Performance of automated platelet quantification using different analysers in comparison with an immunological reference method in thrombocytopenic patients.

Abstract

Rapidly available and accurate platelet counts play an important role in the evaluation of haemorrhagic status and in assessing the need for platelet transfusions. We, therefore, evaluated platelet counting performance of haematology analysers using optical, impedance and immunological methods in thrombocytopenic patients. We considered 99 patients with a platelet (plt) count under 50 x 10(9) plt/L. We compared the platelet counts obtained using ADVIA 2120 (optical method), Cell-Dyn Sapphire (optical, impedance and immunological methods with CD61) and a reference, double staining (CD41+CD61) immunological method. The platelet counts of all the considered methods showed good correlation with those of the reference method, despite an overestimation in platelet quantification. The degree of inaccuracy was greater for platelet counts under 20 x10(9) plt/L. Clinicians who use platelet thresholds below 20 x10(9) plt/L for making clinical decisions must be aware of the limitations in precision and accuracy of cell counters at this level of platelet count. Inaccurate counts of low platelet numbers could create problems if attempts are made to reduce the threshold below 20 x 10(9) plt/L.