PURPOSE: Although splenectomy remains the most effective treatment of chronic idiopathic thrombocytopenia (ITP), the results after splenectomy vary, with the prognostic factors for the postoperative result being debatable. This study investigated the response and predictors of the outcome after splenectomy. METHODS: Forty five adult patients, older than 15 years of age, who underwent splenectomy for ITP between 1995 and 2004, were retrospectively reviewed. The clinical characteristics, therapeutic modality, response rate and predictive factors of the response were analyzed. RESULTS: There were 12 and 33 male and female patients, respectively, with a mean age of 48 years. The preoperative platelet count ranged between 1000 to 47, 000/mm3 (mean: 12, 600). Remission was accessed 6 months after the surgery. Complete remission was observed in 31 patients (65.9%), and a partial response in 4 (8.5%), 12 patients (25.5%) did not respond to the splenectomy. Patients with an elevation of the platelet count to more than 250, 000/mm3 within 2 weeks after the surgery had better responses (P=0.03). However, patients who had splenectomy more than 36 months after the initial diagnosis had poor responses (P=0.02). Age, sex, previous response to glucocorticoids and intravenous immunoglobulin, the initial platelet count and platelet transfusion showed no clinical significance for predicting the response. A single patient had pneumonia postoperatively, there was no mortality. There was no relapse in the complete response group during the follow up period (mean: 27.8 months). CONCLUSION: Splenectomy for patients with ITP was safe and effective treatment. The early recovery of the platelet count and early operation following diagnosis were related with better responses to splenectomy.