The density histograms-derived computerized integrated index (CII) predicts mortality in idiopathic pulmonary fibrosis

Quantitative assessment of the extent of radiological alterations in interstitial lung diseases is a promising field of application that goes beyond the limitations of qualitative scoring. Analysis of density histograms, i.e., skewness, kurtosis, and mean lung attenuation, is among the most studied approaches. We recently proposed their integration in a single parameter, the computerized integrated index (CII), to reduce their redundancy. The CII has proven effective in detecting subclinical lung involvement, correlates with lung function/disease activity, and predicts mortality in systemic sclerosis patients. Seventy-three newly diagnosed and therapy-naive IPF patients (M = 50; median age: 70.2 years) were prospectively enrolled from January 2014 to December 2022, and followed till December 2023. At baseline, all underwent lung function testing and volumetric high resolution chest CT. Density histograms were analyzed with an open-source automatic platform (Slicer 3D) and CII derived by means of Principal Component Analysis, as previously described. During a median follow-up of 5.8 years, 39 (53.4%) subjects died. Median overall survival (OS) was 4.9 years (95% CI 3.7 years-not estimable). The CII was significantly associated with OS (HR 0.49; 95% CI 0.35-0.68; P < 0.001) and correlated with lung function (r = 0.41; 95% CI 0.19 to 0.60; P < 0.001 for FVC, and r = 0.62; 95% CI 0.44 to 0.75; P < 0.001 for DLCO_sb). Patients stratification according to CII tertile, showed a consistent reduction in the hazard of death. After adjusting for body mass index, smoking, GAP stage, and anti-fibrotic therapy, the CII preserved a significant association with the hazard of death (HR 0.35; 95% CI 0.2-0.63; P < 0.001). CII is a proxy marker of IPF severity worthy of use for prognostication purposes in daily practice.