The aim of the current study was to investigate whether there are differences in amplitudes and intrapatient variability of motor evoked potentials to five pulses of transcranial electrical stimulation between ketamine/N2O- and propofol/N2O-based anesthesia. Patients in the propofol group (n = 13) and the ketamine group (n = 13) were anesthetized with 50% N2O in oxygen, fentanyl, and 4 mg/kg/hr of propofol or 1 mg/kg/hr of ketamine, respectively. The level of neuromuscular blockade was maintained at an M-response amplitude of approximately 50% of control. Motor evoked potentials in response to multipulse transcranial electrical stimulation were recorded from the right adductor pollicis brevis muscle, and peak-to-peak amplitude and onset latency of motor evoked potentials were evaluated. To estimate intrapatient variability, the coefficient of variation (standard deviation/mean × 100%) of 24 consecutive responses was determined. Motor evoked potential amplitudes in the ketamine group were significantly larger than in the propofol group (mean, 10th–90th percentile: 380 &mgr;V, 129–953 &mgr;V; 135 &mgr;V, 38–658 &mgr;V, respectively;P< .05). There were no significant differences in motor evoked potential latency (mean ± standard deviation: 20.9 ± 2.2 msec and 21.4 ± 2.2 msec, respectively) and coefficient of variation of amplitudes (median [range]: 32% [22–42%] and 26% [18–41%], respectively) and latencies (mean ± standard deviation: 2.1 ± 0.7% and 2.1 ± 0.7%, respectively) between the ketamine and propofol groups. In conclusion, intrapatient variability of motor evoked potentials to multipulse transcranial stimulation is similar between ketamine/N2O- and propofol/N2O-based anesthesia, although motor evoked potential amplitudes are lower during propofol/N2O-based anesthesia than ketamine/N2O-based anesthesia.