A new analysis of the Swedish Obese Subjects (SOS) study shows that bariatric surgery is associated with about a 3-year increase in lifespan, compared with obese patients who do not undergo surgery. Still, surgery did not restore normal lifespan: Surgical patients’ lifespan remained less than that of a sample from the general Swedish population. The study follows other reports suggesting reduced mortality after bariatric surgery, but with a longer follow-up.
“These data add even more evidence to the growing literature showing that patients who undergo bariatric surgery experience a reduction in all-cause long-term mortality. In making decisions around bariatric surgical procedures and care, patients and their health care providers need to understand the trade-offs between improved weight, health, and longer-term survival versus the surgical risks and problems over time,” said Anita P. Courcoulas, MD, MPH, chief of minimally invasive bariatric and general surgery at the University of Pittsburgh Medical Center, said in an interview. Dr. Courcoulas was not involved in the study.
The results appeared in the.
The SOS study drew from 25 surgical departments and 480 primary health care centers in Sweden. The researchers examined data from 2,007 patients who underwent bariatric surgery between 1987 and 2001, and compared their outcomes to 2,040 matched controls. All were between age 37 and 60 years, with a body mass index (BMI) of at least 34 kg/m2 for men and 38 for women. They also compared outcomes with 1,135 randomly sampled from the Swedish population registry.
Procedures included banding (18%), vertical banded gastroplasty (69%), and gastric bypass (13%). After an initial BMI reduction of about 11, the surgery group stabilized by year 8 at a BMI about 7 lower than baseline, and there was little change in BMI among controls.
After a mean follow-up of 24 years (interquartile range, 22-27 years), there were 10.7 deaths per 1,000 person-years in the surgery group, 13.2 among obese controls, and 5.2 in the general population (hazard ratio, 0.77 for surgery versus no surgery; P < .001). The general population had a lower mortality than nonsurgical controls (HR, 0.44; P < .001).
The surgery group had a higher median life expectancy than controls (median, 2.4 years; adjusted difference, 3.0 years; P < .001). The general population group had a median life expectancy that was 7.4 years higher than the control group (adjusted difference, 8.5 years; P < .001). The surgery group’s median life expectancy was still shorter than the general population reference (adjusted difference, 5.5 years; P < .001).
Cardiovascular disease risk was lower in the surgery group (HR, 0.70; 95% confidence interval, 0.57-0.85), as was risk of MI (HR, 0.51; 95% CI, 0.33-0.79), heart failure (HR, 0.52; 95% CI, 0.31-0.88), and stroke (HR, 0.45; 95% CI, 0.24-0.84). Cancer mortality was also lower (HR, 0.77; 95% CI, 0.61-0.96).
In the surgery group, causes of death that were elevated over the general population included cardiovascular causes (HR, 2.64; 95% CI, 1.78-3.91) and noncardiovascular causes, mainly infections; postsurgical complications; and factors such as alcoholism, suicide, or trauma (HR, 1.50; 95% CI, 1.18-1.91).
The study is limited by its retrospective nature, and because the surgical techniques used at the time are less effective than those used today, and could lead to weight gain over time. As a result, many patients who underwent surgery remained heavier than the general population. It’s also possible that negative health effects accumulated before surgery and persisted afterwards, according to Dr. Courcoulas.
The findings are likely generalizable to people with obesity, many of whom choose not to undergo bariatric surgery despite the potential benefits. “The population studied in SOS had a similar profile of underlying medical diseases to those groups who undergo bariatric surgery today and in the U.S. and around the world,” said Dr. Courcoulas.
The study was funded by the Swedish Research Council and others. Dr. Courcoulas has no relevant financial disclosures
SOURCE: Carlsson L et al. N Engl J Med. 2020 Oct 15. .