High but decreasing rates of reconstruction after total proctocolectomy for ulcerative colitis, and evidence of a direct volume outcome relationship

Abstract

BackgroundTotal (procto)colectomy is indicated in 15%–20% of ulcerative colitis(UC) patients during their disease course. Reconstruction options to avoid a permanent ileostomy include an ileoanal pouch anastomosis (IPAA) or ileorectal anastomosis (IRA). This study aimed to investigate reconstruction rates using Australian‐based population‐level data, and factors influencing reconstruction.MethodsA retrospective data linkage study of the NSW population over a 19‐year period was performed. Patients with UC who underwent total (procto)colectomy with a minimum of 1‐year follow up were included. The main outcome was reconstruction with either IPAA or IRA. The influence of hospital and patient factors on reconstruction rates was assessed by Cox regression.ResultsOverall, 1047 patients underwent a (procto)colectomy for UC (mean age 45.9 years [SD ± 18.3], 640 [61.1%] male). The 5‐year reconstruction rate was 55% (IPAA 89%). Advanced age, emergent colectomy, higher comorbidity burden, and geographical remoteness were significantly associated with lower reconstruction rates. A lower reconstruction rate was also observed in the most recent time‐period (2014–2019) (aHR 0.68[95% CI 0.54–0.86]), and where index (procto)colectomy was performed in low‐volume (<1 pouch/year) pouch hospitals (aHR 0.60 [95% CI 0.43–0.82]).ConclusionsNSW Australia has the highest reported rate of reconstruction following UC (procto)colectomy globally. However, rates reduced in the most recent time‐period. There was variation in reconstruction rates across centres, with primary and overall reconstruction rates proportionate to hospital pouch volume. Reconstruction rates were also lower for patients living outside major cities. To ensure equitable opportunities for reconstruction, patients being considered for IBD pouch surgery should be centralized to a limited number of specialist pouch centres.