Bleeding risk of terminally ill patients hospitalized in palliative care units: the RHESO study

Background The value of primary thromboprophylaxis in patients admitted to palliative care units is debatable. Moreover, the risk of bleeding in these patients is unknown. Objectives Our primary aim was to assess the bleeding risk of patients in a real-world practice setting of hospital palliative care. Our secondary aim was to determine the incidence of symptomatic deep vein thrombosis and to identify risk factors for bleeding. Patients/Methods In this prospective, observational study in 22 French palliative care units, 1199 patients (median age, 71 years; male, 45.5%), admitted for the first time to a palliative care unit for advanced cancer or pulmonary, cardiac or neurologic disease were included. The primary outcome was adjudicated clinically relevant bleeding (i.e. a composite of major and clinically relevant non-major bleeding) at 3 months. The secondary outcome was symptomatic deep vein thrombosis. Results The most common reason for palliative care was cancer (90.7%). By 3 months, 1087 patients (91.3%) had died and 116 patients had presented at least one episode of clinically relevant bleeding (fatal in 23 patients). Taking into account the competing risk of death, the cumulative incidence of clinically relevant bleeding was 9.8% (95% confidence interval [CI], 8.3-11.6). Deep vein thrombosis occurred in six patients (cumulative incidence, 0.5%; 95% CI, 0.2-1.1). Cancer, recent bleeding, antithrombotic prophylaxis and antiplatelet therapy were independently associated with clinically relevant bleeding at 3 months. Conclusions Decisions regarding the use of thromboprophylaxis in palliative care patients should take into account the high risk of bleeding in these patients.