The role of pulsed high-dose dexamethasone (DXM) in the treatment of patients with chronic idiopathic thrombocytopenic purpura (ITP) is still uncertain. Following an early report in which it was described as an effective and well-tolerated treatment with a sustained platelet response in 100% of cases, a number of subsequent studies have failed to confirm such favorable results. As all these studies were conducted on small numbers of patients, we investigated further the effectiveness and side effects of this therapeutic modality in a larger cohort. Thirty-two patients with chronic ITP were scheduled to receive six monthly courses of intravenous DXM at the dose of 40 mg/day for 4 consecutive days. All patients had ITP that had been resistant to between two and five different therapeutic regimens, including 9 patients who had already failed splenectomy. All patients had to be seen 2 weeks after each cycle to asses their response as well as secondary effects. Three patients failed to respond and clinically required other therapy. Thirteen patients (41%) had a partial (platelet count between 50 and 100 x 10(9)/liter) or complete (platelet count >100 x 10(9)/liter) response to treatment, responses being mostly transient. Responses were observed early during the course of treatment, usually right after the first cycle of DXM. There were no late responses. Side effects were mild and did not require discontinuation of treatment. No clinical or laboratory parameter was found to predict treatment outcome. We conclude that high-dose DXM has a limited effect in patients with chronic ITP. Novel approaches and controlled multicenter trials may help identify new therapeutic strategies for this disease.