Leukocytosis as an Independent Risk Factor for Thrombosis in Polycythemia Vera.

The antithrombotic strategy of Polycythemia Vera is usually modulated according to age and thrombotic history but there is a clear need for a more accurate stratification of the vascular risk in the individual patient. The present study exploited the large database collected within the cohort study of European Collaboration on Low-dose Aspirin in Polycythemia Vera (ECLAP) to investigate how haematological variables as well as cardiovascular risk factors may be associated with the thrombotic risk. We used prospectively ascertained information among 1,638 polycythemic patients recruited in the ECLAP observational study and followed for a maximum of 5 years (mean [SD] 2.7 [1.3] years). Information on clinical history, treatment, hematological parameters, and cardiovascular risk factor were assessed at baseline and updated at 1, 2, 3, 4, and 5 years. Multivariable, time-dependent analyses were fitted to assess the association between white blood cell count and thrombotic events during follow-up. Over this period there were 205 thromboses in 169 patients. The time-dependent analysis adjusted for potential confounders confirmed that age and previous thrombotic event are independent predictive factors. As compared to patients with white blood cell count below 10,000x10⁶/L, those with white blood cell count above 15,000x10⁶/L showed a significant increase of the risk of thrombotic events (Hazard Ratio 1.71; 95% confidence interval 1.10 to 2.65; P=0.0171) mainly deriving from an increased risk of myocardial infarction (Hazard Ratio 2.84; 95% confidence interval 1.25 to 6.46; P=0.0127). Patients with white blood cell count above 15,000x10⁶/L had an increased, though statistically non significant, risk of venous thromboembolism (Hazard Ratio 1.81; 95% confidence interval 0.84 to 3.89; P=0.1272). Subjects with hypertension had a mild, non significant increase of the risk of arterial thrombosis while smoke was significantly associated with an increased risk of arterial thrombotic events (Hazard Ratio 1.90; 95% confidence interval 1.15 to 3.14; P=0.0120) and was non significantly associated with the risk of of peripheral artery disease (Hazard Ratio 2.86; 95% confidence interval 0.92 to 8.86; P=0.0692) and of stroke/TIA (Hazard Ratio 1.89; 95% confidence interval 0.89 to 4.01; P=0.0978). Hypercholesterolemia and diabetes had a limited prevalence among our patients and could not be evaluated as to their impact on the vascular risk.

The association of thrombotic events with leukocytosis provided novel information on epidemiology as well as on possible pathogenetic mechanisms of thrombosis in Polycythemia vera. Leukocytosis has to be considered for risk stratification and for choosing and targeting treatment options particularly in patients with previous myocardial infarction or venous thromboembolism. Finally, the high prevalence of smoke and its impact on thrombotic risk call for an effective lifestyle intervention in polycythemic patients.