We have studied the effects of nitrous oxide on EEG burst suppression patterns during stable isoflurane anaesthesia in 13 ASA I patients. After induction of anaesthesia with propofol, the concentration of isoflurane was increased with continuous EEG monitoring to burst suppression level (mean end-tidal concentration of isoflurane, 1.7 (SD 0.2)%), and kept constant during the study. During surgery, isoflurane in air and oxygen (FIO2 0.35), or isoflurane in 65% nitrous oxide in oxygen were given to each patient for 30 min, in random order. EEG was recorded and digitized off-line. The proportion of EEG suppression time was measured after a washin or washout period of at least 15 min for nitrous oxide. There was a significant decrease in the proportion of EEG suppression time (from 69.5 to 43.7%) when air was replaced by nitrous oxide. We conclude that the EEG effects of isoflurane and nitrous oxide are not additive and that nitrous oxide opposes the depression of isoflurane on the central nervous system.