Prognostic implications of regional lung impairment evaluation in quantitative computed tomography imaging of COVID-19

Abstract

Aim: To enhance the understanding of COVID-19 regional lung damage pattern by analyzing the organ in subregions, beyond the typical lobe segmentation. Materials and Methods: This study used semiautomatic computed tomography (CT) imaging segmentation and quantification to investigate regional lung impairments in patients with COVID-19. Each lung was divided into 12 regions, and the anatomical impairments obtained from the CT image (emphysema, ground glass opacity, and collapsed tissue) were quantified. Then, the results for every region were correlated with clinical outcomes. This research encompassed 333 individuals, both COVID positive (n = 190) and COVID negative (n = 143), whose medical reports were checked for the need for ventilatory support and outcome (cure or deceased). Results: Findings indicate a strong association between the extent of lung damage and COVID-19 diagnosis, the level of ventilatory assistance required, and patient survival rates. Notably, the medial posterior lung region exhibited increased opacities and collapse in COVID-positive patients (p < 0.05), particularly those requiring invasive ventilation or who succumbed to the illness. Conclusion: The results expand the knowledge of COVID-19 regional impact beyond typical lobe segmentation and indicate that COVID-19 impairments in the lungs are localized. The most affected region identified was the medial posterior of both right and left lungs. Early detection of quantifiable lung damage can serve as a valuable prognostic tool, helping to pinpoint patients at heightened risk of severe complications or mortality.