IBC Oxford University Poster Abstract 22 - Definition and predictors of weight regain after metabolic and bariatric surgery: a retrospective cohort study

Abstract Background Despite the benefits of metabolic and bariatric surgery (MBS), not all patients can maintain sustained weight loss in the long term. There is increasing awareness that weight regain (WR) is a common issue following MBS. However, WR is defined using various criteria, which makes it challenging to understand its prevalence and clinical impact after MBS. As a result, the factors that contribute to WR before, during, and after surgery are not well understood. This study aims to assess WR at two-years after bariatric surgery, using four definitions, and to identify clinically significant predictors of WR to improve patient counselling regarding prevention and optimize intraoperative bariatric procedures. Methods We conducted a single-centre retrospective cohort study with 1056 consecutive patients submitted to MBS between 2019 and 2021 and followed for at least 2 years. At two-years of follow-up, WR was defined based on four definitions: (1) an increase in weight of more than 10 kg from nadir, (2) an increase in excess weight loss (%EWL) higher than 30% from nadir, (3) an increase in BMI of 5 kg/m2 from nadir, (4) any weight regain. Nadir’s weight was determined based on all the postoperative weight measures available. Univariate and multivariate analyses were performed using logistic regression to compute the odds ratio (OR) and the 95% confidence intervals (95%c.i.). Results and Discussion A total of 1056 patients were included, with a mean age of 45.8 (±10.3) years, of which 82.9% were female. The proportion of WR ranged from 1.0 to 55.0% depending on the definition, being ‘any weight regain’ the definition with the highest proportion of patients (n = 581). Considering %EWL as the definition for WR, after multivariable adjustment, patients with lower presurgical BMI (adjusted odds ratio [adjOR]: 0.81, 95% c.i.:0.70–0.92), diabetes (adjOR: 3.72, 95% c.i.: 1.23–11.26) and history of intragastric balloon placement (adjOR: 9.07, 95% c.i.: 1.34–61.32) had higher risk of WR. Across the 4 definitions, there were no shared independent predictors of WR after MBS. However, in multivariable analysis, history of sleeve gastrectomy was a predictor of WR for 3 definitions, except for %EWL >30% from nadir. Moreover, presurgical BMI was a predictor for 3 definitions, but the direction of effect was not even. Conclusion In this cohort of patients, we found that identifying one single definition of clinically significant WR is challenging. The predictors of WR varied depending on the definition applied. No common predictors of WR after MBS were found, across all four definitions.