Nitrous oxide does not produce a clinically important sparing effect during closed-loop delivered propofol-remifentanil anaesthesia guided by the bispectral index: a randomized multicentre study†‡

Background Nitrous oxide (N2O) offers both hypnotic and analgesic characteristics. We therefore tested the hypothesis that N2O administration decreases the amount of propofol and remifentanil given by a closed-loop automated controller to maintain a similar bispectral index (BIS). Methods In a randomized multicentre double-blind study, patients undergoing elective surgery were randomly assigned to breathe 60% inspired N2O (N2O group) or 40% oxygen (AIR group). Anaesthesia depth was evaluated by the proportion of time where BIS was within the range of 40-60 (BIS40-60). The primary outcomes were propofol and remifentanil consumption, with reductions of 20% in either being considered clinically important. Results A total of 302 patients were randomized to the N2O group and 299 to the AIR group. At similar BIS40-60 [79 (67-86)% vs 76 (65-85)%], N2O slightly decreased propofol consumption [4.5 (3.7-5.5) vs 4.8 (4.0-5.9) mg kg−1 h−1, P=0.032], but not remifentanil consumption [0.17 (0.12-0.23) vs 0.18 (0.14-0.24) µg kg−1 min−1]. For the subgroups of men, at similar BIS40-60 [80 (72-88)% vs 80 (70-87)%], propofol [4.2 (3.4-5.3) vs 4.4 (3.6-5.4) mg kg−1 h−1] and remifentanil [0.19 (0.13-0.25) vs 0.18 (0.15-0.23) µg kg−1 min−1] consumptions were similar in the N2O vs AIR group, respectively. For the subgroups of women, at similar BIS40-60 [76 (64-84)% vs 72 (62-82)%], propofol [4.7 (4.0-5.8) vs 5.3 (4.5-6.6) mg kg−1 h−1, P=0.004] and remifentanil [0.18 (0.13-0.25) vs 0.20 (0.15-0.27) µg kg−1 min−1, P=0.029] consumptions decreased with the co-administration of N2O. Conclusions With automated drug administration titrated to comparable BIS, N2O only slightly reduced propofol consumption and did not reduce remifentanil consumption. There was a minor gender dependence, but not by a clinically important amount. Clinical trial registration This study was registered at ClinicalTrials.gov, number NCT00547209