POS0322 CORRELATION BETWEEN QUANTITATIVE COMPUTED TOMOGRAPHY AND DISEASE ACTIVITY IN SYSTEMIC SCLEROSIS

Abstract

Background:High Resolution Computed Tomography (HRCT) is the gold standard to evaluate Interstitial Lung Diseases (ILDs) extent and severity. Quantitative Computed Tomography (QCT) is a promising tool as it provides an operator-independent assessment of ILD extent. Even if there are emerging data on QCT in Systemic Sclerosis (SSc), its correlation with disease activity (DA) has not been yet studied.Objectives:To evaluate the correlation between QCT score and DA in an Italian cohort of SSc patients.Methods:A multicentric, observational study was conducted in three Italian rheumatological centers.Adult SSc patients classified according to the ACR/EULAR 2013 criteria [1] were assessed with pulmonary function tests, HRCT and for DA. CT images were analyzed quantitatively with the densitometric radiomic data obtained through a free open software – Mean lung attenuation (MLA), Standard Deviation (SD), Kurtosis, Skewness and Lung volume. DA assessment was conducted according to EUSTAR index [2]: a score ≥2.5 was considered indicative of high disease activity.Age below 18 and pregnancy were considered exclusion criteria. We used Student’s T test to evaluate the means of the parameters, Pearson’s r test for correlations, receiver operating characteristics curve to define the cutoff values of the significant details, and linear regression with collinearity test to define the role of the details. P value <0.05 was considered statistically significant.Results:Sixty patients were enrolled (male 8, female 52), with mean age 53.2 years (SD 15.6) and mean disease duration 5.3 years (SD 4.2). QCT indexes distribution was different in high DA vs low DA SSc patients. In particular mean lung attenuation (MLA, -834.7 vs -812.1, p =0.03), standard deviation (95.9 vs 102, p =0.03), skewness (2.2 vs 1.7, P =0.006) and kurtosis (5.5 vs 3.3, p =0.009) of the parenchymal lung and skewness (3.1 vs 2.8, p =0.03) of the whole lung were statistically different. DA correlates with MLA (r =0.28, p =0.003), standard deviation (r =0.21, p =0.02), skewness (r =-0.32, p =0.001) and kurtosis (r =-0.29, p =0.001) of the parenchymal lung and MLA (r =0.25, p =0.006), skewness (R =-0.27, p =0.003), kurtosis (r =-0.21, P =0.02) of the whole lung. The skewness of the parenchymal lung was the QCT index with the best performance in identifying high DA SSc patients (cutoff value ≤1.85; area under the curve 0.74, p =0.005; sensitivity 79.5%, specificity 68.7% accuracy 76.6%).Conclusion:To our knowledge, this is the first study which correlate the QCT score with DA in SSc patients. Our results suggest that QCT can identify SSc patients with high DA score. This could open a scenario of new applications as an operator-independent contribution in DA scores with a potential role in clinical practice. Further studies are needed to confirm the data and to better identify the most suitable parameters for the purpose.