Medical Therapy vs. Bariatric Surgery in Type 2 Diabetes

Posted by • April 27th, 2012

In the 91 years since the discovery of insulin therapy for diabetes by Banting and Best in 1921, the prevalence of type 1 and especially type 2 diabetes has skyrocketed, tied to the ongoing epidemic of obesity.

With the global prevalence of type 2 diabetes now expected to hit nearly 10 percent of the world’s population by 2030, the need for new strategies to help patients achieve glycemic control has never been greater. Diabetes has traditionally been perceived as a medical disease, managed with insulin and oral hypoglycemics. However, two studies published in this week’s issue of NEJM report that bariatric surgery – pioneered as a last-ditch effort for weight loss in obese patients – may improve glycemic control more successfully than medication alone.

Both single-center studies compared methods of bariatric surgery to intensive medical therapy for glycemic control in obese patients with poorly controlled diabetes.

In one study, reported by Philip Schauer and colleagues, 150 patients were randomly assigned to either intensive medical therapy for diabetes or medication plus either a Roux-en-Y gastric bypass or sleeve gastrectomy. These were mainly obese patients, with a BMI between 27 to 43 and an average HgbA1C level of 9.2%. While all three groups showed improved glycemic control, those who were randomly assigned to surgery were much more likely to have reached the target HgbA1C goal of 6% or less by 12 months of follow-up. Only 12% of those receiving medical treatment reached this goal, compared to 42% in the gastric-bypass group, and 37% in the sleeve gastrectomy group.

Not surprisingly, weight loss was greater in the surgical than the medical therapy group. However, those who received a bariatric intervention were also less likely to need diabetic medications, lipid-lowering agents, or anti-hypertensive meds by the study’s end.

Four patients in the surgical group suffered complications requiring reoperation, though there were no deaths or complications that were considered life-threatening.

Interestingly, this striking effect of bariatric surgery on glucose control appeared to go beyond the benefits of weight loss itself, according to the authors of an accompanying study.

Geltrude Mingrone and colleagues randomized 60 patients with a BMI greater than 35 and poorly controlled diabetes to receive either conventional medical therapy, a Roux-en-Y gastric bypass, or biliopancreatic diversion. Their endpoint was “diabetes remission,” defined as fasting plasma glucose less than 100 mg per deciliter and a HgbA1C of less than 6.5% without active pharmacologic therapy. Seventy-five percent of patients in the gastric-bypass group and 95% of those with a biliopancreatic diversion reached this goal.  None of the patients assigned to medical therapy met this endpoint of diabetes remission.

Weight loss appears not to be the only mechanism by which diabetes improves in this population. In fact, the authors found no correlation between normalization of fasting glucose and the weight loss achieved after the surgical intervention.  This suggests, Mingrone and colleagues write, that the surgery itself may be affecting insulin sensitivity beyond the changes in weight.

Will these studies usher in bariatric surgery as part of the regimen for diabetes?

In an accompanying editorial Paul Zimmet from Melbourne and K. George Alberti  from London note that  the International Diabetes Federation has  issued a position statement recognizing bariatric surgery as a treatment for obese patients with hard-to-control diabetes.   But important questions still remain: How long do the positive effects of bariatric surgery last? Are there unforeseen side-effects of the surgical procedures? And, finally, when should physicians start considering bariatric surgery for their diabetic patients?

3 Responses to “Medical Therapy vs. Bariatric Surgery in Type 2 Diabetes”

  1. Sumaya Hammami, MD says:

    Mingrone et al randomized severely obese patients with type-2 diabetes (≥5 years) and a glycated hemoglobin ≥7.0% to medical therapy or bariatric surgery. The primary end point was 2-year remission rate (fasting glucose <100 mg/dl and glycated hemoglobin <6.5% in the absence of pharmacologic therapy). The remission relative risk was ≥7.5 in the surgical compared to medical group.

    We have an issue with the choice of the primary end point. Medical therapy was optimized to reach a glycated hemoglobin < 7% (not <6.5%). Further, it is not appealing to apply "the absence of pharmacologic therapy" condition to the medical group. As expected, there were no remissions in the medical group; in order to calculate risk ratios, the authors assumed that remission had occurred in the 2 drop-out patients of the medical group.

    Although bariatric surgery achieved a remarkable remission rate, the comparison with medical therapy for achieving the stated end point is not fair. A more reasonable endpoint for this design would be the percentage of patients who achieved a glycated hemoglobin level <7%.
    Sumaya Hammami, MD & Muhammad Hammami, MD, PhD

  2. B.A.Kottke MDPhDFACC says:

    An excellent editorial. The mechanism is probably an alteration of secretion of known or unknown peptides due to diversion of food away from the lower stomach. If before and after blood samples are available , running them on an HPLC column designed to separate peptides and comparing the peaks in the before and after samples would identify which peaks changed with the remission of the diabetes and could identify significant new peptides,This could provide major new insights into the mechanisms of type II diabetes and lead ti the development of a medical therapy resulting in the same dramatic remission as seen with these specific types of bariatric surgery

  3. Bruce A. Kottke says:

    Typically the remission of the diabetes occurs a fed days after the Rous en y surgery long before any et loss is apparent. A new approach to consider exploring is rather or not left gastric embolization with microbeads to destroy gastric cells producing GLP-1,grehlin, and possibly other unknown peptides would also cause remission of diabetes. This void be done by a transvascular route with much less risk than bariatric surgery