Diagnostic significance of lung ultrasound in evaluation of heart failure: A cross-sectional study in comparison with clinical assessment, proBNP, chest X-ray, and echocardiography

Lung ultrasound (LUS) detection of B-lines is a useful bedside tool for evaluating pulmonary congestion in acute heart failure. This study aimed to assess the diagnostic performance of LUS compared to clinical assessment, N-terminal pro-B-type natriuretic peptide, chest X-ray, and echocardiography in patients with systolic heart failure. A cross-sectional study was conducted on 50 patients diagnosed with heart failure based on Framingham criteria and confirmed by echocardiography at Al Yarmook General Teaching Hospital, Baghdad (May-October 2021). LUS was performed across 16 lung zones to detect B-lines. Diagnostic validity was assessed using receiver operating characteristic curve analysis, referencing ejection fraction ≤ 30%, mitral inflow to annular velocity ratio (E/eé) ≥ 15, and N-terminal pro-B-type natriuretic peptide > 450 pg/mL. From the 16 lung zones, zone number 1 in the left side was with B lines findings in all (46) ultrasound positive patients, followed by zone number 1 in the right side were with positive B lines in (44) patients. Using ≥10 lung zones with B-lines as the diagnostic cutoff for acute heart failure, LUS demonstrated a sensitivity of 64.7% and specificity of 63.6% when ejection fraction ≤30% was used as the reference standard, and a sensitivity of 63.6% and specificity of 88.2% when E/e' ≥ 15 was used as the reference. While using pro-B-type natriuretic peptide > 450 pg/mL as reference for acute heart failure it showed that 4 lung zones with B lines will be the cutoff value with a sensitivity of 86.67% and specificity of 80.00%. Lung sonography could be considered a reliable tool for the assessment of pulmonary congestion in patients with acute heart failure. Using 10 lung zones with B-lines by ultrasound as a cutoff value for the diagnosis of acute heart failure. However, its accuracy may be limited by operator dependency and small sample size.