Conference Coverage

Seven years after bariatric surgery, more than 40% still off insulin


Key clinical point: In the largest study to follow insulin-dependent patients after bariatric surgery, substantial benefits persist after median 7 years of followup.

Major finding: Among 252 insulin-dependent patients followed for a minimum of 5 years, 44% remain off insulin and 15% are off all anti-diabetic medications.

Data source: Retrospective single-center analysis.

Disclosures: Dr. Aminian reported having no relevant financial relationships.



– Forty-four percent of insulin-dependent patients with type 2 diabetes mellitus (DM2) were at their glycemic target without insulin a median of seven years after surgery. The data from the largest study to evaluate long-term outcomes in this population were presented at Obesity Week 2017.

“These data confirm that the impressive metabolic effects of bariatric surgery in patients with type 2 diabetes are sustained beyond five years,” reported Ali Aminian, MD, a surgeon who specializes in bariatric procedures at the Cleveland Clinic, Cleveland, Ohio. He said that long-term efficacy has not been well characterized previously.

Dr. Ali Aminian

Bariatric surgery patients treated at the Cleveland Clinic were included in this analysis if they had DM2, were taking insulin at the time of their procedure, and had been followed for at least five years. The median follow-up was 7 years with a range out to 12 years. Of the 252 patients included, 194 underwent roux-en-y gastric bypass (RYGB) and 58 underwent sleeve gastrectomy.

Reaching the glycemic target, defined as less than 7% HbA1c, without insulin was only one of the primary endpoints. The other was diabetes remission, which was defined as HbA1c less than 6.5%, fasting blood glucose less than 126 mg/dL, and being off all diabetes medications. This was observed in 15% of the patients after a median of 7 years followup.

Contrasting short-term results, defined as outcomes one to two years after bariatric surgery with the long-term followup, Dr. Aminian was able to show that declines were relatively modest over time. For example, 51% were at the glycemic target off insulin at the short-term mark, which translates into an absolute decline of only 7% relative to the 44% observed at the long-term followup assessment.

Similarly, 70% had achieved the American Diabetes Association (ADA) goal of less than 7% within the first two years of surgery, while 59% remained at this goal at the most recent followup. The proportion taking insulin at the short-term mark was 36% rising only to 40% long-term.

When data were stratified by procedure, results favored RYGB over sleeve gastrectomy. For example, 47% of the RYGB patients versus 33% of the sleeve gastrectomy patients were able to reach the ADA goal without insulin at the end of the study. The proportions in diabetes remission were 17% and 10%, respectively. RYGB was also associated with greater improvement in BMI (median -12 vs. - 8 kg/m2) and reduced late weight gain (median 20% vs. 31%).

However, Dr. Aminian, who did not provide statistical calculations for these differences, cautioned that higher risk patients might have been preferentially selected for sleeve gastrectomy. He noted that difference in median HbA1c levels was significantly lower in the RYGB group two years after surgery (P less than .001) but the numerical advantage had lost significance at the last followup (P = .32).

In an evaluation of predictors for glycemic control, a shorter duration of diabetes (less than 10 years) and good glycemic control prior to surgery were both predictors of achieving the primary outcomes on the basis of a multivariate analysis, according to Dr. Aminian. Younger age was a marginal predictor, but Dr. Aminian said that neither type of procedure nor presurgical BMI predicted outcomes from the multivariate analysis.

Relative to baseline, there were significant improvements in median LDL (P = .001). In addition, HDL, triglyceride levels, systolic, and diastolic blood pressure measurements were all significantly improved, both short-term and long-term after bariatric surgery (all P values less than .001), according to Dr. Aminian. When expressed as ADA goals, 82% of participants had blood pressure less than 140/90 mm Hg 7 years after surgery relative to 44% at baseline (P less than .001). The proportion with LDL less 100 mg/dL approached, but did not reach clinical significance (61% vs. 70%; P=0.06).

“When you consider all three parameters [ADA targets for glycemic control, blood pressure control, and lipid control], only 3% of patients met all three targets at baseline but 32% [P< less than .001] were at these targets at long-term followup,” Dr. Aminian reported.

Dr. Aminian reported having no relevant financial relationships.

As the invited discussant on these data, Raul Rosenthal, MD, Director, Bariatric and Metabolic Institute, Cleveland Clinic Florida, Weston, Florida, reiterated that time with diabetes prior to bariatric surgery may be an important predictor of postsurgical control of metabolic parameters.

“I published a paper about 10 years ago on outcomes in patients with diabetes, and in our experience 5 years was the limit. If you have a history of 5 years or less with diabetes, the chance of going into remission were 80%, and if it was more than 5 years, the likelihood dropped dramatically,” Dr. Rosenthal noted. He indicated duration of diabetes deserves further evaluation for its potential relevance to the optimal timing of bariatric surgery.

Next Article:

   Comments ()