BACKGROUND: Ulcerative colitis (UC) is characterized by a life-long chronic course with remissions and exacerbations. Approximately 15% of patients have a severe attack requiring hospitalization at some time during their illness. These patients are traditionally treated with intravenous corticosteroids, with a response rate of approximately 60%. The patients who do not respond to steroid treatment usually require surgical removal of the large bowel (proctocolectomy or colectomy with an anal pouch). This surgical procedure essentially cures the patient from the disease but is associated with complications such as pouchitis. Few alternative treatments exist for severe ulcerative colitis: immunosuppressive medications (such as azathioprine) have a slow onset of action and are therefore usually ineffective. Antibiotics are not proven to be effective and biological treatments such as infliximab are still under investigation. The introduction of cyclosporine-A (CsA) for use in patients with severe ulcerative colitis (UC) has provided an alternative to patients previously facing only surgical options. Cyclosporine acts mainly by inhibiting T lymphocyte function, which is essential for the propagation of inflammation. Unlike most other immunosuppressive agents, CsA does not suppress the activity of other hematopoietic cells, does not cause bone marrow suppression and has a rapid onset of action. This reviews aims to systematically assess the effectiveness and safety of CsA for severe UC. OBJECTIVES: This review aimed to evaluate the effectiveness of cyclosporine A for patients with severe ulcerative colitis. SEARCH STRATEGY: Electronic searches of The Cochrane Library (Issue 1, 2004), EMBASE (1980-2004), and MEDLINE (1966-2004); hand searching the references of all identified studies; contacting the first author of each included trial. SELECTION CRITERIA: Randomised clinical trials comparing cyclosporine A with placebo or no intervention to obtain and maintain remission of idiopathic ulcerative colitis. DATA COLLECTION AND ANALYSIS: Two reviewers independently appraised the quality of each trial and extracted the data from the included trials. Relative risks (RR) with 95% confidence intervals (CI) were estimated. The reviewers assumed an intention to treat analysis for the outcome measures. MAIN RESULTS: Only two randomized controlled trials were identified that satisfied the inclusion criteria. These two trials could not be pooled for analysis because of major differences in design and patient populations. In the first trial, 11 patients received intravenous cyclosporine (4 mg/kg) and 9 received placebo. Two of 11 in the treatment group failed to respond to therapy compared with nine of nine in the placebo group (RR 0.18, 95% CI 0.05 - 0.64). However, 3/11 and 4/9 eventually underwent colectomy in the treatment and placebo groups respectively and follow-up was less than a month. In the second trial 15 patients were treated with intravenous cyclosporine and 15 with intravenous methylprednisolone. Five of 15 patients in the cyclosporine group failed to respond to therapy as compared to 7/15 in the methylprednisolone group (RR 0.71, 95% CI 0.29 - 1.75). After 1 year 7/9 responders in the cyclosporine group were still in remission compared with 4/8 in the steroid group (p > 0.05) and the colectomy rate was similar in both groups. The mean time to response in the cyclosporine group in the 2 trials was short (7 days and 5.2 days). These results should be interpreted with caution given the small numbers of trials and patients evaluated for comparison, and limited follow-up (few weeks in one trial to a year in the other). The precise assessment of the occurrence of adverse events was difficult because the trials described different adverse reactions, which reversed after discontinuation of cyclosporine. There was no evidence in the trials reviewed that cyclosporine was more effective than standard treatment for preventing colectomy but this effect cannot be excluded due to the small sample size and rarity of this outcome. Additional limitations of current research include lack of data on quality of life, costs and long-term results of cyclosporine therapy. AUTHORS' CONCLUSIONS: There is limited evidence that cyclosporine is more effective than standard treatment alone for severe ulcerative colitis. The relatively quick response makes the short-term use of cyclosporine potentially attractive, but the long-term benefit is unclear, when adverse events such as cyclosporine-induced nephrotoxicity may become more obvious. There is a need for additional research on quality of life, costs and long-term results from cyclosporine therapy in severe ulcerative colitis.