Early predictors of responses and clinical outcomes of corticosteroid treatment for severe ulcerative colitis

Abstract

Background. Patients with severe ulcerative colitis (SUC) have a high risk of requiring colectomy or resorting to a second-line treatment. However, neither clinical outcomes nor factors predictive of poor response have been clearly established in the treatment of SUC. Objective. To assess prospectively the effects and predictors of corticosteroids (CS) use in clinical outcomes of SUC during 1 year of follow-up. Material and methods. Consecutive inpatients with SUC, who had been treated with intravenous CS, were enrolled. Patients were monitored by clinical, laboratory, and endoscopic examinations, and the data were recorded for 1 year. Univariate and multivariate analyses were performed at 1 week. Results. There were 22.6% (14/62) nonresponders at 7 days. Several predictors were associated with nonresponse to CS. These included Mayo Score at baseline (p = 0.007), partial Mayo Score, number of bowel movements, blood presence in stool, abdominal pain, and levels of C-reactive protein (CRP), hemoglobin (Hgb), platelet count (PLT), and erythrocyte sedimentation rate (ESR) on day 3 (p < 0.05). Multiple logistic regression analysis identified the Partial Mayo Score at day 3 as an independent predictor of outcome (p = 0.012). A total of 12 patients underwent colectomy within 1 year. The short-term response rates to intravenous cyclosporin (CsA) and infliximab (IFX) in SUC were 71.4% (5/7) and 77.8% (7/9), respectively. Conclusions. Many patients with SUC eventually became refractory to or dependent on CS. The Mayo score and laboratory characteristics were factors useful in predicting short-term outcome of CS treatment. Secondary medical therapy can help avoid emergency surgery.