Outcomes in the Secondary Prevention of Cerebral Venous Thrombosis in Cancer Only Population

Abstract BACKGROUND AND PURPOSE: There are scant data on the outcomes of cerebral venous thrombosis (CVT) in the cancer only population. Current anticoagulation guidelines exist for the general population but the effect of these strategies in cancer subgroups are not known. We examined the clinical outcomes such as recanalization, thrombosis recurrence and major bleeding during cerebral venous thrombosis treatment in an exclusively cancer population. METHODS: A retrospective cohort study of cancer only patients with cerebral venous thrombosis identified through an institutional data warehouse at MD Anderson Cancer Center between January 2002 and June 2017. Patients with other indications for long term anticoagulation, chronic cerebrovascular thrombosis or less than 12 month follow up were excluded from the primary analysis. The primary end point was thrombus recanalization. Secondary end points were major bleeding and recurrence of cerebral or venous thrombosis. Demographic data, cancer diagnosis and cancer-therapy was also obtained at the time of initial CVT for further analysis. Chi-square test or Fisher exact test were used to assess the association between categorical variables. RESULTS: The population comprised predominantly male adults (55.6%) with median age 54.5 years, IQR [41.5 to 62.4]. The underlying malignancies were non-hematologic in 64.4% of cases. Anticoagulation was used in the treatment of 73.3% of cases, predominantly low molecular weight heparin (LMWH), and a median duration of 7.9 months, IQR [3.1 to 30.8]. Partial and complete recanalization were achieved in 33.3% and 17.8% of cases, respectively. The CVT recurrence rate at 12 months was 15.6% and 31% of patients suffered major bleeding. Most of the major bleeding events (71.1%) occurred outside the brain. Anticoagulation therapy with LWMH showed a trend to a higher rate of thrombus recanalization (p=0.196) without an increment in the overall bleeding events. We observed a lower rate of major bleeding events in patients treated with LWMH when compared with oral anticoagulation (5/14 versus 18/31, respectively; p=0.049). CONCLUSIONS: Our findings suggest that LMWH might represent an effective and safe strategy for the management of CVT in cancer patients. The findings are consistent with results from large randomized trials for outcomes of cancer associated venous thrombosis in other anatomical locations. The impact of thrombus recanalization in recurrent thrombosis and bleeding outcomes deserves further exploration. Limitations to our study are the retrospective and non-randomized design for the adjudication of treatment agents. Disclosures No relevant conflicts of interest to declare.