Clin Lab. 2025 Oct 1;71(10). doi: 10.7754/Clin.Lab.2025.250331.
ABSTRACT
BACKGROUND: Ethylenediaminetetraacetic acid-induced pseudothrombocytopenia (EDTA-PTCP) is a well-documented in vitro artifact characterized by platelet clumping, leading to spuriously low platelet counts despite normal platelet quantity and function. This phenomenon can result in inappropriate clinical management, including unnecessary platelet transfusions. To address this issue, we propose a post-analytical correction strategy that integrates kanamycin treatment with event-based analysis in the PLT-F channel of the Sysmex XN-20 automated hematology analyzer to improve platelet count accuracy.
METHODS: The platelet surface markers CD41a and CD61 were evaluated using flow cytometry to compare their expression on single platelets and platelet clumps. Forward and side scatter characteristics, along with dot plot distributions, were analyzed to determine the extent of overlap between large platelet clumps and the neutrophil gating region. Additionally, blood smears were examined before and after kanamycin treatment to evaluate the degree of platelet clump dissociation. A regression model utilizing Gating-4 event counts from the PLT-F scatter-gram was developed to estimate corrected platelet counts. The diagnostic performance of Gating-4 as a surrogate marker for platelet clumps was assessed using receiver operating characteristic (ROC) curve analysis.
RESULTS: Flow cytometry revealed moderate CD41a and CD61 expression on single platelets, while platelet clumps exhibited markedly increased expression of these markers. Larger platelet clumps demonstrated increased forward scatter intensities, overlapping with the neutrophil population in dot plots. Blood smear analysis confirmed complete platelet clump dissociation in 87 out of 96 samples following kanamycin treatment, whereas residual clumping persisted in 9 cases, leading to unreliable platelet counts. Regression analysis demonstrated that Gating-4 event counts were strong predictors of corrected platelet counts (R² = 0.705, p < 0.001), with high statistical significance (t = 12.377, p < 0.001). A Gating-4 event cutoff of 138.5 optimally distinguished clump presence, with an area under the ROC curve of 0.969 (p < 0.001).
CONCLUSIONS: This study introduces a regression model utilizing Gating-4 events from the PLT-F scattergram to improve platelet count estimation in clumped samples. The model demonstrated high accuracy in moderate clumping cases; however, additional correction was required for severe clumping. To enhance accuracy, we propose a stepwise approach that incorporates platelet clump size assessment and kanamycin intervention.
PMID:41078198 | DOI:10.7754/Clin.Lab.2025.250331
