Abstract
<title>Abstract</title> <p>Background Whether long-term outcomes after bariatric surgery are primarily determined by sustained weight loss or by procedure-specific physiological mechanisms remains uncertain. We investigated the relative contribution of surgical technique and long-term weight-loss maintenance to endoscopic, metabolic, nutritional, and inflammatory outcomes after sleeve gastrectomy (SG) and Roux-en-Y gastric bypass (RYGB). Methods This retrospective comparative cohort included 234 adults who underwent SG (n = 74) or RYGB (n = 160) between 2003 and 2013, with a minimum follow-up of 10 years. Patients were stratified according to long-term total weight loss (< 10%, 10–25%, and > 25%). Endoscopic, metabolic, nutritional, and inflammatory outcomes were compared between procedures within each weight-loss category. Multivariable logistic regression and inverse probability of treatment weighting (IPTW) analyses were performed to identify independent associations. Results Weight-loss maintenance was not significantly associated with most long-term endoscopic, nutritional, or inflammatory outcomes. SG remained independently associated with postoperative gastritis (adjusted OR 5.46, 95% CI 2.04–14.58; P < 0.001), and this association persisted after IPTW adjustment (OR 12.60, 95% CI 6.41–24.79; P < 0.001). RYGB consistently demonstrated lower LDL cholesterol concentrations across all weight-loss categories. Nutritional differences were limited to lower zinc and calcium concentrations after SG among patients maintaining > 25% total weight loss. Inflammatory markers did not differ according to surgical technique or weight-loss maintenance. Conclusions More than 10 years after bariatric surgery, surgical technique exerted a greater influence than long-term weight-loss maintenance on several postoperative outcomes. Procedure-specific physiological mechanisms appear to persist long after weight stabilization and should be considered during surgical decision-making and long-term postoperative surveillance.</p>