Impact of Platelet Count on Bleeding in the Setting of Anti-Platelet Therapy

Anti-platelet therapy (APT) is used for secondary prevention of thrombosis. The most commonly prescribed anti-platelet drugs are aspirin and P2Y12 inhibitors, including clopidogrel, prasugrel and ticagrelor. Dual anti-platelet therapy (DAPT) consisting of aspirin and a P2Y12 inhibitor is often used in the first 1-12 months after an initial thrombotic event and has a greater anti-thrombotic effect than single agents, but is also associated with a higher risk of bleeding. Due to this risk of hemorrhage, the appropriate use of DAPT in patients requiring percutaneous coronary intervention (PCI) with baseline or periprocedural thrombocytopenia remains unclear. To study the impact of thrombocytopenia on bleeding with APT, we used intravital imaging in a murine hemostasis model and adoptive platelet transfer to generate mice with specific platelet counts with or without platelet inhibition. To generate experimental mice, we used transgenic mice in which platelets express a chimeric GPIb receptor with the extracellular domain replaced with a domain of the human IL-4R (hIL-4R/GPIb-Tg). Endogenous platelets were depleted by injection of anti-hIL-4R antibody, and the recipient mice were then transfused with wild-type (WT) platelets from donor mice treated, or not, with single or dual APT (aspirin 20 mg/kg; clopidogrel 25 mg/kg) to achieve specific platelet counts ranging from 50,000 to 400,000 platelets/μL. We also compared these mice with WT mice (with normal platelet counts, ~1,200,000 platelets/μL) treated with APT. Platelet inhibition was confirmed prior to performing in vivo experiments. Hemostasis was determined by intravital imaging in our saphenous vein laser injury model, in which a 50 μm injury was induced by laser ablation. Real-time top-down epifluorescence imaging was used to determine time to initial hemostasis, rebleeding events, and platelet and fibrin accumulation. In each mouse, 3-5 injuries were induced at different sites and each injury was visualized for 10 minutes. Following real-time imaging, spinning disk confocal Z-stacks of platelet plugs were obtained for 3D reconstruction to compare platelet plug volume. In untreated WT mice, hemostasis was achieved in ~20 seconds. In WT mice treated with DAPT, initial hemostasis was often rapidly achieved but this was followed by significant rebleeding events. Paradoxically, platelet accumulation was increased in WT + DAPT mice due to extravascular accumulation of platelets which occurred during bleeding. However, in plugs that stabilized, plug volume was reduced in WT + DAPT mice. In hIL-4R/GPIb-Tg mice with reduced platelet counts, untreated platelets were able to form a stable hemostatic plug even at 50,000/μL, although time to hemostasis was slightly prolonged. However, as platelet counts decreased in mice with DAPT-treated platelets, initial hemostasis became more prolonged and many injuries never achieved initial hemostasis. These results suggest that DAPT may not be safe in the setting of severe thrombocytopenia.