152 Correlation of Point of Care Lung Ultrasound and CT Scan Findings in Patients with COVID-19

Abstract

Chest CT scan (CT) is often considered the gold-standard imaging modality to evaluate pulmonary pathology, and thus is used to assess patients with COVID-19. While CT offers higher resolution images, Point-of-care ultrasound (POCUS) has the advantages of being rapid, low cost, low radiation exposure, and offers the ability for monitoring real-time disease progression. As such, POCUS has also been used to assess patients with COVID-19, and characteristic POCUS findings of COVID-19 are described. In the present study, we compare chest CT to lung ultrasound findings in patients with COVID-19 and examine consistency in pathological findings between the two imaging modalities. 125 patients presenting to an urban emergency department in Tehran, Iran with symptoms concerning for COVID-19 were prospectively enrolled. Participants underwent lung POCUS following a 12-zone protocol assessing each zone for pleural line irregularities, alveolar interstitial syndrome (eg, B-lines), and presence of consolidations including subpleural consolidations (SCs). Patients also received chest CT read by a radiologist evaluating for ground glass opacity, crazy paving patterns, or consolidations. For POCUS and CT, each zone was scored using a 4-point measure, then aggregated total lung involvement scores were calculated for each patient and imaging modality. Descriptive statistics were performed to asses consistently between POCUS and CT findings. POCUS findings overall corresponded well with abnormalities seen on CT, without a significant difference in lung involvement scores between the modalities. On CT, COVID-19 patients showed greater incidence of crazy paving in the AI, PS, PI, AX, PLAPS, CS, and CI distributions (p adj= .00293, 0, .000600, .000533, .00272, .0004) and effusion in CI (p adj= .0216), and on POCUS patients had increased B-lines in the AS, AX, and PLAPS distributions (p adj= .0086, .0012, .0024 respectively), increased pleural thickening in all lung regions (AS, AI, PS, PI, AX, PLAPS; p adj= .0182, .0014, .0375, .0328, .0003, 0), and SCs in AS, AX, and PLAPS (p adj= .0312, .0398, .0324). Both CT and POCUS demonstrated more right-sided findings as a whole, though sidedness of findings was not statistically significant. Both CT and POCUS demonstrated differences in finding densities between lung regions (for CT- ARDS: p adj= 1.00e+ 0; consolidation: p adj= 1.04e- 1; Crazy Paving: p adj= 1.92e-16; effusion: p adj= 8 4.94e- 8; GGO: p adj= 8.73e- 2; interstitial: p ajd= 68e- 6; POCUS- atelectasis: p adj= 1.92e- 7; B-lines: p adj= 6.77e -7; consolidation: p adj= 0.00119; effusion: p adj= 7.20e- 9; pleural thickening: p adj= 7.20e -9; SCs: p adj= 4.39e -5) with the highest concentration of positive findings in the PLAPS region on both modalities. Given comparable findings in the presence and distribution of abnormalities between POCUS and chest CT, POCUS may be a viable alternative to chest CT for diagnosis and risk stratification in patients with suspected COVID-19.