Using computed tomography to measure the site of airflow obstruction

The role of airway structure in chronic lung disease is central to our understanding of airflow obstruction.1,2 However, the actual analysis of airway structure has proven to be very complicated because this usually requires the removal of tissue from the body, which is both invasive and impractical for longitudinal studies. It is for these reasons that ‘non-invasive’ techniques such as CT have become such important tools in lung research. Over the last two decades there has been a surge in quantitative imaging studies involving CT, so much so that CT is now one of the most important techniques used in large-scale studies, including EXACTLE,3,4 ECLIPSE,5,6 COPDGene7,8 and SPIROMICS, to name a few. However, nothing is ever as straightforward as it seems. Although quantitative CT is a powerful tool, it does have some limitations and caveats when used in general clinical studies.9,10 One of the biggest limitations with analysis of airways is that there are almost as many different algorithms as there are centres using them. Furthermore, even though new multidetector CT scanners acquire images with near isotropic voxel resolution within a single breath hold, measurement of the airway wall dimensions of small airways is problematic because they are still at or below the resolution of the CT scanner. With this in mind, studies such as that by Nakano et al. show that medium sized airways may be a surrogate for small airway remodelling.11 However, other studies, including those by Hasagawa et al.12 and Coxson et al.13 showed that the airway wall dimensions of the smallest airways that were measureable, that is, fifth or sixth generation, demonstrated the strongest correlation with FEV1. In an attempt to move beyond these limitations to airway measurements, investigators have recently started to measure the density of the lung at expiration, as an indication of ‘gas trapping’ due to remodelling of the small airways. Recent data from these types of studies suggest that such measurements provide a good surrogate for small airway disease, particularly in asthmatic patients.14 In this issue of Respirology, Kurashima and colleagues have measured not only the airway wall dimensions, but also the internal diameter of the airways.15 Although it could be argued that the airway wall is the site of action in diseases such as COPD and asthma, it may be that lumen measurements will provide important information that is normally not analysed because of issues with the resolution of CT scanners. Furthermore, lumen measurements may be less susceptible to errors associated with the resolution of the CT scanner, due to the relative size of the lumen compared with that of the airway wall. This is a reasonable assumption, even though other data does show that the error in these measurements does increase substantially when smaller airways are measured.16 In the present study, the authors showed that the internal diameter of the third (subsegmental) and fourth generation airways differed in subjects with asthma or asthma plus emphysema compared with those with COPD or normal control subjects. The authors report that in subjects with COPD, the airway walls were thicker even though the lumen was not narrowed. However, in subjects with asthma, the airway walls were thicker and the lumen was also narrower.15 These are interesting findings because asthma is considered to be more a pure airway disease, whereas COPD is known to have both airway and parenchymal components. The fact that there are differences in the airway dimensions between these diseases is not surprising; however, it does yield valuable information about how these diseases differ. Other studies have shown that loss of airways does occur in patients with COPD,17 providing further evidence that the site and nature of the inflammatory process differs between these diseases. The authors conclude that the therapeutic targets for pure asthma and a mixture of asthma and emphysema should be different. This conclusion is what most researchers in the field want to believe; however, the data from this study shows that a detailed description of the lung anatomy is vital to understanding physiological impairment. Nevertheless, as mentioned previously, there are limitations to CT studies that should not be forgotten.9,10 First and foremost is the fact that airway measurement algorithms are still under development and there is no consensus on the most appropriate method for measuring the airways. Second, there is a lack of large cohort, multi-institutional, longitudinal CT studies, so it is hard to know whether all these airway measurement tools will be appropriate for these complex studies. Finally, there is increasing concern worldwide, about medical exposure to radiation and it must be remembered that CT does expose subjects to ionizing radiation. However, despite these caveats, studies such as that by Kurashima et al.15 demonstrate the importance of careful anatomical description of the lung. It is possible to measure the lung structure non-invasively and these measurements do provide valuable information about disease pathogenesis. Studies such as this, combined with studies that examine the number and branching pattern of the airways in disease, are likely to be important in the future. Well designed studies that minimize differences between centres, as well as radiation exposure, will provide useful information on lung disease and, hopefully, lead to the identification of therapeutic targets, as well as methods to assess the efficacy of such treatments. HOC is funded in part by the Dr Roberta R. Miller Fellowship in Thoracic Imaging from the British Columbia Lung Association and the Pittsburgh COPD SCCOR NIH 1P50 HL084948 and R01 HL085096 from the National Heart, Lung and Blood Institute, National Institutes of Health, Bethesda, MD to the University of Pittsburgh.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.