Drug Interactions & Pharmacokinetics
Drug interactions are among the most challenging aspects of epilepsy management. Antiseizure medications (ASMs) span the full spectrum of pharmacokinetic behavior—from potent enzyme inducers (carbamazepine, phenytoin, phenobarbital) that reduce the efficacy of numerous co-medications, to drugs with virtually no interactions (levetiracetam, gabapentin, pregabalin). Understanding these interactions is critical for medication selection, especially in patients who take oral contraceptives, anticoagulants, chemotherapy, immunosuppressants, or transplant medications. The consequences of unrecognized interactions range from ASM failure and seizure breakthrough to toxicity of co-medications or the ASM itself. This topic systematically reviews ASM pharmacokinetic properties, enzyme effects, and clinically important interactions.
Bottom Line
- Enzyme inducers (carbamazepine, phenytoin, phenobarbital, primidone): Potently induce CYP3A4, CYP2C9, and UGT enzymes; reduce levels of oral contraceptives, warfarin, immunosuppressants, and many other drugs; should be avoided when possible in patients on complex medication regimens
- Enzyme inhibitors (valproate, felbamate, stiripentol, cenobamate, cannabidiol): Valproate inhibits UGT, doubling lamotrigine levels; cenobamate and cannabidiol inhibit CYP2C19, raising phenytoin and N-desmethylclobazam; dose adjustments of co-medications are essential
- No/minimal interaction ASMs: Levetiracetam, gabapentin, pregabalin, and lacosamide have negligible pharmacokinetic interactions—ideal for polypharmacy patients, elderly, transplant recipients, and cancer patients
- Oral contraceptive interactions: Enzyme-inducing ASMs reduce contraceptive efficacy—a failure rate approximately double that of non-users; alternative contraception or ASMs without enzyme induction are critical for women of childbearing potential
- Therapeutic drug monitoring: Most useful for phenytoin (nonlinear kinetics), lamotrigine (pregnancy, valproate interaction), and when ASM interactions are suspected
Enzyme-Inducing ASMs
Mechanism of Enzyme Induction
Enzyme-inducing ASMs upregulate hepatic cytochrome P450 (CYP) enzymes and uridine diphosphate glucuronosyltransferases (UGTs), accelerating the metabolism and reducing the plasma concentrations of drugs metabolized by these pathways. The primary inducers are:
| ASM | Enzymes Induced | Induction Potency | Key Co-medications Affected |
|---|---|---|---|
| Carbamazepine | CYP3A4, CYP2C9, CYP1A2, UGT | Potent | Oral contraceptives, warfarin, lamotrigine, valproate, perampanel, cyclosporine, tacrolimus, corticosteroids, many chemotherapy agents, statins, DOACs |
| Phenytoin | CYP3A4, CYP2C9, CYP2C19, UGT | Potent | Same as carbamazepine; also induces its own metabolism at high doses |
| Phenobarbital | CYP3A4, CYP2C9, CYP1A2, UGT | Potent | Same as carbamazepine |
| Primidone | Same as phenobarbital (25% converted to phenobarbital) | Potent | Same as phenobarbital |
| Oxcarbazepine | CYP3A4 (weak) | Weak | Oral contraceptives at doses >900 mg/d; phenytoin (via CYP2C19 inhibition) |
| Eslicarbazepine | CYP3A4 (weak) | Weak | Oral contraceptives; similar to oxcarbazepine but less pronounced |
| Topiramate | CYP3A4 (mild, at ≥200 mg/d) | Mild | Oral contraceptives at ≥200 mg/d |
| Cenobamate | CYP3A4 | Moderate | Oral contraceptives; may reduce lamotrigine levels |
| Felbamate | CYP3A4 (weak) | Weak | Carbamazepine levels (decreased); oral contraceptives |
Clinical Consequences of Enzyme Induction
- Oral contraceptive failure: Enzyme-inducing ASMs reduce estrogen and progestin levels, causing contraceptive failure; use non-hormonal contraception (copper IUD) or switch to a non-inducing ASM; the progestin-only pill and subdermal implants are also unreliable with potent inducers
- Transplant rejection: Reduced cyclosporine or tacrolimus levels can precipitate organ rejection; levetiracetam, gabapentin, or pregabalin are preferred
- Cancer therapy failure: Enzyme inducers reduce levels of many chemotherapy agents, targeted therapies, and immunotherapy co-medications; non-inducing ASMs strongly preferred
- Anticoagulant failure: Warfarin, DOACs, and many antithrombotic agents are affected; increased INR monitoring with warfarin; consider alternative anticoagulation or non-inducing ASM
- Decreased bone density: Enzyme-inducing ASMs accelerate vitamin D metabolism via CYP3A4 induction, contributing to reduced bone density and increased fracture risk with long-term use
Enzyme-Inhibiting ASMs
| ASM | Enzymes Inhibited | Key Effects on Co-medications |
|---|---|---|
| Valproate | UGT (glucuronidation), CYP2C9, epoxide hydrolase | Doubles lamotrigine levels (halve lamotrigine dose and titration rate); increases phenobarbital levels; increases carbamazepine-epoxide (active metabolite causing toxicity); increases rufinamide levels by 70% |
| Felbamate | CYP2C19, CYP1A2, β-oxidation | Increases phenobarbital, phenytoin, valproate, carbamazepine-epoxide, N-desmethylclobazam, and warfarin levels |
| Cenobamate | CYP2C19 | Increases phenytoin, phenobarbital, and N-desmethylclobazam (active metabolite of clobazam) levels—dose reductions of these agents often necessary |
| Cannabidiol | CYP2C19 | Increases N-desmethylclobazam levels (significant clinical interaction: sedation, hepatotoxicity); requires clobazam dose reduction (typically by 50%) |
| Stiripentol | CYP2C9, CYP2C19 | Increases N-desmethylclobazam and valproate levels; dose reductions of both recommended on initiation |
| Oxcarbazepine | CYP2C19 (weak, at high doses) | May raise phenytoin levels at high oxcarbazepine doses |
The Valproate-Lamotrigine Interaction: A Double-Edged Sword
- Valproate inhibits lamotrigine glucuronidation, approximately doubling the lamotrigine half-life (from ~24 hours to ~48–60 hours)
- When adding lamotrigine to valproate: start at HALF the usual dose (12.5–25 mg/d or every other day) and titrate at HALF the usual rate
- When adding valproate to existing lamotrigine: reduce lamotrigine dose by approximately 50% preemptively
- Conversely, removing valproate from a lamotrigine regimen will approximately halve lamotrigine levels—requiring lamotrigine dose increase to avoid seizure breakthrough
- Despite this interaction, the lamotrigine-valproate combination is the best-documented synergistic ASM combination, with supra-additive efficacy
ASMs With No or Minimal Interactions
| ASM | Metabolism/Elimination | Protein Binding | Interaction Potential | Clinical Advantage |
|---|---|---|---|---|
| Levetiracetam | 66% unchanged renal; 30% nonhepatic hydrolysis | Low | None or minimal | Ideal for polypharmacy, elderly, transplant, cancer patients |
| Gabapentin | 100% unchanged renal | None | None (antacids may impair absorption) | Safe in complex medical patients; dose adjust for renal impairment |
| Pregabalin | 100% unchanged renal | None | None | Same as gabapentin; superior bioavailability |
| Lacosamide | 60% hepatic (inactive metabolites); 40% unchanged renal | Low | None or minimal | Favorable for polypharmacy; IV formulation; rapid titration |
| Vigabatrin | 100% unchanged renal | None | None or minimal (weak CYP2C9 inducer) | Limited use due to visual toxicity; no hepatic interaction |
Interactions With Oral Contraceptives
This is one of the most clinically important drug interaction categories in epilepsy management, given that approximately half of people with epilepsy are female and many are of reproductive age.
| ASM Category | Effect on Hormonal Contraception | Recommended Approach |
|---|---|---|
| Potent inducers (carbamazepine, phenytoin, phenobarbital, primidone) | Significantly reduce estrogen and progestin levels; combined oral contraceptive failure rate approximately doubled | Non-hormonal contraception (copper IUD preferred); if hormonal methods necessary, use preparation with ≥50 μg ethinyl estradiol or levonorgestrel IUD; depot medroxyprogesterone (DMPA) may still be effective |
| Weak/moderate inducers (oxcarbazepine at >900 mg/d, eslicarbazepine, topiramate at ≥200 mg/d, cenobamate, felbamate, perampanel at 12 mg/d) | May reduce contraceptive efficacy, particularly at higher ASM doses | Consider copper IUD or levonorgestrel IUD; if using combined OCP, use higher-dose formulations; discuss back-up methods |
| No effect (levetiracetam, lamotrigine, valproate, lacosamide, gabapentin, pregabalin, zonisamide, brivaracetam) | Do not reduce contraceptive efficacy | Standard contraceptive methods are appropriate |
Lamotrigine and Oral Contraceptives: A Bidirectional Interaction
- Lamotrigine does NOT reduce contraceptive efficacy (estrogen levels are unaffected)
- However, estrogen increases lamotrigine clearance by inducing glucuronidation—reducing lamotrigine levels by up to 50%
- During the pill-free week of combined OCP, estrogen levels drop and lamotrigine levels rise—potentially causing side effects cyclically
- Clinical implications: Lamotrigine dose may need to be increased when starting OCP; lamotrigine levels may rise during the hormone-free interval; monthly monitoring of lamotrigine levels during pregnancy (clearance increases markedly)
- This bidirectional interaction is one reason some clinicians prefer continuous (non-cyclic) OCP use in women on lamotrigine
Interactions With Common Co-medications
| Co-medication | Interaction | Clinical Management |
|---|---|---|
| Warfarin | Enzyme inducers reduce warfarin levels; valproate and felbamate increase warfarin levels | More frequent INR monitoring when starting/stopping ASMs; consider non-inducing ASM or alternative anticoagulant |
| DOACs (rivaroxaban, apixaban) | Enzyme inducers (CYP3A4) reduce DOAC levels substantially | Avoid potent inducers with DOACs when possible; if necessary, consider DOAC level monitoring or alternative anticoagulation |
| Cyclosporine / Tacrolimus | Enzyme inducers dramatically reduce levels—risk of transplant rejection | Use non-inducing ASMs (levetiracetam, lacosamide, gabapentin, pregabalin); if inducer required, frequent level monitoring |
| Chemotherapy agents | Many agents metabolized by CYP3A4; inducers reduce efficacy | Non-inducing ASMs strongly preferred in neuro-oncology; levetiracetam is the standard choice |
| Statins | Enzyme inducers reduce simvastatin and atorvastatin levels (CYP3A4) | May need higher statin doses or switch to pravastatin/rosuvastatin (less CYP3A4 dependent) |
| Corticosteroids | Enzyme inducers accelerate dexamethasone/prednisone metabolism | May need higher steroid doses; important in neuro-oncology and transplant medicine |
| CYP3A4 inhibitors (erythromycin, fluoxetine, grapefruit juice, azole antifungals) | Increase carbamazepine levels (may cause toxicity); do NOT affect oxcarbazepine or eslicarbazepine | Monitor carbamazepine levels when adding these agents; consider oxcarbazepine as alternative |
| Amiodarone, isoniazid, fluoxetine, fluvoxamine | Inhibit phenytoin metabolism, causing accumulation and toxicity | Monitor phenytoin levels closely; small dose adjustments due to nonlinear kinetics |
Comprehensive Pharmacokinetic Reference
| ASM | Oral Bioavailability | Protein Binding | Metabolism | Half-life | Interaction Potential |
|---|---|---|---|---|---|
| Brivaracetam | Good | Low | Extensive hepatic | ~7–8 h | Moderate |
| Carbamazepine | Good | Intermediate (~75%) | Extensive hepatic (CYP3A4) | 12–17 h (post-autoinduction) | High |
| Cannabidiol | Low (improved with fat) | High (>94%) | Extensive hepatic (CYP2C19, 3A4) | 56–61 h | High |
| Cenobamate | Good (~88%) | Intermediate (60%) | Extensive hepatic | 50–60 h | High |
| Eslicarbazepine | Good | Low | ~40% hepatic, ~60% renal unchanged | 13–20 h | Moderate |
| Gabapentin | Low (dose-dependent ↓) | None | None (100% renal) | 5–7 h | None/minimal |
| Lacosamide | Good | Low | ~60% hepatic, ~40% renal unchanged | ~13 h | None/minimal |
| Lamotrigine | Good | Intermediate (~55%) | Extensive hepatic (glucuronidation) | ~24 h (12 h with inducers; 48–60 h with VPA) | Moderate |
| Levetiracetam | Good | Low | ~30% nonhepatic hydrolysis; 66% renal unchanged | 6–8 h | None/minimal |
| Oxcarbazepine | Good | Low (MHD) | Extensive hepatic | 8–10 h (MHD) | Moderate |
| Perampanel | Good | High (95%) | Extensive hepatic | ~105 h | Moderate |
| Phenobarbital | Good | Low | >70% hepatic, 20–25% renal unchanged | 80–100 h | High |
| Phenytoin | Variable | High (~90%) | Extensive hepatic, nonlinear (CYP2C9, 2C19) | ~22 h (longer with toxicity) | High |
| Pregabalin | Good | None | None (100% renal) | ~6 h | None/minimal |
| Topiramate | Good | Low | ~30% hepatic, ~70% renal unchanged | ~21 h | None/minimal |
| Valproate | Good | High (~90%) | Extensive hepatic | 13–16 h | High |
| Zonisamide | Good | Low | ~65% hepatic | ~60 h | Moderate |
When to Monitor Drug Levels
Indications for Therapeutic Drug Monitoring
- Phenytoin: Always monitor—nonlinear kinetics make dose-level prediction impossible; check free phenytoin in hypoalbuminemia, renal/hepatic failure, pregnancy, concurrent valproate, and age >50
- Lamotrigine: Monitor during pregnancy (monthly—clearance increases markedly); when adding/removing valproate or enzyme inducers; when starting/stopping oral contraceptives
- Carbamazepine: During the first month (autoinduction changes levels); when adding CYP3A4 inhibitors; when toxicity is suspected; check carbamazepine-epoxide if co-administered with valproate or felbamate
- Valproate: Check free level when total concentration is high (>80 μg/mL), in hepatic/renal disease, during pregnancy, or when combined with phenytoin
- Any ASM: To establish an individual therapeutic reference range during the period of best seizure control (obtain 2–3 trough levels); to assess adherence; when there is a change in seizure control without obvious explanation
- Levetiracetam, gabapentin, pregabalin, lacosamide: Routine monitoring generally not necessary; consider for adherence assessment or unusual clinical situations
Special Populations
Neuro-oncology
Patients with brain tumors frequently require ASMs but also receive chemotherapy, targeted agents, and corticosteroids that are metabolized by CYP3A4. Enzyme-inducing ASMs (carbamazepine, phenytoin, phenobarbital) are strongly discouraged in this population. Levetiracetam is the most widely used ASM in neuro-oncology due to its absence of drug interactions, and lacosamide is an increasingly used alternative.
Organ Transplant Recipients
Cyclosporine and tacrolimus levels are critically affected by enzyme inducers. Non-inducing ASMs are mandatory. Levetiracetam, gabapentin, pregabalin, and lacosamide are preferred.
HIV/Antiretroviral Therapy
Many antiretrovirals (protease inhibitors, non-nucleoside reverse transcriptase inhibitors) are CYP3A4 substrates. Enzyme-inducing ASMs can reduce antiretroviral efficacy and increase viral load. Non-inducing ASMs are preferred; if an inducing ASM is necessary, close viral load and CD4 monitoring is essential.
References
- Abou-Khalil B. Update on antiseizure medications 2025. Continuum (Minneap Minn) 2025;31(1):123–165.
- Patsalos PN, Spencer EP, Berry DJ. Therapeutic drug monitoring of antiepileptic drugs in epilepsy: a 2018 update. Ther Drug Monit 2018;40(5):526–548.
- Tomson T, Battino D, Bromley R. Management of epilepsy in pregnancy: a report from the International League Against Epilepsy Task Force on Women and Pregnancy. Epileptic Disord 2019;21(6):497–517.
- Sabers A, Ohman I, Christensen J, Tomson T. Oral contraceptives reduce lamotrigine plasma levels. Neurology 2003;61(4):570–571.
- Riss J, Cloyd J, Gates J, Collins S. Benzodiazepines in epilepsy: pharmacology and pharmacokinetics. Acta Neurol Scand 2008;118(2):69–86.
- Patsalos PN, Perucca E. Clinically important drug interactions in epilepsy: interactions between antiepileptic drugs and other drugs. Lancet Neurol 2003;2(8):473–481.
- Ficker DM, Privitera M, Krauss G, et al. Improved tolerability and efficacy in epilepsy patients with extended-release carbamazepine. Neurology 2005;65(4):593–595.
- Abou-Khalil B, Klein P, Shah A, et al. Tolerability of adjunctive eslicarbazepine acetate according to concomitant lamotrigine or carbamazepine use. Epilepsy Res 2018;147:80–86.
- Gidal BE, French JA, Grossman P, Le Teuff G. Assessment of potential drug interactions in patients with epilepsy: impact of age and sex. Neurology 2009;72(5):419–425.
- Verrotti A, Mencaroni E, Cofini M, et al. Valproic acid metabolism and its consequences on sexual functions. Curr Drug Metab 2016;17(6):573–581.