Diagnostic performance of lung ultrasound in patients with clinically suspected interstitial lung disease

Objective To assess the diagnostic performance of lung ultrasound (LUS) in patients with clinically suspected interstitial lung disease (ILD) and its association with high-resolution computed tomography (HRCT) Warrick scores. Methods This prospective cross sectional study consecutively enrolled patients with clinical features suggestive of ILD. Participants underwent LUS, HRCT, and multidisciplinary clinical assessment. The final ILD diagnosis was established using a comprehensive multidisciplinary clinical reference standard. LUS was considered positive when the total B-line count was > 5. Diagnostic performance of LUS was evaluated against the final diagnosis. Receiver operating characteristic analysis assessed the discriminative performance of total B-line count, while Spearman correlation evaluated associations between LUS parameters and HRCT Warrick scores. Results Of 81 patients with suspected ILD, 71 cases were confirmed to have ILD, while the remaining 10 served as a non-ILD comparison group. LUS identified 65 of the 71 confirmed ILD cases and demonstrated 91% sensitivity, 70% specificity, 95% positive predictive value, 53% negative predictive value, and a negative likelihood ratio of 0.12. Total B-line count, pleural thickness, and pleural line irregularity were positively correlated with HRCT Warrick scores (ρ = 0.94, 0.69, and 0.57, respectively; all P < .001). Conclusion LUS demonstrated high sensitivity for detecting ILD in patients with clinical suspicion of disease and may serve as an initial adjunctive or triage tool to help identify patients requiring further radiological evaluation with HRCT. However, a negative LUS examination cannot reliably exclude ILD.