Atsushi Matsunashi, Tomoki Maetani, Tomohiro Handa, Kohei Ikezoe, Ryuji Uozumi, Naoya Tanabe, Ryo Sakamoto, Michihiro Uyama, Kiminobu Tanizawa, Toyohiro Hirai
Background Idiopathic pleuroparenchymal fibroelastosis (IPPFE) is radiologically characterized by dense subpleural consolidation with traction bronchiectasis and upper lobe volume loss. However, automated quantitative evaluation of the bronchi or lung parenchyma using high-resolution computed tomography (HRCT) images of IPPFE has not been previously performed. Methods Consecutive patients with IPPFE who underwent chest HRCT were enrolled. For comparison, 120 patients with idiopathic pulmonary fibrosis (IPF) were also enrolled. Volume fraction of consolidation, reticulation, honeycombing, and intrapulmonary bronchi to total lung volume were automatically quantified using AI-based quantitative computed tomography image analysis software (AIQCT). Moreover, after airway tree segmentation, CT indices for central airway morphology, including bronchial bifurcation angle, tracheal tortuosity, extrapulmonary airway lumen area, and the wall and lumen area of the segmental/subsegmental intrapulmonary airways (WA intra , LA intra ) were calculated, and their associations with survival were assessed. Longitudinal changes in central airway and AIQCT parameters were also analyzed in patients with follow-up CT data. Results HRCT images from 71 patients with IPPFE were analyzed. IPPFE showed greater consolidation and larger intrapulmonary bronchial volumes in the upper lobes compared with IPF, and both consolidation and intrapulmonary bronchial volume fractions increased significantly over time. Patients with IPPFE had a larger bronchial bifurcation angle, greater tracheal tortuosity, and higher WA intra and LA intra values. In a multivariable model, the bronchial volume fraction was independently associated with survival (hazard ratio, 1.50; 95% confidence interval, 1.14–1.99). Conclusions Quantitative chest CT analysis in IPPFE revealed that intrapulmonary bronchial volume was associated with prognosis and increased with disease progression.