High resolution computed tomographic assessment of airway wall thickness in chronic asthma: reproducibility and relationship with lung function and severity

Open Access
Publisher Website
Google Scholar
Abstract
Background: In some patients chronic asthma results in irreversible airflow obstruction. High resolution computed tomography (HRCT) has been advocated for assessing the structural changes in the asthmatic lung and permits investigation of the relationships between airway wall thickening and clinical parameters in this condition. Methods: High resolution CT scanning was performed in 49 optimally controlled asthmatic patients and measurements of total airway and lumen diameter were made by two independent radiologists using electronic callipers. Wall area as % total airway cross sectional area (WA%) and wall thickness to airway diameter ratio (T/D) were calculated for all airways clearly visualised with a transverse diameter of more than 1.5 mm, with a mean value derived for each patient. Intra- and inter-observer variability was assessed for scope of agreement in a subgroup of patients. Measurements were related to optimum forced expiratory volume in 1 second (FEV1), forced mid expiratory flow, carbon monoxide gas transfer, two scores of asthma severity, airway inflammation as assessed with induced sputum, and exhaled nitric oxide. Results: Neither observer produced a statistically significant difference between measurements performed on two occasions but there was a significant difference between observers (limits of agreement –2.6 to 6.8 for WA%, prS=0.29 and 0.30, respectively, for ATS score), and an inverse association with gas transfer coefficient was observed (rS=–0.43 for WA% and rS=–0.41 for T/D). No association was identified with FEV1 or airway inflammation. Conclusions: The airway wall is thickened in more severe asthma and is associated with gas transfer coefficient. This thickening does not relate directly to irreversible airflow obstruction as measured with FEV1.

This publication has 35 references indexed in Scilit: