Isoflurane compared with nitrous oxide anaesthesia for intraoperative monitoring of somatosensory-evoked potentials.

Intraoperative monitoring of somatosensory-evoked potentials is a routine procedure. To determine the depressant effect of nitrous oxide relative to isoflurane, the authors recorded the scalp, cervical and brachial plexus-evoked responses to stimulation of the median nerve under different anaesthetic conditions. Eight subjects, age 35 +/- 6 (SD) yr, weight 68 +/- 12 kg, were studied. Following recording of awake control responses, anaesthesia was induced with thiopentone 5 mg.kg-1 and fentanyl 3 micrograms.kg-1 and was followed by succinylcholine 1 mg.kg-1. During normocapnia and normothermia, and with a maintenance infusion of fentanyl 3 micrograms.kg-1.hr-1, evoked potential recording was repeated under three different anaesthetic conditions; 0.6 MAC nitrous oxide, 0.6 MAC nitrous oxide +/- 0.6 MAC isoflurane, and 0.6 MAC isoflurane. Among the anesthetic conditions, the combination of nitrous oxide-isoflurane had the most depressant effect on the cortical amplitude (67 +/- 4% reduction, P < 0.05). Nitrous oxide decreased the cortical amplitude more than an equipotent dose of isoflurane (60 +/- 4% vs 48 +/- 7%, P < 0.05). The latency was unchanged by nitrous oxide, but increased slightly by isoflurane and isoflurane-nitrous oxide anaesthesia (1.0 and 0.9 msec respectively, P < 0.05). We conclude that somatosensory-evoked potential monitoring is feasible both during nitrous oxide anaesthesia and isoflurane anaesthesia, but the cortical amplitude is better preserved during 0.6 MAC of isoflurane alone relative to 0.6 MAC of nitrous oxide alone. The depressant effect is maximal during nitrous oxide-isoflurane anaesthesia but less than the predicted additive effect.