Outcomes Research in Review

Bariatric Surgery Leads to 3-Year Resolution of Diabetes in 24% to 38% of Patients


Schauer PR, Bhatt DL, Kirwan JP, et al. Bariatric surgery versus intensive medical therapy for diabetes—3 year outcomes. N Engl J Med 2014;370:2002–13.


Study Overview

Objective. To examine the 3-year efficacy of bariatric surgery on resolution of diabetes.

Design. Randomized controlled trial.


Setting and participants. Patients were participants in the STAMPEDE trial, a single-center study with enrollment from March 2007 to January 2011. 150 patients aged 20 to 60 years with a hemoglobin A1cof > 7% and a BMI of 27 to 43 kg/m were studied. Patients were excluded for a history of bariatric surgery or complex abdominal surgery and poorly controlled medical or psychiatric conditions [1]. Patients were randomized to intensive medical therapy, Roux-en-Y gastric bypass, or sleeve gastrectomy. All participants received intensive medical therapy, including lifestyle education, diabetes medical management, and cardiovascular risk reduction administered by a diabetes specialist every 3 months for 2 years and every 6 months thereafter. All surgeries were performed by a single surgeon, using equipment by Ethicon (a sponsor of the study, along with the National Institutes of Health, LifeScan, and the Cleveland Clinic).

Main outcome measure. HbA1c of ≤ 6% at 3 years.

Main results. At baseline, 68% were women and 74% were white. Participants had a mean age of 48 years (SD 8), mean A1c of 9.3% (1.5%), and mean BMI of 36 (3.5). 43% required insulin at baseline. Follow-up at 3 years was 91% (9 participants dropped out after enrollment, 4 lost to follow-up), and at this time, A1c levels were ≤ 6% for 5% of intensive medical therapy participants, 38% who had gastric bypass (P < 0.001 compared with medical therapy), and 24% who had sleeve gastrectomy (P = 0.01 compared with medical therapy); the difference between bypass and sleeve gastrectomy arms was not significant (P = 0.17). Nearly all of the participants reaching the primary outcome in the bariatric surgery arms achieved this goal A1c without using diabetic medications (35% and 20%). For the secondary outcome of A1c ≤ 7% without using diabetic medications, 0%, 58%, and 33% reached this endpoint in the medical therapy, bypass, and sleeve gastrectomy arms, respectively (P < 0.001 for both surgery arms compared to medical therapy; P = 0.01 comparing gastric bypass to sleeve gastrectomy). At 3 years, 2%, 69%, and 43% of participants were not taking any diabetic medications; 55% of medical therapy participants were taking insulin compared with 6% and 8% in the surgery arms. Weight loss was significantly greater in the gastric bypass and sleeve gastrectomy arms (24.5% and 21.1% of baseline body weight compared with the medical therapy arm with 4.2%). HDL cholesterol was higher and triglycerides were lower in both surgery arms, compared with medical therapy, but LDL cholesterol and blood pressure were not significantly different. Surgery participants also were taking fewer cardiovascular medications at 3 years. Quality of life was improved in 5 of 8 domains for the bypass arm compared with medical therapy and in 3 of 8 domains for the sleeve gastrectomy arm.

Conclusion. Gastric bypass and sleeve gastrectomy surgery leads to substantial resolution of diabetes compared to medical therapy.



Over the last several decades, bariatric surgery has emerged as important treatment for obesity. Observational studies have demonstrated sustained weight loss persisting up to 15 years, as well as reductions in cardiovascular risk, diabetes, and even mortality [2–5]. In the Swedish Obesity Study, a nonrandomized study of 2010 participants undergoing bariatric surgery and 2037 matched controls, gastric bypass led to a 32% reduction from baseline body weight at 1–2 years after surgery with sustained weight loss of 27% at 15 years [2,3]. Patients undergoing gastric banding lost a bit less weight, with 20% weight loss at 1–2 years and 13% at 15 years. Control subjects lost very little.

Among diabetic Swedish Obesity Study participants, bariatric surgery led to a much higher rate of remission from diabetes over 10 years compared with control patients (36% after surgery, 13% among controls) [2] and lower rates of microvascular and macrovascular complications [6]. Among participants who were not diabetic at baseline, the incidence of diabetes was just 7% in the surgery arm and 24% in the control arm [2]; this difference in incidence persisted for 15 years of follow-up [4].

Among randomized controlled trials, several studies have found short-term resolution of diabetes after surgery. A study of 60 patients (age 30 to 60 years, BMI ≥ 35, A1c ≥ 7%) found that 75% of patients undergoing gastric bypass and 95% of patients undergoing

Next Article: