Pulmonary embolism risk factors for intensive care unit anticoagulated COVID-19 patients undergoing computed tomography angiography

Gonzalo Patricio Briceño-Mayorga, Rocío Gutiérrez, Celine Sotomayor, Matías Ebner, Felipe Allende, Rodrigo Assar

To assess pulmonary embolism incidence, its relationship with D-dimer levels and other possible associated factors in addition to anticoagulation and contrast medium adverse effects. A retrospective observational cohort study at a Chilean public hospital was performed. Intensive care unit mechanically ventilated COVID-19 patients older than 18 years old between March and June 2020 were included. All patients received heparin thromboprophylaxis, which was increased to the anticoagulation dose with D-dimer greater than 3µg/mL. A total of 127 patients were followed up, of whom 73 underwent pulmonary computed tomography angiography (mean age, 54 ± 12 years; 49 men). Sixty-two of the 73 patients (84.9%) received full anticoagulation before computed tomography angiography. In addition, 18 of the 73 patients had pulmonary embolism (24.7%). When comparing patients with and without pulmonary embolism, no significant differences were observed in age, sex, obesity, smoking, Wells and revised Geneva scores, D-dimer or mortality. Anticoagulant use was similar in both groups. Days from the start of anticoagulation until computed tomography angiography were significantly lower in the pulmonary embolism group (p = 0.002). Three patients presented post contrast-acute kidney injury (4.1%), and one patient had major bleeding. Despite anticoagulation, one in four COVID-19 patients connected to mechanical ventilation and evaluated with pulmonary computed tomography angiography had pulmonary embolism. With a longer the delay in performing computed tomography angiography once empirical anticoagulation was started, significantly less pulmonary embolism was identified.