Computed Tomography Lung Density Analysis: An Imaging Biomarker Predicting Physical Inactivity in Chronic Obstructive Pulmonary Disease: A Pilot Study

Physical inactivity correlates with poor prognosis in chronic obstructive pulmonary disease (COPD) and is suggested to be related to lung hyperinflation. We examined the association between physical activity and the expiratory to inspiratory (E/I) ratio of mean lung density (MLD), the imaging biomarker of resting lung hyperinflation. COPD patients (n = 41) and healthy controls (n = 12) underwent assessment of pulmonary function and physical activity with an accelerometer, as well as computed tomography at full inspiration and expiration. E/I(MLD) was calculated by measuring inspiratory and expiratory MLD. Exercise (EX) was defined as metabolic equivalents × duration (hours). COPD patients had higher E/I(MLD) (0.975 vs. 0.964) than healthy subjects. When dividing COPD patients into sedentary (EX < 1.5) and non-sedentary (EX ≥ 1.5) groups, E/I(MLD) in the sedentary group was statistically higher than that in the non-sedentary group (0.983 vs. 0.972). E/I(MLD) > 0.980 was a good predictor of sedentary behavior in COPD (sensitivity, 0.815; specificity, 0.714). Multivariate analysis showed that E/I(MLD) was associated with sedentary behavior (odds ratio, 0.39; p = 0.04), independent of age, symptomology, airflow obstruction, and pulmonary diffusion. In conclusion, higher E/I(MLD) scores are associated with sedentary behavior and can be a useful imaging biomarker for the early detection of physical inactivity in COPD.