Epilepsia, 46(3):372–377, 2005
Blackwell Publishing, Inc.


C 2005 International League Against Epilepsy
Clinical Research
Phenobarbital Pharmacokinetics in Old Age: A Case-matched
Evaluation Based on Therapeutic Drug Monitoring Data
∗ Sara Messina, ∗ Dina Battino, ∗ Danilo Croci, ∗ Daniela Mamoli, ∗ Sara Ratti and †Emilio Perucca
∗ Carlo Besta National Neurological Institute, Milan; and †Clinical Pharmacology Unit, University of Pavia, and Laboratories for
Diagnostics and Applied Biological Research, Institute of Neurology, IRCCS C. Mondino Foundation, Pavia, Italy
Summary: Purpose: To assess the influence of aging on the
pharmacokinetics of phenobarbital (PB) at steady state in pa-
tients receiving long-term therapy.
Methods: Serum PB concentrations from the database of the
therapeutic drug monitoring service of a large neurological hos-
pital were used to calculate apparent clearance values (CL/F) in
224 patients aged 65 years and older (mean, 73 ± 6.1 years).
CL/F values in these patients were compared with those deter-
mined in an equal number of controls aged 20 to 50 years (mean,
35.7 ± 7.9 years) and matched for gender, body weight, and
type of anticonvulsant comedication. Correlations of CL/F with
age, body weight, gender, and comedication also were explored
within each age group.
Results: PB CL/F values were significantly lower in elderly
patients than in controls (3.2 ± 0.8 vs. 4.1 ± 1.2 ml/h/kg; p <
About 25% of new cases of epilepsy occur in elderly
people, many of whom have concomitant cerebrovascu-
lar, neurodegenerative, or neoplastic disorders (1). Inci-
dent and prevalent cases of epilepsy in the elderly also are
expected to increase over the next decades as a result of
the steady increase in life expectancy and proportion of
aged people in the general population, particularly in de-
veloped countries (2,3). Because a diagnosis of a seizure
disorder in old age leads almost invariably to the institu-
tion of antiepileptic drug (AED) therapy (4,5), it is not sur-
prising that AEDs are among the most widely prescribed
therapeutic agents in an elderly population. In a study
from the United States, 10.5% of nursing home residents
aged 65 and older were found to receive AED therapy
(6). Likewise, a survey of five U.S. pharmacy provider
organizations found that 892 (8.8%) of 10,168 nursing
home residents were prescribed AEDs for the treatment
Accepted November 6, 2004.
Address correspondence and reprint requests to Dr. Dina Battino at
Carlo Besta National Neurological Institute, Via Celoria 11, 20133 Mi-
lan, Italy. E-mail: dbattino@istituto-besta.it
372
0.0001). Age was identified as a statistically significant predic-
tor of CL/F at multiple regression analysis, but it accounted for
only a modest component of the interindividual pharmacokinetic
variation. Comedication with carbamazepine (CBZ) and pheny-
toin (PHT) was associated with a moderate decrease in PB CL/F,
which reached statistical significance in the elderly group (p <
0.01 for CBZ comedication; p < 0.001 for PHT comedication).
Conclusions: Aging is associated with a significant decrease in
PB clearance, which might be related to a reduction in glomeru-
lar filtration rate or diminished drug-metabolizing capacity in the
liver or both. Because of this, older patients will require lower
dosages to achieve serum PB concentrations comparable with
those found in nonelderly adults. Key Words: Phenobarbital—
Pharmacokinetics—Elderly—Therapeutic drug monitoring—
Epilepsy.
of epilepsy, and an additional 2.2% were receiving the
same drugs for non–seizure-related disorders (7). Among
available AEDs, time-honored phenobarbital (PB) contin-
ues to be widely prescribed. Its low cost makes it the only
affordable option in emerging countries, and its utilization
remains relatively high also in developed nations (8–10).
Although its ability to interfere with cognitive function and
to induce the hepatic drug-metabolizing enzymes makes
it a less than optimal choice, other properties such as its
sedative effects and the feasibility of once-daily adminis-
tration could be advantageous in an older person. Possibly
because of this, PB ranked third, after phenytoin (PHT)
and carbamazepine (CBZ), among the most commonly
used AEDs in a recent U.S. study of community-dwelling
elderly patients with previously diagnosed epilepsy (11)
and in surveys of patients living in U.S. nursing homes
(7,12). PB also was by far the most commonly used AED
in a recent study of elderly nursing home residents in Italy
(13).
Rational use of AEDs requires a full understand-
ing of factors that may affect their pharmacokinetics,
 PHENOBARBITAL PHARMACOKINETICS IN THE ELDERLY
particularly for those populations, such as the elderly, in
whom physiologic and pathologic changes may have pro-
found effects on drug disposition (14,15). In view of the
fact that PB has been used for about a century and is fre-
quently prescribed for the elderly, it is remarkable that
information on the influence of old age on its pharma-
cokinetics is virtually nonexistent (4,16,17). To address
this deficiency, we conducted a comprehensive evaluation
of pharmacokinetic data in a large population of patients
aged 65 to 90 years receiving long-term PB therapy at a
major neurological hospital in Milan, Italy. The data pro-
vide evidence for a significant decrease in PB clearance
in these patients.
PATIENTS AND METHODS
Patients
Selection of patients and methodologic approach were
identical to those used in recent evaluations of the phar-
macokinetics of CBZ (18) and PHT (19) in old age. All
inpatients and outpatients receiving prolonged PB ther-
apy monitored at the therapeutic drug monitoring service
(TDMS) of the Carlo Besta National Neurological Insti-
tute in Milan were included in the study if they met the
following eligibility criteria: (a) age 65 years or older;
(b) availability of at least one serum PB concentration at
steady state (e.g., ≥6 weeks on a stable dosage, in the ab-
sence of changes in associated comedication), during the
period January 1990 to November 2003; (c) availability of
accurate information on demographics, clinical status, and
details of comedication(s) taken, including dose, time of
administration, and duration of therapy; (d) no associated
drugs, other than AEDs, known to affect PB pharmacoki-
netics (e.g., cimetidine, isoniazid, activated charcoal); (e)
no clinical or laboratory evidence of moderate to severe
hepatic or renal insufficiency; and (f) no evidence of poor
compliance as indicated by variable (>25% difference)
serum PB concentrations on repeated measurements on
the same dosage and comedication. Together with this
group, a control group of equal size of younger patients
aged 20 to 50 years, also receiving PB therapy and moni-
tored at the TDMS, was selected for comparison purposes.
Except for the age range, eligibility criteria for controls
were identical to those for the elderly, with the additional
requirement that each control had to provide a match for
each individual elderly patient with respect to gender, body
weight (within 10%), and type of AED comedication. Be-
cause the distribution of dosages and serum PB levels
among elderly patients and controls was not randomized,
dosages and serum concentrations were deliberately dis-
regarded in the matching process. Matching for dosage
would have resulted in a bias if the physician, based on
TDM data, had adjusted dosages to achieve comparable
serum drug levels in the two populations. Likewise, match-
ing for serum concentration would have equally caused a
373
bias if the physician used a differential dosing policy in
the elderly compared with that in the young (18). When
more than one concentration was available for any given
patient, the last concentration determined was used in the
comparison. All the information used for the study was de-
rived from the TDMS electronic database, where relevant
demographic, clinical, and pharmacologic information is
carefully recorded before each patient’s visit.
Blood sampling and drug assay
Blood samples were taken between 8 a.m. and 10
a.m., before the administration of the morning dose. Drug
concentrations were measured by fluorescence polar-
ization immunoassay (TDx; Abbott Diagnostics, Rome,
Italy). Throughout the period of the study, the labora-
tory’s proficiency was monitored through internal quality-
control procedures and participation in national (Regione
Lombardia, Milan, Italy) and international (Heathcontrol,
Cardiff, U.K.) external quality-control schemes. Between-
day precision coefficient of variation (CV) was better than
3.2%.
Pharmacokinetic and statistical analysis
Because PB shows linear (dose-independent) phar-
macokinetics, the confounding effect of differences in
dosages was eliminated by determining apparent oral
clearance (CL/F) values, which were calculated for each
patient by using the following equation:
CL/F(ml/h/kg) = PBdose(mg/kg/day)/
[Css (mg/ml) × 24h]
where CL is the total body clearance of the drug, F is the
oral bioavailability, and Css is the serum PB concentration
at steady state.
For each set of data (demographic variables, drug
dosages, serum drug concentrations, and CL/F values),
compatibility of data with a normal distribution and equal-
ity of group variances were assessed by means of the
Kolmogorov–Smirnov test and the Bartlett test. If data
were consistent with a normal distribution and equal vari-
ances, means and SD were calculated and compared by
Student’s t test, or one-way analysis of variance (ANOVA)
in the case of multiple comparisons. For data deviating
from a normal distribution and/or showing different vari-
ances (as for dose, age, and CL/F), comparisons were
made by using Mann–Whitney’s U test or the Kruskal–
Wallis test in the case of multiple comparisons. A p value
≤0.05 was considered significant.
To test potential correlations between CL/F values and
age, a linear regression model was initially used. Analy-
sis of covariance (ANCOVA) was used to compare slopes
and intercepts of the regression lines generated for sub-
groups homogeneous for comedication. In addition, to as-
sess the relative potential contribution of several clinical
variables to the interpatient variability in PB CL/F, a step-
wise linear-regression model was developed (18,19) in
Epilepsia, Vol. 46, No. 3, 2005
374 S. MESSINA ET AL.

which the following parameters were tested: age (years), TABLE 2. Comparison of phenobarbital CL/F values (means
body weight (kg), gender (men, 0; women, 1), and pres- ± SD) in different groups according to age and comedication.
ence or absence of CBZ comedication (monotherapy, 0; Mean ± SD (ml h −1 kg−1 )
CBZ comedication, 1) or PHT comedication (monother-
Elderly Controls
apy, 0; PHT comedication, 1). In finalizing the multiple
regression equations, variables with a p value >0.05 were All patients 3.2 ± 0.8∗∗∗ 4.1 ± 1.2
Phenobarbital monotherapy 3.3 ± 0.8∗∗∗ 4.2 ± 1.2
eventually removed. Carbamazepine comedication 3.0 ± 0.7∗◦ 3.7 ± 1.1
Statistical tests were performed by using the GraphPad Phenytoin comedication 2.7 ± 0.6∗∗◦◦ 3.9 ± 1.3
Instat and the GraphPad Prism version 3.0a for Macintosh ∗ p < 0.01, ∗∗ p < 0.001, ∗∗∗ p < 0.0001 (vs corresponding controls)
(GraphPad Software, San Diego, CA, U.S.A.). ◦ p < 0.01, ◦◦ p < 0.001 (vs phenobarbital monotherapy)

RESULTS
Although overall, elderly patients received significantly
Characteristics of the study population lower daily doses, their mean serum PB concentration
Of 12,487 patients registered in the database, 243 el- was similar to that found in nonelderly controls. When
derly subjects receiving PB fulfilled the eligibility criteria monotherapy groups were compared, elderly patients also
for inclusion in the study. Of these, 168 (69%) were receiv- were found to receive significantly lower PB dosages than
ing monotherapy, and 75 (31%) were receiving comedi- controls (1.7 ± 0.6 vs. 2.0 ± 0.9 mg/kg/day; p < 0.0001).
cation with other AEDs, including CBZ (n = 35), PHT Similar dosage differences between the elderly and con-
(n = 21), and a variety of other drugs (n = 19). Because trols were found for subgroups homogeneous for type of
the number of patients receiving comedications other than comedication, but these differences did not reach statisti-
CBZ and PHT was insufficient for meaningful assessment, cal significance.
these patients were excluded from the evaluation.
Comparison of phenobarbital CL/F values between
The study population submitted to final analysis con-
and within age groups
sisted of 224 patients (116 men, 108 women) aged 65 to
When all patients (irrespective of comedication) were
90 years (mean, 73.0 ± 6.1 years) and an equal number of
compared, mean PB CL/F values were found to be 22%
matched controls aged 20 to 50 years (mean, 35.7 ± 7.9
lower in the elderly than in controls (Table 2). A lower
years). The characteristics of the two groups are summa-
CL/F in the elderly compared with controls also was found
rized in Table 1.
for subgroups of patients receiving monotherapy (21%)
and for those comedicated with CBZ (19%) and PHT
(31%).
TABLE 1. Characteristics of the patients included in the PB CL/F values also were lower in elderly patients re-
study. Values are means ± SD. ceiving CBZ or PHT comedication than in those stabilized
Elderly Controls on monotherapy (Table 2). Lower PB CL/F values in pa-
tients receiving comedication compared with patients in
Age (years) 73.0 ± 6.1 35.7 ± 7.9
Body weight (kg) 67.7 ± 11.7 67.9 ± 12.0
monotherapy also were found in the control group, but the
Gender distribution (males/females) 116 108 116 108 differences failed to reach statistical significance.
Patient on phenobarbital monotherapy (n) 168 168
Patient on carbamazepine comedication (n) 35 35 Relation between phenobarbital CL/F values
Patient on phenytoin comedication (n) 21 21 and age
Phenobarbital dosage (mg kg−1 day−1 )
All patients 1.6 ± 0.6∗∗ 2.0 ± 0.9 The relation between individual PB CL/F values and
Patients on monotherapy 1.7 ± 0.6∗∗ 2.0 ± 0.9 age in monotherapy and comedicated patients within
Patients comedicated with carbamazepine 1.5 ± 0.7 1.9 ± 1.0 the elderly and the control populations is illustrated
Patients comedicated with phenytoin 1.5 ± 0.6 2.0 ± 0.9
Serum phenobarbital in Fig. 1. Although a trend was noted for PB CL/F
concentration (µg ml−1 ) to decrease moderately with increasing age in some
All patients 21.2 ± 7.8 20.5 ± 7.3 subgroups, none of the correlations reached statistical
Patients on monotherapy 20.9 ± 6.8 20.3 ± 7.4
Patients comedicated with carbamazepine 21.2 ± 9.1 21.0 ± 7.5
significance.
Patients comedicated with phenytoin 23.8 ± 11.8 21.3 ± 5.7
Carbamazepine dosage (mg kg−1 day−1 ) 11.8 ± 6.0∗ 15.8 ± 6.5 Multiple regression analysis
Serum carbamazepine 6.5 ± 2.2 6.5 ± 2.3
concentration (µg ml−1 )
To assess the relative contribution of different variables
Phenytoin dosage (mg kg−1 day−1 ) 3.6 ± 1.5 4.2 ± 1.7 to the interindividual variability in PB CL/F, the influence
Serum phenytoin 11.4 ± 7.7 12.4 ± 9.9 of age, gender, body weight, and type of comedication was
concentration (µg ml−1 ) tested by multiple regression analysis. The analysis indi-
∗ p < 0.01 (vs corresponding controls) cated that CL/F was significantly correlated with age (p <
∗∗ p < 0.0001 (vs corresponding controls) 0.0001), daily dosage expressed as mg/kg (p < 0.0001),

Epilepsia, Vol. 46, No. 3, 2005

 PHENOBARBITAL PHARMACOKINETICS IN THE ELDERLY
body weight (p < 0.001), and PHT comedication (p <
0.05), according to the following equation (adjusted r2 =
0.40; p < 0.0001):
CL/F = 4.4 − 0.02 × Age − 0.01 × Body weight +
0.6 × Dose − 0.2 × PHT comedication
Of the variables tested, gender did not appear to contribute
significantly to the prediction of CL/F values.
375
FIG. 1. Relation between phenobarbital
(PB) CL/F values and age in patients
receiving PB monotherapy (open circles,
thick line) and in patients comedicated with
carbamazepine (solid circles, broken line)
or phenytoin (triangles, thin line). Elderly
patients are illustrated in the upper panel,
and controls, in the lower panel. Equa-
tions of the regressions lines for elderly pa-
tients were CL/F = 3.2 + 0.002 × Age,
n = 168, p > 0.05 (phenobarbital
monotherapy); CL/F = 4.7 – 0.02 × Age, n
= 35, p > 0.05 (carbamazepine comedica-
tion); and CL/F = 5.2 – 0.03 × Age, n = 21,
p > 0.05 (phenytoin comedication). Equa-
tions of the regressions lines for controls
were CL/F = 4.6 – 0.01 × Age, n = 168, p >
0.05 (phenobarbital monotherapy); CL/F =
4.3 – 0.02 × Age, n = 35, p > 0.05 (carba-
mazepine comedication); and CL/F = 5.8
– 0.05 × Age, n = 21, p > 0.05 (phenytoin
comedication).
DISCUSSION
To date, the influence of age on PB pharmacokinetics
has been characterized extensively only in pediatric age
groups (20), and, surprisingly, virtually no information
is available on potential alterations occurring at the other
extreme of age. The possibility that aging could be associ-
ated with reduced PB clearance was suggested by Eadie et
al. (21), who assessed the distribution of serum PB levels
in a large population and found that subjects older than 40
years had lower mean CL/F values than did subjects aged
Epilepsia, Vol. 46, No. 3, 2005
376 S. MESSINA ET AL.
15 to 40 years (2.5 vs. 4.9 ml/h/kg, respectively). However,
only a few of their patients were older than 65 years, and
potential pharmacokinetic changes specific for the elderly
could not be evaluated. Indirect evidence for possible al-
terations in PB disposition in old age was provided in an
earlier study by our group that compared the pharmacoki-
netics of primidone in 10 patients aged 70 to 81 years and
eight controls aged 18 to 26 years receiving comparable
dosages (22). In that study, serum levels of metabolically
derived PB levels were found to be 41% higher in the
elderly, although the difference failed to reach statistical
significance.
The present investigation provides clear evidence that
PB pharmacokinetics is altered in old age. Elderly patients
had comparable serum PB levels to those of nonelderly
adult controls, although the latter were prescribed higher
dosages, and PB CL/F values were on average 22% lower
in the elderly than in the controls. Although age was iden-
tified as a statistically significant predictor of CL/F at mul-
tiple regression analysis, the interindividual variability in
CL/F at any given age was considerable, and the trend for
CL/F to decrease moderately with increasing age in some
subgroups failed to reach significance in simple regression
analyses (Fig. 1). Indeed, differences in age had overall
a modest role in explaining the variation in PB pharma-
cokinetics, suggesting that other factors have greater im-
portance. Among the latter, AED comedication was iden-
tified as significant. In particular, concurrent administra-
tion of PHT was associated with a modest decrease in PB
CL/F, which reached statistical significance within the el-
derly group. A decrease in PB CL/F by PHT is consistent
with results of previous studies, in which PHT was found
to increase serum PB concentration, presumably by in-
hibiting PB metabolism (23,24). In our study, CBZ also
appeared to have a modest lowering effect on PB CL/F,
although again this reached statistical significance only
in the elderly population. Although an earlier report sug-
gested that CBZ has no major effect on serum PB levels
(21), in at least one study, CBZ was associated with an
elevated serum concentration of PB derived from prim-
idone (25), an observation that could be explained by a
decrease in PB clearance. Among other factors identified
at multiple regression analysis as potentially affecting PB
CL/F, body weight had only a minor impact, whereas the
apparent influence of dosage is likely to be a spurious find-
ing related to the fact that, based on clinical response and
TDM data, physicians are expected to prescribe higher
dosages in patients with higher drug clearances.
With respect to the mechanisms by which old age
may affect PB pharmacokinetics, these are likely to in-
volve a reduced efficiency of the processes responsible
for the elimination of the drug. In nonelderly subjects,
PB is cleared partly by renal excretion in unchanged
form (7–55% of the administered dose) and partly by
cytochrome CYP2C9-mediated oxidation to p-hydroxy-
Epilepsia, Vol. 46, No. 3, 2005
phenobarbital (6–40% of the dose) and N-glycosidation
to 9-D-glucopyranosyl-phenobarbital (25% of the dose), a
reaction catalyzed by uridine diphosphate glucuronosyl-
transferase (17). Because aging is accompanied by a phys-
iologic decrease in glomerular filtration rate, the aging-
associated reduction in PB CL/F rate may be explained,
at least in part, by a decrease in renal clearance of un-
changed drug, as observed with other AEDs that are elim-
inated primarily by renal excretion, such as gabapentin
(26), levetiracetam (27) and vigabatrin (28). A reduction
in metabolic clearance also could contribute to the de-
crease in CL/F, and it is noteworthy that the unbound drug
clearance of AEDs that are eliminated mainly by hepatic
metabolism, such as CBZ (18,29), valproic acid (30,31),
and PHT (19,32) also has been found to be decreased in
elderly patients.
Irrespective of the mechanism(s) involved, the aging-
related reduction in PB CL/F observed in the present study
may have clinical implications. On average, elderly pa-
tients require smaller dosages to achieve serum PB con-
centrations comparable with those found in nonelderly
adults, and careful monitoring of clinical response and
serum PB levels may be especially warranted in this age
group. If the reduction in PB CL/F is indeed caused by a
decrease in renal or metabolic clearance or both, the half-
life of the drug also may be longer in the elderly, leading
to prolongation of the time required to reach steady state
after initiation of therapy or adjustments in dosage.
REFERENCES
1. Stephen LJ, Brodie MJ. Epilepsy in elderly people. Lancet
2000;355:1441–6.
2. Faught E. Epidemiology and drug treatment of epilepsy in elderly
people. Drugs Aging 1999;15:255–69.
3. Willmore LJ. The effect of age on pharmacokinetics of antiepileptic
drugs. Epilepsia 1995;36:S14–21.
4. Bernus I, Dickinson RG, Hooper WD, Eadie MJ. Anticonvulsant
therapy in aged patients: clinical pharmacokinetic considerations.
Drugs Aging 1997;10:278–89.
5. Tallis R, Boon P, Perucca E, Stephen L. Epilepsy in elderly people:
management issues. Epileptic Disord 2002;4:S33–9.
6. Lackner TE, Cloyd JC, Thomas LW, et al. Antiepileptic drug use in
nursing home residents: effect of age, gender, and comedication on
patterns of use. Epilepsia 1998;39:1083–7.
7. Schachter SC, Cramer GW, Thompson GD, et al. An evaluation of
antiepileptic drug therapy in nursing facilities. J Am Geriatr Soc
1998;46:1137–41.
8. Baldy-Moulinier M, Covanis A, D’Urso S, et al. Therapeutic strate-
gies against epilepsy in Mediterranean countries: a report from an
international collaborative survey. Seizure 1998;7:513–20.
9. Kwan P, Brodie MJ. Phenobarbital for the treatment of epilepsy
in the 21st century: a critical review. Epilepsia 2004;45:1141–
9.
10. Kale R, Perucca E. Revisiting phenobarbital for epilepsy. BMJ 2004
(in press).
11. Pugh MJ, Cramer J, Knoefel J, et al. Potentially inappropriate
antiepileptic drugs for elderly patients with epilepsy. J Am Geri-
atr Soc 2004;52:417–22.
12. Cloyd JC, Lackner TE, Leppik IE. Antiepileptics in the elderly:
pharmacoepidemiology and pharmacokinetics. Arch Fam Med
1994;3:589–98.
 PHENOBARBITAL PHARMACOKINETICS IN THE ELDERLY
13. Galimberti CA. Issues in nursing home settings. Epilepsia
2003;44(suppl 8):23.
14. Parker BM, Cusack BJ, Vestal RE. Pharmacokinetic optimisation of
drug therapy in elderly patients. Drugs Aging 1995;7:10–8.
15. Leppik IE. Epilepsy in the elderly. Curr Neurol Neurosci Rep
2001;1:396–402.
16. King-Stephens D. The treatment of epilepsy in the elderly. CNS
Drugs 1999;12:21–33.
17. Anderson GD. Phenobarbital and other barbiturates: chemistry, bio-
transformation and pharmacokinetics. In: Levy R, Mattson R, Miel-
drum B, et al., eds. Antiepileptic drugs, 5th ed. Philadelphia: Lip-
pincott Williams & Wilkins, 2002:496–503.
18. Battino D, Croci D, Rossini A, et al. Serum carbamazepine con-
centrations in elderly patients: a case-matched pharmacokinetic
evaluation based on therapeutic drug monitoring data. Epilepsia
2003;44:923–9.
19. Battino D, Croci D, Mamoli D, et al. Influence of aging on serum
phenytoin concentrations: a pharmacokinetic analysis based on ther-
apeutic drug monitoring data. Epilepsy Res 2004;59:155–65.
20. Battino D, Estienne M, Avanzini G. Clinical pharmacokinetics of
antiepileptic drugs in paediatric patients, Part I: phenobarbital, prim-
idone, valproic acid, ethosuximide and mesuximide. Clin Pharma-
cokinet 1995;29:257–86.
21. Eadie MJ, Lander CM, Hooper WD, et al. Factors influencing
plasma phenobarbitone levels in epileptic patients. Br J Clin Phar-
macol 1977;4:541–7.
22. Martines C, Gatti G, Sasso E, et al. The disposition of primidone in
elderly patients. Br J Clin Pharmacol 1990;30:607–11.
377
23. Windorfer A Jr, Sauer W. Drug interactions during anticonvulsant
therapy in childhood: diphenylhydantoin, primidone, phenobarbi-
tone, clonazepam, nitrazepam, carbamazepin and dipropylacetate.
Neuropadiatrie 1977;8:29–41.
24. Lambie DG, Johnson RH. The effects of phenytoin on phenobar-
bitone and primidone metabolism. J Neurol Neurosurg Psychiatry
1981;44:148–51.
25. Callaghan N, Feely M, Duggan F, et al. The effect of anticonvul-
sant drugs which induce liver microsomal enzymes on derived and
ingested phenobarbitone levels. Acta Neurol Scand 1977;56:1–6.
26. Boyd RA, Turck D, Abel RB, et al. Effects of age and gen-
der on single-dose pharmacokinetics of gabapentin. Epilepsia
1999;40:474–9.
27. Patsalos PN. The pharmacokinetic characteristics of levetiracetam.
Methods Find Exp Clin Pharmacol 2003;25:123–9.
28. Haegele KD, Huebert ND, Ebel M, et al. Pharmacokinetics of viga-
batrin: implications of creatinine clearance. Clin Pharmacol Ther
1988;44:558–65.
29. Graves NM, Brundage RC, Wen Y, et al. Population pharmacoki-
netics of carbamazepine in adults with epilepsy. Pharmacotherapy
1998;18:273–81.
30. Perucca E, Grimaldi R, Gatti G, et al. Pharmacokinetics of valproic
acid in the elderly. Br J Clin Pharmacol 1984;17:665–9.
31. Bauer LA, Davis R, Wilensky A, et al. Valproic acid clearance:
unbound fraction and diurnal variation in young and elderly adults.
Clin Pharmacol Ther 1985;37:697–700.
32. Bachmann KA, Belloto RJ Jr. Differential kinetics of phenytoin in
elderly patients. Drugs Aging 1999;15:235–50.
Epilepsia, Vol. 46, No. 3, 2005