Platelet count evolution as a predictor of outcome after splenectomy for immune thrombocytopenic purpura

Splenectomy is the definitive second-line therapy for refractory immune thrombocytopenic purpura (ITP), and has a reported response rate of 50-80%. Medical attention should be reconsidered when there is no evidence of accessory spleen in refractory ITP patients after splenectomy. The purpose of this study was to determine whether platelet count evolution differs between patients with a successful or unsuccessful result after splenectomy for ITP. Archived records of 104 consecutive patients that underwent splenectomy for ITP were reviewed. Patients were divided into two groups (failures and successes) using a final follow-up platelet count of 100,000/mu L as a cut-off. Platelet count evolutions in these two groups were compared using the Student's t test. Successes and failures were found to have significantly different platelet counts from two days postoperatively (P = 0.016). The area under the receiver operating characteristic curve was 0.630 (95% confidence interval, 0.518-0.741, P = 0.030), and when a cut-off value of 100,000/mu L was used, sensitivity and specificity were 68.2 and 51.2%, respectively. To obtain positive and negative predictive values exceeding 50%, additional platelet counts were required at one week and one month after splenectomy. We propose a protocol for ITP follow-up after splenectomy.