Relationships between type 2 diabetes remission after gastric bypass and different weight loss metrics: arguments against excess weight loss in metabolic surgery.

BACKGROUND: Percentage excess weight loss (%EWL) outcome of bariatric surgery is distorted by deviations in baseline body mass index (BMI). It has been reported that this can lead to false conclusions, most likely because bariatric weight loss in fact is baseline-BMI independent.

OBJECTIVES: If the metabolic effect of bariatric surgery is baseline-BMI independent as well, could %EWL also lead to false conclusions on metabolic surgery?

SETTING: Bariatric Center of Excellence, general hospital, Netherlands.

METHODS: Retrospective analysis of 1-year outcome of all consecutive primary gastric bypass patients with type 2 diabetes (T2DM). Metabolic outcome (glycated hemoglobin [HbA1c], T2DM medication) was compared with bariatric outcome (weight loss) using 3 different metrics: %EWL, the most popular weight loss metric among bariatric surgeons; percentage (total) weight loss (%WL), most commonly used by nonsurgical professionals; and percentage alterable weight loss (%AWL), the only metric rendering weight loss outcome independent of baseline BMI. Metabolic success (HbA1c≤6.0%, T2DM remission) was compared with different definitions of bariatric success (≥50 %EWL, BMI<35 kg/m(2), %AWL percentiles; Mann-Whitney test; P< .05).

RESULTS: Until May 2014, 2001 patients underwent primary laparoscopic Roux-en-Y gastric bypass (LRYGB), of whom 449 had T2DM with baseline BMI 43.3 kg/m(2), mean 1.6 number of T2DM medication and HbA1c 7.5%. At 1 year 95% follow-up, with BMI 30.5 kg/m(2), 52.1% T2DM remission, 86.9% HbA1c<7.0%, and 63.6% without T2DM medication. No significant differences in T2DM outcome and weight loss were found with different baseline BMI, except for %EWL (P<.001). Weight loss was significantly better with better T2DM outcome, but for %EWL contradictory relationships were found in baseline-BMI subgroups. T2DM outcome was not less successful for patients with<50 %EWL.

CONCLUSION: In T2DM patients, weight loss after gastric bypass does not depend on BMI, HbA1c, or T2DM medication at baseline. The popular %EWL metric and the 50 %EWL success criterion are problematic in comparing bariatric and metabolic outcome of gastric bypass surgery. They should be abandoned. The %WL metric is the best and most commonly used alternative, whereas %AWL is ideal for selected logistics in bariatric research. Weight loss percentiles are best suited for defining bariatric success in metabolic surgery.