Platelet clump assessment using the Cellavision peripherical blood application – do we need manual microscopy?

Abstract Objectives This study aims to investigate the frequency, detection, and distribution of platelet clumps, assess the performance of automated digital microscopy (Cellavision) for detecting platelet clumps and explore strategies to optimize detection efficiency. Methods We conducted a retrospective analysis of 987,586 hemograms to evaluate the frequency of platelet clumps, and a study of 246 hemograms with platelet clumps for manual and digital microscopic reviews of blood smears. We investigated the locations and sizes of these clumps along the smear, and evaluated the detection capacity of the Cellavision system. Results Platelet clumps were found in 0.29 % of cases, with a higher incidence in pediatric and elderly populations. Platelet clumps were more frequent and larger at the smear periphery. Cellavision achieved 93 % sensitivity when combining the leukocyte and red blood cell observation fields. The strategy of reviewing only selected cases (platelet count <50 × 10 9 /L or history of clumps) detected 97 % of platelet clumps, while reducing manual reviews by fourfold. Conclusions Automated digital microscopy is an effective tool for detecting platelet clumps, but it requires manual review in specific cases. Expanding image acquisition to the feather edge could further improve detection. A combined approach maximizes efficiency and ensures diagnostic accuracy, particularly in critical cases with low platelet counts.