Platelet function analyzer (PFA-100) as a useful tool for the prediction of transfusion requirements during aortic valve replacement

Abstract

Background: Shear stress-induced hemostatic abnormalities, particularly loss of the hemostatically most competent, highest molecular weight von Willebrand factor multimers, are common in patients with aortic valve stenosis. Although controversially discussed, these hemostatic defects might be associated with an increased risk of bleeding during aortic valve replacement. Since the determination of closure times with a platelet function analyzer is sensitive for the detection of defects of primary hemostasis including shear stress-induced von Willebrand factor abnormalities, this study was performed to evaluate a method to predict intraoperative transfusion requirements in this setting.

Methods: Fifty patients (mean age ± SD: 68 ± 9 years, range 40-85 years) admitted for aortic valve replacement were enrolled in the study. Closure times of epinephrine/collagen and ADP/collagen cartridges were determined with a platelet function analyzer in the absence of antiplatelet agents. Results were compared to those obtained in healthy individuals without medication. The probability that a patient would require a transfusion of packed red cells (RBC) and fresh frozen plasma (FFP) was calculated for each obtained closure time using a multiple regression model.

Results: Compared to controls, patients undergoing aortic valve replacement had a significantly higher incidence of prolonged closure in the platelet function analyzer. The prolonged closure time of both epinephrine/collagen and ADP/collagen cartridges was significantly correlated with intraoperative transfusion of RBC, but not FFP.

Conclusions: In patients undergoing aortic valve replacement, prolongation of closure times as determined by a platelet function analyzer is frequently observed, indicating the presence of shear stress-induced defects of primary hemostasis. Since the prolongation of closure times is significantly correlated to the probability of intraoperative transfusion, this method might offer a significant contribution to the preoperative risk stratification of patients.