By George Ochoa

Studies have shown that gastric bypass surgery can achieve diabetes remission, but a new, large, long-term study suggests the remission may not always be permanent.

In the current study of patients who initially had complete remission, 35.1% redeveloped diabetes within five years. When the patients who never remitted and those who relapsed were added together, more than half of the patients (56%) did not have durable remission of diabetes (Obes Surg 2012 Nov 18; doi 10.1007/s11695-012-0802-1). However, there was evidence that patients who received earlier surgical intervention for their diabetes might have better outcomes.


In a study of 4,434 adults, 68.2% of patients who had gastric bypass had complete diabetes remission within five years postsurgery.

Of the 68.2% of patients who were in remission five years after gastric bypass surgery, 35% then redeveloped diabetes within five years.

Factors significantly associated with higher relapse rate included longer diabetes duration, insulin use and poor preoperative glycemic control.

“What’s new and different [about this trial] is that we focused on relapse of diabetes after patients have remitted,” David E. Arterburn, MD, associate investigator, Group Health Research Institute, Seattle, and lead researcher on the study, told General Surgery News. “We followed them longer—beyond their initial remission—and found that by five years, 35% of patients had redeveloped diabetes. No prior studies have examined this question on this scale.”

Motivated by recent literature indicating the benefits, at least short term, of bariatric surgery for diabetes control, the researchers conducted a retrospective cohort study of adults with type 2 diabetes who received Roux-en-Y gastric bypass from 1995 to 2008 in three integrated health care delivery systems, one in Minnesota and two in California.

“We involved patients in real-world health care systems, not academic medical centers with highly specialized surgical teams,” Dr. Arterburn said. “Our study is a look at what happens to the average patient in routine clinical care.”

The study included 4,434 adults. Overall, 68.2% (95% confidence interval [CI], 66%-70%) had complete diabetes remission within five years postsurgery. Among these, 35.1% (95% CI, 32%-38%) redeveloped diabetes within five years. The median duration of remission was 8.3 years (3,019 days; 95% CI, 2,507-3,281 days). Factors significantly associated with higher relapse rate included longer diabetes duration, insulin use and poor preoperative glycemic control (hemoglobin [Hb]A1c ≥6.5%).

In secondary analyses, weight trajectories after surgery differed significantly among never remitters, relapsers and durable remitters (P=0.03). Patients who never remitted had slightly less weight loss and greater weight regain than those who remitted. Those who relapsed had similar if not slightly superior body mass index (BMI) maintenance after surgery than those with durable remission. Preoperative BMI values did not differ significantly (P=0.93) among the three groups.

“For most patients with type 2 diabetes, gastric bypass is not a cure,” said Dr. Arterburn.

Dr. Arterburn and his colleagues identified three factors that they believe most strongly affect durable remission of diabetes after bariatric surgery:

  1. How long did the patient have diabetes at the time of surgery? The longer the period, the less likely they were to have a remission and the more likely to relapse.
  2. Was the patient on insulin therapy? Patients were less likely to remit and more likely to relapse if they were on insulin at the time of surgery.
  3. Was the patient’s blood sugar controlled at the time of surgery? If blood sugar was poorly controlled at the time of surgery, the patient was less likely to remit and more likely to relapse.

“These are all markers of how severe one’s diabetes is. We concluded that patients with more severe diabetes are likely to benefit less than patients with less severe diabetes. Patients with early diabetes appear to benefit more,” Dr. Arterburn said.

“The biggest message is early intervention in diabetic patients with obesity,” said Jaime Ponce, MD, FACS, FASMBS, president, American Society for Metabolic & Bariatric Surgery, Gainesville, Fla. “Evaluate them for bariatric surgery at BMI of 35 [kg/m2] and at 30 to 35 with diabetes requiring more medications.”

“Because patients who had earlier disease achieved a higher percentage of long-term remission, physicians and patients should consider earlier intervention”—within five to 10 years of diagnosis, said Philip Schauer, MD, director, Bariatric and Metabolic Institute, Cleveland Clinic, in Ohio. “If they wait until their diabetes is more advanced, they’ll have less of a chance for long-term remission.”

Of the new research, Dr. Schauer said, “This study supports what other studies have shown: Not all patients will have a long-term remission.” However, he noted, “a lot of patients who are so-called relapsing are still in good blood sugar control. If their HbA1c is 6% to 7% compared with higher than 7% before surgery, their control is relatively good.”

The potential benefits of a period of remission for patients who relapsed remain unclear, according to Dr. Ponce. “We still don’t know, in patients who got their diabetes back, whether they got a benefit in the years diabetes was in remission,” he said. “Out-of-control diabetes can increase the risk for retinopathy, nephropathy and liver damage, and cause vascular changes and cardiac damage. They didn’t study the benefit out of those years. We need a longer study with a longer follow-up. Even a few years of benefit would be valuable.”

“We do believe and hope to confirm with a future study that even a short period of diabetes remission [in patients who eventually relapse] will have long-lasting benefits compared with those who didn’t have remission or didn’t get surgery,” Dr. Arterburn said.

Limitations of the study noted by Dr. Ponce included not looking at results per variations in technique of bypass, and not having sufficient data to analyze differences in outcomes by race or ethnicity. Dr. Schauer noted the low five-year retention rate of 67.8%. “They’re missing a lot of patients at the five-year mark,” he said. “We would like to see retention rates of 80% to 90%.”

Dr. Schauer stated that a definitive multicenter randomized controlled trial comparing medical and surgical treatment is needed to determine the true effects of bariatric surgery on patients with diabetes.

“[The study] should evaluate not just blood sugar control but occurrence of diabetic complications. Some might interpret this study as ‘surgery doesn’t cure everybody,’ which is true. But the real value of surgery is not just in achieving remission in some patients, which is remarkable, but its ability to substantially improve even those patients who don’t achieve remission.”