ABSTRACT
INTRODUCTION: Patients with moderate thrombocytopenia and comorbidities requiring anticoagulation are currently sub-optimally treated because of bleeding concerns. Guidance on anticoagulating such patients is currently lacking because of limited data on safety and efficacy of anticoagulation in such patients. METHODS: This retrospective study compared the incidence of bleeding and thrombosis in a cohort of warfarinized patients with sustained platelet counts below 100×109/L against a cohort with normal platelet counts (>140×109/L). Primary outcomes of safety and efficacy were determined by incidence rate ratios (IRR) of bleeding and thrombotic events. International normalized ratio (INR) and platelet counts during adverse events in thrombocytopenic arm were secondary outcomes. RESULTS: 137 thrombocytopenic patients (104,985 patient-exposure days) were compared against 939 normal patients (715,193 patient-exposure days). IRR of minor, major bleeding and thrombosis among thrombocytopenic patients were 3.03 (95% CI: 1.57-5.60), 1.48 (95% CI: 0.44-3.98), and 0.807 (95% CI: 0.09-3.43) respectively. Median INR and platelet count readings during minor and major bleeds were 3.60 (IQR: 2.70-4.12) and 3.12 (IQR: 2.82-4.22), and 99×109/L (IQR: 77.0-147.0×109/L) and 115×109/L (IQR: 107.5-169.5×109/L) respectively. CONCLUSION: Warfarinized thrombocytopenic patients are at higher risk of minor bleeding complications with a higher tendency for major bleeding but derive similar benefits against thrombotic events compared to normal patients. Bleeding events are associated with higher INRs. A narrow INR target with an upper limit below 2.5 together with closer anticoagulation monitoring may improve safety of patients.