Nitrous oxide and risk of surgical wound infection: A randomised trial

Background: Nitrous oxide inactivates vitamin B12 and methionine synthase, thereby impairing DNA formation and, consequently, new cell formation. The gas also inhibits methionine production, which can reduce scar formation and depresses chemotactic migration by monocytes. Therefore, we assessed whether nitrous oxide increases the incidence of surgical wound infection. Methods: We recruited 418 patients aged 18-80 years, scheduled for colon resection that was expected to last more than 2 h, at three hospitals in Austria and Hungary. Patients were randomly assigned 65% intraoperative nitrous oxide (n=208) or nitrogen (n=206), with remifentanil and isoflurane. The primary outcome was the incidence of clinical postoperative wound infection, analysed by intention to treat. Findings: 206 patients in the nitrous oxide group and 202 in the nitrogen group were included in the final analysis. Duration of surgery was longer in the nitrogen group (3·4 h [1·5]) than in the nitrous oxide group (3·0 h [SD 1·3]) and arterial pressure (84 mm Hg [10] vs 81 mm Hg [9]), bispectral index values (53 [9] vs 44 [8]), and end-tidal isoflurane concentration (0·64% [0·14] vs 0·56% [0·13]) were greater in patients given nitrogen than in those given nitrous oxide. Infection rate was 15% (31/206) in patients given nitrous oxide and 20% (40/202) in those given nitrogen (p=0·205). Additionally, the ASEPSIS wound healing score, wound collagen deposition, number of patients admitted to critical care unit, time to first food ingestion, duration of hospital stay, and mortality did not differ between treatment groups. Interpretation: Nitrous oxide does not increase the incidence of surgical wound infection.