Study designs of evaluations included in the review
The authors did not state any inclusion or exclusion criteria relating to the study design. Only studies reporting data from five or more patients were eligible.
Specific interventions included in the review
Interventions aimed at establishing a durable complete response (normal platelet count for at least 3 months) were eligible. Articles describing treatment with glucocorticoids (except for regimens of intermittent high-dose dexamethasone), intravenous immunoglobulins and anti-Rh(D) were excluded. The interventions included in the review were azathioprine, vinca alkaloids, danazol, cyclophosphamide, high-dose dexamethasone, rituximab, interferon, cyclosporine, accessory splenectomy, vitamin C, dapsone, anti-(Rh)D-opsonised D+ erythrocytes, cyclophosphamide/stem cell support, mycophenolate mofetil, interleukin 11, 2-chlorodeoxyadenosine, colchicine, plasma exchange, combination chemotherapy, WEB 2086 BS, Campath-1H and protein A immunoadsorption. The number of studies per treatment ranged from 1 to 34.
Participants included in the review
Eligible participants were adults (older than 16 years) with ITP of over 3 months' duration who had undergone a splenectomy and whose platelet count was less than 50E9 cells/litre. Of the patients included in the review, 56% had pre-treatment platelet counts of less than 30E9 cells/litre and 17% had counts of less than 10E9 cells/litre. The number of eligible treated patients ranged from 2 to 109.
Outcomes assessed in the review
The primary outcome assessed was platelet count response, defined as:
complete (normal platelet count of 150E9 cells/litre or as defined in the original report, maintained for at least 3 months without further treatment),
partial (platelet count of more than 50E9, 30E9 or 10E9 cells/litre depending on the patient's baseline count, maintained for any duration with or without additional treatment), or
none.
The occurrence of bleeding and mortality were also assessed.
How were decisions on the relevance of primary studies made?
Two authors independently reviewed studies for relevance. Any disagreements were resolved by consensus among all the authors.