Can weight-loss surgery prevent type 2 diabetes? The SOS study

Why this study matters

Type 2 diabetes is one of the most common weight-related health conditions. Preventing it matters for long-term health. The Swedish Obese Subjects (SOS) study is one of the longest studies of bariatric surgery, which is surgery on the stomach or intestine to help with weight loss. This report asked whether surgery lowered the chance of developing diabetes over as long as 15 years. It is a cohort study, not a randomized trial, so it must be read with care.

Who was in the study

This analysis included 3,429 adults with no diabetes at the start. Of these, 1,658 had bariatric surgery and 1,771 were matched controls who received usual care. Participants were 37 to 60 years old. BMI, which is weight in kilograms divided by height in metres squared, was 34 or more in men and 38 or more in women. Matching was done at the group level rather than person by person. At the time of this analysis, on January 1, 2012, people had been followed for up to 15 years.

What the researchers did

This was a nonrandomized, prospective, controlled study. Prospective means people were enrolled first and then followed forward in time. Nonrandomized means people were not assigned to surgery or usual care by chance. Instead, those who had surgery were compared with a matched group who did not. Because the groups were not chosen by chance, they can differ in ways that affect the result. In fact, the surgery group had a higher starting weight and more pronounced risk factors than the control group.

The surgeries were gastric banding (19%), vertical banded gastroplasty (69%) and gastric bypass (12%). The control group received usual care. New type 2 diabetes was a secondary endpoint that was planned before the study began.

What they found

During follow-up, type 2 diabetes developed in 110 people in the surgery group and 392 in the control group. That works out to 6.8 cases per 1000 person-years after surgery and 28.4 cases per 1000 person-years with usual care (R32-C97). A person-year is one person followed for one year, so this rate accounts for people being followed for different lengths of time.

The adjusted hazard ratio was 0.17 (95% confidence interval 0.13 to 0.21). A hazard ratio compares the rate at which an event happens in one group with the rate in another. A value of 0.17 means the rate in the surgery group was about 17% of the rate in the control group, after adjusting for differences between the groups. The confidence interval is the range within which the true value would likely fall.

The effect of surgery differed depending on whether a person had impaired fasting glucose at the start, which is a blood sugar level above normal but below the diabetes range. It did not differ by BMI. Sensitivity analyses that filled in missing endpoints did not change the overall conclusion.

Side effects and people who stopped

The paper reports that 0.2% of people died after surgery and 2.8% needed a repeat operation within 90 days because of complications. By 15 years, 36.2% of the original participants had dropped out of the study, and 30.9% had not yet reached the time for their 15-year visit. The summary does not report dropout separately by group.

What this study does not tell you

• This was not a randomized trial. The surgery group started heavier and with more risk factors, so other differences between the groups could affect the result.
• More than a third of participants had dropped out by 15 years, which weakens the long-term estimate.
• The surgeries used reflect Swedish practice at the time. Vertical banded gastroplasty, the most common one here, is rarely used today, and gastric bypass was only 12% of cases.
• The study did not compare surgery with weight-loss medicines, and it cannot be used to rank surgery against them.
• This report is about preventing diabetes, not treating it. Other SOS reports cover other outcomes.
• The BMI entry rules were high. Results may not apply to people with a lower BMI.

What it means in Canada

This study describes what happened with bariatric surgery in Sweden over many years. It says nothing about how surgery compares with today's medicines, which were not available at the time. Readers looking at non-surgical options can read about semaglutide in Canada and tirzepatide in Canada. For coverage questions, see public drug coverage by province. To find bariatric and lifestyle programs, see compare services.

Related reading

Other studies in this library that bear on the same question.

Source