Selection of Automated Platelet Counting Methods Based on Mean Platelet Volume (MPV)

BACKGROUND: Impedance-based detection and optic detection with fluorescence are common platelet counting methods used by modern hematology analyzers. There are few studies to compare the accuracy of platelet counts using these methods in case of increased MPV. METHODS: Sixty patients with immune-related thrombocytopenia (IRTP) and 60 healthy controls were included. Platelet counts were obtained by BC-6900 analyzer using impedance detection (PLT-I) and optic detection with fluorescence (PLT-O). Flow cytometry was used as the reference (FCM-ref). RESULTS: The platelet counts in patients using PLT-I were significantly lower than those using PLT-O or FCM-ref by an average of 13.3%. The platelet counts by PLT-O compared to FCM-ref were not statistically significant. MPV inversely affected the platelet counts. When MPV was < 13 fL, platelet counts by all three methods were not statistically different. When MPV was ≥ 13 fL, platelet counts by PLT-I were significantly lower (-15.8%) than those by PLT-O or FCM-ref. Furthermore, when MPV was ≥ 15 fL, platelet counts using PLT-I were further decreased (-23.6%) compared to the counts obtained by PLT-O or FCM-ref. CONCLUSIONS: The platelet counts by PLT-O in patients with IRTP is as accurate as by FCM-ref. When MPV is < 13 fL, platelet counts by all three methods are comparable. However, when MPV is ≥ 13 fL, platelet counts by PLT-I can erroneously decrease by as many as 23.6%. Therefore, in case of IRTP, or any cases when MPV ≥ 13 fL, platelet counts obtained by PLT-I method should be carefully checked by other methods, such as PLT-O to ensure a more accurate platelet count.