The interaction between leukocytosis and other risk factors for thrombosis in essential thrombocythemia

Three recent papers have reported an association between leukocytosis and thrombosis in patients with myeloproliferative disorders; a leukocyte count of more than 15 × 10⁹/L, as opposed to 10 × 10⁹/L or fewer, was associated with myocardial infarction in polycythemia vera,¹  and that of either 15 × 10⁹/L² or more or more than 8.7 × 10⁹/L³ with thrombosis in general in essential thrombocythemia (ET). In the current study, we used the database from a previously published cohort of 605 patients with ET⁴  in order to clarify associations at diagnosis and prognostic relevance for subsequent thrombotic events, using both of the aforementioned leukocyte targets in ET. Arterial and venous thromboses were analyzed separately and in the context of other significant risk factors. We also looked into the interaction between leukocyte count and current smoking; previous epidemiologic studies have suggested that smoking might account for the association between leukocyte count and stroke in the general population.⁵ 

The study population included 605 patients (median age, 57 years; 66% women) with ET whose presenting clinical and laboratory features have previously been communicated in a paper that addressed risk factors for survival and leukemic transformation.⁴  In the current study, variables at diagnosis were evaluated for significant association with thrombosis occurring either at diagnosis or during follow-up (median 84 months). A leukocyte count of 15 × 10⁹/L or more or more than 8.7 × 10⁹/L was registered in 86 (14%) and 348 patients (58%) at diagnosis, respectively. Smoking history at diagnosis was documented in 128 patients (21%) and was more likely to be encountered in patients with leukocyte counts that exceeded either one of the aforementioned leukocyte targets (P = .002 and P < .001, respectively). Numbers of patients with arterial thrombosis history before diagnosis, at diagnosis, and during follow-up were 27, 104, and 133, respectively. The corresponding numbers for venous events were 36, 35, and 45. Information on JAK2V617F mutational status was available in 187 patients.

Table 1 outlines significance levels (ie, P values) during univariate and multivariable analysis of parameters at diagnosis. For arterial thrombosis at diagnosis, independent associations were demonstrated only for higher hemoglobin level and leukocytosis (using either 1 of the aforementioned leukocyte targets). A leukocyte count of 15 × 10⁹/L or more was also associated with venous thrombosis at diagnosis. However, unlike the observations from another recent study,³  leukocyte count at presentation did not predict subsequent vascular events. However, accurate interpretation of the relationship between leukocyte count at diagnosis and subsequent thrombosis is confounded by treatment effect on both variables, and prospective studies are needed to clarify the issue further. The current study also underscores the importance of accounting for previous history of vascular events during risk factor assessment for thrombosis and discloses the specific pattern of the association in this regard.

Correspondence: Ayalew Tefferi, Mayo Clinic, 200 First Street SW, Rochester, MN 55905; e-mail: tefferi.ayalew@mayo.edu.