Attenuation Based Quantification Of Interstitial Lung Disease Using High Resolution Computed Tomography And Correlation With Pulmonary Function Tests

Interstitial lung disease (ILD) is a chronic and progressive pulmonary disease of the lung parenchyma resulting in fibrotic scar formation of the pulmonary alveoli. The pathologic changes in the lung result in restrictive impairment of lung function. High resolution computed tomography (HRCT) of the chest has become an essential technique in diagnosing and assessing the extent of ILD. Clinical evaluation by HRCT is usually qualitative, leading to inter-observer variability particularly in detecting early disease. We propose a quantitative attenuation-based analysis to standardize the assessment of regional lung disease. We hypothesize that early ILD is associated with greater regional heterogeneity than advanced ILD. To test the hypothesis a method of voxel-wise HRCT image analysis is used to quantify regional lung tissue and air volume within and among lobes. We analyzed HRCT (0.625mm intervals from apex to base) obtained at prone end-inspiration, supine end-inspiration and supine end-expiration in 29 patients with ILD. Each lobe was reconstructed separately. Regional air and tissue volumes, and fractional tissue volume (FTV=tissue/ [air + tissue] volume) were expressed along standard x,y,z axes. FTV increased with increasing ILD severity especially in the periphery. Lobar FTV correlates inversely with global lung function in all lobes. Early ILD is associated with greater FTV heterogeneity within lobes. With increasing ILD severity, FTV heterogeneity within lobes decreases while FTV heterogeneity among lobes increases. We conclude that quantitative analysis of FTV can provide clinically relevant markers of regional ILD. ILD severity is associated with decreasing intra-lobar and increasing inter-lobar FTV heterogeneity, best assessed at prone end-inspiration.