Restorative surgery for ulcerative colitis in the elderly: an analysis of ileal pouch-anal anastomosis procedures from the American College of Surgeons National Surgical Quality Improvement Program.

Pellino, G; Reif de Paula, T; Lawlor, G; Keller, D S

BACKGROUND Ileal pouch-anal anastomosis (IPAA) for ulcerative colitis (UC) can be performed as either 2- or 3-stage procedure. IPAA in the elderly has been reported as safe and feasible, but little work to date has assessed outcomes by procedure. The aim of our study was to assess use and short-term outcomes of 2- and 3-stage IPAA in older adults. METHODS The American College of Surgeons National Surgical Quality Improvement Program (ACS-NSQIP) Database was searched for ≥ 65-year-old patients who underwent IPAA for UC in 2- or 3-stage from 2012 to 2016. The primary endpoint was the rate and trends of the two approaches over time. Secondary endpoints included 30-day adverse events and complication-associated costs. RESULTS Overall, 123 patients were included: 77.5% (n = 83) 2-stage and 40 (32.5%) 3-stage IPAA. Mean age was 68.7 ± 3.9 years, with 43 (34.9%) women. The use of the 3-stage IPAA increased over time (18.8% in 2012 vs. 33.3% in 2016), with decreasing use of 2-stage IPAA (81.3% vs. 66.7%, p < 0.001). The morbidity associated with the procedures decreased over time, overall (81.3% in 2012 and 51.5% in 2016, p < 0.001) and in each group individually. No differences were observed in postoperative complications across groups (45.8% 2-stage, 32.5% 3-stage). The overall mean costs of care when no postoperative complications occurred was $25,910, vs. $38,577 when any complication occurred (p < 0.001), but no differences were observed between groups. CONCLUSION In a national analysis, there was a trend of increasing 3-stage vs. 2-stage IPAA for UC in older Americans. Complications and complication-associated costs were comparable across approaches, suggesting that the choice of procedure type should be based on the specific patient comorbidities and surgeon preferences.