Day 3 Oxford criteria predict steroid non-response for acute severe ulcerative colitis in the post biologic era

Outcomes of patients admitted with acute severe ulcerative colitis (ASUC) in the post biologic era are under explored, as well as the ability of scoring indices to predict early steroid non-response.

This retrospective cohort study included adults hospitalized with ASUC (2010-2022) at London Health Sciences Centre, Canada. Steroid response, need for rescue therapy, colectomy during index hospitalization, and colectomy and hospitalization at 3- and 12-months following discharge was assessed. Logistic regression identified predictors of steroid non-response, defined as need for rescue therapy or colectomy during hospitalization.

Of 261 adults hospitalized with ASUC (male: 51.7%, mean age: 40.6 years), 71.2% had extensive colitis. After intravenous corticosteroid therapy during index admission, 55.7% (n = 147) had a response, 37.9% (n = 99) received rescue therapy (infliximab: 98, tofacitinib: 1, and cyclosporine: 0), and 8% (21/261) underwent colectomy. Additionally, 11.6% (28/240) of patients discharged from hospital underwent colectomy within the first 12 months (8.3% at 3-months and 3.3% between 3 and 12 months). There was no difference between steroid responders and non-responders for colectomy (11% vs 12.6%) or hospitalization (33.5% vs 32.6%) at 12 months. The overall cumulative probabilities of colectomy for the entire cohort at 1 year, 3 years, and 5 years were 13.5%, 16.1%, and 17.4%, respectively. On multivariate analysis, Day 3 Oxford criteria was the only factor found to be statistically significant in predicting steroid non-response (odds ratio 4.70, 95%CI [1.06-20.80]).

Day 3 Oxford criteria was an independent predictor of steroid non-response. The risk of colectomy remains substantial after discharge despite low in-hospital colectomy rates following an episode of ASUC. Initial steroid response did not affect long-term colectomy rate at 12 months.