Comparison of automated platelet counts and potential effect on transfusion decisions in cancer patients.

Objectives: To evaluate the accuracy of platelet counts from various hematology analyzers using a reference immunologic method.

Methods: We tested 403 samples with platelet counts less than 50 × 10(9)/L with the Advia (Siemens, Tarrytown, NY), Sysmex (Mundelein, IL), and Abbott (Santa Clara, CA) analyzers.

Results: All methods showed a positive bias, especially at less than 20 × 10(9)/L and less than 10 × 10(9)/L. Undertransfusion risk ranged from 9.1% to 43.3 % in the groups below 20 × 10(9)/L and below 10 × 10(9)/L, respectively. For patients with optical counts more than 10 × 10(9)/L and CD61 less than 10 × 10(9)/L, 64.5% were transfused within 24 hours of the reported count, while 35.5% were transfused in more than 24 hours, after a subsequent optical platelet count of 10 × 10(9)/L or less was reported.

Conclusions: Although optical and impedance methods were shown to be falsely increased in severely thrombocytopenic samples, further studies are needed to determine if more accurate methods would be clinically useful.