BACKGROUND: The utility of bone marow examination (BME) in suspected idiopathic thrombocytopenic purpura (ITP) is an unresolved issue. We have assessed the laboratory findings and the usefulness of routine of BME in children and adults with ITP. METHODS: Data were collected from hospital medical records and laboratory results of 27 children and 36 adults who had undergone BME for investigation of isolated thrombocytopenia between 1996 and 1999. RESULTS: Fifty-nine (94%) of the 63 patients had normal or increased number of megakaryocytes on BME. Fifty-six of them were diagnosed as ITP. One patient (4%) in children and 3 patients (8%) of adults had decreased number of megakaryocytes suggestive of other disorders. In ITP, platelet associated immunoglobulin and reticulated platelet had low sensitivities, 23% and 38% respectively. Mean platelet volume (MPV) waslower than normal control group (P<0.001), but higher than aplastic anemia group (P= 0.025). There was no significant difference in platelet distribution width (PDW) between ITP group and aplastic anemia group. Platelet count and MPV were positively correlated (r=0.606, P<0.001). CONCLUSION: The sensitivities of laboratory tests with peripheral blood for ITP were generally low. However, in childhood and adult patients with suspected ITP the routine performance of BME may be able to be decreased provided that BME should be performed only when the possibility of other disorders is high through a thorough clinical history, physical examination and laboratory findings of peripheral blood.