Multiparametric assessment of decongestion in patients admitted with acute heart failure: dynamic changes and correlation with natriuresis

Abstract Background Heart Failure (HF) is a major public health problem, expected to increase in the next few decades. The vast majority of acute HF (AHF) episodes are precipitated by congestion with volume overload and/or fluid redistribution. Current Guidelines stress on the importance of relieving congestion before discharge, recommending anyway only daily assessment of signs and symptoms. Lung ultrasound (LUS) B-lines and other ultrasound markers of congestion has emerged as valuable tools to improve diagnostic accuracy and help clinicians in fluid management. Purpose To assess the dynamic changes of B-lines and other markers of congestion and their relationship with the clinical evaluation, chest X-ray (CXR) and natriuretic peptides in patients admitted for AHF. Methods Fifty-eight patients (age 72±11 ys, 35% women) admitted with acute HF were enrolled. Patients underwent clinical evaluation (Clinical Congestion Score, CCS, including rales, pathological S3 heart sound, pedal oedema, jugular vein distension; range 0-4), and CXR (CXR score, including enlargement and loss of definition of hilar structures, peribronchial and perivascular cuffing, cardiomegaly, pleural effusion; range 0-4). An integrated multiorgan ultrasound exam including LUS B-lines, echocardiography, and venous excess ultrasound (VExUS) score was performed at admission (T0), between 24 and 48 hours from T0 (T1), and at discharge (T2). Follow-up data at 30 days were collected. Results Mean CCS was 1.52±0.99 at T0, with a reduction at T1 (0.86±0.66; p<0.01), and at T2 (0.34±0.66; p<0.01). Sixteen percent of patients with a CCS=0 at T0 showed, however, persistent B-lines on LUS (mean B-lines 19±6). The mean number of B-lines was 26±12 at T0 with a reduction at T1 (17±11; p<0.001), and at T2 (13±11; p<0.001). Other echocardiographic non-invasive hemodynamic parameters, such as pulmonary artery systolic pressure (PASP) and E/e’ showed a reduction only at T2, but not at T1 (figure). The number of B-lines correlated at all time points with CCS, CXR score, VExUS score, and NT-proBNP values (all p<0.05). We divided patients into two groups: "responders" (with ≥15 reduction of B-lines between T0 and T2) and "non-responders" (with <15 reduction of B-lines between T0 and T2). Responders had higher chlorine and sodium urinary excretion at T1 compared to non-responders. The number of B-lines at all time points was associated to 30-days cardiovascular mortality at univariable analysis. Neither CCS nor CXR score were associated with 30-day events. Conclusions In patients with acute HF, B-lines quantification is related to standard congestion assessment based on clinical evaluation, CXR and NT-proBNP, showing, however, a more dynamic behaviour in the assessment of decongestion. B-lines variations are also related to diuresis and chlorine urinary output, a recognized predictor of diuretic response. Persistent congestion is related to an increased risk of hospitalization and death.