Sir.—In their article "Steroid Responsiveness: A Predictor of the Outcome of Splenectomy in Children With Chronic Immune Thrombocytopenic Purpura" (Journal 1982; 136:1064-1066), Weinblatt and Ortega raised an important question about the role of steroids as a predictor of the outcome of splenectomy in children with chronic immune thrombocytopenic purpura (ITP). They advocated the use of steroids as a "highly reliable method of correctly identifying those patients with chronic ITP who would benefit from splenectomy." The same sort of conclusion has already been established in pediatric textbooks1 that claim, "History of an earlier platelet response to prednisone . . . [is] a factor that favors recovery after splenectomy." Our experience, in a way, contradicts these notions.
In a survey that we conducted at the Beilinson Medical Center in Petah Tiqva, Israel,2 we found 93 patients with ITP. Twenty-three of these patients failed to recover within 12 months, and their conditions were regarded