Bariatric Surgery And Mortality: Long-term Follow-up Data
Two interesting studies on bariatric surgery and an accompanying editorial were published in the New England Journal of Medicine yesterday (Volume 357, August 23, 2007). The first study was entitled “Effects of Bariatric Surgery on Mortality in Swedish Obese Subjects,” and written by Lars Sjostrom and colleagues from a group of academic institutions in Sweden. The second study was entitled “Long-term Mortality after Gastric Bypass Surgery,” written by Ted Adams and colleagues at the University of Utah School of Medicine. The editorial was entitled “The Mising Link-Lose Weight, Live Longer,” and written by George Bray at Louisiana State University in Baton Rouge.
In Summary
Taken together the two studies show quite convincingly that bariatric surgery (from an earlier posting you may remember that the term “bariatric” comes from the Greek word “baros” which means weight) has a statistically significant effect on mortality from cardiovascular diseases, diabetes, and cancer. The Swedish study had a rather impressive duration of follow-up with a mean of almost 11 years. The data support earlier cross-sectional studies showing clear-cut benefits from bariatric surgery in people with obesity (BMIs over 40 or over 35 with obesity complications).
Any pitfalls?
I was a bit surprised to find that the Swedish study which was a non-randomized longitudinal design showed so many deaths in the surgery group (101 vs. 129 in the controls) and many more deaths from non-typical obesity-related conditions (e.g., accidents, suicide) than the control group. I have no idea what this means. My main concern with the data is that the only way to really tell how much the surgery works and what the non-obesity-related deaths are all about is to have a randomized trial. This would be a study design where potential study volunteers agree to be in a study about bariatric surgery and would agree to be in either the surgery or control group, as determined by a “flip of the coin.” The problem with both the Swedish and Utah studies was that controls were not selected on a randomized basis. Regardless of their limitations, these data are important and support the use of bariatric surgery in selected patients.
We still have much to learn about this subject. I fear that there will be a steady increase in surgery for patients with lower and lower BMIs, where the long-term benefits are unknown. To all you bariatric surgeons out there- temper your enthusiasm with a dose of good judgment in selecting cases.