Structure and Function of Asthma Evaluated Using Pulmonary Imaging

Rachel L. Eddy, The University of Western Ontario

Abstract

Asthma has been understood to affect the airways in a spatially heterogeneous manner for over six decades. Computational models of the asthmatic lung have suggested that airway abnormalities are diffusely and randomly distributed throughout the lung, however these mechanisms have been challenging to measure in vivo using current clinical tools. Pulmonary structure and function are still clinically characterized by the forced expiratory volume in one-second (FEV₁) – a global measurement of airflow obstruction that is unable to capture the underlying regional heterogeneity that may be responsible for symptoms and disease worsening. In contrast, pulmonary magnetic resonance imaging (MRI) provides a way to visualize and quantify regional heterogeneity in vivo, and preliminary MRI studies in patients suggest that airway abnormalities in asthma are spatially persistent and not random. Despite these disruptive results, imaging has played a limited clinical role because the etiology of ventilation heterogeneity in asthma and its long-term pattern remain poorly understood. Accordingly, the objective of this thesis was to develop a deeper understanding of the pulmonary structure and function of asthma using functional MRI in conjunction with structural computed tomography (CT) and oscillometry, to provide a foundation for imaging to guide disease phenotyping, personalized treatment and prediction of disease worsening. We first evaluated the biomechanics of ventilation heterogeneity and showed that MRI and oscillometry explained biomechanical differences between asthma and other forms of airways disease. We then evaluated the long-term spatial and temporal nature of airway and ventilation abnormalities in patients with asthma. In nonidentical twins, we observed a spatially-matched CT airway and MRI ventilation abnormality that persisted for seven-years; we estimated the probability of an identical defect occurring in time and space to be 1 in 130,000. In unrelated asthmatics, ventilation defects were spatially-persistent over 6.5-years and uniquely predicted longitudinal bronchodilator reversibility. Finally, we investigated the entire CT airway tree and showed that airways were truncated in severe asthma related to thickened airway walls and worse MRI ventilation heterogeneity. Together, these results advance our understanding of asthma as a non-random disease and support the use of MRI ventilation to guide clinical phenotyping and treatment decisions.