P5588Early versus delayed oral anticoagulation in patients with acute pulmonary embolism: determinants and outcome

Abstract Background Whether early oral anticoagulant treatment is appropriate for patients with acute pulmonary embolism (PE) regardless of PE severity is undefined. The aim of this study in patients with acute PE at intermediate risk of death were: I) to assess the determinants for the use of early vs delayed vs no oral anticoagulants in patients with acute PE and II) to assess the association between timing of oral anticoagulation and in-hospital mortality. Methods Prospective cohorts of patients with acute PE at intermediate risk of death according to the European Society of Cardiology Guidelines 2014 were merged in a collaborative database. The initiation of oral anticoagulation was classified as early (≤3 days) or delayed (between day 3 and 10 from diagnosis). Patients treated with parenteral anticoagulants for longer than 10 days were also included. In-hospital death was the primary study outcome. Results Overall, 557 patients were included in the study, 23 received thrombolytic treatment during the hospital stay. The mean duration of parenteral anticoagulation was 7±8 days (5 median), 348 patients were initiated on a direct oral anticoagulant and 79 on a vitamin K antagonist during the hospital stay. Initiation of oral anticoagulants occurred early or delayed in 209 (37%) and 218 (39%) patients, respectively and never occurred during the first 30 days in 130 (23%). Intermediate-low risk patients more commonly received early and intermediate high delayed oral anticoagulation. Simplified PESI score of zero (OR 1.9, 95% CI 1.3–2.7) was independently associated with early oral anticoagulation; among sPESI components absence of cancer (OR 5.9, 95% CI 3.3–10) and heart rate <110 (OR 1.8, 95% CI 1.01–3.16) were independent predictors of early initiation of oral anticoagulants. The presence of both right ventricle dysfunction and injury was associated with delayed initiation of oral anticoagulants. The incidence of death was 5.5%. Death occurred in 32 patients and was not related to the duration of parenteral anticoagulation (OR 1.01 per day, 95% CI 0.98–1.06) nor to right ventricle dysfunction but to sPESI 1 (OR 3.32, 95% CI 1.14–9.66). These results were partially confirmed in the 435 intermediate risk patients without cancer (OR 1.03, 95% CI 0.99–1.08 for days of parenteral treatment; OR 4.17, 95% CI 0.95–18 for sPESI 1). Conclusion The clinical severity of PE and not the timing of initiation of oral anticoagulants are associated with in-hospital death in patients with intermediate risk PE. Randomized studies are needed to definitively assess the role of heparin lead-in in patients with PE at intermediate risk for death.