Airway remodelling in asthma remains topical in research and clinical practice, and a very poorly understood abnormality despite a large body of published work. It is important both from a pathophysiological and patient management point of view because there is evidence that airway remodelling relates to airway hyperresponsiveness,1 to accelerated lung function decline resulting in fixed airway obstruction2 and severe disease defined by suboptimal symptom control despite high doses of inhaled corticosteroids (ICS).3
One of the major difficulties in studying and understanding airway remodelling in relation to disease mechanisms and clinical expression is the inherent problems in its measurement. It is arguable whether histopathology should be the gold standard for airway remodelling. Nevertheless, measurements from post-mortem studies show that remodelling is present in airways of all sizes and that the severity of remodelling in large airways is indeed reflective of that in the small cartilaginous airways.4 Airway remodelling is worse in patients who have fatal asthma compared with the lungs of patients who have milder stable asthma.5 However, bronchial biopsy is very difficult to undertake in longitudinal studies and therefore is not suitable for intervention studies in large cohorts.
High-resolution computed tomography (HRCT) has been developed as a non-invasive measure of airway remodelling with the parameters of airway wall thickness and area as surrogate indices of remodelling. Although the radiation exposure associated with HRCT imaging has previously limited its use in longitudinal studies, modern hardware has allowed increasingly reduced radiation exposure to the point where it may be possible to acquire four or five scans over the course of a treatment period and still remain within radiation limits that are recognized to have a negligible risk of malignancy. However, there are many characteristics of HRCT measurements that need to be considered when interpreting the results of such studies.6 Factors that may affect the measurement include the degree of lung hyperinflation, scanning technique such as the strength of the beam (mAs) and table speed (pitch), image analysis technique, effect of bronchodilators, the number and site of airways sampled, and standardization against subject size to name a few. Nevertheless, HRCT measurements have provided much insight into the importance of airway remodelling in asthma. Airway wall thickness correlates with asthma duration and severity7 and with forced expiratory volume in 1 s,8 and is reduced by ICS treatment.9 Interestingly, results of one study suggest that airway wall thickness may be protective against airway hyperresponsiveness10 possibly by stiffening of the wall, which limits airway narrowing. Although the resolution of HRCT imaging is limited such that only large airways can be accurately studied, these measurements have been shown to be representative of small airway remodelling in radiological/pathological correlative studies.11
The study by Hoshino et al. in this edition of the Journal is one in a long series of HRCT studies of airway wall thickness in asthma. The authors report results from a study of steroid naïve asthmatic subjects in whom a greater decrease in airway wall thickness with budesonide/formoterol treatment was measured by HRCT compared with budesonide treatment alone for 6 months. The subjects' asthma was uncontrolled in that they had high symptom scores, high sputum eosinophil counts and moderate-to-severe airway hyperresponsiveness. The type of patient group studied raises the questions of what was the underlying change that is represented by a reduction in airway wall area and whether this cohort would be more steroid-responsive than subjects who have more severe asthma (incompletely controlled on regular ICS treatment). It is likely that it represented a reduction in oedema and steroid-responsive inflammation given the high eosinophil count in sputum and high exhaled nitric oxide levels, as well as the correlation between change in sputum eosinophils and change in wall area. Whether there was also a reduction in parameters such as basement membrane thickness, matrix proteins, and elastin and collagen structure is unknown. In a biopsy study of mild-to-moderate asthma, the inflammatory parameters improved after 3 months of treatment, but reticular basement membrane thickness decreased only after 12 months of ICS treatment.1 Wall thickness has been previously shown to decrease with ICS treatment alone, and the relatively small numbers in the ICS treatment group in the Hoshino study explains the apparent lack of improvement in wall area. The main findings from this study suggest that the combination of ICS and long-acting beta agonist (LABA) may be more effective in reducing remodelling than inhaled steroid alone. This is a very intriguing finding that needs independent confirmation ideally with other disease markers that are relevant to remodelling, such as by bronchial biopsy, airway distensibility measurements12 or biochemical markers such as fibulin13 and tumstatin.14 The mechanism for this potential synergistic effect is unknown, but the authors discuss several possibilities, including whether deposition of both drugs, at the same location and at the same time, could be crucial in influencing cellular processes that reduce or reverse remodelling. If these findings could be confirmed, it would also be clinically relevant to determine whether a similar improvement in airway remodelling measurements can be attained with ICS/LABA combination at a lower total steroid dose compared with ICS alone.
HRCT resolution is such that only the large airways can be studied, and the effect of ICS/LABA combination treatment as opposed to ICS alone on remodelling in the small airways remains unknown. There is increasing appreciation of the importance of small airways disease, particularly because it relates to refractory asthma,15 to risk of exacerbations16 and to airway hyperresponsiveness.17,18 Furthermore, small airway dysfunction relates to symptom control, that is, it may directly influence the day-to-day symptoms experienced by patients.19 Hence, it would seem important that treating the small airways effectively will be beneficial, particularly in severe disease.15,20 However, measuring airway remodelling in the small airways in vivo has not been possible despite excellent tools such as HRCT and non-invasive measurements of lung function that do not involve radiation exposure.12,21
What does this all mean for our patients? At present, there is insufficient evidence to suggest that combination ICS/LABA should be given to prevent long-term remodelling or to reduce remodelling in the short term. Studies of longer duration would be useful where the effects of treatment on inflammation can be separated from effects on presumably remodelling. There is still a lack of long-term studies of airway remodelling in asthma in terms of treatment, identification of predictors and of basic mechanisms. A comparison of the response of remodelling to combination ICS/LABA treatment in refractory asthma to milder disease also deserves attention. Given the importance of small airways, the benefit of combination ICS/LABA treatment over ICS treatment alone also requires future study. For now, the hope is that combination ICS/LABA treatment for those in whom it is clinically indicated, confers benefit in terms of better treatment of remodelling. At least until we know how to measure remodelling better, what are the mechanisms and where in the lung it matters.