Hematocrit and the Risk of Recurrent Venous Thrombosis: A Prospective Cohort Study

BACKGROUND Venous thromboembolism (VTE) is a multicausal disease which recurs. Hematocrit is associated with a thrombotic risk. We aimed to investigate if hematocrit is associated with the recurrence risk. METHODS Patients with a first VTE were followed after anticoagulation. Patients with VTE provoked by a transient risk factor, natural inhibitor deficiency, lupus anticoagulant, homozygous or double heterozygous defects, cancer, or long-term antithrombotic treatment were excluded. The study endpoint was recurrent VTE. RESULTS 150 (23%) of 653 patients had recurrence. Only high hematocrit was significantly associated with recurrence risk [hazard ratio (HR) for 1% hematocrit increase with the third tertile 1.08; 95% CI 1.01-1.15]. No or only a weak association for hematocrits within the first and second tertile was seen (HR 1.03; 95% CI 0.97-1.09, and 1.07; 95% CI 1.00-1.13). Hematocrit was associated with recurrence risk only among women. After five years, the probability of recurrence was 9.9% (95% CI 3.7%-15.7%), 15.6% (95% CI 9.7%-21.2%) and 25.5% (95% CI 15.1%-34.6%) in women, and was 29.2% (95% CI 21.1%-36.5%), 30.1% (95% CI 24.1%-35.7%) and 30.8% (95% CI 22.0%-38.7%) in men for hematocrits in the first, second and third tertile, respectively. Men had a higher recurrence risk (1.9; 95% CI 1.1-2.7; p = 0.03), which dropped by 23.5% after adjustment for hematocrit. Hematocrit was not a significant mediator of the sex-difference in recurrence risk (p = 0.223). CONCLUSIONS High hematocrit is associated with the recurrence only in women. The different recurrence risk between men and women is possibly partly explained by hematocrit.