Applied Chest Imaging Laboratory

Clinical Implications of Bronchiectasis in Smokers

Grant support: 1R01HL133137

PI: Dr. Alejandro Diaz

Project Information:

Chronic obstructive pulmonary disease (COPD) affects approximately 26 million people in the United States. It is estimated that 16%-56% of these people have bronchiectasis (pathologic dilation of the airways). The presence of bronchiectasis in COPD has been linked to longer hospital stays, prolonged recovery from exacerbations, and increased mortality. The visual assessment of bronchiectasis on high-resolution computed tomography (CT) scans is the standard diagnostic method. In this investigation the visual assessment will be used to further develop, validate, refine, and apply an automated imaging tool to objectively assess the burden of bronchiectasis in smokers using baseline and follow-up chest CT scans from the COPDGene Study (a large cohort of smokers with and without COPD). In Aim 1, we will perform a visual scoring and marking of bronchiectasis on a subset of baseline and 5-yr follow- up CT scans. We will then assess the associations between baseline bronchiectasis and acute respiratory disease episodes, health-related quality of life, and death. In aim 2, we will perform point-based (at a single slice) objective measurements of bronchial dilation and airway morphology in bronchiectatic and non- bronchiectatic airway sections labeled for Aim 1 to identify CT features that distinguish bronchiectasis. We will build receiver operating characteristics (ROC) curves to determine the predictive ability of these objective measurements to detect bronchiectasis. Finally, in Aim 3 we will fully develop, validate, and refine an automated imaging tool with the features found to be discriminatory to detect bronchiectasis in Aims 1 and 2. We will then determine the burden of bronchiectasis using this tool on all baseline and follow-up CT scans (estimated N= 16,300) and assess its association with clinically relevant disease outcomes including acute respiratory disease episodes, quality of life, and mortality. We will also use ROC analysis as well as the integrated discrimination index (IDI) and net reclassification index (NRI) measures to assess the contribution of our imaging tool to predict those outcomes.

People Involved:

Dr. Alejandro A. Diaz, Dr. Jose Orejas, Dr. Raúl San José Estépar, Rubén San José Estépar