Abstract
Introduction
Patients hospitalized with COVID-19 have an increased incidence of venous thromboembolism (VTE) and arterial thromboembolism (ATE) events. These thrombotic events increase readmission and mortality rate in COVID-19 survivors who are recently discharged from hospital. To lower the risk of VTE, a short course of post-discharge anticoagulation at either prophylactic or therapeutic dose has been variably prescribed among different facilities to COVID-19 patients. This practice, however, is challenged by less than 3% incidence of VTE in unselected patients. The net clinical benefit of extended thromboprophylaxis beyond hospitalization remains unclear.
Methods
We conducted a retrospective multicenter observational study of 5613 hospitalized COVID-19 patients. After applying the inclusion and exclusion criteria, 2838 patients were included in statistical analysis. Patients were excluded if they had negative SARS-CoV-2 PCR, remained hospitalized at the time of analysis, or were discharged to hospice service. The first symptomatic ATE and VTE events up to 90 days after patients' discharge from their index admission for COVID-19 were identified using ICD-10 codes, and subsequently validated by chart review. The predictors for post-discharge VTE were identified using multivariate logistic regression. The average protective effect of anticoagulation was assessed using inverse propensity score weighting.
Results
The mean age (SD) of our cohort was 63.4 (16.7) years old and 47.6% were male. Black, white and other races were 38.9%, 50.7% and 10.3%, respectively. Thirty-six (1.3%) patients developed post-discharge VTE events that require hospital visits (18 deep vein thromboses, 16 pulmonary embolisms and 2 portal vein thromboses). Fifteen (0.5%) patients developed post-discharge ATE events (14 acute coronary syndromes and 1 transient ischemic attack). The incidence of VTE decreased with time (p <.001) with the median event time of 16 days (Figure 1). The incidence of ATE was unchanged with time (p =.369) with the median event time of 37 days (Figure 1). Patients who had a history of VTE (OR=3.24, 95% CI 1.34-7.86), peak D-dimer >3 µg/mL (OR=3.76, 95% CI 1.86-7.57), and predischarge C-reactive protein >10 mg/dL (OR=3.02, 95% CI 1.45-6.29) were at a high risk of developing VTE after hospital discharge (Figure 2). A short course of prophylactic or therapeutic anticoagulation after hospital discharge markedly reduced VTE (OR=0, 95% CI 0-0, p<.001, and OR=0.176, 95% CI 0.04-0.75, p=.02, respectively).
Conclusions
Although extended thromboprophylaxis in unselected COVID-19 patients is not recommended, post-discharge anticoagulation may be considered in high-risk patients who have a history of VTE, peak D-dimer >3 µg/mL and predischarge C-reactive protein >10 mg/dL if their bleeding risk is low. Our study has provided the first evidence to guide the selection of hospitalized COVID-19 patients who may benefit from post-discharge anticoagulation.
Kaatz: Gilead: Consultancy; Novartis: Consultancy; CSL Behring: Consultancy; Bristol Myer Squibb: Consultancy, Research Funding; Alexion: Consultancy; Pfizer: Consultancy; Janssen: Consultancy, Research Funding; Osmosis Research: Research Funding.
