Assessment of the causal effect of lifestyle factors on postoperative infections based on Mendelian randomization

Determining causal relationships between modifiable lifestyle factors and postoperative infections is crucial for guiding preventive strategies, yet establishing causality in observational studies remains challenging. This study aimed to investigate whether educational attainment, sleep duration, alcohol consumption, smoking, and body mass index (BMI) are causally linked to postoperative infection risk using Mendelian randomization (MR). We conducted a 2-sample MR analysis utilizing genome-wide association study data for 5 modifiable lifestyle factors as exposures and 2 postoperative infection genome-wide association study datasets as outcomes. Primary causal estimates were derived using inverse-variance weighted (IVW), with sensitivity analyses performed via MR-Egger regression, weighted median, and Mendelian randomization pleiotropy residual sum and outlier methods to assess pleiotropy and outliers. Heterogeneity was evaluated using Cochran Q statistic, and reverse MR analyses were employed to test for reverse causation. All analyses were implemented in R with rigorous statistical thresholds. BMI exhibited a positive causal association with postoperative infection risk [IVW method: dominance ratio (odds ratio) = 1.002, 95% CI (1.001-1.003), P = 2.60E-05], while higher educational attainment (years of schooling) was inversely associated [IVW method: odds ratio = 0.998, 95% CI (0.997-1.000), P = .009]. Sensitivity analyses confirmed robustness, with no evidence of horizontal pleiotropy or influential outliers. No significant associations were observed for sleep duration, alcohol use, or smoking. This first genome-wide MR study demonstrates that elevated BMI increases, and longer education decreases, the risk of postoperative infections. These findings underscore BMI and education as modifiable targets for infection prevention, providing a foundation for clinical and public health interventions.