Longitudinal follow-up study of smoking-induced lung density changes by high-resolution computed tomography.

To evaluate the ability of high-resolution computed tomography (HRCT) to detect longitudinal changes in structural abnormalities caused by smoking, HRCT and pulmonary function tests were used to examine nonsmokers, current smokers, and past smokers annually for 5 yr. Inspiratory HRCT was taken for the upper, middle, and lower lung fields, while expiratory images were obtained for the upper and lower lung fields only. We estimated the three quantitative CT parameters including MLD (mean CT value), HIST (CT value with the most frequent appearance), and %LAA (relative area of low attenuation with CT values less than -912 HU). Most of the pulmonary function tests, excepting FEV(1), did not change annually, whereas many of the inspiratory HRCT parameters did. In nonsmokers, only %LAA in the middle or lower lung fields exhibited an annual increase. In current smokers, %LAA in the upper lung field was augmented, while inspiratory MLD or HIST in the middle or lower lung field became more positive. In past smokers, %LAA in any lung field examined increased. The annual change in %LAA in the upper lung field was larger for past smokers than nonsmokers, with little difference between past and current smokers. Expiratory CT parameters showed few annual changes in all groups. In conclusion, (1) aging increases airspace abnormalities, mainly in the lower lung field; (2) although continuous smoking worsens airspace abnormalities mainly in the upper portion of the lung, this trend does not seem to slow down even after smoking cessation; and (3) inspiratory HRCT images are superior to expiratory images for longitudinal estimation of structural abnormalities caused by aging and smoking.