Bedside lung ultrasound versus chest CT in critically ill patients: a cross-sectional diagnostic accuracy study

Abstract

Study aim: Evluate bedside lung ultrasound (LUS) diagnostic accuracy versus chest computed tomography (CT) in adult ICU patients with respiratory distress. Design and methods: Single-center, cross-sectional study of 200 paired LUS-CT examinations. LUS was performed within 12 h, scanning six zones per hemithorax and classifying findings into A-lines, C1 (pneumonia), C2 (atelectasis), B1, B2 (interstitial patterns), pleural effusion, and emphysema. Diagnostic performance metrics included sensitivity, specificity, predictive values, likelihood ratios, and ROC analysis. Key results: CT showed consolidations (60.5%), pleural effusions (42.5%), interstitial syndrome (28.5%), ground-glass opacities (26%), and emphysema (23.5%). LUS detected C1 in 66.5% (posterior-inferior predominance), B2 in 40.5%, B1 in 40%, pleural effusions in 38.5%, C2 in 13.5%, and emphysema in 16.5%. Overall concordance with CT was 81%. LUS had very high specificity for pleural effusion (Sp = 96.5%, PLR = 24.5, accuracy = 91.9%) and atelectasis (Sp = 93.7%, NPV = 97.6%, PLR = 12.6, accuracy = 92.3%), very high sensitivity for alveolar-interstitial syndrome (Se = 96.2%, NPV = 98.3%, NLR = 0.05), and good sensitivity for consolidations (Se = 89.3%). Conclusion: LUS is a reliable bedside tool that can reduce CT use, radiation exposure, and enable dynamic monitoring; structured training is essential.