Antiepileptic Drugs, 5th Edition



Drug Interactions

Barry E. Gidal PharmD

Associate Professor, School of Pharmacy and Department of Neurology, University of Wisconsin, Madison, Wisconsin

Treatment with the traditional antiepileptic drugs (AEDs) such as phenytoin (PHT), carbamazepine (CBZ), phenobarbital, and sodium valproate (VPA) is complicated by clinically significant, pharmacokinetic interactions. A potential advantage of the newer-generation AEDs is an improved drug interaction profile.

Topiramate (TPM) is a sulfamate-substituted monosaccharide compound. Because TPM is frequently used in combination with other AEDs, it is important to evaluate potentially reciprocal interactions between this drug and the traditional medications. In addition, given the increased use of many of the newer AEDs in conditions other than epilepsy such as bipolar-affective disorder, evaluation of the potential interactions between TPM and various psychiatric medications is also warranted. Although TPM is generally considered to have an improved drug interaction profile, clinically evident pharmacokinetic interactions have been reported.


The clinically relevant pharmacokinetic properties of TPM are summarized here to form a basis for discussion and evaluation of TPM drug interactions. After oral administration, TPM is rapidly (time to maximum concentration or Tmax of 1.75 to 4.3 hours) and extensively absorbed, with an oral bioavailability of 80% (1,2). Dose-proportionality studies (100 mg to 1,200 mg) carried out in healthy volunteers suggest that increases in maximum concentration (Cmax) and area under the curve (AUC), although linear, are not dose proportional (3). Although ingestion with a high-fat meal slows TPM absorption (1.4 versus 3 hours at a 100-mg dose and 2.7 versus 4.8 hours at a 400-mg dose), overall completeness of absorption was not impaired (3). Changes in rate of absorption after food ingestion as well as an apparent prolongation in Tmax with increasing doses are unlikely to be clinically relevant.

Binding to plasma proteins is minimal for TPM (13% to 17%) (1), thus, one would not predict that TPM would be subject to protein binding displacement interactions. The apparent volume of distribution is 0.6 to 0.8 L/kg. Values for apparent volume of distribution do appear to be inversely related to TPM dose; saturable binding of TPM to erythrocytes may provide a possible explanation for this observation, as well as the apparent lack of dose proportionality seen at higher dose ranges (4).

When TPM is given as monotherapy, the fraction of TPM that is metabolized is relatively low. In this setting, ~70% of an oral TPM dose is excreted unchanged in the urine. The fraction metabolized is substantially increased, however, when TPM is coadministered with enzyme-inducing AEDs, a finding suggesting that, in certain polytherapy settings, significant drug interactions may exist (see later). Metabolites thus far identified include two hydroxy and two diol metabolites, as well as several glucuronide conjugates (1). TPM elimination half-life in monotherapy ranges from 19 to 23 hours, versus 12 to 15 hours when TPM is given with enzyme-inducing drugs.


To characterize the potential inhibitory activity of TPM, in vitro studies using human liver microsomes were performed (5). Substrates metabolized by seven specific cytochrome P450 isozymes—GYP1A2, CYP2A6, CYP2C9, CYP2C19, CYP2D6, CYP2E1, and CYP3A4—were incubated with TPM at concentrations ≤1,000 µmol/L. Significant inhibition was not seen for any CYP isoform except CYP2C19. Formation of 4-OH-(S)-mephenytoin was reduced by 29%. Based on these experimental findings, it can therefore be predicted that TPM coadministration would only be expected to result inhibitory interactions for drugs in which CYP2C19 is responsible for a significant fraction of substrate metabolism.



The potential for interactions between TPM and CBZ, PHT, and VPA was evaluated in a series of studies in patients receiving one of these three AEDs as monotherapy. Each of these studies was conducted using a common study design (Table 79.1). Determination of pharmacokinetic parameters for each of these traditional AEDs was conducted at baseline and after the introduction of TPM during a stepwise dose escalation over a 6-week period to a maximum possible dose of 800 mg/day. Specifically, pharmacokinetic analysis of the original AED was performed after stabilization at TPM doses of 100, 200, and 400 mg every 12 hours given for 2 weeks at each target dose. Two weeks after the TPM dose escalation period, the background AED was systematically withdrawn in 25% weekly decrements. TPM pharmacokinetic parameters were evaluated before and after withdrawal of the background AED. Using this unique clinical design, evaluation of potential reciprocal interactions between TPM and the background AED could be determined.

Carbamazepine and Topiramate

In an early clinical report, Wilensky et al. (6) noted no significant effects of adjunctive treatment with TPM on CBZ or CBZ-10,11-epoxide (CBZE) plasma concentrations. In a more recent evaluation (7), Sachdeo and coworkers reported that average steady-state plasma concentrations of both total and unbound CBZ and its principal metabolite, CBZE, were unchanged among baseline values, during TPM dose escalation, and after attainment of maximal TPM doses in 11 evaluable patients. Likewise, specific pharmacokinetic parameters including AUC(0-8), Cmax, Tmax, and apparent oral clearance (Cl/F) were unchanged by comedication with TPM. Patients in this trial were receiving CBZ in doses of 300 to 800 mg every 8 hours.

Changes in TPM pharmacokinetics were evaluated in 10 of these patients at the beginning and at the end of the CBZ withdrawal. TPM Cl/F was again evaluated in patients in whom TPM monotherapy was successfully maintained. During CBZ dose reduction, TPM Cl/F was reduced by 22% when CBZ daily dose was reduced by approximately 61%. In the three patients who achieved TPM monotherapy, mean TPM plasma concentrations were substantially altered after withdrawal of CBZ. Specifically, TPM Cmax (3.4±1.4 versus 5.5±0.6 µg/mL) and Cmin (2.2±0.4 versus 3.7±0.9 µg/mL) mean concentrations were ~60% to 68% higher during TPM monotherapy as compared with comedication with CBZ. Likewise, TPM apparent Cl/F was reduced by ~50% during monotherapy as compared with baseline polytherapy values (33.2±4.8 versus 63.7±33.5 mL/min). TPM renal clearance was unaffected by comedication with CBZ. In addition, the mean fraction of TPM dose excreted unchanged in urine was approximately 40% less during concomitant treatment with CBZ as compared with monotherapy. These findings suggest that effect of CBZ on TPM apparent Cl/F is the result of enzyme induction.


Antiepileptic Drug

Change in Antiepileptic Drug Plasma Concentration

Change in Topiramate Plasma Concentration



40% decrease


0-25% increase

48% decrease


13% decrease

14% decrease

Phenytoin and Topiramate

In an earlier brief report, Floren et al. (8) reported no apparent change in PHT plasma concentrations in six patients receiving adjunctive treatment with TPM. Conversely, when TPM (200 to 800 mg/day) was added to 12 patients receiving PHT monotherapy (six patients receiving PHT 130 to 300 mg every 12 hours and six patients receiving 360 to 480 mg PHT once daily), using the study design described previously, six of 12 patients had increases in PHT AUC of approximately 25% as compared with baseline values (9,10). This somewhat variable interaction is consistent with TPM ability to inhibit CYP2C19. Although these modest elevations in PHT plasma concentrations are not likely to be clinically significant in most patients, clinicians should nevertheless be alert for signs of PHT intoxication in patients receiving this AED combination.

In this study, the effect of PHT on TPM pharmacokinetics was compared with TPM monotherapy. TPM plasma concentrations are ~48% lower during comedication with PHT. Likewise, TPM apparent Cl/F is ~52% to 59% lower during TPM monotherapy as compared with adjunctive treatment with PHT. Definitive statistical analysis of these data are confounded, however, because only four patients were successfully converted to TPM monotherapy. However, these data suggest that, after discontinuation of PHT, one would expect to see an approximate doubling of TPM plasma concentrations.

Valproate and Topiramate

Rosenfeld and colleagues (11) evaluated VPA and TPM pharmacokinetics in 12 patients receiving monotherapy with VPA after the addition of TPM; the previously described clinical design was used. In this study, all 12 patients achieved the final target TPM dose of 800 mg/day, with data from 10 patients evaluable for VPA pharmacokinetics.


VPA doses ranged from 1,000 to 4,500 mg/day. After addition of TPM, statistically significant increases in VPA clearance were noted at all TPM doses. At a daily TPM dose of 800 mg/day, VPA apparent Cl/F was approximately 13% greater and VPA average steady-state concentrations were approximately 12% lower. Although the total percentage of the VPA dose excreted in urine was unaffected by concomitant TPM treatment, the mean formation clearance of the glucuronide conjugate pathway was decreased by 35%, whereas significant increases in the formation clearances of 4-ene VPA (+60%), ω-oxidation (+36%), and β-oxidation (+42%) pathways were seen. Although the overall effects of TPM on VPA pharmacokinetics is only modest, and will most likely not require VPA dosage adjustments, the changes in fractions of VPA metabolic pathways are noteworthy. In particular, 4-ene VPA has been implicated as a potential hepatotoxin (12).

While the clinical significance of these changes is still unclear, however, although the combination of TPM and VPA was implicated in the development of hyperammonemic encephalopathy in several patients (13). With respect to the effect of VPA on TPM pharmacokinetics. Data from eight evaluable patients receiving TPM 800 mg/day suggested that TPM Cmax (5.8±0.8 versus 6.8±1.1 µg/mL, p < .05) and Cmin (3.9±0.6 versus 4.6±0.9 µg/mL, p < .05) values were ~17% to 18% higher during concomitant treatment with VPA as compared with TPM monotherapy. TPM Cl/F was 13% greater during comedication with VPA (29.8±4.2 versus 25.9±4.6 mL/min). TPM renal clearance, however, was unaffected by VPA treatment. These findings taken together suggest induction of TPM metabolism by VPA. These changes are quite modest, however, and would not be expected to be clinically meaningful.


Evaluation of phenobarbital and primidone plasma concentrations during several placebo-controlled clinical trials suggested no significant pharmacokinetic effect of TPM comedication on these drugs (14). Currently, prospective data are lacking regarding potential pharmacokinetic interactions between TPM and other newer AEDs such as lamotrigine, gabapentin, zonisamide, or oxcarbazepine. Based on TPM's known in vitro interaction profile, as well as results from clinical pharmacokinetic trials, one would not predict interactions between TPM and gabapentin, or levetiracetam, drugs that are eliminated by predominantly renal mechanisms and non hepatic metabolism. Preliminary evidence also suggests that no interaction exists between TPM and the investigational agent, retigabine (15). Given that both zonisamide and TPM are modest carbonic anhydrase inhibitors, the potential for additive adverse effects such as nephrolithiasis exists.

In one case series, 4 of 7 patients comedicated with lamotrigine and topiramate had lamotrigine concentrations 40%-50% lower as compared to monotherapy values (16).


Increasingly, some AEDs, including TPM, are being used in settings other than epilepsy, including adjunctive treatment of bipolar-affective disorder. At present, however, only limited data exist regarding the potential for pharmacokinetic interactions between TPM and various psychotropic drugs.

Topiramate and Haloperidol

An open-label crossover study evaluated the effect of TPM on haloperidol pharmacokinetics in 12 healthy volunteers (17). Subjects received a single 2-mg dose of haloperidol before and after receiving TPM, titrated to a maximal dose of 200 mg/day over 14 days. Mean haloperidol AUC increased by 15%, with the greatest individual increase in haloperidol AUC being 28%. The authors concluded that this modest increase in haloperidol is unlikely to be of clinical significance in most patients.

Topiramate and Lithium

Results from one study in 12 healthy adults suggest that, after the addition of TPM, serum lithium concentrations declined approximately 11% to 16% (unpublished data, Ortho-McNeil Pharmaceutical). In one patient, a 70% reduction in lithium AUC was noted after TPM administration. Mechanistically, these modest reductions in lithium concentrations may reflect TPM's ability to inhibit carbonic anhydrase activity weakly. The administration of acetazolamide has been reported to increase lithium excretion by ~30%, and significantly reduced lithium concentrations have been reported (18). Although the clinical significance of this interaction with TPM is unclear, clinicians using this combination would be advised to monitor lithium concentrations after TPM administration during medication introduction.


Oral Contraceptives

Understanding the potential for interactions between AEDs and oral contraceptive drugs is clinically imperative. Although many of the newer AEDs appear to have a reduced capacity for interactions with oral contraceptives, TPM coadministration may have clinically meaningful effects.



The effect of TPM administration on oral contraceptive kinetics was evaluated in 12 women with epilepsy who were receiving VPA monotherapy (19). Patients received a combination product containing norethindrone 1.0 mg and ethinyl estradiol 35 µg daily for 21 days, followed by inert pills to complete a 28-day cycle. Hormonal concentrations were evaluated on day 20 for four cycles. TPM was coadministered at escalating doses of 200 to 800 mg/day, starting on cycle 2.

Although TPM coadministration had no significant effect on plasma concentrations of norethindrone, norethindrone Cl/F was increased 22% at the highest TPM dose. Plasma concentrations of ethinyl estradiol significantly declined in a dose-related manner. Specifically, at TPM doses of 200, 400, and 800 mg/day, estradiol concentrations declined by 18%, 21%, and 30%. At TPM doses of 200 and 800 mg/day, mean estradiol Cl/F was significantly increased by 14.7% and 33%, respectively, as compared with baseline values.

Although the magnitude of this interaction is substantially less than that seen with either PHT or CBZ, these findings nonetheless suggest that oral contraceptive effectiveness may be compromised in the presence of TPM. Patients receiving preparations with <35 µg estrogenic component should be advised to report changes in bleeding patterns. Alternatively, substitution of an oral contraceptive drug with higher estrogenic dosage may be appropriate in some patients.

Topiramate and Digoxin

Digoxin pharmacokinetics was determined in healthy persons before and after 10-day treatment with TPM at 200 mg/day (20). As compared with baseline values, treatment with TPM resulted in a 13% decline in digoxin Cl/F. There were no significant changes noted in digoxin elimination half-life or renal clearance. Based on this study design, it is unclear whether this interaction is dose dependent or whether it is a result of enhanced digoxin clearance or reduced bioavailability. Although this interaction appears to be modest, because of digoxin's narrow therapeutic index, monitoring of digoxin serum concentrations may be prudent.


As compared with the traditional AEDs, the drug interaction profile of TPM appears to be relatively modest. In the studies designed to evaluate TPM and CBZ, CBZE, and VPA, TPM comedication had negligible effects. In some patients, treatment with TPM may result in modest elevations in plasma PHT concentrations. Clinical experience would suggest, however, that this inhibitory interaction would seldom require PHT dosage adjustment. In contrast, treatment with the enzyme-inducing AEDs such as PHT or CBZ can be expected to result in a two- to threefold increase in TPM clearance. These interactions can be clinically significant, and TPM dosage reduction may be required after the removal or reduction of one of these enzyme inducers. Because of the obvious clinically important consequences of oral contraceptive failure, the interaction of these agents with TPM requires close clinical monitoring.


  1. Gamett WR. Clinical pharmacology of topiramate: a review. Epilepsia2000;41[Suppl 1]:S61-S65.
  2. Bourgeois BED. Pharmacokinetics and metabolism of topiramate. Drugs Today1999;35:43-48.
  3. Doose DR, Scott VV, Margul BL, et al. Multiple-dose pharmacokinetics of topiramate in healthy male subjects. Epilepsia1988; 29:662.
  4. Gidal BE, Lensmeyer GL. Therapeutic monitoring of topiramate: evaluation of the saturate distribution using an optimized hill method. Ther Drug Monit1999;21:567-576.
  5. Levy RH, Bishop F, Streeter AJ, et al. Explanation and prediction of drug interactions with topiramate using a GYP4SO inhibition spectrum. Epilepsia1995;36[Suppl 4]:47.
  6. Wilensky AJ, Ojemann LM, Chmelir T, et al. Topiramate pharmacokinetics in epileptic patients receiving carbamazepine. Epilepsia1989;30:645(abst).
  7. Sachdeo RG, Sachdeo SK, Walker SA, et al. Steady-state pharmacokinetics of topiramate and carbamazepine in patients with epilepsy during monotherapy and concomitant therapy Epilepsia1989;30:646(abst).
  8. Floren KL, Graves NM, Leppik IE, et al. Pharmakokinetics of topiramate in patients with partial epilepsy receiving phenytoin or valproate. Epilepsia1989;30:646(abst).
  9. Gisclon LG, Curtin GR, Kramer LD, et al. The steady-state (SS) pharmacokinetics (PK) of phenytoin (Dilantin) and topiramate (Topamax) in epileptic patients on monotherapy, and during combination therapy. Epilepsia1994;35[Suppl 8]:54.
  10. Gisclon LG, Curtin GR, Kramer LD, et al. A comparative study of the steady-state pharmacokinetics of phenytoin (Dilantin) kapseals and topiramate (Topamax) in male and female patients on monotherapy and during combination therapy. Pharm Res1994;11[Suppl 8]:54.
  11. Rosenfeld WE, Liao S, Kramer LD, et al. Comparison of the steady-state pharmacokinetics of topiramate and valproate in patients with epilepsy during monotherapy and concomitant therapy. Epilepsia1997;38:324-333.
  12. Levy RH, Rettenmeier AW, Anderson GD, et al. Effects of polytherapy with phenytoin, carbamazepine, and stiripentol on formation of 4-ene-valproate, a hepatotoxic metabolite of valproic acid. Clin Pharmacol Ther1990;48:225-235.
  13. Hamer HM, Knake S, Schomburg U, et al. Valproate-induced hyperammonemic encephalopathy in the presence of topiramate. Neurology 2000;54:230-232.
  14. Doose DR. Walker SA, Pledger G, et al. Evaluation of phenobarbital and primidone/phenobarbital (primidones active metabolite) plasma concentrations during administration of add-on topiramate therapy in five multicenter, double-blind, placebo-controlled trials in outpatients with partial seizures. Epilepsia1995;36[Suppl 3]:158(abst).



  1. Sachdeo RG, Ferron GM, Partiot AM, et al. An early determination of drug-drug interaction between valproic acid, phenytoin, carbamazepine or topiramate and retigabine in epileptic patients. Neurology2001;(abst).
  2. Wnuk W, Volanski A, Foletti G. Topiramate decreases lamotrigine concentrations. Ther Drug Monit1999;21:4-449.
  3. Doose DR, Kohl KA, Desai-Krieger D, et al. No clinically significant effect of topiramate on haloperidol plasma concentration. Eur Neuropsychopharmacol1999;9[Suppl 5]:S357.
  4. Thomsen K, Schou M. Renal lithium excretion in man. Am J Physiol1968;215:823-827.
  5. Rosenfeld WE, Doose DR, Walker SA, et al. Effect of topiramate on the pharmacokinetics of an oral contraceptive containing norethindrone and ethinyl estradiol in patients with epilepsy. Epilepsia1997;38:317-323.
  6. Liao S, Palmer M. Digoxin and topiramate drug interaction study in male volunteers. Pharm Res1993;10[Suppl]:405.