Background

The longer-term risks of bariatric surgical procedures are poorly characterized limiting shared decision making.

Methods

Adults underwent a primary adjustable gastric band (AGB), Roux en Y gastric bypass (RYGB), or sleeve gastrectomy (SG) between 1/1/05 and 9/30/15 in 10 health systems within the National Patient-Centered Clinical Research Network (PCORnet®). Information was extracted from electronic health records (EHRs) using a common data model. The primary outcome was reoperation/re-intervention (RORI) with and without endoscopy, and secondary outcomes included rehospitalization and mortality at 1, 3, and 5-year follow-up.

Results

34,714 adults at 10 centers in 4 clinical data research networks underwent AGB 1,154 (3.3%), RYGB 18,056 (52%), and SG 15,504 (44.7%). The mean age was 45 years and BMI 49 kg/m2. The cohort was predominantly White (66%) with 25% of Hispanic ethnicity. RORI was less likely for SG than RYGB (Hazard Ratio: 0.89; CI 0.83, 0.96; p=0.002), more likely for AGB than RYGB (HR: 1.45; CI 1.28, 1.63; p<0.0001), and more likely for AGB than SG (HR: 1.62; CI 1.42,1.85; p<0.0001). The estimated cumulative incidence of RORI for the average patient at 1, 3, and 5 years was 10.2%, 20.9%, and 27.5% for AGB, 7.2%, 15.0%, and 20.0% for RYGB, and 6.4%, 13.4% and 18% for SG. With endoscopy included at 5 years, RORI was 30.7%, 30.2%, and 22.7% for AGB, RYGB, and SG. The estimated cumulative incidence of rehospitalization at 1, 3, and 5 years was 16.3%, 31.8%, and 42.3% for AGB, 14.4%, 28.5%, and 38.3% for RYGB, and 12.0%, 24.2%, and 32.8% for SG. There were no significant differences in all-cause mortality between procedures.

Conclusions

After bariatric surgery reoperation, reintervention, and rehospitalization events are commonly experienced over time, and occur most often after AGB, followed by RYGB, and then SG. There were no differences in long-term mortality between procedures.