Insulin Cessation and Diabetes Remission After Bariatric Surgery in Adults With Insulin-Treated Type 2 Diabetes

Abstract

OBJECTIVE: The impact of bariatric surgeries on insulin-treated type 2 diabetes (I-T2D) in the general population is largely undocumented. We assessed changes in insulin treatment after bariatric surgery in a large cohort of I-T2D patients, comparing Roux-en-Y gastric bypass surgery (RYGB) with laparoscopic adjustable gastric banding (LAGB), controlling for differences in weight loss between procedures. RESEARCH DESIGN AND METHODS: Of 113,638 adult surgical patients in the Bariatric Outcomes Longitudinal Database (BOLD), 10% had I-T2D. Analysis was restricted to 5,225 patients with I-T2D and at least 1 year of postoperative follow-up. Regression models were used to identify factors that predict cessation of insulin therapy. To control for differences in weight loss patterns between RYGB and LAGB, a case-matched analysis was also performed. RESULTS: Of I-T2D patients who underwent RYGB (n = 3,318), 62% were off insulin at 12 months compared with 34% (n = 1,907) after LAGB (P < 0.001). Regression analysis indicated that RYGB strongly predicted insulin cessation at both 1 and 12 months postoperatively. In the case-matched analysis at 3 months, the proportion of insulin cessation was significantly higher in the RYGB group than in the LAGB group (P = 0.03), and the diabetes remission rate was higher at all time points after this surgery. RYGB was a weight-independent predictor of insulin therapy cessation early after surgery, whereas insulin cessation after LAGB was linked to weight loss. CONCLUSIONS: I-T2D patients have a greater probability of stopping insulin after RYGB than after LAGB (62% vs. 34%, respectively, at 1 year), with weight-independent effects in the early months after surgery. These findings support RYGB as the procedure of choice for reversing I-T2D.