Clinical outcomes and predictive factors differ in acute severe ulcerative colitis for East Asian and Western patients.

Abstract

Background: There are limited comparisons of clinical outcomes of acute severe ulcerative colitis (ASUC) between East Asian (EA) and Western countries and many currently used scoring systems are based on Western populations. We developed a predictive model using ASUC data from EA patients and assessed its ability to predict clinical outcomes in an Australia/New Zealand (ANZ) cohort. Method(s): This retrospective international study was conducted across 23 referral hospitals in EA and ANZ. Patients who met Truelove and Witts criteria for ASUC between January 2015 and December 2022 were included. We compared the 1-year colectomy rates and non-response to corticosteroid therapy (NRS) between the EA and ANZ cohorts. Logistic regression analysis was employed to develop predictive models for 1-year colectomy and NRS. Each variable found as an independent predictor in the logistic regression analysis was weighted as score 1 making the scoring system range 0 to 3 or 4. Result(s): A total of 826 patients with ASUC (411 EA and 415 ANZ) were included. The 1-year colectomy rate was significantly lower in the EA group (3.9%) compared to the ANZ group (22.7%, p<0.001), as was the NRS rate (25% vs. 58.9%, p<0.001). In the EA cohort, independent risk factors for 1-year colectomy included female sex, previous exposure to tumor necrosis factor inhibitors, and albumin levels <3 g/dL at admission. For NRS in the EA cohort, independent risk factors included age at diagnosis <37 years, baseline steroid use, albumin levels <2.5 g/dL at admission, and presence of extraintestinal manifestations. The scoring system based on this model did not predict colectomy and NRS risk effectively in the ANZ cohort (p=0.106 and p=0.012, respectively) compared to the EA cohort (p<0.0001 and p=0.001, respectively). In contrast, previously developed predictive models from Europe, including the French colectomy score and the ADMIT-ASC index for NRS, successfully predicted outcomes in the ANZ cohort (p=0.007 and p<0.0001, respectively) but were less effective in the EA cohort (p=0.106 and p=0.026, respectively). Conclusion(s): Clinical outcomes and predictors of ASUC differ between EA and ANZ patients. Further investigation is needed to determine whether these differences arise from variations in management strategies or disease behavior between these groups.