We studied the feasibility of recording motor evoked responses to transcranial electrical stimulation (tce-MERs) during partial neuromuscular blockade (NMB). In 11 patients, compound muscle action potentials were recorded from the tibialis anterior muscle in response to transcranial electrical stimulation during various levels of vecuronium-induced NMB. The level of NMB was assessed by accelerometry of the adductor pollicis muscle after train-of-four stimulation of the ulnar nerve. The compound muscle action potential was also recorded from the tibialis anterior muscle after direct stimulation of the peroneal nerve (M-response) as an alternative means of assessing the degree of NMB. In all patients, tce-MERs could be recorded reliably during anesthesia with N2O and a continuous infusion of sufentanil (0.5 micrograms.kg-1.h-1). An intact train-of-four was present in all patients, and the amplitude of the first twitch was recorded and designated as the control value. Before administration of vecuronium, the M-response amplitude was 9.6 +/- 3.6 (mean +/- SD) mV, and the tce-MER amplitude was 1.21 +/- 0.66 mV. Although administration of vecuronium (0.05 mg/kg) resulted in loss of the mechanical adductor pollicis response in 8 of the 11 patients, the M-response and the tce-MER remained recordable. Subsequently, during an infusion of vecuronium, adjusted to maintain one or two mechanical responses to train-of-four stimulation, the average M-response to peroneal nerve stimulation was 5.2 +/- 2.5 mV (53% of the control value), and tce-MER amplitude was 0.59 +/- 0.36 mV (59% of the control value).(ABSTRACT TRUNCATED AT 250 WORDS)