Thromboprophylaxis after discharge improves clinical outcomes for certain high-risk patients hospitalized with COVID-19 (MICHELLE)

Reference

Ramacciotti E, Agati LB, Calderaro D, et al, for the MICHELLE investigators. Rivaroxaban versus no anticoagulation for post-discharge thromboprophylaxis after hospitalisation for COVID-19 (MICHELLE): an open-label, multicentre, randomised, controlled trial. Lancet 2022;399(10319):50-59.

Synopsis

In this industry-funded study from Brazil, the authors investigated the use of extended thromboprophylaxis in patients who were hospitalized with COVID-19 and were at increased risk for venous thromboembolism (VTE) based on a validated risk score (IMPROVE VTE) and D-dimer levels. Patients with an IMPROVE VTE score of at least 4 and patients with a score of 2 to 3 plus an increased D-dimer level were included. Patients with active cancer, gastrointestinal ulcers, chronic kidney disease, and recent bleeding and those using dual antiplatelet therapy were excluded. All patients received standard prophylactic doses of enoxaparin, unfractionated heparin, or fondaparinux while in the hospital. Upon discharge, those in the treatment group (n = 159) received rivaroxaban 10 mg daily for 35 days while those in the control group (n = 159) received no anticoagulation. The 2 groups were similar at baseline: mean age was 57 years, 40% were women, median hospital stay was 8 days, and approximately half in each group required a stay in the intensive care unit. Although the study protocol called for computed tomography angiogram and lower limb ultrasound to assess for VTE in all patients at day 35, the percentage of patients who were actually evaluated was higher in the treatment group. Despite this, more clots were found in the control group. For the composite outcome of symptomatic and asymptomatic VTE, symptomatic arterial thromboembolism, and cardiovascular death at day 35, the treatment group fared better (3.1% vs 9.4%; relative risk [RR] 0.33; 95% CI 0.13 - 0.90; P = .03). This was driven primarily by a lower rate of symptomatic and fatal VTE in the treatment group (0.6% vs 5.0%; RR 0.13; 0.02 - 0.99). There were no major bleeding events in either group.