Volatile versus intravenous anesthesia and postoperative neurocognition: A meta-analysis and trial-sequential analysis.

BACKGROUND

Postoperative delirium and delayed neurocognitive recovery are common and consequential complications of anesthesia and surgery. Maintenance of general anesthesia with propofol-based intravenous drugs may reduce postoperative neurocognitive complications compared with volatile anesthesia.

METHODS

Major databases were searched through November 2025. The primary outcome was postoperative delirium within 7 postoperative days. Secondary outcomes included delayed neurocognitive recovery (formerly termed postoperative cognitive decline) and Mini-Mental State Examination scores on postoperative days 1, 3, and 7 (range 0 to 30, higher scores better, with 2 points being meaningful).

RESULTS

Across 29 trials (N = 11,896 patients), the incidence of postoperative delirium was similar with volatile and intravenous anesthesia RR: 0.94 [95% CI, 0.71 to 1.25]. Volatile anesthesia provoked delayed neurocognitive recovery RR: 1.35; 95% CI, 1.07 to 1.70), although effects were inconsistent across assessment periods. Postoperative MMSE scores were slightly lower with volatile anesthesia (mean difference: -1.4 points [95% CI, -2.2 to -0.6]), a difference we did not consider clinically meaningful. Both delayed neurocognitive recovery and Mini-Mental State Examination scores analyses displayed substantial heterogeneity. Trial sequential analysis indicated that more trials are needed for each outcome.

CONCLUSIONS

The incidence of postoperative delirium was similar with volatile and intravenous anesthesia. Volatile anesthesia provoked small, heterogeneous, and time-dependent differences in delayed neurocognitive recovery, but not of a meaningful magnitude. Current evidence suggests that intravenous anesthesia does not meaningfully improve postoperative neurocognitive outcomes; however, available information remains limited, heterogeneous, and imprecise.