Prediction of airflow limitation in smokers using multiple CT imaging indices: airway wall thickness, quantitative emphysema, and visually identified centrilobular emphysema

Abstract

Background: Visually diagnosed centrilobular emphysema (CLE) and airway wall thickness and quantitative emphysema on computed tomography (CT) are associated with airflow limitation (AFL) in smokers. The aim of this study was to determine whether a combination of CT indices can be used to accurately detect AFL in smokers.

Methods: We retrospectively included male subjects aged ≥40 years with a smoking history of at least 10 pack-years who participated in health checkups and underwent lung cancer screening CT and spirometry at two Japanese hospitals. The percentage of low-attenuation areas < -950 HU (LAA%) was quantified. CLE and paraseptal emphysema were visually assessed according to the Fleischner Society criteria. Airway wall thickness was calculated as the mean percentage of wall area (WA%) of prespecified segmental and subsegmental bronchi from each lung lobe. Logistic regression models were used to develop diagnostic predictions for AFL with a forced expiratory volume in 1 s/forced vital capacity <0.7, and the area under the receiver operating characteristic curve (AUC) was determined.

Results: Among the 513 smokers included in this study, 55 had AFL. In multivariable regression analysis models using the stepwise backward variable selection method, CLE presence, high LAA%, and high subsegmental WA% were independently associated with AFL. A logistic regression model using CLE, LAA%, and subsegmental WA% predicted AFL with a sensitivity of 83.6 %, specificity of 87.8 %, and AUC of 0.908.

Conclusions: A model using airway wall thickness, quantitative emphysema, and visually-identified CLE may facilitate the detection of AFL in smokers.

Keywords: Airflow limitation; Airway wall thickness; Centrilobular emphysema; Chronic obstructive pulmonary disease; Emphysema.