The Annals of Internal Medicine has just published a review of surgical interventions to treat type 2 diabetes, "Effect of Bariatric Surgery on Type 2 Diabetes Mellitus." The authors reviewed published English-language studies where 10 or more diabetic patients had been enrolled, and where diabetes-related outcomes were reported. Ten such studies were identified.

The authors point out that "Nearly 30% of patients who undergo bariatric surgery have type 2 diabetes, and for many of them, diabetes resolves after surgery (84% to 98% for bypass procedures and 48% to 68% for restrictive procedures). Glycemic control improves in part because of caloric restriction but also because gut peptide secretion changes."

That's the good news.

But the bad news is that "Conclusions about bariatric surgery and diabetes resolution come with an important qualifier: The studies had serious methodological weaknesses. Few are randomized, controlled trials; most surgical outcome studies are uncontrolled case series with considerable missing data. In 1 meta-analysis, one quarter of the studies did not report enrolling consecutive patients and fewer than 50% reported how many enrolled patients provided follow-up data. "

In other words, the patients seemed to improve, but the study designs and reporting of results were so vague that it's difficult to guess what might have happened to the patients who apparently had the surgery, and didn't improve. One would like to hope they had the same good outcomes, but I guess I'll be skeptical if the data isn't available.

Which is why I continue to say that any person with diabetes who might be contemplating such surgery should be enrolled in a study that's well-designed, and which has appropriate staff at the study site to assist with perioperative problems, and where you as the patient sign an "Informed Consent Form" (ICF) that you know that you are participating in a study. Such an ICF is different from the usual surgical consent forms, and should describe clearly the anticipated benefits and possible risks, whether the sponsor of the trial will pay for your medical and surgical care, that you have freedom to withdraw from the study at any time, confidentiality and privacy issues, and lots more. An Institutional Review Board (IRB) should be overseeing the conduct of the trial, and have authority to demand changes in the trial if patient safety is at risk.

Ideally, the best study from a scientific viewpoint (although perhaps difficult to justify ethically) would be to randomize patients into several groups, and compare the outcomes. If you signed the informed consent, and met the inclusion and exclusion criteria for the study, you would be randomly assigned to one of the following groups:

1) receiving dietary and lifestyle advice but no surgical intervention.
2) receiving dietary and lifestyle advice and surgical intervention.
3) receiving dietary and lifestyle advice and sham surgical intervention. If in this group, you would be wheeled into the operating room, and have an incision in your belly identical to if you were going to have "real" surgery, but the surgeon would not change anything internally.