Re‐evaluation of the role of azathioprine in the treatment of adult chronic idiopathic thrombocytopenic purpura: a report on 53 cases

Publisher Website
Google Scholar
Abstract
Summary: We treated 53 adults (mean age 54 years, range 17‐89; 37 females and 16 males) with chronic idiopathic thrombocytopenic purpura (ITP) by azathioprine. All patients had received at least one form of therapy (including splenectomy in 40 patients) and had less than 50 x 109/1 platelets. The duration of ITP before azathioprine was started ranged from 6 to 350 months (median 19). All patients initially received 150 mg/d of azathioprine. This was associated with a short initial course of prednisone (0.3–0.5 mg/kg d) in 10 of them, who were refractory to prednisone alone. 34 (64%) patients responded, including 24 (45%) complete remissions (CR), three (6%) partial remissions (PR) and seven (13%) minor responses (MR). Median time to achieve response was 4 months. 17 of the CR persisted after 7–182 months, 10 of them after discontinuation of azathioprine. Seven patients relapsed after 4–26 months, five of them after azathioprine was stopped or its dose was reduced. PR were short and the median duration of MR was 8 months. Overall, 21 patients (40%) had responses lasting 1 year or more and 17 (32%) lasting 2 years or more. Median duration of treatment was 18 months (range 3–84). Five patients died of bleeding while on treatment. No prognostic factors for response to azathioprine were found. Mild leucopenia was seen in seven patients and a moderate (x3) increase in transaminases in two patients. No opportunistic infections were seen and no malignancy has occurred since the onset of azathioprine. We conclude that azathioprine gives a relatively high incidence of durable responses and very limited side effects in chronic ITP, when splenectomy has failed or is contraindicated. This efficacy, in our experience, is superior to that obtained with other therapeutic approaches. As responses may be delayed, a course of azathioprine of 4 months is required before one can infer a failure to respond. In responding patients, however, the optimum duration of treatment remains to be established.