Transcranial motor evoked potentials (tc-MEPs) are used to monitor spinal cord integrity intraoperatively. We compared myogenic motor evoked responses with electrical and magnetic transcranial stimuli during nitrous oxide/opioid anesthesia. In 11 patients undergoing spinal surgery, anesthesia was induced with i.v. etomidate 0.3 mg/kg and sufentanil 1.5 microg/kg and was maintained with sufentanil 0.5 microg x kg(-1) x h(-1) and N2O 50% in oxygen. Muscle relaxation was kept at 25% of control with i.v. vecuronium. Electrical stimulation was accomplished with a transcranial stimulator set at maximal output (1200 V). Magnetic transcranial stimulation was accomplished with a transcranial stimulator set at maximal output (2 T). Just before skin incision, triplicate responses to single stimuli with both modes of cortical stimulation were randomly recorded from the tibialis anterior muscles. Amplitudes and latencies were compared using the Wilcoxon signed rank test. Bilateral tc-MEP responses were obtained in every patient with electrical stimulation. Magnetic stimulation evoked only unilateral responses in two patients. With electrical stimulation, the median tc-MEP amplitude was 401 microV (range 145-1145 microV), and latency was 32.8 +/- 2.3 ms. With magnetic stimulation, the tc-MEP amplitude was 287 microV (range 64-506 microV) (P < 0.05), and the latency was 34.7 +/- 2.1 ms (P < 0.05). We conclude that myogenic responses to magnetic transcranial stimulation are more sensitive to anesthetic-induced motoneural depression compared with those elicited by electrical transcranial stimulation. CONCLUSIONS Transcranial motor evoked potentials are used to monitor spinal cord integrity intraoperatively. We compared the relative efficacy of electrical and magnetic transcranial stimuli in anesthetized patients. It seems that myogenic responses to magnetic transcranial stimulation are more sensitive to anesthetic-induced motoneural depression compared with electrical transcranial stimulation.