BACKGROUND: For the diagnosis of essential thrombocythemia (ET), no single clinical or laboratory finding of diagnostic value is available and a differential diagnosis of other myeloproliferative neoplasms or reactive thrombocytosis (RT) is needed. Following recent developments in automated blood cell analyzers, various platelet indices can now be measured. In this study, we analyzed whether platelet counts and 6 platelet indices can be used for the differentiation of ET from RT in patients with a platelet count of 600x10(3)/microL or more. METHODS: The subjects studied were 31 patients with ET and 224 patients with RT. The platelet counts, mean platelet volume (MPV), plateletcrit (PCT), platelet distribution width (PDW), mean platelet mass (MPM), mean platelet component concentration (MPC) and large platelets (LPLT) were measured by ADVIA 120 (Bayer Diagnostics, USA). The mean values of each item were compared between the two patient groups and the sensitivity and specificity of each item in the diagnosis of ET were determined by ROC curve analysis. RESULTS: In essential thrombocythemia, all parameters except MPC were significantly higher than in reactive thrombocytosis. For the diagnosis of ET, the sensitivity and specificity were: 74.2% and 84.4%, when the platelet count was > or = 820 x 10(3)/microL; 80.6% and 80.0%, when the plateletcrit was > or = 0.63%; and 64.5% and 99.1%, respectively, when LPLT was > or = 23 x 10(3)/microL. CONCLUSIONS: The platelet counts and platelet indices are useful for the differential diagnosis of thrombocytosis. The plateletcrit and LPLT are particularly useful for the diagnosis of ET when the platelet count is markedly increased.
BACKGROUND: For the diagnosis of essential thrombocythemia (ET), no single clinical or laboratory finding of diagnostic value is available and a differential diagnosis of other myeloproliferative neoplasms or reactive thrombocytosis (RT) is needed. Following recent developments in automated blood cell analyzers, various platelet indices can now be measured. In this study, we analyzed whether platelet counts and 6 platelet indices can be used for the differentiation of ET from RT in patients with a platelet count of 600x10(3)/microL or more. METHODS: The subjects studied were 31 patients with ET and 224 patients with RT. The platelet counts, mean platelet volume (MPV), plateletcrit (PCT), platelet distribution width (PDW), mean platelet mass (MPM), mean platelet component concentration (MPC) and large platelets (LPLT) were measured by ADVIA 120 (Bayer Diagnostics, USA). The mean values of each item were compared between the two patient groups and the sensitivity and specificity of each item in the diagnosis of ET were determined by ROC curve analysis. RESULTS: In essential thrombocythemia, all parameters except MPC were significantly higher than in reactive thrombocytosis. For the diagnosis of ET, the sensitivity and specificity were: 74.2% and 84.4%, when the platelet count was > or = 820 x 10(3)/microL; 80.6% and 80.0%, when the plateletcrit was > or = 0.63%; and 64.5% and 99.1%, respectively, when LPLT was > or = 23 x 10(3)/microL. CONCLUSIONS: The platelet counts and platelet indices are useful for the differential diagnosis of thrombocytosis. The plateletcrit and LPLT are particularly useful for the diagnosis of ET when the platelet count is markedly increased.