Predictors of Venous Thromboembolism in Severe COVID-19 Patients: A Retrospective Multi-Center Cohort Study

Background: Diagnosis of venous thromboembolism (VTE) in severe COVID-19 patients is challenging. D-dimer was reported to be the strongest independent factor that can predict VTE. Methods: This multicenter retrospective study describes the factors that might predict VTE diagnosis in 219 critically-ill COVID-19 patients. The primary outcome of interest was D-dimer cutoff values to predict VTE diagnosis in anticoagulation vs. no-anticoagulation cohorts. Findings: Of all included patients, 57 (26%) patients were diagnosed with VTE. At baseline, patients in the VTE cohort were more likely to be male and to age between 50-70 years. The optimum cutoff value for D-dimer to predict VTE in all patients was 2000 ng/ml (sensitivity of 86·0% and NPV of 91·3%). Patients who didn’t receive thrombo-prophylaxis, D-dimer value of less than 500 ng/ml can be used to rule-out DVT (sensitivity of 100·0%, NPV of 100·0%, and negative LR of 0). Patients who received thrombo-prophylaxis, D-dimer value of less than 1000 ng/ml can be used to rule-out DVT (sensitivity of 94·4%, NPV of 93·9%, and negative LR of 0·2). A total of 36 (16%) deaths occurred during the follow-up period. Elevated D-dimer (>2000 mg/ml, OR 1·7 [0·8-3·8]) was not associated with an increased risk of death. The use of intensified prophylactic anticoagulation was not significantly associated with a better survival outcome (OR 0·801, CI:0·67-1·748). Interpretation: In patients with severe COVID-19, D-dimer levels can be used to predict and rule-out VTE. The use of intensified prophylactic anticoagulation was not associated with better survival outcomes; this finding requires validation by randomized clinical trials. Funding Statement: This study was not funded by any source. Declaration of Interests: The authors declare that they have no conflict of interest. Ethics Approval Statement: This study was approved by the institutional review board of Cleveland Clinic.