A patient’s body mass index before bariatric surgery is unrelated to remission of his or her diabetes afterward, according to a report published online in Annals of Surgery.

In a meta-analysis of 94 observational and interventional clinical studies, the efficacy of the surgery as a metabolic procedure had nothing to do with baseline body mass index (BMI) and instead depended on the degree of diabetic compromise at baseline. “We believe therefore that the use of baseline BMIs to gauge the metabolic effect of surgery is misleading and should be avoided,” said Simona Panunzi, Ph.D., of the Italian National Council of Research-Institute of Systems Analysis and Computer Sciences, Biomathematics Laboratory, Rome, and her associates.

At present, baseline BMI is considered the only valid selection criterion for bariatric surgery in diabetic adults, even though numerous studies have reported that neither BMI nor body weight are good predictors of diabetes remission. In contrast, many other studies have reported that body fat distribution, and visceral obesity in particular, directly affects glycemic control, insulin resistance, and the metabolic syndrome, the investigators commented ( Ann. Surg. 2014 Oct. 30 [E-pub ahead of print]).

To assess which factors best predict diabetes remission after bariatric surgery, Dr. Panunzi and her associates performed a systematic review and meta-analysis of the literature. They screened 1,437 articles and focused on 94: 35 studies included patients with a baseline BMI of less than 35 kg/m2, 56 studies included patients with a baseline BMI of 35 kg/m2 or more, and 3 studies included patients with any baseline BMI. There were a total of 94,579 patients, of whom 4,944 had type 2 diabetes.

The mean age of the study participants was approximately 46 years. Bariatric procedures included sleeve gastrectomy, Roux-en-Y gastric bypass, biliopancreatic diversion, laparoscopic gastric banding, and duodenal-jejunal bypass. Across all the studies, the percent reduction in mean BMI after surgery was 17% for patients with a baseline BMI of less than 35 kg/m2 (group 1) and 35% for those with a baseline BMI of 35 kg/m2 or more (group 2).

Diabetes remitted in 72% of the patients in group 1 and 71% of those in group 2, a nonsignificant difference. The only factor found to predict remission, defined as the normalization of HbA1c, was the severity of diabetes as measured by fasting levels of glucose, insulin, and HbA1c at baseline, the investigators said.

“The appropriateness of the BMI criterion proposed by the National Institutes of Health for determining the eligibility of patients with diabetes for bariatric surgery has already been challenged by other authors. However, until now there was no direct evidence of its lack of relevance. It seems clear from the results of [our] study that baseline BMI does not carry prognostic information for diabetes resolution after bariatric surgery,” they added.

This meta-analysis was limited in that few of the reviewed studies reported diabetes duration, a well-known predictor of remission. And the definition of diabetes remission varied widely among the studies, from simple withdrawal of diabetes medications to various specific plasma glucose or HbA1c levels, the investigators noted.