Quantitative CT in COPD MAP: Emphysema and small airways disease independently contribute to FEV1

Background: Quantitative CT can describe COPD heterogeneity; Percentile 15(perc15) describes emphysema, Mean Lung DensityE/I (MLD E/I) is a surrogate measure of small airways disease and Percentage Wall Area(%WA) reflects bronchial wall thickness. We hypothesised that CT measures independently influence disease severity defined by FEV1%predicted (FEV1%) Methods: All patients had a full inspiratory/expiratory CT and pulmonary function tests. Associations between CT parameters and lung function were explored. Results: 65 patients; 68% were male, with mean(SEM) age 68(1), FEV1% 52(2)% and FEV1/FVC ratio 51(1)%. Significant correlations (p<0.0001) were observed between MLD E/I and RV% predicted (r=0.47), FRC% (r=0.491), FEV1% (r=-0.66) and FEV1/FVC ratio (r=-0.52). There were also significant correlations between Perc15 and RV% (r=-0.35,p=0.008), FRC% (r=-0.37, p=0.003), FEV1% (r=0.41, p=0.001), FEV1FVC ratio (r=0.51, p<0.0001) and KCO% (r=0.42, p=0.001). There were no significant correlations between %WA in RB1 and lung function measures. Standard multiple regression was performed with FEV1% as the dependent variable and MLD E/I and perc15 as independent variables. The model showed no violation of multicollinearity or homoscedasticity. MLD E/I and perc15 explained 36% of the variance in FEV1%. MLD E/I made the largest unique contribution (beta=-0.45, p<0.001), although perc15 also made a statistically significant contribution (beta=0.30, p=0.008). Conclusions: Emphysema and small airway measures independently contribute to disease severity defined by FEV1%, with small airways measures (MLD E/I) having a larger role.