Lung ultrasound and transthoracic impedance for noninvasive evaluation of pulmonary congestion in heart failure

Background: In patients with heart failure, many indexes are available for noninvasive identification of pulmonary congestion: E/E′ at echocardiography; plasma levels of brain natriuretic peptide (BNP) (pg/ml); number of B-lines at lung ultrasound; and transthoracic conductance [thoracic fluid content (TFC)TT = 1/Ω] at impedance cardiography (ICG). Methods: We obtained 75 measures from 50 patients (72 ± 10 years, NYHA 2.4 ± 0.7, ejection fraction 31 ± 7%), 25 of them studied before and after intravenous diuretics, in whom we assessed the following: E/e′ from Doppler echocardiogram; BNP plasma levels; presence and number of B-lines at lung ultrasound; and TFCTT from ICG. We determined the relationship among these indexes and their change with treatment, and compared B-lines and TFC for the diagnosis of pulmonary congestion. Finally, we considered the timing and the personnel required for performing and interpreting each test. Results: A mutual relationship was observed between all the variables. After clinical improvement, changes in each variable were of similar direction and magnitude. Congestion (estimated by chest radiograph) was present in 59% of the patients: TFC value and B-line number had the best sensitivity and specificity for its detection. BNP determination and ICG assessment were performed by a nurse (15 min), and echocardiography and lung ultrasound were performed by a cardiologist (15 min). Conclusion: The correlation between all indexes and their consensual change after improvement of the clinical status suggests that they all detect pulmonary congestion, and that using at least two indexes improves sensitivity and specificity. The choice among the methods may be determined by the patient characteristics or by the clinical setting.