COPD phenotypes by computed tomography and ventilatory response to exercise

Introduction

Computed tomographic (CT) phenotypic patterns of chronic obstructive pulmonary disease (COPD) identify different clinical features of disease. The impact of these variables on the physiological response to exercise has been the focus of a great deal of research as it allows more individualized clinical approaches. The aim of our study was to evaluate the relationships between CT phenotyping of subjects with COPD and the ventilatory response during cardiopulmonary exercise testing (CPET).

Methods

Subjects with COPD were classified into four phenotypes based on CT metrics of emphysema (low attenuation area less than a threshold of -950 Hounsfield [%LAA-950]) and airwall thickness (bronchial wall area percentage [%WA]).

Results

Eighty COPD patients (78.8% males, median age 65±11.3 years) were enrolled in the study. Based on CT phenotype, 25 (31.3%) patients were classified as normal, 27 (33.8%) air dominant, 17 (21.3%) emphysema dominant and 11 (13.8%) mixed type. The emphysema and mixed phenotypes showed the highest ventilatory equivalent for carbon dioxide (V_E/VCO₂) and V_E/VCO₂ slope (p<0,05). In all phenotypes, %LAA was positive correlated with V_E/VCO₂ and V_E/VCO₂ slope (r = 0.437, p = 0.006 and r = 0.503, p<0.001, respectively). %WA also showed a positive correlation with V_E/VCO₂ and V_E/VCO₂ slope (r = 0.541, p<0.001 and r = 0.299, p = 0.033, respectively). In multivariate regression models, after adjustment for age, BMI, sex and FEV1, %LAA was the only independent predictor of V_E/VCO₂ and V_E/VCO₂ slope (β 0.343, SE 0.147, 95% CI 0.009/0.610, p = 0.044 and β 0.496, SE 0.081, 95% CI 0.130/0.455, p = 0.001, respectively).

Conclusion

Emphysema (%LAA) and airways metrics (%WA) had strong relationships with the different characteristics of ventilatory response to exercise in subjects with mild to moderate COPD. In particular, %LAA seemed to play an important role as an independent predictor of V_E/VCO₂ and V_E/VCO₂ slope. These results suggested that CT phenotyping may help predicting ventilatory response to exercise in subjects with COPD.