Predicting Early Weight Loss Failure Using a Bariatric Surgery Outcomes Calculator and Weight Loss Curves

Abstract

Context: Weight loss after bariatric surgery can be accurately predicted using an outcomes calculator; however, outliers exist that do not meet the 1 year post-surgery weight projections.

Objective: Our goal was to determine how soon after surgery these outliers can be identified.

Design: We conducted a retrospective cohort study.

Setting, patients, and intervention: Using a bariatric surgery outcomes calculator formulated by the Michigan Bariatric Surgery Collaborative (MBSC), predicted weight loss at 1 year post-surgery was calculated on all patients who underwent primary bariatric surgery at a single-center academic institution between 2006 and 2015 who also had a documented 1-year follow-up weight (n = 1050).

Main outcome measures: Weight loss curves were compared between high, low, and non-outliers as defined by their observed-to-expected (O:E) weight loss ratio based on total body weight loss (TBWL) %.

Results: Mean predicted weight loss for the study group was 39.1 ± 9.9 kg, while mean actual weight loss was 39.7 ± 17.1 kg resulting in a mean O:E 1.01 (± 0.35). Based on analysis of the O:E ratios at 1 year post-surgery, the study group was sub-classified. Low outliers (n = 188, O:E 0.51) had significantly lower weight loss at 2 months (13.1% vs 15.6% and 16.5% TBWL, p < 0. 001) and at 6 months (19% vs 26% and 30% TBWL, p < 0.001) when compared to non-outliers (n = 638, O:E 1.00) and high outliers (n = 224, O:E 1.46), respectively.

Conclusions: Weight loss curves based on individually calculated outcomes can help identify low outliers for additional interventions as early as 2 months after bariatric surgery.

Keywords: Bariatric surgery; Patient-reported outcomes; Weight loss; Weight loss failure; Weight-related comorbidities.