See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/20416154

Phenytoin withdrawal and seizure frequency

Article  in  Neurology · August 1989

DOI: 10.1212/WNL.39.7.905 · Source: PubMed

CITATIONS READS

22 219

5 authors, including:

Orrin Devinsky Frank J Nice

NYU Langone Medical Center Nice Breastfeeding
830 PUBLICATIONS   35,367 CITATIONS    28 PUBLICATIONS   333 CITATIONS   

SEE PROFILE SEE PROFILE

William H Theodore
National Institutes of Health
384 PUBLICATIONS   16,822 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Cohort of HIV-Associated Seizures and Epilepsy (CHASE) Study View project

SUDEP prevention View project

All content following this page was uploaded by Frank J Nice on 21 November 2018.

The user has requested enhancement of the downloaded file.

 Phenytoin withdrawal and seizure frequency
E. B. Bromfield, J. Dambrosia, O. Devinsky, F. J. Nice and W. H. Theodore
Neurology 1989;39;905-

This information is current as of April 6, 2007

The online version of this article, along with updated information and services, is
located on the World Wide Web at:
http://www.neurology.org

Neurology is the official journal of AAN Enterprises, Inc. A bi-monthly publication, it has been
published continuously since 1951. Copyright © 1989 by AAN Enterprises, Inc. All rights
reserved. Print ISSN: 0028-3878. Online ISSN: 1526-632X.

Downloaded from www.neurology.org by WILLIAM H THEODORE on April 6, 2007

 . Phenytoin withdrawal
and seizure frequency
E.B. Bromfield, MD; J. Dambrosia, PhD; 0. Devinsky, MD; F.J. Nice, MS; and W.H. Theodore, MD

Article abstract-We withdrew phenytoin from 17 inpatients maintained on combination therapy with carbamazepine for
complex partial seizuresand analyzed seizure occurrence in relation to plasma levelsand time from initiation of withdrawal. The
ratio of maximum to mean weekly seizure frequency did not vary with initial level or rate of withdrawal. The week with most
frequent seizures began a median of 10 days after phenytoin levels became undetectable, and mean daily seizure frequency was
higher at undetectable than at falling levels for the entire 2- to 10-weekstudy period. Four patients had a total of 6 clusters of
generalized tonic-clonic seizures; only 2 occurred while levels were falling and the other 4 at 3,9,28, and 42 days after reaching
undetectable levels. Our data argue against the occurrence of withdrawal seizures in these patients and suggest that worsening of
seizures followingphenytoin discontinuation more likely reflects loss of therapeutic drug effect than a true abstinence phenome-
non.
NEUROLOGY 198339905-909

Although phenytoin ( P H T ) remains a first-line drug for
the treatment of both complex partial seizures (CPS)
a n d secondary generalized tonic-clonic seizures
(GTCS),concerns about cognitive and cosmetic side
effects, as well as a recent trend toward single-drug
have led some practitioners to withdraw
PHT even from many patients whose seizures are not
adequately controlled. Because of its relative freedom
from similar side effects, carbamazepine (CBZ) has
been a widely-used alternative.' Evaluation for possible
surgical therapy is another common context for drug
. ~this
d i s c ~ n t i n u a t i o n ~in ; situation, an increase in sei-
zure frequency is desired. Little is known, however,
about the effects of PHT withdrawal on seizure fre-
quency and severity.
We now report our results in withdrawing PHT from
17 hospitalized patients with uncontrolled partial and
generalized seizures on concomitant therapy with CBZ.
Methods. Patients. Patients had all been referred to the Clin-
ical Epilepsy Section at the National Institutes of Health for
uncontrolled partial seizures. Seventeenpatients (11men and
6 women, ages 19 to 51) were included, with clinical charac-
teristics as shown in table 1.All had CPS, and 13had GTCS as
well; seizure diagnosis was verified by video-EEG telemetry.
Eleven had a history of seizureclusters or of status epilepticus.
All had been on chronic therapy with PHT for years, and most
had been on CBZ for months to years as well. PHT withdrawal
was undertaken as part of either a positron emission tomogra-
phy protocol designed to study effects of PHT on cerebral
glucose metabolism or in preparation for an investigational
drug trial. Most patients had been on additional drugs prior to
admission, but these were withdrawn days to weeks before
PHT; 4 patients were on subtherapeutic doses of other drugs
which were withdrawn simultaneously (table 1).
Procedure. Seizures were recorded by nursing personnel in
a specialized epilepsy unit. Blood levels were measured 2 or 3
times weekly using an enzyme-multiplied immunoassay tech-
nique. Rate of withdrawal was determined clinically by the
treating physician, as were adjustments in CBZ dose; the
latter were minor, since all patients had been stabilized on
therapeutic levels ( > 6 mg/l) of CBZ before beginning PHT
withdrawal. The observation period ended when the patient
began acute treatment with PHT or chronic therapy with any
other active drug, or was discharged.
Analysis.For each patient, the daily frequency of CPS and
GTCS (omittingsimplepartial seizures,whose subjectivenature
precluded accurate verification) was computed for days when
PHT level was high (>lo mg/l), low (between10 and 2.5 mg/l),
and undetectable ( t 2 . 5 mg/l). Levels on days with no mea-
sured value were estimated by linear interpolation.Two patients
(5 and 7) typically had clusters lasting several minutes of brief
CPS with motor phenomena, each only a few seconds in dura-
tion; these were counted as single seizures. Average daily seizure
rate at each of the 3 ranges was ranked and analyzed by a
modiiied Friedman test7;the wide intersubject variability in
seizure fresuency necessitated use of nonparametrictechniques.
In addition, weekly seizure frequencieswere determined for the
entireobservationperiod. The ratio of maximum to mean weekly
seizure frequency wm taken as an index of seizure exacerbation
and correlated with initial PHT level, rate of PH" withdrawal,
and changein CBZ level; rank-transformedvalues were analyzed
by linear regression techniques. Finally,we constructedactuarial
m e s , using as an endpoint the start of the week with each
patient's maximum number of seizures.Subjectswere stratified
by length of taper, and life tables compared using the Mantel-
Haenszel method.
Resulte. Patients were observed for a median of 6.1
weeks (range, 2.3 to 9.6) after start of PHT taper.Mean

From tbe Clinical Epilepsy Section, Medical Neurology Branch (Drs. Bromfield, Devinsky. Theodore, and Mr.Nice), and Biometry and Field Studies Branch (Dr.
Darnbrosia), National Institute of Neurological Disorders and Stroke. Beth&. MD.
Received November 29,1988. Accepted for publication in final form February 2,1989.
Address correspondence and reprint requeststo Dr. Bromfield. Building 10. Room 5N-246, National Institutes of Health. Bethesda, MD 20892.
July 1989 NEUROLOGY 39 906
Downloaded from www.neurology.org by WILLIAM H THEODORE on April 6, 2007
Table 1. Patient characteristics

Duration Initial Initial

Age Seizure of epilepsy phenytoin carbamazepine Other
Pt (yrs) Sex type Etiology (Y=4 Neurologic exam level (mg/l) level (mgfl) medications

1 29 M CP.GTC Unknown 28 Normal 9.9 9.5

2 32 F CP.CTC Unknown 19 Normal 14.1 8.8
3 32 F CP, GTC Infection 20 Normal 16.0 7.1
4 31 M CP Unknown 27 Normal 13.6 7.5
5 33 M CP,GTC Unknown 25 Normal 15.7 10.6
6 33 F CP.GTC Unknown 29 Static encephalopathy. 6.2 7.6
mild right hemiparesis
after cortical resection
7 31 M CP.GTC Unknown 25 Normal 22.0 8.7
8 25 M CP.GTC Unknown 23 Cortically blind after 19.0 10.0
intraoperative occipital
hemorrhage
9 51 M CP.CTC Unknown 33 Normal 12.5 9.8
10 31 M CP,GTC Unknown 20 Normal 12.9 9.2
11 19 M CP Infection 14 Normal 13.4 8.2
12 42 M CP,CTC Unknown 27 Normal 21.0 7.8
13 29 F CP Trauma 9 Normal 10.6 9.9 -
14 25 F CP Unknown 10 Normal 8.6 6.1 Phenobarbital
(admission level 9 mgb,
undetectable on day 5)
15 27 F CP. GTC Infection 14 Normal 20.1 6.3 Clonazepam 0.5 mg/d
(discontinued on day 4)
16 26 M CP,GTC Unknown 14 Normal 15.0 1.5 Valproic acid (admission
level 36 mgb,
undetectable on day 4)
17 33 M CP,GTC Trauma 17 Normal 19.5 9.4 Clorazepate 3.25
mg/d (discontinued
on day 13)
31 ? 7‘ 21 ? 7 14.7 ? 4.5 8.5 5 1.5

CP Complex partial.
GTC Generalized tonic-clonic.
Means 5 SD.

initial PHT level was 14.7 f 4.5 mg/l (mean f SD) and
became undetectable after 10.1 f 4.9 days (range,5 to 21
days). Rate of fall was 1.7 f 0.6 mg/l/d. CBZ level rose
from 8.5 f 1.5mg/l to 11.8 f 1.8 mg/l over the course of
the observation period ( t = 6.17, df = 15,p < 0.0001).
Seizure frequency was higher at undetectable levels
than at either high or low detectable levels as previously
defined; adjusted ranks were 1.74 for high, 1.71 for low,
and 2.29 for undetectable levels 03 = 0.061, modified
Friedman test). Furthermore, the 1st week of undetec-
table levels did not carry any greater risk than subse-
quent weeks; 12 of the 17 patients, in fact,, had more
frequent seizures during the latter period (Z-score with
Yates’correction = 1 . 4 2 , ~= 0.16,2-tailed). Among the
13 patients with GTCS, 10 had such seizures more fre-
quently after withdrawal than while levels were falling
(corrected Z-score = 1.67, p = 0.095, 2-tailed).
Table 2 shows the weekly seizure counts for each
patient. Median frequency was 3.9 seizures per week,
but ranged from 1.0to 30.8. Exacerbation index, or ratio
of maximum to mean weekly seizure frequency,was 2.2
f 0.7 (calculatedwithout extrapolating for incomplete
final weeks in patients 2, 9, 11, and 12). Regression
analysis of the rank-transformed exacerbation indices
showed no significant relationship to initial PHT level,
rate of PHT withdrawal, initial CBZ level, or change in
CBZ level (p > 0.10 for all comparisons without regard
906 NEUROLOGY 39 July 1989
Downloaded from www.neurology.org by WILLIAM H THEODORE on April 6, 2007
to extrapolation for incomplete final weeks).
Figure 1 shows the relationship among time to un-
detectable PHT, time to beginning of the week with the
most seizures, and length of observation period for each
patient. (For patients 1, 4, and 8, who had 2 such
“worst” weeks, the 2 first days were averaged.) The
median interval between end of withdrawal and start of
that week was 10 days (range, -9 to $26 days; 95%
confidence interval, 3 to 16 days). Figure 1 shows no
clustering at any particular time relative to completion
of withdrawal. Of the 4 patients whose worst week
included or preceded the end of PHT withdrawal,3 were
on additional drugs a t subtherapeutic doses, and 2 had
falling levels of the other drug during that week (phe-
nobarbital in patient 14 and clorazepate in patient 17).
Figure 2 shows actuarial survival curves using the start
of the week with the greatest number of seizures as the
endpoint. Life-table analysis revealed no difference in
time to worst week when patients were stratified by
time to undetectable levels (figure 2).
Several patients experienced episodes of repetitive
seizures requiring treatment with additional medica-
tions. All such patients had experienced similar epi-
sodes in the past, although often not at a rate that would
have led one to expect their occurrence during such a
limited observation period. In many cases these oc-
curred days to weeks after completion of PHT with-
Table 2. Seizure frequency during and after phenytoin withdrawal
Maximum
Pt Week1 Week2 Week3 Week4 Week5 Week6 Week7 Week8 Week9 Week10 Mean Mean

1 0 1 (1) 0 l ( 1 ) Ildl 1 .o 2.1

2 4 (1) 2 5.2 1.2
3 6 (5). 2 3 .O 2.0
4 1 2 2.7 1.9
5 0 0 2.4 2.9
6 22 (2) 25 (1) 30.8 1.6
7 11 11 14.1 2.1
8 3 (3) 3 (2) 4.7 1.5
9 1 2 3.9 2.2
10 1 (1) 3 (1) 3.6 2.8
11 2 0 1.9 1.6
12 0 0 1.4 3.6
13 29 34' 17.8 1.9
14 10. 4 6.8 1.5
15 2 6 (3) 0 144 2.6 2.7
16 0 15 (12). 4.1 3.6
17 2 6. 3.9 1.5

Numbers show weekly totals of complex partial and generalized tonic-clonic seizures (generalized in parentheses).
Number of days observed in brackets. if <7.
Asterisk (*) shows number of seizures during worst week.

drawal. Patient 2 had 3 CPS with prolonged postictal
confusion on day 15and was restarted on PHT. Patient
3 had 5 GTCS on day 5 and received oral diazepam; no
further clusters occurred during the observation period.
Patient 7 had 7 GTCS on day 49 and was treated for
status epilepticus with IV diazepam and PHT. Patient
9 was given IV diazepam and PHT on day 25 after 4
GTCS; while the formal observation period stopped at
that time, chronic PHT therapy was not maintained,
and the patient had no further clusters during the ensu-
ing 6 weeks of CBZ monotherapy. Patient 16 had 3
discrete 2-day clusters of 3 or more GTCS per day; these
occurred on days 9 and 10, 14 and 15,and 55 and 56.
Comparison of these study days with the corresponding
first days of undetectable PHT levels shown in figure 1
confirms that, of these 6 GTCS clusters, 2 occurred
while PHT levels were falling and the other 4 took place
3, 9, 28, and 42 days after undetectable levels were
reached.
Discussion. Using several distinct though related
measures, we failed to find an association between sei-
zure exacerbation and falling PHT levels. Over the en-
tire observation period, there was a strong trend toward
more frequent CPS and GTCS after levels became un-
detectable than either early or late in the withdrawal
process. Thus, although we could not reject the hypoth-
esis of no difference between withdrawal and post-
withdrawal periods for the combined seizure scores, it is
highly unlikely that the true frequency would be the
opposite of that observed. In addition, each patient's
week with poorest seizure control began on average 10
days after withdrawal was complete (figure l),and the
95% confidence interval for the start of this week began
3 days after completion of withdrawal, again arguing
against an association of seizure worsening with falling
Downloaded from www.neurology.org by WILLIAM H THEODORE on April 6, 2007
PHT levels. While there were individual exceptions to
this generalization, in only 1 patient (patient 3) did a
severe exacerbation occur during withdrawal that was
not equaled or surpassed during the subsequent phase of
undetectable PHT, while 2 patients (7 and 9)had severe
exacerbations only well after the withdrawal phase.
Animal studies suggest that seizure threshold is
lowered when chronically administered PHT is ab-
ruptly discontinued.8-10 PHT withdrawal in normal rats
also produces an increase in nonseizure stereotyed be-
havior, suggesting CNS hyperactivity." In an analo-
gous human study, however, Rosenblum12 found no
clinical or electrographicevidence of seizure activity, or
other adverse effects, in 38 nonepileptic neurologic pa-
tients who received 200 to 400 mg of PHT daily for l to 8
months. Although serum levels were not measured, sev-
eral patients experienced toxicity during their PHT
course. These results differ from those obtained with
barbiturates and benzodiazepines, whose discontinua-
tion may result in seizures even in individuals with no
prior seizure history.13J4
Several studies have commented on P H T with-
drawal in patients with well-controlled seizures. Schob-
bed5 found a lesser tendency to relapse among those
withdrawn from PHT than among those previously on
barbiturates, valproic acid, or CBZ. Patients without
seizures for 2 or more years reported by Callaghan et all6
were most likely to relapse after having been withdrawn
from valproic acid and least likely after CBZ; PHT was
associated with an intermediate risk. Whether relapses
occurred during or after drug withdrawal was not ex-
plicitly stated. Of 36 patients changed from 2-drug to
single-drug therapy by Schmidt,' 5 were withdrawn
from PHT combinations, 4 from PHT/CBZ and 1 from
PHT/primidone. After discontinuation of PHT, sei-
zure frequency was decreased in 1, unchanged in 1, and
July 1989 NEUROLOGY 39 907
 PATIENT OBSERVATION PERIODS RANKED BY INTERVAL
BETWEEN END OF WITHDRAWAL AND START OF WORST
WEEK
Palienl No I
6 1
0- - I
*----‘5
I
I
I
- -- L.- - - - - - - ---- -
I I
I
I I I 1 I I
Figure 1. Each horizontal line represents the entire
observation period of an individual patient, identified by
number on the solid vertical axis. Day 0 is the 1st day on
which phenytoin was undetectable. Patients are ordered by
the interval between day 0 and the 1st day of the week with
the most seizures. Negative values indicate thut the worst
week began during withdrawal. Dashed vertical line shows
the median interval for the 17patients.
increased in 2 of the CBZ group, and increased in the
patient withdrawn to primidone; again, however, no
information was provided regarding seizures during the
actual withdrawal period. Leppik et all7reported 2 pa-
tients whose PHT was withdrawn in the setting of
combination therapy with CBZ and progabide; the au-
thors refer to a transient increase in seizure frequency
associated with a decrease in PHT levels from 33% of
baseline to zero, but details are not given.
Two groups have looked at seizure exacerbations in
patients withdrawn sequentiallyfrom multiple drugs as
a part of an evaluation for surgical treatment. Of 3
patients whose PHT was abruptly discontinued by
Spencer et al,5 2 had a total of 4 seizures during the 12
defined withdrawal days (day 2 through day 5 after
cessation of PHT for each of the 3 patients), in contrast
to 6 seizures occumng during the days of steady thera-
peutic or undetectable levels; the study focused on type
and localization of seizures rather than on frequency,
however, and the length of the baseline period was not
given. Marciani et a16defined PHT withdrawal periods
as the time between 1and 4 22-hour half-lives after an
abrupt decrease to a lower dose, and found that 6 of 19
such periods were accompaniedby GTCS, as opposed to
2 of 19 instances where seizures occurred only after
levels were presumably stable or undetectable; appar-
ently, the 11remaining withdrawal periods were neither
accompanied nor followed by GTCS. Co-medication
908 NEUROLOGY 39 July 1989
Downloaded from www.neurology.org by WILLIAM H THEODORE on April 6, 2007
TIME TO WORST WEEK STRATIFIED BY LENGTH OF
WITHDRAWAL PERIOD
4 8 12 16 20 24 28
”
32 36 40
-
44
DAYS AFTER BEGINNING PHT TAPER
i Wifhdrawal mmpleled in -9 days (n = 9)
:
0 Withdrawal mrnplelec In 210 days (n = 8 )
Figure 2. Actuarial survival curves using start of the week
with the most seizures as endpoint. Open circles represent 9
patients with shortest withdrawal periods (median, 6 days;
range, 5 to 9 days), closed circles 8 patients with longest
withdrawal (median, 14 days; range, 10 to 21 days). There is
no significant difference in time to worst week (p = 0.30).
was not standardized, however, and seizure frequencies
during withdrawal and postwithdrawal phases were not
directly compared. These studies show that patients do
have seizures while PHT levels are falling; however, the
studies do not clearly demonstrate that more seizures
occur when blood levels are falling than during an ap-
propriate comparison period.
Our results provide no evidence for PHT withdrawal
seizures. This may be contrasted with barbiturate with-
drawal, where a statistically significant rise in seizure
rates occurred when levels passed through the range of
15 to 20 mg/l.18 It is possible that the concurrent use of
CBZ in the present study protected against seizures
that would otherwise have occurred, especially since
CBZ levels rose. However, neither PHT nor CBZ pro-
tects against barbiturate ~ i t h d r a w a l . ~ ~ J ~
The effects of drug withdrawal could persist beyond
the period of detectable blood levels. Brain-plasma dif-
ferences in PHT concentration or kinetics, however,
are insufficient to produce such a phenomenon; the
brain:plasma ratio is approximately 1.5 (less in gray
matter), and PHT freely diffuses from the CNS when
plasma levels Another possibility is that PHT
could induce changes in structural or biochemical neu-
ronal elements, such as membrane transport channels
or neurotransmitters,21*22 that could take days or weeks
to reverse. Thus, although the withdrawal syndrome of
opioids, barbiturates, and benzodiazepines usually be-
gins within 4 half-lives after discontinuation, it may
continue for days or weeks beyond that point.14Figure 1,
however, shows no convincing peak period of vul-
nerability. Assuming that the observed seizure frequen-
cies were representative of these patients’ ongoing
experiencewith CBZ monotherapy, the apparent wors-
ening would reflect only loss of therapeutic PHT effect.
The implication that the CBZ/PHT combination may
be more efficaciousfor some patients than CBZ alone is
contro~ersial,2~.~~and the disagreement cannot be re-
solved by the present data, since most patients were not
observed for an extended period of time on combination
therapy.
The current study suggests that PHT may be rapidly
withdrawn in relative safety if CBZ is first introduced
and brought to a therapeutic level. Patients with intrac-
table epilepsy may show subsequent seizure exacerba-
tions, but these are as likely to occur after, as during,
drug taper; careful follow-up beyond the immediate
withdrawal period is required. The existence of a PHT
withdrawal syndrome in humans has yet to be demon-
strated.
Acknowledgments
Ms. Patricia Reeves provided invaluable assistance with record-keep-
ing. The 5-West nursing staff of the NIH Clinical Center made the
study possible by verifying seizure occurrence and assisting in care of
the patients.
References
1. Schmidt D. Reduction of two-drug therapy in intractable epilepsy.
Epilepsia 1983;24:368-376.
2. Lesser RP, Pippenger CE, Lllders H,Dinner DS. High-dose
monotherapy in the treatment of intractable seizures. Neurology
1984;34707-711.
3. Albright P, Bruni J. Reduction of polypharmacy in epileptic pa-
tients. Arch Neurol1985;42:797-799.
4. Hirtz DG, Nelson KB. Cognitive effects of antiepileptic drugs. In:
Pedley TZ, Meldnun BS, eds. Recent advances in epilepsy. New
York Churchill Livingstone, 1985:161-181.
5. Spencer SS, Spencer DD, Williamson PD, Mattson RH. Ictal
effects of anticonvuleant medication withdrawal in epileptic pa-
tients. Epilepsia 1981;22297-307.
6. Marciani MG, Gotman J , Andermann F, Olivier A. Patterns of
seizup activation after withdrawal of antiepileptic medication.
Neurology 1985,351537-1543.
7. Skillings JH, Mack GA. On the use of a Friedman-type statistic in
balanced and unbalanced block designs. Technometrics
Downloaded from www.neurology.org by WILLIAM H THEODORE on April 6, 2007
1981;23:171-177.
8. DeLima TCM, Neto JP. Central nervous system supersensitivity
and withdrawal from long-term diphenylhydantoin treatment.
Brazilian J Med Biol Res 1983;16:241-245.
9. DeLima TCM, Neto JP. Effects of withdrawal from long-term
diphenylhydantoin treatment on audiogenic and maximal elec-
troshock-induced seizures in rats. Acta Neurol Scand
1981;63189-196.
10. Craig CR, Chiu P, Colasanti BK. Effects of diphenylhydantoin
and trimethadione on seizure activity during cobalt experimental
epilepsy in the rat. Neuropharmacology 1976;15:485-489.
11. DeLima TCM, Neto JP. The effects of diphenylhydantoin on rat
behavior. Psychopharmacology 1980,69183-185.
12. Rosenblum JA. Diphenylhydantoin sodium (Dilantin) with-
drawal in non-epileptics. Curr Ther Res 1971;13:369-373.
13. Essig CF. Clinical and experimental aspects of barbiturate with-
drawal convulsions. Epilepsia 1967821-30.
14. Petursson H,Lader M. Dependence on tranquilizers. New York
Oxford University Press,1984:6-36.
15. Schobben AFAM. Pharmacokinetics and therapeutics in epi-
lepsy. Nijmegen, The Netherlands: Stichting Studentenpers,
1979~251-258.
16. Callaghan N, Garrett A, Coggin G. Withdrawal of anticonvuleant
drugs in patients free of seizures for two years. N Engl J Med
1988;318942-946.
17. Leppik IL, Brundage RC, Krall R, et al. Double-blind withdrawal
of phenytoin and carbamazepine in patients treated with pro-
gabide for partial seizures. Epilepsia 1986,27:563-568.
18. Theodore WH, Porter RJ,Raubertas RF. Seizuresduring barbitu-
ra te withdrawal: relation t o blood level. Ann Neurol
198722644-617.
19. Essig CF, Carter WW. Failure of diphenylhydantoin in preventing
barbiturate withdrawal convulsions in the dog. Neurology
1962;12:481-484.
20. Woodbury DM. Phenytoin: absorption, distribution, and excre-
tion. In: WoodburyDM, Penry JK, Pippinger CE, eds.Antiepilep-
tic drugs, 2nd ed. New York: Raven Press, 1982:191-207.
21. Eadie MJ, Tyrer JH. Anticonvulsant therapy: pharmacologic
baais and practice, 2nd ed. Edinburgh Churchill Livingstone,
1980:45-161.
22. Woodbury DM. Phenytoin: mechanisms of action. In: Woodbury
DM, Penry JK, Pippinger CE, eds. Antiepileptic drugs, 2nd ed.
New York Raven Press,1982269-282.
23. Lorenu, NY,Bromfield EB, Theodore WH. Phenytoin and car-
bamazepine: combination versus single-drugtherapy for intracta-
ble partial seizures. Abstract. Ann Neurol1988,24:136.
24. Morris JC, Dodson WE, Hatlelid JM, Ferrendelli JA. Phenytoin
and carbamazepine, alone and in combination: anticonvuleant
and neurotoxic effects. Neurology 1987;37:1111-1118.
July 1989 NEUROLOGY 39 909
 Phenytoin withdrawal and seizure frequency
E. B. Bromfield, J. Dambrosia, O. Devinsky, F. J. Nice and W. H. Theodore
Neurology 1989;39;905-
This information is current as of April 6, 2007

Updated Information including high-resolution figures, can be found at:

& Services http://www.neurology.org
Permissions & Licensing Information about reproducing this article in parts (figures,
tables) or in its entirety can be found online at:
http://www.neurology.org/misc/Permissions.shtml
Reprints Information about ordering reprints can be found online:
http://www.neurology.org/misc/reprints.shtml

Downloaded from www.neurology.org by WILLIAM H THEODORE on April 6, 2007

View publication stats