Gastric bypass helps poorly controlled type 2 diabetes

The 22 serious adverse events in the surgery group included 4 perioperative and 6 late postoperative complications. The most serious complications involved anastomotic leakage that was not detected on routine postoperative upper-GI contrast studies. One of these patients eventually required extracorporeal membrane oxygenation; she developed anoxic brain injury, required lower-extremity amputation, and remains permanently disabled.

Nutritional deficiencies also developed relatively often in the bypass group, despite supplementation. These included iron, vitamin B, and vitamin D deficiencies, as well as hypoalbuminemia.

Other adverse events known to develop more than 1 year after gastric bypass include weight regain, internal hernias, anastomotic ulcers, hypoglycemia, anastomotic strictures, bleeding, bowel obstruction, kidney stones, and fractures.

Overall, these findings show that "the merit of gastric bypass treatment of moderately obese patients with type 2 diabetes depends on whether potential benefits make [these] risks acceptable," the investigators said.

In their systematic review of the literature, Dr. Maggard-Gibbons and her associates reviewed 1,291 studies but could find only 3 randomized trials that came close to directly comparing surgical against nonsurgical interventions in patients with BMIs of 30-35. The total number of patients in these trials was only 290, and individual-level data were available only for 13 patients with type 2 diabetes in this weight category.

They then altered their search criteria to include studies of patients with BMIs of 35-40 and studies that did not directly compare outcomes between surgical and nonsurgical approaches. Even then, "many of the studies were from single surgeons at single academic institutions and may not be representative of results in a general population of patients and surgeons."

Of greatest concern was the fact that "we found no long-term studies of postsurgery adverse events in patients with diabetes within our target BMI range," they said (JAMA 2013;309:2250-61).

The limited studies that were available consistently found that short-term weight loss and glycemic control were better for patients treated with bariatric surgery than for those who were not. However, the durability of these results is uncertain, and the amount of weight loss did not always correlate with the degree of improvement of metabolic factors. And it is not yet known whether these outcomes translate into reduction in the important macrovascular and microvascular effects of diabetes.

Moreover, although surgical complications were not common, those that did occur often were serious and required significant intervention. And of the few studies that did track later complications, most were not designed to do so and covered only "surgeon-reported outcomes from selected, experienced centers."

Thus, "the evidence is insufficient to reach conclusions about the appropriate use of bariatric surgery in this patient population, [and] performance of these procedures in this target population should be under close scientific scrutiny," Dr. Maggard-Gibbons and her colleagues said.

Dr. Irkamuddin’s study was supported by Covidien and the National Center for Advancing Translational Sciences. Dr. Irkamuddin reported ties to Novo Nordisk, USGI, and other companies; his associates reported ties to numerous industry sources. Dr. Maggard-Gibbons’ study was supported by the Agency for Healthcare Research and Quality; she and her associates reported no relevant financial conflicts of interest.