Study Suggests Bariatric Surgery Does Not Decrease Mortality among Obese Veterans
Between 2000 and 2005, the prevalence of obese patients with a body mass index (BMI) greater than 35 increased by 39%; in addition, severe obesity (BMI >40) increased by 50%, and super obesity (BMI >50) increased by 75%. Obesity is difficult to treat, and bariatric surgery is the most effective means to induce weight loss for the severely obese. Existing evidence of the survival associated with bariatric surgery is based on cohort studies of predominantly younger women with a relatively low inherent obesity-related risk of mortality. The association between survival and bariatric surgery for older men is less clear. This retrospective cohort study sought to determine whether bariatric surgery is associated with reduced mortality among Veterans, who are older and predominantly male compared to prior studies. Using VA data, investigators identified 850 Veterans who had bariatric surgery in a VA facility between 1/00 and 12/06 – and who were considered to be “high risk” due to older age (mean age 49.5) and higher body mass index (mean BMI 47.4) compared to patients in previous bariatric mortality assessments. Investigators compared all-cause mortality through 12/08 between the 850 obese Veterans who had received bariatric surgery and 41,244 Veterans with a BMI between 35 and 80 who had not received surgery but had used VA outpatient services. In addition, investigators used propensity score matching to improve equivalence in baseline characteristics and reduce confounding among the surgical and non-surgical groups, and compared mortality among these groups.
- In a propensity-matched cohort of obese, high-risk, predominantly male Veterans (847 who underwent surgery and 847 non-surgical controls), bariatric surgery was not significantly associated with a survival benefit during a median of 6.7 years of follow-up.
- In unmatched comparisons of 850 Veterans who underwent bariatric surgery and 41,244 Veterans who did not, those in the surgical group were significantly younger (49.5 years vs. 54.7 years) had higher BMIs (47.4 vs. 42.0), and had greater comorbidity burden. Surgical patients also were more likely to be super-obese (BMI >50; 31.3% vs. 7.0%).
- However, analyses after matching reduced the significant differences in characteristics between surgical and control patients (e.g., age: 49.5 vs. 49; BMI: 47.3 vs. 47.1; super-obesity 31.3% vs. 41.3%). These analyses also controlled more closely for time of follow-up and showed that the protection conferred by surgery was small and not statistically significant after 6.7 years. Crude mortality only differed statistically between the matched case patients and controls at 6 years: 1-year, 1.5% vs. 2.0%; 2-years, 2.2% vs. 3.5%, and 6-years, 6.7% vs. 12.8%.
- This study did not include patients who had undergone laparoscopic banding procedures.
- These findings did not account for unobserved confounding that may persist even after propensity score matching.
- As suggested in some other studies, survival benefits from surgery-induced weight loss may take longer than 6 years to emerge.
This study was funded by HSR&D (RCS 10-391, IIR 05-201, SHP 08-137). Dr. Maciejewski is supported by an HSR&D Research Career Scientist Award. Dr. Maciejewski and Ms. Smith are part of HSR&D’s Center for Health Services Research in Durham, NC.
Maciejewski M, Livingston E, Smith V, et al. Survival among High-Risk Patients after Bariatric Surgery. JAMA June 15, 2011;305(23):2419-26.