INT J TUBERC LUNG DIS 8(6):749–759

© 2004 IUATLD

Psychiatric issues in the management of patients

with multidrug-resistant tuberculosis

P. Vega,* A. Sweetland,† J. Acha,† H. Castillo,†‡ D. Guerra,† M. C. Smith Fawzi,§ S. Shin†¶

* Neuropsychiatric Service, Department of Medicine, Ministry of Health, Lima, † Socios En Salud, Lima, ‡ Department of
Health and Social Sciences, Universidad Peruana Cayetano Heredia, Lima, Peru; § Program in Infectious Disease and
Social Change, Department of Social Medicine, Harvard Medical School, Boston, ¶ Division of Social Medicine and
Health Inequalities, Brigham and Women’s Hospital, Boston, Massachusetts, USA

SUMMARY

I N T R O D U C T I O N : Psychiatric issues present a challenge
in the treatment of patients with multidrug-resistant
tuberculosis (MDR-TB). Both baseline psychiatric dis-
orders and development of psychiatric complications
related to anti-tuberculosis drugs and psychosocial fac-
tors require aggressive management.
S E T T I N G : A community-based non-governmental health
organization in Lima, Peru.
O B J E C T I V E : To review the literature for psychiatric com-
plications associated with anti-tuberculosis medications,
to describe the incidence and prevalence of depression,
anxiety and psychosis among individuals receiving MDR-
TB therapy, and to detail the management approach used
in this cohort.
M E T H O D S : A retrospective case series was performed
among the first 75 patients to receive individualized MDR-
TB therapy in Lima, Peru, between 1996 and 1999.
R E S U L T S : Baseline depression and baseline anxiety were
observed in respectively 52.2% and 8.7% of this cohort.
Most individuals with baseline depression experienced
improvement of depressive symptoms during the course
of TB therapy. The incidence of depression, anxiety and
psychosis during MDR-TB treatment was 13.3%, 12.0%
and 12.0%, respectively. While the majority of individu-
als with depression, anxiety and psychosis required psy-
chiatric pharmacotherapy, cycloserine was successfully
continued in all but one case.
C O N C L U S I O N : Psychiatric comorbidities are not a contra-
indication to MDR-TB therapy. Management of psychi-
atric complications is possible without compromising
anti-tuberculosis treatment.
K E Y W O R D S : multidrug-resistant tuberculosis; psycho-
sis; depression; anxiety; cycloserine; Peru; review

MORE THAN 50 YEARS after the advent of effec-
tive therapy, tuberculosis (TB) remains one of the
leading causes of adult deaths in the world, dispro-
portionately affecting people in poor countries.1 Indi-
viduals at highest risk for exposure and illness are
thus precisely those with the least means to overcome
the disease. Furthermore, drug-resistant strains are
contributing increasingly to a global public health
disaster. In a survey performed in 2000, multidrug-
resistant TB (MDR-TB) was identified in each of the
72 countries surveyed.2 Current estimates indicate
that 273 000 new MDR-TB cases occur each year.3 Al-
though recently removed from the list of high-burden
countries,4 Peru continues to have one of the highest
rates in the Americas, with 15% of reported cases,
despite representing only 3% of the population.5
By definition, MDR-TB refers to strains of Myco-
bacterium tuberculosis that are resistant to at least
isoniazid (INH) and rifampicin, the two most powerful
anti-tuberculosis agents. Second-line anti-tuberculosis
drugs, including cycloserine (CS), the fluoroquiono-
lones, ethionamide/prothionamide (ETH), kanamycin/
amikacin, capreomycin and para-aminosalicylic acid,
are generally weaker and more toxic than first-line
agents. For this reason, prolonged therapy (18–24
months) and frequent adverse reactions are significant
challenges to successful treatment of MDR-TB patients,
particularly in resource-poor settings. Management
of these side effects is complicated by the fact that
discontinuation of the drug or treatment interruption
may compromise treatment efficacy, especially in cases
of high-grade drug resistance. Rather than suspending
the suspected agent, which may be one of few thera-
peutic resources available to treat the infection, med-
ical providers are required to develop strategies to
aggressively manage the symptoms through secondary
medications and other biosocial interventions.
While adverse effects associated with MDR-TB

Correspondence to: Ms Annika Sweetland, Socios En Salud—Sucursal Perú, Avda Merino Reyna 575, Carabayllo, 06 Lima,
Peru. Tel: (51) 1-612-5200. Fax: (51) 1-547-2121. e-mail: asweetland@pih.org
Article submitted 22 July 2003. Final version accepted 23 November 2003.
750 The International Journal of Tuberculosis and Lung Disease
treatment may be managed effectively,6,7 certain side
effects require special attention. In particular, psychi-
atric complications, such as anxiety, depression, and
psychosis, can greatly impact patient quality of life, as
well as physicians’ attitudes toward MDR-TB ther-
apy. Successful control of psychiatric symptoms is
therefore crucial not only for favorable patient out-
come, but also for patients’ overall well-being and
physicians’ comfort with managing MDR-TB ther-
apy. The most commonly reported management strat-
egy to control psychiatric symptoms is to remove the
offending agent.8–10 However, a few published reports
describe management strategies that avoid the discon-
tinuation of the drug, for example lowering the dose
or simultaneously administering antidepressant or
antipsychotic therapy.11–13 The use of B6 supplemen-
tation to prevent neurotoxicity associated with ETH,
INH and CS has also been advocated. Similar protec-
tive effects may be observed with psychiatric symp-
toms induced by these medications.10,14–17
Psychiatric complications have been associated
with anti-tuberculosis therapy since the 1950s.18,19
Although significant psychiatric symptomatology has
been most commonly associated with CS and INH,
other drugs implicated at the case report level include
ETH, ethambutol (EMB) and the fluoroquinolones.
Severe psychiatric manifestations—including halluci-
nations, anxiety, depression, euphoria, behavioral dis-
orders, and suicidal ideation and/or attempts—have
been reported to occur in 9.7–50% of individuals
receiving CS.14,19–27 CS-associated neurotoxicity is
likely due to diminished central nervous system
(CNS) production of -aminobutyric acid (GABA)
caused by inhibition of glutamic decarboxylase and
facilitated by effective penetration of the blood-brain
barrier.28 In the majority of these cases, the drug was
discontinued, with rapid recovery of mental status and
no recurring symptoms. Psychiatric symptoms appear
most likely to present within the first 3 months of
treatment.22,24,26,29 Increased risk of CNS toxicity may
be associated with supratherapeutic levels of CS,30
concomitant use of ETH,31 INH,31 or fluoroquino-
lones,32 and ethanol ingestion.27
Less commonly, INH has been associated with
neuropsychiatric side effects, including depression, irri-
tability, obsessive-compulsive neurosis, and attempted
suicide.8–10,33–36 In the Boston Collaborative Drug
Surveillance Program performed in 1974, more than
35% of adverse effects associated with INH were psy-
chiatric in nature, with an incidence of 1.9%.37 Simi-
larly, in Peru, severe psychiatric syndromes associated
with INH occurred in approximately 1.0% of tuber-
culosis cases between 1991 and 1999.38 Pallone,
Goldman and Fuller performed a review of the litera-
ture on INH-induced psychosis in 1993.39 They found
that the most common psychiatric symptoms associ-
ated with INH were delusions, generally presenting
after approximately 4 weeks of taking the drug, and
among patients of an average age of 35 years (range
17–53). They summarize risk factors as receiving a
dose above 5 mg/kg; age 50 years or older; co-
morbid disease including diabetes mellitus, hepatic
insufficiency, alcoholism, and hyperthyroidism; and
past psychiatric history. Several mechanisms have
been proposed to account for INH-associated toxic-
ity. INH may act as an monamine oxidase (MAO)
inhibitor; alternatively, psychiatric effects may be
caused by INH-induced pyridoxine deficiency, with
subsequently diminished production of norepineph-
rine, serotonin, dopamine, and GABA.39 Further
review has uncovered several more case reports de-
tailing psychotoxic reactions to INH with similar
observations.13,40–43
Several case reports associate the use of ETH with
occurrence of depression, anxiety, psychosis, and sui-
cide.9,14,44–46 The mechanism for INH and ETH neuro-
toxicity is likely the same.37 EMB may also be associ-
ated with mania,47 confusion,15,31 and psychosis,48
although the mechanism for this effect is unknown.
Finally, the fluoroquinolones have been implicated in
rare occurrences of psychosis,49,50 depression,51 delir-
ium,52,53 and nightmares.54 A large-scale retrospective
study by Hollweg et al. reported psychiatric distur-
bance in 0.7% of 4189 individuals treated with either
ofloxacin or ciprofloxacin.55 In this study, while el-
derly individuals were more likely to experience delir-
ium or paranoia, younger individuals had greater
rates of depressive and manic syndromes. Underly-
ing hepatic or renal dysfunction, concomitant anti-
biotics or immunosuppressants, and baseline psychi-
atric disorders or psychosocial stressors, were identified
as risk factors for quinolone-associated psychiatric
disturbance.
In addition to drug toxicity, psychosocial factors
contribute to psychiatric complications during MDR-
TB therapy, and consequently patients’ adherence to
these regimens. Two early reports on sanatorium care
discuss several emotional aspects associated with TB,
including depression, anxiety, and suicidal ideation.56,57
Considering that these reports pre-date the develop-
ment of effective antibiotics for the treatment of TB,
it is apparent that psychiatric complications occur
even in the absence of drug therapy. Some of the psy-
chosocial issues that are often prominent concerns of
individuals with MDR-TB include: social stigma and
discrimination; fear and guilt associated with infec-
tious risk; the socio-economic and psychological bur-
dens of living with a chronic, life-threatening illness;
increased dependence on others; multiple treatment
failures and being told in health centers that no fur-
ther therapy was available; losing family members to
the disease; and concomitant poverty. The impact of
social stigma has been examined with regard to other
infectious diseases, such as the human immunodefi-
ciency virus and the acquired immune-deficiency syn-
drome (HIV/AIDS), the impact of which include

depression, low self esteem, and shame.58–61 Fear of
infection is one of the factors contributing to social
stigma, which may produce social isolation, dimin-
ished marriage prospects, limited social support, and
result therefore in the denial of diagnosis and conse-
quent rejection of treatment.62,63 Despite the fact that
TB infection is not necessarily associated with specific
‘risk behaviors’, TB patients are still generally held
responsible for their illness and blamed for not taking
better care of themselves.63 Further, poverty alone has
been demonstrated to have a clear association with
increased risk for mental illness.64 Several authors
have described how these psychosocial factors com-
plicate adherence to drug regimens, and emphasize
the importance of attention to mental health needs to
ensure positive treatment outcomes.62,63,65,66 Adher-
ence is especially important in the case of MDR-TB,
as this is often a patient’s last treatment option; fail-
ure to complete this treatment leads to a high rate of
fatality, in addition to ongoing transmission of highly
drug-resistant strains.
Despite such challenges, a community-based initia-
tive in Lima, Peru, has demonstrated successful out-
comes in treating MDR-TB patients,67 in part through
comprehensive management of psychiatric side effects
and psychosocial factors.63 One of the fundamental
components of this program has been the aggressive
management of adverse reactions using protocols and
community-based outreach to minimize the need for
treatment interruption or discontinuation of anti-
tuberculosis drugs.68,69 Here, we describe the inci-
dence, characteristics, and management of psychosis,
depression, and anxiety occurring among patients
receiving individualized treatment for MDR-TB in
Lima, Peru.
METHODS
Study population
All patients had documented MDR-TB and were
enrolled in treatment between 1 August 1996 and 31
January 1999 through collaborative effort between
two non-governmental organizations (Partners In
Health and Socios En Salud—Sucursal Perú) and the
Peruvian Ministry of Health’s National Tuberculosis
Program (NTP). The catchment area included three
districts of Northern Lima (Carabayllo, Comas and
Independencia), a shanty town hosting roughly 762 000
people in 1997 with high rates of poverty, violence,
unemployment, and mental illness.70–73 In addition,
through active case finding, this area was identified as
a ‘hot spot’ for the disease in Peru, with estimated TB
rates higher than the national incidence.74
Treatment protocol
All patients were referred by the NTP for suspicion of
MDR-TB based on history of treatment failure or
household contact with a known MDR-TB patient.
Psychiatric issues in MDR-TB therapy 751
Comorbidities, including psychiatric conditions, were
not a contraindication to treatment. Written informed
consent was obtained from all patients and one family
member prior to initiating therapy. Multidrug resis-
tance was confirmed by conventional or BACTEC
methods performed by the Massachusetts State Lab-
oratory Institute on sputum specimens.74 Individual-
ized therapy was based on each patient’s resistance
pattern. Of note, individuals with resistance to INH
only at low concentrations received high-dose INH (900
mg twice a week) as part of their regimens. Pyridoxine
(150–300 mg/day) was administered with MDR-TB
regimens. All patients participated in an unstructured
clinical interview with a psychiatrist to screen for base-
line psychiatric disorders prior to initiation of indi-
vidualized therapy. Patients received treatment under
the auspices of the Peruvian NTP: community-based
directly observed therapy delivered in health centers
and in patients’ houses by community health work-
ers.75,76 All treatment was provided to patients free of
charge.
Patients were evaluated at least once a month by
an NTP physician trained in the management of
MDR-TB. These evaluations included screening for
psychosis, depression, and anxiety. Patients were addi-
tionally assessed for any side effects identified by
health care workers. Patients with psychiatric symp-
toms were referred to a psychiatrist at the physician’s
discretion. The psychiatrist used clinical criteria based
on DSM-IV77 to diagnose psychiatric syndromes. In
general, symptoms were managed according to proto-
cols that were developed by the physicians trained at
Socios En Salud in the management of MDR-TB (see
Appendices 1 and 2). Anti-depressants, anxiolytics,
anti-psychotics, and psychotherapy were prescribed
by both primary MDR-TB physicians and psychia-
trists, and provided free of charge.
Study design and data collection
A case series with a retrospective chart review was
conducted among all patients who had initiated indi-
vidualized treatment between 1 August 1996 and 31
January 1999. All patients had completed treatment
at the time of chart review. The follow-up period for
all patients who were still alive was at least 18
months after completion of treatment. All charts were
reviewed by a physician trained in the management
of MDR-TB, then all cases with suspected new onset
of psychiatric disorders were further reviewed by the
treating psychiatrist. The case definition of psychosis
was based on DSM-IV criteria by a physician.77 Cate-
gorization of depression was based on DSM-IV criteria
by a psychiatrist, and included the following diag-
noses: major depressive disorder, dysthymia, adjust-
ment disorder with depressed mood, and substance-
induced mood disorder with depressive features.
The diagnosis of anxiety was also based on DSM-IV
criteria for generalized anxiety disorder or substance-
752 The International Journal of Tuberculosis and Lung Disease

induced anxiety disorder by a psychiatrist. Other vari- Table 1 Prevalence and incidence of psychosis, depression
ables extracted from the chart review for all patients and anxiety among MDR-TB patients
included age, sex, drugs in the individualized regimen, Prevalence Persistent Incidence
number of previous treatment regimens, treatment prior to symptoms during Prevalence
outcome, baseline psychiatric illness using DSM-IV MDR-TB among those MDR-TB during
therapy with baseline therapy MDR-TB
criteria according to a psychiatrist, history of alcohol (n  69) diagnosis (n  75) therapy
or drug abuse, hypothyroidism (before or during n (%) n (%) n (%) n (%)
treatment), HIV, diabetes mellitus, and malnutrition. Psychosis 0 (0) 0 (0) 9 (12.0) 9 (12.0)
The charts of all cases of psychiatric disorders were Depression 36 (52.2) 12/36 (33.3) 10 (13.3) 22 (29.3)
reviewed in more detail with respect to timing and Anxiety 6 (8.7) 3/6 (50.0) 9 (12.0) 12 (16.0)
duration of presentation of symptoms, clinical man- MDR-TB  multidrug-resistant tuberculosis (defined as resistance to at least
agement, and treatment outcome. isoniazid and rifampicin).

Statistical analysis
The following risk factors were analyzed for associa-
tion with incidence of depression, psychosis or anxi-
ety: sex; age; baseline depression; education level;
unemployment; marital status; presence of dependent
children; presence of TB household contacts; pres-
ence of household contacts who died of TB; hypothy-
roidism; presence of comorbidities (i.e., diabetes, HIV,
malnutrition, history of alcoholism or drug abuse);
history of institutionalization (i.e., incarceration or
extended hospitalization) or homelessness; receipt of
medications associated in the literature with psychiat-
ric side effects (i.e., CS, INH, EMB, ETH, fluoroqui-
nolones); and CS dose. Multivariable analysis for
occurrence of each side effect was also performed
using a multiple regression model including all vari-
ables associated with a P value 0.05 or an odds ratio
(OR) 2.0 on univariate analysis.
All data were entered in Microsoft Excel 97
(Microsoft Corporation, Seattle, WA, USA); all statis-
tical analysis was performed using SAS, version 8.2
(SAS Institute, Cary, NC, USA). Reported P values
were based on two-sided Fishers exact tests, or
for continuous variables, t-tests or, if non-parametric,
the Wilcoxon test. For binary variables, ORs with
95% confidence intervals (95%CIs) were also
calculated.
RESULTS
Baseline characteristics, incidence of psychiatric
complications, and treatment outcomes
Between August 1996 and January 1999, 75 patients
initiated treatment. Charts were available and reviewed
for all cases. The baseline prevalence, as well as inci-
dence during treatment of psychosis, depression and
anxiety, are summarized in Table 1. Of note, roughly
half of the cohort had a baseline diagnosis of depression,
including three individuals who also had a co-existing
anxiety disorder. None had baseline psychosis. Dur-
ing the course of MDR-TB therapy, 10 (13.3%)
patients developed depression, nine (12.0%) devel-
oped anxiety disorders, and nine (12%) experienced
psychosis. While psychosis tended to occur earlier in
MDR-TB treatment (mean 3.6 months, range 1–8),
the onset of depression and anxiety was more vari-
able. For both depression and anxiety, the mean time
of presentation was 7.3 months, with a range of 1–17
and 2–16 months, respectively.
Table 2 summarizes the demographic and clinical
characteristics of this cohort and has been described in
more detail elsewhere.67 MDR-TB regimens included
CS in all but one case, with a median daily dose of
1000 mg. Twenty-five patients (33.3%) received high
doses of INH, 52 (69.3%) received ETH, 11 (14.7%)
received EMB, and 72 (96.%) received a fluoroquino-
lone. While the occurrence of psychiatric complica-
tions during MDR-TB therapy was not significantly
associated with adverse treatment outcome, there
was a trend toward increased risk of default among
individuals who developed psychosis (OR  6.0,
95%CI 0.9–42.3).
Risk factors and management
of psychiatric complications
Psychosis
Among the nine patients (12%) who developed psy-
chosis during treatment, younger age (average 24.1 vs.
29.7 years among those with and those without psy-
chosis, P  0.008) was identified as a risk factor. Fur-
thermore, none of the individuals who developed psy-
chosis were married vs. 36.4% among those who did
not de-velop psychosis (P  0.05). In a multivariable
analysis, including female sex, age, baseline depres-
sion, unmarried status, number of dependent children,
and presence of an MDR-TB household contact, no
variable was found to be significantly associated with
psychosis. In three (33.3%) cases CS was suspended
temporarily, in two (22.2%) the CS dose was de-
creased, and in one (11.1%) the drug was suspended
then resumed at a lower dose (Table 3). It was never
necessary to discontinue CS indefinitely. No changes in
other TB drugs were required. Only one (11.1%)
patient was hospitalized due to psychosis, while the
rest were managed through close monitoring by com-
munity health promoters, nurses, and doctors in the
patients’ homes. All but one individual received anti-
psychotic drug therapy. Psychotic episodes recurred
in two (22.2%) individuals, but were effectively
 Psychiatric issues in MDR-TB therapy 753

Table 2 Baseline characteristics among 75 patients receiving individualized MDR-TB therapy

Characteristic n (%) Median (range)

Age, years 26.8 (11.8–65.1)
Sex
Male 37 (49.3)
Female 38 (50.7)
Education level (n  53)
Completed elementary school or less 7 (13.2)
Some high school or more 46 (86.8)
Household size (n  70) 7 (2–20)
Marital status
Married/living with partner 24 (34)
Unmarried (single, divorced, widowed, separated) 47 (66)
Occupation
Not working 12 (16.0)
Student 19 (25.3)
Homemaker 14 (18.7)
Worker, aside from professional 30 (40.0)
Professional 0 (0.0)
Number of dependent children 0 (0–7)
TB and MDR-TB household contacts 58 (77.3)
Household contacts who died of TB 28 (37.3)
Number of previous treatment regimens 3.0 (0–8)
Number of drugs to which patient is resistant 6.0 (2–12)
Comorbidity* 23 (30.7)
Hypothyroidism before or during ITR 11 (14.7)
Prior institutionalization or homelessness 9 (12.0)
Maximum cycloserine dose (mg) 1000 (750–1000)
Cycloserine 74 (98.7)
Isoniazid (high dose) 25 (33.3)
Ethionamide 52 (69.3)
Ethambutol 12 (16.0)
Fluoroquinolones 72 (96.0)

* Chronic disease included diabetes mellitus (1), HIV (1), malnutrition (15), and history of alcoholism or drug use (7).
MDR-TB  multidrug-resistant tuberculosis; TB  tuberculosis; ITR  individualized treatment regimen; HIV  human
immunodeficiency virus.

managed in both cases. While patients receiving anti-
psychotic medications at the conclusion of therapy
generally continued their use for an additional 30–45
days, in only one (11.1%) case was it necessary for the
patient to continue taking anti-psychotic medications
beyond this window. Psychosis resolved in eight
(88.9%) of the cases, with a median duration of psy-
chotic symptoms of 4 weeks (range 0.71–28). Two
(22.2%) patients who experienced psychosis aban-
doned treatment, one (11.1%) of whom was psychotic
at the time of default. Of note, three (33.3%) of these
cases occurred among sisters in one family.

Table 3 Characteristics and management of psychiatric symptoms (n  75)

Baseline Baseline
depression with anxiety with
persistent persistent
New psychosis New depression New anxiety symptomatology symptomatology
(n  9) (n  10) (n  9) (n  12) (n  3)
Management n (%) n (%) n (%) n (%) n (%)
Reduced dose of cycloserine 3 (33.3) 1 (10.0) 4 (44.4) 10 (8.3) 0 (0)
Temporary suspension of cycloserine 4 (44.4) 0 (0) 0 (0) 0 (0) 0 (0)
Terminated dose of cycloserine 0 (0) 0 (0) 0 (0) 0 (0) 0 (0)
Required psychiatric medications 9 (88.9) 8 (80) 8 (88.9) 6 (50) 3 (100)
Persistent psychiatric symptoms
throughout MDR-TB treatment
requiring psychiatric medications 3 (33.3) 1 (10.0) 3 (33.3) 2 (16.7) 0 (0)
Required hospitalization due to
psychiatric symptoms 1 (11.1) 0 (0) 0 (0) 0 (0) 0 (0)

MDR-TB  multidrug-resistant tuberculosis.

754 The International Journal of Tuberculosis and Lung Disease
Depression
Among the 36 (52.2%) patients with baseline depres-
sion, two thirds (66.7%) did not require psychiatric
attention or medications for depressive symptoms
during therapy due to mild symptomatology or remis-
sion. Ten (13.3%) patients without baseline symp-
tomatology experienced a new onset of depression
during TB therapy; no significantly associated risk
factor was identified on univariate analysis. In a
multivariable analysis including number of depen-
dent children, hypothyroidism, and presence of a TB
household contact, no variable was identified to be
significantly associated with depression. The dose of
CS was lowered in one (10.0%) new case for the
duration of treatment; no other modification of TB
therapy was made for other patients presenting with
depression during treatment. Anti-depressive medica-
tions were indicated in 80% of the new cases, although
symptoms were generally effectively managed and they
were not needed throughout treatment. Finally, no
patients in this cohort were hospitalized for depression.
Anxiety
Of the six (8.7%) patients with baseline anxiety, half
(50.0%) had persistent symptoms requiring psychiatric
intervention, and one (16.7%) experienced a manic
episode. Furthermore, nine (12%) patients had a new
onset of anxiety during treatment. Risk factors signif-
icantly associated with onset of anxiety during MDR-
TB treatment were having more dependent children
(P  0.007) and a higher average dose of CS (972 mg
vs. 884 mg among those with and those without anx-
iety, P  0.04). Eight of nine patients (88.9%) were
receiving CS at the maximum dose of 1000 mg. The
multivariable analysis included the following factors:
age, married status, number of dependent children,
CS dose, TB household contact, and lack of chronic
disease. Among these variables, only higher dose of CS
(P  0.01) and greater number of dependents (P 
0.002) were significantly associated with occurrence
of anxiety. In four (44.4%) cases, the dose of CS was
lowered for the duration of treatment; no other changes
in anti-tuberculosis therapy were made. While anxi-
olytic medications were necessary in eight (88.9%)
of the nine cases, only three (33.3%) required med-
ications for the duration of treatment. There were no
hospitalizations due to anxiety.
DISCUSSION
The baseline prevalence of depression observed in our
cohort was greater than the prevalence in the general
population of Lima, Peru (52.2% vs. 6.7%).71 On the
other hand, baseline rates of anxiety and psychosis
were comparable to those of the general population
of Lima.71 High rates of depression and anxiety
among tuberculosis patients have been reported else-
where,78 and are likely related to social stigma, inad-
equate social support, and the physiologic impact
of chronic disease.62,63,66,79 Interestingly, two thirds of
those with baseline depression actually experienced
an improvement in their depressive symptoms during
treatment, presumably as they recovered from tuber-
culosis and were able to resume important social roles
in their family and community.
Development of depression during treatment, ob-
served in 13.3% of our cohort, could be influenced by
similar psychosocial forces, as well as drug-related
factors. However, a clearer association can be observed
between anti-tuberculosis drugs and both anxiety and
psychosis: anxiety occurred more frequently among
those receiving higher doses of CS, and psychosis
responded to temporary suspension of CS and/or
reduction of CS dose. Nevertheless, the role of psy-
chosocial factors should not be underestimated. For
instance, having more dependent children was a risk
factor for anxiety, likely reflecting the stress of fulfill-
ing a caregiving role while sick. Indeed, the combined
influence of polypharmacy and psychosocial factors
highlights the need for integrated responses to psychi-
atric complications, including both pharmacological
and psychosocial components.
Three main strategies have been used to manage
psychiatric complications during MDR-TB therapy in
this program. First, community health workers have
been crucial in the identification, referral and man-
agement of psychiatric disorders. All community health
workers are trained to recognize early manifestations
of adverse reactions, including psychiatric symptoms.
Importantly, they provide direct supervision of psy-
chiatric medications, in addition to emotional sup-
port and counselling, for patients and family members.
The second strategy has been the use of psychiatric
medications in the majority of incident cases, often
managed by a psychiatrist. For psychosis, low-dose ris-
peridone has replaced haloperidol as the favored ther-
apy, due to fewer extrapyramidal symptoms associ-
ated with long-term use. For depression, serotonin-
uptake inhibitors, such as sertraline and fluoxetine, are
well tolerated and effective, with minimal pharmacolog-
ical interactions. Among anxiolytics, lorazepam offers
the advantage of a short half-life, while clonazepam
may be appropriate in individuals with concomitant
psychotic or depressive symptoms.31 Adequate psy-
chopharmacotherapy has allowed CS to be successfully
continued in all but one case, in which CS was sus-
pended only because an adequate regimen remained.
Finally, many patients are referred to a psycho-
social support group comprised of patients in treat-
ment and cured individuals. This methodology,
described elsewhere,63 has been effective in combat-
ing the psychosocial impact of the disease, including
marginalization and stigma, hopelessness and griev-
ing, as well as contemplation of suicide or treatment
default.
There are several limitations to this study. First, the

small sample size limits the analysis of risk factors
and effect of psychiatric complications on treatment
outcome. A larger cohort will likely be necessary to
identify risk factors associated with development of
depression and psychosis. Moreover, given the retro-
spective nature of this study, the incidence of psychi-
atric complications could be underestimated because
of lack of reporting or appropriate diagnosis.
CONCLUSIONS
Baseline prevalence of depression, anxiety, and psy-
chosis in our cohort was 52.5%, 8.7% and 0%,
respectively. Nonetheless, the presence of a baseline
psychiatric disorder was not a contraindication to
MDR-TB therapy nor to the use of CS. Furthermore,
individuals with initial depression and/or anxiety often
experienced an improvement in their symptoms dur-
ing the course of MDR-TB treatment. Depression,
anxiety and psychosis each occurred in approxi-
mately 12–13% of our cohort during treatment.
Aggressive use of psychiatric medications (in particu-
lar for psychosis), in addition to psychosocial sup-
port, has permitted successful resolution of symp-
toms, usually without the need for hospitalization.
Compared with other MDR-TB cohorts, our cohort
experienced higher or comparable rates of psychiatric
complications, and yet discontinuation of medications,
in particular CS, occurred less frequently.12,80–82 Thus,
psychiatric disorders among patients with MDR-TB can
be successfully managed without endangering effective
MDR-TB therapy.
Acknowledgements
The authors wish to thank the Bill & Melinda Gates Foundation,
as well as Thomas White, for their generous support. In addition,
we would like to thank the Peruvian Ministry of Health for their
collaboration, and the community health workers, whose valiant
efforts made this work possible. Finally, we wish to thank our
patients for the inspiration that they have given us to continue pro-
viding high quality TB care for patients living in poverty.
References
1 Benatar S R. Prospects for global health: lessons from tubercu-
losis. Thorax 1995; 50: 487–489.
2 Espinal M A, Simonsen L, Laszlo A, et al. for the World Health
Organization/International Union Against Tuberculosis and
Lung Disease Global Working Group on Anti-tuberculosis
Drug Resistance Surveillance. Anti-tuberculosis drug resistance
in the world. Report No. 2. WHO/TB/2000.278. Geneva, Swit-
zerland: WHO, 2000. http://www.who.int/gtb/publications/
drugresistance/ Accessed March 2004.
3 Dye C, Espinal M A, Watt C J, Mbiaga C, Williams B. World-
wide incidence of multidrug-resistant tuberculosis. J Infect Dis
2002; 185: 1197–1202.
4 World Health Organization. WHO Report 2002. Global
tuberculosis control: surveillance, planning, and financing.
WHO/CDS/TB/2002.295. Geneva, Switzerland: WHO, 2002.
5 World Health Organization. Health, a key to prosperity: suc-
cessful stories in developing countries. WHO/CDS/2000.4.
Geneva, Switzerland: WHO, 2002.
Psychiatric issues in MDR-TB therapy 755
6 Shin S S, Hyson A M, Castañeda C, et al. Peripheral neuro-
pathy associated with treatment for multidrug-resistant tuber-
culosis. Int J Tuberc Lung Dis 2003; 7: 347–353.
7 Furin J J, Mitnick C D, Shin S S, et al. Occurrence of serious
adverse effects in patients receiving community-based therapy
for multidrug-resistant tuberculosis. Int J Tuberc Lung Dis
2001; 5: 648–655.
8 Weidorn W S, Ervin F. Schizophrenic-like psychotic reactions
with administration of isoniazid. Arch Neurol Psychiat 1954;
72: 321–324.
9 Lees A W. Ethionamide and isoniazid in previously untreated
cases of pulmonary tuberculosis. Dis Chest 1964; 45: 247–
250.
10 Reilly D K. Isoniazid-related CNS toxicity. Drug Ther 1979; 9:
187–188.
11 Aceto J N, Covert D F. Observations on cycloserine-isoniazid
in pulmonary tuberculosis. Antibiot Med Clin Ther 1960; 7:
705–712.
12 Pasechnikov A D, Malinovskaya T Z, Kostornoy O S. Side
effects in multiple drug resistant TB treatment in Tomsk
Oblast. Moscow, Russian Federation: 17th Russian Congress
on Tuberculosis, 2003.
13 Duggal H S, Nizamie S H. Novel antipsychotic drugs and INH-
related psychosis. Aust N Z J Psychiatry 2000; 34: 343–344.
14 Johnson D A. Drug-induced psychiatric disorders. Drugs
1981; 22: 57–69.
15 Patel A M, McKeon J. Avoidance and management of adverse
reactions to antituberculosis drugs. Drug Safety 1995; 12: 1–
25.
16 Ministerio de Salud 1995. Actualización de la doctrina, nor-
mas y procedimientos para el control de la tuberculosis en el
Perú. Rev Soc Per Tis Neu E T 1996; 40: 19–41.
17 Ministerio de Salud. Actualización de la doctrina, normas y
procedimientos para el control de la tuberculosis en el Perú:
2001. Lima, Peru: Ministerio de Salud del Peru, 2001: pp 71–
77.
18 Weidorn W S, Ervin F. Schizophrenic-like psychotic reaction
with administration of isoniazid. Arch Neurol Psychiat 1954;
72: 321.
19 Lewis W C, Calden G, Thurslan J R, Gibson W E. Psychiatric
and neurological reaction to cycloserine in the treatment of
tuberculosis. Dis Chest 1957; 32: 172–177.
20 Riska N. Tolerance to cycloserine. Scand J Respir Dis 1970; 71
(Suppl): 209–216.
21 Stephanopoulos C, Zoumbouloglou H. Clinical tolerance to
cycloserine. Scand J Respir Dis 1970; 71 (Suppl): 235–238.
22 Dissmann E. Experience with cycloserine. Scand J Respir Dis
1970; 71 (Suppl): 239–243.
23 Bethlem N. Results of treatment of pulmonary tuberculosis
with cycloserine in association with other drugs. Scand J
Respir Dis 1970; 71 (Suppl): 244–249.
24 Helmy B. Side effects of cycloserine. Scand J Respir Dis 1970;
71 (Suppl): 220–225.
25 Pasargiklian M, Biondi L. Neurologic and behavioral reactions
of tuberculosis patients treated with cycloserine. Scand J Respir
Dis 1970; 71 (Suppl): 201–208.
26 Leston J M, Rey J C, Gonzalez Montaner L J, Grondona A,
Zavalla P N. Psychosomatic reactions to cycloserine in the
treatment of tuberculosis. Scand J Respir Dis 1970; 71 (Suppl):
231–234.
27 Nariman S. Adverse reactions to drugs used in the treatment of
tuberculosis. Adverse Drug React Acute Poisoning Rev 1988;
4: 207–227.
28 Berning S E, Peloquin C A. Antimycobacterial agents: Cyclo-
serine. In: Yu V L, Merigan J R, Barriere S L, eds. Antimicrobial
therapy and vaccines. Baltimore, MD: Williams & Wilkins,
1999: pp 638–642.
29 Snavely S R, Hodges G R. The neurotoxicity of antibacterial
agents. Ann Intern Med 1984; 101: 92–104.
756 The International Journal of Tuberculosis and Lung Disease
30 Sweetman S. Martindale—The complete drug reference. 33rd
ed. London, UK: Pharmaceutical Press, 2002.
31 Medical Economics Staff, PDR Staff, eds. Physicians’ Desk Refer-
ence 2003. 57th ed. Oradell, NJ: Medical Economics Co, 2003.
32 Yew W W, Wong C F, Wong P C, Lee J, Chau C H. Adverse
neurological reactions in patients with multidrug-resistant pul-
monary tuberculosis after coadministration of cycloserine and
ofloxacin. Clin Infect Dis 1993; 17: 289–290.
33 Goldman A L, Braman S S. Isoniazid: a review with emphasis
on adverse effects. Chest 1972; 62: 71–77.
34 Reilly D K. Isoniazid-related CNS toxicity. Drug therapy 1979;
9: 187–188.
35 Bhatia M S. Isonizid-induced obsessive compulsive neurosis. J
Clin Psych 1990; 51: 387.
36 Madan A, Jain N K, Sharma T N, Sharma D K, Madan R. At-
tempted suicide following treatment with isoniazid. Tubercle
1989; 70: 147–149.
37 Holdiness M R. Neurological manifestations and toxicities of
the antituberculosis drugs. Med Toxicol 1987; 2: 33–51.
38 Bonilla C, Portocarrero P G, Suárez P G, Molero D. Reac-
ciones adversas a fármacos antituberculosos (RAFA) en trata-
mientos directamente observados (DOTS). Peru 1991–1999.
Tuberculosis en el Perú: Informe 1999. Lima, Peru: Ministerio
de Salud del Perú, 2000: pp 99–106.
39 Pallone K A, Goldman M P, Fuller M A. Isoniazid-associated
psychoses: case report and review of the literature. Ann Phar-
macother 1993; 27: 167–169.
40 Ibrahim Z Y, Menke J J. Comment: Isoniazid-induced psycho-
sis. Ann Pharmacother 1994; 28: 1311.
41 Gnam W, Flint A, Goldbloom D. Isoniazid-induced hallucino-
sis: response to pyridoxine. Psychosomatics 1993; 34: 537–
539.
42 Alao A O, Yolles J C. Isoniazid-induced psychosis. Ann Phar-
macother 1998; 32: 889–891.
43 Bourgeois J A, Zelenko M, Waraich B S. Psychotic disorder
associated with isoniazid. Mil Med 1996; 161: 707.
44 Sharma G S, Gupta P K, Jain N K, Shanker A, Nanawati V.
Toxic psychosis to isoniazid and ethionamide in a patient with
pulmonary tuberculosis. Tubercle 1979; 60: 171–172.
45 Lansdown F S, Beran M, Litwak T. Psychotoxic reaction during
ethionamide therapy. Am Rev Respir Dis 1967; 95: 1053–
1055.
46 Narang R K. Acute psychotic reaction probably caused by
ethionamide. Tubercle 1972; 53: 137–138.
47 Pickles R W, Spelman D W. Suspected ethambutol-induced
mania. Med J Aust 1996; 164: 445–446.
48 Hsu C W, Chu K A, Lu T, Lai R S, Lu J Y. Ethambutol-induced
psychosis: a case report [abstract]. Zhonghua Yi Xue Za Zhi
1999; 62: 724–727.
49 Mulhall J P, Bergmann L S. Ciprofloxacin-induced acute psy-
chosis. Urology 1995; 46: 102–103.
50 Zaudig M, von Bose M, Weber M M, Bremer D, Zieglgans-
berger W. Psychotoxic effects of ofloxacin. Pharmacopsychia-
try 1989; 22: 11–15.
51 Feinberg S S. Fluoroquinolone-induced depression. Am J Psy-
chiatry 1995; 152: 954–955.
52 Jay G T, Fitzgerald J M. Ciprofloxacin-induced delirium. Ann
Pharmacother 1997; 31: 252.
53 Fennig S, Mauas L. Ofloxacin-induced delirium. J Clin Psychi-
atry 1992; 53: 137–138.
54 Dey S K. Nightmare due to ciprofloxacin in young patients.
Indian Pediatr 1995; 32: 918–920.
55 Hollweg M, Kapfhammer H P, Krupinski M, Moller H J. Psy-
chopathological syndromes in treatment with gyrase inhibitors
[abstract]. Nervenarzt 1997; 68: 38–47.
56 Todd G S, Wittkower E. The psychological aspects of sanato-
rium management. Lancet 1948; 254: 49–53.
57 Ludwig A O. Emotional factors in tuberculosis. Public Health
Rep 1948; 63: 883–888.
58 Lang N G. Stigma, self-esteem and depression: psychosocial
responses to risk of AIDS. Hum Org 1991; 51: 66–72.
59 Ingram D, Hutchinson S A. HIV-Positive women and stigma.
Health Care Women Int 1999; 20: 93–103.
60 Fife B L, Wright E. The dimensionality of stigma: a comparison
of its impact on the self of persons with HIV/AIDS and cancer.
J Health Soc Behav 2000; 41: 50–67.
61 Lawless S, Kippax S, Crawford J. Dirty, diseased, and unde-
serving: the positioning of HIV positive women. Soc Sci Med
1996; 43: 1371–1377.
62 Liefooghe R, Michiels N, Habib S, Moran M B, De Muynck A.
Perception and social consequences of tuberculosis: a focus
group study of tuberculosis patients in Sialkot, Pakistan. Soc
Sci Med 1995; 41: 1685–1692.
63 Sweetland A, Acha D, Guerra D. Enhancing adherence: the
role of group psychotherapy in the treatment of MDR-TB in
urban Peru. In: Cohen A, Kleinman A, Saraceno B, eds. World
Mental Health Casebook: Social and Mental Health Inter-
ventions in Low-Income Countries. New York, NY: Kluwer
Academic/Plenum Press, 2002: pp 51–79.
64 Patel V, Kleinman A. Poverty and common mental disorders
in developing countries. Bull World Health Organ 2003; 81:
609–615.
65 Fullilove M T, Young R, Panzer P G, Muskin P. Psychosocial is-
sues in the management of patients with tuberculosis. J Law
Med Ethics 1993; 21: 324–331.
66 Barnhoorn F, Adriaanse H. In search of factors responsible for
noncompliance among tuberculosis patients in Wardha Dis-
trict, India. Soc Sci Med 1992; 34: 291–306.
67 Mitnick C, Bayona J, Palacios E, et al. Community-based therapy
for multidrug-resistant tuberculosis in Lima, Peru. N Engl J
Med 2003; 348: 119–128.
68 Partners In Health, Harvard Medical School, Bill and Melinda
Gates Foundation. A DOTS-Plus Handbook: Guide to the
Community-Based Treatment of MDR-TB. Boston, MA: Part-
ners In Health, 2002.
69 Partners In Health/Program in Infectious Disease and Social
Change, Harvard Medical School. The PIH Guide to the Med-
ical Management of Multidrug-Resistant Tuberculosis. Bos-
ton, MA: PIH/PIDSC, 2003.
70 Instituto Nacional de Estadística e Informática. Proyecciones
de población por años calendario según departamentos, pro-
vincias, y distritos (Periodo, 1990–2002). Lima, Peru: Instituto
Nacional de Estadística e Informática, 2001. Boletín Especial
No. 16: p. 63.
71 Saavedra J E, Instituto Especializado de Salud Mental “Hono-
rio Delgado—Hideyo Noguchi.” Estudio epidemiológico met-
ropolitano en salud mental 2002. Anales de Salud Mental
2002; 18: 1–197.
72 Minobe K, Perales A, Sogi C, Warthon D, Llanos R, Sato T.
Prevalencia de vida de trastornos mentales en Independencia.
Anales de Salud Mental 1990; 6: 9–20.
73 Instituto Nacional de Salud Mental “Honorio Delgado—Hideyo
Noguchi.” Prevalencia de vida de trastornos mentales en Indepen-
dencia (Lima, Peru). Anales de Salud Mental 1985; 1: 206–222.
74 Becerra M C, Freeman J, Bayona J, et al. Using treatment failure
under effective directly observed short-course chemotherapy
programs to identify patients with multidrug-resistant tubercu-
losis. Int J Tuberc Lung Dis 2000; 4: 108–114.
75 Farmer P E, Bayona J, Shin S, et al. Preliminary results of com-
munity-based MDR-TB treatment in Lima, Peru. Int J Tuberc
Lung Dis 1998; 2 (Suppl 2): S371.
76 Farmer P E, Walton D A, Becerra M C. International tubercu-
losis control in the 21st century. In: Friedman L N, ed. Tuber-
culosis: current concepts and treatment. 2nd ed. Boca Raton,
FL: CRC Press, 2000: pp 475–496.
77 American Psychiatric Association. Diagnostic and Statistical
Manual of Mental Disorders. 4th ed. Washington, DC: Amer-
ican Psychiatric Association, 1994.
 Psychiatric issues in MDR-TB therapy 757

78 Aghanwa H S, Erhabor G E. Demographic/socioeconomic fac-
tors in mental disorders associated with tuberculosis in South-
west Nigeria. J Psychosom Res 1998; 45; 353–360.
79 Aydin I O, Uluşahin A. Depression, anxiety comorbidity, and
disability in tuberculosis and chronic obstructive pulmonary
disease patients: applicability of GHQ-12. Gen Hosp Psychia-
try 2001; 23: 77–83.
80 Goble M, Iseman M D, Madsen L A, Waite D, Ackerson L,
Horsburgh C R Jr. Treatment of 171 patients with pulmonary
tuberculosis resistant to isoniazid and rifampin. N Engl J Med
1993; 328: 527–532.
81 Yew W W, Chan C K, Chau C H, et al. Outcomes of patients
with multidrug-resistant pulmonary tuberculosis treated with
ofloxacin/levofloxacin-containing regimens. Chest 2000; 117:
744–751.
82 Tahaoglu K, Torun T, Sevim T, et al. The treatment of multi-
drug-resistant tuberculosis in Turkey. N Engl J Med 2001; 345:
170–174.

APPENDICES

Appendix 1 Management of psychosis in individualized MDR-TB therapy (adapted from

reference 69). PO  per os; IV  intravenous; IM  intramuscular; TID  thrice daily; QD 
four times daily; BID  twice daily.
758 The International Journal of Tuberculosis and Lung Disease
Appendix 2 Management of depression in individualized MDR-TB therapy (adapted from refer-
ence 69).
INTRODUCTION : Les problèmes psychiatriques repré-
sentent un défi lors du traitement des patients atteints de
tuberculose à germes multirésistants (TB-MR). Une prise
en charge vigoureuse s’impose à la fois pour les maladies
psychiatriques préexistantes et pour le développement de
complications psychiatriques liées aux médicaments anti-
tuberculeux et aux facteurs psycho-sociaux.
C O N T E X T E : Une organisation de santé non-gouverne-
mentale basée sur la collectivité à Lima, Pérou.
O B J E C T I F : Revue de la littérature concernant les compli-
cations psychiatriques associées aux médicaments anti-
tuberculeux, description de l’incidence et de la prévalence
de la dépression de l’anxiété et de la psychose parmi les
individus traités pour TB-MR et détails de l’approche de
prise en charge utilisée dans cette cohorte.
M É T H O D E S : Une série rétrospective de cas a été formée
par les 75 premiers patients qui ont bénéficié d’un trai-
tement individualisé pour TB-MR à Lima, Pérou, entre
1996 et 1999.
RÉSUMÉ
RESULTATS : On a observé une dépression et une anxiété
initiales chez respectivement 52.2% et 8.7% des patients
de cette cohorte. La plupart des individus ayant eu une
dépression initiale ont ressenti une amélioration des
symptômes de dépression au cours du traitement de la
tuberculose. L’incidence de la dépression, de l’anxiété et
de la psychose au cours du traitement de la TB-MR a été
respectivement de 13,3%, de 12,0% et de 12,0%. Alors
que la majorité des individus atteints de dépression,
d’anxiété et de psychose ont nécessité une pharmacothé-
rapie psychiatrique, on a pu poursuivre avec succès l’admi-
nistration de cyclosérine dans tous les cas sauf un.
C O N C L U S I O N : Les comorbidités psychiatriques ne repré-
sentent pas une contre-indication au traitement de la
TB-MR. La prise en charge des complications psychia-
triques est possible sans compromettre le traitement
antituberculeux.

M A R C O D E R E F E R E N C I A : Las complicaciones psiquiá-
tricas presentan un desafío en el tratamiento de pacientes
con la tuberculosis multidrogo-resistente (TB-MDR).
Trastornos psiquiátricos de inicio y también la presenta-
ción de nuevas complicaciones psiquiátricas relaciona-
das con las drogas anti-tuberculosas y factores psico-
sociales requieren un manejo agresivo.
L U G A R : Una organización no-gubernamental basada en
la comunidad en Lima, Perú.
O B J E T I V O : Revisar la literatura para complicaciones
psiquiátricas asociadas con los medicamentos anti-
tuberculosos, describir la incidencia y prevalencia de la
depresión, ansiedad y psicosis entre los individuos reci-
biendo terapia TB-MDR, y resumir con detalle los pro-
tocolos de manejo usados con este cohorte.
M É T O D O S : Una serie retrospectiva de casos fue reali-
zada entre los primeros 75 pacientes que recibían tera-
Psychiatric issues in MDR-TB therapy 759
RESUMEN
pia TB-MDR individualizado en Lima, Peru, de 1996 a
1999.
R E S U L T A D O S : Depresión y ansiedad de inicio fueron
observadas en 52,2% y 8,7%, respectivamente, de este
cohorte. La mayoría de individuos con depresión de ini-
cio experimentaron una mejoría de síntomas depresivos
durante el curso de su terapia antituberculosa. La inci-
dencia de depresión, ansiedad y psicosis durante el trata-
miento TB-MDR fue de 13,3%, 12,0%, y 12,0%, respec-
tivamente. Aunque en la mayoría de individuos con
depresión, ansiedad y psicosis requerían psicofármacos,
se logró continuar el uso de cicloserina en todos menos
un caso.
C O N C L U S I Ó N : Comorbididades psiquiátricas no son
contraindicación para terapia TB-MDR. Manejo de
complicaciones psiquiátrica es posible sin comprometer
tratamiento anti-tuberculoso.