Bariatric Surgery to Cure Diabetes: Two Compelling Studies But There Are Still Four Reasons for Healthy Skepticism
Posted Mar 27 2012 10:19pm
A cure for diabetes?
Mrs. Jones (name changed) was obese. Her weight remained persistently high despite education and entreaties about diet and exercise. She hated taking all those medications. She dreaded bathing suits.
And then.... she had bariatric surgery. She shed pounds faster than Supreme Court justices spanking a health insurance mandate. Instead of having a corpulent and unhealthy patient, the Disease Management Care Blog had a svelte and healthy patient.
Based on witnessing first hand patient transformations like this, the DMCB knows that bariatric surgery for obesity works.
Despite clinical anecdotes, however, obesity surgery skeptics have pointed out that the evidence has been marred by the lack of any prospective randomized clinical trials. Looking backwards at outcomes data can't rule out the possibility that something else was going on to account for the surgery's apparent success. Most of all, this includes self-selection bias where patients, who are destined to independently do well, select surgery. By leaving assignment to chance, docs and patients are out of the decision-making. This randomization helps researchers be more confident that the surgery, and nothing else, accounts for any observed outcomes.
Enter the New England Journal of Medicine, which published the results from two landmark prospective and randomized trials that compared obesity surgery to conservative medical management 1. Geltrude Mingrone and colleagues screened 72 and then randomly assigned 60 diabetic persons with a BMI of 35 to either a) conventional medical therapy (targeting an A1c of 7% using a multidisciplinary team with visits every 3 months for a year and then one additional visit at two years) or b) gastric bypass or c) biliopancreatic diversion . The study occurred at Rome's Catholic University. Follow-up lasted two years. At the end of the study, 56 patients' data were available for analysis. 15 of the 20 patients who had gastric bypass, 19 of the 20 patients who had the diversion and zero of the medically treated patients were off diabetes medications and had normal blood glucoses. As expected, surgery resulted in a whopping decrease in the BMI down to approximately 29. In contrast the mean BMI was 43 in the medically treated group. Two patients had the surgical complications of hernia and obstruction
2. Philip Schauer and colleagues screened 218 patients and randomly assigned 150 diabetic obese persons with BMIs ranging from 27 to 43 to either a) medical therapy (life style counseling, weight management home glucose monitoring and medications with diabetes specialist clinic visits every three months that targeted an A1c of 6%) or b) " Roux en Y" bypass or c) sleeve gastrectomy . The study occurred at Cleveland Clinic. Follow-up lasted one year and 140 patients' data were available for analysis. 5 of 41 patients in the medical therapy group vs. 21 in the 50 assigned to gastric bypass and 18 of 49 who had the sleeve achieved the A1c of 6%. What's more, most of the surgery patients who achieved the targeted A1c were off all diabetes medications. As expected, the surgery groups decreased their BMI down to the 26-27 range, while the medically treated patients' BMI remained essentially unchanged. In contrast to Rome, there was a wider range of complications that included 4 reoperations and 1 patient that developed a gastrointestinal leak with peritonitis.
Based on these data plus less pristine studies, its clinical experience and common sense, the DMCB is convinced that bariatric surgery works. These two studies are an important step forward in building the case for the use of this approach in persons with obesity and diabetes.
That being said, there is still room for some skepticism. That lingering doubt could be enough for a commercial insurer to limit coverage. It may be enough to prompt a PCP to recommend that an obese patient with diabetes should still hold off on surgery a bit longer. It may be enough for patients and families to wait another year until there are more confirmatory studies.
Here's why 1. The studies were not "blinded." The purpose of "blinding" is to keep patients and doctors from being swayed by a " placebo effect ." While that's intuitively silly, there is a possibility that having abdominal surgery made those patients believe they were going to lose weight and be cured of diabetes. After all, sham surgery has been known to help angina chest pains.
2. The studies are not necessarily "generalizable." Both studies were conducted by teams of surgeons from single institutions. While we can take Rome and Cleveland's word for it, we don't know if obesity surgery done at Bumkinville's Our Mother of Holy Deficit Hospital will have the same success and low rates of complications.
3. Speaking of complications, both studies were not adequately "powered" to fully assess mishap rates. While there were small single digit differences in the rates of complications, the small numbers may not tell the whole story. Having more patients enrolled in these studies would have increased the ability to meaningfully quantify all the possible bad outcomes . That was one of the lessons of the Vioxx catastrophe .
4. Last but not least, the success of the surgery may have been inflated by the relatively poor performance of the non-surgical comparison groups. We know very little about the "intensity" of the medical treatment, other than they had the benefit of accessing a multidisciplinary clinic every three months. Population health management experts know that lifestyle change requires an intense program that includes engagement in a personalized and multidimensional care plan that includes far more frequent in-person and telephonic coaching. We don't know if the medical therapy group achieved this level of excellence.
Despite these limitations, however, the DMCB is more convinced that, for patients in whom nothing else works, bariatric surgery can reverse diabetes.