Commentary by Geltrude Mingrone, MD, PhD and Frank Greenway, MD
Predictors of diabetes remission varied by type of bariatric surgery, according to a study published online ahead of print in Diabetes Care. The relationship between bariatric surgery and diabetes remission appears to be independent of baseline BMI, the study authors reported.
“The criteria that are presently used to select people with diabetes for bariatric surgery do not select those most likely to have remission of that disease, and different criteria should be created that do select people with diabetes most likely to go into remission,” commented Frank Greenway, MD, Vice President of the Obesity Treatment Foundation and Medical Director of the Pennington Biomedical Research Center, Baton Rouge, LA.
The researchers merged data from the Swedish Obese Subjects (SOS) study1 with two randomized controlled studies2,3 to investigate a combined group of 727 patients with type 2 diabetes who underwent conventional medical therapy and a combined group that underwent bariatric surgery (n=415).
The medical therapies differed among the 3 studies. In the randomized trials, patients were asked to reduce their calorie intake to <30%, were encouraged to eat low-glycemic index and high-fiber foods, and were encouraged to engage in 10,000 steps/day and 200 minutes/week of physical activity. The medical intervention was not standardized in the SOS group.
Four types of bariatric surgery were evaluated in the 3 studies: four different surgical techniques: vertical banded gastroplasty (VBG; n=227), adjustable or nonadjustable gastric banding (GB; n=91), Roux-en-Y gastric bypass (RYGB; n=77), and biliopancreatic diversion (BPD; n=20).
Bariatric Surgery Increases Likelihood of Diabetes Remission Over Pharmacotherapy
A significantly greater proportion of patients in the surgical group achieved diabetes remission (fasting glycemia <5.6 mmol/L without pharmacological therapy) compared with patients in the medical group (63.7% vs 14.4%; P<0.001). When the surgery patients were divided into gastric only (VBG and GB) or gastric with diversion (RYGB and BPD) subgroups, patients in the GD subgroup were found to have higher remission rates than the GO subgroup (60% vs 76%; P=0.016).
Tight glycemic control (fasting glycemia <7 mmol/L) at 2 years was found in 85% and 78% of the GD and GO groups without use of pharmacological therapy and in 91% and 88% of the respective groups with use of concomitant pharmacological therapy. In contrast, less than 40% of medically treated patients achieved tight glycemic control.
Patients who achieved remission lost more weight (25% vs. 17%), showed a greater decrease in waist circumference (18% vs. 13%), and experienced better insulin sensitivity than nonresponders. In the GO subgroup, diabetes duration, fasting glycemia, and therapies for diabetes were inversely correlated with diabetes remission. In contrast, only fasting glycemia was a significant predictor of diabetes remission in the GD group.
Should BMI Be Used as a Cutpoint for Bariatric Surgery?
“Our study demonstrates that while BMI is not at all a predictor of diabetes remission and glycemic control, measures of insulin resistance such as HOMA-IR [homeostasis model assessment of insulin resistance] and waist circumference are good predictors,” Dr. Mingrone said.
Patients with a BMI >40 kg/m2 had a significantly greater likelihood for achieving remission than patients with a BMI ≤35 kg/m2 (P=0.003). In contrast, the likelihood of achieving diabetes remission was not significantly different between subjects with a body mass index (BMI) ≤35 kg/m2 and 35-40 kg/m2.
“Our and many other studies have clearly shown that BMI does not predict diabetes remission or improvement in glycemic control nor macro- and microvascular complication occurrence,” Dr. Mingrone noted. “Thus, a new randomized controlled study should be carried out using waist circumference instead of BMI in uncontrolled type 2 diabetes as an enrollment criterion to prove that measures of adiposity distribution are more effective than mere measures of body weight.”
“The article shows that best criteria to select patients with diabetes for bariatric surgery who will be most likely to experience remission of their diabetes are not BMI or uncontrolled diabetes, the present criteria,” Dr. Greenway commented. “Instead, the study showed that possible new criteria that might be better predictors of diabetes remission could include a type of surgery, better diabetes control at baseline, a shorter duration of diabetes, greater sensitivity to insulin and a greater waist circumference.”
“More research is needed to determine which of these potential criteria are best and where the optimal cut points might be,” Dr. Greenway said. “Once those considerations are defined, a set of criteria could be defined to rationally select the diabetic subjects for obesity surgery who are most likely to experience remission of their diabetes.”
“From a medical perspective, a major reason to do obesity surgery on a person with diabetes is to have the diabetes go into remission,” Dr. Greenway said. “Therefore, defining the criteria to select the people with diabetes who will be most likely to go into remission is very important and will optimize the benefits in relation to the risks by operating on the people most likely to have the greatest benefits in relation to their diabetes. This in turn will likely reduce the expense to society by minimizing or delaying the complications related to diabetes, which are expensive to the healthcare system.”
December 17, 2015
Panunzi S, Carlsson L, De Gaetano A, et al. Determinants of diabetes remission and glycemic control after bariatric surgery. Diabetes Care. 2015 Dec 1. pii: dc150575. [Epub ahead of print]