Recurrent Thrombosis in Patients with Polycythemia Vera or Essential Thrombocythemia: Efficacy of Treatment in Preventing Rethrombosis in Different Clinical Settings.

Background: Thrombosis is a common cause of morbidity in polycythemia vera (PV) and essential thrombocythemia (ET). Prior thrombosis is a major risk factor for rethrombosis; yet optimal strategy for prevention of recurrence has been scarcely addressed.

Objective: To assess in PV or ET patients with prior thrombosis the rate of recurrence, the risk factors for recurrence, and the efficacy of antithrombotic or cytoreductive strategies in preventing rethrombosis.

Methods: We retrospectively investigated a cohort of 470 patients (M/F 227/243, PV/ET 235/235, median age at diagnosis 62 yrs, range 19–88) with a first arterial (n=328, 69.8%) or venous thrombosis (n=142, 30.2%). First thrombosis was cerebrovascular disease (CVD) in 184 cases, acute coronary syndrome (ACS) in 102, venous thromboembolism of limbs (VTE) in 102, and venous thrombosis of unusual sites (un-VTE) in 40. Microcirculatory events were not computed. Multivariate time-dependent models were developed having first recurrence as dependent variable and putative predictive or protective factors as covariates.

Results: 158 patients (33.6%) had one or more recurrences during a total time from the first thrombosis of 2,821 pt-yrs (median 5.3 yrs); the overall incidence of events was 7.54% pt-yrs (4.60% arterial and 2.94% venous recurrences). Major bleeding occurred in 25 patients (84% receiving antithrombotic prophylaxis) with an incidence of 0.88% pt-yrs. The cumulative incidence of recurrence at 10 years after the first event was 64.7% (95%CI 49.1–80.3) in VTE patients, 51.9% (95%CI 37.8–66.0) in ACS, 46.6% (95%CI 35.1–58.1) in CVD, and 30.2% (95%CI 8.8–51.6) in un-VTE. In VTE patients analysis restricted to venous recurrences showed at 10 years a rate of 49.4% (95%CI 32.8–66.1). Sex, diagnosis (PV or ET), and presence of vascular risk factors did not affect the overall probability of recurrence. Age >60 years predicted recurrence (multivariate hazard ratio 1.76, 95%CI 1.24–2.49); arterial or venous first thrombosis predicted either arterial (HR 4.43, 95%CI 1.22–16.07) or venous recurrence (HR 8.48, 95%CI 1.62–44.48), respectively. Cytoreductive treatment halved the probability of all types of recurrence in the overall cohort (HR 0.54, 95%CI 0.38–0.76). In the overall model long-term antiplatelet or anticoagulant treatment did not result as independent protective factors. However, analysis of patient groups showed that a significant efficacy in preventing rethrombosis was achieved in VTE only by long-term oral anticoagulation (HR 0.42, 95%CI 0.19–0.91), in ACS by cytoreduction (HR 0.37, 95%CI 0.17–0.78), and in CVD by antiplatelet agents (HR 0.34, 95% CI 0.16–0.69).

Conclusions: In PV or ET the rate of recurrent VTE seems higher than the usually observed in patients with unprovoked first VTE. Moreover our findings suggest that strategies for preventing rethrombosis and based on cytoreduction or antithrombotic prophylaxis should be tailored according to the type of first clinical event. Further studies are needed to confirm such hypothesis.