290. RISK OF ESOPHAGEAL CANCER AFTER BARIATRIC SURGERY: COMPARISON BETWEEN SLEEVE GASTRECTOMY AND GASTRIC BYPASS AT 10 YEARS, A NATIONWIDE STUDY

Abstract

Abstract Backgroud The risk of esophageal cancer after bariatric surgery is a matter of debate. Several studies have reported that bariatric surgery patients have a lower risk of cancer. However, esophageal cancer is associated with gastroesophageal reflux disease, which may increase after bariatric surgery and in particular after sleeve gastrectomy. Objectives The objective of this study is to evaluate the risk of esophageal cancer after two bariatric procedures: sleeve gastrectomy (SG) and gastric bypass (GBP). Methods Data from patients operated on for bariatric surgery between 2007 and 2020 in France were extracted from a national discharge database (PMSI). Adult patients operated on for SG or GBP were included and followed up until December 2020. Patients were divided into two groups according to the procedure initially performed. The endpoints were the occurrence of esophageal and cardia cancers. To assess the effect of the type of bariatric procedure on the development of cancer, a multivariate analysis was performed using a marginal Cox model after matching on propensity score (PS). In the sensitivity analysis, we used different models: standard Cox regression, Cox regression adjusted on propensity score, and Cox model with adjustment for inverse probability of treatment weighting. Results Of the 367,169 patients included, 68.7% were SGs and 31.3% were GBPs. The two groups differed at inclusion in terms of age, sex, BMI, and comorbidities. Median follow-up was 6.0 years (IQR 3.0–8.0 years) for the entire cohort, and approximately 35,000 patients were followed for at least 10 years. A total of 88 esophageal cancers were identified, including 64 after SG and 24 after GBP. In multivariate analysis, no significant difference was found between SG and GBP on cancer incidence (HR 1.6, 95%CI 0.9–2.7, p = 0.09 for OS versus GBP). No differences were found in sensitivity analyses. Conclusion In this large national cohort of bariatric surgery patients, no significant differences were found in the incidence of esophageal and cardia cancer between OS and GBP.