Dual-energy CT angiography reveals high prevalence of perfusion defects unrelated to pulmonary embolism in COVID-19 lesions

Insights Imaging. 2021 Feb 17;12(1):24. doi: 10.1186/s13244-021-00972-0.

Abstract

Background: Lung perfusion defects (PDs) have been described in COVID-19 using dual-energy computed tomography pulmonary angiography (DE-CTPA). We assessed the prevalence and characteristics of PDs in COVID-19 patients with suspected pulmonary embolism (PE) and negative CTPA.

Methods: This retrospective study included COVID-19 and non-COVID-19 pneumonia groups of patients with DE-CTPA negative for PE. Two radiologists rated the presence of PD within the lung opacities and analyzed the type of lung opacities and PD pattern (i.e. homogeneous or heterogeneous). The clinical, biological, radiological characteristics including time from first symptoms and admission to DE-CTPA, oxygen requirements, CRP, D-dimer levels, duration of hospital admission and death were compared within the COVID-19 group between patients with (PD +) or without PD (PD-).

Results: 67 COVID-19 and 79 non-COVID-19 patients were included. PDs were more frequent in the COVID-19 than in the non-COVID-19 group (59.7% and 26.6% respectively, p < 0.001). Patterns of PDs were different, with COVID-19 patients exhibiting heterogenous PDs (38/40, 95%) whereas non-COVID-19 patients showed mostly homogeneous perfusion defects (7/21 heterogeneous PDs, 33%), p < 0.001. In COVID-19 patients, most consolidations (9/10, 90%) exhibited PDs while less than a third of consolidations (19/67, 28%) had PDs in non-COVID-19 patients. D-dimer, oxygen levels and outcome were similar between COVID-19 PD + and PD- patients; however, time between admission and DE-CTPA was longer in PD + patients (median [IQR], 1 [0-7] and 0 [0-2]; p = 0.045).

Conclusion: Unlike in bacterial pneumonia, heterogeneous PDs within lung opacities are a frequent feature of COVID-19 pneumonia in PE-suspected patients.

Keywords: COVID-19 pneumonia; Computed tomography angiography; Dual-energy computed tomography.