Abstract
BACKGROUND: Guidelines recommend treatment with direct oral anticoagulants (DOACs) over unfractionated heparin (UFH) or low-molecular-weight heparin (LMWH) among ambulatory patients, including patients in the emergency department, with pulmonary embolism (PE) at low risk for mortality; however, recent evidence suggests that patients with low-risk PE are usually admitted to the hospital from the emergency department rather than discharged on DOACs. RESEARCH QUESTION: Among hospitalized patients with low- and low-intermediate-risk PE, how do patterns in anticoagulation and outcomes vary between institutions? STUDY DESIGN AND METHODS: This multicenter retrospective cohort study used the PINC AI enhanced administrative database (2016-2022). Eligible adult patients were admitted to a general ward, had an International Classification of Diseases, 10th Revision, diagnosis code for PE present on admission, were initiated on anticoagulation (UFH, LMWH/fondaparinux, or a DOAC) on but not before day 1, and had troponin and brain natriuretic peptide below the upper limit of normal. Initial anticoagulation practices were summarized overall and by hospital. Regression modeling was used to determine associations between initial anticoagulation and median length of stay. RESULTS: Among 2,369 eligible patients, the percentage of patients initiated on UFH was 54%, initiated on LMWH/fondaparinux was 41%, and initiated on DOACs was 4%. Anticoagulation with DOACs decreased median length of stay by 0.62 days (95% CI, -1.04 to -0.20) compared with those who initially received UFH. INTERPRETATION: Our results showed that hospitalized patients with low- and low-intermediate-risk PE generally do not receive initial DOACs. However, use of initial DOAC therapy was associated with shorter hospital length of stay compared with other initial anticoagulation strategies.