Journal of Thrombosis and Circulation: Open Access

Abstrait

Rethinking D-dimer’s Role in the Diagnosis of Pulmonary Thromboembolism in Patients with COVID-19

Nieto Estrada Victor Hugo, Martinez Del Valle Anacaona, Valencia Moreno Albert Alexander, Molano Franco Daniel Leonardo, Calle Alvarez Elsy Sofia, Osorio Perdomo Daniela, Castaneda Ramirez Carlos Hernan, Gonzalez Zarate Natalia Andrea, Jaramillo Caceres Dayang Sulai, Bernal Salazar Tatiana Andrea

Background: D-dimer is a non-specific inflammatory marker that elevates in infections, thrombosis, and pregnancies. COVID-19 is a prothrombotic inflammatory disease, and it is common to find elevation of D-dimer in severe COVID-19 cases. The usefulness of D-dimer for the diagnosis of Pulmonary Thromboembolism (PE) in SARS CoV- 2 has not been determined.

Objective: To determine the operational characteristics of D-dimer as a diagnostic method for PE in patients with COVID-19 treated at a university hospital in Bogotá, Colombia.

Methods: Study of diagnostic tests that included data from patients with COVID-19 with suspected PE who were screened with the index test (D-dimer measured by turbidimetric immunoassay technique) and reference test (Angiotomography of pulmonary arteries).

Results: Among the 209 patients analyzed, the prevalence of PE was 14.4%, D-dimer levels were significantly higher in the group of PE cases (2888 ng/Dl vs. 1114 ng/Dl; p=0.002). 80% of PE cases were submassive and 53% segmental. The operating characteristics for the reference cut-off point of the technique (>499 ng/mL) was Sensitivity: 93.9%, Specificity: 8.9%, Positive predictive value: 14.7%, Negative predictive value: 8.9%, proportion of false positives: 91.1% with a Youden J- index of 0.02. The area under the curve was 0.684. The coordinates of the curve showed a Youden J- index of 0.367 for a value of 2.281 ng/mL (4.5 times the reference value), using this cut-off point, we obtained a sensitivity of 60%, a specificity of 76%, PPV of 30%, NPV of 92%, and a proportion of false Negatives of 40%.

Conclusion: D-dimer does not have appropriate characteristics to be used alone for the diagnosis PE in patients with severe COVID 19. It can be used as part of a rational diagnostic process, being just as specific as the patient’s signs and symptoms.