Patients receiving anticonvulsants are resistant to nondepolarizing muscle relaxants (NDMR). This study examines the effect of plasma anticonvulsant levels on pipecuronium-induced neuromuscular blockade. Twenty adult patients scheduled for neurosurgery were assigned to one of two groups. Group 0 (G0) consisted of 10 patients not on anticonvulsant therapy; group 1 (G1) included 10 patients treated either with phenytoin or carbamazepine. G1 patients were further divided into G1u (n = 4) and G1w (n = 6) subgroups, according to the plasma anticonvulsant level measured the day before surgery--under (G1u) or within (G1w) the therapeutic range. Neuromuscular transmission was monitored with a Biometer International A/S Accelograph. Anesthesia was induced and maintained using propofol and sufentanil. After calibration of the accelograph, a bolus of pipecuronium 0.08 mg/kg was given IV. The time from pipecuronium injection to the peak reduction of T1 was taken as the onset time. The time in min from pipecuronium injection to recovery of T1% (first accelograph response/baseline response) x 100 and TR% (fourth accelograph response/first accelograph response) x 100 were recorded at 25, 50, and 75% of baseline. The recovery index (RI) was taken as the time from 25 to 75% of baseline. The recovery index (RI) was taken as the time from 25 to 75% recovery of the baseline response. The onset time was not different in G0 (203 +/- 60.4 s), G1 (230.5 +/- 79.3 s), and G1u (181.8 +/- 60.4 s) but prolonged in G1w (279.2 +/- 67.7 s).(ABSTRACT TRUNCATED AT 250 WORDS)