Clinical, Laboratory and Lung Ultrasound Assessment of Congestion in Patients with Acute Heart Failure

Congestion is the main cause of hospitalization in patients with acute heart failure (AHF), however its precise assessment by simple clinical evaluation remains elusive. The recent introduction of the lung ultrasound scan (LUS) allowed to physicians to more precisely quantify pulmonary congestion. The aim of this study was to compare clinical congestion (CC) with LUS and B-type natriuretic peptide (BNP) in order to achieve a more complete evaluation and to evaluate the prognostic power of each measurement. Methods: All patients were submitted to clinical evaluation for blood sample analysis and LUS at admission and before discharge. LUS protocol evaluated the number of B-lines for each chest zone by standardized eight site protocol. CC was measured following ESC criteria. The mean difference between admission and discharge congestion logBNP and B-lines values were calculated. Combined end points of death and rehospitalization was calculated over 180 days. Results: 213 patients were included in the protocol; 133 experienced heart failure with reduced ejection fraction (HFrEF), and 83 presented with heart failure with preserved ejection fraction (HFpEF). Patients with HFrEF had a more increased level of BNP (1150 (812–1790) vs. 851 (694–1196); p = 0.002) and B lines total number (32 (27–38) vs. 30 (25–36); p = 0.05). A positive correlation was found between log BNP and Blines number in both HFrEF (r = 0.57; p < 0.001) and HFpEF (r = 0.36; p = 0.001). Similarly, dividing B-lines among tertiles the upper group (B-lines ≥ 36) had an increased clinical congestion score. Among three variables at admission only B-lines were predictive for outcome (AUC 0.68 p < 0.001) but not LogBNP and CC score. During 180 days of follow-up, univariate analysis showed that persistent ΔB-lines <−32.3% (HR 6.54 (4.19–10.20); p < 0.001), persistent ΔBNP < −43.8% (HR 2.48 (1.69–3.63); p < 0.001) and persistent ΔCC < 50% (HR 4.25 (2.90–6.21); p < 0.001) were all significantly related to adverse outcome. Multivariable analysis confirmed that persistent ΔB-lines (HR 4.38 (2.64–7.29); p < 0.001), ΔBNP (HR 1.74 (1.11–2.74); p = 0.016) and ΔCC (HR 3.38 (2.10–5.44); p < 0.001 were associated with the combined end point. Conclusions: a complete clinical laboratory and LUS assessment better recognized different congestion occurrence in AHF. The difference between admission and discharge B-lines provides useful prognostic information compared to traditional clinical evaluation.