The Swedish Obese Subjects (SOS) study is a continuing, nonrandomized, controlled, prospective study comparing health outcomes in 2,010 obese patients who had bariatric surgery with a matched control cohort of 2,037 patients who received only the standard obesity treatment (the nature of which was up to individual facilities). The original results of the study, regarding patient mortality, were published in 2007. The current study's outcome measures were the incidence of fatal or nonfatal cardiovascular events, specifically myocardial infarction (MI) and stroke.

From 1987 to 2001 the researchers enrolled patients between the ages of 37 and 60 years with a high body mass index (BMI; 34 or higher in men and 38 or higher in women). Among surgery patients, the most common procedure performed was vertical banded gastroplasty (68%), followed by banding (19%) and gastric bypass (13%). The follow-up time was approximately 15 years.

There were a total of 234 cardiovascular events in the control group, compared with 199 events in the surgery group and 49 cardiovascular deaths in the control group, compared with 28 in the surgery group. After adjustment for baseline patient variables, the patients who underwent bariatric surgery had significantly lower numbers of deaths from MI and stroke and of total (fatal and nonfatal) MI and stroke than patients in the control group. Surprisingly, patients' actual weight loss wasn't associated with a reduction in cardiovascular events.

A secondary analysis also found that patients with higher plasma insulin levels at baseline were significantly more likely to benefit from the bariatric surgery, in terms of both MI and stroke, over the long term. Other variables, such as BMI and triglyceride levels, weren't associated with the same benefit.

The study authors concluded that bariatric surgical intervention for obesity reduces the long-term incidence of MI and stroke more than standard obesity treatment. Additionally, practitioners should consider the findings related to plasma insulin levels when identifying patients who might benefit from these surgical treatments.-AK

Reference