Diagnostic accuracy of lung ultrasound for interstitial lung disease in patients with connective tissue diseases: a meta-analysis

Abstract

Objective The purpose of this study was to evaluate and compare the diagnostic performance of lung ultrasound (US) in respect to high-resolution computed tomography (HRCT) findings in patients with connective tissue disease-associated interstitial lung disease (CTD-ILD). Methods We searched the Pubmed, Embase, and Cochrane Library databases, and performed a meta-analysis on the diagnostic accuracy of lung US according to B-lines (comet tail sign) and on the correlation coefficients between lung US scores and HRCT Warrick scores in CTD-ILD patients. Results Five studies that included a total of 349 patients were available for this meta-analysis. The pooled sensitivity and specificity of lung US were 91.5% (95% confidence interval [CI]: 84.5-96.0) and 81.3% (95% CI: 74.6-86.9), respectively. The positive likelihood ratio, negative likelihood ratio, and diagnostic odds ratio were 4.100 (2.133-7.879), 0.176 (0.006-0.363), and 34.73 (10.10-99.66), respectively. The area under the curve was 0.915 and the Q* index was 0.848, indicating a high diagnostic accuracy. When all four studies with systemic sclerosis were considered together, the pooled sensitivity and specificity of lung US were 89.5% (95% CI 80.3-95.3) and 79.6% (69.9-87.2), respectively. A significant correlation was found between lung US B-line scores and HRCT Warrick scores in CTD-ILD (correlation coefficient: 0.783; p-value <1 x 10(-9)). Conclusion Our meta-analysis of published studies demonstrates that lung US has a high diagnostic accuracy, correlates well with HRCT findings, and plays an important role in the diagnosis of CTD-ILD.