To compare mortality and organ injury associated with intraoperative hyperoxia and normoxia in cardiac surgery patients.

Systematic review and meta-analysis of randomized clinical trials (RCTs).

Tertiary care.

Patients who underwent cardiac surgery with cardiopulmonary bypass or off-pump procedures.

Intraoperative hyperoxia or normoxia.

A systematic search of PubMed, Embase, and the Cochrane Library identified RCTs comparing hyperoxia versus normoxia in cardiac surgery. Primary outcomes were in-hospital and 30-day mortality; secondary outcomes included postoperative clinical and biochemical measures. Risk ratios (RRs), mean differences (MDs), and standardized mean differences (SMDs) with 95% confidence intervals (CIs) were calculated. Trial sequential analysis and subgroup analysis based on the variable used to define oxygenation (PaO₂-based v FiO₂-based thresholds) were performed to address heterogeneity in oxygenation strategies. Nineteen RCTs (2,001 patients) were included, with 982 patients (49.1%) experiencing hyperoxia. In-hospital (9/798 [1.1%] v 10/799 [1.3%], RR 0.84, 95% CI 0.36 to 1.94, p = 0.684) and 30-day mortality (1/282 [0.4%] v 4/280 [1.4%], RR 0.41, 95% CI 0.08 to 2.28, p = 0.311) were not different between the groups. However, the normoxia group had a higher postoperative PaO₂/FiO₂ ratio (MD –31.49, 95% CI –47.13 to –15.85, p < 0.01), whereas creatine kinase-MB (SMD 1.65, 95% CI 0.12 to 3.18, p = 0.03) and malondialdehyde (SMD 3.77, 95% CI 2.99 to 4.55, p < 0.01) were higher in the hyperoxia group.

Compared with normoxia, hyperoxia during cardiac surgery does not impact in-hospital or 30-day mortality but is associated with worsening physiological and biochemical parameters.