IBC Oxford Poster Abstract 15 - Defining preoperative weight change thresholds and their impact on postoperative weight and comorbidity outcomes following bariatric surgery: a retrospective cohort study in a high-volume centre

Abstract Introduction Preoperative weight loss is commonly recommended before metabolic and bariatric surgery, but its true impact on long-term outcomes remains uncertain. This study aims to evaluate the association between preoperative weight trajectory—categorized as weight maintenance, weight loss, or weight gain—and postoperative outcomes. Materials and methods A retrospective cohort study including patients who underwent Roux-en-Y gastric bypass (RYGB) or sleeve gastrectomy (SG) with follow-up at two years. Patients were stratified according to preoperative weight trajectory. Primary outcomes were TWL 20% and EWL 50%. Secondary outcomes included remission of type 2 diabetes, hypertension, dyslipidemia, and obstructive sleep apnoea, as well as the determination of optimal thresholds for preoperative weight loss and gain using receiver operating characteristic (ROC) analysis and identification through the Youden index method. Logistic regression models were adjusted for sex, presurgical BMI, hypertension, and hospital response time (which were the statistically significant variables between groups, Table 1). Results A total of 1094 patients were included, of whom 82.2% were female. The median presurgical BMI was 41.6 kg/m², and 71.8% underwent RYGB (Table 1). At two years, patients who maintained their weight preoperatively had the highest success rates: 97% achieved TWL 20%, compared to 94.2% (weight loss) and 91.8% (weight gain) (P = 0.025). For EWL 50%, the rates were 97.0%, 94.2%, and 92.8%, respectively (P = 0.074). Preoperative weight gain was independently associated with a lower likelihood of achieving both TWL 20% (adjusted OR 0.41; 95% c.i.: 0.18–0.93) and EWL 50% (adjusted OR 0.41; 95% c.i.: 0.18–0.95). Although preoperative weight loss was not significantly associated with improved weight outcomes in the total cohort, stratified analysis showed significant benefits at 1 year: for TWL 20% in RYGB (P = 0.015) and EWL 50% in both surgeries (P = 0.049). Regarding comorbidities, only preoperative weight loss was significantly associated with diabetes remission at 1 year, both in the total cohort and in RYGB. No significant associations were found for the other comorbidities, though remission was analyzed across all (Table 2). ROC analysis at two years identified a minimum preoperative weight loss of 10.91% (for TWL) and 4.96% (for EWL) as the optimal thresholds to maximize success, and a maximum tolerable weight gain of 6.67% (TWL) and 5.51% (EWL) beyond which outcomes were negatively affected (Table 4). Additional thresholds were also calculated for comorbidity remission. Conclusion Preoperative weight gain is associated with poorer long-term weight outcomes, while modest weight loss (≥5–10%) may improve early results and diabetes remission. These findings support an individualized preoperative strategy that avoids weight gain and encourages targeted loss when appropriate, rather than enforcing universal targets. The identified cut-offs may assist clinicians in tailoring preoperative counselling to optimize postoperative success.