Background: Pulmonary fibrosis is the main death cause in systemic sclerosis (SSc). Lung involvement is assessed with spirometry (which evaluates pulmonary function) and chest Computed Tomography (CT) scan (that identifies interstitial fibrosis). FVC < 70 % and DLco < 75 % are considered pulmonary functional values associated with a remarkable interstitial lung disease (ILD) and, therefore, they are adverse prognostical indicators. Both qualitative and semiquantitative radiologic ILD assessment have a considerable interobserver variability. To overcome this problem quantitative scores (called CT indexes) correlating with ILD extent detectable on chest CT have been proposed. The aims of this work are to find: a) whether there is a correlation between pulmonary functional tests and CT indexes , b) which CT indexes have the best performance in discriminating patients with a pulmonary function indicative of an extensive ILD. Methods: Chest TC and spirometry of 90 SSc patients (31 from Parma, 30 from Turin, 29 from Piacenza) meeting ACR criteria were performed. Digital Imaging and COmmunications in Medicine (DICOM) images of chest CT were processed with OsiriX (a free and user-friendly DICOM-viewer) in order to obtain patients' CT indexes. The Spearman rank test was used to verify the correlations between CT indexes and spirometrical measures. CT indexes discriminative ability was verified using ROC analysis. A p-value < 0.05 was considered statistically significant. Results: Whole lung kurtosis (tKurt) is the best FVC correlating CT index (rho = 0,623; p < 0,0001). Parenchymal lung skewness (nSkew) is the best CT index correlating with DLco (rho = -0,582; p < 0,0001). ROC analysis showed that tKurt = 6,32 can discriminate very well patients with FVC < 70% (sensibility 80,0%, specificity 74,3%). Similarly nSkew = 2,2 distiguishes subjects with DLco < 75% (sensibility 85,7%, specificity 52,2%). Conclusions: Spirometry and CT indexes correlations are consistent with literature. The identification of CT index values corresponding to spirometric cutoff indicative of a considerable limitation of lung function makes TC ILD quantification useful in establishing SSc patients prognosis. Obtaining CT indexes with a free and user-friendly software can contribute to widespread in clinical practice this new SSc ILD assessment.
PS164 SYSTEMIC SCLEROSIS INTERSTITIAL LUNG DISEASE EVALUATION: COMPARISON BETWEEN SEMIQUANTITATIVE AND QUANTITATIVE CT ASSESSMENTS / Ariani, Alarico; Lumetti, Federica; Santilli, D; Mozzani, F; Silva, Mario; Delsante, G; Sverzellati, Nicola. - (2014). (Intervento presentato al convegno 3rd Systemic Sclerosis World Congress, 2014 tenutosi a Rome (IT) nel February 6-8, 2014).
PS164 SYSTEMIC SCLEROSIS INTERSTITIAL LUNG DISEASE EVALUATION: COMPARISON BETWEEN SEMIQUANTITATIVE AND QUANTITATIVE CT ASSESSMENTS
SILVA, Mario;SVERZELLATI, Nicola
2014-01-01
Abstract
Background: Pulmonary fibrosis is the main death cause in systemic sclerosis (SSc). Lung involvement is assessed with spirometry (which evaluates pulmonary function) and chest Computed Tomography (CT) scan (that identifies interstitial fibrosis). FVC < 70 % and DLco < 75 % are considered pulmonary functional values associated with a remarkable interstitial lung disease (ILD) and, therefore, they are adverse prognostical indicators. Both qualitative and semiquantitative radiologic ILD assessment have a considerable interobserver variability. To overcome this problem quantitative scores (called CT indexes) correlating with ILD extent detectable on chest CT have been proposed. The aims of this work are to find: a) whether there is a correlation between pulmonary functional tests and CT indexes , b) which CT indexes have the best performance in discriminating patients with a pulmonary function indicative of an extensive ILD. Methods: Chest TC and spirometry of 90 SSc patients (31 from Parma, 30 from Turin, 29 from Piacenza) meeting ACR criteria were performed. Digital Imaging and COmmunications in Medicine (DICOM) images of chest CT were processed with OsiriX (a free and user-friendly DICOM-viewer) in order to obtain patients' CT indexes. The Spearman rank test was used to verify the correlations between CT indexes and spirometrical measures. CT indexes discriminative ability was verified using ROC analysis. A p-value < 0.05 was considered statistically significant. Results: Whole lung kurtosis (tKurt) is the best FVC correlating CT index (rho = 0,623; p < 0,0001). Parenchymal lung skewness (nSkew) is the best CT index correlating with DLco (rho = -0,582; p < 0,0001). ROC analysis showed that tKurt = 6,32 can discriminate very well patients with FVC < 70% (sensibility 80,0%, specificity 74,3%). Similarly nSkew = 2,2 distiguishes subjects with DLco < 75% (sensibility 85,7%, specificity 52,2%). Conclusions: Spirometry and CT indexes correlations are consistent with literature. The identification of CT index values corresponding to spirometric cutoff indicative of a considerable limitation of lung function makes TC ILD quantification useful in establishing SSc patients prognosis. Obtaining CT indexes with a free and user-friendly software can contribute to widespread in clinical practice this new SSc ILD assessment.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.