Adjusted D-dimer to Exclude Concurrent Pulmonary Embolism in Patients with COVID-19

OBJECTIVES: To determine if elevated D-dimers in Severe Acute Respiratory Syndrome Coronavirus-2 (COVID-19) positive patients were predictive of concurrent pulmonary embolism (PE), and what, if any, adjusted D-dimer cutoff would be predictive of PE in COVID-19 patients. BACKGROUND: COVID-19 has led to over 500,000 death the United States. COVID-19 has been linked to a prothrombotic state, increasing the affected individual's propensity to form blood clots which is reflected in elevated D-dimers levels. While elevated D-dimers have been associated with 28-day mortality (making them useful in tracking the disease), it is difficult, in the acute setting, to determine if an elevated D-dimer is due to COVID-19, a pulmonary embolism (PE), or both conditions simultaneously. METHODS: This was a retrospective chart review of emergency department patients who underwent a computed tomography pulmonary angiogram (CTPA) who had documented COVID-19 infections between March and December of 2020. Charts were abstracted for COVID-19 polymerase chain reaction (PCR) results, D-dimers levels, and CTPA results. Kruskal-Wallis rank sum one-way analysis of variance and sensitivity analyses were completed via the software R. RESULTS: Of 346 COVID-19 positive patients by nasal swab PCR, 24 (6.9%) had a PE on CTPA. The average D-dimer in the subset was 22,768 ng/dL vs 3,620 ng/dL in COVID-19 positive but PE negative patients (p=.034). As a screening test for concurrent PE, a D-dimer cut off of 1,000ng/dL or, for those over 50, [age in years]*20 ng/dL, had a sensitivity of 96%, specificity of 48% and negative predictive value (NPV) of 99.4%. A cut off of 1,000ng/dL had a sensitivity of 100%, specificity of 44% and NPV of 100%. CONCLUSION: Emergency department patients with COVID-19 and a PE have a significantly higher elevation in their D-dimer when compared to COVID-19 positive patients without a PE. A flat cut off of 1000ng/dL or a two tiered cut off of 1,000ng/dL for those <50 years old and [age in years]*20 ng/dL for those >50 years old proved to be reasonable screening tests for concurrent PE in COVID-19 positive patients.