Inspiratory and Expiratory Computed Tomography Imaging Clusters Reflect Functional Characteristics in Chronic Obstructive Pulmonary Disease

Abstract

Purpose: Disease probability measure (DPM) is a useful voxel-wise imaging assessment of gas-trapping and emphysematous lesions in patients with chronic obstructive pulmonary disease (COPD). To elucidate the progression of COPD, we performed a cluster analysis using the following DPM parameters: normal (DPM_Normal), gas-trapping (DPM_GasTrap), and emphysematous lesions (DPM_Emph). Our findings revealed the characteristics of each cluster and the 3-year disease progression using imaging parameters.

Patients and methods: Inspiratory and expiratory chest computed tomography (CT) images of 131 patients with COPD were examined, of which 84 were followed up for 3 years. The percentage of low attenuation volume (LAV%) and the square root of the wall area of a hypothetical airway with an internal perimeter of 10 mm (√Aaw at Pi10) were quantitatively measured using inspiratory chest CT. A hierarchical cluster analysis was performed using the DPM parameters at baseline. Five clusters were named according to the dominant DPM parameters: normal (NL), normal-GasTrap (NL-GT), GasTrap (GT), GasTrap-Emphysema (GT-EM), and Emphysema (EM).

Results: Women were predominantly diagnosed with GT. Forced expiratory volume in 1 s gradually decreased in the following order: NL, NL-GT, GT, GT-EM, and EM. DPM_Emph correlated well with LAV%. Four clusters other than NL showed significantly higher values of √Aaw at Pi10 than NL; however, no significant differences were observed among them. In all clusters, DPM_Emph increased after 3 years. DPM_Normal only increased in the GT cluster.

Conclusion: Clusters using DPM parameters may reflect the characteristics of COPD and help understand the pathophysiology of the disease.

Keywords: COPD; DPM; cluster analysis; computed tomography; emphysema; gas-trapping.