Predictors of venous thromboembolic events in hospitalized patients with COVID-19.

Abstract

COVID-19 is associated with an increased risk of venous thromboembolism (VTE) in hospitalized patients. Although prior studies have attempted to identify predictors of VTE, restricted sample size and use of administrative claims data have limited such analyses. We utilized data from hospitalized patients in the CORONA-VTE Network, a United States multicenter registry of adult patients with PCR-confirmed COVID-19 (N = 3,844). The primary outcome was time-to-first event for a composite of adjudicated pulmonary embolism or deep vein thrombosis during 90-day follow-up. The candidate variables were selected by a priori clinical consensus. We conducted cause-specific Cox regression analysis adjusted for the selected variables for each imputed dataset and pooled the estimated HRs for reporting (p < 0.05 for significance). VTE occurred in 206 patients, with a cumulative incidence of 5.3% at 90 days. The covariates associated with increased risk of VTE were history of VTE (HR: 1.71; 95% CI: 1.11-2.63), corticosteroid therapy (HR: 1.76; 95% CI: 1.32-2.33) and known thrombophilia (HR: 3.56; 95% CI: 1.54-8.21) while therapeutic anticoagulation at baseline (HR: 0.42; 95% CI: 0.26-0.69), antecedent use of statins (HR: 0.67; 95% CI: 0.50-0.90), and prophylactic anticoagulation during hospitalization (HR: 0.52; 95% CI: 0.38-0.71) were associated with reduced risk of VTE. While prior VTE, corticosteroid therapy, and known thrombophilia were associated with an increased risk of VTE, prescriptions of prophylactic and therapeutic anticoagulation, and statins were associated with a decreased risk. Once externally validated, these findings may inform risk assessment in hospitalized patients with COVID-19.