Assessment of interstitial lung disease in systemic sclerosis using the quantitative CT algorithm CALIPER

Interstitial lung disease (ILD) remains a major cause of morbidity and mortality in systemic sclerosis (SSc). Study aim is to characterize and quantify SSc-ILD by using Computer-Aided Lung Informatics for Pathology Evaluation and Rating (CALIPER). Secondly, our objective is to evaluate which radiological pattern is predictive of lung function decline at 12 months follow-up. In the prospective study (IRB 5435), 66 SSc patients underwent high-resolution computerized tomography (HRCT) at baseline. HRCT was performed according to standard protocol using a CT 64GE light speed VCT power scanner. CALIPER classified lung parenchyma on volume units. Every volume unit was classified into radiological parenchymal patterns (honeycombing, reticular and ground glass). Pulmonary function tests (PFTs) were performed at baseline and after 12 months of follow-up. Cigarette smoking and other lung diseases unrelated to SSc are exclusion criteria. CALIPER analysis showed normal lung parenchyma 87.4 +/- 9.8%, ground glass 2.8 +/- 5.3%, reticular 4 +/- 5.7%, and honeycombing 1 +/- 1%. In multiple regression analysis, FEV1 (p < 0.0001), FVC (p = 0.001), and DLCO (p < 0.0001) measurements at baseline showed a negative correlation with the reticular pattern percentage. At follow-up, DLCO reduction showed a positive correlation (p < 0.001) with the percentage of ground glass pattern (r = 0.33, beta coefficient = 0.51). In the ROC curve analysis, ground glass score is a good predictor (0.75, p = 0.009; 95% CI 0.59-0.91) of DLCO worsening, defined as a decrease of more than 10% of DLCO. Using a cutoff >= 4.5 for ground glass score, the RR for DLCO worsening is 6.8 (p < 0.01; 95% CI 1.6-29.2). The results of this study show that CALIPER is useful not only for quantifying lung damage but also for assessing worsening PFTs, but larger studies are needed to confirm these preliminary data.