Warfarin reduces circulating white blood cell count : post-hoc analysis of two randomized trials

Abstract

Warfarin is a widely used vitamin K antagonist (VKA) with known pleiotropic effects beyond anticoagulation. Preclinical and case-control evidence suggests that warfarin may affect hematopoiesis, but longitudinal human evidence is lacking. To explore this potential effect, we conducted a post-hoc analysis of participants in the Hokusai-VTE and ENGAGE AF-TIMI 48 trials, which randomized patients to warfarin or the direct oral anticoagulant edoxaban with routine laboratory testing at predefined follow-up visits. We analyzed changes in total circulating white blood cells (WBC) and subpopulations (lymphocytes, monocytes, granulocytes) using linear regression and mixed-effects models, adjusting for baseline counts, age, sex, and time. Among 23,618 patients enrolled in the two phase 3 trials, warfarin use was associated with a modest but statistically significant reduction in WBC count (−2.3 %, 95 % CI −2.9 % to −1.7 %) and granulocyte count (−3.6 %, 95 % CI −4.5 % to −2.7 %) compared with edoxaban, while lymphocyte and monocyte counts did not differ. The associations remained consistent across multiple sensitivity analyses. No increase in clinically relevant granulocytopenia was observed. In the context of two large randomized trials, these findings support a subtle hematologic effect of warfarin, particularly in granulocytes, that aligns with preclinical findings and warrants further investigation into the long-term impact of VKAs on hematopoiesis.