Aims & Objectives:
To evaluate verification criteria for reflex ordering of platelet scans for automated platelet counts generated by Beckman Coulter LH-780 Hematology Analyzer.
Materials & Methods: The study uses automated platelet counts <100x103/µL generated by the two Beckman Coulter LH-780 analyzers as part of evaluation and monitoring of thrombocytopenia. The cases are grouped into 4 based on platelet flags generated by the analyzer as (1)positive for giant platelets, (2)positive for platelet clumps (CLP), (3)positive for both giant platelets & clumps (GP+CLP) and (4)without platelet flags. Corresponding smears were reviewed to determine if the automated platelet counts were acceptable and to note down the positive findings such as presence of giant platelets, clumps, fibrin strands, microclots. Positive and Negative predictive values (PPV and NPV) were calculated. The automated platelet count was accepted and the same reported if the manual platelet count was within 10% of the automated count for counts >/=40x103/µL and within 20% for counts <40x103/µL.
Results: Of the total 1004 smears studied for thrombocytopenia, the category CLP had the least PPV of 0.27, followed by categories GP with 0.74 and GP+CLP with 0.92. The group with no platelet flags comprised of 645 samples. In these 645 samples, 13 samples had unacceptable counts - 12 smears showed giant platelets of which 8 were cases with initial presentation, 2 with a platelet count <20x103/µL and 2 cases with a positive smear finding noted on previous smear examination. One sample had no smear findings, but a repeat sample was requested in view of delta check failure which had unacceptable platelet count.
Conclusion: Based on the findings of (a) PPV of 0.74 and 0.92 respectively for the groups with GP and GP+CLP (b) NPV of 0.98 for the group with no flags, one could also understandably exclude all the platelet count <100 × 103/μL as a criteria for a reflex order of a platelet scan and instead limit it to only those with platelet flags; and for those negative for flags, reflex smears may be done for the initial presentation, counts with <20x103/µL, a delta check failure or a positive smear finding on a previous smear examination.
With the revised policy, we expect a significant reduction in the number of platelet scans performed daily in our laboratory.