Weight loss percentile charts of large representative series: A benchmark defining sufficient weight loss challenging current criteria for success of bariatric surgery

摘要

Background: The frequently used 35 kg/m2 body mass index (BMI) and 50 % excess weight loss (%EWL) criteria are no longer adequate for defining the success of a bariatric or metabolic surgery. It is not clear whether they are still useful to simply determine the sufficiency of a patient's postoperative weight loss. An alternative way of defining sufficient weight loss is presented, using weight loss percentile charts of large representative series as a benchmark. Methods: Gastric bypass weight loss results from the Bariatric Outcomes Longitudinal Database (BOLD) with ≤2 years of follow-up are presented with percentiles in function of postoperative time and their nadir results in function of initial BMI using different outcome metrics. These percentiles are compared with the BMI35 and 50 %EWL criteria. Results: Of 49,098 patients eligible for ≤2 years of follow-up, 8,945 had reported weight loss at ≤2 years (20.0 % male, mean initial BMI 47.7 kg/m2). They reached nadir BMI at a mean of 603 days. Their 50th percentiles surpassed both 50 %EWL and BMI35 after 135 days. More than 95 % achieved 50 %EWL; more than 75 % achieved BMI35. BMI and %EWL results are influenced more by initial BMI than total weight loss (%TWL) results. Conclusions: BOLD gastric bypass weight loss data are presented with percentile curves. BMI and %EWL are clearly not suited for this purpose. Provided that follow-up data are solid, %TWL-based percentile charts can constitute neutral benchmarks for defining sufficient postoperative weight loss over time. Criteria for overall success, however, should consider clear goals of health improvement, including metabolic aspects. Frequently used criteria 50 %EWL and BMI35 are inadequate for both. Their static weight loss components do not match the found percentiles and their health improvement components do not match known metabolic criteria.