Search

Purpose Usual interstitial pneumonia (UIP) pattern of interstitial lung disease (ILD) can have varied etiology, with connective tissue disease (CTD) being a common known cause. The anterior upper lobe (AUL) sign, exuberant honeycombing (EHC), and straight edge (SE) sign are recently described computed tomography (CT) signs in CTD-related UIP. We test the diagnostic value of these CT signs for CTD in patients with UIP and compare the incidence of these signs between CTD-related UIP and non–CTD-related UIP. We also evaluated the interobserver agreement in detection of these CT signs. Methods Retrospective study of all patients who had UIP pattern of ILD on CT thorax done from January 1, 2016 to January 31, 2019, was grouped into two: non–CTD-related UIP or CTD-related UIP. CT thorax was reviewed for the presence of these signs—AUL, SE, and EHC. The diagnostic values of these signs in diagnosing CTD-related UIP was assessed. For assessment of interobserver agreement, another radiologist reviewed a subset of 30 randomly selected cases and looked for the presence of these signs. Results Of the 156 patients included, 76 had CTD. The incidence of CT signs were significantly higher in CTD-related UIP. The specificities of AUL, EHC, and SE were 82.5, 75, and 85%, respectively. The EHC sign had highest sensitivity of 48.7%. Inclusion of more than one sign increased the specificity of diagnosis of CTD-related UIP; however, the sensitivity decreases. There was excellent interobserver agreement (0.81–0.87) for each of these signs. Conclusion The presence of SE, AUL, and EHC signs in cases with UIP pattern are specific imaging markers to diagnose underlying CTD; however, due to its low sensitivity, the absence of these signs cannot exclude the same. Because of its excellent interobserver agreement, these signs are reliable in the evaluation of CTD-related ILD.

Abstract The aim of this study was to compare whole lung CT density histograms to predict critical illness outcome and hospital length of stay in a cohort of 80 COVID-19 patients. CT chest images on segmented lungs were retrospectively analyzed. Functional Principal Component Analysis (FPCA) was used to find the main modes of variations on CT density histograms. CT density features, the CT severity score, the COVID-GRAM score and the patient clinical data were assessed for predicting the patient outcome using logistic regression models and survival analysis. ROC analysis predictors of critically ill status: 87.5th percentile CT density (Q875)—AUC 0.88 95% CI (0.79 0.94), F1-CT—AUC 0.87 (0.77 0.93) Standard Deviation (SD-CT)—AUC 0.86 (0.73, 0.93). Multivariate models combining CT-density predictors and Neutrophil–Lymphocyte Ratio showed the highest accuracy. SD-CT, Q875 and F1 score were significant predictors of hospital length of stay (LOS) while controlling for hospital death using competing risks models. Moreover, two multivariate Fine-Gray regression models combining the clinical variables: age, NLR, Contrast CT factor with either Q875 or F1 CT-density predictors revealed significant effects for the prediction of LOS incidence in presence of a competing risk (death) and acceptable predictive performances (Bootstrapped C-index 0.74 [0.70 0.78]).