Life expectancy after surgery: the Swedish Obese Subjects study

Why this study matters

Obesity shortens life expectancy. Earlier reports from the Swedish Obese Subjects (SOS) study showed that bariatric surgery (an operation on the stomach or gut to help with weight loss) was linked to a lower risk of death over about 10 years. What had not been clear was what that means in years of life. This 2020 report answers that question with more than two decades of follow-up and adds a comparison with people from the general population. Canadian readers may find it useful when thinking about the long-term picture, keeping in mind that this was not a randomized trial.

Who was in the study

The analysis included 2007 adults who had bariatric surgery and 2040 adults who received usual obesity care. A third group, the reference cohort, was a random sample of 1135 people from the general population in Sweden. At the time of the analysis (December 31, 2018), the median follow-up for deaths was 24 years in the surgery group (interquartile range 22 to 27 years) and 22 years in the usual care group (interquartile range 21 to 27 years). The interquartile range is the span that contains the middle half of people. Death records were available for 99.9% of study patients. The reference cohort had a median follow-up of 20 years (interquartile range 19 to 21) with records for 100% of participants.

What the researchers did

This was a prospective, controlled cohort study. People were not randomized; those who chose surgery were matched with similar people who received usual obesity care, and both groups were followed forward in time. The published summary does not describe usual care in detail.

The researchers used a statistical model (called a Gompertz proportional hazards regression) to compare death rates and to estimate life expectancy in each group. The extracted data show the analysis was "as-treated" and that people were censored (their follow-up stopped counting) at emigration, change of treatment, withdrawal of consent, or the end of follow-up. This matters because the life expectancy figure is an estimate from a model, not a direct count of years lived.

What they found

In total, 457 people (22.8%) in the surgery group died, compared with 539 (26.4%) in the usual care group. The hazard ratio was 0.77 (95% CI 0.68 to 0.87; P less than 0.001). A hazard ratio compares how often an event happens in two groups over time; a value below 1 means deaths were less frequent in the surgery group. The 95% confidence interval (CI) is the range in which the true value is likely to fall.

The adjusted median life expectancy in the surgery group was 3.0 years longer than in the usual care group (95% CI 1.8 to 4.2 years) (R32-C96). At the same time, it was 5.5 years shorter than in the general population. In other words, surgery was associated with longer life, but people in both study groups still had higher death rates than people in the reference cohort.

Side effects and people who stopped

The 90-day postoperative mortality, meaning deaths within 90 days of surgery, was 0.2%. Repeat surgery was needed in 2.9% of people in the surgery group. The published summary we reviewed does not report other complications or side effects, and it does not give numbers of people who withdrew.

What this study does not tell you

• It shows an association, not proof of cause. Because people chose their treatment rather than being randomized, differences between the groups may partly explain the results.
• The 3.0-year figure is a model estimate for a group. It is not a prediction for any individual.
• Most operations were older procedures done decades ago, not the procedures most common today.
• The "as-treated" analysis stopped counting people when they changed treatment, which can influence the comparison.
• The study does not compare surgery with modern weight-management medicines, which did not exist for most of the follow-up period.
• The 2007 mortality report and the 2012 cardiovascular report use the same people. They are overlapping analyses, not separate confirmations.

What it means in Canada

This is the longest follow-up of bariatric surgery in a controlled study, and Canadian clinicians and economic models often cite it. It does not tell you whether surgery is right for you, and it cannot say how surgery compares with medicines used today. For information on medicines, see semaglutide in Canada and tirzepatide in Canada. For questions about what provincial plans pay for, see public drug coverage by province.

Related reading

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

Source