Airway Remodelling in Asthma and COPD: Findings, Similarities, and Differences Using Quantitative CT

Abstract

Airway remodelling is a well-established feature in asthma and chronic obstructive lung disease (COPD), secondary to chronic airway inflammation. The structural changes found on pathological examination of remodelled airway wall have been shown to display similarities but also differences. Computed tomography (CT) is today a remarkable tool to assess airway wall morphology in vivo since submillimetric acquisitions over the whole lung volume could be obtained allowing 3D evaluation. Recently, CT-derived indices extracted from CT images have been described and are thought to assess airway remodelling. This may help understand the complex mechanism underlying the remodelling process, which is still not fully understood. This paper summarizes the various methods described to quantify airway remodelling in asthma and COPD using CT, and similarities and differences between both diseases will be emphasized.

Figure 1

Top left. Thin section CT image, perpendicular to the fourth generation of the right apical bronchus from a random patient. Top right. Sample of manual airway wall thickness extraction. Green doublehead arrow indicates external diameter D, and red doublehead arrow shows internal diameter L. Bottom left. Sample of manual delineation of the external layer (green line) and internal layer (red line), using the mouse on the CT scan image seen in top left. Bottom right. Automatic quantification of airway wall dimension (white arrow) using a Laplacian-of-Gaussian algorithm.

Figure 2

Right. Thin-section CT image perpendicular to the third generation of the right segmental apical bronchus, from a random patient. Red line indicates the external wall contour, and green line the internal layer. Left. Theoric single intensity curve (blue line) representing voxel attenuation variation along the blue arrow seen in right image.

Figure 3

Segmented axial thin section image of right and left lung, from which the mediastinum and lung contours were removed. Using the density technique, dark blue areas indicate voxels between −1000 HU and −950 HU.