Background: Observational studies suggest that bariatric surgery procedures promote weight loss and increase diabetes remission rates in patients who are very obese. The adjusted relative risk of developing type 2 diabetes is 93 for women (95% confidence interval [CI], 81 to 107) and 42 for men (95% CI, 22 to 81) when the body mass index (BMI) is greater than 35 kg per m². Despite the high correlation between obesity and type 2 diabetes, no randomized trial on the possible benefits of gastric banding procedures has been performed. Although weight-loss surgery appears to be effective, there are concerns about the lack of formal evidence and the risks and costs of surgery. Laparoscopic adjustable gastric banding (LAGB), which is less invasive and safer than earlier bypass methods, leads to a weight loss of approximately 20 percent. Dixon and colleagues compared the effects of LAGB and conventional therapy on weight loss and diabetes remission in persons who are obese.

The Study: The randomized trial included patients 20 to 60 years of age with a BMI of 30 to 40 kg per m². Participants were diagnosed with type 2 diabetes within the preceding two years, but were otherwise healthy. During a three-month run-in period, behavioral and medical management of participants' diabetes was optimized. Patients were randomized to receive LAGB or conventional therapy. Conventional therapy emphasized state-of-the-art diabetes education with additional tailored pharmacologic management conducted by an expert team. The surgical group participated in the same lifestyle intervention within one month after randomization. The primary end point was glycemic control at two years, based on A1C and fasting plasma glucose levels. Secondary outcomes included percentage change in A1C level, weight, blood pressure, abdominal girth, and lipid levels. Follow-up occurred at a minimum of six-week intervals over two years.

Results: Of the 30 patients in each arm of the study, 29 in the surgical group and 26 in the conventional therapy group completed the study. At two years, diabetes remission occurred in 73 percent of the surgical group and in 13 percent of the conventional therapy group. Percentage of weight loss accounted for the remission in most patients. At two years, patients in the surgical group lost a mean of 20 percent body weight, compared with 1.4 percent in the conventional therapy group. BMI was reduced from 36.9 to 29.5 kg per m² in the surgical group, and from 37.1 to 36.6 kg per m² in the conventional therapy group. Differences in weight loss between groups increased over time. Physical activity was associated with weight loss, but did not independently predict diabetes remission. The surgical group had significantly better glycemic control than the conventional therapy group at two years (P < .001).

At baseline, all but two patients in the surgical groups and four in the conventional therapy group were using pharmacotherapy. At two years, 26 patients in the surgical group were not using hypoglycemic medication compared with eight in the conventional therapy group. Improvements in other secondary outcome measures were also apparent, but the study was not sufficiently powered to interpret the results. Surgical patients were also using fewer antihypertensive and lipid-lowering agents at the end of the trial. Adverse effects of surgery included superficial wound infection (one patient); gastric pouch enlargement requiring revision (two patients); eating difficulties and regurgitation requiring band removal (one patient); and a self-resolving febrile episode (one patient).

Conclusion: After two years, patients who received LAGB had a fivefold greater diabetes remission rate and a fourfold greater reduction in A1C levels than those who received conventional therapy. To achieve remission, patients generally had to lose more than 10 percent of their body weight. The authors point out that any therapy leading to sufficient weight loss is likely to result in remission, but this may be difficult to achieve with conventional therapy. Early insulin use and intensive hypoglycemic therapy appear to achieve similar remission rates as LAGB. This study showed benefit in patients who were obese, rather than very obese. The study had a short duration; therefore, long-term benefits and clinical end points, such as cardiovascular events and mortality, could not be assessed.