Preoperative weight loss is linked to improved mortality and leaks following elective bariatric surgery: an analysis of 548,597 patients from 2015-2018.

Abstract

The effects of preoperative weight loss on bariatric surgery outcomes are still unclear, despite the practice being adopted by bariatric centers worldwide. Ongoing studies are needed for routine adoption of this practice given the multiple issues patients face with following difficult preoperative weight loss protocols. The aim of this study was to characterize the prevalence of preoperative weight loss and evaluate its impact on outcomes following elective bariatric surgery. This retrospective study was conducted using the Metabolic and Bariatric Surgery Accreditation and Quality Improvement Program (MBSAQIP) data registry from 2015-2018. All primary Roux-en-Y (RYGB) and sleeve gastrectomy (SG) procedures were included, whereas prior revisional surgeries and emergency surgeries were excluded. Cases were then divided into preoperative weight loss (PWL) and control cohorts. PWL was defined categorically if the highest 30-day preoperative weight was greater than the closest recorded weight before surgery. Primary outcomes included identifying the impact of PWL on postoperative complications. Multivariable logistic regression modelling was used to examine the influence of PWL on serious complications and mortality after adjusting for patient co-morbidities and procedure type. A total of 548,597 patients were identified with the majority experiencing preoperative weight loss (n= 459,500; 83.8%). The PWL cohort was older (44.8 ± 12.0 versus 43.2 ± 11.9 yr), had a reduced body mass index (BMI) (45.0 ± 7.4 versus 46.1 ± 7.6 kg/m2), and was more likely to be male (20.3% versus 18.7%). Patients with preoperative weight loss also were more likely to have metabolic co-morbidities including medication and insulin-dependent diabetes (27.0% versus 23.2%), hypertension (HTN) (48.9% versus 44.7%), dyslipidemia (DLP ) (24.6% versus 21.0%), and sleep apnea (39.6% versus 32.3%). No clinically significant differences were observed for operative length between cohorts (85.3 ± 46.9 min PWL versus 83.9 ± 46.2 min control). The protective benefit was found to be most significant for patients experiencing greatest weight loss with those experiencing a >10% PWL showing 30% decreased odds of leak (OR .68%; 95% CI [confidence interval] .56-.84; P < .0001) and a 40% decrease in odds of mortality versus those with no PWL (OR .60; 95% CI .39-.92; P = .02). Preoperative weight loss before bariatric surgery is common, occurring in >80% of elective cases. Our findings suggest that preoperative weight loss is associated with improved odds of 30-day mortality and leaks but no differences in bleeds or overall serious complications. Additional prospective trials are needed to further evaluate the role of routine PWL in addition to ongoing development of tolerable preoperative weight-loss protocols.