Multicenter evaluation of analytical performances of platelet counts and platelet parameters: Carryover, precision, and stability

Introduction The correctness of the results of automated platelet analysis is still highly debated. The aim of this multicenter study, conducted according to international guidelines, was to verify the analytical performance of nine different types of hematology analyzers (HAs) in the automated platelet analysis. Methods Four hundred eighty‐six peripheral blood samples (PB), collected in K3EDTA tubes, were analyzed by ABX Pentra, ADVIA2120i, BC‐6800, BC‐6800 Plus, Cell‐DYN Sapphire, DxH800, XE‐2100, XE‐5000, XN‐20 with PLT‐F App. Within‐run imprecision and between‐run imprecision were carried out using PB and material control, respectively. The carryover, low limit of quantification (LoQ), and the PB stability were evaluated. Results The carryover was absent for all HAs. The LoQ of PLT ranged between 2.0 (Cell‐Dyn Sapphire) and 25.0 × 109/L (ADVIA 2120i), while immature platelet fraction (IPF) ranged between 1.0 (XN‐20) and 12.0 × 109/L (XE‐5000). The imprecision (%CV) increases as the platelet count decreases. No HAs showed desirable CVAPS for PLT counts less than 50.0 × 109/L, with the exception of Cell‐DYN Sapphire (CV 3.0% with PLT‐O mean value of 26.7 × 109/L), XN‐20 (CV 2.4% with PLT‐F mean value of 21.5 × 109/L), and BC‐6800 Plus (CV 1.9% with PLT‐O mean value of 26.5 × 109/L). The sample stability ranged between under two hours for MPV by ADVIA2120i and 8 hours for other PLT parameters and HAs. Conclusion The findings of this study may provide useful information regarding carryover, precision, and stability of platelet counts and parameters, especially in thrombocytopenic samples. Moreover, the stability of sample for platelet analysis is conditioned by the HA and by temperature and storage time.