Evidence Based Practice and Quality Assurance
Background and Goal of Study: Supplemental oxygen has been shown to decrease the incidence of postoperative nausea and/or vomiting (PONV). In order to estimate the efficacy of an increased (˜80%) oxygen concentration versus routine oxygen administration (˜30%) administered in the perioperative period we performed a quantitative systematic review.
Materials and Methods: Systematic search (MEDLINE, EMBASE, CENTRAL, bibliographies, all languages, up to October 2004) for randomised comparisons of supplemental oxygen versus routine oxygen in surgical patients. Relevant outcomes were the incidences of postoperative nausea (PN), postoperative vomiting (PV), PONV and rescue treatment. Combined data were analysed using relative risk (RR) with 95% confidence intervals (CI) calculated with a random effects model.
Results and Discussions: In 7 trials 742 patients received supplemental oxygen with 50-80% oxygen, balance nitrogen and 752 patients routine oxygen with 30% oxygen, balance nitrogen. These regimen were administered in the perioperative period and were maintained up to 2 hours in the postoperative period. Pooled RR for PN, PV and PONV with supplemental oxygen versus routine oxygen were 0.91 (95%-CI: 0.73-1.15), 0.78 (95%-CI: 0.62-0.98) and 0.90 (95%-CI: 0.71-1.15), respectively. RR for rescue treatment was 0.88 (95%-CI: 0.66-1.19). Due to considerable heterogeneity (I2-Test ˜ 70% for outcome “PONV”) we performed a sensitivity analysis. The marginal effect on PV depends on the inclusion of 2 initial trials that were performed in patients undergoing abdominal surgery and gynecologic laparoscopy. Restricting the analysis to the remaining 5 trials with a total of 1.104 patients revealed RR for PN, PV and PONV of 1.08 (95%-CI: 0.95-1.23), 0.86 (95%-CI: 0.67-1.09) and 1.09 (95%-CI: 0.96-1.23).
Conclusion(s): In accordance with the IMPACT study (1) the perioperative administration of supplemental oxygen in order to decrease the incidence of PONV cannot be regarded as valid concept so far. Potential effects in abdominal surgery need further analyses.
1 Apfel CC et al. NEJM 2004; 350: 2441-2451.