Background: Intensive intravenous treatment remains the first line therapy of severe, uncomplicated attacks of ulcerative colitis.
Aim: To predict the failure of intensive intravenous treatment by combining clinical and laboratory parameters with endoscopy findings.
Methods: Retrospective study conducted in a tertiary care referral centre. Failure of intensive intravenous treatment was defined as colectomy before day 30, intravenous cyclosporin, or death. Predictive factors of outcome were assessed using univariate and multivariate prognostic analysis.
Results: Between January 1990 and May 1997, 85 consecutive patients were treated with intensive intravenous treatment for non-response to oral corticosteroids (n=59) and/or severe attack of ulcerative colitis (n=26). There were 41 successes and 44 failures (including 1 death, 13 cyclosporin and 30 colectomies before day 30). Multivariate prognostic analysis found that the presence of Truelove and Witts' criteria (P=0.018), an attack that had lasted more than 6 weeks (P=0.001), and severe endoscopic lesions (P=0.007) were associated with an increased risk of failure. Patients with severe endoscopic lesions and Truelove and Witts' criteria, or an attack of more than 6 weeks had a failure rate of 85-86%.
Conclusion: Clinical, laboratory and endoscopic findings can predict the risk of failure of intensive intravenous treatment. A prospective study is required to confirm these results.