Lung ultrasound for the evaluation of pulmonary congestion in outpatients: a comparison with clinical assessment, natriuretic peptides, and echocardiography.

OBJECTIVES The aim of this study was to define the performance of lung ultrasound (LUS) compared with clinical assessment, natriuretic peptides, and echocardiography, to evaluate decompensation in patients with systolic heart failure (HF) in an outpatient clinic. BACKGROUND Evaluation of pulmonary congestion in chronic HF is challenging. LUS has been recently proposed as a reliable tool for the semiquantification of extravascular lung water through assessment of B-lines. METHODS This was a cohort study of patients with moderate to severe systolic HF. Receiver-operating characteristic (ROC) analyses were performed to compare LUS with a previously validated clinical congestion score (CCS), amino-terminal portion of B-type natriuretic peptide (NT-proBNP), E/e' ratio, chest x-ray, and 6-min walk test. RESULTS Ninety-seven patients were enrolled. Decompensation was present in 57.7% of patients when estimated by CCS, 68% by LUS, 53.6% by NT-proBNP, and 65.3% by E/e' ≥15. The number of B-lines was correlated to NT-proBNP (r = 0.72; p < 0.0001), E/e' (r = 0.68; p < 0.0001), and CCS (r = 0.43; p < 0.0001). In ROC analyses, considering as reference for decompensation a combined method (E/e' ≥ 15 and/or NT-proBNP >1,000 pg/ml), LUS yielded a C-statistic of 0.89 (95% confidence interval: 0.82 to 0.96), providing the best accuracy with a cutoff ≥ 15 B-lines (sensitivity 85%, specificity 83%). A systematic approach using CCS, E/e', NT-proBNP, chest x-ray, and 6-min walk test in different combinations as reference for decompensation also corroborated this cutoff and found a similar accuracy for LUS. CONCLUSIONS In an HF outpatient clinic, B-lines were significantly correlated with more established parameters of decompensation. A B-line ≥15 cutoff could be considered for a quick and reliable assessment of decompensation in outpatients with HF.