Search in the Abstract Database

Search Abstracts 2013

* = Presenting author

P486. Intravenous corticosteroids in moderate active ulcerative colitis not responding to oral corticosteroids

J. Llaó1, J.E. Naves2, A. Ruiz-Cerulla3, L. Marín2, M. Mañosa4, L. Rodríguez-Alonso3, E. Cabré4, E. Garcia-Planella1, J. Guardiola3, E. Domènech4, 1Hospital Santa Creu i Sant Pau, Gastroenterology, Barcelona, Spain, 2Hospital Universitari Germans Trias I Pujol, Dept. of Gastroenterology, Badalona, Spain, 3Hospital Universitari Bellvitge-IDIBELL, Gastroenterology, L'Hospitalet de Llobregat, Spain, 4Hospital Universitari Germans Trias I Pujol-CIBEREHD, Dept. of Gastroenterology, Badalona, Spain


Oral corticosteroids (CS) remain the mainstay of conventional therapy for moderate flares of ulcerative colitis (UC). In patients who fail to respond to oral CS, it is recommended to attempt the intravenous (i.v.) route before starting rescue therapies, although no evidence supports this strategy.

Aims: To evaluate the efficacy of i.v. CS in moderate attacks of UC according to previous failure to oral CS or not, and to identify differences in long-term outcomes.


UC patients admitted to 3 University hospitals between January 2005 and December 2011 were identified from electronic databases. Disease severity was defined according to the Montreal classification, and only patients with moderate active UC treated with i.v. CS were included. Patients were grouped depending on previous treatment with oral CS for the index flare. Main end-points were: initial efficacy (defined as mild or inactive disease according to Montreal classification at day 7 after starting i.v. CS without rescue therapy) and the long-term clinical outcome (steroid dependency, colectomy).


110 attacks were included (89 patients), with a median time from UC diagnosis of 28 months (IQR, 3–108). 45% of them were initially treated with oral CS without response, with a median dose of 60 mg/d (IQR, 50–60) and during a median time of 10 days (IQR, 7–17). No differences in clinical features and biological parameters between both groups, except for younger age and lower C-reactive protein levels at the beginning of i.v. CS in the group of patients initially treated with oral CS. The i.v. CS dose was 60 mg/d (IQR, 50–60) and the median concentration of C-reactive protein at the beginning of i.v. treatment was 44 mg/L (IQR, 16–88). Initial response was achieved in 75%, without differences between the both study groups (78% vs. 75%). Rescue therapy during the admission was required in 26% of cases, with a colectomy rate of 3%. No predictive factors to initial response to i.v. CS were found. After a median follow-up of 12 months (IQR, 4–24), 35% of initial responders developed steroid-dependency and up to 13% required colectomy. The unsuccessful response to oral CS was the only factor associated to steroid-dependence in the long-term (54% vs. 18%, P = 0.001).


Intravenous CS are efficient for inducing remission in moderately active UC not responding to oral CS, but almost half of the patients develop steroid-dependency later on. Alternative therapeutic strategies should be assessed in this clinical setting.