You are on page 1of 5


Phenytoin: A Guide to Therapeutic Drug Monitoring

Melissa Faye Wu, BSc (Pharm) (Hons), Wai Hing Lim, B.Pharm (Hons), M Clin Pharm
Department of Pharmacy, Singapore General Hospital, Singapore

Therapeutic drug monitoring of phenytoin is carried out to ensure effective and safe levels. Some of the factors
complicating phenytoin dosing include a narrow therapeutic window, high degree of protein binding, and nonlinear pharmacokinetics. However, serum drug levels should only be taken when there is a clear indication to
guide patient management. Scenarios where monitoring levels may be clinically useful include: (1) establishing an
individual therapeutic concentration, (2) aiding in diagnosis of clinical toxicity, (3) assessing patient compliance,
and (4) guiding dosage adjustments in patients likely to have greater pharmacokinetic variability. This paper
aims to provide an overview on of the correct timing and interpretation of phenytoin levels. It will also address
phenytoin dosing, other monitoring parameters, and management of phenytoin adverse effects.
Keywords: Pharmacokinetics, Phenytoin, Therapeutic drug monitoring

Phenytoin is an anticonvulsant licensed for the
management of generalised tonic-clonic (grandmal) seizures and complex partial (psychomotor,
temporal lobe) seizures1. It is also approved by
the United States Federal Drug Authority for the
prevention and treatment of seizures occurring
during or after neurosurgery2.
Phenytoin has a narrow therapeutic window hence
a fine balance must be found between efficacy and
dose-related side effects3. Since phenytoin is highly
protein-bound and free (unbound) phenytoin is the
component producing the pharmacological effect,
any factor which changes the protein binding of
phenytoin would be expected to alter the free
drug levels. As such, interactions with other drugs
(drug-drug interactions) or with diseases e.g. renal
impairment, uraemia, and critical illness (drugdisease interactions) may give rise to changes
in phenytoin pharmacokinetics and/or efficacy
and toxicity.


Furthermore, phenytoin demonstrates non-linear

pharmacokinetics even within the therapeutic
range. The enzyme system involved in phenytoin
metabolism gradually becomes saturated,
resulting in a decrease in the rate of elimination
of phenytoin as the dose is increased4. This means
that once the enzyme system becomes saturated
with phenytoin, even a small change in dose can
lead to a large change in phenytoin levels. The
phenytoin concentrations leading to enzyme
saturation vary considerably between individuals.
Thus, patients taking the same dosage can have
up to a 50-fold difference in plasma phenytoin
concentration (inter-individual variability)4. For
these reasons, monitoring of phenytoin levels may
be clinically valuable to ensure therapeutic efficacy
in individual patients.
Laboratories in Singapore commonly monitor total
phenytoin concentration in blood. Free phenytoin
levels are available in some facilities worldwide,
but access remains limited and costly. Furthermore,
studies show that the use of total phenytoin
levels is adequate in most clinical cases5,6 and free
phenytoin levels may be required only for patients

Proceedings of Singapore Healthcare Volume 22 Number 3 2013

Phenytoin TDM

expected to have altered protein binding or where

adverse effects are experienced at unexpectedly
low total phenytoin concentrations6.
TDM of phenytoin levels should only be done
when there is a clear indication to guide patient
management. Situations where measured serum
phenytoin levels may be clinically useful7 are
summarised below:
To Establish an Individual Therapeutic
The reference range for phenytoin levels may
not be applicable to all patients. The difference
can be due to seizure type, the severity of the
underlying disorder, or genetic abnormalities.
Ultimately, phenytoin therapy may be best guided
by the identification of the individual therapeutic
concentration the concentration which has
been found to produce optimal response in the
individual patient (i.e. a balance between seizure
control and acceptable adverse effects).
To Aid in Diagnosis of Clinical Toxicity
Phenytoin toxicity may be variable in presentation
(Table 1). Some dose-related adverse effects may
mimic symptoms of other diseases. Drug levels
may assist in ruling out phenytoin toxicity as a
differential diagnosis.
To Assess Patient Compliance (e.g. Situations of
Uncontrolled or Breakthrough Seizures)
Breakthrough seizures or uncontrolled epilepsy
may be due to a genuinely refractory disease
or intercurrent illness. However, patient non-

compliance leading to sub-therapeutic phenytoin

levels is common enough that checking the
phenytoin level to ensure effective drug
concentrations would be a logical first step in
patients presenting with breakthrough seizures or
worsening epilepsy.
To Guide Dosage Adjustments in Situations
Associated with Greater Pharmacokinetic
Factors altering the pharmacokinetics (absorption,
distribution, metabolism, and elimination) of
phenytoin may result in unpredictable changes
in drug levels. For example, the phenytoin
volume of distribution is higher in obesity,
implying a need for higher loading doses but also
possibly prolonging the half-life of phenytoin in
obese patients8. Other patient groups in which
pharmacokinetics are altered include the elderly,
the critically ill, those with renal or liver failure, or
pregnant women7. Patients who are on concurrent
drugs which interact with phenytoin will also
benefit from drug level monitoring. Changing the
dosage forms or salt forms (phenytoin sodium
versus phenytoin base) and/or interchanging
brands may also lead to changes in bioavailability
monitoring the phenytoin level during the
period of changeover will also help to optimise the
anticonvulsant efficacy.
The average half-life of oral phenytoin is 22
hours. Using normal maintenance doses, the drug
commonly reaches a steady state in 710 days9. In
emergency situations such as status epilepticus
when phenytoin must be initiated urgently for
a patient, the use of a loading dose will help the

Table 1. Phenytoin Side Effects9,10.

Common dose-related side effects
Central nervous system effects: somnolence, fatigue, dizziness,
confusion, visual disturbances, nystagmus, ataxia
Gastrointestinal side effects: nausea, vomiting, anorexia
> 20 mg/L: far lateral nystagmus
> 30 mg/L: 45O lateral gaze nystagmus and ataxis
> 40 mg/L: decreased mentation
> 100 mg/L: death

Non-dose related side effects

Gingival hyperplasia hirsutism
thickening of facial features
vitamin D deficiency
folic acid deficiency
peripheral neuropathy

Proceedings of Singapore Healthcare Volume 22 Number 3 2013



medication reach the necessary therapeutic levels

faster. Giving a loading dose of phenytoin should
be done in an inpatient setting with close followup and monitoring of levels. Loading may not be
recommended in patients with significant renal
and/or hepatic impairment.
Phenytoin can be given intravenously or orally.
The loading dose is generally 1020 mg/kg10.
If given orally, this dose should be divided into
three doses (e.g. 1000 mg given as 400 mg initially,
300 mg two hours later and 300 mg two hours
after that) to ensure optimal oral absorption. If
given intravenously, the full dose can be given at
a maximum rate of 50 mg/min (higher rates can
lead to hypotension and cardiac compromise).
Subsequent maintenance doses are usually in the
range of 57 mg/kg/day (300400 mg/day).
The Importance of Taking Concurrent Albumin
Total serum phenytoin levels reflect both bound
and unbound drug. However, as protein-bound
phenytoin cannot cross the blood-brain barrier,
only free phenytoin is active. In healthy adults,
approximately 90% of phenytoin is bound to
albumin. Thus, phenytoin levels must be corrected
according to albumin levels.
Formula for Corrected Phenytoin3
For patients with good renal function:
Corrected phenytoin (mg/L)= Observed phenytoin (mg/L)
(O.2 x albumin [g/dL]) + 0.1

In end-stage renal failure patients, the binding

of phenytoin to albumin is impaired. Thus, the
formula is modified to:
Corrected phenytoin (mg/L)= Observed phenytoin (mg/L)
(O.1 x albumin [g/dL]) + 0.1

N.B. Errors in calculation often occur because of the

use of the wrong units for albumin. Laboratories
often report the value in g/L instead (g/dL = g/L x
Scenario 1: Newly Started Patient on Phenytoin
After a patient has received a loading dose of
intravenous phenytoin, levels can be checked
one hour after the dose. If loading is achieved by


oral dosing, phenytoin levels can be checked 24

hours after the last dose10. This level can aid in
determining maintenance dose or need to reload.
After initiation of a maintenance dose, levels
can be taken within three to four days to ensure
that the patients metabolism is not remarkably
different from the average literature-derived
pharmacokinetic parameters10.
Scenario 2: Suspected Non-compliance/
Breakthrough Seizures
Because of the long half-life of phenytoin during
long-term administration, the diurnal fluctuation
in its plasma concentration is relatively small. The
timing of blood samples in relation to the time of
dosing is therefore of little importance for correctly
interpreting the plasma concentration4. Levels can
be drawn at the point of admission, regardless of
the patients normal dosing time.
Scenario 3: After Dose Adjustment
In normal healthy subjects after dose adjustment,
the phenytoin level should be drawn within six to
seven days with subsequent doses of phenytoin
adjusted accordingly. In these situations of routine
monitoring (unlike the emergency situation
of breakthrough seizures in scenario two), it is
preferable to draw the phenytoin level just prior to
the next dose (a trough level) or at least eight hours
after the last dose10,11.
Scenario 4: Suspected Phenytoin Toxicity
Upon suspicion of phenytoin toxicity, a blood
level can be drawn immediately and correlated
clinically to the patient presenting symptoms. High
phenytoin levels with corresponding symptoms
will warrant immediate discontinuation of
phenytoin. A second phenytoin level may be useful
to guide when to restart phenytoin. The lapse
time in rechecking the phenytoin level should be
determined by how high the first toxic level was, as
phenytoin clearance dramatically slows with very
toxic concentrations.
The reference range for corrected phenytoin is 10
20 mg/dL9. When it comes to TDM of anti-epileptic
drugs, there is a principle which says: Treat the
patient, not the numbers. Certain patients may
not reach the recommended reference range yet
remain seizure-free, while other patients may
exhibit concentration-related side effects within

Proceedings of Singapore Healthcare Volume 22 Number 3 2013

Phenytoin TDM

the normal range. It is not recommended to increase

the dose of phenytoin in patients who are seizurefree with low therapeutic levels. Conversely, the
dose should be decreased in patients exhibiting
symptoms of phenytoin toxicity with therapeutic
levels. This brings to light the importance of
establishing individual therapeutic concentrations
for different patients.
How Often Should I Check Phenytoin Levels in a
Stable Patient i.e. in the Outpatient Setting?
Depending on the state of the patient, phenytoin
levels are generally monitored at 3- to 12-months
intervals10. Generally, situations which alter
phenytoin pharmacokinetics (e.g. body weight
changes, initiation of interacting drugs, changes in
renal or liver function) should prompt the physician
to revisit the phenytoin regimen and review the
need to check drug levels.
How Much Should I Increase the Phenytoin Dose if
my Patient is not Responding to Treatment?
Guidelines for dosing adjustments based on
phenytoin plasma concentrations have been
proposed for adults with epilepsy without clinically
significant renal or hepatic disease: for plasma
phenytoin concentrations less than 7 g/ml, a
dosage increase of 100 mg/day is recommended;
for plasma concentrations between 7 and 12 g/
ml, the dose may be increased by 50 mg/day; if
the plasma concentration is greater than 12 g/
ml, the dose may increase by 30 mg/day9. Dosage
increases when the plasma level is above 16 g/ml
should only be done with caution as even a small
increase may result in toxicity9.
Dose changes should be more conservative in
patients with reduced protein binding (e.g. in
hypoalbuminemia or in renal impairment) as
changes in drug concentration will be exaggerated
in these cases.
Besides monitoring therapeutic drug levels of
phenytoin when such monitoring is of clinical
value, there are other parameters which should be
considered for all patients on phenytoin therapy.
Monitoring full blood count and liver function tests
as well as for any suicidality (suicidal thoughts,
depression, behavioural changes) should be
performed on a routine basis10.


The complex pharmacokinetics of phenytoin
and the need to titrate phenytoin dosage based
on drug levels and patient response should not
be a deterrent to using phenytoin for seizure
control. Phenytoin, when used appropriately, is
safe and has well-documented efficacy for many
seizure types.
Dose-related adverse effects (Table 1) are common
during initiation of therapy, but may improve or
resolve over time, even without a change in dose12.
Titration of dosage regimen based on phenytoin
levels may also help with dose-related adverse
effects, to achieve an effective drug level with
tolerable adverse effects.
The main reason for switching from phenytoin
would usually be as a result of non-dose related
adverse reactions. Idiosyncratic reactions usually
manifest within the first few weeks or months
of therapy12, and patients newly started on
phenytoin should be counselled to be vigilant in
looking out for adverse effects and to raise any
concerns quickly.
Hypersensitivity manifesting as cutaneous
reactions ranging from rash, Stevens-Johnson
syndrome (SJS), toxic epidermal necrolysis and
drug reaction with eosinophilia and systemic
symptoms, may warrant discontinuation of
phenytoin9. Cross-sensitivity of skin rash has been
noted between phenytoin and carbamazepine13,
and carbamazepine should be avoided in patients
with rash due to phenytoin. Of note, phenytoinrelated SJS seems to be increased in Asians with
the HLA-B*1502 allele14.
Hepatitis may result from phenytoin use. Generally,
sustained transaminitis or clinical hepatitis
(jaundice, hepatomegaly, acute liver failure) should
spur a change of anti-epileptic9. When considering
alternative anti-epileptics in this situation,
precautions with pre-existing liver impairment
may need to be considered, for example,
valproate or carbamazepine (both of which are
metabolised hepatically).
Drug interactions between phenytoin and other
crucial drugs may warrant switching to a non-

Proceedings of Singapore Healthcare Volume 22 Number 3 2013



interacting drug. Phenytoin induces cytochrome

P450 enzyme activity, and may alter the effects
of medicines such as immunosuppressants
(tacrolimus, cyclosporine) and anticoagulants
(warfarin, rivaroxaban)15. Where no alternative to
these medications exist, phenytoin may need to be
switched to an alternative non-interacting drug.


Dilantin [package insert]. New York, New York: Pfizer Inc;

Aug 2012. Accessed 3 June 2013.
2. Dilantin [package insert]. New York, New York: Parke-Davis
(Division of Pfizer Inc); Jan 2012. http://www.accessdata.
Accessed 3 June 2013.
3. von Winckelmann SL, Spret I, Willems L. Therapeutic
drug monitoring of phenytoin in critically ill patients.
Pharmacotherapy 2008;28(11):1391400.
4. Aronson JK, Hardman M, Reynolds DJM. ABC of monitoring
drug therapy: phenytoin. BMJ 1992;305(6863):12158.
5. Rimmer EM, Buss DC, Routledge PA, Richens A. Should we
routinely measure free plasma phenytoin concentration?
Br J Clin Pharmacol 1984;17(1):99102.
6. Theodore WH, Yu L, Price B, Yonekawa W, Porter RJ,
Kapetanovic I, et al. The clinical value of free phenytoin
levels. Acta Neurol 1985;18(1):903.



Patsalos PN, Berry DJ, Bourgeois BF, Cloyd JC, Glauser TA,
Johannessen SI, et al. Antiepileptic drugsbest practice
guidelines for therapeutic drug monitoring: A position
paper by the subcommission on therapeutic drug
monitoring, ILAE Commission on Therapeutic Strategies.
Epilepsia 2008;49(7):123976.
8. Abernethy DR, Greenblatt DJ. Phenytoin disposition in
obesity. Determination of loading dose. Arch Neurol
9. Phenytoin. In: DRUGDEX System [database on the
Internet]. Ann Arbor (MI): Truven Health Analytics;
2013 [cited 24 May 2013]. Available from: www. Subscription required to
10. Lexi-comp Online. Drug monograph: Phenytoin.
Accessed 24 May 2013.
11. Brophy GM, Tesoro EP, Schrote GL, Garnett WR. Pharmacist
impact on posttraumatic seizure prophylaxis in patients
with head injury. Pharmacotherapy 2002;22(2):2515.
12. Greenwood RS. Adverse effects of antiepileptic drugs.
Epilepsia 2000;41(Suppl2):S42S52.
13. Hirsch LJ, Arif H, Nahm EA, Buchsbaum R, Resor SR Jr, Bazil
CW. Cross-sensitivity of skin rashes with antiepileptic
drug use. Neurology 2008;71(19):152734.
14. Man CB, Kwan P, Baum L, Yu E, Lau KM, Cheng AS, et al.
Association between HLA-B*1502 allele and antiepileptic
drug-induced cutaneous reactions in Han Chinese.
Epilepsia 2007;48(5):10158.
15. Nation RL, Evans AM, Milne RW. Pharmacokinetic Drug
Interactions with Phenytoin (Part I). Clin Pharmacokinet

Proceedings of Singapore Healthcare Volume 22 Number 3 2013