Carbamazepine is metabolized by CYP3A4 and several other cytochrome P450 enzymes. The potential effects of zonisamide on carbamazepine pharmacokinetics (PK) have not been well characterized, with contradictory literature reports. Hence, an in vitro study was designed to evaluate the cytochrome P450 inhibition spectrum of zonisamide using human liver microsomes. Further, an in vivo steady-state study was performed to measure the effect of zonisamide on carbamazepine PK in epileptic patients, and monitor zonisamide PK. In vitro human liver microsomes were incubated with zonisamide (200, 600 or 1000 microM) in the presence of appropriate probe substrates to assess selected cytochrome P450 activities. In vivo, the effect of zonisamide, up to 400 mg/day, on the steady-state PK of carbamazepine and carbamazepine-epoxide (CBZ-E) was studied in 18 epileptic patients. In vitro, zonisamide did not inhibit CYP1A2 and 2D6, and only weakly inhibited CYP2A6, 2C9, 2C19, and 2E1. The estimated Ki for zonisamide inhibition of CYP3A4 was 1076 microM, 12 times higher than typical unbound therapeutic serum zonisamide concentrations. In vivo, no statistically significant differences were observed for mean Cmax, Tmax, and AUC0-12 of total and free carbamazepine and CBZ-E measured before and after zonisamide administration (300-400 mg/day for 14 days). However, CBZ-E renal clearance was significantly (p < 0.05) reduced by zonisamide. The observed mean zonisamide t1/2 (36.3h), relative to approximately 65 h reported in subjects on zonisamide monotherapy, reflects known CYP3A4 induction by carbamazepine. Based on the lack of clinically relevant in vitro and in vivo effects, adjustment of carbamazepine dosing should not be required with concomitant zonisamide administration.