Unexpected variation in outcomes following total (procto)colectomies for ulcerative colitis in New South Wales, Australia: a population-based 19-year linked-data study.

Abstract

Total (procto)colectomy for ulcerative colitis (UC) is associated with significant morbidity, which is increased in the emergency setting. This study aimed to evaluate the outcomes following total (procto)colectomies at a population level within New South Wales (NSW), Australia, and identify case mix and hospital factors associated with these outcomes. A retrospective data linkage study of patients undergoing total (procto)colectomy for UC in NSW over a 19-year period (2001-2020) was performed. The primary outcome was 90-day mortality. The influence of hospital level factors (including annual volume) and patient demographic variables on outcomes was assessed using logistic regression. Temporal trends in annual volume and evidence for centralization were assessed. In all, 1418 patients (mean 47.0 years [SD 18.7], 58.7% male) underwent total (procto)colectomy during the study period. The overall 90-day mortality rate was 3.2% (emergency 8.6% and elective 0.8%). After adjusting for confounding, increasing age at total (procto)colectomy, higher comorbidity burden, public health insurance (Medicare) status, emergency operation and living outside a major city were significantly associated with increased mortality. Hospital volume was significantly associated with mortality at a univariate level, but this did not persist on multivariate modelling. Outcomes of UC patients undergoing total (procto)colectomy in NSW Australia are comparable to international experience. Whilst higher mortality rates are observed in low volume and public hospitals, this appears attributable to case mix and acuity rather than surgical volume alone. However, as inflammatory bowel disease surgery is not centralized in Australia, only one NSW hospital performed >10 UC total (procto)colectomies annually. Variation in mortality according to insurance status and across regional/remote areas may indicate inequality in the availability of specialist inflammatory bowel disease treatment, which warrants further research.