Corticosteroid exposure prior to admission and predicting need for rescue therapy in acute severe ulcerative colitis

Background: Steroid exposure has been associated with poorer outcomes following colectomy in acute severe ulcerative colitis (ASUC). Aim: To examine the effect of prolonged oral corticosteroid therapy immediately prior to admission on the likelihood of requiring rescue therapy along with predictors of intravenous corticosteroid failure on Day 1 of admission. Methods: A retrospective case note and electronic record review was conducted at a tertiary inflammatory bowel disease referral centre of admissions for ASUC meeting Truelove and Witts criteria from 2013 to 2019.The data was analysed for the effect of pre-admission steroid exposure on need for rescue therapy and for predictors of intravenous corticosteroid failure. Results: Ninety-two admissions were identified for ASUC meeting Truelove and Witts criteria. Over 1 week of steroid therapy prior to admission was associated with need for rescue therapy and trended to significance for colectomy at admission and at 12 months. A generalised linear model was constructed with multivariate regression significant for over 1 week of steroid therapy prior to admission, endoscopic Mayo score and albumin. The area under the receiver operator curve for this model was 0.86. Conclusion: Prolonged steroid use prior to ASUC admission is a significant predictor of need for rescue therapy. A generalised linear model incorporating steroid prior to admission, endoscopic Mayo score and albumin was highly accurate at predicting failure of corticosteroid. Consideration should be given for commencement of rescue therapy prior to Day 3, especially in those with prolonged steroid prior to admission.