Abdominal vein thrombosis in essential thrombocythemia: prevalence, clinical correlates, and prognostic implications

Among 460 consecutive patients with essential thrombocythemia (ET) seen at our institution, 19 cases (4%) of abdominal vein thrombosis (AVT) were documented either at (n = 9) or after (n = 10) diagnosis. Women (P = 0.03) and the young (P = 0.002) were preferentially affected. Accordingly, clinical comparisons were performed among three groups of female patients: those with AVT (group A; n = 17), a control group without AVT but closely matched to group A in terms of age and year of diagnosis (group B; n = 34), and all female patients without AVT (group C; n = 288). As expected from the consequences of AVT-associated portal hypertension and anticoagulant therapy, patients in group A experienced significantly higher rates of hemorrhage, palpable splenomegaly, and anemia. Unexpectedly, however, compared with group B, group A displayed both a higher conversion rate into myelofibrosis/acute leukemia (P = 0.0008) and a shorter median survival (116 vs. 156 months; P = 0.0012). Multivariable analysis including all female patients with ET identified AVT, along with advanced age, leukocytosis, and tobacco use, as an independent risk factor for inferior survival. Groups A, B, and C did not differ in either JAK2(V617F) mutational frequency or incidence of non-abdominal thrombosis. We conclude that AVT in ET is a marker of aggressive disease biology.