EDTA-associated pseudothrombocytopenia: definition and real-world occurrence

Abstract Objectives To better characterize occurrence and extent of anticoagulant-associated pseudothrombocytopenia (PTCP) in the daily routine of a high-throughput clinical laboratory in order to draw conclusions on a more precise definition of this phenomenon. Methods Concomitant platelet counts in both EDTA and citrate whole blood (WB) performed in our laboratory over a period of four years and 9 months, were analyzed, calculating the correlation, as well as the absolute difference in the results obtained from both materials, cross-referencing these measures with automated flags for platelet aggregates and the results of the visual examination for platelet aggregates of peripheral blood smears. Results Platelet counts in both materials were strongly correlated (ρ=0.86; p<0.0001) but are on average significantly higher in EDTA WB than in citrate WB (median difference: 11 ± 14.8/nL, p<0.0001). This is in spite of numerous instances of EDTA-associated PTCP recorded in our data, where the opposite is the case. The automated flag for possible platelet aggregates was shown to be very unspecific, while a machine-learning algorithm suggested the difference in platelet counts between EDTA and citrate WB as a predictor of platelet aggregates. Conclusions EDTA-associated PTCP is a regular occurrence. Differences in platelet counts between EDTA and citrate WB appear to be a far better predictor of PTCP than automated flags. A clear and useful definition of PTCP is still missing, however, and cannot be derived from our data either, indicating the need for further research.