There is a large controversy as to whether nitrous oxide (N2O) added to the anaesthetic gas mixture is harmful or harmless for postoperative cognitive function recovery. We performed a nested study in the ENIGMA-II trial and compared postoperative neurocognitive recovery of patients randomly receiving N2O (70%) or Air (70%) in 30% O2 during anesthesia. We included adults having non cardiac surgery. We compared recovery scores for episodic memory, decision making/processing speed and executive functions measured with the computerised Cambridge Neuropsychological Test Automated Battery (CANTAB). Assessments were performed at baseline, seven and ninety days. At first interim analysis, following recruitment of 140 participants, the trial was suspended. We found that the mean (95%CI) changes of scores for episodic memory were in the Pocock futility boundaries. Decision making/processing speed did not differ either between groups (P > 0.182). But for executive functions at seven days, the mean number (95% CI) of problems successfully solved and the number of correct box choices made was higher in the N2O group, P = 0.029. N2O with the limitations of an interim analysis appears to have no harmful effect on cognitive functions (memory/processing speed). It may improve the early recovery process of executive functions. This preliminary finding warrants further investigations.