Abstract
Abstract Background In today's era of automation driven by Artificial Intelligence, even the most efficient automated hematology analyzer may struggle to accurately determine platelet count in patients with low platelet count and microcytosis. The Sysmex XN-series analyzer (Sysmex, Kobe, Japan) uses optical density, fluorescence, and electronic impedance methods for platelet counting. This study aimed to evaluate the accuracy of optical, fluorescence, and electrical impedance methods in estimating precise platelet counts in thrombocytopenic and microcytic patient samples. Methods A cross-sectional study was conducted in the Department of Laboratory Medicine at AIIMS, New Delhi, for 30 days consecutively. A total of 200 blood samples collected, of which 100 samples were of thrombocytopenia (<100x10^9/L) and 100 samples were of microcytosis (Mean corpuscular volume <76 femtoliters). MCV<76 fl was further categorized into 2 groups- MCV 60-70 fl and MCV 71-76 fl accordingly. Males and females were categorized and analyzed distinctly for thrombocytopenia and microcytosis. Statistical analysis was performed using SPSS 16.0 for Windows (SPSS Inc., Chicago, IL, USA) and the Bland Altman plot was analyzed. Results Of the 100 patients with thrombocytopenia (platelet<1 lakh/L) females were predominantly 54% while males were 46% and the median age was 29 years (0.02-70). Of the 100 patients with low MCV (<76 fl) females were 53%, while males were 47% and the median age was 29 years (0.3-69). The findings were compared against the standard manual platelet count method in peripheral smears verified by two expert pathologists. The Bland Altman plot for Platelet <1 lakh/L and MCV <76 fl were compared with standard manual platelet count method and the following findings were obtained i) In female with PLT- <1 lakh/L, PLT-O showed least bias (9.8) followed by PLT-F (10.4) and PLT-I (12.1) ii) In males with PLT-<1 lakh/L, PLT-O showed least bias(14.1) followed by PLT-F(16.5) and PLT-I(17.0) iii) For MCV 61-70fl, PLT-O showed the least bias (4.4) followed by PLT-F (6.7) and PLT-I (28.5) iv)For MCV 71-76fl, PLT-F showed least bias (-0.2) followed by PLT-O (1.5) and PLT-I(31.2). Conclusions Platelet count by optical method (PLT-O) detects platelet count more accurately with minimum bias when compared with PLT-F and PLT-I in patients with thrombocytopenia. In patients with microcytosis, PLT-I is less reliable in comparison to PLT-O and PLT-F. A limitation of the automated hematology analyzer is the inability to detect platelet clumps which is appreciated in peripheral smears.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call