Magdolna Nagy, Sepanta Fazaeli, René van Oerle, Hugo Ten Cate, Marcel Schemmann, John Sherry, Gillian Kelleher, Henri M H Spronk
Tromb J. 2021 May 4;19(1):29.
PMID: 33947405
Manuscript in collaboration with 2M Engineering (Valkenswaard).
https://www.2mel.nl/nl/projects/bloedplaatjes-trombocyten-teller/
Introduction: Platelet count can be altered in various diseases and treatments and measuring it may provide better insight into the expected outcome. So far, quantification of platelet count is done within laboratory conditions by using established hematology analyzers, whereas a point-of-care device could be used for this purpose outside of the clinical laboratories.
Aim: Our aim was to assess the closeness of agreement between a newly developed point-of-care PC100 platelet counter and two reference methods (Sysmex® XP-300, Sysmex® XN-9000) in measuring platelet counts in whole blood and platelet-rich-plasma (PRP).
Method: Whole blood was obtained from 119 individuals, of which 74 were used to prepare PRP samples. Whole blood platelet count was measured by the two reference methods and the PC100 platelet counter. PRP was prepared from the whole blood and platelet count was adjusted to the range of 250-3600 × 103/μl and measured with the PC100 platelet counter and Sysmex® XP-300.
Results: A median difference of – 1.35% and – 2.98% occurred in whole blood platelet count between the PC100 platelet counter and the Sysmex® XP-300 and Sysmex® XN-9000, respectively. A strong linear correlation (r ≥ 0.98) was seen in both cases and regression equations indicated neither a constant nor a proportional bias between the methods. Direct comparison of the two reference methods revealed a median difference of – 1.15% and a strongly linear relationship (r = 0.99). Platelet count in PRP resulted in a median difference of 1.42% between the PC100 platelet counter and the reference method, Sysmex® XP-300. While the difference between two methods increased with concentration of platelets in PRP, a strong linear relationship remained throughout the whole measuring interval indicated by the high correlation coefficient (r = 0.99). Assessment of the predicted bias at predefined platelet counts showed that the bias in platelet counts falls within the acceptance criterion for both whole blood and PRP measurements.
Conclusions: Our results show that the PC100 platelet counter can be used interchangeably with the reference methods for determining platelet counts.
Keywords: Hematology analyzer; Method comparison; Platelet counter; Point-of-care device.