Roux-en-Y gastric bypass: the impact of tobacco nicotine, nontobacco nicotine, and marijuana use on postoperative outcomes.

Enagala P.; Wang J.; Dhalla Z.; Hsu C.; Samreen S.; McTaggart L. · Surgery for obesity and related diseases : official journal of the American Society for Bariatric Surgery · 2026

**Background & Methods** This retrospective cohort study analyzed the TriNetX Research Network to evaluate postoperative complications in adults (>18 years) undergoing Roux-en-Y gastric bypass (RYGB) stratified by preoperative substance use within 90 days of surgery. Four propensity-matched cohorts were compared: tobacco nicotine users, nontobacco nicotine users, marijuana users, and nonusers, with one-year follow-up outcomes including infection, readmission, seroma, anastomotic leak, deep vein thrombosis, internal hernia, marginal ulcer, and nutritional deficiencies. **Key Findings** • Tobacco nicotine users demonstrated increased risk of readmission (RR 1.29), internal hernia (RR 1.29), marginal ulcer (RR 1.69), and nutritional deficiency (RR 1.04), with paradoxically reduced anastomotic leaks (RR 0.65). • Nontobacco nicotine users showed similar patterns with increased readmission (RR 1.29), internal hernia (RR 1.30), and marginal ulcer (RR 1.57), alongside reduced anastomotic leaks (RR 0.71), with no clinically meaningful differences compared to tobacco users. • Marijuana users exhibited increased marginal ulcer risk (RR 1.74) but no significant differences in readmission rates compared to nonusers, with no clinically meaningful differences observed between marijuana and nicotine-using groups. **Dosage & Administration** Not reported. **Safety & Adverse Effects** All substance use categories (tobacco nicotine, nontobacco nicotine, and marijuana) were associated with increased postoperative morbidity across multiple complications including readmission, internal hernia formation, marginal ulceration, and nutritional deficiencies. The most consistent adverse outcome across all groups was marginal ulcer development, with marijuana users demonstrating the highest risk (RR 1.74). **Evidence Quality** This retrospective cohort study represents moderate-level evidence with inherent limitations including reliance on documented preoperative substance use within a specific 90-day window, potential for selection bias, and inability to establish causal mechanisms. The propensity-score matching strengthens comparability between groups but cannot control for unmeasured confounders such as dosage, frequency, duration of use, or timing relative to surgery. The authors acknowledge that findings do not demonstrate superiority of any substance type and call for prospective studies and standardized preoperative cessation protocols. The large database provides statistical power but lacks mechanistic insight into why these substances impact specific outcomes differently.

More recent cannabis research