Deep Learning Assessment of Progression of Emphysema and Fibrotic Interstitial Lung Abnormality.

Abstract

While studies have evaluated emphysema and fibrotic interstitial lung abnormality (ILA) individually, less is known about their combined progression. To define clinically meaningful progression of fibrotic ILA in smokers without interstitial lung disease and evaluate the effects of fibrosis and emphysema progression on mortality. Emphysema and pulmonary fibrosis were assessed on baseline and 5-year follow-up computed tomography scans of 4450 smokers in the COPDGene study using deep learning algorithms. Emphysema was classified as absent, trace, mild, moderate, confluent, or advanced destructive. Fibrosis was expressed as a percentage of lung volume. Emphysema progression was defined as an increase by at least one grade. A hybrid distribution and anchor-based method was used to determine minimally clinically important difference (MCID) in fibrosis. The relationship between progression and mortality was evaluated using multivariable shared frailty models using an age timescale. The MCID for fibrosis was 0.58%. Based on this threshold, 2822 (63%) had progression of neither emphysema or fibrosis, 841 (19%) had emphysema progression alone, 512 (12%) had fibrosis progression alone, and 275 (6.2%) had progression of both. Compared to non-progressors, hazard ratios for mortality were 1.42 (95% CI: 1.11-1.82) in emphysema progressors, 1.49 (1.14-1.94) in fibrosis progressors, and 2.18 (1.58-3.02) in those with progression of both emphysema and fibrosis. In smokers without known interstitial lung disease, small changes in fibrosis may be clinically significant, and combined progression of emphysema and fibrosis is associated with increased mortality.