Intraoperative recording of myogenic motor potentials evoked by transcranial electrical stimulation (tcMEP) is a method of monitoring the integrity of the vulnerable motor pathways during thoracoabdominal aortic aneurysm (TAAA) surgery. Deflation of the left lung during TAAA surgery may result in impairment of arterial oxygenation. Ventilation with nitrous oxide may cause further desaturation. We studied the effects of 20%, 40% and 60% nitrous oxide in oxygen on within-patient variability and magnitude of tcMEP in response to six pulse transcranial electrical stimulation during fentanyl-low-dose propofol anaesthesia with partial neuromuscular block. Ten patients (two females; aged 63-74 yr) were studied. After achieving a stable anaesthetic state and before surgery, 10 tcMEP were recorded from the right tibialis anterior muscle during addition of 20%, 40% and 60% nitrous oxide in oxygen in random order. When ventilation with 40% or 60% nitrous oxide in oxygen was performed, there was 50-70% depression of tcMEP amplitude (P < 0.05) and 40-60% reduction in tcMEP area under the curve (P < 0.05) compared with 20% nitrous oxide in oxygen. There was no significant difference in the coefficients of variation for tcMEP between the three nitrous oxide anaesthetic regimens. Our results suggest that increasing doses of nitrous oxide reduce the MEP waveform to six pulse transcranial electrical stimulation, but even with 60% nitrous oxide in oxygen, the tcMEP were recordable and as reproducible as with 20% and 40% nitrous oxide regimens. The method is sufficiently robust for use in aortic surgery.