Bariatric Surgery and MI risk

October 28, 2009 (Washington, DC) — A long-term prospective study of bariatric surgery for weight loss in obese individuals showed the procedure has no effect on the rates of MI when compared with individuals who underwent conventional care.

While the rates of MI were equivalent in follow-up out to 20 years, investigators suggest there was a significant effect of surgery on fatal and nonfatal MI in subjects with elevated baseline glucose levels.

"Patients with elevated glucose levels are at higher risk, and it is reduced with surgery," lead investigator Dr Lars Sjöström (Gothenburg University, Sweden) told heartwire . "For patients with lower glucose levels, those below the median, glucose levels aren't reduced as much, so we're unable to reduce the outcome on myocardial infarction in the same way when compared with those who have higher levels."

He cautioned, however, against making too much of the MI benefit with surgery in patients with elevated glucose levels, because the analysis was post hoc and needs to be confirmed in other studies.

The results of the study, an analysis of the Swedish Obese Subjects (SOS) study, were presented here this week during the late-breaking clinical-trials session at Obesity 2009.

Mortality Benefits Observed Earlier

The SOS study was initiated in 1987 and is a prospective, matched, surgical-interventional trial that includes 4047 obese subjects at 25 surgical departments and 480 primary-healthcare centers.

Regarding the primary end point of the study, investigators showed that 10 years after bariatric surgery, the procedure was associated with a 29% reduction in total mortality when compared with conventional treatment [1]. These results were published in the New England Journal of Medicine in 2007. Other analyses have shown that surgery was also associated with reductions in the incidence of new diabetes mellitus and reductions in cancer in women. As recently reported by heartwire , there is also growing enthusiasm for performing bariatric surgery, even in people who do not meet the definition of obese, in response to evidence suggesting that the procedure can "cure" diabetes.

In this analysis, investigators presented data on the secondary end point, the effect of bariatric surgery on fatal and nonfatal MI. Patients were well matched regarding clinical characteristics, including baseline body-mass index. As reported previously, patients who underwent surgery lost significantly more weight after one, two, and 15 years, although the weight loss occurred primarily in the first year. From baseline, surgical patients lost approximately 30 kg in the first year compared with just 1 kg in the conventional-treatment arm.

After a median follow-up of 12.9 years, investigators showed there to be essentially no difference in event rates between the two treatment groups. The cumulative incidence curves were separated maximally after 11 years but later converged. Overall, there were 104 MIs in the surgery arm and 113 events in the control group. As noted, only patients with elevated blood glucose levels at baseline--those with levels higher than the median 4.72 mmol/L--had a reduction in MI events, and this finding was driven by the benefit in men only.

Sjöström said the overall results are "discouraging" and pointed to the low event rates in SOS as a possible reason for the negative findings. He said other studies have shown a benefit of bariatric surgery on cardiovascular disease event rates, including one that suggested a 50% to 60% reduction in risk. He cautioned, however, that these are retrospective studies, and the field is littered with conflicting findings.

"There are a number of studies that have shown that weight loss is associated with improved cardiovascular risk factors, suggesting perhaps that weight loss would also be associated with decreased incidence of cardiovascular disease," said Sjöström. "However, the controlled, prospective interventions proving that weight loss is in fact associated with decreased cardiovascular disease have so far been lacking."

References