater cumulative neuroleptic treatment. It is also more likely in older patients because
changes in the aging CNS predispose them to developing tardive dyskinesia. Advanced
age is now known to be the most consistent and most likely predictor of the develop-
ment of tardive dyskinesia following use of neuroleptic drugs.
Since all these side effects may have more severe consequences in the older
patient, the clinician should avoid combining MAOIs and cyclics in the elderly unless
they do not respond to the usual therapies or have had a previous good response to
combination therapy.
Coadministration of ethchlorvynol (Placidyl®) and cyclic antidepressants may
cause a transient delirium and also should be avoided.
B. Cardiac
Unwanted cardiovascular effects from psychotropic drugs and from the combina-
tion of psychoactive and medical agents include changes in cardiac conduction, cardiac
rhythm, and blood pressure. Although less frequent than CNS sedation or anti-
cholinergic effects, cardiovascular changes are particularly hazardous in the elderly,
especially if they have preexisting heart disease.
Orthostatic hypotension in the elderly probably results from several causes: block-
ade of central vasomotor centers, negative inotropic effects (decreased cardiac contrac-
tility), and peripheral a-adrenergic blockade. This adverse reaction occurs most often
with the use of cyclic antidepressants (particularly amitriptyline and imipramine) and
antipsychotics (particularly chlorpromazine and thioridazine).
In clinical practice, transient hypotension may cause dizziness, falls and fractures,
myocardial infarcts, or strokes. In one study, about 40% of geriatric patients reported
dizziness and falling. In the elderly, hypotension may occur at night when patients get
up to urinate. Therefore, each patient should be checked for orthostatic changes and
told to change positions slowly. Sometimes, support stockings can also help.
Cyclic antidepressants decrease the antihypertensive properties of many drugs,
particularly guanethidine (Ismelin® and others), bethandine, and debrisoquine. Cyclics
also inhibit the effects of clonidine (Catapres®) but probably not reserpine (Serpasil®
and others) or a-methyldopa (Aldomet® and others). Antipsychotic agents, like cyclic
antidepressants, can also inhibit the effect of antihypertensive agents by blocking their
uptake. Low-potency antipsychotics may be more inhibitory than their high-potency
counterparts.
In the elderly patient, there is an increased risk of cardiac effects from antipsycho-
tics and cyclic antidepressants. Because of their anticholinergic properties and effects
on intracardiac conduction, these agents cause tachycardia. They also may produce an
increased incidence of premature ventricular contractions (although at some doses they
decrease), heart block, atrial and ventricular arrhythmias, heart failure, or worsening of
congestive heart failure [see Chapters 2 and 4 for electrocardiogram (EKG) changes].
Spontaneously occurring atrial and ventricular arrhythmias are commonly seen in
the geriatric population. Because cyclic antidepressants retard cardiac conduction, their
combination with drugs that prolong conduction, such as quinidine, procainamide
8. GERIATRIC PSYCHOPHARMACOLOGY 375
(Pronestyl® and others), or disopyramide (Norpace® and others), may result in toxic
cardiac effects. Digoxin (Lanoxin® and others), when combined with lithium, may
cause severe nodal bradycardia with slow atrial fibrillation or sinoatrial block. The
combination of thiazide diuretics (which deplete sodium and potassium), lithium, and
digoxin is particularly hazardous; sodium depletion leads to lithium retention, which, in
tum, worsens potassium depletion and increases digoxin toxicity. Permanent ventricu-
lar pacemakers are not affected by the tendency of cyclic antidepressants to interfere
with intraventricular conduction.
B. Clinical Use
1. Agitated Behavior and Psychotic Thinking
produces fewer and less severe extrapyramidal reactions in the elderly. The selection of
an appropriate antipsychotic medication, therefore, rests on a consideration of un-
wanted effects. For example, nonsedating antipsychotic drugs are more likely to pro-
duce extrapyramidal symptoms; conversely, sedating antipsychotics are less likely to
produce extrapyramidal symptoms.
In addition, the clinician can administer other antipsychotics if the older individu-
al has previously responded favorably and safely and if no new physical conditions
preclude their use. Antipsychotics should be started in low oral doses (e.g., 1-2 mg of a
high-potency drug or 10-25 mg of thioridazine) 2-3 times per day and increased
slowly.
Recent experience suggests that the atypical neuroleptics, cIozapine and ris-
peridone, are also useful for controlling severe agitation in elderly demented
nursing home residents. Although sedation, respiratory depression, and
agranulocytosis may complicate the use of cIozapine, and extrapyramidal symp-
toms and activation interfere with the therapeutic etTect of risperidone, both
drugs are etTective and provide a useful alternative to typical neuroleptic agents.
Doses of cIozapine that are helpful range from 25 mg/day to 200 mg/day although
the lowest etTective dose is recommended. Similarly, low doses of risperidone, such
as 0.5-1 mg/day may be therapeutic. Higher doses, up to 4 mg/day, may be
required and tolerated by some older patients.
b. Non-Neuroleptic Treatment of Acute Psychosis or Behavior Problems. A
number of non-neuroleptic treatments are now available to the clinician to control
severe disruptive behavior in elderly patients. Although research studies in this area are
virtually nonexistent, growing clinical experience suggests that the following may be
quite useful:
1. Trazodone (Desyrel® and others). This antidepressant drug in low doses (75-
150 mg) is remarkably effective in controlling severe agitation and disruptive
behavior. Side effects other than sedation are unusual at this low dose.
2. Buspirone (BuSpar®). This new antianxiety agent (5-15 mg/day) effectively
controls disruptive behavior. Experience with buspirone at this time is consid-
erably less than with other drugs, but reports suggest that toxicity is very low,
and the therapeutic benefit substantial.
3. Anticonvulsants. Carbamazepine and anticonvulsants, when used in very low
doses (e.g., carbamazepine, 100-200 mg/day), are sometimes effective in
controlling disruptive behavior that has not been managed by other medica-
tions.
4. Lithium. Low doses of lithium carbonate (e.g., 75-300 mg/day) given daily
on a maintenance basis have also been found effective in controlling or
preventing outbursts of yelling, agitation, and disruptive behavior. Low-dose
lithium must be used with extreme caution in older people (see below),
especially since symptoms of lithium CNS toxicity may be worsening of
confusion, agitation, and disruptive behavior.
5. J3-Blockers. The use of doses of J3-blocking drugs such as propranolol (10-
100 mg/day) may dramatically reduce agitation, wandering, and screaming
378 CARL SALZMAN, M.D.
behavior associated with dementing illness. In most casei, adverse effects are
minimal. The older patient must have an adequate cardiovascular evaluation
before l3-blocking drugs are used and must not have a disorder such as asthma
that would contraindicate the use of these drugs.
2. Dementia
Chronic brain syndromes may result from many etiologies. Dementias caused by
tumors or increased intracranial pressure may be reversible. However, the majority of
the elderly with dementia have progressive, diffuse loss of brain cells (Alzheimer's
disease accounts for 50-75% of cases) and ongoing deterioration of mental capacities
and functioning (see Section II.A.2). Less often, older patients may develop dementia
from arteriosclerosis. Multiple infarcts often cause a clinical picture in the elderly
marked by sudden onset and stepwise progression with a deteriorating mental status
and focal neurological signs. About one-third of demented patients have both Alzhei-
mer's dementia and multiple infarcts. Older patients also may present with "pseudode-
mentia." Although not a specific disorder, this term describes a decline in cognitive
processes associated with depression. The patient appears somewhat confused and
disoriented and has impaired recent memory. In severe cases, there are behavioral
disturbances and a deterioration of self-care. Adequate antidepressant treatment often
brings about a surprising restoration of thinking, memory, and behavior. Factors associ-
ated with this condition include:
Older patients with well-established organic dementia also may have significant
depressive symptoms that compromise thinking and behavior. Administering an anti-
depressant often improves the patient's depression, memory, orientation, and behavior.
Patients with senile dementia of the Alzheimer's type may have a selective deteri-
oration of acetylcholine neurons. Precursors of acetylcholine such as lecithin and
choline have been administered to these patients with equivocal results. One such
agent, tacrine (tetrahydroaminoacridane), acts by interfering with the enzymatic
destruction of acetylcholine. Research studies have demonstrated modest effec-
tiveness of this drug after several months of treatment. Tacrine, however, may
cause considerable hepatic toxicity. Consideration of tacrine treatment, therefore,
requires a careful appraisal of the patient's liver function status. Family members
should also be informed that in most cases, cognitive improvement, if it occurs at
all, may be relatively modest, although an occasional patient has been reported to
have significant improvement in cognitive status. Some late-life dementias have
been thought to result from decreased CNS oxygenation. Clinical and research experi-
ence with cerebral vasodilators as well as drugs that increase cerebral blood flow
[heparin (Panheprin® and others), papaverine (Cerebid® and others), cyclandelate (Cy-
clospasmol® and others)] and hyperbaric oxygen has also produced questionable re-
sults. The ergot alkaloid combination ergoloid mesylates (Hydergine® and others) has
been shown to have slightly positive effects in the mildly demented older patient, but
these results have not been striking. These drugs are no longer commonly used as
cognitive enhancers. Central nervous system stimulants have had no positive thera-
peutic effect on thinking or memory. Recent research has also focused on new protein
complexes, steroids, and metabolic agents such as piracetam, ACTH4 _ 1O, procaine
(Novocain® and others), and RNA, but the effects of these agents have been either
negative or equivocal.
3. Anxiety
compromised. Because these agents tend to accumulate, however, they should be given
only once a day or perhaps every other day. Unfortunately, the long-acting ben-
zodiazepines have several disadvantages for the elderly: they take longer to reach
steady-state plasma concentrations, and, should toxicity occur, elimination of these
drugs may take days to weeks, thus subjecting the older patient to prolonged unwanted
effects.
The other group of benzodiazepines has a much shorter duration of activity, has
no active metabolites, and tends not to accumulate. These agents include oxazepam
(Serax® and others), lorazepam (Ativan® and others), and alprazolam (Xanax®). They
must be administered throughout the day, since their duration of action is measured in
hours. Symptoms of rebound anxiety and insomnia are common several hours or days
after the last dose of a short-half-life benzodiazepine. These rebound symptoms may be
interpreted as a return of anxiety, prompting the continuation of drug treatment. Should
toxicity occur, however, the drugs will be more rapidly eliminated than the long-acting
benzodiazepines. Because the older patient is more sensitive to the benzodiazepines'
unwanted effects and to the problems of polypharmacy, the use of short-acting drugs
may be preferred.
As a general clinical recommendation, benzodiazepines should not be pre-
scribed to older patients on a regular maintenance basis for the treatment of anxiety or
of insomnia. Long-term benzodiazepine use, even at therapeutic doses, can be associ-
ated with subtle but progressive increase in confusion, disorientation, and forgetful-
ness. Research has demonstrated the reversibility of such benzodiazepine-induced
cognitive dysfunction, but its presence may be mistaken for an endogenous de-
menting disorder. Older patients may become dependent on benzodiazepines, espe-
cially short-half-life drugs, and have considerable difficulty in giving up their use.
When an older patient reacts adversely to benzodiazepines or exhibits persistent,
severe, or disorganizing anxiety, the practitioner should avoid other sedative-hypnotic
drugs. These medications, particularly the barbiturates, may cause respiratory depres-
sion, eNS impairment, and frequent paradoxical responses. Instead, the practitioner
might consider using low doses of an antipsychotic agent.
4. Insomnia
Old people commonly suffer from insomnia. The clinician should carefully evalu-
ate the patient, since sleep disorders frequently are secondary to various other condi-
tions (e.g., depression, anxiety states, physical pain, dyspnea, daytime napping, urinary
urgency, recent emotional trauma, or caffeine). If medication is appropriate, the practi-
tioner should only administer a hypnotic for a short period. Flurazepam, 15 mg at
bedtime, is effective; with longer use, this drug may accumulate and cause sundowning
and disinhibition. The shorter-acting benzodiazepines (e.g., oxazepam, 5-10 mg, or
lorazepam, 0.5-1.0 mg) offer other alternatives. Temazepam (Restoril® and others), 15
mg at bedtime, has strong hypnotic properties and a short half-life. Although there is
little experience with this drug in the elderly, it may have significant advantages.
Triazolam (Halcion®) is a benzodiazepine with an extremely short elimination
half-life (4 hr) that is also rapidly absorbed. For short-term treatment of sleep distur-
8. GERIATRIC PSYCHOPHARMACOLOGY 381
bance, triazolam has been found to help older patients. Doses must be kept low (0.125-
0.25 mg/night), because the older patient is extremely sensitive to higher doses of this
drug. Increased confusion, agitation, and forgetfulness have been observed at higher
doses. Because triazolam is so rapidly eliminated, rebound insomnia may develop 4 or
5 hr after the last dose is taken. It is not unusual for older patients who take triazolam at
bedtime to reawaken at 2:00 or 3:00 in the morning. It is difficult to distinguish
between such awakening on the basis of endogenous sleep disorder problems versus
triazolam rebound insomnia. Nevertheless, some clinicians prescribe a second dose of
triazolam for such awakenings. This practice leads to long-term higher-dose triazolam
use, dependence, and, in some older patients, progressive cognitive impairment. Triaz-
olam treatment of sleep disturbance, even at very low doses, is recommended only for
very brief periods of time; as a general clinical rule, this hypnotic should not be
routinely prescribed to older patients. Other short-half-life benzodiazepine hyp-
notics such as temazepam and estazolam may be useful, but the same warnings
given for triazolam apply to these drugs as well. A non-benzodiazepine drug,
zolpidem, mimics the pharmacologic activity of benzodiazepines and is an effec-
tive hypnotic. Its efficacy in older populations has been demonstrated in research
studies, but clinical experience is still insufficient to offer recommendations on its
use in this population.
5. Depression
Cyclic antidepressants have been the drugs of choice for the severely depressed
elderly patient. However, older patients are more susceptible to adverse effects than
younger individuals. In particular, orthostatic hypotension, cardiac irritability, and
anticholinergic toxicity occur more often in the elderly. Since no data suggest clinical
superiority of one cyclic antidepressant over another, the clinician should choose a
cyclic on the basis of its adverse reactions. In general, tertiary tricyclic amines such as
imipramine and amitriptyline produce more severe untoward effects than their de-
methylated secondary-amine metabolites, desipramine (Norpramin® and others) and
nortriptyline, respectively. Tertiary amines also last longer and, because of altered
hepatic metabolism with age, tend to accumulate more than secondary amines.
Therapeutic blood levels of nortriptyline or desipramine are not different in
older people compared with younger adults. However, these therapeutic blood levels
are usually achieved with doses that are substantially lower than those given to young-
er adult patients. For example, desipramine may be therapeutic at doses as low as 30-
75 mg/day; nortriptyline may be therapeutic at doses as low as 20-50 mg/day. It must
be remembered, however, that an occasional older patient may require a full therapeutic
dose to achieve full antidepressant effect. Therefore, although cyclic antidepressants
should be prescribed in lower doses for older patients, the actual dose necessary to treat
an individual elderly patient has to be based on therapeutic and toxic response.
In addition to desipramine and nortriptyline, trazodone, a sedating antidepressant
with no anticholinergic activity, is extremely useful as a hypnotic for the depressed
older patient who has difficulty falling or staying asleep.
Serotonin-selective reuptake inhibitors (SSRI), as a class of antidepressants, are
382 CARL SALZMAN, M.D.
The manic elderly patient presents a difficult therapeutic challenge. In acute manic
states, antipsychotic medication may be necessary to control hyperactivity, which can
rapidly lead to exhaustion. Most commonly, high-potency agents such as haloperidol,
are used, although recent experience suggests that the atypical neuroleptics
clozapine and risperidone may also be therapeutic.
Since lithium may produce serious adverse effects in the elderly, its use in the
treatment of mania requires special care. The elderly are sometimes more sensitive to
lithium toxicity, which is commonly manifested by a confusional state with memory
loss, disorientation, fear, and agitation. Since lithium excretion tends to be delayed in
older patients, the likelihood of such toxicity is increased, leading to an erroneous
diagnosis of dementia or delirium.
Lithium use, therefore, should be restricted to the treatment of the acute manic
episode and for prophylaxis only in patients with well-dbcumented recurrent bipolar
illness. Acute mania may be treated with an initial dose of 150 mg of lithium bj.d.;
dosages should be increased by 150-mg increments. Because older people are more
sensitive to the effects of lithium, plasma levels should be determined by the
clinical response of the patient; low plasma levels (0.2-0.4 mEq/L) may be effec-
384 CARL SALZMAN, M.D.
tive; levels above 0.8 mEq/L, which are recommended for younger adults, should
not be routine. If high levels are necessary, the clinician should carefully monitor the
patient's electrolytes, fluid status, and mental status. For prophylaxis, a blood level of
0.2-0.6 mEq/L is often adequate. Cardiac and renal function should be carefully
checked, and serum lithium levels, EKGs, and thyroid and kidney function should be
reviewed frequently.
The anticonvulsants carbamazepine and valproate have demonstrated efficacy in
the treatment and prophylaxis of bipolar affective disorder, regardless of the age of a
patient. These two compounds represent an alternative to lithium treatment, although
controversy still exists regarding the equivalent efficacy of the three compounds. Most
clinicians still favor lithium as the medication of first choice, especially because
plasma levels can be monitored. The anticonvulsants may be used to augment lithium's
effect or may be substituted for lithium in cases of toxicity or unusual sensitivity. Both
carbamazepine and valproate are potential hepatic toxins, so physical health clearance
and evaluation of hepatic function are essential prior to treatment. Carbamazepine also
may cause agranulocytosis, so pretreatment white blood cell counts are necessary and
should be followed early in treatment. Plasma levels of these drugs approximately
correlate with therapeutic effect (8-12 ng/mL for carbamazepine; 50-100 ng/mL for
valproate), although starting doses should be low and dosage increments gradual. Both
carbamazepine and valproate alter hepatic metabolic enzymes. Carbamazepine induces
hepatic enzymes and lowers blood levels of itself and other drugs; valproate inhibits
these enzymes and raises blood levels of other drugs. Caution must be exercised,
therefore, when these drugs are administered concurrently with other medica-
tions.
IV. CONCLUSION
Medicating the elderly presents many difficulties for the clinician. The marked
heterogeneity of this population and their special problems (particularly polypharmacy,
altered pharmacokinetics, and CNS sensitivity) must be confronted with each patient.
Often, serious adverse reactions develop before the clinician can give adequate thera-
peutic doses of medication. Therefore, the practitioner should carefully monitor each
drug and change the regimen if the initial trial fails. Even with skilled pharmacological
and non biological interventions, older patients may have limited responses. Despite
these problems, caregivers who provide comprehensive treatment, including psychoso-
cial, medical, and psychopharmacological approaches, often see gratifying results.
9
Pediatric Psychopharmacology
JANET WOZNIAK, M.D., JOSEPH
BIEDERMAN, M.D., THOMAS SPENCER,
M.D., and TIMOTHY WILENS, M.D.
I. INTRODUCTION
Janet Wozniak, M.D. • Pediatric Psychopharmacology Unit, Massachusetts General Hospital, Boston,
Massachusetts 02114; Department of Psychiatry, Harvard Medical School, Boston, Massachusetts
02115. Joseph Biederman, M.D. • Joint Program in Pediatric Psychopharmacology, Massachusetts
General Hospital, Boston, Massachusetts 02114; McLean Hospital, Belmont, Massachusetts 02178; Depart-
ment of Psychiatry, Harvard Medical School, Boston, Massachusetts 02115. Thomas Spencer,
M.D. • Pediatric Psychopharmacology Unit, Massachusetts General Hospital, Boston, Massachusetts
02114; Harvard Medical School, Boston, Massachusetts 02115. Timothy Wilens, M.D. • Pediatric
Psychopharmacology Unit, Massachusetts General Hospital, Boston, Massachusetts 02114; Harvard Medical
School, Boston, Massachusetts 02115.
The Practitioner's Guide to Psychoactive Drugs (Fourth Edition), Gelenberg and Bassuk, eds.
Plenum Publishing Corporation, New York, 1998
385
386 JANET WOZNIAK, M.D., ET AL
gram. Certain adverse effects can be anticipated based on known pharmacologic prop-
erties of the drug (e.g., the anticholinergic effects of tricyclic antidepressants), whereas
others, generally rare, are unexpected (idiosyncratic) and are difficult to anticipate.
Treatment should be started at the lowest possible dose, with frequent reevaluation
during the initial phase of treatment. Following a sufficient period of clinical stabiliza-
tion (6-12 months), it is prudent to reevaluate the need for continued medication
treatment. 5
Pharmacokinetic principles of psychotropic medications in children and adoles-
cents are similar to those for adults. As is true for adults, treatment in children must
take into account individual differences in binding, metabolism, and elimination of
medications. This interindividual variability generally plays a greater role in the psy-
chopharmacologic management of children than variability due to age; however, few
studies have addressed specific pharmacokinetic differences between juveniles and
adults. The few studies comparing half-lives of selected medications generally have
found shorter half-lives in children than in adults. While generally smaller in size than
adults, children are often more efficient at metabolizing psychotropics, requiring the
use of higher than expected weight-corrected doses to achieve desired effects. How-
ever, in the case of neuroleptic use in children, therapeutic effects are noted at lower
weight-corrected doses than in adults. Additionally, the psychopharmacologic treat-
ment of children and adolescents must take into special account the issues of drug
effects on growth and cognition. Until more is known, generalizations about dosing in
children must be extrapolated from the adult literature. 6 Table 1 summarizes, by drug
class and name, the main indications, weight-corrected dosing, and adverse effects of
the common psychotropics used in pediatric psychiatry.
Daily Daily
dose dosage Preparations
Drug Brand name Main indicationsb (mg/kg) schedule Adverse effects available
Stimulant
Dextroamphetamine Dexedrine® ADHD;ADHD + 0.3-\.5 Twice or Insomnia; decreased 5 mg; sustained
comorbidity/ three appetite; weight loss; release, 5, 10, and
developmental times depression; psychosis 15 mg; elixir
disabilities; adjunct (rare, with very high
therapy in refractory doses); increase in heart
depression rate and blood pressure
(mild); possible reduction
in growth velocity with
long-term use; withdrawal
effects and rebound
phenomena; rarely, tic
disorders
Methylphenidate Ritalin® As above 0.3-2.0 Twice or As above 5, 10, and 20 mg;
three sustained release,
times 20 mg
Magnesium Cylert® As above \.0-3.0 Once Same as other stimulants; 18.75,37.5, and 75
pemoline abnormal liver function mg
tests
Antidepressants
Tricyclics (TCAs) MDD; enuresis; ADHD; 2.0-5.Qc Once or Anticholingergic (dry mouth, 10, 25, and 50 mg
Imipramine Tofranil® tic disorder + OCD; twice constipation, blurred (hydrochloride);
anxiety disorders vision); weight loss; 75, 100, 125, and
cardiovascular: mild 150 mg (pamoate)
increase in diastolic blood
pressure and EKG
conduction parameters
with daily doses >
mg/kg; treatment requires
serum-level and EKG
monitorinQ.
Table I. (Continued)
Daily Daily
dose dosage Preparations
Drug Brand name Main indicationsb (mg/kg) schedule Adverse effects available
Desipramine Norpramin®, As above 2.0-5.()c Once or As above 10, 25, 50, 75, 100,
Pertofrane® twice and 150 mg
Nortriptyline Pamelor® As above 1.0-3.()c·d Once or As above 10, 25, 50, and 75
twice mg; elilxir
Clomipramine Anafranil® ()cO 2.0-3.0 Once or No known long-term side 25, 50, and 75 mg
twice effects; withdrawal effects
(severe gastrointestinal
symptoms, malaise) can
occur; overdoses can be
fatal
Nonselective
monoamine oxidase-
AlB inhibitors
(MAOIs)
Phenelzine Nardil® MOD: atypical 0.5-1.0 Twice or Severe dietary restrictions 15 mg
depression; ADHD; three (high-tyramine foods);
anxiety disorders times Hypertensive crisis with
dietetic transgression or
with certain drugs; weight
gain; drowsiness; changes
in blood pressure;
insomnia; liver toxicity
(remote)
Tranylcypromine Pamate® As above 0.5-1.0 Twice or As above IOmg
three
times
Selective monoamine
oxidase-B inhibitor
Selegiline Eldepryl® ADHD in Tourette's syn- 70-214 f.L/kg, Twice (A.M., Tyramine reaction (hyperten- 5 mg
drome :$ 15mg noon) sive crisis with dietetic
transgression or with cer-
tain drugs) may occur at
high doses; nausea; dizzi-
ness; changes in blood
pressure
Serotonin-selective
reuptake inhibitors
(SSRIs)
FIuoxetine Prozac® MDD; oeD; anxiety 0.5-1.0 Once (in the Irritability; insomnia; gastro- 10 and 20 mg; elixir
A.M.) intestinal symptoms; head-
aches
Sertraline Zoloft® As above 1.5-3.0 Once (in the As above 50 and 100 mg
A.M.)
Paroxetine Paxil® As above 0.25-0.70 Once (in the As above 20 and 30 mg
A.M.)
FIuvoxamine Luvox® As above 1.5-4.5 Once (in the As above 50 and 100 mg
A.M.)
Bupropion Wellbutrin® ADHD;MDD 3-6 Three times Irritability; insomnia; drug- 75 and 100 mg
induced seizures (in doses
> 6 mg/kg); contraindi-
cated in bulimics
Trazodone Desyrel® MDD; aggression; in- 2-5 Twice Orthostatic hypotension; pri- 50, 100, 150, and
somnia apism; sedation 300 mg
Venlafaxine Effexor® MDD; oeD; anxiety 1-3 Twice or Similar to those of SSRIs; 25, 37.5, 50, 75, and
three nausea; sleepiness; dizzi- 100 mg
times ness; sustained diastolic
hypertension
(continued)
Table 1. (Continued)
Daily Daily
dose dosage Preparations
Drug Brand name Main indicationsb (mg/kg) schedule Adverse effects available
Antipsychotics
Low-potency
phenothiazines
Chlorpromazine Thorazine® Psychosis; mania; as last 3-6 Once or Anticholinergenic (dry 10, 25, 50, 100, and
resource for aggressive twice mouth, constipation, 200 mg; elixir;
behavior, severe blurred vision); weight suppositories;
agitation, severe gain; extrapyramidal injectable
insomnia, and severe reactions (dystonia,
self-abuse; Tourette's rigidity, tremor, akathisia);
disorder drowsiness; risk for
tardive dyskinesia with
chronic administration;
withdrawal dyskinesia
Thioridazine Mellaril® As above 3-6 Once or As above 10, 15, 25, 50, 100,
twice 150, and 200 mg;
elixir
High-potency
phenothiazines
Auphenazine Prolixin® As above 0.1-0.5 Once or As above; anticholinergic 1,2.5,5, and 10 mg;
twice effects less common than elixir; injectable;
with low-potency long-acting 2, 4, 8,
phenothiazines; higher risk and 16 mg; elixir;
for extrappyramidal injectable
reactions
Perphenazine Trilafon® As above 0.1-0.5 Once or As above 2,4,8, and 16 mg;
twice elixir; injectable
Butyrophenone
Haloperidol Haldol® As above 0.1-0.5 Once or Same as for phenothiazines 0.5, I, 2, 5, 10, and
twice 20 mg; elixir;
injectable
Thioxanthene
Thiothixene Navane® As above 1-3 Once or Sane as for phenothiazines I, 2, 5, 10, 20 mg
twice elixir, injectable
Molindone Moban® As above 1-3 Once or Same as for phenothiazines; 5, 10, 25, 50, 100 mg
(indole derivative) twice lower risk of weight gain elixir
Pimozide Orap® Tourette's disorder < 0.2(:S 10 Once or Caridac arrhythmias (elon- 2mg
mg/day) twice gated QTc); seizures; ex-
trapyramidal reaction;
drowsiness; tardive dys-
kinesia; withdrawal dys-
kinesia
Atypical
anti psychotics
Clozapine Clozaril® Treatment-refractory psy- 3-5 Twice or Low incidence of extra- 25 and 100 mg
( dibenzodiazepine chosis three pyramidal adverse effects;
atypical antipsychotic times does not induce dystonia;
with serotonergic, low risk for tardive dys-
adrenergic, and kinesia; granulocytopenia/
histaminergic agranulocytosis (treatment
activity requires constant
mmonitoring of blood
count); higher risk of sei-
zures (dose related)
Risperidone Risperdal® Psychosis (positive and 0.2 Once or Low incidence of extra- I, 2, 3, and 4mg
[atypical negative symptoms) twice pyramidal adverse effects
antipsychotic with
dopaminergic (D2 )
and serotonergic
activity]
(continued)
Table 1. (Continued)
Daily Daily
dose dosage Preparations
Drug Brand name Main indicationsb (mg/kg) schedule Adverse effects available
Antimanic agents
Lithium carbonate Eskalith<B>, Bipolar disorder; mania; 1O-3()C Once or Polyuria; polydipsia; tremor; 300 mg; sustained
Lithane<B> prophylaxis of BPD; twice nausea; diarrhea; weight release, 300 and
MDD; hyperaggressive gain; drowsiness; skin 450 mg; elixir
behavior; adjunct abnormalities; possible
therapy in refractory effects on thyroid and
MDD renal functioning with
chronic administration;
therapy requires
monitoring of lithium
levels and thyroid and
renal tests; lithium toxicity
(level> 2mEq/L) can be
life-threatening
Lithium citrate Cibalith-S<B> As above 1O-3()C Once or As above elixir
twice
Carbamazepine Tegretol<B> Complex partial seizures; 10-20<' Twice Bone marrow suppression 100 mg chewable;
bipolar disorder; (baseline and close 200 mg; elixir
adjunct therapy in monitoring of blood
refractory MDD counts required);
dizziness; drowsiness;
rashes; nausea; liver
toxicity (uncommon)
Valproic acid Depakene<B> Absence seizures; bipolar 15-6()c Twice or Sedation; nausea; liver 250 mg; elixir
disorder; adjunct three toxicity (baseline and
therapy in refractory times close monitoring
MDD required); bone marrow
suppression
High-potency
benzodiazepines
Clonazepam Klonopin® Anxiety disorder; adjunct 0.01-0.04 Once or Drowsiness; disinhibition; 0.5, I, and 2 mg
therapy in refractory twice agitation; confusion; de-
psychosis; adjunct in pression; withdrawal reac-
mania; severe agita- tions; potential risk for
tion; Tourette's disor- abuse and dependence;
der; severe insomnia; less risk for rebound and
MDD + anxiety; withdrawal reactions
akathisia
Alprazolam Xanax® As above 0.02-0.06 Three times Same as for other ben- 0.25,0.5, I, and 2
(short-acting) zodiazepines; higher risk mg
for rebound and with-
drawal reactions
Lorazepam Ativan® Same as for other high- 0.04-0.09 Three times Same as for other ben- 0.5, 1, and 2 mg; in-
(short -acting) potency ben- zodiazepines; higher risk jectable
zodiazepines; tempor- for rebound and with-
ary use in severe drawal reactions
adjustment disorder
with anxious mood
Atypoical anxiolytic
Buspirone BuSpar® Anxiety disorders; ad- 0.2-0.6 Three times Drowsiness; disinhibition; no 5 and 10 mg
junct therapy in refrac- cross-tolerance with ben-
tory OCD zodiazpines
Noradrenergic agents
C10nidine Catapres® Tourette's disorder; 3-10 fLg/kg Twice or Sedation (very frequent); hy- 0.1, 0.2, and 0.3 mg;
(nonspecific a-2 ADHD; agression/self- three potension (rare); dry transdermal patch
agonist) abuse; severe agita- times mouth; confusion (with
tion; ?anxiety disor- high dose); depression; re-
ders; adjunct inmania bound hypertension; 10-
and schizophrenia; calized irritation with
withdrawal syndromes transdermal preparation
(continued)
Table 1. (Continued)
Daily Daily
dose dosage Preparations
Drug Brand name Main indications b (mg/kg) schedule Adverse effects available
Guanfacine Tenex® Tourette's disorder; 15-43IJ.g/kg Once or Same as for cIonidine; less I and 2 mg
(selective a-2A ADHD twice sedation and hypotension
agonist)
Propranolol Inderal® Tourette's disorder; 2-8 Twice Similar to those for 10, 20, 40, 60, and
(l3-blocker) ADHD; cIonidine; higher risk for 80 mg; long-
aggression/self-abuse; bradycardia and acting, 60, 80, 120,
severe agitation; hypotension (dose and 160 mg
akathisia dependent) and rebound
hypertension;
bronchospasm
(contraindicated in
asthmatics); rebound
hypertension on abrupt
withdrawal
Antihistamine, anti-
cholinergic agents
Diphenhydramine Benadryl® Sleep disorders; agita- 1-4e Three or Sedation; cognitive impair- 25 and 50 mg; elixir,
tion; acute dystonic re- four times ment; anticholinergic (dry injectable
actions mouth, constipation,
blurred vision); delirium
(rare)
Benztropine Cogentin® Sleep disorders; agita- 43-86 fLg/kg e Twice or Sane as for diphenhydramine 0.5, I, and 2 mg;
tion; extrapyramidal three elixir; injectable
reactions (dystonia, ri- times
gidity, tremor,
akathisia)
Other agents
Fenfluramine Pondimin® ?PDD 1-2 Twice or Anorexia; irritability; 20mg
three drowsiness; weight loss; in-
times somnia (when discontinued)
Naltrexone Trexan® Self-abuse 1-2 Once or Long-acting opiod antagonis- 50mg
twice tic; minimal adverse ef-
fects; hepatotoxicity (rare)
Desmopressin DDAVP® Enuresis 3-10 fLg/kg Intranasal Headache; nausea Nasal spray; rhinal
(0.1-0.2 once or tube; injectable
mL) twice
of these compounds, and they generally are not used as a first-line treatment in children
for ADHD.9
TCAs include the tertiary amines amitriptyline, imipramine, doxepin and tri-
mipramine and the secondary amines desipramine, nortriptyline, and protriptyline. The
mechanism of action of this class of drugs appears to be due to their blocking effects on
the reuptake of brain neurotransmitters, especially norepinephrine. Although these
agents have similar spectra of action, the various TCAs differ in their inhibitory effects
on transmitter reuptake, anticholinergic effects, and side-effect profiles.
Recent work suggests that children and adolescents metabolize TCAs more effi-
ciently than adults and that, despite higher weight-corrected dosing, children often have
lower plasma TCA levels than adults. lO Because a substantial interindividual vari-
ability in metabolism and elimination has been demonstrated in children, dose should
always be individualized. Treatment with a TCA should be initiated with a lO-mg or
25-mg dose, and the dose should be increased slowly every 4-5 days by 20-30%.
When an effective dose or daily dose of 3 mg/kg (1.5 mg/kg for nortriptyline) is
reached, steady-state serum levels and an electrocardiogram (EKG) should be obtained.
Subjective adverse effects such as dry mouth or dizziness tend to be unrelated to drug
serum levels. Typical dose ranges for the TCAs are 2.0-5.0 mg/kg (1.0-3.0 mg/kg for
nortriptyline).
Common short-term adverse effects of the TCAs include anticholinergic effects,
such as dry mouth, blurred vision, and constipation. There are no known negative
effects associated with chronic administration of these drugs. Gastrointestinal symp-
toms and vomiting may occur when these drugs are discontinued abruptly; thus, slow
tapering of these medications is recommended. Because the anticholinergic effects of
TCAs limit salivary flow, they may promote tooth decay in some children.
Evaluations of short- and long-term effects of therapeutic doses of TCAs on the
cardiovascular system in children have found TCAs to be generally well tolerated; only
minor EKG changes are associated with TCA treatment in daily oral doses as high as 5
mg/kg. TCA-induced EKG abnormalities (conduction defects) have been consistently
reported in children at doses higher than 3.5 mg/kg (1.0 mg/kg for nortriptyline).8
Although of unclear hemodynamic significance, the development of conduction de-
fects in children receiving TCA treatment merits closer EKG and clinical monitoring,
especially when relatively high doses of these medicines are used. In the context of
cardiac disease, conduction defects may have potentially more serious clinical implica-
tions. When there is doubt about the cardiovascular state of the patient, a more compre-
hensive cardiac evaluation is suggested prior to initiation of treatment with a TCA,
including 24-hr EKG (Holter monitor) and pediatric cardiology consultation to help
determine the risk-benefit ratio of such an intervention.
Several recent case reports of sudden death in children being treated with de-
sipramine have raised renewed concern about the potential cardiotoxic risk associated
with TCAs in the pediatric population. II Despite uncertainty and imprecise data, a
recent epidemiologic evaluation of this issue l2 suggested that the risk of desipramine-
associated sudden death may be slightly elevated but not much greater than the baseline
risk of sudden death in children not on medication. Treatment with a TCA should be
preceded by a baseline EKG, with serial EKGs at regular intervals throughout treat-
9. PEDIATRIC PSYCHOPHARMACOLOGY 399
ment. Because of the potential lethality of TCA overdose, parents should be advised to
carefully store a child's medication in a place inaccessible to the child and his or her
siblings.
In addition to TCAs, other antidepressants have been evaluated in the treatment of
ADHD. One study has reported that monoamine oxidase inhibitors (MAOIs) (Par-
nate®, Nardil®) are helpful in reducing ADHD symptoms. However, MAOI treatment
is problematic in children because of potential drug-drug interactions (including inter-
actions with over-the-counter cold medications) and strict tyramine-free diet restric-
tions required to avoid hypertensive crisis. Also, the novel dopaminergic antidepressant
bupropion (Wellbutrin®) was found to be superior to placebo and was well tolerated in
the treatment of ADHD children. However, bupropion has been reported to be associ-
ated with exacerbation of seizures and tic disorders.
In addition to stimulant medication and TCAs, the antihypertensive medications
clonidine (Catapres®) (see Section VII) and guanfacine (Tenex®) have been used for
ADHD. Guanfacine is an <x 2 -noradrenergic agonist that has been available as an FDA-
approved hypotensive agent for more than two decades. An initial open study in 13
ADHD children has indicated improvement in hyperactivity and attention. 14 Although
not well delineated, dosing in school-aged children should be started at 0.5 mg/day and
gradually increased as necessary to a maximum of 4 mg/day in two or three divided
doses. The available clinical data, although preliminary, indicate that guanfacine may
be longer acting, less sedating, and more effective for attentional problems than
clonidine. Like clonidine, the occurrence of hypotension in children treated with guan-
facine is not problematic. The most common adverse effects include transient som-
nolence, headaches, stomachaches, early-morning sedation, and enuresis.
While these common events are associated with psychosocial stressors and improve
spontaneously after the distress is removed, depressive disorders do not.
Major depression increases in prevalence with age. Major depression is estimated
to affect 0.3% of preschoolers, 1-2% of elementary-school-age children, and 5% of
adolescents. 16.17 Gender representation is equal until adolescence, when the adult
pattern emerges with approximately two-thirds of cases affecting females.
Research findings suggest that high levels of psychiatric comorbidity are the rule
and not the exception in juvenile major depression. 18 As many as 80% of juvenile
major depressive cases present with at least one, and often multiple, concurrent psychi-
atric illnesses. Puig-Antich l9 reported that 30% of boys fitting research diagnostic
criteria for major depressive disorder also met criteria for conduct disorder of the third
edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-III).
Substance abuse also commonly co-occurs with depression in older children and ado-
lescents. As in adults, the literature also supports comorbidity with anxiety disorders in
33% of depressed children. 20 Comorbidity of major depression with ADHD has been
reported in numerous studies but is still considered controversial. Findings reported by
Biederman et aJ.21 based on familial risk analyses support the hypothesis that ADHD
and at least some forms of major depression may share common familial vul-
nerabilities.
There is some evidence that juvenile mood disorders may be more refractory to
pharmacologic intervention than the adult mood disorders. Although double-blind
studies of antidepressants in the treatment of childhood depression have failed to
document superiority to placebo, this may be, in part, due to methodological issues.
Furthermore, open studies of children and adolescents do suggest that as many as half
of those treated will respond. Despite the failure of studies of juvenile depression to
document the efficacy of antidepressants, the watchful introduction of these drugs in
children with severe depression could be lifesaving. Because of the lack of demon-
strated efficacy for any class of antidepressant in juvenile depressive disorders,22,23
owing to safety concerns recent attention has been focused on the use of the serotonin-
selective reuptake inhibtors (SSRIs). In depressive adults, bipolarity, atypicality, and
psychotic symptoms are associated with poor response to treatment. Juvenile anti-
depressant nonresponders may benefit from treatment strategies that include (a) the use
of higher doses of the antidepressant if tolerated and if serum levels are not high; (b)
the use of a different class of medication; and (c) combined treatment approaches such
as use of two antidepressants of different classes, or additional low-dose lithium car-
bonate, stimulants, thyroid hormone (T3)' or antianxiety medications.
There are four main families of antidepressant medications: the tricyclic anti-
depressants (TCAs) (see Section II), the serotonin-selective reuptake inhibitors
(SSRIs), the monoamine oxidase inhibitors (MAOIs), and the atypical antidepressants
such as bupropion. The SSRIs include fluoxetine (Prozac®), paroxetine (Paxil®), ser-
traline (Zoloft®), and fluvoxamine (Luvox®). Because of their pharmacologic profile,
these medications have less anticholinergic, sedative, cardiovascular (blood pressure
and EKG changes), and weight-affecting adverse side effects than the TCAs. Unlike
TCAs, the SSRIs are structurally dissimilar to each other. Further, these agents vary in
their pharmacokinetics and side-effect profiles. Fluoxetine comes in capsules of 10 and
9. PEDIATRIC PSYCHOPHARMACOLOGY 401
20 mg and a liquid preparation (20 mg/5 mL). Suggested daily doses in pediatric
subjects are approximately 0.5-1.0 mg/kg. Fluoxetine and its active metabolite have a
long half-life of approximately 7-9 days. In contrast, paroxetine and sertraline have
medium half-lives of approximately 24 hr. Paroxetine comes in 20- and 30-mg cap-
sules, and the usual dose range is 0.25-0.70 mg/kg daily. Sertraline comes in 50- and
100-mg tablets, and the usual daily dose range is 1.5-3.0 mg/kg. Common adverse
effects of SSRIs include agitation, gastrointestinal symptoms, irritability, and insom-
nia. Fluoxetine and, to a lesser extent, paroxetine have been found to be potent inhibi-
tors of hepatic enzymes and thereby increase serum levels of TCAs and other similar
compounds. In contrast, sertraline is less likely than the other SSRIs to interact with
other drugs. Fluvoxamine appears similar to the other SSRIs with a relatively greater
incidence of sedation.
There are several other chemically unrelated antidepressants. Venlafaxine (Effex-
or®) possesses both SSRI and TCA properties. With venlafaxine, unlike the SSRIs,
there is a need to monitor for potential cardiac effects including diastolic hypertension.
Bupropion hydrochloride is a novel antidepressant of the aminoketone class, related to
the phenylisopropylamines but pharmacologically distinct from known antidepres-
sants. Although bupropion seems to have an indirect dopamine agonist effect, its
specific site or mechanism of action remains unknown. Bupropion is rapidly absorbed,
with peak plasma levels usually achieved after 2 hr and an average elimination half-life
of 14 hr (8-24 he). The usual dose range is 4.0-6.0 mg/kg per day in divided doses.
Side effects include irritability, anorexia, and insomnia and, rarely, edema, rashes, and
nocturia. Exacerbation of tic disorders also has been reported with bupropion. Bupro-
pion appears to have a somewhat higher (0.4%) rate of drug-induced seizures than the
other antidepressants, particularly when it is administered in daily doses higher than 6
mg/kg and in patients with preexisting seizure disorder or bulimia nervosa.
The MAOIs include hydrazine (phenelzine) and nonhydrazine (tranylcypromine)
compounds. MAOIs can be helpful in the treatment of atypical depressive disorders
with reverse endogenous features and depressive disorders with prominent anxiety
features. Daily doses should be carefully titrated based on response and adverse effects
and range from 0.5 to 1.0 mg/kg. MAOIs are generally not a good choice for use in
children owing to the dietetic and other restrictions imposed by the need to avoid
tyramine-containing foods (e.g., most cheeses), pressor amines (i.e., sympathomimetic
substances), or drug interactions (e.g., most cold medicines, amphetamines), which can
induce a hypertensive crisis. Short-term adverse effects include orthostatic hypoten-
sion, weight gain, drowsiness, and dizziness. Concurrent use of SSRIs is contraindi-
cated because of reports of potentially fatal hyperserotoninism.
v. ANXIETY DISORDERS
important antianxiety agents are the benzodiazepines. Other available compounds in-
clude sedative antihistamines (i.e., diphenhydramine, hydroxyzine, and promethazine)
and the atypical nonbenzodiazepine antianxiety drug buspirone (BuSpar®). Owing to
their pharmacologic properties (clinical effects), toxicologic properties (comfortable
margin of safety), and minimal pharmacokinetic interactions with other drugs, ben-
zodiazepines, along with antihistamines, are often used to treat the sometimes poorly
diagnosed symptoms of agitation and insomnia in children and adolescents. Although
potentially helpful in the treatment of anxiety disorders, antianxiety agents should be
used with caution in the management of developmentally disordered children because
they may produce disinhibition, which could result in increased restlessness and more
disturbed behavior.
Benzodiazepines are lipophilic and highly bound to plasma proteins and have
active metabolites that dominate their course of activity. Most benzodiazepines [e.g.,
diazepam (Valium®) and lorazepam (Ativan®)] are absorbed at an intermediate rate,
with peak plasma levels appearing 1-3 hr after ingestion. Benzodiazepines and related
sedative drugs can produce tolerance (and cross-tolerance with most other ben-
zodiazepines) and dependence [physiologic (addiction) and psychologic (habituation)].
The primary site of action of benzodiazepines in the central nervous system is at
GABA A receptors. The fact that benzodiazepines, barbiturates, and ethanol all have
related actions on a common receptor type explains their pharmacologic synergy and
cross-tolerance.
In recent years, the high-potency benzodiazepines alprazolam (Xanax®) and
clonazepam (K10nopin®) have received increasing attention as effective and safe treat-
ments for adult panic disorder with and without agoraphobia. Recent reports suggest
that children also can manifest adult-type anxiety disorders such as panic disorder and
agoraphobia which may also respond to high-potency benzodiazepines. Clonazepam, a
long-acting, high-potency benzodiazepine, is given 1-3 times a day, with a total daily
dose of 0.01-0.04 mg/kg. In contrast, the short-to-intermediate-acting, high-potency
benzodiazepines [alprazolam, lorazepam, and oxazepam (Serax®)] must be given 3 or 4
times a day because of the higher risk for rebound and withdrawal reactions. The total
daily dose of alprazolam is 0.02-0.06 mg/kg, that of lorazepam is 0.04-0.09 mg/kg,
and that of oxazepam is 0.5-1.7 mg/kg.
In general, the clinical toxicity of the benzodiazepines is low. The most commonly
encountered short-term adverse effects are sedation, drowsiness, and decreased mental
acuity. When given high doses, patients can become confused. In adults, ben-
zodiazepines have been reportedly associated with depressogenic adverse effects. With
the exception of the potential risk for tolerance and dependence (risk suspected but not
well studied in adults, risk unknown in children), there are no known long-term adverse
effects associated with benzodiazepines. Adverse withdrawal effects can occur, and
benzodiazepines should always be tapered slowly. Lorazepam and oxazepam do not
have active metabolites and do not tend to accumulate in tissue, making them prefer-
able for short-term symptomatic use. When long-term use is anticipated, longer-acting
benzodiazepines, such as clonazepam, are preferable.
Buspirone is a novel nonbenzodiazepine anxiolytic without anticonvulsant, seda-
tive, or muscle relaxant properties. It is thought that the anxiolytic effects of buspirone
may relate to reduction in serotonergic neurotransmission. Clinical experience with this
406 JANET WOZNIAK, M.D., ET AL.
drug suggests limited antianxiety efficacy. Daily dose is estimated to range from 0.3 to
0.6 mg/kg. It has been suggested that buspirone may be effective in the treatment of
aggressive behaviors in children with developmental disorders, including pervasive
developmental disorders.
Because of their relatively benign adverse-effects profile, high-potency ben-
zodiazepines are the drug of choice for uncomplicated juvenile anxiety disorders. In
some individuals, combined treatment (TCAs + benzodiazepines) may be necessary
for treatment of severe or comorbid anxiety conditions. Although the antidepressants
(see Sections II and III) have been found to be effective in the treatment of adult
anxiety disorders, controlled studies of TCAs for the treatment of childhood anxiety
disorders have had mixed results. While most of the available evidence suggests that
TCAs and MAOIs have antianxiety properties, the SSRIs appear to share these benefits
as well. Although TCAs and MAOIs can be effective in most anxiety disorders, they
are associated with substantial adverse effects, and the MAOIs require serious dietetic
restrictions. When TCAs are used, careful monitoring of serum levels and EKGs is
necessary. Imipramine has been reported to be superior to placebo in children with
school phobia.
OCD may need additional phannacotherapy with serotonergic blocking drugs such as
clomipramine, fluoxetine, or sertraline.
VIII. PSYCHOSIS
Both their in vitro receptor-binding properties and their in vivo effects confinn
that most antipsychotic drugs in current use block the binding of dopamine at its D2
receptor. Molindone, although structurally distinct, exhibits many similarities to other
neuroleptics, including dopamine receptor blockade, antipsychotic effects, and anti-
cholinergic adverse effects. Clozapine, an atypical antipsychotic agent, has a relatively
strong antagonistic interaction with central ex I-adrenergic, cholinergic, histaminic (H I)'
and serotonergic (S-HT 2 ) receptors. While clozapine exerts only weak antagonism of
D2 dopaminergic transmission, it has a high affinity for D4 receptors, with a greater
specificity for mesolimbic and mesocortical tracts. Thus, this compound lacks acute
extrapyramidal adverse effects and can be effective in treatment-resistant cases or in
children who develop tardive dyskinesia. Risperidone, recently approved, is another
novel antipsychotic medication which combines dopaminergic (D 2) antagonist and
serotonergic (S-HT 2) properties.
The usual oral dosage of antipsychotic drugs ranges between 3 and 6 mg/kg per
day for the low-potency phenothiazines and between 0.1 and O.S (up to 1.0) mg/kg per
day for the high-potency phenothiazines, butyrophenones, thioxanthenes, and indol one
derivatives. The dose of clozapine is 3.0-S.0 mg/kg, and that of risperidone is up to 0.2
mg/kg. Antipsychotic medications have a relatively long half-life, and, therefore, they
should be administered not more than twice daily. Most antipsychotic preparations are
available in either tablet or capsule fonn. In addition, at least one compound from each
class of anti psychotics is available in a liquid concentrate fonn. Several compounds,
including chlorpromazine, haloperidol, and fluphenazine, are available in an injectable
fonn for intramuscular administration.
Common short-tenn adverse effects of antipsychotic drugs are drowsiness, in-
creased appetite, and weight gain. Anticholinergic effects like dry mouth, nasal conges-
tion, and blurred vision are more commonly seen with the low-potency phenothiazines.
Extrapyramidal effects such as acute dystonia, akathisia (motor restlessness), and par-
kinsonism (bradykinesia, tremor, facial inexpressiveness) are more commonly seen
with the high-potency compounds (phenothiazines, butyrophenones, and thioxan-
thenes).
As in adults, the long-tenn administration of antipsychotic drugs in children and
adolescents may be associated with tardive dyskinesia, although children appear gener-
ally less vulnerable than adults to develop this adverse effect. Tardive dyskinesia
should be distinguished from the more common, generally benign, withdrawal dyski-
nesia associated with the abrupt cessation of antipsychotic drugs, which tends to
subside after several months of drug discontinuation. In children with mental retarda-
tion and pervasive developmental disorders, tardive dyskinesia should be differentiated
from the commonly occurring stereotypies. One approach to minimizing withdrawal
reactions is to taper antipsychotic drugs very slowly over several months. As there is no
treatment available for tardive dyskinesia, regular monitoring aimed at early detection
is an important standard of care.
The novel antipsychotic clozapine has less extrapyramidal side effects and a
reduced risk of tardive dyskinesia. In America as well as Europe, there has been
considerable experience with clozapine in adolescents. 36 An open 6-week trial of
clozapine in II adolescents (mean sixth week daily dose, 370 mg) resulted in marked
410 JANET WOZNIAK, M.D., ET AL
and clonazepam, can facilitate the management of the patient and may lead to the use
of lower doses of antipsychotics. In recent years, the syndrome of postpsychotic de-
pression has received increasing attention. Initial trials with antidepressant drugs added
to the antipsychotic treatment appear to be promising in relieving psychotic patients
from associated depression, thus fostering rehabilitation efforts. Postpsychotic depres-
sion should be distinguished from akinesia, which is an adverse extrapyramidal effect
that can respond to anti parkinsonian agents.
[3-Blockers and clonidine (see Section VII) have been increasingly reported to be
useful in controlling aggression and self-injurious behaviors in patients with develop-
mental disorders, which may include mental retardation, pervasive developmental dis-
orders (autistic and autistic-like disorders), and the specific developmental disorders
(learning disabilities). Considering the relatively low toxicologic profile of these drugs
compared with the antipsychotics, they are the preferred first treatment for the manage-
ment of these complications. Also, although not systematically investigated in children,
[3-blockers (propranolol) have been reported to be helpful in adults with ADHD and
severe impulsive and dyscontrol symptoms.
Among the [3-blockers, propranolol is the most commonly used. Propranolol is a
nonselective [3-adrenergic antagonist (i.e., it affects both [3\ and [32 receptors). Pro-
pranolol's effects are mediated through its ability to block [3-adrenergic receptors at
multiple sites in the body. It also crosses the blood-brain barrier, and this probably
accounts for some of its efficacy in psychiatric disorders but also contributes to con-
cerns regarding potential CNS toxicity. As much as a 20-fold interindividual variability
in hepatic elimination can occur. The half-life after chronic administration is about 4 hr.
The dose range used in pediatric disorders requiring propranolol is approximately 1-5
mg/kg per day. Short-term adverse effects of propranolol are usually not serious and
generally abate upon cessation of drug administration. Nausea, vomiting, constipation,
and mild diarrhea have been reported. Psychiatric side effects appear to be relatively
infrequent but can occur and include vivid dreams, depression, and hallucinations.
Allergic reactions manifested by rash, fever, and purpura are infrequent but have been
reported and warrant discontinuation of the drug. Propranolol can cause bradycardia
and hypotension as well as increased airway resistance and is contraindicated in asth-
matic and certain cardiac patients. There are no known long-term effects associated
with chronic administration of propranolol. Because abrupt cessation of this drug may
be associated with rebound hypertension, gradual tapering is recommended. Pro-
pranolol has also received considerable attention for its usefulness in drug-induced
akathisia and anxiety disorders, as well as the aggressive and self-abusive behavioral
disorders.
Fenfluramine has been evaluated as a possible treatment for the autistic syndrome,
in part because of the finding that the 30-40% of autistic patients have elevated blood
serotonin levels. Fenfluramine is a sympathomimetic amine related to the amphet-
412 JANET WOZNIAK, M.D., ET AL
X. SLEEP DISTURBANCES
Clonidine (see Section VII) has been described as a useful medication for sleep
disturbances in children with ADHD.38 The common side effect of sedation can be
utilized to clinical advantage by administering clonidine at bedtime to facilitate falling
asleep. Clonidine appears to be a safe and effective companion medication for the
stimulants, which commonly produce the side effect of insomnia.
XI. ENURESIS
Increasingly, multiple agents have been used in both clinical practice and research
for the treatment of child and adolescent psychiatric disorders. The need for combined
pharmacotherapy has arisen out of an emerging awareness of the high rates of comor-
bidity described in clinical and epidemiological studies of juvenile psychiatric disor-
ders. 39.40 In addition, controlled investigations of single agents have produced incon-
clusive findings. Enhanced response rates have been reported when traditional agents
are combined. 41 ,42 Previously, the law of parsimony dictated a single cause for each
symptom complex. This led to the use of large doses of individual agents for a given
9. PEDIATRIC PSYCHOPHARMACOLOGY 413
disorder, often resulting in intolerable adverse effects. In contrast, the use of combined
pharmacotherapy has permitted more targeted treatment and has allowed greater effi-
cacy to be achieved, often with lower doses and fewer adverse effects.43
XIII. CONCLUSIONS
REFERENCES
1. Institute of Medicine: Research on Children and Adolescents with Mental, Behavioral, and Developmen-
tal Disorders. Washington, D.C., National Institute of Mental Health, 1990.
2. FDA: Use of Approved Drugs of Unlabeled Indications. Psychopharma 18:5, 1982.
3. Gittelman-Klein R.: Diagnosis and drug treatment of childhood disorders, in Klein D. E, Gittelman R.,
Quitkin E, and Rifkin A. (eds): Diagnosis and Drug Treatment of Psychiatric Disorders: Adults and
Children. Baltimore, Williams & Wilkins, 1980, p 576.
4. Biederman J.: Psychopharmacology in children and adolescents, in Wiener 1. (eds): Comprehensive
Textbook of Child and Adolescent Psychiatry. Washington, D.C., American Psychiatric Press, 1991,
p 545.
5. Spencer T., Wilens T., Biederman J.: Psychotropic medication for children and adolescents. Child
Adolesc Psychiatr Clin North Am 1:97, 1995.
6. Clein P., Riddle M.: Pharmacokinetics in children and adolescents. Child Adolesc Psychiatr Clin North
Am 1:59, 1995.
7. Barkley R. A.: Attention Deficit Hyperactivity Disorder: A Handbookfor Diagnosis and Treatment. New
York, Guilford Press, 1990.
8. Biederman J., Baldessarini R., Wright V., et al: A double-blind placebo controlled study of desipramine
in the treatment of attention deficit disorder II: Serum drug levels and cardiovascular findings. J Am
Acad Child Adolesc Psychiatry 28:903, 1989.
414 JANET WOZNIAK, M.D., ET AL
9. Biederman 1.: Sudden death in children treated with a tricyclic antidepressant: A commentary. JAm
Acad Child Adolesc Psychiatry 30:495, 1991.
10. Wilens T. E., Biederman J., Baldessarini R J., et al: Developmental changes in serum concentrations of
desipramine and 2-hydroxydesipramine during treatment with desipramine. J Am Acad Child Adolesc
Psychiatry 31 :691, 1992.
II. Riddle M. A., Nelson J. C., Kleinman C. S., et al: Sudden death in children receiving norpramin: A
review of three reported cases and commentary. JAm Acad Child Adolesc Psychiatry 30:104, 1991.
12. Biederman 1., Thisted R., Greenhill L., et al: An epidemiological evaluation of the association between
tricyclic antidepressants and the risk for sudden death in five to 14 year-old children. J CUn Psych
56:87-93, 1995.
13. Casat C. D., Pleasants D. Z., Schroeder D. H., et al: Bupropion in children with attention deficit disorder.
Psychopharma 25:198, 1989.
14. Hunt R, Arnsten A., Asbell M.: An open trial of guanfacine in the treatment of attention-deficit
hyperactivity disorder. JAm Acad Child Adolesc Psychiatry 34:50, 1994.
15. Ryan N. D., Puig-Antich J., Ambrosini P., et al: The clinical picture of major depression in children and
adolescents. Arch Gen Psychiatry 44:854, 1987.
16. Anderson J. c., Williams S., McGee R., et al: DSM-III disorders in preadolescent children: Prevalence in
a large sample from the general population. Arch Gen Psychiatry 44:69, 1987.
17. Kashani J. H., Orvaschel H.: Anxiety disorders in mid-adolescence: A community sample. Am J
Psychiatry 145:960, 1988.
18. Biederman J., Faraone S., Mick E., et al: Psychiatric comorbidity among referred juveniles with major
depression: Fact or artifact. JAm Acad Child Adolesc Psychiatry 34:579, 1995.
19. Puig-Antich J.: Major depression and conduct disorder in prepuberty. JAm Acad Child Adolesc Psychia-
try 21: 118, 1982.
20. Kovacs M., Feinberg T., Crouse-Novak M., et al: Depressive disorders in childhood: II. A longitudinal
study of the risk for a subsequent major depression. Arch Gen Psychiatry 41 :643, 1984.
21. Biederman 1., Faraone S. V., Keenan K., et al: Further evidence for family-genetic risk factors in
attention deficit hyperactivity disorder (ADHD): Patterns of comorbidity in probands and relatives in
psychiatrically and pediatrically referred samples. Arch Gen Psychiatry 49:728, 1992.
22. Geller B., Cooper T. 8., Graham D. L., et al: Pharmacokinetically designed double-blind placebo-
controlled study of nortriptyline in 6-12 year olds with major depressive disorder: Outcome; nortiptyline
and hydroxy-nortriptyline plasma levels; EKG, BP and side effects measurements. JAm Acad Child
Adolesc Psychiatry 31 :34, 1992.
23. Puig-Antich 1., Perel J. M., Lupatkin W., et al: Imipramine in prepubertal major depressive disorders.
Arch Gen Psychiatry 44:81, 1987.
24. Wozniak J., Biederman J., Kiely K., et al: Mania-like symptoms suggestive of childhood onset bipolar
disorder in clinically referred children. JAm Acad Child Adolesc Psychiatry 34:867, 1995.
25. Alessi N., Naylor M., Ghaziuddin M., et al: Update on lithium carbonate therapy in children and
adolescents. JAm Acad Child Adolesc Psychiatry 33:291, 1994.
26. Leonard H. L., Swedo S. E., Rapoport J. L., et al: Treatment of obsessive compulsive disorder with
clomipramine and desipramine in children and adolescents: A double-blind crossover comparison. Arch
Gen Psychiatry 46:1088, 1989.
27. Rapoport 1., Swedo S., Leonard H.: Childhood obsessive compulsive disorder. J CUn Psychiatry 53:11,
1992.
28. Riddle M., Scahill L., King R, et al: Double-blind, crossover trial of fluoxetine and placebo in children
and adolescents with obsessive-compulsive disorder. J Am Acad Child Adolesc Psychiatry 31: 1062,
1992.
29. Geller D. A., Biederman 1., Reed E. D., et al: Similarities in response to fluoxetine in the treatment of
children and adolescents with obsessive-compulsive disorder. J Am Acad Child Adolesc Psychiatry
34:36, 1994.
30. Leckman J. E, Hardin M. T., Riddle M. A., et al: Clonidine treatment of Gilles de la Tourette's
syndrome. Arch Gen Psychiatry 48:324, 1991.
31. Riddle M. A., Hardin M. T., Cho S. C., et al: Desipramine treatment of boys with attention-deficit
9. PEDIATRIC PSYCHOPHARMACOLOGY 415
hyperactivity disorder and tics: Preliminary clinical experience. J Am Acad Child Adolesc Psychiatry
27:811, 1988.
32. Singer H. S., et al: The treatment of attention-deficit hyperactivity disorder in Tourette syndrome: A
double-blind placebo-controlled study with c10nidine and desipramine, Pediatrics 95:74-81, 1995.
33. Spencer T., Biederman J., Kerman K., et al: Desipramine in the treatment of children with Tic disorder or
Tourette's Syndrome and attention deficit hyperactivity disorder. JAm Acad Child Adolesc Psychiatry
32:354, 1993.
34. Jankovic J.: Deprenyl in attention deficit associated with Tourette's syndrome. Arch Neural 50:286,
1993.
35. Campbell M., Spencer E. K.: Psychopharmacology in child and adolescent psychiatry: A review of the
past five years. JAm Acad Child Adolesc Psychiatry 27:269, 1988.
36. Rapoport J.: Clozapine and child psychiatry. J Child Adolesc PsychopharmacoI4:1, 1994.
37. Frazier 1., Gordon C. T., McKenna K., et al: An open trial of c10zapine in II adolescents with childhood
onset schizophrenia. JAm Acad Child Adolesc Psychiatry 33:658, 1994.
38. Wilens T., Biederman 1., Spencer T.: Clonidine for sleep disturbances associated with attention-deficit
hyperactivity disorder. JAm Acad Child Adolesc Psychiatry 33:424, 1994.
39. Biederman J., Newcom J., Sprich S.: Comorbidity of attention deficit hyperactivity disorder with
conduct, depressive, anxiety, and other disorders. Am J Psychiatry 148:564, 1991.
40. Bird H. R., Canino G., Rubio-Stipec M., et al: Estimates of the prevalence of childhood maladjustment in
a community survey in Puerto Rico. Arch Gen Psychiatry 45: 1120, 1988.
41. Rapport M. D., Carlson G. A., Kelly K. L., et al: Methylphenidate and desipramine in hospitalized
children: I. Separate and combined effects on cognitive function. JAm Acad Child Adolesc Psychiatry
32:333, 1993.
42. Wilens T. E., Biederman J., Geist D. E., et al: Nortriptyline in the treatment of attention deficit
hyperactivity disorder: A chart review of 58 cases. JAm Acad Child Adolesc Psychiatry 32:343, 1993.
43. Wilens T., Spencer T., Biederman J., et al: Combined pharmacotherapy: An emerging trend in pediatric
psychopharmacology. JAm Acad Child Adolesc Psychiatry 34: 110, 1994.
10
Psychotropic Drug Use
in Pregnancy
I. INTRODUCTION
Treating psychiatrically ill female patients during pregnancy with psychotropic drugs
presents the physician with many clinical dilemmas. 1 The U.S. Food and Drug Admin-
istration (FDA) has not approved any psychotropic use during pregnancy though the
use of these medications is common. 2,3 Once considered a time of emotional well-
being for women, a growing literature suggests that pregnancy may not convey the
"protection" against psychiatric disorder once believed. 4 - 7 The marked prevalence of
psychiatric disorders during pregnancy and the growing evidence that psychiatric pa-
tients may be at high risk for relapse when medications are discontinued8- 11 make
treatment planning for psychiatrically ill pregnant women and for those who may wish
to conceive much more difficult. Balancing the risk of fetal exposure with the risk of
untreated psychiatric disorder places the clinician "between a teratologic rock and a
clinical hard place."12
Various reviews written during the last decade 12 - 16 describe the outcome of fetal
exposure to antipsychotics 17 - 2o antidepressants 21 ,22 mood stabilizers23 - 27 and ben-
zodiazepines. 28 - 3o While most studies suggest low teratogenic potential following in
Lee S. Cohen, M.D. • Perinatal and Reproductive Psychiatry Clinical Research Program, Clinical
Psychopharmacology Unit, Massachusetts General Hospital, Boston, Massachusetts 02114; Department of
Psychiatry, Harvard Medical School, Boston, Massachusetts 02114. Lori Altshuler, M.D. • UCLA
Neuropsychiatric Hospital and Center for Behavioral Sciences, Los Angeles, California 90024; VA
Medical Center, West Los Angeles, Los Angeles, California 90073. Vicki L. HeUer, M.D. • Brigham
and Women's Hospital, Department of Obstetrics and Gynecology, Boston, Massachusetts 02115.
Jerrold Rosenbaum, M.D. • Massachusetts General Hospital, Boston, Massachusetts 02114.
The Practitioner's Guide to Psychoactive Drugs (Fourth Edition), Gelenberg and Bassuk, eds.
Plenum Publishing Corporation, New York, 1998
417
418 LEE S. COHEN, M.D., ET AL
utero exposure to many of these agents, data are insufficient to determine absolute risk
of organ dysgenesis. Even less is known about the potential for long-term subtle
neurobehavioral sequelae associated with prenatal exposure to psychotropics. Thus,
clinicians must make decisions that are based on partially calculated risk. The purpose
of this chapter is to review the risks of prenatal exposure to psychotropics and to
provide disorder-based treatment guidelines for women who plan to conceive and for
psychotropic drug use during pregnancy.
A. Risks of Pharmacotherapy
Risks associated with prenatal exposure to psychotropics include (1) teratogen-
icity, (2) direct neonatal toxicity leading to a spectrum of perinatal syndromes, and (3)
"behavioral teratogenesis" or long-term behavioral sequelae. The effects of psycho-
tropics on the breast-feeding infant represent a fourth type of risk.
Teratogenic risk refers to the likelihood of gross organ malformation. Organ
malformation occurs during the first 12 weeks of gestation. 32 A drug is considered a
"teratogen" when prenatal exposure during this time results in an increased risk of
congenital malformations compared to the baseline incidence of congenital malforma-
tions in the absence of drug exposure. 33 The estimated baseline incidence of congenital
malformations in the United States is 3-4%.34 Toxic exposure prior to 2 weeks gesta-
tion is not typically associated with an increased risk for organ dysgenesis. Toxic injury
during this very early period of development (corresponding to the time prior to the
first missed menses) is more likely to result in a nonviable blighted ovum. 35 This may
have important implications for women maintained on psychotropic medications with
10. PSYCHOTROPIC DRUG USE IN PREGNANCY 419
regarding the use of psychotropic drugs during pregnancy. To date, the impact of
neuroendocrinologic dysregulation associated with psychiatric disorders 70,71 or the
precise impact of untreated psychiatric symptoms on the fetoplacental unit72 - 74 is
unclear, but it is of some concern.
A. Antipsychotics
Although an early case report describing limb malformations raised concerns
regarding first-trimester exposure to haloperidol,75 several studies have failed to dem-
onstrate increased teratogenic risk with respect to high- or low-potency neurolep-
tics. 17,19 In a large retrospective study, Rumeau-Rouguette et al. identified 315 cases of
first-trimester exposure to neuroleptics; 3.5% of these women gave birth to malformed
infants. 18 These investigators did report an increased risk of congenital malformations
in offspring exposed to phenothiazines with three-carbon aliphatic chains (e.g.,
chlorpromazine) but not in those exposed to phenothiazines with two-carbon side
chains (piperazine or piperidine derivatives). A recent metaanalysis of studies of prena-
tal exposure to antipsychotics supprts higher risk of congenital malformations follow-
ing first-trimester exposure to low-potency neuroleptics. 76
In another multicenter study by Slone and colleagues,17 describing 50,282 preg-
nant women including 1309 cases of phenothiazine exposure, researchers did not
document an increased rate of malformations in the offspring of the women exposed to
drug. These findings are consistent with those of Milkovich and Van den Berg,I9 who
failed to demonstrate an increased risk of malformations in the offspring of women
taking prochlorperazine during the first trimester to treat symptoms of nausea. 19 How-
ever, a report that subsequently reanalyzed the data from this study described a trend
toward increased numbers of anomalies in the offspring of women who received this
agent between the 6th and 10th week of gestation. 20
Whereas several investigators have failed to describe a significant increase in risk
associated with first-trimester exposure to phenothiazines, fewer studies have exam-
ined the effect of prenatal exposure to high-potency neuroleptics, such as haloperidol,
or atypical antipsychotics, such as clozapine. Two small retrospective studies have
found no association between fetal exposure to haloperidol and resultant congenital
deformities. 77 ,78 Similarly, in one case report of clozapine use during pregnancy, no
adverse effects were described with respect to mother or fetus. 79
Treatment-emergent adverse effects evident during the perinatal period have been
described in neonates exposed to neuroleptics during pregnancy. No systematic study
of the incidence of these perinatal syndromes has been conducted. However, case
reports of motor restlessness, tremor, difficulty with oral feeding, hypertonicity, dys-
tonic movements, and parkinsonian-like effects have all been described. 45 ,51,54,80
These symptoms have typically been of short duration although some have lasted up to
10 months, with ultimate resolution. 81
Data regarding potential long-term neurobehavioral sequelae associated with pre-
natal exposure to neuroleptics have been lacking. Data from animal studies suggest a
10. PSYCHOTROPIC DRUG USE IN PREGNANCY 421
B. Antidepressants
Although pregnancy has frequently been described as a time of emotional well-
being for women, suggesting "protection" from psychiatric disorders,90 recent research
suggests that as many as 10% of women may meet research diagnostic criteria (RDC)
for major or minor depression during pregnancy.6,91-93 In addition, patients with
histories of recurrent major depression appear to be at particularly high risk for relapse
when antidepressants are discontinued. 8 Given epidemiologic data that suggest a high
prevalence of mood disorders in women during childbearing years 94 and the absence of
data supporting a salutary effect of pregnancy on mood, women may present with
emergent or recrudescing symptoms of depression during pregnancy.
1. Tricyclic Antidepressants
Early anecdotal reports in the 1970s described a possible association between
first-trimester exposure to tricyclic antidepressants (TCAs) and limb malformations. 95
However, subsequent studies have not confirmed these findings. To date, 3 prospective
studies and over 10 retrospective studies have been published in which risk for
organ dysgenesis has been evaluated following first-trimester exposure to tri-
cyclics. 2I ,22,86,96-105 When these studies are pooled, an excess of 500,000 births have
been evaluated, and over 400 cases of first-trimester exposure to tricyclics have been
documented. No single study nor group of studies consistently supports an increased
risk of congenital malformations following first-trimester exposure to TCAs.
Various case reports describe perinatal syndromes following exposure to TCAs.
These have included TCA withdrawal syndromes with characteristic symptoms of
jitteriness, irritability, and seizures following exposure to these agents during labor and
delivery46-49 Symptoms of functional bowel obstruction and urinary retention, pre-
sumably secondary to anticholinergic effects of TCAs, have also been reported. 50,53
While several animal studies describe a spectrum of behavioral abnormalities during
the first 30 days of life following prenatal exposure to TCAs,45,106-lll no systemat-
ically derived data are available regarding long-term sequelae of such exposure in
humans. The significance of changes at the receptor level in animals exposed to TCAs
prenatally, including decreased adrenergic receptor binding and decreased density of
serotonin receptors, is unclear and remains to be studied.43,1 10,1 12
422 LEE S. COHEN, M.D., ET AL
affective disorders. Although amphetamine abuse has been associated with growth
retardation and premature delivery, 102 use of amphetamines for medical purposes does
not appear to increase risk for organ dysgenesis in most,86.118-120 but not all,121.122
studies. In one study, for example, no overall increase in congenital malformations was
noted following prenatal exposure to amphetamine, but an excess of oral clefts was
noted in the offspring of mothers who used the drug in the first trimester. 122
Data regarding long-term sequelae of in utero exposure to stimulants are lacking.
While symptoms of irritability, apnea, and jerking movements have been reported
during the neonatal period in children whose mothers used amphetamine, these find-
ings were specific to cases of amphetamine abuse and not to use of the stimulant for
medical reasons or for treating affective disorders.123.124
C. Electroconvulsive Therapy
The use of electroconvulsive therapy (ECT) during pregnancy dates back over 50
years. 125 Several clinical reports describe the safe use of ECT during pregnancy,
particularly in settings in which risk from psychiatric disorders (Le., impulsivity associ-
ated with mania, psychosis noted in psychotic depression) demanded expeditious treat-
ment. 126,127 Two recent reviews of ECT use during pregnancy describe the efficacy
and safety of the procedure.126-129 While one case of placental abruption following
ECT use in pregnancy has been described,l3° other reports note the safety of the
procedure, particularly in the setting of comprehensive treatment by a psychiatrist,
obstetrician, and anesthesiologist.127.131.132
D. Mood Stabilizers
1. Lithium
Women with bipolar disorder have typically been counseled either to defer preg-
nancy, because of the need for mood stabilizers, or to terminate pregnancies following
exposure of the fetus to drugs such as lithium carbonate. Concern regarding the extent
to which prenatal exposure to lithium increases risk for congenital, and specifically
cardiovascular, malformations dates back to the early 1970s and the first reports from
the International Register of Lithium Babies. 133.134 This voluntary physician reporting
system included women with histories of first-trimester exposure to lithium carbonate.
Though the reporting system had methodologic limitations, with an inherent bias
toward overreporting of adverse outcome, initial and subsequent reports from the
Register described increased rates of cardiovascular malformations, most notably Eb-
stein's anomaly.133-135 This anomaly is characterized by right ventricular hypoplasia
and congenital downward displacement of the tricuspid valve into the right ventricle.
Frequently, other cardiac anomalies are present as well. In one study, in which 36
children born with Ebstein's anomaly were followed, approximately half presented in
the first week of life with congestive heart failure and cyanosis. Half of the 36 children
died within days of presentation, though the others went on to have surgery for correct-
able lesions and had a good prognosis for long-term survival. 136
424 LEE S. COHEN, M.D., ET AL
Several recent epidemiologic studies suggest a more modest teratogenic risk asso-
ciated with first-trimester exposure to lithium than proposed by previous investiga-
tions. In two cohort studies, for example, investigators have described the outcome of
pregnancies in women who used lithium during at least the first trimester of pregnan-
cy.24,137 Both of these studies did not reveal the strong association between first-
trimester exposure to lithium and congenital malformations noted in the initial reports
from the Lithium Register. Four case control studies have complemented the cohort
studies,138-141 In these reports, maternal lithium treatment during pregnancy was
evaluated in children with Ebstein's anomaly compared to a group of controls. These
case control studies also failed to demonstrate a significant association between first
trimester use of lithium and Ebstein's anomaly as suggested in the initial Lithium
Register reports.
A recent pooled review of the two cohort studies and four case control studies that
have followed the sealing of the Lithium Register supports the possibility that first-
trimester exposure to lithium is associated with an increased risk of cardiovascular
malformations. 23 However, the risk has been revised, with estimates of Ebstein's
anomaly following first-trimester exposure to lithium ranging from 10 to 20 times that
noted in the general population. With a baseline risk for Ebstein's anomaly estimated at
1 in 20,000, the revised risk for this congenital malformation lies between 1 in 2000
(0.05%) and I in 1000 (0.1 %). In sum, although the relative risk for Ebstein's anomaly
may be somewhat increased, the absolute risk of having a child with this anomaly after
first-trimester exposure to lithium remains quite low.
Concerns regarding lithium exposure have extended beyond risk for organ dys-
genesis. Neonatal toxicity in offspring exposed to lithium around the time of labor and
delivery has also been reported. A "floppy baby" syndrome has been reported, charac-
terized by cyanosis and hypertonicity.142,143 A case of neonatal hypothyroidism and
nephrogenic diabetes insipidus has also been reported. 144 One recent naturalistic study
of bipolar women followed during pregnancy and the puerperium described a group of
women maintained on lithium during pregnancy and the postpartum period. No evi-
dence of neonatal toxicity was noted in newborns whose mothers were taking lithium
either during pregnancy or during labor and delivery.145
Limited data are available regarding long-term follow-up of children exposed to
lithium during pregnancy. In one study by Schou,146 a five-year follow-up investiga-
tion was performed on 60 children exposed to lithium during the second and third
trimesters of pregnancy and born without physical malformations. 146 Parents were
given questionnaires, and exposed children were compared with 57 nonexposed sib-
lings. No significant behavioral problems were noted in the exposed group, and the two
groups were not noted to be developmentally distinct.
2. Anticonvulsants
psychiatric illness. While it is often suggested that epileptic women give birth to
children with greater numbers of malformations as compared to the general population,
the incidence of malformations in offspring exposed to anticonvulsants in utero re-
mains higher than that in nonexposed controls even after controlling for the effects of
epilepsy. 147-149
First-trimester exposure to carbamazepine has been associated with spina bifida,
with an estimated risk of 1%.150 Valproic acid has similarly been associated with
increased risk of spina bifida, with estimates of risk ranging from 1 to 3%.151-153
Several studies report an increased risk with the use of mUltiple anticonvulsants and
possibly in the setting of higher maternal drug plasma levels. 154-156 One recent study
suggested an association between first-trimester exposure to anticonvulsants and orofa-
cial clefts,157 although no specific anticonvulsant appeared to be responsible for the
increased risk. Several investigators have described a syndrome including minor mal-
formations (i.e., rotated ears, flat nasal bridge, fingernail hypoplasia) in infants exposed
to anticonvulsants. 26 ,147,158,159 These minor malformations tended to disappear with
the process of maturation. 155
Data regarding the long-term impact of prenatal exposure to anticonvulsants have
been somewhat conflicting. Interpretation of neurobehavioral outcome following in
utero exposure to anticonvulsants has been confounded by difficulty in controlling for
factors including parental IQ, severity of epilepsy, seizures during pregnancy, and a
number of psychosocial influences. 148, 160 Nonetheless, one recent study has suggested
that prenatal exposure to carbamazepine is not associated with more neurobehavioral
abnormalities than noted in nonexposed controls. 161
E. Benzodiazepines
Concern regarding first-trimester exposure to benzodiazepines and risk for con-
genital malformations dates back over 20 years.162.163 Although early reports described
an increased risk of oral clefts following first-trimester exposure to drugs such as
diazepam, 162, 164 other studies do not support the association. 29 ,30 To date, 12 studies
have evaluated the relationship between prenatal exposure to benzodiazepines and risk
for congenital anomalies. 29 ,30,86,162,163,165-171 Interpretation of these studies is made
particularly difficult given the differences in sampling schemes and overall study
design employed in each investigation (L. Altshuler et aI, unpublished).
With respect to the risk of congenital malformations following first-trimester
exposure to benzodiazepines, three prospective studies appear to support the absence of
significant increased risk of congenital malformations following such exposure to
benzodiazepines. 167,170,171 Somewhat more controversial has been the issue of whether
first-trimester exposure to benzodiazepines increases risk for specific malformations,
such as cleft lip or cleft palate. While several recent studies do not support the associ a-
tion 29 ,30 a recent metaanalysis suggests that first-trimester exposure to benzodiazepines
may increase risk of oral clefts significantly76 as compared to the risk for oral clefts
noted in the general population of 6 in 10,000 (0.06%). These investigators describe a
risk for oral clefts of 0.7%, which, although significantly higher than that in nonex-
posed offspring, is still a relatively modest risk.
426 LEE S. COHEN, M.D., ET AL
Whereas several published studies describe the relative risk of in utero exposure to
benzodiazepines such as diazepam and alprazolam, data regarding risk of clonazepam
exposure are sparse. No controlled studies exist in which risk of teratogenesis has been
assessed following clonazepam exposure in humans. Limited data from animal studies
describe a smaller risk of teratogenesis associated with clonazepam exposure as com-
pared to that associated with exposure to other anticonvulsants such as carbamazepine,
phenobarbital, primidone, and ethosuximide. In Clonazepam was associated with a
fetal defect rate of 2.7%, which is two times higher than in control animals; the
significance of this is unclear. In one prospective study of 38 pregnant women with
panic disorder, clonazepam (0.5-4 mg per day) was used for varying periods. No
evidence of congenital malformations was noted. In addition, Apgar scores were uni-
formly high, and no children showed signs of neonatal withdrawal syndromes, hypo-
tonia, temperature dysregulation or other perinatal difficulties (C. Weinstock et al.,
unpublished data).
Neonatal toxicity has been reported following benzodiazepine exposure at or
about the time of delivery in several case reports 55 - 59,173-175 Symptoms have included
impaired temperature regulation, apnea, depressed APGAR scores, muscular hyper-
tonicity, and failure to feed. Studies of neurobehavioral function following prenatal
exposure to benzodiazepines have been conflicting. Several studies describe develop-
mental and motor delays, but these studies have frequently been accused of suffering
from severe ascertainment biases. 169 ,176 The balance of the data does not support a
significant effect on neurobehavioral function 167; however, the data are extremely
limited.
Treatment guidelines for psychotropic drug use during pregnancy vary as a func-
tion of the severity of the underlying disorder and the extent to which it threatens
maternal and fetal well-being. Relative risk of pharmacotherapeutic intervention must
be weighed against risk of psychiatric disorder. Risk can only be relatively quantified.
Ultimately, patients and physicians must work collaboratively as they weigh risks of
fetal exposure against the suffering and potential dangers of untreated psychiatric
illness. The inability to absolutely quantify risk does not preclude thoughtful treatment
planning and appropriate decisions regarding psychotropic drug use during pregnancy.
A. Psychosis
Psychosis during pregnancy is an obstetrical and medical emergency. Psychotic
symptoms may interfere with a woman's ability to obtain appropriate and necessary
prenatal care. 177 ,178 In addition, disorganization or psychotic denial may increase the
risk for impulsivity and dangerous behaviors.179 Whereas psychotic symptoms in
patients who have discontinued neuroleptics may reflect recrudescing illness, new
onset of psychosis during pregnancy demands an aggressive differential diagnostic
evaluation comparable to that performed in nongravid patients.
10. PSYCHOTROPIC DRUG USE IN PREGNANCY 427
Despite the abundant data that support the relative safety of high-potency neuro-
leptics during pregnancy, the use of these agents should be minimized if possible. Mild
intermittent symptoms of psychosis may be treated with "as needed" neuroleptics as
opposed to regular daily use, particularly in the first trimester. This clinical situation is
distinguished from those involving patients with chronic psychosis and particularly
those with histories of repeated decompensations either when tapered from or when
noncompliant with pharmacotherapy. Maintenance of antipsychotics in pregnant pa-
tients with chronic psychosis may limit overall prenatal drug exposure by obviating the
need for reintroduction of treatment with higher doses of drug following potential
relapse.
Although reports describe neonatal extrapyramidal symptoms in the setting of
neuroleptic exposure at or around the time of delivery, the rationale for discontinuing
antipsychotics just prior to labor and delivery is lacking. Because risk for decompensa-
tion is typically high in the setting of abrupt discontinuation of drug I 80 and postpartum
decompensation can occur within weeks of delivery,181.182 altering pharmacotherapy
just prior to delivery places the mother at risk for relapse.
B. Mood Disorders
1. Major Depression
Several studies suggest that women are not "protected" from mood disorders
during pregnancy. Investigators estimate the prevalence of depression during pregnan-
cy to be comparable to rates of depression in matched nongravid women. 6.93 With the
growing awareness that maternal mood and anxiety disorder may impact on child
development,183-185 the significance of depression during pregnancy and the puer-
perium becomes that much clearer.
The diagnosis of depression during pregnancy can be difficult since the distur-
bances in sleep and appetite as well as the symptoms of fatigue and change in libido do
not necessarily suggest an evolving mood disorder. Use of antidepressants during
pregnancy should be reserved for those cases in which neurovegetative symptoms
begin to interfere with the well-being of the mother.
The most appropriate treatment algorithm for management of major depression
during pregnancy depends on the severity of the underlying disorder. Women who
develop mild depressive symptoms during pregnancy may benefit from nonphar-
macologic treatments instead of somatotherapy. Prior to institution of antidepressants
in this population, cognitive therapy l86 or interpersonal therapy187 may be particularly
helpful. These nonpharmacologic strategies may also be appropriate for patients with
histories of mild to moderate episodes of depression who become pregnant or who are
on antidepressants and wish to conceive; tapering or discontinuing medication may be
appropriate for these patients. Cognitive-behavioral strategies may help eliminate the
need for medication.
Pharmacotherapeutic intervention is appropriate for pregnant patients who suffer
from moderate to severe symptoms of depression. These patients typically have symp-
toms including diminished oral intake, suicidality, or psychosis. Continuation of pre-
428 LEE S. COHEN, M.D., ET AL.
2. Bipolar Disorder
between episodes, thereby increasing the likelihood of lost productivity for women
suffering from chronic illness. 68
Women with histories of bipolar disorder who are maintained on lithium should
attend to family planning, as planned pregnancy affords the greatest number of options.
Reproductive risk counseling also involves the patient in the process of weighing the
relative risks of various treatment options. The decision to use lithium during pregnan-
cy depends on the severity of illness and the apparent need for prophylaxis with a mood
stabilizer. Women who have histories of single episodes of mania with long periods of
interepisode affective well-being or women with bipolar II disorder may be able to
taper and to discontinue lithium prior to trying to conceive. Taper of lithium should be
gradual to avoid relapse.
Women with bipolar illness who have histories of more frequent episodes of
mania and depression present a greater clinical challenge. Lithium discontinuation in
this population should proceed only after documentation of pregnancy, which needs to
be as early as possible. This clinical course minimizes fetal exposure and affords
prophylactic benefit while the patient is trying to conceive. With a growing number of
women choosing to conceive at an older age, maintenance of lithium therapy until early
documentation of pregnancy is prudent since the time required to conceive may be
more unpredictable than in younger patients.
For women with the most severe forms of bipolar disorder, maintenance of lithium
treatment before and during pregnancy is advisable. These patients are at highest risk
for clinical deterioration in the absence of antimanic prophylaxis. Assumption of a
modest increase in teratogenic risk associated with first-trimester exposure to lithium
seems particularly justified in this "high-risk" group. Patients with bipolar disorder
whose clinical status deteriorates require aggressive treatment. Treatment may include
hospitalization, use of adjunctive strategies such as treatment with neuroleptics and
clonazepam, or ECT. For patients who may have discontinued lithium, the drug should
be reintroduced regardless of the trimester.
Several earlier reports suggested the value of switching from therapy with lithium
to an alternative mood stabilizer, such as carbamazepine or valproic acid, during
pregnancy.12.25 These recommendations follow from concerns about the risk of first-
trimester exposure to lithium. Recent studies, however, describe higher rates of malfor-
mations following prenatal exposure to carbamazepine than previously thought. 26
Rates of neural tube defects following in utero exposure to valproic acid have also been
noted to be as high as 5%.153 These estimates are high compared to the revised risk
estimates for first-trimester exposure to lithium. Moreover, with fewer data supporting
comparable efficacy of these agents as mood stabilizers, it is hard to justify their use
during pregnancy. Their use may be appropriate in very select situations such as in
rapid-cycling bipolar patients who have not responded to lithium but who respond to
valproic acid. Given data from at least one study that suggest that administration of
folate to women during pregnancy reduces the risk of neural tube defects in off-
spring,197 women who remain on valproic acid or carbamazepine during pregnancy
should receive prophylactic folate.
Women who use lithium during the first trimester of pregnancy should be coun-
430 LEE S. COHEN, M.D., ET AL
seled about the slightly increased relative risk of congenital malformations associated
with such exposure, although reaasurance regarding the low absolute risk associated
with such exposure should be offered. 23 Fetal cardiac ultrasonography may be very
reassuring to patients with histories of first-trimester exposure to lithium and can be
performed at 16-18 weeks of gestation. For those patients maintained on lithium
during pregnancy, the drug may be tapered by 25-30% just prior to delivery.l2 This
reduces the likelihood of toxicity for the mother, given the rapid fluid shifts characteris-
tic of the puerperium. However, lithium should not be discontinued entirely as the
postpartum period appears to be a time of heightened risk for bipolar women. A body
of evidence consistently estimates risk for puerperal decompensation in bipolar women
at rates of 30_50%.192.195 Given this heightened risk for postpartum worsening of
mood, several investigators have evaluated the extent to which postpartum prophylaxis
with lithium attenuates such risk.l45.198 Results of this work point to significant reduc-
tion in rates of relapse in women who receive lithium during the first 48 hours postpar-
tum compared with women who do not receive postpartum lithium therapy. In addition,
anecdotal reports of neonatal toxicity including lethargy, and hypotonia,l44 have not
been seen consistently in follow-up studies of children whose mothers have received
prophylactic lithium.
c. Anxiety Disorders
1. Panic Disorder
Results from a number of studies suggest that panic disorder has a chronic and
recurrent course. 69 Some data suggest that certain patients can successfully taper and
discontinue anti-panic medication for a period of time during the course of their
illness.l 99 While early anecdotal and retrospective reports described a "protective"
effect of pregnancy on symptoms of panic disorder,2oo-203 several recent studies4,5
describe persistence or worsening of panic symptoms during pregnancy. Apparently,
pregnancy may ameliorate symptoms of panic in some patients,203-206 affording them
a chance to discontinue medication. Others patients may note persistence or exacerba-
tion of symptoms requiring anti panic pharmacotherapy. Factors that distinguish these
two groups are unclear.
Patients with panic disorder who are maintained on anti panic medications and
who wish to conceive should be advised to taper these drugs slowly. Adjunctive
cognitive-behavioral therapy may be of some benefit in helping patients discontinue
anti panic agents and may increase the period of time prior to relapse. 207 If taper is
unsuccessful, and symptoms recur, reinstitution of TeAs or benzodiazepines may be
required. Given the first-trimester risk of benzodiazepine-associated oral cleft, the use
of TeAs is preferable. However, the extent to which many patients with panic disorder
do not respond to TeAs alone makes the use of benzodiazepines a reasonable alterna-
tive. The use of newer antidepressants such as the SSRIs also represents a viable
treatment alternative. Nonetheless, prescription of TeAs in this population is still
preferable, given the longer history of safe use of these agents during pregnancy.
Some patients may inadvertently conceive on antipanic drugs and present for
10. PSYCHOTROPIC DRUG USE IN PREGNANCY 431
2. Obsessive-Compulsive Disorder
Several studies have documented an association between pregnancy and onset of
obsessive-compulsive disorder (OCD).208-210 In one study,210 52% of the women
developed the onset of OCD during their first pregnancy. However, other studies have
not confirmed this finding.211 Currently, there is no biochemical explanation for these
high rates of association of OCD with pregnancy-related life events. Behavioral tech-
niques (cognitive therapy) for OCD during the first trimester should be considered as
an alternative to medication. For OCD, tricyclic antidepressants and fluoxetine repre-
sent a reasonable pharmacologic approach during or particularly after the first trimes-
ter. Alternatively, clomipramine can be used if previously successful. Clomipramine is
not considered teratogenic; however, it may aggravate orthostatic hypotension, and
thus its use may be limited in pregnancy. Clomipramine is also best avoided at the time
of labor and delivery as there are is at least anecdotal data describing neonatal seizures
following clomipramine exposure. 48
V. BREAST FEEDING
VI. CONCLUSION
atric disorder. While concern regarding the effects of prenatal exposure to psycho-
tropics has kindled appropriate vigilance, data regarding the effect of discontinuing
drug on the well-being of patients with psychiatric illness indicate significant mor-
bidity. Thus, women who are maintained on psychotropics and become pregnant, as
well as women with the new onset of psychiatric symptoms during pregnancy, should
be carefully reassessed.
Data for some phenothiazines, lithium, benzodiazepines, and the anticonvulsants
valproic acid and carbamazepine suggest an increase in the relative risk of developing a
congenital malformation in the children of mothers exposed to these agents during the
first trimester of pregnancy. However, for some patients, the decision to accept a
modest increase in teratogenetic risk may be appropriate, given the need to ensure
stable maternal mental health during pregnancy. Data supporting safety of the TCAs
and SSRIs are reassuring, although the number of exposures to these drugs during
pregnancy remains relatively small.
Information regarding the potential for long-term behavioral changes following
prenatal exposure to psychotropics is sparse. Animal studies suggest changes in brain
receptor number and function in the offspring of mothers exposed in utero to various
psychotropic drugs. Large epidemiologic follow-up studies of children exposed to
medication in utero may help clinicians to weigh more completely the relative risk of
psychotropic drug use during pregnancy. Given the prevalence of psychotropic use
during pregnancy, it is critical that the clinician have a prepared approach to adminis-
tration of these agents. Coordinated care among patient, husband or partner, obstetri-
cian, and psychiatrist is essential, as is careful medical record documentation. Psycho-
tropics are generally used in pregnancy without discernible adverse consequences.
Pending good controlled prospective data on the impact of drugs on fetal and later
development, the clinician will continue to have to act in a state of potential uncertain-
ty, weighing partially calculated risks to manage individual clinical dilemmas.
REFERENCES
1. Cohen L. S., Rosenbaum 1. F. and Heller V. H.: Psychotropic drug use in pregnancy, in Ge1enberg A. 1.,
Bassuk E. L., Schoonover S. C. (eds): The Practioner's Guide to Psychoactive Drugs, 3rd ed. New
York, Plenum Medical Book Company, 1991.
2. Doering 1. C., Stewart R. B.: The extent and character of drug consumption during pregnancy. JAMA
239:843-846, 1978.
3. Kasilo 0., Romero M., Bonati M., Tognoni G.: Information on drug use in pregnancy from the
Viewpoint Regional Drug Information Center. Eur J Clin Pharmacol 35:447-453, 1988.
4. Cohen L. S., Sichel D. A., Dimmock 1. A., Rosenbaum 1. F.: Impact of pregnancy on panic disorder: A
case series. J Clin Psychiatry 55:284-288, 1944.
5. Northcott C. 1., Stein M. B.: Panic disorder in pregnancy. J Clin Psychiatry 55:539-542, 1994.
6. O'Hara M. w.: Social support, life events, and depression during pregnancy and the pueperium. Arch
Gen Psychiatry 43:569-573, 1986.
7. Frank E., Kupfer D. J., Jacob M., Blumenthal S. J., Jarrett D. B.: Pregnancy related affective episodes
among women with recurrent depression. Am J Psychiatry 144:288-293, 1987.
8. Kupfer D., Frank E., Perel J., Comes C., Mallinger A., Thase M.,McEachran A., Grochocinski V.:
Five-year outcome for maintenance therapies in recurrent depression. Arch ofGen Psychiatry 49:769-
773, 1992.
10. PSYCHOTROPIC DRUG USE IN PREGNANCY 433
9. Suppes T., Baldessarini R. 1., Faedda G. L., Tohen M.: Risk of recurrence following discontinuation of
lithium treatment in bipolar disorder. Arch Gen Psychiatry 48: 1082-1088, 1991.
10. Roy-Byrne P. P., Dager S. R., Cowley D. S., Vitaliano P., Dunner D. L.: Relapse and rebound following
discontinuation of benzodiazepine treatment of panic attacks: Alprazolam versus diazepam. Am 1
Psychiatry 146:860-865, 1989.
11. Dencker S.J., MaIm U., Lepp M.: Schizophrenic relapse after drug withdrawal is predictable. Acta
Psychiatr Scand 73: 181-185, 1986.
12. Cohen L. S., Heller V. L., Rosenbaum J. E: Treatment guidelines for psychotropic drug use in
pregnancy. Psychosomatics 30:25-33, 1989.
13. Calabrese J. R., Gulledge A. D.: Psychotropics during pregnancy and lactation: A review. Psychoso-
matics 26:413-426, 1985.
14. Miller L.: Clinical strategies for the use of psychtropic drugs during pregnancy. Psychiatr Med 9:275-
98, 1991.
15. Mortola J.: The use of psychotropic agents in pregnancy and lactation. Psychiatr Clin North Am 12:69-
87, 1989.
16. Robinson G. E., Stewart D. E., Flak E.: The rational use of psychotropic drugs in pregnancy and
postpartum. Can 1 Psychiatry 31: 183-190, 1986.
17. Slone D., Siskind v., Heinonen O. P., Monson R R, Kaufman D. W, Shapiro S.: Antenatal exposure to
the phenothiazines in relation to congenital malformations, perinatal mortality rate, birth weight, and
intelligence quotient score. Am Obstet Gynecol 128:486-488, 1977.
18. Rumeau-Rouguette C., Goujard J., Huel G.: Possible teratogenic effect of phenothiazines in human
beings. Teratology 15:57-64, 1977.
19. Milkovich L., Van den Berg B. 1.: An evaluation of the teratogenicity of certain anti nauseant drugs. Am
1 Obstet Gynecol 125:244-248, 1976.
20. Edlund M. J., Craig T. J.: Antipsychotic drug use and birth defects: An epidemiologic reassessment.
Compr Psychiatry 25:32-37, 1984.
21. Kuenssberg E. v., Knox J. D.: Imipramine in pregnancy. Br Med 1 2:292, 1972.
22. Idanpaan-Heikkila 1., Saxen L.: Possible teratogenicity of imipramine-chloropyramine. Lancet 2:282-
284, 1973.
23. Cohen L. S., Friedman 1. M., Jefferson 1. W, Johnson E. M., Weiner M. L.: A reevaluation of risk of in
utero exposure to lithium. lAMA 271: 146-150, 1994.
24. Jacobson S. J., Jones K., Johnson K., Ceolin L., Kaur P., Sahn D., Donnenfeld A. E., Rieder M., Santelli
R, Smythe 1., Pastuszak A., Einarson T., Koren G.: Prospective multicentre study of pregnancy
outcome after lithium exposure during first trimester. Lancet 339:530-533, 1992.
25. Markovitz P. J., Calabrese J. R: Use of anticonvulsants for manic depression in pregnancy (letter).
Psychosomatics 31: 118, 1990.
26. Jones K. L., Lacro R v., Johnson K. A., Adams J.: Pattern of malformations in the children of women
treated with carbamazepine during pregnancy. N Engl 1 Med 320:1661-1666,1989.
27. Lindhout D., Schmidt D.: In utero exposure to valproate and neural defects. Lancet 1:329-393. 1986.
28. Czeizel A., Lendvay A.: In-utero exposure to benzodiazepines (letter). Lancet 1:628, 1987.
29. Rosenberg L., Mitchell A. A., Parsells 1. L., Pashayan H., Louik C., Shapiro S.: Lack of relation of oral
clefts to diazepam use during pregnancy. N Engl 1 Med 309:1282-1285,1983.
30. Shiono P. H., Mills I. L.: Oral clefts and diazepam use during pregnancy (letter). N Engl 1 Med
311:919-920, 1984.
31. Coverdale J. H., Aruffo J. A.: Family planning needs of female chronic psychiatric outpatients. Am 1
Psychiatry 146:1489-1491, 1989.
32. Dicke 1. M.: Teratology: Principles and practice. Med Clin North Am 73:567-581, 1989.
33. American Medical Association: Drug interactions and adverse drug reactions, in AMA Drug Evalua-
tion. Chicago, American Medical Association, 1983.
34. Fabro S. E.: Clinical Obstetrics. New York, John Wiley, 1987.
35. Langman J.: Human development-normal and abnormals, in Langman J. (ed): Medical Embryology.
Baltimore, Williams & Wilkins, 1985.
36. Vernadakis A., Parker K.: Drugs and the developing central nervous system. Pharmacol Ther 11 :593-
647, 1980.
434 LEE S. COHEN, M.D., ET AL.
37. Vorhees C" Brunner R., Butcher R.: Psychotropic drugs as behavioral teratogens. Science 205:1220-
1225, 1979.
38. Coyle I., Wayner M., Singer G.: Behavioral teratogenesis: A critical evaluation. Pharmacol Biochem
Behav 4:191-200, 1976.
39. Robertson R. T., Majka I. A., Peter C. P., Bokelman D. I.: Effects of prenatal exposure to
chlorpromazine on postnatal development and behavior of rats. Toxicol Appl PharmacoI53:541-549,
1980.
40. Kellogg C., Ison I., Miller 1.: Prenatal diazepam exposure: Effects on auditory temporal resolution in
rats. Psychopharmacology 79:332-337, 1983.
41. Kellogg C.K.: Benzodiazepines: Influence on the developing brain. Prog Brain Res 73:207-228,1988.
42. Miller G. C., Friedhoff A. G.: Prenatal neurotransmitter programming of postnatal receptor function.
Prog Brain Res 2:509-522, 1987.
43. Montero D., DeCebailos M., DelRio I.: Down-regulation of 3D-imipramine binding sites in rat cerebral
cortex after prenatal exposure to antidepressants. Life Sci 46: 1619-1626, 1990.
44. Lauer 1. A., Adams P. M., lohnson K. M.: Perinatal diazepam exposure: Behavioral and neurochemical
consequences. Neurotoxicology and Teratology 9:213-219, 1989.
45. Auerbach 1. G., Dans S. L., Marcus I., Maeir S.: Maternal psychotropic medication and neonatal
behavior. Neurotoxicol TeratoI14:399-406, 1992.
46. Eggermont E.: Withdrawal symptoms in neonates associated with maternal imipramine therapy. Lancet
2:680, 1972.
47. Schimmel M., Katz E., Shaag Y., Pastuszak A., Koren G.: Toxic neonatal effects following maternal
clomipramine therapy. Clin ToxicoI29:479-84, 1991.
48. Cowe L., Lloyd D., Dawling S.: Neonatal convulsions caused by withdrawal from maternal
clomipramine. Br Med J 284:1837-1838, 1982.
49. Webster P. A. C.: Withdrawal symptoms in neonates associated with maternal antidepressant therapy.
Lancet 2:318-319, 1973.
50. Shearer W. T., Schreiner R. L., Marshall R. E.: Urinary retention in a neonate secondary to maternal
ingestion of nortriptyline. J Pediatr 81 :570-572, 1972.
51. Dill R. M., Desmond M. M., Kay I. L.: Extrapyramidal dysfunction in an infant of a schizophrenic
mother. J Pediatr 69:589-595, 1966.
52. Scokel P. W., Jones W. D.: Infant jaundice after phenothiazine drugs for labour: An enigma. Obstet
GynecoI20:124-127, 1962.
53. Falterrnan L. G., Richardson D. 1.: Small left colon syndrome associated with maternal ingestion of
psychotropics. J Pediatr 97:300-310, 1980.
54. Tamer A., McKey R., Arias D., Worley L., Fogel B. I.: Phenothiazine-induced extrapyramidal dysfunc-
tion in the neonate. J Pediatr 75:479-480, 1969.
55. Fisher I., Edgren B., Mammel M.: Neonatal apnea associated with maternal clonazepam therapy.
Obstet Gynecol66:34, 1985.
56. Athinarayanan P., Peirog S., Nigam S.: Chlordiazepoxide withdrawal in the neonate. Am J Obstet
Gynecol 124:212-213, 1976.
57. Gillberg C.: "Floppy infant death syndrome" and maternal diazepam. Lancet 2:244, 1977.
58. Mazzi E.: Possible neonatal diazepam withdrawal: A case report. Am J Obstet GynecoI129:586-587,
1977.
59. Whitelaw A., Cummings A., McFadyen I.: Effect of maternallorazepam on the neonate. Br Med J
282: 11 06-11 08, 1981.
60. Gelenberg A. I.: Pregnancy, psychotropics, and psychiatric disorders. Psychosomatics 27:216-217,
1986.
61. Gelenberg A. I.: Psychotropic drugs during pregnancy and the perinatal period. Bioi Ther Psychiatry
2:41-42, 1979.
62. Nahas C., Goujard I.: Phenothiazines, benzodiazepines and the fetus, in Scarpelli E. M., Cosmi E. V.
(eds): Reviews in Perinatal Medicine. New York, Raven Press, 1978.
63. Riordan I.: Drugs and breastfeeding, in Riordan I., Auerbach K. G. (eds): Breastfeeding and Human
Lactation. Boston, lones and Bartlett, 1993.
10. PSYCHOTROPIC DRUG USE IN PREGNANCY 435
64. Buist A.. Norman T. R .• Dennerstein L.: Breastfeeding and the use of psychotropic medication: A
review. J Affective Disord 19:197-206, 1990.
65. Matheson I., Panele H.. Alertson A. R: Respiratory depression caused by N-desmethyldoxepin in
breats milk. Lancet 2:1124,1985.
66. Tunnessen W. W.• Hertz C. G.: Toxic effects of lithium in newborn infants: A commentary. J Pediatr
81:804-807, 1972.
67. Ananth J.: Side effects in the neonate from psychotropic agents excreted through breastfeeding. Am J
Psychiatry 135:801-804, 1978.
68. Tohen M., Waternaux C. M .• Tsuang M. T.: Outcome in mania. A 4-year prospective follow-up of 75
patients utilizing survival analysis. Arch Gen Psychiatry 47:1106-1111, 1990.
69. Pollack M. H., Otto M. w.. Rosenbaum J. F.. Sachs G. S .• O'Neil c.. Asher R .• Meltzer-Brody S.:
Longitudinal course of panic disorder: Findings from the Massachusetts General Hospital Naturalistic
Study. J Clin Psychiatry 51: 12-16. 1990.
70. Charney D .• Heninger G.: Abnormal regulation of noradrenergic function in panic disorder. Arch Gen
Psychiatry 43:1042-1054,1986.
71. Siever L.. Uhde T.: New studies and perspectives on the noradrenergic receptor system in depression:
Effects of the alpha-2 adrenergic agonist c1onidine. Bioi Psychiatry 19:131-156, 1984.
72. Cohen L. S., Rosenbaum J. F., Heller V. L.: Panic attack-associated placental abruption: A case report. J
Clin Psychiatry 50:266-267. 1989.
73. Perkin M. R. Bland J. M., Peacock J. L.. Anderson H. R: The effect of anxiety and depression during
pregnancy on obstetric complications. Br J Obstet Gynaecol 100:629-634, 1993.
74. Steer R A., Scholl T. 0 .. Hediger M. L.. Fischer R. L.: Self-reported depression and negative pregnan-
cy outcomes. J Clin EpidemioI45:1093-1099, 1992.
75. Kopelman A. E., McCullar F. W., Heggeness L.: Limb malformations following maternal use of
haloperidol. JAMA 231:62-64, 1975.
76. Altshuler L. L., Cohen L. S .• Szuba M. P.• Burt V. K.. Gitlin M .• Mintz 1.: Pharmacologic management
of psychiatric illness in pregnancy: Dilemmas and guidelines. Am J Psychiatry 153:592-606, 1996.
77. Hanson G .• Oakley G.: Haloperidol and limb deformity. JAMA 231:26. 1975.
78. van Waes A., van de Velde E.: Safety evaluation of haloperidol in the treatment of hyperemesis
gravidum. J Clin Psychopharmacol 224-?, 1969.
79. Waldman M .• Safferman A.: Pregnancy and c1ozapine. Am J Psychiatry 150:168-169, 1993.
80. Levy W., Wisniewski K.: Chlorpromazine causing extrapyramidal dysfunction in newborn infants of
psychotic mothers. N Y State J Med 74:684-685. 1974.
81. Desmond M. M .• Rudolph A. 1., Hill R M.: Behavioral alterations in infants born to mothers on
psychoactive medication during pregnancy. in Farrell, G. (ed): Congenital Mental Retardation. Univer-
sity of Texas, 1967.
82. Clarke c.. Gorman D., Vernadakis A.: Effects of prenatal administration of psychotropic drugs on
behavior of developing rats. Dev Psycho I 3:225-235, 1970.
83. Golub M., Kornetsky c.: Seizure susceptibiity and avoidance conditioning in adult rats treated pre-
natally with chlorpromazine. Dev Psychobiol 7:79-88. 1974.
84. Spear L. P., Shalaby I. A .. Brick J.: Chronic administration of haloperidol during development. Behav-
ioral and psychopharmacological effects. Psychopharmacology 70:47-58. 1980.
85. Cagiano R, Barfield R, White N.: Subtle behavioral changes produced in rat pups exposed in utero to
haloperidol. Eur J Pharmacol 157:45-50, 1988.
86. Heinonen 0 .• Sloan D .• Shapiro S.: Birth Defects and Drugs in Pregnancy. Littleton, Mass., Publishing
Services Group, 1977.
87. Saxen I.: Cleft palate and maternal diphenhydramine intake. Lancet 1:407-408, 1974.
88. Nelson M., Forfar J.: Associations between drugs administered during pregnancy and congenital
anomalies in the fetus. Br Med J 1:523-527, 1971.
89. Aselton P., Jick H., Milunsky A., Hunter 1.. Stergachis A.: First-trimester drug use and congenital
disorders. Obstet Gynecol 65:451-455, 1985.
90. Zajicek E.: Psychiatric problems during pregnancy. in Wolkind S .• Zajicek E. (eds): Pregnancy: A
Psychological and Social Study. London. Academic. 1981.
436 LEE S. COHEN, M.D., ET AL
91. O'Hara M. W.: Postpartum Depression: Causes and Consequences. New York, Springer-Verlag, 1995.
92. O'Hara M. w., Zekoski E. M., Phillips L. H., Wright E. 1.: A controlled prospective study of postpar-
tum mood disorders: Comparison of childbearing and nonchildbearing women. 1 Abnorm Psychol
99:3-15, 1990.
93. Gotlib I. H., Whiffen V. E., Mount J. H., Milne K., Cordy N. I.: Prevalence rates and demographic
characteristics associated with depression in pregnancy and the postpartum period. 1 Consult Clin
Psychol 57:269-274, 1989.
94. Kessler R c., McGonagle K. A., Swartz M., Blazer D. G., Nelson C. B.: Sex and depression in the
National Comorbidity Survey I: Lifetime prevalence, chronicity and recurrence. 1 Affect Dis 29:85-96,
1993.
95. McBride W. G.: Limb deformities associated with ininodibenzyl hydrochloride [letter]. Med 1 Aust
1:492, 1972.
96. Misri S., Sivertz K.: Tricyclic drugs in pregnancy and lactation: A preliminary report. Int 1 Psychiatry
Med 21:157-171, 1991.
97. Pastuszak A., Schick-Boschetto B., Zuber c., Feldkamp M., Pinelli M., Sihn S., Donnenfeld A.,
McCormack M., Leen-Mitchell M., Woodland c., Gardner A., Hom M., Koren G.: Pregnancy outcome
following first-trimester exposure to fluoxetine (Prozac). lAMA 269:2246-2248, 1993.
98. Crombie D., Pinsent R J., Felming D.: Imipramine and pregnancy. Br Med 1 1:745, 1972.
99. Rachelefsky G., Flynt 1., Ebbin A.: Possible teratogenicity of tricyclic antidepressants. Lancet 1:838,
1972.
100. Sim M.: Imipramine and pregnancy. Br Med 1 2:45, 1972.
101. Scanlon F: Use of antidepressant drugs first trimester [letter]. Med 1 Aust 2:1077, 1969.
102. Briggs G., Freeman R., Yaffe S.: Drugs in Pregnancy and Lactation. Baltimore, Williams & Wilkins,
1994.
103. Banister P., Dafoe C., Smith E. S. 0., Miller J.: Possible teratogenicity of tricyclic antidepressants
[letter]. Lancet 1:838-839, 1972.
104. Morrow A.W.: Imipramine and congenital abnormalities. N Z Med lournaI75:1972, 1972.
105. Jacobs D.: Imipramine (tofranil) [letter]. S Afr Med Journal 46: 1023, 1972.
106. Ali S., Buelkesam J., Newport L.: Early neurobehavioral and neurochemical alterations in rats pre-
natally exposed to imipramine. Neurotoxicol 7:365-380, 1986.
107. DelRio 1., Montero D., DeCeballos M.: Long lasting changes after perinatal exposure to antidepres-
sants. Prog Brain Res 73:173-187, 1988.
108. Coyle L. R: Changes in developing behavior following prenatal administration. Pharmacol Biochem
Behav 3:799-807, 1975.
109. File S. E., Tucker J. c.: Prenatal treatment with clomipramine: Effects on the behavior of male and
female adolescent rats. Psychopharmacology 82:221-224, 1984.
110. Jason K., Cooper T., Friedman E.: Prenatal exposure to imipramine alters early beahavioral develop-
ment and beta adrenergic receptors in rats. 1 Pharmacol Exp Ther 217:461-466, 1981.
111. DeCeballos M., Benedi A., DeFelipe C., DelRio 1.: Prenatal exposure of rats to antidepressants
enhances agonist affinity of brain dopamine receptors and dopamine-mediated behavior. Eur 1 Phar-
macoI116:257-262, 1985.
112. DeCeballos M., Benedi A., Urdin C., DelRio 1.: Prenatal exposure of rats to antidepressant drugs down-
regulates beta-adrenoreceptors and 5-HT2 receptors in cerebral cortex. Lack of correlation between
5HT2 receptors and seratonin-mediated behavior. Neuropharmacol 23:947-952, 1985.
113. Inman W., Kobotu K., Pearce G., et al: Prescription event monitoring of paroxetine. PEM Reports PXL
1206:1-44, 1993.
114. Cooper G. L.: The safety of fluoxetine: An update. Br 1 Psychiatry 153:77-86, 1989.
115. Poulson E., Robson 1. M.: Effect of phenelzine and some related compounds in pregnancy. 1 Endo-
crinol 30:205-215, 1964.
116. Biomedical Information Corporation: Compendium of Drug Therapy (The Obstetrician and Gynecolo-
gist's). New York, Biomedical Information Corporation, 1984.
117. Chiarello R., Cole J.: The use of psychostimulants in general psychiatry: a reconsideration. Arch Gen
Psychiatry 44:286-295, 1987.
10. PSYCHOTROPIC DRUG USE IN PREGNANCY 437
118. Nora 1., McNamara D., Fraser E: Dextroamphetamine sulphate and human malformations. Lancet
1:570-571, 1967.
119. Kalter H., Warkang 1.: Congenital malformations. N Eng J Med 308:491-497, 1983.
120. Zierler S.: Maternal drugs and congenital heart disease. Obstet GynecoI65:155-165, 1985.
121. Nora 1., Vargo T., Nora A: Dexamphetamine: A possible environmental trigger in cardiovascular
malformations. Lancet 1:1290-1291, 1970.
122. Milkovich L., van den Berg B. 1.: Effects of antenatal exposure to anorectic drugs. Am J Obstet Gynecol
121:637-642, 1977.
123. Ramer c.: The case history of an infant born to an amphetamine addicted mother. Clin Pediatr 13:596-
597, 1974.
124. Oro A, Dixon S.: Perinatal cocaine and methamphetamine exposure: Maternal and neonate correlates.
J Pediatr 111:571-578,1987.
125. Goldstein H., Weinberg 1., Sankstone M.: Shock therapy in psychosis complicating pregnancy, a case
report. Am J Psychiatr 98:201-202, 1941.
126. Impasato D. 1., Gabriel A R., Lardara M.: Electric and insulin shock therapy during pregnancy. Dis
Nerv Syst 25:542-546, 1964.
127. Remick R. A, Maurice W. L.: ECT in pregnancy [letter]. Am J Psychiatry 135:761-762, 1978.
128. Ferrill M. 1., Kehoe W. A., lacisin 1. 1.: ECT during pregnancy: Physiologic and pharmacologic
considerations. Convulsive Ther 8: 186-200, 1992.
129. Miller L. 1.: Use of electroconvulsive therapy during pregnancy. Hosp Community Psychiatry 45:444-
450, 1994.
130. Sherer D. M., D' Amico L. D., Warshal D. P., Stern R. A, Grunert H. E, Abramowicz 1. S.: Recurrent
mild abruptio placentae occurring immediately after repeated electroconvulsive therapy in pregnancy.
Am J Obstet Gynecol 165:652-653, 1991.
131. Wise M. G., Ward S. C., Townsend-Parchman w., Giltrap L. c., III, Hauth 1. C.: Case report of ECT
during high-risk pregnancy. Am J Psychiatry 141:99-101, 1984.
132. Repke 1. T., Berger N. G.: Electroconvulsive therapy in pregnancy. Obstet Gynecol 63:39S-40S,
1984.
133. Weinstein M. R.: The International Register of Lithium Babies. Drug In! J 50:81-86, 1976.
134. Schou M., Goldfield M. D., Weinstein M. R., Villaneuve A: Lithium and pregnancy, I: Report from the
register of lithium babies. Br Med J 2:135-136, 1973.
135. Nora 1. 1., Nora A H., Toews W. H.: Lithium, Ebstein's anomaly and other congenital heart defects.
Lancet 2:594-595, 1974.
136. Kirklin 1. w., Barratt-Boyes B. G.: Ebstein's malformation, in Kirklin 1. W., Barratt-Boyes B. G. (eds):
Cardiac Surgery New York, John Wiley, 1986.
137. Kallen B., Tandberg A: Lithium and pregnancy: A cohort study on manic-depressive women. Acta
Psychiatr Scand 68:134-139, 1983.
138. Edmonds L. D., Oakley G. P.: Ebstein's anomaly and maternal lithium exposure during pregnancy.
Teratology 41 :551-552, 1990.
139. Kallen B.: Comments on teratogen update: Lithium. Teratology 38:597, 1988.
140. Sipek A.: Lithium and Ebstein's anomaly. Cor Vasa 31:149-156,1989.
141. Zalstein E., Koren G., Einarson T., Freedom R. M.: A case control study on the association between
1st-trimester exposure to lithium and Ebstein's anomaly. Am J Cardio165:817, 1990.
142. Schou M., Amdisen: Lithium and the placenta [letter]. Am J Obstet Gynecol 122:541, 1975.
143. Woody 1., London w., Wilbanks G.: Lithium toxicity in a newborn. Pediatrics 47:94-96, 1971.
144. Ananth 1.: Side effects of fetus and infant of psychotropic drug use during pregnancy. Pharmaco-
psychiatry 11:246-260, 1976.
145. Cohen L. S., Sichel D. A, Robertson L. M., Heckscher E., Rosenbaum 1. E: Postpartum prophylaxis
for women with bipolar disorder. Am J Psychiatry 152:1641-1645, 1995.
146. Schou M.: What happened later to the lithium babies: A follow-up study of children born without
malformations. Acta Psychiatr Scand 54:193-197, 1976.
147. Delgado-Escueta A., lanz D.: Consensus guidelines: Preconception counseling, management, and care
of the pregnant woman with epilepsy. Neurology 42(Suppl 5):149-160, 1992.
438 LEE S. COHEN, M.D., ET AL
148. Fisher J., Vorhees C.: Developmental toxicity of antepileptic drugs: Relationship to postnatal dysfunc-
tion. Pharmacol Res 26:207-21, 1992.
149. So E.: Update on epilepsy. Med Clin North Am 77:203-214, 1993.
150. Rosa F.w.: Spina bifida in infants of women treated with carbamazepine during pregnancy. N Engl J
Med 324:674-677, 1991.
151. Lindhout D., Meinardi H., Meijer 1.: Antiepileptic drugs and teratogenesis in two consecutive cohorts:
Changes in prescription policy paralleled by changes in pattern malformations. Neurology 42(Suppl
5):94-110, 1992.
152. Larnmer E. J., Sever L. E., Oakley G. P.: Teratogen update: Valproic acid. Teratology 35:465-473,
1987.
153. Omtzigt J. G. C., Los F. J., Grobbee D. E., Pijpers L., Jahoda M. G. J., Brandenburg H., Stewart P. A.,
Gaillard H. L. J., Sachs E. S., Wladimiroff 1. W., Lindhout D.: The risk of spina bifida aperta after first-
trimester exposure to valproate in a prenatal cohort. Neurology 42:119-125, 1992.
154. Battino D., Binelli S., Caccamo M.: Malformation in offspring of 305 epileptic women: A prospective
study. Acta Neurol Scand 85:204-207, 1992.
155. Koch S., Losche G., Jager-Roman E.: Major and minor birth malformations and antiepileptic drugs.
Neurology 42(Suppl 5):83-88, 1992.
156. Nakane Y., Okuma T., Takahashi R., Sato Y., Wada T., Sato T., Fukushima Y., Kumashiro H., 000 T.,
Takahashi T., et al: Multi-institutional study on the teratogenicity and fetal toxicity of antiepileptic
drugs: A report of a collaborative study group in Japan. Epilepsia 21:663-680, 1980.
157. Shaw G. M., Wasserman C. R., 0' Malley C. D., Lammer E. J., Finnell R. H.: Orofacial clefts and
maternal anticonvulsant use. Reprod ToxicoI9:97-98, 1995.
158. Jager-Roman E., Deichi A., Jakob S.: Fetal growth, major malformations, and minor anomalies in
infants born to women receiving valproic acid. J Pediatr 108:997-1004, 1986.
159. Gaily E., Granstrom M.: Minor anomalies in children of mother with epilepsy. Neurology 42(Suppl
5):128-131, 1992.
160. Granstrom M. L., Gaily E.: Psychomotor development in children of mothers with epilepsy. Neurology
42:144-148, 1992.
161. Scolnik D., Nulman I., Rovet J., Gladstone D., Czuchta D., Gardner H. A., Gladstone R., Ashby P.,
Weksberg R., Einarson T., Koren G.: Neurodeve\opment of children exposed in utero to phenytoin and
carbamazepine monotherapy. JAMA 271:767-770, 1994.
162. Safra M. J., Oakley G. P.: Association between cleft lip with or without cleft palate and prenatal
exposure to diazepam. Lancet 2:478-480, 1975.
163. Aarskog D.: Association between maternal intake of diazepam and oral clefts. Lancet 2: 1975.
164. Saxen I.: Association between oral clefts and drugs taken during pregnancy. 1m J EpidemioI4:37-44,
1975.
165. Czeizel A.: Diazepam, phenytoin and aetiology of cleft lip and/or cleft palate. Lancet 1:810, 1976.
166. Crombie D., Pinsent R., Fleming D.: Fetal effects of tranquilizers in pregnancy [letter]. N Eng J Med
293:198-199, 1975.
167. Hartz S., Heinonen 0., Shapiro S.: Antenatal exposure to meprobamate and chlordiazepoxide in
relation to malformations, mental development, and childhood mortality. N Eng L Med 292:726-728,
1975.
168. Bergman U., Rosa F. w., Baum C., Wilholm B. E., Faich G. A.: Effects of exposure to benzodiazepine
during fetal life. Lancet 340:694-696, 1992.
169. Laegreid L., Olegard R., Conradi N., Hagberg G., Wahlstrom J., Abrahamsson L.: Congenital malfor-
mations and maternal consumption of benzodiazepines: A case control study. Dev Med Child Neurol
32:432-441, 1990.
170. Milkovich L., van den Berg B. J.: Effects of prenatal meprobamate and chlordiazepoxide hydrochloride
on human embryonic and fetal development. N Engl J Med 291:1268-1271, 1974.
171. St. Clair S. M., Schirmer R. G.: First-trimester exposure to alprazolam. Obstet Gynecol 80:843-846,
1992.
172. Sullivan F. M., McElhatton P. R.: A comparison of the teratogenic activity of the antiepileptic drugs
carbamazepine, clonazepam, ethosuximide, phenobarbital, phenytoin, and primidone in mice. Toxicol
Appl Pharmacol 40:365-378, 1977.
10. PSYCHOTROPIC DRUG USE IN PREGNANCY 439
173. Rementaria J. L., Blatt K.: Withdrawal symptoms in neonates from intrauterine exposure to diazepam. J
Pediatr 90:123-126, 1977.
174. Rowlatt R: Effect of maternal diazepam on the newborn. Br J Anaethesiol 1:985, 1978.
175. Speight A.: Floppy infant syndrome and maternal diazepam and/or nitrazepam. Lancet 2:878, 1977.
176. Viggedal G., Hagberg B. S., Laegreid L., Aronsson M.: Mental development in late infancy after
prenatal exposure to benzodiazepines: A prospective study. J Child Psychol Psychiatr 34:295-305,
1993.
177. Spielvogel A., Wile 1.: Treatment and outcomes of psychotic patients during pregnancy and childbirth.
Binh 19: 1992.
178. Wrede G., Mednick S., Huttenen M.: Pregnancy and delivery complications in the births of unselected
series of Finnish children with schizophrenic mothers. Acta Psychiatr Scand 62:369-381, 1980.
179. Miller L. J.: Psychotic denial of pregnancy: Phenomenology and clinical management. Hosp Commu-
nity Psychiatry 41:1233-1237, 1990.
180. Carpenter W., Hanlon T., Heinrichs D., Summerfelt A., Kirkpatrick B., Levine J., Buchanan R.:
Continuous vs. targeted medication in schizophrenic outpatients: Outcome results. Am J Psychiatry
147:1138-1141, 1990.
181. Davidson J., Robertson E.: A follow-up study of postpartum illness. Acta Psychiatr Scand 71 :451-457,
1985.
182. McNeil J. F.: A prospective study of postpartum psychosis in a high risk group: Relationship to
demographic and psychiatric history characteristics. Acta Psychiatr Scand 73:35-43, 1987.
183. Cutrona C. E.: Causal attributions and perinatal distress. J Abnorm PsychoI92:161-172, 1983.
184. Zuckerman B., Bauchner H., Parker S., Cabral H.: Maternal depressive symptoms during pregnancy,
and newborn irritability. J Dev Behav Pediatr 11: 190-194, 1990.
185. Cohn J. F., Tronick E.: Specificity of infants' response to mothers' affective behavior. J. Am Acad Child
Adolesc Psychiatry 28:242-248, 1989.
186. Beck A. T., Rush A. J., Shaw B. E, Emery G.: Cognitive Therapy of Depression. New York, Guilford
Press, 1979.
187. Klerman G. L., Weissman M. M., Rounsaville B. J., et al: Interpersonal psychotherapy of depression.
New York, Basic Books, 1984.
188. Stoukides 1. A., Stoukides C. A.: Extrapyramidal symptoms upon discontinuation of fluoxetine [letter].
Am J Psychiatr 148:1263, 1991.
189. O'Hara M. w., Neunaber D. 1., Zekoski E. M.: Prospective study of postpartum depression: Prevalence,
course, and predictive factors. J Abnorm PsychoI93:158-171, 1984.
190. Wisner K. L., Wheeler S. B.: Prevention of recurrent postpartum major depression. Hosp Community
Psychiatry 45:1191-1196, 1994.
191. Brockington I. E, Cernik K. E, Schofield E. M., et al: Puerperal psychosis: Phenomena and diagnosis.
Psychiatry 38:829, 1981.
192. Kendell R. E., Chalmers J. C., Platz C.: Epidemiology of puerperal psychoses. Br J Psychiatry
150:662-673, 1987.
193. Kendell R. E., Wainwright S., Hailey A., Shannon B.: The influence of childbirth on psychiatric
morbidity. Psychol Med 6:297-302, 1976.
194. Paffenbarger R. A.: Epidemiological aspects of mental illness associated with childbearing, in
Brockington I. E, Kumar R. (eds): Motherhood and Mental Illness. New York, Grune and Stratton,
1982.
195. Reich T., Winokur G.: Postpartum psychosis in patients with manic depressive disease. J Nerv Ment Dis
151:60-68, 1970.
196. Faedda G. L., Tondo L., Baldessarini R. J., Suppes T., Tohen M.: Outcome after rapid vs gradual
discontinuation of lithium treatment in bipolar disorders. Arch Gen Psychiatry 50:448;-455, 1993.
197. MRC Vitamin Study Research Group: Prevention of neural-tube defects: Results of the medical
research council vitamin study. Lancet 338:131-137, 1991.
198. Stewart D. E., Klompenhouwer J. L., Kendall R. E., Van Hurst A. M.: Prophylatic lithium in puerperal
psychosis: The experience of three centers. Br J Psychiatry 158:393-397, 1991.
199. Wheeler E. 0., White P. D., Reed E. W., et al: Neurocirculatory anthenia (anxiety neurosis, effort
syndrome, neurasthenia): A twenty-year follow-up study of 173 patients. JAMA 142:878-890, 1950.
440 LEE S. COHEN, M.D., ET AL
200. Barron w., Mujais S., Zinaman M.: Plasma catacholamine responses to physiologic stimuli in normal
human pregnancy. Am J Obstet Gynecol 154:80-84, 1986.
201. Pernoll M., Metcalfe J., Kovach p.: Ventilation during rest and exercise in pregnancy and postpartum.
Respir Physiology 25:295-310, 1975.
202. Klein D. F.: False suffocation alarms, spontaneous panics, and related conditions. Arch Gen Psychiatry
50:306-317, 1993.
203. Cowley D. S., Roy·Byrne P. P.: Panic disorder during pregnancy. J Psychosom Obstet Gynaecol
10:193-210, 1989.
204. George D. T., Ladenheim J. A., Nutt D. J.: Effect of pregnancy on panic attacks. Am J Psychiatry
144:1078-1079, 1987.
205. Klein D.: Delineation of two drug responsive anxiety syndromes. Psychopharmacology 5:397-408,
1964.
206. Villeponteaux V. A., Lydiard R. B., Laraia M. T., Stuart G. W., Ballenger 1. C.: The effects of pregnancy
on pre-existing panic disorder. J Clin Psychiatry 53:201-203, 1992.
207. Robinson L., Walker J. R., Anderson D.: Cognitive-behavioural treatment of panic disorder during
pregnancy and lactation. Can J Psychiatry 37:623-626, 1992.
208. Buttolph M. L., Holland A.: Obsessive compulsive disorders in pregnancy and childbirth, in Jenike M.,
Baer L., Minichiello W. E. (eds): Obsessive Compulsive Disorders, Theory and Management. Chicago,
Yearbook Medical Publishers, 1990.
209. Neziroglu F., Anemone R., Yaryura-Tobias J. A.: Onset of obsessive-compulsive disorder in pregnan-
cy. Am J Psychiatry 149:947-950, 1992.
210. Ingram I. M.: Obsessional illness in mental hospital patients. J Ment Sci 107:382-402, 1961.
211. Lo W. H.: A follow-up study of obsessional neurotics in Hong Kong Chinese. Br J Psychiatry 113:823-
832, 1967.
212. Breyer-Pfaff U., Nill K., Entenmann A., Gaertner H. J.: Secretion of amitriptyline and metabolites into
breast milk. Am J Psychiatry 152:812-813, 1995.
213. American Academy of Pediatrics Committee on Drugs: The transfer of drugs and other chemicals into
human milk. Pediatrics 93:137-149, 1994.
11
Psychopharmacology of DIV-
Related Psychiatric Disorders
FRANCISCO FERNANDEZ, M.D.,
and JORGE MALDONADO, M.D.
I. INTRODUCTION
Over the last 15 years, the annual number of reported cases of the acquired immu-
nodeficiency syndrome (AIDS) in the United States has risen steadily. According to the
1995 year-end edition of the HIV / AIDS Surveillance Report issued by the Centers for
Disease Control (CDC), a total of 513,486 cases of adult and pediatric AIDS were
diagnosed in the United States between 1980 and 1995. Approximately 60% of these
individuals have died. In fact, we have witnessed this infection and its related disorders
becoming the leading cause of death in adults aged 25 to 44 years, only midway
through the second decade of the human immunodeficiency virus (HIV)/AIDS pandemic.
Unfounded fears about acquiring and transmitting HIV infection continue to ham-
per education and prevention efforts. The scientific evidence indicating that transmis-
sion of HIV occurs exclusively by contact with infected cells is overwhelming and
conclusive. Sexual intercourse of any type, inoculation by infected blood or blood
products, and intrapartum, perinatal, or passive transmission of HIV via breast milk are
the only modes of transmitting HIV. The minimal dose of HIV that is required to
produce active infection is unknown. However, the extremely low rates of infection
after needlestick accidents in the health care setting as well as the lower rates of
infection by other fluids where the virus has been cultured in small amounts (i.e.,
saliva, tears, urine, bronchial secretions, etc.) suggest that the efficiency of transmis-
sion is greatest for inoculation with infected blood or blood products, followed by
perinatal transmission and sexual intercourse. All the epidemiological and sero-
prevalence data exclude airborne, fecal-oral, and insect transmission. Casual transmis-
Francisco Fernandez, M.D. • Department of Psychiatry, Stritch School of Medicine, Loyola University
Medical Center, Maywood, Illinois 60153. Jorge Maldonado, M.D. • Psychosomatic Medicine and
Consultation Liaison Psychiatry, St. Luke's Episcopal Hospital, Houston, Texas, 77030; Baylor College of
Medicine, Houston, Texas 77030.
The Practitioner's Guide to Psychoactive Drugs (Fourth Edition), Gelenberg and Bassuk, eds.
Plenum Publishing Corporation, New York, 1998
441
442 FRANCISCO FERNANDEZ, M.D., AND JORGE MALDONADO, M.D.
The HIV antibody test and the cell ratios of CD4/CD8 lymphocytes remain the
basis for diagnosing HIV infection. HIV antigen measures and the polymerase chain
reaction (PCR) detection methods are not indicated for routine diagnosis. Even viral
culture is reserved for selected cases such as suspected neonatal infection.
The enzyme-linked immunoabsorbent assay (ELISA) has become the standard
screening procedure for HIV infection. It is based on antibody detection (otherwise
known as seroconversion following acute infection) specific for HIY. The timing for
the development and detection of antibodies to HIV infection is known as the window
period. This is usually 6-8 weeks although it is occasionally prolonged up to 3 months
and infrequently to 18 months.
The Western blot or immunoblot test is a technique used for confirming the
diagnosis of HIV infection after the antibody test is positive. This electrophoretic
technique separates specific viral antigens into bands that have characteristic orienta-
tion on the nitrocellulose strips and can be readily identified. Accurate diagnosis can be
made in almost all instances, and the additional techniques of antigen detection, HIV
culture, and nucleic acid enhancement by PCR provide complementary techniques for
indeterminant cases.
Because the primary target of HIV infection is the CD4 T-cell lymphocyte, it is
important to understand its dysfunction and depletion in HIV disease. This will help us
better understand some of the neuropsychiatric manifestations of HIV infection and
AIDS. T cells exist in two varieties based on their functional capacities: "help-
erlinducer" cells (CD4 cells) and "suppressor/cytotoxic" cells (CD8) cells. The CD4
cells facilitate antigen recognition and induce a cascade of immune responses critical to
processing foreign antigens internally. They activate the immediate and long-term
immune response when reexposure or rechallenging occurs at a later date by the same
antigens. CD4 cells are infected by HIV and destroyed by programming their own cell
death (i.e., apoptosis) via an aberrant transduction signal to the nucleus. Over time,
there is a gradual progressive decline of CD4 lymphocytes. This edition of CD4 cells
has become the centerpiece for classifying the disease as well as for monitoring its
progression. The slope of the CD4 drop is highly predictive of the course of the disease
for each individual. In fact, in 1993 the CDC case definition for AIDS was modified to
include individuals with CD4 cell counts less than 200/mL. The great majority of
patients remain clinically asymptomatic during the decline of CD4 cells (the so-called
clinical latency period), and only when the number of cells drops below 200/mL does
the patient becomes susceptible to a host of opportunistic infections (OIs), neoplastic
diseases, and neurological complications. There are other immunological abnormalities
that will not be discussed here, and the reader is referred to Harrison s Principles of
Internal Medicine, Chapter 279, for more information.
11. IDV-RELATED PSYCHIATRIC DISORDERS 443
A. General Considerations
During the acute HIV infection, also known as the primary HIV infection, a
mononucleosis-like syndrome can occur in as many as half of the patients infected with
HIV. Neuropsychiatric signs and symptoms include headache, photophobia, and a
meningoencephalitis. HIV is known to enter the central nervous system during this
phase of the infection and lay dormant for years.
Following this period, individuals typically will enter a clinical latency period that
may last 10 or more years (CDC clinical stage A). Some of these individuals will
demonstrate signs of neuropsychological impairment. The most common signs and
symptoms include mild problems with abstraction, psychomotor speed of information
processing, and impaired verbal memory. Estimates of the number of affected individu-
als range between 0 and 45% of those infected. Recently, one study has suggested that
low educational level may be an important cofactor denoting limited functional reserve
capacity. Though scientific opinions vary on whether there is any true impairment or
not, neuropsychological impairment on tests of cognitive and motor function does
occur. This may not necessarily translate into clinically or functionally relevant
changes, and it is not the same as a fully developed cognitive disorder or dementia.
During the early symptomatic phase of HIV infection (CDC stage B), clinicians
and investigators agree that there may be significant neuropsychological impairment.
This, however, does not necessarily imply clinically relevant impairment. However, the
AIDS Task Force of the American Academy of Neurology has recently published
clinical guidelines for defining some of the early neurological manifestations of HIV
disease (see Table 1). At least two of the following must be present for at least one
month: impaired attention or concentration; mental slowing; impaired memory; slowed
movements; incoordination; and personality change or emotional lability. In order to
confirm the diagnosis, there must be some work-related impairment or change in the
ability of the individual to carry out his/her activities of daily living. This neuro-
behavioral dysfunction is classified as HIV-I-associated minor cognitive/motor disor-
der (HMCMD). The relationship of the cognitive impairment described by some inves-
tigators of HMCMD is unknown.
There is almost complete agreement that significant impairment occurs in late
symptomatic infection (CDC stage C). HIV-I-associated dementia complex (HAD)
may occur in as many as 40% during this stage and may be clinically differentiated into
early and late stages (see Table 1).
B. Neuropsychiatric Complications
Early in the course of the infection, HIV enters the brain by different mechanisms,
including penetrance through endothelial gaps in brain capillaries, the choroid plexus,
and in latently infected monocytes. HIV is thought to destroy subcortical and cortical
areas by direct effect on neural cells, and indirectly by neurotoxic effects. Neurotox-
icity is known to be caused, in part, by the surface glycoprotein gpI20, which sensitizes
TABLE 1. American Academy of Neurology AIDS Task Force Nomenclature and
Research Case Definitions for Neurologic Manifestions of Human Immunodeficiency
Virus-Type 1 (HIV-l) Infectiona
"Adapted from Janssen R. S., Saykin A. J., Cannon L., et al.: American Academy of Neurology AIDS Task Force.
Nomenclature and research case definitions for neurologic manifestations of human immunodeficiency virus-type-I (HIV-I)
infection. Neurology 41 :778-785, 1991.
11. HIV-RELATED PSYCHIATRIC DISORDERS 445
1. Dementia
a. Epidemiology. HIV infection causes a wide spectrum of cognitive impair-
ments that vary from subclinical neuropsychological deficits in otherwise asymptomat-
ic patients to HIV-associated dementia complex (see Section II.A). It is estimated that
60-90% of patients during late symptomatic infection will have impaired cognitive
functioning, and as many as 20-30% may have dementia at some point in this stage.
2. Depression
b. Clinical Features. Patients may present with mood changes that range from
normal sadness and grief to a fully developed depressive disorder including substance-
induced depression, depression due to a general medical condition, or a major depres-
sive disorder. An accurate diagnostic formulation of the depressive disorders becomes
difficult as the usual diagnostic indicators of depression are also common to both
systemic disease and HIV with CNS involvement. One cannot rely on the usually
vegetative signs and symptoms as these are also associated with progression of HIV
disease. As with other medically ill populations, cognitive and affective symptoms
such as tearfulness, low self-esteem or self-worth, helplessness, and hopelessness are
considered to be more reliable for making a definitive diagnosis. Clinically, one should
avoid "splitting hairs" and use an inclusive approach (i.e., count all symptoms toward
the diagnosis, whether vegetative or cognitive) in order to avoid the all-too-frequent
underdiagnosis of depression in the medically ill. An important factor in diagnosing
HIV-related depression is the high frequency of bereavement secondary to multiple
losses that create a cumulative burden predisposing to depression. Clinically, one
should have a low threshold for diagnosing depression in the context of multiple losses
so as to ensure appropriate treatment.
11. HIV·RELATED PSYCHIATRIC DISORDERS 447
Name
Age and birthday
Handedness
First language at home
Educational backgnlund
Best subjects, grades
Worst subjects
Occupational background
How long
Medical history
Childhood diseases or injuries
Head injuries with loss of consciousness
Strokes
High fevers
Toxin exposure
Major illnesses, injuries, or surgeries
Medicines: prescription, nonprescription
Duration of diagnosis of HIV infection; AIDS
Current problem
Change in thinking functions: how long, or, over what period of time
Any change in ability to concentrate
Any periods of confusion or mental "fuzziness"
When talking with people, or on the phone, watching TV or a movie, reading
Any pnlblem with following the train of thought
Any difficulties with handwriting
Any word-finding pnlblems, difficulties with slurring or stammering
Any slowing of thinking or understanding, trouble with mental arithmetic, like making change or
balancing checkbook
Wear glasses
Any blurring vision, double vision, or flashing lights in eyes
Any change in understanding what is seen; do things look right in their relation to each other
Overlook things when right in front of you
Hear any unusual sounds; see unusual things; have any strange feelings?
Any changes in any other senses:
Decreased hearing, ringing, or buzzing sounds
Change in smell or taste
Any numbness, "pins or needles," loss of feeling, tingling, or burning feelings
Any severe pain
Memory:
Any areas of memory that are better or worse
Memory for recent information?
Information fnlm way back in life
Any difference in memory for situations versus rote facts and figures
Kinds of things most easily forgotten: names, addresses, directions, reading
How long can things be remembered, more notes written than used to
Any lapses noted
Any getting lost or forgetting where one is
Any new difficulties with thinking through pnlblems or solving them, decision making, staying
organized-on job, at home
How is sleep: any trouble getting to sleep, night versus daytime, any awakenings from which one
cannot immediately return to sleep
(continued)
448 FRANCISCO FERNANDEZ, M.D., AND JORGE MALDONADO, M.D.
TABLE 2. (Continued)
aAdapted from Levy 1. K., Fernandez E: HIV infection of the eNS: Implications for neuropsychiatry, in Yudofsky S. c.,
Hales R. E. (eds): The American Psychiatric Press Textbook of Neuropsychiatry, 3rd ed. American Psychiatric Press, 1997,
pp. 663-692.
"Adapted from Butters N., Grant I., Haxby J., et al.: Assessment of AIDS-
related cognitive changes: Recommendations of the NIMH Workgroup on
Neuropsychological Assessment Approaches. J c/in Exp Neuropsychol
12:963-978, 1990.
well as a full inventory of medications from all sources is required for assessing the
HIV-infected depressed patient.
3. Anxiety
a. Epidemiology. Anxiety frequently accompanies serostatus notification and
disease progression. About 20% of HIV patients suffer from chronic anxiety. Homo-
sexual men suffer more frequently from preexisting anxiety disorders than heterosexual
450 FRANCISCO FERNANDEZ, M.D., AND JORGE MALDONADO, M.D.
men matched on demographic factors. Drug users may also experience increased
anxiety. These groups, those who have had anxiety disorders before becoming HIV-
infected, and individuals with limited psychosocial support are more likely to experi-
ence increased levels of anxiety during the course of their illness. Among the anxiety
disorders, adjustment disorders with anxious mood are the most common; panic disor-
der and a posttraumatic stress disorder (PTSD)-like syndrome in relation to becoming
HIV-positive are also frequent.
d. Evaluation. The same medical evaluation as for delirium and dementia can be
used when evaluating the HIV-infected patient who presents with anxiety. Baseline
laboratory evaluation is required; it should include blood chemistry, complete blood
count, thyroid function tests, toxicology screen, arterial blood gases, RPR/VDRL,
folate, and B 12' Depending on symptoms profile, also consider brain imaging studies,
lumbar puncture, EEG, chest X-rays, and electrocardiogram. A detailed history of the
current medical status, medications from all sources, and use of illicit drugs is fre-
quently helpful in determining the source of the anxiety.
4. Delirium
TABLE 4. Neuropsychiatric
Infections/Malignancies Associated with HIVa
Aspergillus
Candidiasis
Coccidioidomycosis
Cryptococcus neoformans
Toxoplasma gondii
Atypical mycobacteria
Mycobacterium tuberculosis
Kaposi's sarcoma
Primary or secondary CNS lymphoma
5. Psychosis
"Adapted from Fernandez E. Levy J. K.: Adjuvant treatment of HIV dementia with psychostimuJants. in Ostrow D. (ed):
Behavioral Aspects of AIDS and Other Sexually Transmitted Diseases. New York. Plenum, 1990. pp. 279-286.
stood. In one survey, the incidence of psychosis in symptomatic patients with HIV
infection was reported to be 8%. The distribution of psychosis in HlV disease seems to
be bimodal; early in the disease, psychosis may represent a primary psychiatric disor-
der (such as bipolar affective disorder or schizophrenia), and later on it may result
directly from HIV CNS involvement.
progressive course, with many patients dying within a short period of time « 1 year)
after the onset of the psychotic illness. There are also case reports of brief reactive
psychosis associated with notification of a positive antibody test; in our opinion, this is
a rare event.
6. Insomnia
c. Differential Diagnosis. All primary sleep and other psychiatric disorders need
to be ruled out. In hospitalized patients, delirium should be screened for because either
insomnia or hypersomnia may be prodromal features of a disturbance in cerebral
metabolism. Insomnia may also be associated with an acute stress reaction and be-
reavement. Likewise, the neurotoxic effects of psychoactive substances or prescribed
medications, including HIV / AIDS treatments, need to be carefully evaluated.
A. Dementia
1. Primary Therapy
Zidovudine (ZDV) has shown in vitro and in vivo efficacy in inhibiting retroviral
replication and, therefore, in reducing the morbidity and mortality of AIDS patients.
ZDV has also been effective in treating asymptomatic HIV-infected individuals who
have a CD4 count of less than 500 cells/mL. It has also been suggested that therapy
with ZDV during the primary infection may improve the subsequent clinical course and
increase the CD4 count.
Cognitive function has improved in patients with HMCMD and HAD treated with
ZDY. Significant gains in cognitive functioning occurred with higher doses of ZDV
(2000 mg/day); however, low doses (600 mg/day) have also been associated with
improvement on neuropsychological tests. Didanosine (ddI), another nucleoside now
used in the treatment of AIDS, has been effective in improving HIV-related cognitive
impairment in a small number of cases; however, controlled studies are lacking. Other
nucleosides-zalcitabine (ddC), stavudine (D4t), and lamivudine (3TC)-and the non-
nucleoside reverse transcriptase inhibitors have efficacy alone or in combination in the
11. HIV-RELATED PSYCHIATRIC DISORDERS 455
treatment of AIDS. Unfortunately, controlled studies are lacking regarding their effects
on cognitive function. The new protease inhibitors that prevent maturation of HIY
particles are showing promise in combination with the reverse transcriptase agents in
ameliorating the cognitive deficits of the HIY infection. Again, controlled studies
advocating the use of these agents are lacking.
2. Adjunctive Therapy
Other pharmacotherapies used specifically for HIY CNS infection include pep-
tide-T, which blocks the binding of the glycoprotein gpl20 to the CD4 receptors in the
CNS. Memantine, MK-801, and nitroglycerin are NMDA receptor antagonists and may
be beneficial in preventing the neurotoxic effect caused by HIY infection. Calcium-
channel blockers (nimodipine and nifedipine) decrease the influx of calcium mediated
by the NMDA receptor; however, verapamil has been found to increase viral replica-
tion in vitro and therefore is not clinically used. The tumor necrosis factor-a (TNF-a)
blockers pentoxifylline and thalidomide, interleukin-l receptor blockers, N-acetylcys-
teine, vitamins E, B6 , and B 12 , zinc, nerve growth factor, and others are being studied,
but their role in the treatment of HIY-I-associated cognitive disorder is not yet known.
3. Palliative Therapy
Besides antiviral therapy used to recover cognitive function or to prevent further
cognitive decline in HIY-infected individuals, stimulants have been used to improve
HIY-related cognitive impairment. Methylphenidate ranging from 10 to 90 mg/day in
divided doses is the recommended agent in this group of medications. An initial dose of
5-15 mg/day is progressively increased according to the patient's response. The mean
dose for maximal response is 30-45 mg/day, and higher doses are only occasionally
required. A typical schedule would be 5-10 mg of methylphenidate administered at 7
A.M., 10 A.M., and 1 P.M. Dextroamphetamine is twice as potent as methylphenidate and
is useful when the latter has failed. Initial doses of 5-10 mg daily are increased to up to
45 mg/day in a bj.d. schedule (usually 8 A.M. and noon). The mean dose for maximal
response is 10-20 mg/day. Pemoline is useful when once-a-day dosing is required or
for the patient who has difficulty swallowing or other problems associated with oral
dosing. It is available in a chewable form, which can even be used sublingually;
however, it may be more difficult to titrate than the other stimulants, and it should be
started at one-half of an 18.5-mg tablet every morning and increased slowly.
4. Nonbiological Treatments
Neurorehabilitative strategies based on cognitive skills training for traumatic brain
injury patients are particularly effective as adjunctive therapy in HMCMD and HAD.
Use of attention drills, mnemonic strategies, and other behavioral prostheses are also
effective compensatory strategies. Some have found the use of cognitive-enhancing
stimuli along with f3-EEG biofeedback helpful. Controlled clinical trials are lacking in
the rehabilitative efforts for HMCMD.
456 FRANCISCO FERNANDEZ, M.D., AND JORGE MALDONADO, M.D.
For late HAD, the recommended interventions are the same as for Alzeheimer's
disease. The use of structured environmental strategies utilizing familiar orienting
stimuli has been reported to be useful.
B. Depression
1. Heterocyclic Antidepressants
Since all antidepressants are equally effective, selecting a particular agent solely
depends on the side-effect profile. Antidepressants with significant affinity for the
central muscarinic receptor can result in undesirable anticholinergic effects that can
aggravate HIV-related cognitive impairment or precipitate delirium and seizures. Dry-
ing of mucous membranes likewise can become a problem, predisposing patients to
infections with oral candidiasis.
If tricyclic antidepressants (TCAs) are used to initiate pharmacotherapy for HIV-
infected patients, even those in the asymptomatic phase of infection, a low dose of 10-
25 mg at bedtime is recommended with increases every 1-2 days by 10-25 mg until a
positive symptom response is achieved. These patients, like other medically ill patients,
may respond to lower doses than those usually required by otherwise healthy individu-
als; doses ranging from 25 to 100 mg/day are often sufficient to achieve a good
antidepressant response. However, studies suggest that asymptomatic patients can tol-
erate standard therapeutic regimens of doses up to 100-300 mg of imipramine or its
equivalent. Clinically, the TCAs that have sedating effects, such as amitriptyline,
trimipramine, and doxepin, are used when patients have significant agitation or severe
insomnia. Patients with psychomotor retardation may benefit better from the least-
sedating TCAs, such as the secondary amines desipramine and protriptyline. TCAs are
also used in the treatment of neuropathies, and patients who have both depression and
peripheral neuropathies may benefit from this group of antidepressants.
Intravenous administration of amitriptyline or imipramine has proved to be effec-
tive and offers an alternative for the patient who is unable to take oral medications.
Although intravenous use of antidepressants has not been approved by the U.S. Food
and Drug Administration (FDA), approval by one's own pharmacy and review board
for compassionate use may be sought.
Whenever possible, one should avoid TCAs in advanced HIV disease. Trazodone
has very low anticholinergic effects, and it is very well tolerated by patients with AIDS.
The starting dose for trazodone is usually 25-50 mg at bedtime, increased by 25-50
mg every 3-5 days. Therapeutic response to treatment has been reported with doses of
trazodone as low as 150 mg/day. Major drawbacks of treatment with trazodone are
sedation and orthostatic hypotension. Once-daily dosing with trazodone at bedtime
should be avoided with doses greater than 100 mg to avoid a typical hangover effect the
next morning. Another low-anticholinergic agent is desipramine, which has been re-
ported in one study to have activating effects equivalent to those of methylphenidate.
Dosing can be initiated at 10-25 mg t.i.d. or q.i.d. and increased by 25 mg every 3-5
days as tolerated. Treatment response can occur with total doses of desipramine of 75-
150 mg.
11. HIV-RELATED PSYCHIATRIC DISORDERS 457
2. New Agents
All serotonin-selective reuptake inhibitors (SSRIs) have little, if any, anti-
cholinergic effects. Fluoxetine, 20-40 mg daily, is a good alternative for treating
depression in HIV-infected patients. The efficacy of sertraline and paroxetine has not
been confirmed in large randomized controlled trials; however, reports on small num-
bers of patients reveal these agents to be safe and effective antidepressants in HIV
disease. Only fluvoxamine was poorly tolerated in one trial with HIV-infected de-
pressed patients. Bupropion has an activating effect, and therefore its use offers an
advantage for the withdrawn or anhedonic patient; however, it needs to be used cau-
tiously in patients with CNS disease, because seizures have been precipitated in such
patients at doses greater than 200 mg/day. No immune enhancement or suppression has
been noted with SSRIs used to treat depression in HIV disease. Newer antidepressants,
such as venlafaxine and nefazodone, have been well tolerated and effective.
3. Other Agents
Patients who have been treated with lithium carbonate or monoamine oxidase
inhibitors (MAOIs) before HIV / AIDS was diagnosed should continue taking them.
Closer monitoring is indicated because the patients may develop toxicity more easily
and titration may be required. Patients with infectious complications causing diarrhea
or significant fluid loss from any source may develop neurotoxicity or nephrotoxicity
with lithium carbonate. Alternate agents, such as anticonvulsants, may be required.
Likewise, MAOIs may be continued when patients have been taking them before the
onset of HIV disease. In theory, MAOIs are incompatible with ZDV therapy, which is
reported to have catechol-O-methyltransferase-inhibiting effect.
Psycho stimulants are especially useful in patients who have coexisting depression
and cognitive impairments or in those in whom other antidepressants have proven to be
ineffective as single agents or intolerable due to severe or unusual side effects. Meth-
ylphenidate is an especially safe and effective treatment for depression in HIV disease.
Response may occur within hours after the first dose is administered, producing psy-
chomotor activation, appetite stimulation, and quantitative as well as qualitative im-
provement of higher cognitive functions. Similar dosing as that used adjunctively for
dementia has been reported to have antidepressant efficacy equivalent to that of de-
sipramine. The role of psycho stimulants in treating depression in patients with current
or previous history of drug abuse is questionable, and psycho stimulants should be
avoided in this subgroup of patients.
4. Nonbioiogicai Treatments
C. Anxiety
1. Anxiolytics
Anxious states that are intense enough to be intolerable for the patient may
effectively be treated with anxiolytic agents. Anxiety associated with pain, respiratory
distress, organic mental disorders, or toxicity from other medications also responds
well to anxiolytics, but in these more complex situations lower doses of anxiolytics
may be required.
When anxiolytic therapy is required, short-to-intermediate-acting benzodiaz-
epines are preferred. Long-acting benzodiazepines should be avoided, particularly in
patients with any degree of cognitive impairment, disinhibition, frontal lobe dysfunc-
tion, delirium, or confusion. Oxazepam, lorazepam, and alprazolam are recommended
for HIV-related anxiety. Oxazepam and lorazepam have no active metabolites, which
makes them particularly useful in patients with multi organ disease. For generalized,
free-floating anxiety, lorazepam, 0.5-1 mg, oxazepam, 10-15 mg, or alprazolam,
0.125-0.5 mg t.i.d. or q.i.d., is usually sufficient. Always indicate to patients that
treatment with benzodiazepines should be time-limited (2 weeks) and possibly com-
bined and eventually replaced with more specific pharmacotherapy. For example, some
have advocated combining benzodiazepines with buspirone, 5-10 mg PO t.i.d. Busp-
irone can be increased by 5-10 mg every 3 days until a therapeutic dose (30-60
mg/day) is attained over the 2-week course of benzodiazepine therapy. A ben-
zodiazepine taper can then be initiated. Buspirone can then be continued as long as it is
clinically indicated because it does not have the negative side-effect profile associated
with the benzodiazepines. Patients with advanced HIV disease may experience more
confusion when taking buspirone, but usually it is well tolerated by HIV-infected
patients who are asymptomatic or have limited symptomatic disease. When used for
patients who are taking ZDV, buspirone is better tolerated when ZDV is administered
in low doses than in doses greater than 600 mg/day. Of interest, buspirone has been
reported to curb drug-seeking behavior in anxious patients with addictions.
Trazodone is perhaps more useful than benzodiazepines and buspirone in anxiety
disorders associated with HIV disease. Doses of 25-200 mg of trazodone per day are
helpful as primary anxiolytic drug or as an adjuvant medication for these patients. It
can be used safely in patients with neuromuscular or neuropsychiatric complications,
can be used over long periods of time, and has the advantage of giving an immediate
relaxing and sedating effect, which is experienced by patients as salutary, and it is
therefore preferable to other antidepressants or to buspirone. As with bus pirone,
trazodone can be combined with other antidepressants or with benzodiazepines accord-
ing to the patient's needs.
2. Nonbiological Treatments
tion, impair motor function, and worsen cognitive efficiency, and some may even cause
dependence. Furthermore, reducing anxiety symptoms through behavioral techniques
will support the patient's sense of control, self-esteem, and confidence. As with depres-
sion, cognitive-behavioral therapy has also been reported to be effective. For some
patients, supportive psychotherapy and group therapy may be sufficient to reach a new
level of adaptation and avoid pharmacotherapy as a first line of treatment.
D. Delirium
Pharmacological treatment for delirium should be similar to treatment for an acute
psychosis, but with some modifications. Reports on the use of high-potency neuroJep-
tics, administered by oral or intramuscular routes, in patients with delirium have
demonstrated these agents to be safe and effective and without serious adverse effects.
Neuroleptic malignant syndrome (NMS) is the most ominous potential adverse effect
in any medically ill patient, and it has been reported to occur more frequently in
patients with HIY infection. Likewise, various reports have noted an increased frequen-
cy of extrapyramidal reactions (EPRs) in HIY-infected patients. Because diseased
subcortical gray-matter structures frequently involved in HIY infection include the
basal ganglia, thalamus, and temporolimbic structures, these patients may be more
vulnerable to side effects of high-potency neuroleptics. Results from a controlled study
on HIY-related delirium by Breitbart and colleagues 14 suggest that low doses of halo-
peridol (2.8 ± 2.4 mg/day) and chlorpromazine (50 ± 23 mg/day) were highly effec-
tive in treating delirious patients without significant EPRs. This report also included a
subgroup of delirious HIY patients treated with lorazepam alone. Lorazepam as a
single agent was poorly tolerated; in fact, some patients deteriorated with this treat-
ment.
In delirious HIY-infected patients who are more sensitive to neuroleptics, lower
doses of haloperidol and chlorpromazine or other equivalent agents should be tried. In
HIY-infected patients who cannot tolerate these commonly available neuroleptics,
molindone in doses up to 225 mg/day has been reported effective without significant
side effects. Even in cases in which there has been a previous history of EPRs as well as
NMS, molindone has been found effective as an alternative agent for agitation, psycho-
sis, or delirium.
When using oral or intramuscular agents for treating delirium, these agents are
usually administered on a pm basis depending on the level of agitation. This, in our
opinion, is tantamount to treating sepsis with pm antibiotics depending on the level or
degree of the accompanying fever. Our experience is that when delirium is treated with
neuroleptics on a pm basis, the patient is generally undertreated. A more effective
strategy is to treat on a time-contingent basis (as in pain management) until the patient
is calm (not sedated). One can then calculate the total dose necessary to achieve
calmness and administer it on a scheduled basis for another 24 hr and then reduce the
dose by 20-25% daily until the medication can be safely discontinued.
When oral or intramuscular administration of haloperidol or other neuroleptics has
failed, intravenous pharmacotherapy should be considered. Although the intravenous
administration of haloperidol is considered investigational and is not approved by the
460 FRANCISCO FERNANDEZ, M.D., AND JORGE MALDONADO, M.D.
aAdapted from Sanders K. M., Cassem E. H.: Psychiatric complications in the critically ill cardiac patient. Texas Hean [nSf J
20:180-187, 1993.
FDA, the relative rarity of treatment-related adverse effects must be weighed clinically
against the dangers of untreated delirium. Pharmacologically, intravenous haloperidol
is no different from intravenous droperidol (marketed as a preanesthetic agent for
intravenous use as Inapsine). The major difference is that haloperidol is a halogenated
compound which has less a-I blockade than droperidol. Thus, in medically ill patients
who are either dehydrated or hemodynamically unstable, intravenous haloperidol
would be preferred because it does not further compromise the hemodynamic status of
the patient. The guidelines for intravenous use of haloperidol have been established by
the Massachusetts General Hospital Psychiatric Consultation Service (see Table 6).
Another methodology, advocated by Adams and colleagues,l is to use combina-
tion chemotherapy with haloperidol and lorazepam with or without hydromorphone to
achieve rapid control in severely agitated patients (see Table 7). Adams's approach
should not be used unless one is familiar with the use of intravenous psychotropics in
critically ill patients because hypotension, respiratory depression, and other complica-
tions could occur with the use of this combination of agents. Rarely does one see EPRs
with combination intravenous pharmacotherapy. Lorazepam's possible protective influ-
ence against haloperidol's propensity to cause extrapyramidal side effects has been
postulated, but it must be delineated with further controlled trials. The utilization of
these agents intravenously for delirium can be approved by one's hospital pharmacy
committee or the institutional review board.
E. Psychosis
1. Neuroleptics
The treatment for HIV-related psychosis is much like that for delirium. Initial
treatment of an agitated, psychotic patient with HIV infection requires rapid tranquiliz-
ation with neuroleptics. Unfortunately, EPRs and NMS have been reported with the
high-potency neuroleptics, and confusion and seizures have been reported with the
low-potency neuroleptics.
11. mV-RELATED PSYCHIATRIC DISORDERS 461
1. Initial procedure is the same as in the MGH protocol (Table 6, step I).
2. After two consecutive doses, 30 min apart, if the patient is still agitated, then fix the IV haloperidol
at 10 mg and add lorazepam for residual agitation as follows:
• Mild: 0.5-1.0 mg
• Moderate: 2.0-5.0 mg
• Severe: 6.0-10.0 mg
For example, if a patient fails to respond to haloperidol 10 mg IV x 2, then after 30 min of
observation, if the patient is still agitated, administer haloperidol 10 mg IV immediately followed by
lorazepam, 0.5-1.0 mg IV. If after 2 consecutive doses of IV haloperidol and lorazepam, the patient
remains agitated, then administer haloperidol, 10 mg IV, immediately followed by lorazepam, 2.0-
5.0 mg IV.
3. At each hour interval, if the patient is not calm, continue to administer a fixed dose of 10 mg of
haoperidol IV; lorazepam becomes the variable until the maximum of 10 mg of lorazepam IV is
reached.
4. If the patient fails to respond to two consecutive IV doses of 10 mg of haloperidol and 10 mg of
lorazepam, then add low doses of hydromorphone as follows:
• Mild: 1.5-2.0 mg
• Moderate: 3.0-5.0 mg
• Severe: 6.0-10.0 mg
At this point, both the haloperidol and the lorazepam doses become fixed at 10 mg each, and the
hydromorphone becomes the variable.
5. If a patient fails the above protocol, arrangements should be made with anesthesia to intubate the
patient and utilize propofol (Diprivan®) as per routine.
aAdapted from Fernandez F., Levy J. K., Mansell P. W. A.: Management of delirium in terminally ill AIDS patients. Int J
Psychiatry Med 19(2):165-172 1989.
Various reports have described the use of molindone in AIDS patients with psy-
chosis of varying etiologies who were unable to tolerate other therapies. Treatment
with molindone had infrequent adverse effects. The usual dosage rangers from 30 mg
to 225 mg daily. While the clinical experience with risperidone is limited, we have
noted clinically significant improvement with no sedation and no measurable negative
impact on cognitive functions. Treatment can be initiated with risperidone at 0.5 mg
b.i.d. and adjusted slowly to a maximum of 6 mg daily in divided doses. One report on
the use of risperidone, 1-2 mg b.i.d., in acute mania resulting from HIV disease found
that it effectively reduced psychotic symptoms within a week, without any EPRs.
2. Lithium
Lithium carbonate can be used safely in patients with a preexisting history of
mood disorder, new-onset organic mood disorder, organic personality disorder, or
agitation secondary to neurotoxicity of steroids or antivirals. A reduced dose may be
required as the systemic disease progresses, and possible neurotoxicity should be
monitored more frequently. Doses ranging from 150 to 1800 mg daily can be safely
used, with precaution in more medically compromised patients. As lithium can stimu-
late the production of granulocyte-macrophage colony-stimulating factor (GM-CSF),
462 FRANCISCO FERNANDEZ, M.D., AND JORGE MALDONADO, M.D.
3. Anticonvulsants
Anticonvulsants have also been studied in psychotically ill HlV-infected patients.
A single report on the treatment of mania in HlV disease found a preferential response
to valproic acid (750-1250 mg daily) in patients with abnormal MRI scans of the brain.
In fact, in this series, an abnormal MRI predicted a poor response to lithium and
neuroleptics. Carbamazepine has been avoided in treating HlV-infected patients receiv-
ing antivirals because of the potential for added hematopoietic toxicity from this
combination. Anecdotal reports suggest that it is also beneficial in the treatment of
mania of varying etiologies.
4. Nonbiological Treatments
Although pharmacological treatment is mandatory in patients with psychosis,
individual, family, and group psychotherapy have been recommended in the general
management of the patient with psychosis. The patients benefit from supportive psy-
chotherapy, especially when combined with social casework, rehabilitation measures,
and public health nursing. Family therapy and psychoeducation may be beneficial for
the patients and their families. Group psychotherapy may also be helpful, especially
with a supportive approach.
F. Insomnia
1. Pharmacotherapy
Sedative-hypnotic agents remain the primary treatment for insomnia associated
with HIV disease. Short-to-intermediate-acting benzodiazepines (lorazepam, te-
mazepam, and oxazepam) without active metabolites are the preferred treatments.
When insomnia is severe and refractory to treatment with these benzodiazepines,
trazodone is an effective alternative to hypnotic agents or as adjunctive treatment in
combination with the benzodiazepines. Other antidepressants can have undesirable
anticholinergic effects that may aggravate cognitive dysfunction and lower the seizure
threshold. Antihistamines, chloral hydrate, and other long-acting benzodiazepines
should be avoided. A nonbenzodiazepine hypnotic, zolpidem (5-10 mg/daily), has a
short half-life, has no metabolites, and appears safer than other hypnotics even in
cognitively impaired patients.
2. Nonbiological Treatments
Sleep hygiene rules should be recommended as first-line therapy for most pa-
tients. Similar strategies to those employed in the treatment of anxiety, such as relax-
ation training, self-hypnosis, and tapes with relaxation strategies, are also helpful.
11. HIV-RELATED PSYCHIATRIC DISORDERS 463
V. CONCLUSIONS
SELECTED READINGS
I, Adams E: Neuropsychiatric evaluation and treatment of delirium in the critically ill cancer patient.
Cancer Bull 36(3): 156-160, 1984.
2. Adle-Biassette H., Colombel L. M., Poron E, et al: Neuronal apoptosis in HIV infection in adults.
Neuropathol Appl NeurobioI21:218-227, 1995.
3. Alberts G. W., Saenz R. E., Moses J. A.: tolerability of oral dextromethorphan in patients with a history
of brain ischemia. Clin NeuropharmacoI15:509-514, 1992.
4. American Academy of Neurology: Working Group of the AAN and AIDS Task Force. Nomenclature
and research definition for neurologic manifestations of HIV-l infection. Neurology 41 :778-785, 1991.
5. Atkinson 1. H., Grant I., Kennedy C. J., et al: Prevalence of psychiatric disorders among men infected
with human immunodeficiency virus. Arch Gen Psychiatry 45:859-864, 1988.
6. Aylward E. H., Brettschneider P. D., McArthur J. c., et al: Magnetic resonance imaging measurement of
gray matter volume reductions in HIV dementia. Am J Psychiatry 152:987-994, 1995.
7. Beason-Hazen S., Nasrallah H. A., Bornstein R. A.: Self-report of symptoms and neuropsychological
performance in asymptomatic HIV-positive individuals. J Neuropsychiatry Clin Neurosci 6:43-49,
1994.
8. Becker 1. T., Caldrarao R., Lopez O. L., Dew M. A., Dorst S. K., Banks G.: Qualitative features of the
memory deficit associated with HIV infection and AIDS: Cross-validation of a discriminant function
classification scheme. J Clin Exp Neuropsychol 17: 134-142, 1995.
9. Berger J. R., Kaszovitz B., Post M. J., et al: Progressive multifocalleukoencephalopathy associated with
human immunodeficiency virus infection. A review of the literature with a report of sixteen cases. Ann
Intern Med 107:78-87, 1987.
10. Blaney N. T., Goodkin K., Morgan R. 0., et al: A stress moderator model in early HIV-I infection:
Concurrent analysis of life events, hardiness, and social support. J Psychosom Res 35:297-305, 1991,
1I, Bolan G.: Management of syphilis in HIV-infected persons, in Sande M. A., Volberding P. A. (eds): The
Medical Management of AIDS, 4th ed. Philadelphia, Saunders, 1995, pp 537-554.
464 FRANCISCO FERNANDEZ, M.D., AND JORGE MALDONADO, M.D.
12. Bornstein R. A., Nasrallah H. A., Para M. E, Whitacre C. C., Rosenberger P., Fass R. J.: Neuropsycho-
logical performance in symptomatic and asymptomatic HIV infection. Acquired Immune Defic Syndr
7:519-524, 1993.
13. Bredesen D. E., Levy R. M., Rosenblum M. L.: The neurology of human immunodeficiency virus
infection. Q J Med 68:665-667, 1988.
14. Breitbart w., Marotta R., Platt M., et al: A double-blind trial of haloperidol, chlorpromazine, and
lorazeparn in the treatment of delirium in hospitalized AIDS patients. Am J Psychiatry 153:231-237, 1996.
IS. Brew B. 1.: Central and peripheral nervous system abnormalities. Med Clin North Am 76(1):63-81,
1992.
16. Bridge T. P., Haseltine P. M. R., Parker E. S.: Initial pahse I testing of Peptide-T in AIDS patients and
normal controls. Lancet 1:226, 1989.
17. Brouillette M. J., Chouinard G., Lalonde R.: Didanosine-induced mania in HIV infection. Am J Psychia-
try 151: 1839-1840, 1994.
18. Brown 1. H., Henteleff P., Barakat S., et al: Is it normal for terminally ill patients to desire death. Am J
Psychiatry 143:208-211, 1986.
19. Cavanaugh S., Clark D. C., Gibbons R. D.: Diagnosing depression in the hospitalized medically ill.
Psychosomatics 24:809-815, 1983.
20. Centers for Disease Control: 1993 revised classification system for HIV infection and expanded surveil-
lance case definition for AIDS among adolescents and adults. MMWR 41[44-17]:1-19, 1992.
21. Codazzi E, Menegon A., Zacchetti D., et al: HIV-I gpl20 glycoprotein induces (Ca2+)j responses not
only in type-2 but also type-I astrocytes and oligodendrocytes of the rat cerebellum. Eur J Neurosci
7:1333-1341, 1995.
22. Drew W. L., Buhles w., Erlich K. S.: Management of herpes virus infection (CMV, HSV, VZV), in
Sande M. A., Volberding P. A. (eds): The Medical Management of AIDS. 4th ed. Philadelphia, W. B.
Saunders, 1995, pp 512-536.
23. Dreyer E. B., Kaiser P. K., Offermann 1. T.: HIV-I coat protein neurotoxicity prevented by calcium
channel antagonists. Science 248:364-367, 1990.
24. Anon: Drugs that cause psychiatric symptoms. The Medical Letter 35:901, 1993.
25. Empfield M., Cournos E, Meyer I., et al: HIV seroprevalence among homeless patients admitted to a
psychiatric inpatient unit. Am J Psychiatry 150:47-52, 1993.
26. Falangola M. E, Hanly A., Galvao-Castro B., Petito C. K.: HIV infection of human choroid plexus: A
possible mechanism of viral entry into the eNS. J Neuropathol Exp Neurol 54:497-503, 1995.
27. Fernandez D., Holmes V. E, Levy 1. K., Ruiz P.: Consultation-liaison psychiatry and HIV-related
disorders. Hosp Community Psychiatry 40:146-153, 1989.
28. Fernandez E, Levy 1. K.: Cognitive impairment due to AIDS-related complex and its response to
psychostimulants. Psychosomatics 291(1)38-46, 1988.
29. Fernandez E, Levy 1. K.: The use of molindone in the treatment of psychotic and delirious patients
infected with the human immunodeficiency virus: Case reports. Gen Hosp Psychiatry 15:31-35, 1993.
30. Fernandez E, Levy 1. K.: Psychopharmacotherapy of psychiatric syndromes in asymptomatic and
symptomatic HIV infection. Psychiatry Med 9(3):377-394, 1991.
31. Fishman B., Perry S., Jacobsberg L., et al: Psychological factors predicting distress after HIV testing.
Abstracts of the World Health Organization Fifth International Conference on AIDS, Montreal, Canada.
World Health Organizatin, 1989, p 737.
32. Goethe K. E., Mitchell 1. E., Marshall D. W., et al: Neuropsychological and neurological function of
human immunodeficiency virus seropositive asymptomatic individuals. Arch Neurol46: 129-133, 1989.
33. Grant I., Atkinson 1. H., Hesselink 1. R., Kennedy C. J., Richman D. D., Spector S. A., McCutchan 1. A.:
Evidence for early centran nervous system involvement in the acquired immunodeficiency syndrome
(AIDS) and other human immunodeficiency virus (HIV) infections. Studies with neuropsychological
testing and magnetic resonance imaging. Ann Int Med 107:828-836, 1987.
34. Gyorkey E, Melnick 1. L., Gyorkey P.: Human immunodeficiency virus in brain biopsies of patients with
AIDS and progressive encephalopathy. J Infect Dis 155:870-876, 1987.
35. Hill J. M., Farrar W. L., Pert C. B.: Autoradiographic localization of T4 antigen, the HIV receptor in
human brain. Int J Neurosci 32:687-693, 1987.
36. Holmes V. E, Fricchione G. L.: Hypomania in an AIDS patient receiving amitriptyline for neuropathic
pain. Neurology 39:305, 1989.
11. HIV-RELATED PSYCHIATRIC DISORDERS 46S
37. Jacobsberg L., Frances A, Perry S.: Axis II diagnosis among volunteers for HIV testing and counseling.
Am 1 Psychiatry 152:1222-1224, 1995.
38. Lipton S. A.: Menantine prevents HIV coat protein-induced neuronal injury in vitro. Neurology
42:1403-1405, 1992.
39. Lyketsos C. G., Hanson A L., Fishman M., Rosenblatt A, McHugh P. R., Treisman G. 1.: Manic
syndrome early and late in the course of HIV. Am 1 Psychiatry 150:326-327, 1993.
40. Maldonado 1. L., Fernandez J., Fernandez E: Depresion en el paciente con enfermedades medicas.
Medico Interamericano 1:353-357, 1996.
41. Martin J. L.: Psychological consequences of AIDS-related bereavement among gay men. 1 Consult Clin
PsychoI56:856-862, 1988.
42. Martin 1. L., Dean L.: Effects of AIDS-related and HIV-related illness on psychological distress among
gay men: A 7-year longitudinal study, 1985-1991. 1 Consult Clin PsychoI61:94-103, 1993.
43. Marzuk P. M., Tierney H., TardiffK.: Increased risk of suicide with AIDS. lAMA 259:1333-1337,1988.
44. Maxwell S., Scheftner W. A., Kessler H. A., Busch K.: Manic syndrome associated with zidovudine
treatment. lAMA 259:3406-3407, 1988.
45. McKegney E P., O'Dowd M. A: Suicidal ideation and HIV status. Am 1 Psychiatry 149:396-398, 1992.
46. Meyer I., Empfield M., Engel D., Coumos E: Characteristics of HIV-positive chronically mentally ill
inpatients. Psychiatr 0 66:201-207, 1995.
47. Nath A, Padua R A., Geiger 1. D.: HIV-I coat protein gpl20-induced increases in levels of intrasynap-
tosomal calcium. Brain Res 678:200-206, 1995.
48. Nichols S. E.: Psychosocial reactions of persons with the acquired immunodeficiency syndrome. Ann
Intern Med 103:765-767, 1985.
49. Nichols S. E.: Psychiatric aspects of AIDS. Psychosomatics 24:1083-1089, 1983.
50. O'Dowd M. A, McKegney E P.: Manic syndrome associated with zidovudine. lAMA 260:3587-3588,
1988.
51. Ostrow D., Grant I., Atkinson H.: Assessment and management of the AIDS patient with neuropsychiat-
ric disturbances. 1 Clin Psychiatry 49(Suppl):14-22, 1988.
52. Perry S., Jacobsberg L. B., Fishman B., Frances A., Bobo 1., Jacobsberg B. K.: Psychiatric diagnosis
before serological testing for the human immunodeficiency virus. Am 1 Psychiatry 147:89-93, 1990.
53. Perry S. w.: Organic mental disorders caused by HIV: Update on early diagnosis and treatment. Am 1
Psychiatry 147:696-710, 1990.
54. Perry S. w., Tross S.: Psychiatric problems of AIDS inpatients in the New York Hospital: Preliminary
report. Public Health Rep 99:200-205, 1984.
55. Pert C. B., Smith C. c., RuffM. R., Hill J. M: AIDS and its dementia as a neuropeptide disorder: Role of
VIP receptor blockade by human immunodeficiency virus envelop. Ann Neurol 23(Suppl):S71-S73, 1988.
56. Peters S., Koh J., Choi D. W.: Zinc selectively blocks the action of N-methyl-o-aspartate on cortical
neurons. Science 236:589-593, 1994.
57. Podraza A. M., Bomstein R. A., Whitacre C. C., et al: Neuropsychological performance and CD4levels
in HIV-I asymptomatic infection. 1 Clin Exp Neuropsychol 16:777-783, 1994.
58. Price R. w., Sidtis J. 1., Navia B. H., Pumarola-Sune T., Omitz D. B.: The AIDS dementia complex, in
Rosenblum M. L., Levy R. M., Bredesen D. E. (eds): AIDS and the Nervous System. New York, Raven
Press, 1988, pp 203-220.
59. Price R w., Worley J. M.: Management of neurologic complications of HIV-I infection and AIDS, in
Sande M. A., Volberding P. A. (eds): The Medical Management of AIDS 4th ed. Philadelphia, W. B.
Saunders, 1995, pp 261-288.
60. Riccio M., Pugh K., Jadresic D., et al: Neuropsychiatric aspects of HIV-I infection in gay men:
Controlled investigation of psychiatric, enuropsychological and neurological status. 1 Psychosom Res
37:819-830, 1993.
61. Rundell J. R, Kyle K. M., Brown G. R., Thomason 1. L.: Risk factors for suicide attempts in a human
immunodeficiency virus screening program. Psychosomatics 33:24-27, 1992.
62. Saag M. S.: Cryptococcosis and other fungal infections (histoplasmosis, coccidiodomycosis), in Sande
M. A., Volberding P. A. (eds): The Medical Management of AIDS. 4th ed. Philadelphia, W. B. Saunders.
1995, pp 437-459.
63. Schmitt E A., Bigley J. w., and McKinnis R., et al: Neuropsychological outcome of zidovudine (AZT)
treatment of patients with AIDS and AIDS related complex. N Engl 1 Med 319:1573-1578, 1988.
466 FRANCISCO FERNANDEZ, M.D., AND JORGE MALDONADO, M.D.
64. Seines O. A., Miller E. N., McArthur J. c., Gordon B., Munoz A., Sheridan K., Fox R., Saah A. 1.:
Multicenter AIDS cohort study. HIV-I infection: No evidence of cognitive decline during the asymptom-
atic stages. Neurology 40:204-208, 1990.
65. Seines O. A., Miller E. N.: Development of a screening battery for HIV-related cognitive impairment:
The MACS experience, in Grant I., Martin A. (eds): Neuropsychology of HIV Infection. New York,
Oxford University Press, 1994, pp 176-185.
66. Siegel K.: Psychosocial aspects of rational suicide. Am J Psychother 3:405-418, 1986.
67. Stem R. A .. Silva S. G., Chaisson N., Evans D. L.: Influence of cognitive reserve on neuropsychological
functioning in asymptomatic human immunodeficiency virus-I infection. Arch Neural 53:148-153,
1996.
68. Ushijima H., Nishio 0., Klocking R., et al: Exposure to gp 120 of HlV-I induces an increased release of
arachidonic acid in rat primary neuronal cell culture followed by NMDA receptor-mediated neurotox-
icity. Eur J Neurasci 7:1353-1359, 1995.
69. Van Gorp W. G., Miller E., Satz P., Visscher B.: Neuropsychological performance in HIV-I immu-
nocompromised patients. J Clin Exp Neuropsychol II :35, 1989.
70. Wang E, Vittinghoff E., Malani H., Reingold A., Lewis E., Giordano J., Jannsen R.: Incidence, propor-
tion of, and risk factors for AIDS patients diagnosed with HIV dementia. J AIDS Hum Retravirol 8:75-
82, 1995.
71. Wesley A. L., Gatchel R. 1., Polatin P. B., Kinney R. K., Mayer T. G.: Differentiation between somatic
and cognitive/affective components of depression in commonly used measurements of depression in
patients with chronic low-back pain. Let's not mix apples and oranges. Spine 16(Suppl):S213-S215,
1991.
72. Wiley C. A., Masliah E., Morey M., et al: Neocortical damage during HIV infection. Ann Neural
29:651-657, 1991.
73. Wilkie E, Eisdorfer c., Morgan R., Loewenstein D. A., Szapocznik 1.: Cognition in early human
immunodeficiency virus infection. Arch Neural 47:433-440, 1990.
74. Wilkie E L., Morgan R., Fletcher M. A. et al: Cognition and immune function in HIV-I infection.
Acquired Immune Defic Syndr 6:977-981, 1992.
75. Williams 1. B. w., Rabkin 1. G., Remien R. H., Gorman 1. M.: Multidisciplinary baseline assessment of
homosexual men with and without human immunodeficiency virus infection: II Standardized clinical
assessment and lifetime psychopathology. Arch Gen Psychiatry 48:124-130, 1991.
76. Yarchoan R., Berg G., Brouwers P., et al: Response of human immunodeficiency virus associated
neurological disease to 3' -azido-3' -deoxythymidine. Lancet I: 132- 135, 1987.
77. Zhang J., Dawson V. L., Dawson T. M., Synder S. H.: Nitric oxide activation of poly(ADP-ribose)
synthetase in neurotoxicity. Science 263:687-689, 1994.
78. Mellors J. w., Kingsley L. A., Rinaldo C. R. et al: Quantitation of HlV-I RNA in plasma predicts
outcome after seroconversion. Ann Intern Med 122: 573-579, 1995.
79. Volberding P. A.: HlV quantification: clinical applications. Lancet 347:71-72, 1996.
12
Eating Disorders
ANDREW W. BROTMAN, M.D.,
DAVID B. HERZOG, M.D., and
DAVID C. JIMERSON, M.D.
I. INTRODUCTION
The eating disorders discussed in this chapter include anorexia nervosa (AN), bulimia
nervosa (BN), and a new diagnostic entity called binge eating disorder (BED), which
appears to be prominent among the obese population. AN and BN have increased in
prevalence over the last 20 years and are characterized by disturbed eating behavior
and body image.
The prevalence of AN is between 0.4% and 1%1,2 whereas that of BN is between
I and 2%.3,4 The lifetime prevalence of these two eating disorders may approach 5%,5
making eating disorders a serious public health problem. AN and BN affect predomi-
nantly women, who comprise 90% of the eating-disordered population. BED has a
more balanced female-to-male ratio of 1.5: 1. The affected population for all eating
disorders crosses the lines of race, class, and educational status.
The etiology of the eating disorders is unknown, although biologic, sociocultural,
and psychological frameworks have been explored. Investigations of neurochemical
abnormalities have focused on neurotransmitters, hormones, and polypeptides. The
most robust evidence suggests that bulimics may have a low level of serotonin, causing
binging and perhaps explaining why antidepressant medication may be useful in treat-
ment. Sociocultural theories suggest that the cultural value of slimness has contributed
to the development of eating disorders. Impaired psychological development is fre-
Andrew W. Brotman, M.D. • Psychiatry, Beth Israel Deaconess Medical Center, Boston, Mas-
sachusetts 02215; Department of Psychiatry, Harvard Medical School, Boston, Massachusetts 02215.
David B. Herzog, M.D. • Eating Disorders Unit, Massachusetts General Hospital, Boston, Massachusetts
02114; Department of Psychiatry, Harvard Medical School, Boston, Massachusetts 02115.
David C. Jimerson, M.D. • Department of Psychiatry, Beth Israel Deaconess Medical Center, Boston,
Massachusetts 02215; Department of Psychiatry, Harvard Medical School, Boston, Massachusetts 02115.
The Practitioner's Guide to Psychoactive Drugs (Fourth Edition), Gelenberg and Bassuk, eds.
Plenum Publishing Corporation, New York, 1998
467
468 ANDREW W. BROTMAN, M.D., ET AL
quently viewed as the core problem in eating disorders, although there is little agree-
ment on a specific developmental problem. It is likely that the eating disorders are
heterogeneous in nature, and research will continue on several fronts.
The diagnostic criteria for these disorders from the fourth edition of the Diagnos-
tic and Statistical Manual of Mental Disorders (DSM-IV) are listed in Table 1.
A. Anorexia Nervosa
The anorectic usually seeks help reluctantly, often being asked by family, friends,
or employers to see a physician. She frequently denies any problems. Typically, she
describes herself or perceives herself as an overweight young woman who in her early
to mid-teens began a diet that escalated into an obsession. 6 She continued to lose
weight by severely restricting food intake, by exercising excessively, and/or by purg-
ing. She morbidly fears obesity and perceives herself as fat despite weight loss. An-
orectics who refuse to maintain a weight of at least 85% of expected body weight and
who either binge eat or purge should be diagnosed as "anorexia, binge eating/purging
type," whereas patients who do not binge eat or purge are classified as "anorexia
nervosa, restricting type." This disorder can begin quickly or insidiously and can be
self-limited, cyclical, or chronic. Morbidity and mortality rates for AN are among the
highest for psychiatric disorders. During the course of illness, more than half of these
patients can meet clinical criteria for depression. 7,8 One-third may have recurrent
affective illness, and one-fourth never regain menses or attain 75% of ideal body
weight. 9 A high prevalence of obsessive compUlsive disorder has been found in pa-
tients with anorexia nervosa,1O and this subtype may preferentially respond to serotonin
reuptake inhibitors.
The medical sequelae of anorexia nervosa are similar to those of chronic starva-
tion. Signs and symptoms include hypotension, bradycardia, amenorrhea, hypothermia,
anemia, growth failure, lanugo, and osteoporosis. I I Delayed gastric motility can pro-
duce abdominal pain, constipation, or stomach fullness.
Etiologic theories of AN include sociocultural explanations,12 psychological ex-
planations,13-17 and biological theories.I 8 ,19 Pharmacologic strategies are based on
potential alterations in central monoamine and peptide neurotransmitters. Persistent
changes in central norepinephrine and serotonin function have been noted even in
weight-recovered anorectics. 20,21
B. Bulimia Nervosa
Bulimia usually begins in late adolescence or young adulthood, when the patient
becomes aware that self-induced vomiting, laxative use, ipecac, or diuretics help to
cause weight loss or to maintain weight in the face of binging. Bulimics may binge on
12. EATING DISORDERS 469
huge quantities of food, usually in secret, followed by purging or restriction. The DSM-
IV includes criteria for the "purging type" and "nonpurging type," depending on
whether a patient engages or does not engage in self-induced vomiting or use of
laxatives or diuretics. 22 ,23
Unlike the anorectic, the bulimic may come for help in a distressed state and
ashamed of her symptoms. Bulimics usually look much better than they feel. They may
appear strong, sociable, and assertive but are frequently tormented by low self-esteem,
fear of intimacy, and aggression.
Bulimia can be self-limited but is usually chronic and episodic. Relapse rates
could be as high as 63%,24,25 but the rate ofresponse to treatment can also be greater
than 70%.26
The medical complications of BN are primarily related to purging and binging.
Bulimics may have electrolyte abnormalities, dental enamel erosion, parotidomegaly,
and abrasion of the knuckles from self-induced vomiting.27 About 40% of patients
have menstrual abnormalities (whereas almost all anorectics do), despite normal
weight. Cardiac arrhythmias, particularly when the patient is hypokalemic, can occur.
Chronic ipecac use can cause toxic myocardial dysfunction as a result of emetine
poisoning.
Etiologic factors include psychosocial influences, neurobiologic characteristics,
and a cultural emphasis on a particular body shape. 28 A dysregulation of neuro-
transmitters may exist in BN; serotonin and dopamine have been known to have a role
in satiety and appetite responses. 29 - 33 Neuropeptide dysregulation may also contribute
to BN.34,35
As in AN, more than half of patients with BN have a history of major depressive
episodes. 36 - 39 Anxiety and substance abuse disorder may also be more common
among patients with BN.4o A subgroup of bulimics may be very difficult to treat, with
several concurrent impulsive behaviors. 41
A. Outpatient Treatment
Outpatient treatment is the predominant treatment approach for all three disorders
and should be designed to meet the needs of the individual. An initial assessment
should ideally include a physical exam, psychiatric evaluation, nutritional history, and
perhaps a family evaluation. Some patients can be managed by a single clinician, but a
multidisciplinary team may often be indicated. For all patients, initial therapy should
include medical monitoring and behavioral treatment for weight recovery or control of
binging and purging.
Cognitive-behavioral therapies are common for the treatment of eating disorders,
interpersonal therapy has shown efficacy, and psychodynamic therapy is still widely
used. Many studies comparing cognitive therapy to pharmacotherapy have found simi-
lar outcomes. 45 Psychodynamic approaches are usually longer-term and consider eat-
ing disorder as the patient's compromise solution to unsolvable psychological conflicts.
These concerns can include a concurrent desire for autonomy and dependence and a
need to be in complete control of a body that refuses to cease having appetites.
Cognitive-behavioral treatments are usually-time limited and focus on directly
eradicating symptomatology. Short-term goal contracts, journal keeping, constructing
lists of alternative activities, time-delaying tactics, relaxation, meal planning, and
weight monitoring may all be employed. Treatment modalities are not mutually exclu-
sive so that an individual patient might be in a variety of treatments, including cog-
nitive-behavioral therapy, individual long-term therapy, group therapy, and phar-
macotherapy.
B. Inpatient Treatment
Although there are no specific guidelines for hospitalization, it can provide a
necessary bridge or catalyst to outpatient treatment. Inpatient treatment should be
viewed as an emergency measure and not a routine part of treatment. For example, the
decision to admit an anorectic patient might be based on the criteria listed in Table 2.
A. Anorexia Nervosa
1. Use of Medications in AN
Antidepressants, antipsychotics, and cyproheptadine (an antihistamine) are the
primary medications that have been used for AN in double-blind, placebo-controlled
trials. These trials are listed in Table 3.
Tricyclic antidepressants (TeAs) appear to be limited in their benefit for restora-
tion of low weight in patients with AN. The results of these studies suggest that
472 ANDREW W. BROTMAN, M.D., ET AL.
thymoleptic medication does not generally show superiority over placebo for the treat-
ment of AN. Statistical significance has not consistently translated into clinical signifi-
cance, and most of these studies were short-term, inpatient-based trials in which other
forms of nonpharmacologic treatment were the primary modalities being used. Long-
term effects of psychotropic medications with anorectics are not known, and, at this
point in time, pharmacologic treatment of the disorder is rather discouraging. Inter-
estingly, there have been no controlled trials of predominantly noradrenergic anti-
depressants, such as desipramine, for the treatment of AN.
Open studies with serotonin-selective reuptake inhibitors (SSRIs) such as fluox-
etine have begun to appear. In patients with AN who have recovered most of their
weight, fluoxetine appears to decrease the risk of relapse. In patients with active AN
who have not recovered, some case reports have found fluoxetine to be of help, but it is
unclear what role SSRIs have in the active treatment of AN. Their efficacy in ob-
sessive-compulsive disorder and in affective disorder may have implications for sub-
groups of anorectics who have comorbid disorders treatable with SSRIs.
Case reports using monoamine oxidase inhibitors (MAOIs) and opiate antagonists
have also appeared, but evidence for their efficacy in AN has not been established.
Cyproheptadine, an antihistamine and serotonin antagonist, has been used in
several studies, one of which demonstrated more rapid weight gain among hospitalized
patients who were treated with cyproheptadine as compared with those on placebo or
amitriptyline. Despite this finding, cyproheptadine is not widely used in clinical prac-
tice, and the doses required (up to 32 mg) can lead to sedation. Neuroleptic medication
has been used, especially in England during the 1960s and 1970s, but double-blind
studies have failed to show evidence of superiority over placebo. These results, com-
bined with neuroleptic toxicity, result in the infrequent use of these medications in AN.
Anorectics, by consensus, are not considered to be psychotic despite having attitudes
about food and body image that would be considered delusional. Antipsychotics do not
appear to correct the cognitive distortions inherent in patients with AN. Nonetheless,
some individuals are so anxious about eating that small doses of antianxiety agents or
12. EATING DISORDERS 473
Clomipramine 16 50 mg NSa 47
Amitriptyline 25 3 mg/kg NS 48
Amitriptyline 72 40-160 mg NS for amount of 49
weight gain,
weight gain
somewhat faster
when compared to
placebo
Cyproheptadine 72 32 mg More rapid weight 49
gain vs. placebo
Cyproheptadine 24 12 mg NS 50
Lithium 16 0.9-ng/mL Equivocal 51
level
Clonidine 4 NS 52
Metoclopramide NS 53
THC NS 54
Pimozide 18 4-6 mg NS 55
Sulpiride 18 300-400 mg NS 56
Open Studies
Fluoxetine May prevent relapse 51,58
MAOIs Case reports- 59
equivocal, side
effects
Opiate antagonists Case reports-
equivocal
even antipsychotics might be used prior to meal time as part of a course of behavioral
treatment.
Other medications including lithium, clonidine, metoclopramide, and Ll9-tetrahy-
drocannabinol (THC) have been studied, but clinical efficacy has not been demon-
strated for any of them.
2. Clinical Recommendations
Despite the absence of robust evidence, there is a place for the use of phar-
macologic agents in patients with AN. Generally, medication would be considered
adjunctive to psychotherapeutic treatment. Indications for medication include the pres-
ence of depression; prominent neurovegetative signs; severe anxiety, psychosis, or
obsessive-compulsive disorder; or profound treatment resistance. The use of an SSRI
or TCA is a reasonable first-line approach. The clinician should discuss possible side
effects with the patient and expect initial denial. Once the patient agrees, a medical
screening including complete blood count (CBC), electrolytes, BUN, electrocardio-
gram (EKG), and thyroid function should be done. All levels do not need to be
474 ANDREW W. BROTMAN, M.D., ET AL
completely normal, but none should be in a dangerous range. The clinician should
begin with the lowest dose of medication possible, titrate slowly upward, and obtain
blood levels. Positive outcomes include weight gain, increased interest in eating, de-
creased anxiety or depression, decreased obsessional thoughts, and increased willing-
ness and ability to attend to psychotherapeutic treatment. If the patient is treatment
resistant, the clinician might consider other agents such as cyproheptadine, antianxiety
agents, anti psychotics, MAOIs, and possibly electroconvulsive therapy (ECT).
3. Outcomes
A series of long-term outcome studies have been summarized by Hsu46 ; these
studies show a good outcome in 44% of patients, an intermediate outcome in 28%, a
poor outcome in 24%, and a mortality rate of less than 5%. Causes of death include
profound malnutrition, sudden death due to cardiac causes, and suicide. There is
considerable comorbidity with AN, and medication may have a long-term place in the
treatment of the comorbid disorders.
B. Bulimia Nervosa
1. Use of Medications in BN
Over the last 15 years, there have been numerous reports on the pharmacologic
treatment of BN, many of which have been controlled double-blind studies. Most of the
medications studied have been antidepressants, and the results are much more encour-
aging than those obtained in studies conducted for AN. A summary of controlled trials
can be found in Table 4.
The use of antidepressants in bulimia nervosa was prompted by a link between
eating disorders and affective illness. Bulimic patients have:
1. frequent current and previous symptoms of affective illness,
2. higher than expected current and previous episodes of major depressive disor-
der,
3. familial aggregation and genetic predisposition to affective disorder,
4. neuroendocrine abnormalities, similar to those found in affective probands,
and
5. a response to antidepressant medication.
Overall, antidepressant medication treatment of bulimics reduces binge frequency
by almost 60% compared to a decrease of just over 10% with placebo. It appears that a
reduction in purging follows this pattern although the response may be somewhat less
robust.
Generally, doses of antidepressant medication are similar to those used in depres-
sion. However, studies using the serotonin reuptake inhibitor fluoxetine have shown
that doses on the higher end of the spectrum (Le., 60 mg) seem to be more effective
than doses on the lower end of the spectrum (20 mg). Although formal studies of other
serotonin reuptake inhibitors have not been completed, this class of drug is generally
effective and well tolerated in BN.
12. EATING DISORDERS 475
TABLE 4. Percent Reduction in Binging for Patients on Active Drug and on Placebo
and Percent of Patients Not Binging (Abstinent) During Last Week in Study
Percent reduction in
binging activity Percent
abstinent in
Author(s)a Drug Daily dose Duration N Active drug Placebo study
Looking at the data, one can see that the SSRIs are not necessarily more effective
after a therapeutic trial but that the dropout rates in studies were much lower for
subjects taking SSRIs than for those taking TeAs or MAOIs. Despite the fact that there
seems to be a reduction in binging and purging at the end of these studies, the actual
percentage of patients who have a complete remission in bulimia hovers around 33%.
The long-term course of bulimia is uncertain, but studies to date are listed in Table 5.
There is a suggestion that patients treated with pharmacotherapy may do somewhat
better over the long term than those treated without medication, but these results are
preliminary. 26
476 ANDREW W. BROTMAN, M.D., ET AI..
Drugs not yet available in the United States include d-fenfluramine and selective
MAOIs; both have shown efficacy in double-blind studies. 60,61
The use of anticonvulsant medication has been equivocal and is not a first-line
treatment for BN. Serotonin agonists such as fenfluramine may be helpful, but there is
some controversy about their use because of regulatory problems in many states re-
garding the use of sympathomimetic amines for appetite control. Although bupropion
showed efficacy for BN in one study, several patients in this study had seizures and
therefore the medication is not approved by the U.S. Food and Drug Administration
(FDA) for eating disorders. Early studies of high-dose naltrexone showed some effi-
cacy, but the results of a double-blind study were not significant, suggesting that higher
doses may be necessary.
2. Clinical Recommendations
In approaching the pharmacologic treatment of BN, the presence of comorbid
depression is not necessary for an antidepressant to be used. In fact, there is some
suggestion that bulimics without depression may respond better to an antidepressant in
terms of their eating disorder than patients with comorbid depression. Nonetheless, a
trial of a nondrug treatment may be a reasonable first step, as studies of cognitive-
behavioral therapy and other approaches have shown efficacy. If medication is used, a
medical screening similar to that in AN should be done, with target symptoms being the
amount of binging, amount of vomiting, level of depression, and level of obsessions.
As with AN, we suggest combining pharmacotherapy with other nondrug treatments.
We would suggest the following approach to treatment of BN:
1. Begin treatment with an SSRI or a TCA, such as desipramine or imipramine.
2. Start with low doses of TCAs, titrate slowly, obtain blood levels, and dose the
medication at intervals to maximize the patient's ability to hold it down
without purging (usually before bedtime). SSRIs are best given in the
morning.
3. Titrate TCAs until therapeutic blood levels are maintained, and increase the
dose of SSRIs if there is no response (i.e., to fluoxetine, 60 mg; paroxetine, 60
mg; sertraline, 200 mg).
12. EATING DISORDERS 477
4. If a patient dose not respond by 6-8 weeks, consider switching the class of
antidepressant from SSRI to TeA or vice versa.
5. If a patient does not respond, consider using an MAOI. Inform the patient
thoroughly about the dietary restrictions, and withhold from severely impul-
sive patients or those who binge on tyramine-containing foods.
6. For those who are still treatment resistant, consider combination treatments
such as TeAs and SSRIs, possibly lithium augmentation of TeA or SSRI, and
fenfluramine or naltrexone if other agents have failed.
V. CONCLUSION
REFERENCES
I. Casper R. c.: The pathophysiology of anorexia nervosa and bulimia nervosa. Annu Rev Nutr 6:299-316,
1986.
2. Lucas A. R., Beard C. M., O'Fallon W. M., Kurland L. T.: 50-Year trends in the incidence of anorexia
nervosa in Rochester, MN: A population-based study. Am J Psychiatry 148:917-922, 1991.
3. Schotte D. E., Stunkard A. 1.: Bulimia vs bulimic behaviors on a college campus. JAMA 258: 12\3-1215,
1987.
4. Fairburn C. G., Beglin S. J.: Studies of the epidemiology of bulimia nervosa. Am J Psychiatry 147:401-
408,1990.
5. Pope H.g., Hudson J., Yurgium-Todd D.: Anorexia nervosa and bulimia among 300 women shoppers.
Am J Psychiatry 141:292-294, 1984.
6. Halmi K. A., Casper R. C., Eckert E. D., et al: Unique features associated with the age of onset of
anorexia nervosa. Psychiatr Res 1:209-215, 1979.
7. Cantwell D. P., Steuzenberger S., Burroughs J., Salkin B., Green J. K.: Anorexia nervosa: An affective
disorder? Arch Gen Psychiatry 34:1087-1093, 1977.
8. Gershon E. S., Schreiber J. L., Hamovit J. R., Dibble E. D., Kaye W., Nurnberger J. I., Jr., et al: Clinical
findings iii patients with anorexia nervosa and affective illness in their relatives. Am J Psychiatry
141:1419-1422, 1984.
9. Hsu L. K. G., Crisp A. H., Harding B.: Outcome of anorexia nervosa. Lancet 1:61-65, 1979.
10. Halmi K. A., Eckert E., Marchi P., Sampugnaro V., Apple R., Cohen J.: Comorbidity of psychiatric
diagnoses in anorexia nervosa. Arch Gen Psychiatry 48:712-718, 1991.
II. Brotman AW, Rigotti NA, Herzog DB. Medical complications of eating disorders. Compr Psychiatry
26:258-272, 1985.
12. Garner D. M., Garfinkel P. E., Schwartz D., et al: Cultural expectations of thinness in women. Psychol
Rep 47:483-491, 1980.
13. Bruch H.: Eating Disorders: Obesity, Anorexia Nervosa and the Person Within. New York, Basic Books,
1973.
14. Selvini Palazolli M.: Self-Starvation. London, Chaucer Publishing, 1974.
15. Geist R. A.: Psychotherapeutic dilemmas in the treatment of anorexia nervosa: A self-psychological
perspective. Contemp Psychother Rev 2(Fall), 1984.
16. Gilligan C.: In a Different Voice. Cambridge, Mass.: Harvard University Press, 1982.
17. Steiner-Adair C. The body politic: Normal female adolescent development and the development of
eating disorders. J Am Acad Psychoanal 14( I ):95-114, 1986.
18. Holland A. J., Sicotte N., Treasure J.: Anorexia nervosa: Evidence for a genetic basis. J Psychosom Res
32:561-571, 1988.
19. Strober M., Lampert C., Morrell W., Burroughs J., Jacobs C.: A controlled family study of anorexia
nervosa: Evidence of familial aggregation and lack of shared transmission with affective disorders. Int J
Eating Disord 9:239-253, 1990.
20. Kaye W. H., Jimerson D. C., Lake C. R., Ebert M. H.: Altered norepinephrine metabolism following
long-term weight recovery in patients with anorexia nervosa. Psychiatry Res 14:333-342, 1985.
21. Kaye W. H., Gwirtsman H. E., George D. T., Ebert M. H.: Altered serotonin activity in anorexia nervosa
after long term weight restoration: Does elevated cerebrospinal fluid 5-hydroxyindoleacetic acid level
correlate with rigid and obsessive behavior? Arch Gen Psychiatry 48:556-562, 1991.
22. America Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders. 4th ed. Wash-
ington, D.C., American Psychiatric Press, 1994.
23. Walsh B. T.: Diagnostic criteria for eating disorders in DSM IV; work in progress. Int J Eating Disord
11:301-304, 1992.
24. Herzog D. B., Keller M. B., Lavori P. W., Ott I. L.: Short-term prospective study of recovery in bulimia
nervosa. Psychiatry Res 23:45-55, 1987.
25. Keller M.B., Herzog D.B., Lavori P.W., Bradburn I.S., Mahoney E.M.: The naturalistic history of
bulimia nervosa: Extraordinary high rates of chronicity, recurrence, and psychosocial morbidity. Int J
Eating Disorders 12:1-9, 1992.
12. EATING DISORDERS 479
26. Herzog D. B., Sacks N. R: Bulimia nervosa: Comparison of treatment responders vs nonresponders.
Psychopharmacol Bull 29(1):121-125, 1993.
27. Herzog D. B., Copeland P. M.: Eating disorders. N Engl J Med 313:481-487, 1985.
28. Polivy J., Herman C. P.: Dieting and binging: A casual analysis. Am PsychoI40:193-20I, 1985.
29. Jimerson D. C., Lesem M. D., Hegg A. P., Brewerton T. D.: Serotonin in human eating disorders, in:
Whitaker-Aximita P. M., Peroutka S. 1. (eds): The Neuropharmacology of Serotonin, Annals of the New
York Academy of Sciences, vol 600. New York, New York Academy of Sciences, 1990, pp 532-544.
30. Jimerson D. c., Lesem M. D., Kaye W. H., Brewerton T. D.: Low serotonin and dopamine metabolite
concentrations in cerebrospinal fluid from bulimic patients with frequent binge episodes. Arch Gen
Psychiatry 49: 132-138, 1992.
31. Brewerton T. D., Mueller E. A., Lesem M. D., Brandt H. A., Quearry B., George D. T., et al: Neuroen-
docrine responses to m-chlorophenylpiperazine and L-tryptophan in bulimia. Arch Gen Psychiatry
49:852-861, 1992.
32. Kaye W. H., Ballenger 1. C., Lydiard R B., Stuart G. w., Laraia M. T., O'Neil P., et al: CSF monoamine
levels in normal-weight bulimia: Evidence for abnormal noradrenergic activity. Am J Psychiatry
147:225-229, 1990.
33. Jimerson D. C.: The role of central catecholamine pathways in eating disorders, in Ferrari E., Brambilla
F. (eds): Proceedings of the 2nd International Symposium on Disorders of Eating Behavior. Oxford,
Pergamon Press 1996, pp. 57-62.
34. Morley J. E., Levine A. S., Gosnell B. A., Mitchell J. E., Krahn D. D., Nizielski S. E.: Peptides and
feeding. Peptides 6: 181-192, 1985.
35. Geracioti T. D., Liddle R A.: Impaired cholecystokinin secretion in bulimia nervosa. N Engl J Med
319:683-688, 1988.
36. Hudson 1. I., Pope H. G., Jonas 1. M., Yurgelun-Todd D.: Phenomenologic relationship of eating
disorders to major affective disorder. Psychiatry Res 9:345-354, 1983.
37. Piran N., Kennedy S., Garfinkel P. E., Owens M.: Affective disturbance in eating disorders. J Nerv Ment
Dis 173:395-400, 1985.
38. Walsh B. T., Roose S. P., Glassman A. H., Gladis M., Saidk C.: Bulimia and depression. Psychosom Med
47: 123-130. 1985.
39. Jimerson, D. C., Lesem M. D., Kaye W. H., Hegg A. P., Brewerton T. D.: Eating disorders and
depression: Is there a serotonin connection? Bioi Psychiatry 28:443-454, 1990.
40. Hudson 1. I., Pope H. G. Jr., Yurgelun-Todd D., Jonas J. M., Frankenburg F. R: A controlled study of
lifetime prevalence of affective and other psychiatric disorders in bulimic outpatients. Am J Psychiatry
144:1283-1287,1987.
41. Fichter M. M., Quadflieg N., Rief w.: Course of multi-impulsive bulimia. Psychol Med 24:591-604,
1994.
42. Spitzer R. L., Devlin M., Walsh B. T., Hasin D., Wing R, Marcus M., et al: Binge eating disorder: A
multi-site field trial of the diagnostic criteria. Int J Eating Disord II: 191-203, 1992.
43. Hudson 1. I., Pope G. H., Wurtman J., Yurgelun-Todd D., Mark S., Rosenthal N. E.: Bulimia in obese
individuals: Relationship to normal-weight bulimia. J Nevr Ment Dis 176:144-152, 1988.
44. Devlin M. 1., Walsh B. T., Spitzer R L., Hasin D.: Is there another binge eating disorder? A reviewof the
literature on overeating in the absence of bulimia nervosa. J Eating Disord II :333-340, 1992.
45. Fairburn C. G., Jones R., Peveler R. C., Carr S. 1., Solomon R. A., O'Connor M. E., et al: Three
psychological treatments for bulimia nervosa: A comparative trial. Arch Gen Psychiatry 48:463-469,
1991.
46. Hsu L. K. G.: The outcome of anorexia nervosa: A reappraisal [Editorial], Psychol Med 18:807-812,
1988.
47. Lacey J. H., Crisp A. H.: Hunger, food intake and weight, the impact of clomipramine on a refeeding
anorexia nervosa population. Postgrad Med J 56:79-85, 1981.
48. Biederman J., Herzog D. B., Rivinus T.: Amitriptyline in anorexia nervosa: A double blind study. J Clin
Psychopharmacol 5: 10-16, 1985.
49. Halmi K. A., Eckert E., LaDu T. H., Cohen J.: Anorexia nervosa: Treatment efficacy of cyproheptadine
and amitriptyline. Arch Gen Psychiatry 43:177-181, 1986.
480 ANDREW W. BROTMAN, M.D., ET AL
50. Vigersky R. A, Loriaux D. L.: The effect of cyproheptadine in anorexia nervosa: A double blind trial, in
Vigersky RA (ed): Anorexia Nervosa. New York, Raven Press, 1977, pp 349-356.
51. Gross H. A, Ebert M. H., Faden V. B., et al: A double blind controlled trial of lithium carbonate in
primary anorexia nervosa. J Clin Psychopharmacol 1:376-381, 1981.
52. Casper R. c., Schlemmer R. E Jr., Javaid J. I.: A placebo controlled crossover study of oral clonidine in
acute anorexia nervosa. Psychiatry Res 20:249-260, 1987.
53. Moldofsky H., Jeuniewic N., Garfinkel P. E.: Preliminary report on metoclopramide in anorexia nervosa,
in Vigersky RA (ed): Anorexia Nervosa. New York, Raven Press, 1977, pp 373-376.
54. Gross H., Ebert M. H., Faden V. B., Goldberg S. C., Kaye W. H., Caine E. D., et al: A double blind trial
of delta-tetrahydrocannabinol in primary anorexia nervosa. J Clin PsychopharmacoI3:165-171, 1983.
55. Vandereycken W., Pierloot R.: Pimozide combined with behavior therapy in the short term treatment of
anorexia nervosa: A double blind, placebo-controlled, cross-over study. Acta Psychiatr Scand 66:445-
460,1982.
56. Vandereycken W.: Neuroleptics in the short term treatment of anorexia nervosa: A double blind placebo
controlled study with sulpiride. Br J Psychiatry 144:288-292, 1984.
57. Gwirtsman H. E., Guze B. H., Yager J., Gainsley B.: Fluoxetine treatment of anorexia nervosa: An open
clinical trial. J Clin Psychiatry 51 :378-382, 1990.
58. Kaye W. H., Weltzin T. E., Hsu L. K. G., Bulik C. M.: An open trial of fluoxetine in patients with
anorexia nervosa. J Clin Psychiatry 52:464-471, 1991.
59. Hudson 1. I., Pope H. G., Jonas J. M., et al: Treatment of anorexia nervosa with antidepressants. J Clin
PsychopharmacoI5:17-22,1985.
60. Fahy T. A., Eisler I., Russell G. E: A placebo controlled trial of d-fenfluramine in bulimia nervosa. Br J
Psychiatry 162:597-603, 1993.
61. Kennedy S. H., Goldbloom D. S., Ralevski E., Davis C., Dsouza J. D., Lofchy 1.: Is there a role for
selective monoamine oxidase inhibitor therapy in bulimia nervosa? A placebo controlled trial of
brofaromine. J Clin Psychopharmacol 13:415-422, 1993.
62. Sabine E. T., Yonaie A., Forringten A. T., et al: Bulimia nervosa: A placebo controlled double blind trial
of mianserin. Br J Clin Pharmacol 16:1955-2025, 1983.
63. Pope H. G., Hudson 1. I., Jones J. M., et al: Bulimia treated with imipramine: A placebo controlled
double blind study. Am J Psychiatry 140:554-558, 1983.
64. Walsh B. T., Gladis M., Roose S. P., et al: Phenelzine vs. placebo in 50 patients with bulimia. J Clin
PsychopharmacoI9:254-259, 1989.
65. Mitchell 1. E., Grout R.: A placebo controlled double blind trial of amitriptyline in bulimia. J Clin
Psychopharmacol 4: 193-196, 1984.
66. Hughes, P. L., Wells L. A., Cunningham L. J., et al: Treating bulimia with desipramine: A placebo
controlled double blind study. Arch Gen Psychiatry 43: 182-186, 1985.
67. Agras W. S., Dorian 8., Kirkley B. J., et al: Imipramine in the treatment of bulimia: A double blind
controlled study. Int J Eating Disord 6:29-38, 1987.
68. Kaplan A S.: Anticonvulsant treatment of eating disorders, in Garfinkel P. E., Garner D. M. (eds): The
Role of Drug Treatments for Eating Disorders. New York, Brunner/Mazel, 1987, pp 181-192.
69. Blouin A G., et al: Treatment of bulimia with fenfluramine and desipramine. J Clin Psychopharmacol
8:261-269, 1988.
70. Barlow 1., Blouin 1. H., Blouin A., et al: Treatment of bulimia with desipramine: A double blind
crossover study. Can J Psychiatry 330:129-133, 1988.
71. Home R. L., Ferguson J. M., Pope H. G. Jr., et al: Treatment of bulimia with bupropion: A multi-center
controlled trial. J Clin Psychiatry 49:262-266, 1988.
72. Kennedy S. H., et al: A trial of isocarboxazid in the treatment of bulimia nervosa. J Clin Psychophar-
macoI8:391-396, 1988.
73. Mitchell 1. E., Grout R.: A placebo controlled double blind trial of amitriptyline in bulimia. J Clin
PsychopharmacoI4(4):186-193, 1984.
74. Pope H. G., Keck P. E., et al: A placebo controlled study of trazodone in bulimia nervosa. J Clin
PsychopharmacoI9:254-259, 1989.
75. Mitchell 1. E., Pyle R. L., Eckert E. D., Hatsukami D., Pomeroy c., Zimmerman R.: A comparison study
12. EATING DISORDERS 481
of antidepressants and structured intensive group psychotherapy in the treatment of bulimia nervosa.
Arch Gen Psychiatry 47:149-157, 1990.
76. Walsh B. T., Hadigan C. M., Devlin M. J., Gladis M., Roose S. P.: Longterm outcome of antidepressant
treatment for bulimia nervosa. Am J Psychiatry 148:1206-1212, 1991.
77. Fichter M. M., Leibl K., Rief W, Brunner E., Schmidt-Auberger S., Engle R. R.: Fluoxetine versus
placebo: A double blind study with bulimic inpatients undergoing intensive psychotherapy. Pharmaco-
psychiatry 24(1): 1-7, 1991.
78. Fluoxetine Bulimia Study Group: Fluoxetine in the treatment of bulimia nervosa: A multi-centered
placebo controlled double blind trial. Arch Gen Psychiatry 49:139-147, 1992.
79. McCann U. D., Agras W. S.: Successful treatment of nonpurging bulimia nervosa with desipramine: A
double-blind, placebo controlled study. Am J Psychiatry 147:1509-1513, 1990.
80. Marcus M. D., Wing P. R., Weing L., Kern E., McDermott M., Gooding W: A double-blind, placebo
controlled trial of fluoxetine plus behavior modification in the treatment of obese binge-eaters and non-
binge eaters. Am J Psychiatry 147:876-881, 1990.
81. Abraham S. F., Mira M., Llewellyn-Jones D.: Bulimia: A study of outcome. Int J Eating Disord 2:175-
180, 1983.
82. Fichter N. M., Quadflieg N., Rief W: The German longitudinal bulimia nervosa study I, in Herzog W,
Deter H. C., Vandereycken W., (eds): The Course of Eating Disorders: Long-Time Follow-up Studies of
Anorexia Nervosa and Bulimia Nervosa. Springer-Verlag, Berlin, 1992, pp 591-604.
83. Herzog D. B., Keller M. B., Lavori P. W, Ott I. L.: Short term prospective study of recovery in bulimia.
Psychiatry Res 23:45-55, 1988.
84. Herzog D. B., Keller M. B., Lavori P. W.: Outcome in anorexia nervosa and bulimia nervosa: A review
of the literature. J Clin Psychiatry 53(5):147-152, 1992.
85. Hsu L.K.: Eating Disorders. New York, Guilford Press, 1990.
86. Lacey H.: Bulimia nervosa, binge eating, and psychogenic vomiting: A controlled treatment study and
long term outcome. Brit Med J 63:433-442, 1988.
87. Mitchell J. E., Davis L., Goff G., et al: A follow up study of patients with bulimia. Int J Eat Disord
5:441-450, 1986.
88. Pope H. G., Hudson J. I., Jonas J. M.: Antidepressant treatment of bulimia: Preliminary experience and
practical recommendations. J Clin PsychopharmacoI3(5):274-281, 1983.
89. Swift W J., Kalin N. H., Wamboldt F. S., Kaslow N., Ritholz M.: Depression in bulimia at 2 to 5 year
follow up. Psychiatry Res 16: Ill, 1985.
90. APA Practice Guidelines: Practice guidelines for eating disorders. Am J Psychiatry 150(2):209-228,
1993.
91. Jimerson D. c., Herzog D. B., and Brotman A. W: Pharmacologic approaches in the treatment of eating
disorders. Harvard Rev Psychiatry 1:82-93, 1993.
13
Borderline Disorders
I. INTRODUCTION
Paul H. Soloff, M.D. • Psychiatry, University of Pittsburgh Medical Center, Western Psychiatric Institute
and Clinic, Pittsburgh, Pennsylvania 15213-2593.
The Practitioner's Guide to Psychoactive Drugs (Fourth Edition), Gelenberg and Bassuk, eds.
Plenum Publishing Corporation, New York, 1998
483
484 PAUL H. SOLOFF, M.D.
A. Psychobiologic Models
The etiology of the borderline personality disorder is unknown but includes con-
tributions from both temperament and developmental experience. The modem psycho-
biologic perspective of the borderline disorder views the affective, cognitive, and
impulsive-behavioral dysregulation of the patient as originating in the biology of
temperament, Le., personality dimensions mediated by neurotransmitter function. 6
These biologic dimensions, both trait vulnerabilities and state symptoms, are the true
targets of pharmacotherapy in personality disorders. The psychobiologic perspective of
personality dimensions offers the practicing clinician both a theoretical and on empiri-
cal rationale for prescription. For example, the transient cognitive distortions experi-
enced by the borderline patient under stress may be viewed as state manifestations of
an underlying cognitive trait vulnerability, a dysregulation of dopaminergic neuro-
transmission and cognitive infonnation processing. The referential thinking, paranoid
ideation, perceptual distortions, and other "schizotypal" symptoms of the borderline
patient may share a common origin in dopaminergic neurotransmitter function with
other, more severe, manifestations of thought disorder and respond similarly to neuro-
leptic drugs in clinical trials.
The impulsive-behavioral dyscontrol of the borderline patient, manifested in re-
current destructive behaviors, both parasuicidal and assaultive in nature, has been
strongly related to diminished central serotonergic regulation. 7 Serotonergic function
has also been implicated in the affective dysregulation of the borderline patient, mani-
fested by brief but intense mood swings and rage attacks. The psychobiology suggests
pharmacological trials, which further enhance our knowledge of the biology of person-
ality dimensions through study of symptom responses. [For example, serotonin-selec-
tive reuptake inhibitors (SSRIs) reduce impulsive aggression.]
Other monoaminergic systems (especially norepinephrine and dopamine) may
also be involved in the regulation of mood in BPD, although their roles are less well
understood. Endogenous opioid systems have been implicated in the dissociative states
and stress-related analgesia that often accompanies self-mutilation in the borderline
patient. s
It is simplistic to assume that anyone neurotransmitter is responsible for complex
13. BORDERLINE DISORDERS 485
B. Psychosocial Models
While the dyscontrol states of borderline patients appear to be mediated by neuro-
transmitter function, the maladaptive interpersonal style and intrapsychic dynamics are
generally attributed to developmental influences,especially traumatic early life events.
Empirical studies suggest an etiologic role for early life separation from caretakers by
death and divorce,9 and histories of physical and/or sexual abuse.lO Some specific
symptoms of borderline patients, such as dissociation and habitual self-mutilation, are
strongly related to severe physical and sexual abuse in childhood.
A broader biosocial theory of etiology combines temperamental influences with
childhood experience. In addition to important biologic influences over affect and
impulse, the borderline patient's inability to modulate or cope with strong affect may
originate,in part, from an "invalidating environment" in childhood, an environment that
does not acknowledge the child's emotions as real or relevant in interaction with
caretakers.!! Expressions of strong feeling, especially needs contrary to the caretaker's
desires, are met with threats of rejection and instill a sense of shame and self-hatred.
The patient does not learn to label feelings, regulate arousal, contain or direct strong
affect, and tolerate distress. The resulting emotional vulnerability is seen in the adult's
uncontrolled expressions of anger and depression, in "rejection-sensitive dysphoria,"
and self-destructiveness. The borderline patient learns to use people as external sooth-
ing agents to help modulate strong emotion, leading to a maladaptive interpersonal
style. Many of the diagnostic criteria for borderline personality disorder may be under-
stood from this perspective.
Psychoanalytic descriptions of developmental defects in separation-individuation
in borderline patients deal with the same issues from a more intrapsychic perspective
and remain relevant for understanding the primitive defenses and interpersonal con-
flicts of these patients, especially in the treatment setting. For example, the borderline
patient's defensive "splitting" may have its origins in the biology of anger and, simul-
taneously, in the need to preserve a relationship with a potentially rejecting caretaker.
The patient's developmental deficits in "object constancy" in early life relationships
determine much of their adult interpersonal behavior. To understand the borderline
patient's fear of closeness (experienced as "engulfment") and anxieties at separation
(experienced as "abandonment") in the therapy relationship requires a psychodynamic
framework. Treatment of intrapsychic dynamics and resulting maladaptive interperson-
al relationships requires psychotherapeutic approaches.
The psychosocial and psychobiologic origins of personality dimensions are inter-
related. Chronic early life abuse may result in a posttraumatic stress syndrome which,
over time, distorts both the psychobiology and the developing personality.!2 Animal
studies suggest that neurotransmitter abnormalities observed in adults (monkeys, in this
486 PAUL H. SOLOFF, M.D.
A. Diagnostic Criteria
Table 1 lists diagnostic criteria for BPD derived from empirical studies and
arranged by content area. 16,19,20 (Criteria included in DSM-IV are noted below.)
13. BORDERLINE DISORDERS 487
aFrom Gunderson, Kolb, and Austin,16 Perry,19 and Sheehy, Goldsmith, and Charles,2o,
B. Differential Diagnosis
The differential diagnosis at the time of initial intake may run the full gamut of
axis I and II disorders. Because the symptoms of BPD overlap with its diagnostic "next
neighbors," and comorbidity with other disorders is common, differential diagnosis is
an unusually difficult exercise. Among the axis I affective disorders commonly mas-
querading as-or comorbid with-BPD are major depression and dysthymic disorder.
Lability of mood suggests a need to rule out cyclothymic and bipolar II disorders. At
the psychotic end of the spectrum, early or mild schizoaffective disorders and paranoid
disorders must be considered. Anxiety and panic disorders as well as specialized
symptom presentations, such as bulimia, drug and alcohol abuse, and sexual deviance,
are often comorbid features. A history of chronic physical or sexual abuse often
accompanies a clinical syndrome suggesting chronic posttraumatic stress disorder in
borderline patients, although autonomic reactivity is not prominent and avoidance has
been transferred into interpersonal dynamics. Half of borderline patients meet full
criteria for schizotypal personality disorder (DSM-III SPD). BPD generally overlaps
with the axis II cluster B disorders-histrionic, narcissistic, and antisocial personality
disorders.
Impulsive behavior in the borderline patient may have diverse etiologies. In
addition to being the result of impulsive temperament or behavioral dyscontrol as a trait
vulnerability (e.g., diminished serotonergic regulation), impulsivity in the borderline
patient may be the result of early-life CNS injury (e.g., postencephalitis), attention-
deficit hyperactivity disorder, or epileptiform disorders (e.g., cerebral dysrhythmia)
Careful electroencephalographic (EEG) and neuropsychological study may help re-
solve this differential. A neurobehavioral etiology for BPD has been proposed for such
patients but represents, at best, a small and atypical sample (e.g., male borderline
patients with attention-deficit hyperactivity disorder).21 In consecutively admitted
adult patients (predominantly women), EEG abnormalities are not more frequently
associated with BPD than with non-BPD controls.22 In most cases of impulsive behav-
ior, the etiology remains unknown.
major depression, i.e., much more favorable than had been assumed before longitudinal
systematic study.23 However, 3-9% of patients with BPD die by suicide. 24
BPD. Other trials are currently under way using interpersonal psychotherapy and
Beck's cognitive behavior therapy.
In contrast to supportive therapy, the intensive approach utilizes exploration of the
past and present to make comprehensive interpretations about the origins of the pa-
tient's character style. The therapist is generally neutral, providing the needed ambi-
guity to allow transference to develop. The method is nondirective and time-intensive,
with the patient being seen three or more times weekly. The goal is reconstruction of
character "from the inside out." Because some degree of regression is anticipated with
this approach, hospitalization may be part of the overall treatment plan and may be of
many months' duration. Limit setting is utilized primarily around the mechanics of
treatment (e.g., keeping appointments, paying bills, no violence in office) while a
greater degree of neutrality is maintained toward behavior in the community.
There are currently no reported systematic trials of analytic psychotherapy using
reliable, manualized methods or any systematic studies comparing its efficacy to that of
supportive approaches. Principles and methods of analytic psychotherapy of the bor-
derline patient have been presented by Kemberg et al. in preparation of a manualized
method. 28
2. Hospitalization
Leaving aside the question of duration of hospitalization, one may clearly define
indications for admission to a hospital. These include behaviors injurious to self or
others, especially when social controls are inadequate to prevent significant physical or
social consequences. Such impulsive-destructive behaviors can include overdosing,
self-mutilation, or assaultiveness or socially destructive behavior such as drug and
alcohol abuse or promiscuity. When such behavior is either seriously dangerous to self
or others or clearly out of control, admission to a hospital is warranted. Hospitalization
and inpatient care are important treatment modalities in their own right and have clear
therapeutic effects.
3. Special Risks
Overdose, abuse, and noncompliance with medication are special risks of phar-
macotherapy in borderline outpatients. Although these problems are not restricted to
this patient population, the borderline's penchant for using their medication to "act out"
conflicts makes these risks greater. The borderline inpatient may "act out" conflicts
through abusive or destructive behavior on the ward, "splitting" staff into "good"
13. BORDERLINE DISORDERS 491
and "bad," and testing the limits of the milieu. Because "splitting" of staff is an external
consequence of an internal psychodynamic process, it may be managed by effective
communication among staff (e.g., making the split apparent), setting a unified behav-
ioral standard for the patient, and also teaching the patient about the splitting process.
Behavioral regression should be dealt with quickly by setting firm but fair limits, even
including administrative discharge from the hospital for repeated intentional violations
of rules. Transfer from acute to longer-term care may be required if regression persists
and involves serious physical risks. In a long-term setting, a strict behavioral program
should be implemented in order to set limits and reverse regression (e.g., staff attention
should be strictly contingent on responsible behavior). "Fair" medication trials 4-6
weeks long may be systematically evaluated in this setting.
Borderline patients generally respond to firm limits and a collaborative approach
to pharmacotherapy. A supportive doctor-patient relationship is the sine qua non of
any pharmacotherapy, more so when the method of the treatment is empirical and the
goals are narrowly defined and modest.
Evaluation is often complicated by the patient's lack of cooperation. Clinicians are
often surprised by the entitled, demanding, and hostile presentation of a patient referred
acutely after a suicide attempt. Indeed, this "characterologic" presentation should
strongly suggest borderline disorder in a patient complaining of depression. The most
pressing immediate question is to determine the need for hospitalization. The patient's
impulsiveness, suicidal ideation, and motivation must be thoroughly explored. If social
support is available in the form of family and friends and the acute crisis appears
resolved, outpatient treatment is feasible. Borderline patients are often "well known to
the system." For such patients, suicidal or self-abusive behaviors may be experienced
often enough and be understood well enough to make it possible to entrust the patient
to family and friends. If the patient is not well known, social support is lacking, and
crisis intervention does not resolve self-destructive impulses, hospitalization is clearly
indicated.
IV. PHARMACOTHERAPY
A. Approach to Treatment
Research studies among inpatients suggest that some acute stress symptoms re-
solve in the hospital without any need to resort to medication. Group therapy, family
therapy, and the therapeutic milieu (the "placebo" treatments of most inpatient studies)
result in improvements in suspiciousness, paranoia, schizotypal symptoms, anger/
hostility, observed depression, and interpersonal sensitivity.30,31
Based on empirical data, a medication-free period of at least 7-10 days is strongly
recommended for the evaluation of borderline patients. In some studies, "placebo"
improvement is progressive for the first three weeks of an inpatient stayPO
The most difficult part of the differential diagnosis is ruling out a DSM-IV axis I
affective disorder. Depressed mood is ubiquitous in patients with borderline disorder.
Empirical study indicates that 50-70% of hospitalized borderline patients meet diag-
nostic criteria for major depression. 32,33 However, careful study indicates that the same
492 PAUL H. SOLOFF, M.D.
patients may meet criteria for atypical depressive disorder or hysteroid dysphoria when
these syndromes are systematically investigated. 34 One perspective [whether DSM-IV,
Schedule for Affective Disorders and Schizophrenia-Research Diagnostic Criteria
(SADS-RDC), or Hamilton Rating Scale for Depression] is often insufficient to charac-
terize accurately the borderline patient's depression. Some investigators have shown
that the diagnosis of major depression in patients with concomitant borderline disorder
does not "predict" suppression on the dexamethasone suppression test, blunting of
thyroid-stimulating hormone to thyrotropin-releasing hormone stimulation, or favor-
able clinical response to tricyclic antidepressants. 30,35,36 Many studies use cross-sec-
tional evaluation (by SADS-RDC, DSM-IV, etc.) based on brief interview and patient
report to establish the diagnosis of major depression. Observing the patient on the ward
will quickly reveal if the patient's mood is reactive or autonomous, the depression
labile or pervasive. In general, when a clearly diagnosed and treatable axis I disorder is
defined, it should be given priority and treated separately. For example, when major
depression is diagnosed, with clear endogenous, neurovegetative features, phar-
macotherapy must be directed toward the axis I disorder. Because other depressive
diagnoses-e.g., dysthymic disorder, adjustment disorder with depressed mood, and
chronic and intermittent depression-are insufficiently characterized and cannot be
easily discriminated from borderline pathology, pharmacotherapy of the axis II disor-
der should be considered, Indeed, some authorities argue that the borderline disorder is,
itself, an atypical affective disorder. 37 Patients with a clear diagnosis of BPD and
episodic alcohol abuse, food binges (bulimia), or anxiety attacks may also be consid-
ered candidates for pharmacotherapy of the axis II disorder, especially when these
symptom presentations are secondary to the borderline pathology and do not reflect
independent axis I disorders.
low-dose neuroleptic strategy is most often required in BPD patients with the schizoty-
pal symptom pattern. (Most will be DSM-IV BPD + SPD.)
Low-dose neuroleptics also have significant effects against severity of depressed
mood and behavioral impulsivity. The spectrum of response is broad, suggesting a lack
of specificity. Overall symptom severity and global functioning are improved in affec-
tive, cognitive, and impulsive symptom domains. Although effects are statistically
significant, clinical improvement is modest in magnitude. Patients change from se-
verely to moderately impaired. Treatment effects appear within 2 weeks. Duration of
treatment must be determined by the patient's needs but generally may be between 6
and 12 weeks. Prophylactic efficacy against symptom recurrence has not been estab-
lished by research.
Low-dose neutoleptics should be reserved for severe symptom presentations.
Patients with less severe symptom manifestations, especially those without schizotypal
symptoms or hostile, disruptive behavior, tend not to respond well to neuroleptics, and
the use of neuroleptics in this patient population may aggravate the clinical picture
through side effects.39 Studies have shown that borderline patients do not tolerate low-
dose neuroleptics well over prolonged continuation trials (up to 22 weeks) but report
significant increases in depressive symptoms related to psychomotor slowing. 40 The
clinician must weigh the risks of neuroleptic treatment (including the very small risk of
tardive dyskinesia) against the potential benefits in any prolonged continuation or
maintenance strategy. Continuation treatment should be reserved for the especially
vulnerable schizotypal patient. The danger of abuse or overdose of neuroleptic medica-
tion poses less medical risk than that of any other medication considered here. This
may be a deciding factor in choosing medication for the seriously impulsive outpatient.
The uncontrolled use of street drugs and alcohol by the borderline patient also poses
less medical danger of drug interaction with low-dose neuroleptic treatment. Typical
treatment protocols include haloperidol (Haldol® and others), 1-6 mg daily; thiothix-
ene (Navane® and others), 2-20 mg daily; or perphenazine (Trilafon® and others), 4-
16 mg daily.
The atypical neuroleptic clozapine may offer similar benefits with less risk of
extrapyramidal side effects, though at greater risk of neutropenia. 41 No double-blind,
comparative trials are currently available to judge the efficacy of clozapine compared
to that of less toxic neuroleptics. Newer atypical neuroleptics (e.g., risperidone) may
offer similar benefits with even less risk.
2. Antidepressants
As noted above, most depressed borderline patients demonstrate a mixed profile
of symptoms including both neurovegetative and atypical reactive patterns. If a melan-
cholic endogenous pattern can be clearly demonstrated, the treatment is directed to-
ward axis I major depression and may include tricyclic antidepressant medications.
However, tricyclic antidepressants (e.g., amitriptyline) have produced paradoxical be-
havioral toxicity in patients with BPD, including increased suicidal ideation, paranoid
thinking, and assaultiveness. 42 In the more common borderline presentations, e.g., the
globally depressed and hostile depressed patterns dominated by reactive labile moods
494 PAUL H. SOLOFF, M.D.
3. Anticonvulsants
Neuroleptics and antidepressants comprise the first line of treatment for BPD
because of safety and proven efficacy. Carbamazepine (Tegretol® and others) and
13. BORDERLINE DISORDERS 495
lithium (Eskalith® and others) represent useful second-line treatments but are limited in
use because of difficulties in management as well as a paucity of clear empirical
support for their effectiveness.
Based on one double-blind, crossover study, carbamazepine has been recom-
mended for impulsivity, behavioral dyscontrol, anger, suicidality, and anxiety.49,5o Its
efficacy as an anti manic agent has led to widespread use for mood lability in BPD.
Management is complicated by significant risk of bone marrow depression and the
need to follow hematological parameters as well as blood levels of the drug. Car-
bamazepine is given as 400-1200 mg daily, with dose guided by an optimal blood
level range of 4-10 IJ.g/mL. Limiting factors include sedation, ataxia, nausea, and
blurred vision. Carbamazepine has precipitated melancholic depression in borderline
patients with a history of this disorder. 51
Valproate (Depakote®) has also achieved clinical attention as a treatment for mood
lability in patients with BPD based on its efficacy in bipolar disorder. Although there
are no reported double-blind placebo-controlled studies, open label trials of valproate
in nondepressed borderline patients have demonstrated improvement in half of patients
on measures of mood, anxiety, anger, impulsivity, rejection sensitivity, irritability, and
general symptom severity. Valproate is given in 250-mg increments to achieve a plasma
level of 50-100 IJ.g/mL with an 8-week treatment trial. 52
Despite clinical popularity, empirical support for the efficacy of anticonvulsant
drugs in BPD remains to be firmly established. A clear demonstration of efficacy
would argue for a dimensional definition of mood lability across diagnoses such as
personality disorders, bipolar disorder, and cyclothymic disorder.
4. Lithium Carbonate
Lithium carbonate decreases mood lability in female patients with "emotionally
unstable personality disorder" and decreases impulsive aggression in criminal subjects
and juvenile delinquents. 53 - 56 It stabilizes mood, perhaps through a serotonergic mech-
anism. Empirical study in patients with BPD produced small effects against impul-
sivity.57 In spite ofthis inadequate data base, lithium is still widely prescribed for mood
lability and impulsive aggression in BPD. Daily doses are given in a range of 900-
1200 mg to achieve therapeutic blood levels, e.g., 0.8-1.2 mEq/L. Blood must be
sampled weekly until stability is achieved. Limiting side effects may be tremor, poly-
uria and polydipsia, and gastric irritation. Maintenance lithium therapy poses a risk of
hypothyroidism, which must be periodically assessed. As with carbamazepine, frequent
blood sampling may lead to patient resistance and noncompliance. The greatest draw-
back to using lithium in the borderline patient is the risk of overdose associated with a
drug that has such a narrow therapeutic-to-toxic range.
5. Benzodiazepines
The use of benzodiazepines is problematic in patients with BPD. Borderline
patients often suffer greatly from chronic anxieties ("pan-anxiety") as well as stress-
related exacerbations of anxious symptoms. However, the use of a short-acting ben-
zodiazepine (e.g., alprazolam) has been associated with precipitating serious dyscontrol
in one carefully conducted trial. 58 Abuse potential is significant, and tolerance becomes
496 PAUL H. SOLOFF, M.D.
V. CLINICAL CONSIDERATIONS/APPLICATIONS
Because individuals diagnosed with borderline disorders can exhibit such a wide
array of symptoms, what approach can the clinician adopt for developing the most
reliable medication strategies? Factor-analytic techniques have been used to reduce the
diversity of symptom presentations into a few recognizable clinical patterns. The
baseline symptoms for borderline patients can be organized into four clinical patterns:
1. Global/depression
2. Hostile/depression
3. Schizotypal symptom pattern
4. Impulsive symptom pattern
VI. CONCLUSION
REFERENCES
1. Mack J. E.: Borderline states: An historical perspective, in Mack 1. E. (ed): Borderline States in
Psychiatry. New York, Grune & Stratton, 1975, pp 1-28.
2. Stone, M. H.: The Borderline Syndromes. Constitution, Personality and Adaptation. New York,
McGraw-Hill. 1980, pp 1-33.
3. Swartz M., Blazer D., George L., et al: Estimating the prevalence of borderline personality disorder in
the community. J Pers Disorders 4:257-272, 1990.
4. Soloff P. H., Lis J. A., Kelly T., Cornelius J., Ulrich 1.: Risk factors of suicidal behavior in BPD. Am J
Psychiatry 151:1316-1323, 1994.
5. Soloff P. H., Lis 1. A., Kelly T., Cornelius 1., Ulrich R.: Self mutilation and suicidal behavior in
borderline personality disorder. J Pers Disorder 8:257-267, 1994.
6. Siever L. J., Klar H., Coccaro E.: Psychobiologic substrates of personality, in Klar H. and Siever LJ.
(eds.): Biologic Response Styles. Washington, D.C., American Psychiatry Press, 1985, pp 37-66.
7. Coccaro E. F., Siever L. 1., Klar H. M., et al: Serotonergic studies in patients with affective and
personality disorders: Correlates with suicidal and impulsive aggressive behavior. Arch Gen Psychiatry
46:587-599, 1989.
8. Russ M. 1.: Self injurious behavior in patients with borderline personality disorders. Biologic perspec-
tives. J Per Disord 6:64-81, 1992.
9. Soloff P. H., Millward 1. w.: Development histories of borderline patients. Compr Psychiatry 24:574-
588, 1983.
10. Herman J. L., van der Kolk B. A.: Traumatic antecedents of borderline personality disorder, in: van der
Kolk B. A., (ed): Psychological Trauma. Washington, D.C., American Psychiatric Press, 1987, p 116.
11. Linehan M. M.: Cognitive Behavioral Treatment of Borderline Personality Disorder. New York,
Guilford Press, 1993.
12. Gunderson J. G., Sabo A. N.: The phenomenological and conceptual interface between borderline
personality disorder and PTSD. Am J Psychiatry 150:19-27, 1993.
13. Higley 1. D., Suomi S. 1., Linnoila M.: CSF monoamine metabolite concentrations vary according to age,
rearing, and sex, and are influenced by the stressor of social separation in rhesus monkeys. Psychophar-
macology (Berl) 103:551-556, 1991.
14. Torgersen S.: Genetic and nosologic aspects of schizotypal and borderline personality disorders. Arch
Gen Psychiatry 41:546-554, 1984.
15. Silverman J. M., Pinkham L., Horvath T. B., Coccaro E. F., Klar H., Schear S., Apter S., Davidson M.,
Mohs R. C., Siever L. J.: Affective and impulsive personality disorder traits in the relatives of patients
with borderline personality disorder. Am J Psychiatry 148:1378-1385, 1991.
16. Gunderson 1. G., Ko1b J. E., Austin V.: The diagnostic interview for borderline patients (OIB). Am J
Psychiatry 138:896-903, 1981.
17. Zanarini M. C., Gunderson 1. G., Frankenburg F. R., Chauncey D. L.: The Revised Diagnostic Interview
for Borderlines (OIB-R): discriminating BPD from other axis II disorders. J Pers Disorders 3: 10-18,
1989.
18. American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders, 4th ed.
Washington D.C., American Psychiatric Press, 1994.
19. Perry 1.: The Borderline Personality Disorder Scale. Cambridge, Mass., Cambridge Hospital, 1982.
20. Sheehy M., Goldsmith L., Charles E.: A comparative study of borderline patients in a psychiatric
outpatient clinic. Am J Psychiatry 137:1374-1379, 1980.
21. Andrulonis P. A., Glueck B. c., Stroebel C. F., et al: Organic brain dysfunction and the borderline
syndrome. Psychiatr Clin North Am 4:61-66, 1981.
22. Cornelius 1. R., Brenner R. P., Soloff P. H., et al: EEG abnormalities in borderline personality disorder:
Specific or non-specific. Bioi Psychiatry 21 :974-977, 1986.
23. McGlashen T. H.: The Chestnut Lodge Follow-up Study III. Long term outcome of borderline person-
ality. Arch Gen Psychiatry 43:2-30, 1986.
24. Stone M. H.: The course of borderline personality disorder, in Tasman A., Hales R. E., Frances A. 1.
(eds): Review of Psychiatry 8. Washington, D.C., American Psychiatric Press, 1989, pp 103-122.
13. BORDERLINE DISORDERS 499
25. Linehan M. M.: Skills Training Manual for Treating Borderline Personality Disorder. New York,
Guilford Press, 1993.
26. Linehan M. M., Armstrong H. E., Suarez A., Allmon D., Heard H. L.: Cognitive-behavioral treatment of
chronically parasuicidal borderline patient. Arch Gen Psychiatry 48:1060-1064, 1991.
27. Linehan M. M., Heard H. L., Armstrong H. E.: Naturalistic follow-up of a behavioral treatment for
chronically parasuicidal borderline patients. Arch Gen Psychiatry 50:971-974, 1993.
28. Kernberg 0., Selzer M. A., Koenigsberg H. W, Carr A. c., Applebaum A.: Psychodynamic Psycho-
therapy of Borderline Patients. New York, Basic Books, 1989.
29. Kelly T., Soloff P. H., Cornelius 1., George A., Lis 1. A., Ulrich R.: Can we study (treat) borderline
patients. Attrition from research and open treatment. J Pers Disorders 6:417-433, 1992.
30. Soloff P. H., George A., Nathan R. S., Schulz P. M., Ulrich R. E, Perel J. M.: Progress in phar-
macotherapy of borderline disorders. Arch Gen Psychiatry 43:691-697, 1986.
31. Soloff P. H., Cornelius 1., George A., Nathan R S., Perel 1. M., Ulrich RE: Efficacy of phenelzine and
haloperidol in borderline personality disorder. Arch Gen Psychiatry 50:377-385, 1993.
32. Pope H. G., Jonas J. M., Hudson 1. 1., et al: The validity of DSM-III borderline personality disorder. Arch
Gen Psychiatry 40:23-30, 1983.
33. Baxter L., Edell W., Gerner R., et al: Dexamethasone suppression test and axis I diagnoses of inpatients
with DSM-III borderline disorder. J Clin Psychiatry 45: 150-153, 1984.
34. Soloff P. H., George A., Nathan R S., et al: Characterizing depression in borderline patients. J Clin
Psychiatry 48:155-157, 1987.
35. Soloff P. H., George A., Nathan R. S.: The dexamethasone suppression test in patients with borderline
personality disorder. Am J Psychiatry 139: 1621-1623, 1982.
36. Nathan R S., Soloff P. H., George A., et al: DST and TRH in borderline personality disorder, in Shagass
C., Josiassen R., Bridger W H., et al (eds): Biological Psychiatry, 1985. New York, Elsevier, 1986,
pp 563-565.
37. Akiskal H. S.: Subaffective disorders: Dysthymic, cyclothymic and bipolar II disorders in the "bor-
derline" realm. Psychiatr Clin North Am 4:25-60, 1981.
38. Liebowitz M. R., Klein D. E: Hysteroid dysphoria. Psychiatr Clin North Am 2:555-575, 1979.
39. Soloff P. H., Cornelius J. R, George A., Nathan R. S., Perel 1. M., Ulrich R E: Efficacy of phenelzine
and haloperidol in borderline personality disorder. Arch Gen Psychiatry 50:377-385, 1993.
40. Cornelius J. R., Soloff P. H., Perel 1. M., Ulrich R E: Continuation pharmacotherapy of borderline
personality disorder with haloperidol and phenelzine. Am J Psychiatry 150:1843-1848, 1993.
41. Frankenburg E R., Zanarini M. C.: Clozapine treatment of borderline patients: A preliminary study.
Compr Psychiatry 34:402-405, 1993.
42. Soloff P. H., George A., Nathan R S., Schulz P. M., Perel 1. M.: Paradoxical effects of amitriptyline in
borderline patients. Am J Psychiatry 143: 1603-1605, 1986.
43. Cornelius J. R, Soloff P. H., Perel J. M., Ulrich R. E: Fluoxetine trial in borderline personality disorder.
Psychopharmacol Bull 26: 149-152, 1990.
44. Norden M. J.: Fluoxetine in borderline personality disorder. Prog. Neuro-psychopharmacol. Biol Psychi-
atry 13:885-893, 1989.
45. Markovitz P. S., Calabrese J. R, Schulz S. C., Meltzer H. Y.: Fluoxetine in borderline and schizotypal
personality disorder. Am J Psychiatry 148: I 064-1067, 1991.
46. Salzman C., Wolfson A. N., Schatzberg A., Looper J., Henke R, Albanese M., Schwartz J., Miyawaki
E.: Effect of fluoxetine on anger in symptomatic volunteers with borderline personality disorder. Paper
presented at the American College of Neuropsycho-pharmacology, San Juan, Puerto Rico, Dec.14-18,
1992.
47. Kavoussi R. J., Liu J., Coccaro E. E: An open trial of sertraline in personality disordered patients with
impulsive aggression. J Clin Psychiatry 55:137-141,1994.
48. Parsons B., Quitkin EM., McGrath P. 1., Stewart 1. W, Tricamo E., Ocepek-Welikson K., Harrison W,
Rabkin 1. G., Wager S. G., Nunes E.: Phenelzine, imipramine and placebo in borderline patients meeting
criteria for atypical depression. Psychopharmacol Bull 25:524-534, 1989.
49. Cowdry R W., Gardner D. L.: Pharmacotherapy of borderline personality disorder: Alprazolam, car-
bamazepine, trifluoperazine and tranylcypromine. Arch Gen Psychiatry 45: 111-119, 1988.
500 PAUL H. SOLOFF, M.D.
50. Gardner D. L., Cowdry R W.: Positive effects of carbamazepine on behavioral dyscontrol in borderline
personality disorder. Am J Psychiatry 143:519-522, 1986.
51. Gardner D. L., Cowdry R. W.: Development of melancholia during carbamazepine treatment in bor-
derline personality disorder. J Clin Psychopharmacol 6:236-239, 1986.
52. Stein D. 1., Simeon D., Frenkel M., Islam M., Hollander E.: An open trial of valproate in borderline
personality disorder. J Clin Psychiatry 56:506-510, 1995.
53. Rifkin A., Levitan S. J., Galewski J., Klein D. F.: Emotionally unstable character disorder-a follow-up
study. Bioi Psychiatry 4:65-79, 1972.
54. Shader R. I., Jackson A. H., Dodes L. M.: The anti-aggressive effects of lithium in man. Psychophar-
rnalogia (Berl) 40:17-24, 1974.
55. Sheard M. H.: Lithium in the treatment of aggression. J Nerv Ment Dis 160:108-118, 1975.
56. Tupin J. P., Smith D. B., Clanon T. L., Kim L. I., Nugent A., Groupe A.: Long-term use of lithium in
aggressive disorders. Compr Psychiatry 14:311-317, 1973.
57. Links P. S., Steiner M., Boiagi I., Irwin D.: Lithium therapy for borderline patients: Preliminary findings.
J Pers Disorders 4: 173-181, 1990.
58. Gardner D. L., Cowdry R. W: Alprazolam-induced dyscontro1 in borderline personality disorder. Am J
Psychiatry 142:98-100, 1985.
59. Grinker R R, Sr., Werble B., Drye R. C.: The Borderline Syndrome. New York, Basic Books, 1965.
60. Spitzer R L., Endicott 1., Gibbon M.: Crossing the border into borderline personality and borderline
schizophrenia: The development of criteria. Arch Gen Psychiatry 36:17-24, 1979.
61. George A., Soloff P. H.: Schizotypal symptoms in patients with borderline personality disorders. Am J
Psychiatry 143:212-215, 1986.
62. Soloff P. H., Siever L., Cowdry R, Kocsis J. H.: Evaluation of pharmacologic treatments in personality
disorders, in Prien R F., Robinson D. S. (eds): Clinical Evaluation of Psychotropic Drugs. Raven Press,
New York, 1994, pp 651-674.
14
Medicolegal
Psychopharmacology
I. INTRODUCTION
Clinicians in all fields of medicine, including psychiatry, have become acutely con-
scious of the problems posed by the threat of liability. In psychopharmacology, courts
and attorneys are able to understand and see some of the concrete actions of the
pharmacological agents, in contrast to the relatively cloudy effects of psychotherapy,
thus increasing the likelihood of litigation. In this context, clinicians must understand
the medicolegal issues that surround, limit, and influence their practice of psychophar-
macology.
This chapter discusses three critical legal issues: the informed consent process, the
problem of liability, and the right to refuse treatment. Although this list is not exhaus-
tive, it encompasses the major medicolegal questions and difficulties confronting clini-
cians.
Treating anyone without his or her consent is a battery, which is both a crime and
a civil wrong. Treating someone without informed consent may be a form of malprac-
tice. Because the informed consent process is commonly misunderstood by clinicians,
this section attempts to clarify some of the confusion (see also Ref. 1).
Informed consent has three components: information, voluntarity, and compe-
tence. The information in question refers to what a patient would wish to know to make
Thomas G. Gutheil, M.D. • Program in Psychiatry and the Law, Massachusetts Mental Health Center,
Boston, Massachusesetts 02115.
The Practitioner's Guide to Psychoactive Drugs (Fourth Edition), Gelenberg and Bassuk, eds.
Plenum Publishing Corporation, New York, 1998
501
S02 THOMAS G. GUTHEIL, M.D.
a reasonable (Le., infonned) decision about the particular procedure. In the present
case, it refers to the use of medication in psychiatric treatment.
The legal definition of "infonnation" varies among jurisdictions and is based on
rulings in case law that establish how much infonnation should be given or how
explicit the infonnation should be. Despite this diversity, the following guidelines
concerning the information to be conveyed should be usable for almost all treatments:
• Risks and benefits of the proposed treatment
• Risks and benefits of alternative treatments
• Risks and benefits of no treatment (i.e., the consequences of not taking medica-
tion at all)
"Voluntarity" is the "consent" in infonned consent. This means that compliance
based on threat or coercion is not an example of infonned consent, nor are involuntary
emergency interventions, though they may be justified on other grounds. Perhaps most
importantly true, voluntary consent, by definition, can be revoked at any time. For
example, "living will" arrangements (in which a competent patient attempts to provide
for being treated over his/her own objection during an expected future state of incom-
petence) are notoriously unsuccessful in medication cases, because the patient's present
refusal revokes even a previously arranged consent.
Finally, competence represents the patient's capacity or ability to weigh and
process infonnation necessary to make a reasonable decision about treatment. Compe-
tence may be impaired by organic or functional illness. Also, minors are considered
incompetent by law in most areas.
This documents the patient's active participation in the consent process. It is not
necessary to write down the standard discussion verbatim, since it represents the usual
or customary one given by the clinician. A typical consent form, read to a jury, can
often sound stilted and unintelligible. By contrast, a physician describing his usual
patient information to the jury, and reading his notes of conversation with the plaintiff,
can be very persuasive.
III. LIABILITY
Liability is one of the major concerns facing the clinician who practices or special-
izes in psychopharmacology. Although a theoretical understanding of this issue is not
essential to the practitioner, the following brief review provides a background.
A. Definition
For malpractice to be present, the plaintiff must prove, by preponderance of the
evidence (i.e., that it is more likely than not), that four elements are present. These may
be mnemonically summarized as the "four Ds": duty, dereliction, damages, and direct
causation.
504 THOMAS G. GUTHEIL, M.D.
First, there must exist a duty, usually in the form of a doctor-patient relationship.
In simplest terms, this means that, based on the existence of the professional relation-
ship with that person, only your patient (or your patient's estate) can sue you. This duty
may exist under the circumstances of treatment, of consultation, and-at times-of an
<,>ffer of help.
Second, this duty must be breached by the clinician's dereliction: a deviation
from the standard medical care established by a professional peer group. The clinician
is usually measured against "the average prudent professional in that specialty," though
the wording may vary by jurisdiction. Such deviation represents negligence, the impu-
tation of a "sin of omission" in standard medical practice. The court is informed about
the standard of care by the testimony of expert witnesses for both sides.
Third, damages or harms must have resulted, in the form of actual death, injury,
pain, suffering, emotional distress, and the like.
Last, direct causation must be present. This indicates that the harms must be
directly related to the negligence. The law requires that the negligence be the "proxi-
mate cause" (immediate, direct precipitant) of the damages.
In sum, malpractice does not exist until the patient/plaintiff can prove, by prepon-
derance of the evidence, the four Ds: dereliction of a duty directly causing damages.
However, this technical, legalistic description, though constituting an accurate defini-
tion, does not accurately reflect the realities of malpractice litigation. Our experience
teaches that the actual occurrence of lawsuits for malpractice has little to do with the
standard of care or actual bad practice; instead, lawsuits are generally caused by bad
outcomes in the context of bad feelings.
Bad outcomes include death, injury, and disability. The bad feelings in question
may include guilt (in the survivors of suicide, hoping to alleviate their guilt by transfer-
ring some of the blame to the physician's negligence), rage (at the arrogant conduct of a
physician), grief (over the loss of a person or even the loss of an ideal of a perfect
outcome), and surprise (about an unexpected side effect).
One bad feeling that may be present in most malpractice suits is the feeling of
abandonment, of being left alone with one's calamity. Here the clinician's availabili-
ty-throughout the patient's illness and despite a bad outcome-is an essential ele-
ment in preventing litigation.
because a cause of action in this area is always plausible. The attorneys may then later
decide to drop this as a claim if warranted by the evolving case.
2. Misdiagnosis
As a medicolegal claim, this terms alludes to the allegation that the defendant
clinician failed to detect something that the average clinician would have detected (i.e.,
that a diagnosis was "missed"). Common examples include missing the patient's sui-
cidal state; missing an underlying medical condition that causes or worsens the present-
ing symptoms; and missing a medication side effect that goes undetected and causes
harm.
3. Inadequate Treatment
Inadequate treatment refers to cases in which the plaintiff alleges that the treat-
ment itself deviated from the standards of care. Typical examples include overtreat-
ment, undertreatment, and inappropriate treatment for the specific problem.
Although clinical judgment can and should prevail in these cases, clinicians
attempting more innovative and pioneering (more intrusive or, in some situations, less
aggressive) courses of treatment than usual, or using novel and experimental therapies,
are urged to obtain consultation and to document their own risk-benefit reasoning with
care. Competent patients can consent to necessary, novel, or even experimental treat-
ment as long as the full implications of such treatment are explained and understood.
Exceeding the doses for various medications given in the Physicians' Desk Reference
(PDR) occurs almost universally. In response, plaintiffs' attorneys sometimes suggest
that the PDR is equivalent to the standard of care, a ploy that commonly evokes
clinicians' anxiety. In actuality, the professional literature and standards of practice are
determinative. Thus, novel indications, nonmainstream prescribing of a drug, or unusu-
ally high dosages (contrasted with PDR limits) should be justified and supported by
consultation as needed. Citing specific literature in the record is worth its weight in
"liability-preventive gold."
5. Tardive Dyskinesia
The problems addressed in this section do not differ fundamentally from those
discussed above, but their particular form and common occurrence merit separate
consideration.
Work with tricyclic antidepressants takes place in the shadow of an inescapable
paradox. Tricyclics are among the most lethal medications used in modem psychiatry,
yet suicidal overdoses are most likely to occur when a patient is seriously depressed,
which is when tricyclics are most useful. This paradox cannot be resolved by actual
clinical practice. The possibility of plaintiffs' attorneys second-guessing such treat-
ment, however, has prompted clinicians to attempt certain strategies for dealing with
this problem beyond preferential use of serotonin-selective reuptake inhibitors (SSRIs).
A popular approach involves prescribing only a limited amount of medication for
the patient at any given time. For example, a clinician may prescribe only a week's
supply in an attempt to prevent the patient from accumulating a lethal total amount.
This approach may indeed be useful, especially when the patient tends to be impulsive,
but it is far from a panacea, as it has several limitations. First, a suicidal patient is under
no "contractual obligation" to use only these medications for suicide; they may be
mixed with other substances, or the patient may use completely different methods (e.g.,
hanging, jumping). Second, some patients view short-term prescribing as coercive, as a
sign of the physician's distrust of the patient, or as an approach calculated to protect the
doctor's interests but not necessarily the patient's. Third, both physician and patient
may be lulled into a false sense of security about the matter of suicide. Any patient can
save the weekly supply of medication or obtain more from other sources.
The author recommends that the physician acknowledge to the patient that these
14. MEDICOLEGAL PSYCHOPHARMACOLOGY 507
medications are not only potentially harmful but also tempting for depressed individu-
als. Asking the patient to agree not to overdose on these medications and to call the
physician if feeling impulsive may establish or restore a collaborative atmosphere of
open discussion about suicidal thoughts.
The decision to prescribe limited amounts of medication should be presented and
discussed with the patient with the intent of forming a therapeutic alliance. For exam-
ple, "You have told me that in the past you have been troubled by your sometimes
impulsive decisions; it would be sad if that led you to take too much medication before
it could finish helping you. Would you feel that you were more in control if I prescribed
only a small amount at a time, each week, that would be safe for you?" In this example,
the physician is allying with the part of the patient that wishes to retain control and to
resist impulsive action.
The modem psychopharmacologist treating a depressed patient must factor into
the prescriptive decision an analysis of the differential lethality of the various anti-
depressant agents.
1. Sharing Uncertainty
To the surprise of some clinicians, defensive practice is not only ineffective but
may also be provocative and lead to litigation. Defensive practice converts the patient
into an enemy against whom one "defends," and many patients can readily detect this
quality in the relationship. The patient feels, "This doctor is doing this for his own
interests, not for mine. This is self-protection, not patient care."
An alternative model recognizes the experience of the sick patient in relation to
the physician. Illness fosters regression, often expressed as the return to childhood
magical wishes for certainty and perfection in caregivers, wishes certain to be disap-
pointed. The physician must implicitly recognize this universal desire for a guaranteed
outcome and empathize with it yet gently counter it. This can be done by sharing
uncertainty.
The core problem facing both doctor and patient (and thus, potentially, uniting
them) is the uncertainty inherent in clinical practice. After all, when a patient takes a
medication for the first time, the physician is, in effect, performing an experiment; this
particular combination of person and drug is unique. A realistic appreciation of this
uncertainty reflects the doctor's awareness of unknowable dimensions of clinical care.
When this uncertainty is shared by doctor and patient, the uncertainty is "halved," as
many feelings are when shared. 2 This sharing also alerts us to another powerful factor
in decreasing liability risk-the therapeutic alliance.
508 THOMAS G. GUTHEIL, M.D.
The therapeutic alliance in this context has been described as "participant pre-
scribing," in which a patient is recruited into collaboration with the physician. 3 Even
with psychotic patients, trust may be decisive in ensuring a positive outcome. Most
importantly, the alliance reassures the patient against feared abandonment in the face of
a bad outcome.
3. Selection
A. Basic Issues
The core question posed under the rubric of the right to refuse treatment might be
summarized as: "Under what circumstances may patients refuse treatment (in the
present context, usually antipsychotic medication) prescribed by the physician respon-
14. MEDICOLEGAL PSYCHOPHARMACOLOGY 509
sible for their care?" Two aspects of the response to this question are emergencies and
vicarious consent.
1. Emergencies
Emergencies are the exception to many rules, this use among them. In emergency
situations, no court has taken the position that intervention is proscribed; in fact, all
known judicial holdings have made exceptions for emergencies. The emergency itself
is defined in various ways in different jurisdictions-imminent harm, urgent need,
"requires immediate intervention." One basic element of the emergency does not differ:
in emergencies, the clinician must act. To fail to do so out of fear of being second-
guessed by the law is empirically unrealistic and highly detrimental to the welfare of
patients. In any case of acute danger to self or others posed by a patient's condition, the
need for intervention reliably overcomes the right to refuse treatment.
2. Vicarious Consent
If competent patients can refuse treatment, and if no patient can refuse treatment
in a true emergency, the remaining "subject population" of the right to refuse treatment
issue should be incompetent, or questionably competent, nonemergency patients; in-
deed, this is the case. Thus, we are dealing with decision making for the incompetent
patient.
Historically, a guardian appointed by the court has been the individual most
frequently chosen to decide about treatment on behalf of the incompetent patient. l In
right-to-refuse litigation, however, the courts have chosen diverse individuals or groups
as the appropriate vicar of the decision; judge, guardian, clinical review panel, indepen-
dent psychiatrist, or treating psychiatrist. 4
This last point touches on the core issue underlying both the right to refuse
treatment and the right to treatment: the quality of care. 9 Much of the debate around the
principles of the right to refuse treatment masks issues related to the poor care often
encountered in state hospital systems. When care is fundamentally inadequate, the
questions of whether to offer or withhold treatment, to leave it accepted or refused,
become purely academic. The issue of quality of care of the mentally ill lies beyond the
reach of most doctors and patients; it is a matter for the legislature, for funding by
departments of mental health, and for patient advocates.
Some general guidelines can be extracted for the clinician's benefit. First, attempt
to deliver the highest quality of care. This apparently self-evident goal may involve, in
addition to individual practice, political actions such as boycotting of substandard
institutions, working with parent groups such as the Alliance for the Mentally Ill, and
acting through the American Psychiatric Association's district branches.
Second, note that treatment refusal is a clinical issue, whatever its legal implica-
tions.1O First, intervene clinically, using usual methods of exploration and discussion,
education, information, and permission. Legalizing the intervention may cloud its
clinical core.
Third, know the rulings in your jurisdiction. If there are no rulings, consider
designing regulations that are protective of patients' needs through legislative action by
the district branch of the American Psychiatric Association.
Fourth, carefully document the appropriate clinical evidence (e.g., presence of an
emergency, patients' incompetence) to offer the best likelihood that patients' needs will
prevail in the legal setting.
VI. CONCLUSION
REFERENCES
I. Appelbaum P. S., Gutheil T. G.: Clinical Handbook of Psychiatry and the Law, 2nd ed. Baltimore,
Williams and Wilkins, 1991.
2. Gutheil T. G., Bursztajn H., Brodsky A.: Malpractice prevention through the sharing of uncertainty:
Informed consent and the therapeutic alliance. N Engl J Med 311:49-51, 1984.
3. Gutheil T. G.: The psychology of psychopharmacology. Bull Menninger Clin 46:321-330, 1982.
14. MEDICOLEGAL PSYCHOPHARMACOLOGY Sl1
4. Gutheil T. G.: The right to refuse treatment: Paradox, pendulum and quality of care. Behav Sci Law
4:265-277, 1986.
5. For example, see Stensvad v. Reivitz (D.C. Wisc. 84-C-383-5, Jan. 10, 1985).
6. Youngberg v. Romeo 50 U.S.L.W. 4681 (June IS, 1982).
7. For example, Rivers v. Katz 495 N.E.2d 337 (N.Y., 1986).
8. Hoge S. K., Gutheil T. G., Kaplan E.: The right to refuse treatment under Rogers v. Commissioner:
Preliminary empirical findings and comparisons. Bull Am Acad Psychiatry Law 15:163-169, 1987.
9. Appelbaum P. S., Gutheil T. G.: The right to refuse treatment: The real issue is quality of care. Bull Am
Acad Psychiatry Law 9:199-202, 1981.
10. Appelbaum P. S., Gutheil T. G.: Clinical aspects of treatment refusal. Compr Psychiatry 23:560-566,
1982.
Index
513
514 INDEX
Anxiety states, 4; see also Agitation/excita- Behavioral therapy: see Interpersonal, cognitive-be-
tion/restlessness; Borderline disorders havioral, and psychotherapy
AIDS/HIV-related, 449-450, 458--459 Behavior problems, 4; see also Borderline disorders
with depression, 87-88 geriatric patients, 376--378
drug-responsiveness of, 7 lithium for, 118
geriatric patients, 379--380 pediatric patients, 387, 388, 389, 390, 391, 392,
sleep disorders with, 284 393,394,395,396
with tri- and heterocyclic antidepressants, Benadryl®: see Diphenhydramine
49--50 Benzisoxazoles, 165, 166, 169
Anxiolytics: see Antianxiety drugs Benzodiazepines, 4, 128, 185, 200, 227
Apomorphine, 158, 159 abuse/dependence disorders, 277, 284, 307-317,
Appetite: see Vegetative symptoms 319
Apresoline®, 81 for acute manic episode, 138
Aramine® (metaraminol), 189 adverse effects, 171, 172, 173, 237-241, 273-275
Arrhythmias, 47, 48, 62; see also Cardiac/cardio- for alcohol dependence, 322, 327
vascular effects for anxiety disorders, 219, 248, 250
Artane®: see Trihexylphenidyl adjustment disorders, 249--251
Artificial hibernation, 158 generalized, 261
Asendin®: see Amoxapine panic disorders, 251-253
Aspirin, 122 PTSD,260
Astemizole, 11,48,56,57,370 social phobia, 255-256
Atarax®: see Hydroxyzine for borderline disorders, 495-496
Atenolol, 245--246, 255 in breast milk, 431
Ativan®: see Lorazepam combination therapy/polypharmacy
Atromid-S® (clofibrate), 127 with antidepressants, 49--50
Atropine, 370, 371, 372 with anti psychotics, 196
Attention deficit-hyperactivity disorders pediatric patients, 406
(ADHD), 218, 386, 396--399 for depression, 81, 88
with bipolar disorders, 401-402 drug interactions, 241-242, 275, 276
pediatric patients, 387, 388, 389, 390, 391, with antidepressants, 56, 57
392,393,394,395,396 with carbamazepine, 130
with Tourette syndrome, 407 geriatric patients, 370
Atypical depressions, 88 extrapyramidal reaction management, 171, 172,
Autism, 218, 244, 411-412 173
Autonomic side effects: see Neurological effects geriatric patients, 368, 371, 372, 375, 379--380
Avoidant personality disorders, 218, 230 drug interactions, 370
for insomnia, 380--381
Baclofen (Lioresal®), 184 for HI V-related disorders, 458, 462
Barbiturates, 122, 128 hypomania, 141
abuse/dependence disorders, 309 pediatric patients, 393, 405, 406, 410-411
acute intoxication, 296 pharmacology, 229--242
for anxiety disorders, 229, 230, 242-243 adverse effects, 237-241
and carbamazepine toxicity, 130 drug interactions, 241-242
for ECT, 33 laboratory tests, 242
geriatric patients, 370, 380 mechanism of action, 231
management ofCNS depressant withdrawal, pharmacokinetics, 233--237
314--317 principles of use, 232-233
and sleep disorders, 269 pregnant patients, 417, 419, 425-426, 430, 432
for sleep disorders, 271, 281, 285--286 and sleep disorders, 269
tolerance and dependence, 308 for sleep disorders, 271, 272-278, 284, 285, 286
Basal ganglia disorders, 224 adverse effects, 273--275
Behavioral effects dependence, withdrawal, toxicity, 277
antidepressants, 52, 76 drug interactions, 275, 276
antipsychotics, 157-159 geriatric patients, 380--381
benzodiazepines, 274--275 Benztropine, 52, 171, 172, 175,395,421
INDEX 517
Calcium channel blockers. II. 127, 128, 133, Children: see Pediatric pharmacology
197,455 Chloral hydrate, 81
Capsulotomy, 229, 259 abuse of, 307, 309
Carbamazepine, II, 81, 128, 197, 240, 260 geriatric patients, 370
for bipolar disorders, 99, 104, 106, 132 for HIV-related disorders, 462
acute manic episode, 138, 140 for sleep disorders, 271, 281-282, 285--286
mixed or dysphoric manias, 144 Chloramphenicol, 370
pretreatment evaluation, 136 Chlorazepate, 231, 236, 308-309, 370, 379
rapid cycling, 144 Chlordiazepoxide, 4, 308
for borderline disorders, 494-495 for alcohol withdrawal, 322, 323
drug interactions, 127-129 for anxiety disorders, 231, 234, 236
with antidepressants, 57 phobias, 257
with valproate, 119, 121, 122 situational anxiety, 251
for eating disorders, 475 geriatric patients, 370, 379
geriatric patients, 370, 371, 372, 377, 384 Chlorgyline,71
for HIV-related disorders, 462 Chlorpromazine, 3-4, 38, 156, 158, 159, 166, 168,
maintenance therapy, 143 169,189,190-191,193,194.197,209
pediatric patients, 392, 402, 403-404 alcohol-induced psychosis, 324
pharmacology, 123-130 drug interactions, with zolpidem, 279
adverse reactions, 124-126 geriatric patients, 368, 370, 371, 372, 374
drug interactions, 127-129 for HIV-related disorders, 459
pharmacokinetics, 123-124 pediatric patients, 390, 408
in pregnancy, 130 Chlorpropamide (Diabinese®), 127
preparations and dosage, 130 Chlorprothixene, 163, 164, 166,371,372
toxicity. 129-130 Choledyl® (oxtriphylline), 114, 129
pregnant patients, 117, 425, 429, 432 Choline, 81,184
Carbidopa, 71, 332 Cigarette smoking: see Smoking/tobacco use
Carbonic anhydrase inhibitors. 113, 114 Cimetidine, 37.119,197, 128,275.276.370
Carbon monoxide poisoning, 33 Cingulotomy, 156,229,259
Cardiac arrhythmia suppression trial (CAST), 47 Circadian rhythm disruption, 269, 271
Cardiac/cardiovascular effects Cisapride (Propulsid®), 11,43,48.56
acute intoxication states: see Overdose. over- Clarithromycin (Biaxin®), 128, 276
dose management. toxicity Clinical practice, 6-9
antidepressants, 81 Clofibrate (Atromid-S®), 127
tri- and heterocyclic antidepressants, 45-49, Clomipramine,4, II, 12,37.38,39,44-45,50,55,
53.58-59,60-61,62 64
MAOI contraindications, 77 for anxiety disorders, 259
anti psychotics, 167, 188-190 OCD, 247, 258
benzodiazepines, 275 panic disorders, 246, 252
carbamazepine, 125, 127 drug interactions, 56
c1onidine, 299-300 combination therapy caveats, 91
in geriatric patients, 369, 370, 374-375 with MAOIs, 74, 75
lithium, 110-111 for eating disorders, 473
Cardiazem® (diltiazem), 12, 127, 128 pediatric patients, 388, 406, 408
Catapres®: see Clonidine pregnant patients, 431
Catastrophic stressors, 225--226 Clonazepam,76, 128, 183, 185
Catatonia, 118 abuse of, 309
Catecholamine hypothesis, 5 for acute manic episode, 138
Catecholamines: see Neurobiology for alcohol dependence, 327
Central nervous system: see CNS effects for anxiety disorders, 230, 231, 234, 236
Centrax®: see Prazepam panic disorders, 253
Cerebid® (papaverine), 377 social phobia, 256
Character pathology, 31, 90, 118 benzodiazepine withdrawal management, 240
Cheese effect, 37, 71, 72, 73 pediatric patients, 393, 405, 411
Chemoreceptor trigger zone, 159 for sleep disorders, 272, 273, 278, 285
INDEX 519