Association between quantitative computed tomography metrics and pulmonary function in patients with post-COVID-19 pulmonary fibrosis: a single-center cross-sectional study

Post-COVID-19 pulmonary fibrosis is an important long-term complication of severe SARS-CoV-2 infection. Quantitative computed tomography (QCT) provides observer-independent measurement of parenchymal abnormality. However, the association between structural lung abnormalities and functional impairment remains insufficiently characterized. We quantified the cross-sectional association between a CT-derived density metric and pulmonary function, exercise capacity and dyspnea. Single-center cross-sectional study of 39 adults assessed approximately one year after severe COVID-19 pneumonia; each underwent one chest CT and one pulmonary function assessment, with no serial imaging or serial physiology. QCT was performed in 3D Slicer v5.0.3 with the Lung CT Analyzer extension. Collapse percentage (collapse%) was defined as the volume of intrapulmonary tissue with attenuation above − 400 HU, after exclusion of vessels (> 0 HU), divided by total lung volume (TLV) — the non-aerated compartment of the established CT aeration model. Spirometry followed the 2019 ATS/ERS standards; exercise capacity was assessed by 6-minute walk distance (6MWD) with Borg dyspnea scoring. Correlation coefficients are reported with 95% confidence intervals (CI). Diffusing capacity for carbon monoxide (DLCO) was unavailable. Mean age was 56.7 ± 11.4 years and 19 of 39 participants (48.7%) were male. Mean TLV was 4070.2 ± 1529.5 mL and mean collapse% was 5.4 ± 5.2%. Collapse% was inversely associated with 6MWD ( r = − 0.59, 95% CI − 0.76 to − 0.34; p < 0.0001), with FVC % predicted ( r = − 0.46, − 0.68 to − 0.17; p < 0.001) and, to a lesser extent, with FEV1% predicted ( r = − 0.32, − 0.58 to − 0.00; p < 0.05), corresponding to 35%, 21% and 10% of shared variance. Collapse% was higher in participants with a Borg score of 2–4 than in those scoring 0–1 ( p = 0.03). Quantitative CT provided a structural measure of residual lung involvement, while pulmonary function tests measured the associated physiological impairment; the two were only weakly to moderately associated, indicating that they are complementary rather than interchangeable. As a cross-sectional, single-center study of severe-disease survivors without gas transfer measurement or serial imaging, these findings are hypothesis-generating and require longitudinal validation before quantitative CT can be used to inform diagnosis, monitoring, or management.