Perioperative monitoring of platelet function in paediatric cardiac surgery by thromboelastometry, or platelet aggregometry?

Background Impaired platelet function increases the risk of bleeding complications in cardiac surgery. Reliable assessment of platelet function can improve treatment. We investigated whether thromboelastometry detects clinically significant preoperative, perioperative, and postoperative adenosine diphosphate (ADP)-dependent platelet dysfunction in paediatric cardiac surgery patients. Methods Fifty-seven children were included in a single-centre prospective observational study. Clot formation (modified rotational thromboelastometry with heparinase, HEPTEM) and platelet aggregation (multiple electrode aggregometry) were analysed at five time points before, during, and after surgery. The accuracy of thromboelastometric indices of platelet function [maximal clot firmness (MCF) and clot formation time (CFT)] to detect ADP-dependent platelet dysfunction (defined as ADP-induced aggregation ≤30 units) was calculated with receiver operating characteristics analysis, which also identified optimal cut-off levels. Positive and negative predictive values for the identified cut-off levels (CFT≥166 s; MCF≤43 mm) to detect platelet function were determined. Results The MCF and CFT were highly accurate in predicting platelet dysfunction during cardiopulmonary bypass [CPB; area under the aggregation curve 0.89 (95% confidence interval 0.80–0.97) and 0.86 (0.77–0.96), respectively] but not immediately after CPB [0.64 (0.48–0.79) and 0.67 (0.52–0.82), respectively] or on arrival at the intensive care unit [0.53 (0.37–0.69) and 0.60 (0.44–0.77), respectively]. The positive and negative predictive values were acceptable during CPB (87 and 67%, respectively, for MCF≤43 mm; 80 and 100% for CFT≥166 s) but markedly lower after surgery. Conclusion In paediatric cardiac surgery, thromboelastometry has acceptable ability to detect ADP-dependent platelet dysfunction during, but not after, CPB.