Acute Pulmonary Embolism in Patients with COVID-19 at CT Angiography and Relationship to d-Dimer Levels

Introduction

Reports of acute pulmonary embolism associated with coronavirus disease 2019 (COVID-19) have emerged in the literature. For example, Chen et al (1) described 25 pulmonary CT angiographic examinations from 1008 patients with COVID-19; 10 were positive for pulmonary embolism, mostly as segmental or subsegmental acute pulmonary embolism. In addition, d-dimer levels have been reported as elevated in patients with COVID-19 (2,3), with the suggestion of an independent association between the severity of the disease and the level of d-dimer (4). The purpose of this report is to describe the rate of pulmonary embolus in patients classified as having COVID-19 infection who underwent chest CT at a tertiary referral center.

Materials and Methods

Results

A flowchart of all patients with CT scans obtained from March 1 to March 31, 2020, is shown in Figure 1. During this period, 1696 patients underwent CT for suspicion or follow-up of COVID-19 infection. Dedicated pulmonary CT angiography was performed in 135 of the 1696 patients (8%); 25 additional patients had pulmonary arterial phase images included in chest-abdomen-pelvic CT (total, 160 patients). Of these 160 patients, 106 were classified as having COVID-19 infection (97 patients with RT-PCR and nine with positive CT findings and negative RT-PCR test). In these 106 patients, CT angiography was performed owing to suspicion of pulmonary embolus in 67 patients (63%) and other CT indications in 39 (37%).

Figure 1:

Download as PowerPoint

Thirty-two of the 106 patients (30% [95% confidence interval: 22%, 40%]) with COVID-19 who underwent pulmonary CT angiography were positive for acute pulmonary embolus (Fig 2); 74 were negative for pulmonary embolus at CT. Relevant clinical and biologic data are summarized in the Table.

Figure 2:

Download as PowerPoint

Clinical and Biologic Data for Patients Undergoing Pulmonary CT Angiography Classified as Having COVID-19 Infection

Patients with COVID-19 infection and pulmonary embolus had higher d-dimer levels than those without pulmonary embolus (median: 15 385 µg/L [interquartile range, 8180-22 590 µg/L] vs 1940 µg/L [interquartile range, 410-3470 µg/L], respectively; P < .001), were more likely to be in the intensive care unit (24 of 32 patients [75%] vs 24 of 74 patients [32%], P < .001), and were treated more often with low-molecular-weight heparin before CT angiography (25 of 32 patients [78%] vs 17 of 74 patients [23%], P < .001) (Table). In these patients with COVID-19 infection, a d-dimer level greater than 2660 µg/L had a sensitivity of 100% (32 of 32 patients; 95% confidence interval: 88%, 100%) and a specificity of 67% (49 of 74 patients; 95% CI: 52%, 79%) for pulmonary embolism at CT angiography.

Discussion

Our study demonstrated that of 106 pulmonary CT angiograms performed for patients with COVID-19 over a 1-month period in a tertiary care center, 32 of 106 patients (30%) had acute pulmonary embolus. This rate of pulmonary embolus is higher than usually encountered in critically ill patients without COVID-19 infection (1.3% [6]) or in patients treated in the emergency department (3%–10% [7]). In our patient population, a d-dimer threshold of 2660 µg/L enabled the detection of all patients with pulmonary embolus at chest CT. This threshold of 2660 µg/L is higher than previously reported median levels of 2400 µg/L (8) and 900 µg/L (2) and is higher than cut-off values used to exclude pulmonary embolus in patients not in the intensive care unit (9). High values of d-dimer could be related to a higher activation of blood coagulation in patients with COVID-19 secondary to a systemic inflammatory response syndrome, or as a direct consequence of the SARS-CoV-2 itself. Although a single-center retrospective report, our results of the potential for pulmonary embolism associated with COVID-19 infection may serve to alert the medical community to heighted vigilance of this complication.