-
PDF
- Split View
-
Views
-
Cite
Cite
Nicola Riccardo Pugliese, Pierpaolo Pellicori, Francesco Filidei, Lavinia Del Punta, Nicolò De Biase, Alessio Balletti, Valerio Di Fiore, Alessandro Mengozzi, Stefano Taddei, Luna Gargani, Wilfried Mullens, John G F Cleland, Stefano Masi, The incremental value of multi-organ assessment of congestion using ultrasound in outpatients with heart failure, European Heart Journal - Cardiovascular Imaging, Volume 24, Issue 7, July 2023, Pages 961–971, https://doi.org/10.1093/ehjci/jeac254
- Share Icon Share
Abstract
We investigated the prevalence and clinical value of assessing multi-organ congestion by ultrasound in heart failure (HF) outpatients.
Ultrasound congestion was defined as inferior vena cava of ≥21 mm, highest tertile of lung B-lines, or discontinuous renal venous flow. Associations with clinical characteristics and prognosis were explored. We enrolled 310 HF patients [median age: 77 years, median NT-proBNP: 1037 ng/L, 51% with a left ventricular ejection fraction (LVEF) <50%], and 101 patients without HF. There were no clinical signs of congestion in 224 (72%) patients with HF, of whom 95 (42%) had at least one sign of congestion by ultrasound (P < 0.0001). HF patients with ≥2 ultrasound signs were older, and had greater neurohormonal activation, lower urinary sodium concentration, and larger left atria despite similar LVEF. During a median follow-up of 13 (interquartile range: 6–15) months, 77 patients (19%) died or were hospitalized for HF. HF patients without ultrasound evidence of congestion had a similar outcome to patients without HF [reference; hazard ratio (HR) 1.02, 95% confidence interval (CI) 0.86–1.35], while those with ≥2 ultrasound signs had the worst outcome (HR 26.7, 95% CI 12.4–63.6), even after adjusting for multiple clinical variables and NT-proBNP. Adding multi-organ assessment of congestion by ultrasound to a clinical model, including NT-proBNP, provided a net reclassification improvement of 28% (P = 0.03).
Simultaneous assessment of pulmonary, venous, and kidney congestion by ultrasound is feasible, fast, and identifies a high prevalence of sub-clinical congestion associated with poor outcomes.

(A) Congestion assessment by ultrasound (US): increasing B-lines (≥4), discontinuous renal venous flow (RVF), and dilated inferior vena cava (IVC ≥ 21 mm). (B) The value of measuring congestion by US: Kaplan–Meier for the primary outcome of all-cause death (ACM) and heart failure (HF) hospitalizations (HFHs) in HF patients (black lines) and patients without HF (red line). HF patients with ≥2 US signs had the worst outcome, while those free of congestion at US had a prognosis similar to patients without HF.
Introduction
Excessive renal salt and water retention increases cardiac filling pressures, leading to an expanded intravascular volume and interstitial fluid accumulation, otherwise known as congestion.1 Congestion is a key driver of symptoms, poor quality of life, disease progression, and prognosis for patients with heart failure (HF), irrespective of left ventricle ejection fraction (LVEF).2,3 In patients with HF, clinical signs of congestion (e.g. an elevated jugular venous pressure or presence of peripheral oedema) identify those most at risk.1 However, clinically overt congestion is a late manifestation of HF and might often be missed unless it is severe. Indeed, the assessment of clinical congestion is subjective and highly dependent on clinical skills, experience, and level of training.4,5
Recent research has shown that quantification of venous [measuring the inferior vena cava (IVC) and/or jugular vein distention], lung (counting B-lines), and renal congestion by ultrasound (US) is feasible and identifies patients with HF with a worse prognosis.6 Whether a simultaneous, multi-organ evaluation of congestion by US further improves risk stratification is unknown. Therefore, in the present study, we assessed congestion by US in the great veins, lungs, and kidneys in patients with HF across the LVEF spectrum to understand the prevalence of these signs, their physiopathology, and prognostic relevance.
Methods
Study population
Between September 2020 and December 2021, we prospectively enrolled 326 consecutive patients with a prior clinical diagnosis of HF, attending a routine follow-up visit at the University Hospital of Pisa, Italy. During the same period, we also enrolled 102 consecutive patients with cardiovascular risk factors, including hypertension, Type II diabetes, or chronic ischaemic heart disease and no history of HF. They were required to have an NT-proBNP of <125 ng/L, LVEF of ≥50%, and were not taking loop diuretics.7 All patients underwent a complete clinical assessment, blood and urine tests, 12-lead ECG, and an US evaluation. We excluded eight patients (<2%) whose quantification of US congestion was impossible due to poor quality images (IVC: n = 2; lung: n = 5; kidney: n = 1). All the patients with a history of lung disease underwent spirometry, and those (n = 9, 2%) with more than moderate airflow obstruction were excluded. The final study population (n = 411) consisted 101 patients without HF and 310 patients with HF, who were stratified into HF with reduced (LVEF <50%, HFrEF, n = 159) and preserved LVEF (≥50%, HFpEF, n = 151; Supplementary material online, Figure S1). The study fulfilled the requirements in the Declaration of Helsinki; the protocol was approved by the local ethics committees (number 19204), and written informed consent was obtained from all patients.
Laboratory evaluation
Patients were instructed to fast overnight and not to take any medications before blood and urine sampling on the morning of the tests. Blood samples were drawn after a 30-min supine rest. Detailed laboratory protocol is provided in Supplementary material online, Appendix. After blood sampling, patients were asked to give a urine sample. We estimated the fractional excretion of sodium (FENa) as8:
We defined micro-albuminuria and macro-albuminuria as a urinary albumin-to-creatinine ratio (UACR) >30 and >300 mg/g, respectively. We evaluated the instantaneous estimated plasma volume status (ePVS) in mL/g9 as:
Baseline echocardiography
All patients underwent a comprehensive transthoracic echocardiography examination with a phased array transducer (1–5 MHz, Hitachi Medical Systems LISENDO 880, Tokyo, Japan) according to the international recommendations, including 3D and speckle tracking evaluation (STE).10 We non-invasively estimated echo-derived pulmonary artery wedge pressure (ePAWP) and pulmonary vascular resistance using a previously validated equation.11 A detailed echocardiographic protocol is provided in Supplementary material online, Appendix.
Renal venous flow
Doppler assessment of renal venous flow (RVF) was performed in the left lateral decubitus position, using the same phased array transducer aligned with the lowest intercostal space rendering a longitudinal view of the right kidney. The flow scale of colour Doppler imaging was adjusted to low-flow velocities (<20 cm/s) to optimize the identification of the interlobar veins. The best-aligned vein was then sampled with pulsed-wave Doppler during an end-expiratory breath hold. The scale was adjusted to maximize the signal amplitude (usually around ≈20 cm/s), and the electrocardiographic signal was used to synchronize the RVF signal with the cardiac cycle. We used a semi-quantitative assessment of the effects of central venous pressure (CVP) on renal haemodynamics (Graphical Abstract). In normal conditions, the interlobar RVF is continuous with a small varying amplitude during the cardiac cycle. With worsening congestion, the amplitude variation increases with the minimal velocity gradually approaching zero, eventually leading to a discontinuous, pulsatile, or even biphasic flow. In the most severe cases, RVF is only seen during diastole (monophasic).6 We also measured the renal venous impedance index (VII) and the venous discontinuity index (VDI). The VII is the ratio of the difference between the maximum and minimum velocity to the maximum velocity during a cardiac cycle with a number varying from 0 (no variation in velocity) to 1 (minimum velocity is zero, i.e. the flow becomes discontinuous). VDI expresses the percentage of no-flow time during a cardiac cycle.6
Lung ultrasound
B-lines were measured with a linear transducer in parallel orientation (transverse) to the ribs at an imaging depth of ∼15–18 cm using an eight-region scan.6 In each region, B-lines were counted one by one if distinguishable; if confluent, we estimated their number by the percentage of space occupied on the screen, divided by 10 (up to a max of 10 B-lines/region). The sum of B-lines from the eight scanning regions yielded a score denoting the extent of the extravascular lung water; zero was defined as a complete absence of B-lines.
We defined congestion by US as a discontinuous RVF, a dilated IVC (≥21 mm), or B-lines above or equal to the lower boundary of the highest tertile (≥4).6
Clinical follow-up
The minimum follow-up period was 3 months for the last patient enrolled. The primary outcome was all-cause mortality or HF hospitalization. HF hospitalization was defined as an in-hospital stay >24 h due to HF as the primary diagnosis on the discharge letter. When multiple events occurred, patients were censored at the time of the first event. In our department, follow-up events are regularly adjudicated by an independent trained investigator, blinded to clinical data, and entered into a dedicated database. Detailed statistical analysis protocol is provided in Supplementary material online, Appendix.
Results
Study population
Demographic and clinical characteristics of the HF population according to US signs of congestion are shown in Table 1, while the whole study population, including those without HF, is summarized in Supplementary material online, Table S1. Likewise, the US evaluation is summarized in Table 2 and Supplementary material online, Table S2. The reproducibility of US markers of congestion was excellent (see Supplementary material online, Table S3). Classification of RVF patterns was highly consistent between intra-observer and inter-observer assessments (κ > 0.9). On average, the time of acquisition and interpretation for lung ultrasound (LUS) required <3 min per patient and RVF about 5 min per patient. Congestion by US was nearly twice as common as clinical congestion. Around one-third of patients with HF had pulmonary or renal congestion, while only one in five had evidence of venous congestion; isolated venous congestion was rare (2%). Patients with HF with ≥2 US signs of congestion had higher plasma concentrations of NT-proBNP than those with less congestion by ultrasound; this was true even when only patients in sinus rhythm were considered (Figure 1).

Distribution of patients with heart failure (HF) according to the presence of congestion by ultrasound (US). HF patients with ≥2 US signs of congestion had the highest levels of NT-proBNP (bottom left), also when considering HF patients with sinus rhythm only (bottom centre) or HF patients without sinus rhythm (i.e. n = 122 having atrial fibrillation and n = 8 with atrial pacing; bottom right). IVC, inferior vena cava; RVF, renal venous flow.
Variable . | HF w/o US signs of congestion (n = 149) . | HF with 1 US sign of congestion (n = 82) . | HF with ≥2 US signs of congestion (n = 79) . | P-value . |
---|---|---|---|---|
Demographics (0 missing) | ||||
Age, years | 73 (64–81) | 75 (69–82)* | 78 (72–85)* | 0.001 |
Men | 89 (60) | 56 (68) | 56 (71) | 0.2 |
BMI, kg/m2 | 27.5 ± 4.7 | 28.4 ± 5.1 | 28.6 ± 5.6 | 0.2 |
BSA, m2 | 1.9 ± 0.2 | 1.9 ± 0.3 | 1.9 ± 0.3 | 0.9 |
Smoker | 27 (18) | 16 (20) | 18 (23) | 0.3 |
NYHA class | 0.001 | |||
I | 28 (19) | 12 (15) | 7 (9) | |
II | 97 (65) | 46 (56) | 37 (47) | |
III | 24 (16) | 24 (29) | 35 (44) | |
KCCQ score | 53 ± 22 | 50 ± 27 | 48 ± 29 | 0.1 |
Arterial hypertension | 112 (75) | 65 (79) | 57 (72) | 0.4 |
Stroke/TIA | 18 (12) | 8 (10) | 8 (10) | 0.5 |
Diabetes mellitus | 43 (29) | 25 (30) | 25 (32) | 0.2 |
COPD | 24 (16) | 16 (20) | 15 (19) | 0.3 |
CAD | 48 (32) | 29 (35) | 26 (33) | 0.3 |
Previous MI | 34 (23) | 21 (26) | 18 (23) | 0.2 |
Previous PCI/CABG | 42 (28) | 25 (30) | 22 (28) | 0.5 |
Pacemaker | 31 (21) | 18 (22) | 18 (23) | 0.3 |
ICD | 26 (17) | 12 (15) | 14 (18) | 0.6 |
CRT | 19 (13) | 9 (11) | 4 (5) | 0.1 |
Atrial fibrillation | 39 (26) | 38 (46)* | 45 (57)*,** | 0.01 |
Clinical evaluation (0 missing) | ||||
Brachial systolic BP, mmHg | 131 ± 22 | 130 ± 22 | 121 ± 19*,** | 0.001 |
Brachial diastolic BP, mmHg | 77 ± 14 | 76 ± 14 | 76 ± 13 | 0.8 |
Heart rate, b.p.m. | 72 ± 12 | 73 ± 16 | 73 ± 15 | 0.9 |
No clinical signs of congestion | 129 (87) | 61 (74)* | 34 (43)*,** | <0.0001 |
Pitting oedema (any degree) | 23 (15) | 21 (25) | 35 (44)*,** | 0.001 |
Lung crackles (any degree) | 4 (3) | 7 (9) | 23 (28)*,** | <0.0001 |
Jugular vein distension (any degree) | 0 | 6 (7) | 14 (18)*,** | <0.0001 |
Blood tests (0 missing) | ||||
Haemoglobin, g/dL (men) | 13.5 ± 1.7 | 13.5 ± 1.9 | 13.4 ± 1.6 | 0.1 |
Haemoglobin, g/dL (women) | 12.9 ± 1.8 | 12.8 ± 1.8 | 12.7 ± 2.1 | 0.1 |
Na+, mEq/L | 140 ± 3 | 141 ± 3 | 142 ± 3* | 0.005 |
K+, mEq/L | 4.3 ± 0.5 | 4.3 ± 0.4 | 4.3 ± 0.5 | 0.9 |
Total cholesterol, mg/dL | 163 ± 41 | 163 ± 39 | 164 ± 40 | 0.3 |
Fasting glucose, mg/dL | 98 ± 31 | 99 ± 28 | 102 ± 26 | 0.6 |
HbA1c, mmol/mol | 41 ± 10 | 41 ± 10 | 41 ± 9 | 0.9 |
Uric acid, mg/dL | 7.3 ± 1.9 | 7.7 ± 2.4 | 8.3 ± 2.9*,** | <0.0001 |
hs-CRP, mg/dL | 0.42 (0.24–0.60) | 0.44 (0.25–0.63) | 0.48 (0.27–0.69)* | 0.006 |
Creatinine, mg/dL | 1.01 (0.82–1.23) | 1.05 (0.91–1.30) | 1.08 (0.95–1.32) | 0.1 |
eGFR, mL/min/1.73 m2 | 65 (54–82) | 64 (52–77) | 61 (50–72) | 0.1 |
Urea, mg/dL | 47 (36–57) | 48 (37–58) | 54 (43–72)*,** | 0.001 |
ePVS, mL/g | 4.7 (4.1–5.3) | 4.8 (4.3–5.5) | 5.3 (4.7–6.2)*,** | <0.0001 |
Osmolality, mOsm/kg | 294 (289–298) | 294 (290–299) | 298 (292–301)*,** | <0.0001 |
Norepinephrine, pg/mL | 318 (232–384) | 332 (287–483) | 398 (313–592)*,** | <0.0001 |
Renin, mIU/L | 21.1 (6.1–51.4) | 23.7 (8.2–75.4) | 28.3 (9.1–123.1) | 0.1 |
Aldosterone, ng/dL | 9.5 (6.5–12.7) | 10.3 (7.1–13.8) | 15.1 (9.8 −23.1)*,** | <0.0001 |
NT-proBNP, pg/mL | 409 (308–862) | 936 (627–1671)* | 1862 (1379–2886)*,** | <0.0001 |
NT-proBNP, pg/mL (SR only) | 211 (160–543) | 336 (185–749) | 847 (561–1590)*,** | <0.0001 |
hs-Troponin T, pg/mL | 15 (11–24) | 18 (12–27) | 25 (15–46)*,** | <0.0001 |
Urine test (12 missing) | ||||
Osmolality, mOsm/kg | 540 (417–680) | 507 (408–630) | 456 (334–571)* | 0.001 |
UACR, mg/g | 21 (8–63) | 23 (9–65) | 57 (21–99)*,** | <0.0001 |
Micro-albuminuriaa | 38 (25) | 26 (32) | 51 (65) | <0.0001 |
Macro-albuminuriaa | 7 (5) | 5 (6) | 11 (14) | 0.04 |
Spot urinary sodium, mEq/L | 69 (48–118) | 63 (44–117) | 45 (30–76)*,** | <0.0001 |
FENa, % | 0.58 (0.40–0.91) | 0.53 (0.37–0.89) | 0.40 (0.20–0.66)*,** | <0.0001 |
Therapy (0 missing) | ||||
Beta-Blocker | 118 (79) | 66 (80) | 64 (81) | 0.5 |
DHP CCB | 34 (23) | 18 (22) | 9 (11) | 0.1 |
Non-DHP CCB | 0 | 2 (2) | 0 | 0.2 |
ACEi or ARB | 94 (63) | 50 (61) | 52 (66) | 0.3 |
MRA | 69 (46) | 43 (52) | 45 (57) | 0.2 |
ARNI | 37 (25) | 14 (17) | 15 (19) | 0.3 |
ASA | 67 (45) | 34 (41) | 37 (47) | 0.7 |
Statins | 83 (56) | 42 (51) | 41 (52) | 0.3 |
Thiazides/thiazide-like diuretics | 16 (11) | 11 (13) | 5 (6) | 0.4 |
Loop diuretics | 106 (71) | 62 (75) | 62 (78) | 0.1 |
Furosemide equivalent dose | 0.4 | |||
1–50 mg | 28 (19) | 14 (17) | 12 (15) | |
51–100 mg | 70 (47) | 39 (48) | 42 (53) | |
>100 mg | 8 (6) | 7 (8) | 10 (13) | |
SGLT2i | 19 (13) | 9 (11) | 8 (10) | 0.3 |
Insulin | 15 (10) | 6 (7) | 5 (6) | 0.2 |
Oral anticoagulants | 40 (27) | 39 (48)* | 46 (58)* | 0.01 |
Variable . | HF w/o US signs of congestion (n = 149) . | HF with 1 US sign of congestion (n = 82) . | HF with ≥2 US signs of congestion (n = 79) . | P-value . |
---|---|---|---|---|
Demographics (0 missing) | ||||
Age, years | 73 (64–81) | 75 (69–82)* | 78 (72–85)* | 0.001 |
Men | 89 (60) | 56 (68) | 56 (71) | 0.2 |
BMI, kg/m2 | 27.5 ± 4.7 | 28.4 ± 5.1 | 28.6 ± 5.6 | 0.2 |
BSA, m2 | 1.9 ± 0.2 | 1.9 ± 0.3 | 1.9 ± 0.3 | 0.9 |
Smoker | 27 (18) | 16 (20) | 18 (23) | 0.3 |
NYHA class | 0.001 | |||
I | 28 (19) | 12 (15) | 7 (9) | |
II | 97 (65) | 46 (56) | 37 (47) | |
III | 24 (16) | 24 (29) | 35 (44) | |
KCCQ score | 53 ± 22 | 50 ± 27 | 48 ± 29 | 0.1 |
Arterial hypertension | 112 (75) | 65 (79) | 57 (72) | 0.4 |
Stroke/TIA | 18 (12) | 8 (10) | 8 (10) | 0.5 |
Diabetes mellitus | 43 (29) | 25 (30) | 25 (32) | 0.2 |
COPD | 24 (16) | 16 (20) | 15 (19) | 0.3 |
CAD | 48 (32) | 29 (35) | 26 (33) | 0.3 |
Previous MI | 34 (23) | 21 (26) | 18 (23) | 0.2 |
Previous PCI/CABG | 42 (28) | 25 (30) | 22 (28) | 0.5 |
Pacemaker | 31 (21) | 18 (22) | 18 (23) | 0.3 |
ICD | 26 (17) | 12 (15) | 14 (18) | 0.6 |
CRT | 19 (13) | 9 (11) | 4 (5) | 0.1 |
Atrial fibrillation | 39 (26) | 38 (46)* | 45 (57)*,** | 0.01 |
Clinical evaluation (0 missing) | ||||
Brachial systolic BP, mmHg | 131 ± 22 | 130 ± 22 | 121 ± 19*,** | 0.001 |
Brachial diastolic BP, mmHg | 77 ± 14 | 76 ± 14 | 76 ± 13 | 0.8 |
Heart rate, b.p.m. | 72 ± 12 | 73 ± 16 | 73 ± 15 | 0.9 |
No clinical signs of congestion | 129 (87) | 61 (74)* | 34 (43)*,** | <0.0001 |
Pitting oedema (any degree) | 23 (15) | 21 (25) | 35 (44)*,** | 0.001 |
Lung crackles (any degree) | 4 (3) | 7 (9) | 23 (28)*,** | <0.0001 |
Jugular vein distension (any degree) | 0 | 6 (7) | 14 (18)*,** | <0.0001 |
Blood tests (0 missing) | ||||
Haemoglobin, g/dL (men) | 13.5 ± 1.7 | 13.5 ± 1.9 | 13.4 ± 1.6 | 0.1 |
Haemoglobin, g/dL (women) | 12.9 ± 1.8 | 12.8 ± 1.8 | 12.7 ± 2.1 | 0.1 |
Na+, mEq/L | 140 ± 3 | 141 ± 3 | 142 ± 3* | 0.005 |
K+, mEq/L | 4.3 ± 0.5 | 4.3 ± 0.4 | 4.3 ± 0.5 | 0.9 |
Total cholesterol, mg/dL | 163 ± 41 | 163 ± 39 | 164 ± 40 | 0.3 |
Fasting glucose, mg/dL | 98 ± 31 | 99 ± 28 | 102 ± 26 | 0.6 |
HbA1c, mmol/mol | 41 ± 10 | 41 ± 10 | 41 ± 9 | 0.9 |
Uric acid, mg/dL | 7.3 ± 1.9 | 7.7 ± 2.4 | 8.3 ± 2.9*,** | <0.0001 |
hs-CRP, mg/dL | 0.42 (0.24–0.60) | 0.44 (0.25–0.63) | 0.48 (0.27–0.69)* | 0.006 |
Creatinine, mg/dL | 1.01 (0.82–1.23) | 1.05 (0.91–1.30) | 1.08 (0.95–1.32) | 0.1 |
eGFR, mL/min/1.73 m2 | 65 (54–82) | 64 (52–77) | 61 (50–72) | 0.1 |
Urea, mg/dL | 47 (36–57) | 48 (37–58) | 54 (43–72)*,** | 0.001 |
ePVS, mL/g | 4.7 (4.1–5.3) | 4.8 (4.3–5.5) | 5.3 (4.7–6.2)*,** | <0.0001 |
Osmolality, mOsm/kg | 294 (289–298) | 294 (290–299) | 298 (292–301)*,** | <0.0001 |
Norepinephrine, pg/mL | 318 (232–384) | 332 (287–483) | 398 (313–592)*,** | <0.0001 |
Renin, mIU/L | 21.1 (6.1–51.4) | 23.7 (8.2–75.4) | 28.3 (9.1–123.1) | 0.1 |
Aldosterone, ng/dL | 9.5 (6.5–12.7) | 10.3 (7.1–13.8) | 15.1 (9.8 −23.1)*,** | <0.0001 |
NT-proBNP, pg/mL | 409 (308–862) | 936 (627–1671)* | 1862 (1379–2886)*,** | <0.0001 |
NT-proBNP, pg/mL (SR only) | 211 (160–543) | 336 (185–749) | 847 (561–1590)*,** | <0.0001 |
hs-Troponin T, pg/mL | 15 (11–24) | 18 (12–27) | 25 (15–46)*,** | <0.0001 |
Urine test (12 missing) | ||||
Osmolality, mOsm/kg | 540 (417–680) | 507 (408–630) | 456 (334–571)* | 0.001 |
UACR, mg/g | 21 (8–63) | 23 (9–65) | 57 (21–99)*,** | <0.0001 |
Micro-albuminuriaa | 38 (25) | 26 (32) | 51 (65) | <0.0001 |
Macro-albuminuriaa | 7 (5) | 5 (6) | 11 (14) | 0.04 |
Spot urinary sodium, mEq/L | 69 (48–118) | 63 (44–117) | 45 (30–76)*,** | <0.0001 |
FENa, % | 0.58 (0.40–0.91) | 0.53 (0.37–0.89) | 0.40 (0.20–0.66)*,** | <0.0001 |
Therapy (0 missing) | ||||
Beta-Blocker | 118 (79) | 66 (80) | 64 (81) | 0.5 |
DHP CCB | 34 (23) | 18 (22) | 9 (11) | 0.1 |
Non-DHP CCB | 0 | 2 (2) | 0 | 0.2 |
ACEi or ARB | 94 (63) | 50 (61) | 52 (66) | 0.3 |
MRA | 69 (46) | 43 (52) | 45 (57) | 0.2 |
ARNI | 37 (25) | 14 (17) | 15 (19) | 0.3 |
ASA | 67 (45) | 34 (41) | 37 (47) | 0.7 |
Statins | 83 (56) | 42 (51) | 41 (52) | 0.3 |
Thiazides/thiazide-like diuretics | 16 (11) | 11 (13) | 5 (6) | 0.4 |
Loop diuretics | 106 (71) | 62 (75) | 62 (78) | 0.1 |
Furosemide equivalent dose | 0.4 | |||
1–50 mg | 28 (19) | 14 (17) | 12 (15) | |
51–100 mg | 70 (47) | 39 (48) | 42 (53) | |
>100 mg | 8 (6) | 7 (8) | 10 (13) | |
SGLT2i | 19 (13) | 9 (11) | 8 (10) | 0.3 |
Insulin | 15 (10) | 6 (7) | 5 (6) | 0.2 |
Oral anticoagulants | 40 (27) | 39 (48)* | 46 (58)* | 0.01 |
Values are mean ± standard deviation, n (%), or median (25th quartile, 75th quartile).
ACEi, angiotensin-converting enzyme inhibitor; ARB, angiotensin receptor blocker; ARNI, angiotensin receptor neprilysin inhibitor; ASA, acetylsalicylic acid; BMI, body mass index; BSA, body surface area; CABG, coronary artery bypass graft; CAD, coronary artery disease; COPD, chronic obstructive pulmonary disease; CRT, cardiac resynchronization therapy; DHP CCB, dihydropyridine calcium channel blocker; eGFR, estimated glomerular filtration rate; ePVS, estimated plasma volume status; FENa, fractional excretion of sodium; HbA1c, glycated haemoglobin (available only in patients with diabetes mellitus); HFpEF, heart failure with preserved ejection fraction; HFrEF, heart failure with reduced ejection fraction; hs-CRP, high sensitivity C-reactive protein; ICD, implantable cardioverter defibrillator; KCCQ, Kansas City Cardiomyopathy Questionnaire; MI, myocardial infarction; MRA, mineralocorticoid receptor antagonist; NT-proBNP, N-terminal prohormone of brain natriuretic peptide; PCI, percutaneous coronary intervention; SR, sinus rhythm; SGLT2i, sodium glucose co-transporter 2 inhibitors; TIA, transient ischaemic attack; UACR, albumin-to-creatinine ratio.
Micro-albuminuria and macro-albuminuria were defined as UACR >30 mg/g and >300 mg/g, respectively.
*P < 0.01 vs No US congestion.
**P < 0.01 vs 1 US sign of congestion.
Variable . | HF w/o US signs of congestion (n = 149) . | HF with 1 US sign of congestion (n = 82) . | HF with ≥2 US signs of congestion (n = 79) . | P-value . |
---|---|---|---|---|
Demographics (0 missing) | ||||
Age, years | 73 (64–81) | 75 (69–82)* | 78 (72–85)* | 0.001 |
Men | 89 (60) | 56 (68) | 56 (71) | 0.2 |
BMI, kg/m2 | 27.5 ± 4.7 | 28.4 ± 5.1 | 28.6 ± 5.6 | 0.2 |
BSA, m2 | 1.9 ± 0.2 | 1.9 ± 0.3 | 1.9 ± 0.3 | 0.9 |
Smoker | 27 (18) | 16 (20) | 18 (23) | 0.3 |
NYHA class | 0.001 | |||
I | 28 (19) | 12 (15) | 7 (9) | |
II | 97 (65) | 46 (56) | 37 (47) | |
III | 24 (16) | 24 (29) | 35 (44) | |
KCCQ score | 53 ± 22 | 50 ± 27 | 48 ± 29 | 0.1 |
Arterial hypertension | 112 (75) | 65 (79) | 57 (72) | 0.4 |
Stroke/TIA | 18 (12) | 8 (10) | 8 (10) | 0.5 |
Diabetes mellitus | 43 (29) | 25 (30) | 25 (32) | 0.2 |
COPD | 24 (16) | 16 (20) | 15 (19) | 0.3 |
CAD | 48 (32) | 29 (35) | 26 (33) | 0.3 |
Previous MI | 34 (23) | 21 (26) | 18 (23) | 0.2 |
Previous PCI/CABG | 42 (28) | 25 (30) | 22 (28) | 0.5 |
Pacemaker | 31 (21) | 18 (22) | 18 (23) | 0.3 |
ICD | 26 (17) | 12 (15) | 14 (18) | 0.6 |
CRT | 19 (13) | 9 (11) | 4 (5) | 0.1 |
Atrial fibrillation | 39 (26) | 38 (46)* | 45 (57)*,** | 0.01 |
Clinical evaluation (0 missing) | ||||
Brachial systolic BP, mmHg | 131 ± 22 | 130 ± 22 | 121 ± 19*,** | 0.001 |
Brachial diastolic BP, mmHg | 77 ± 14 | 76 ± 14 | 76 ± 13 | 0.8 |
Heart rate, b.p.m. | 72 ± 12 | 73 ± 16 | 73 ± 15 | 0.9 |
No clinical signs of congestion | 129 (87) | 61 (74)* | 34 (43)*,** | <0.0001 |
Pitting oedema (any degree) | 23 (15) | 21 (25) | 35 (44)*,** | 0.001 |
Lung crackles (any degree) | 4 (3) | 7 (9) | 23 (28)*,** | <0.0001 |
Jugular vein distension (any degree) | 0 | 6 (7) | 14 (18)*,** | <0.0001 |
Blood tests (0 missing) | ||||
Haemoglobin, g/dL (men) | 13.5 ± 1.7 | 13.5 ± 1.9 | 13.4 ± 1.6 | 0.1 |
Haemoglobin, g/dL (women) | 12.9 ± 1.8 | 12.8 ± 1.8 | 12.7 ± 2.1 | 0.1 |
Na+, mEq/L | 140 ± 3 | 141 ± 3 | 142 ± 3* | 0.005 |
K+, mEq/L | 4.3 ± 0.5 | 4.3 ± 0.4 | 4.3 ± 0.5 | 0.9 |
Total cholesterol, mg/dL | 163 ± 41 | 163 ± 39 | 164 ± 40 | 0.3 |
Fasting glucose, mg/dL | 98 ± 31 | 99 ± 28 | 102 ± 26 | 0.6 |
HbA1c, mmol/mol | 41 ± 10 | 41 ± 10 | 41 ± 9 | 0.9 |
Uric acid, mg/dL | 7.3 ± 1.9 | 7.7 ± 2.4 | 8.3 ± 2.9*,** | <0.0001 |
hs-CRP, mg/dL | 0.42 (0.24–0.60) | 0.44 (0.25–0.63) | 0.48 (0.27–0.69)* | 0.006 |
Creatinine, mg/dL | 1.01 (0.82–1.23) | 1.05 (0.91–1.30) | 1.08 (0.95–1.32) | 0.1 |
eGFR, mL/min/1.73 m2 | 65 (54–82) | 64 (52–77) | 61 (50–72) | 0.1 |
Urea, mg/dL | 47 (36–57) | 48 (37–58) | 54 (43–72)*,** | 0.001 |
ePVS, mL/g | 4.7 (4.1–5.3) | 4.8 (4.3–5.5) | 5.3 (4.7–6.2)*,** | <0.0001 |
Osmolality, mOsm/kg | 294 (289–298) | 294 (290–299) | 298 (292–301)*,** | <0.0001 |
Norepinephrine, pg/mL | 318 (232–384) | 332 (287–483) | 398 (313–592)*,** | <0.0001 |
Renin, mIU/L | 21.1 (6.1–51.4) | 23.7 (8.2–75.4) | 28.3 (9.1–123.1) | 0.1 |
Aldosterone, ng/dL | 9.5 (6.5–12.7) | 10.3 (7.1–13.8) | 15.1 (9.8 −23.1)*,** | <0.0001 |
NT-proBNP, pg/mL | 409 (308–862) | 936 (627–1671)* | 1862 (1379–2886)*,** | <0.0001 |
NT-proBNP, pg/mL (SR only) | 211 (160–543) | 336 (185–749) | 847 (561–1590)*,** | <0.0001 |
hs-Troponin T, pg/mL | 15 (11–24) | 18 (12–27) | 25 (15–46)*,** | <0.0001 |
Urine test (12 missing) | ||||
Osmolality, mOsm/kg | 540 (417–680) | 507 (408–630) | 456 (334–571)* | 0.001 |
UACR, mg/g | 21 (8–63) | 23 (9–65) | 57 (21–99)*,** | <0.0001 |
Micro-albuminuriaa | 38 (25) | 26 (32) | 51 (65) | <0.0001 |
Macro-albuminuriaa | 7 (5) | 5 (6) | 11 (14) | 0.04 |
Spot urinary sodium, mEq/L | 69 (48–118) | 63 (44–117) | 45 (30–76)*,** | <0.0001 |
FENa, % | 0.58 (0.40–0.91) | 0.53 (0.37–0.89) | 0.40 (0.20–0.66)*,** | <0.0001 |
Therapy (0 missing) | ||||
Beta-Blocker | 118 (79) | 66 (80) | 64 (81) | 0.5 |
DHP CCB | 34 (23) | 18 (22) | 9 (11) | 0.1 |
Non-DHP CCB | 0 | 2 (2) | 0 | 0.2 |
ACEi or ARB | 94 (63) | 50 (61) | 52 (66) | 0.3 |
MRA | 69 (46) | 43 (52) | 45 (57) | 0.2 |
ARNI | 37 (25) | 14 (17) | 15 (19) | 0.3 |
ASA | 67 (45) | 34 (41) | 37 (47) | 0.7 |
Statins | 83 (56) | 42 (51) | 41 (52) | 0.3 |
Thiazides/thiazide-like diuretics | 16 (11) | 11 (13) | 5 (6) | 0.4 |
Loop diuretics | 106 (71) | 62 (75) | 62 (78) | 0.1 |
Furosemide equivalent dose | 0.4 | |||
1–50 mg | 28 (19) | 14 (17) | 12 (15) | |
51–100 mg | 70 (47) | 39 (48) | 42 (53) | |
>100 mg | 8 (6) | 7 (8) | 10 (13) | |
SGLT2i | 19 (13) | 9 (11) | 8 (10) | 0.3 |
Insulin | 15 (10) | 6 (7) | 5 (6) | 0.2 |
Oral anticoagulants | 40 (27) | 39 (48)* | 46 (58)* | 0.01 |
Variable . | HF w/o US signs of congestion (n = 149) . | HF with 1 US sign of congestion (n = 82) . | HF with ≥2 US signs of congestion (n = 79) . | P-value . |
---|---|---|---|---|
Demographics (0 missing) | ||||
Age, years | 73 (64–81) | 75 (69–82)* | 78 (72–85)* | 0.001 |
Men | 89 (60) | 56 (68) | 56 (71) | 0.2 |
BMI, kg/m2 | 27.5 ± 4.7 | 28.4 ± 5.1 | 28.6 ± 5.6 | 0.2 |
BSA, m2 | 1.9 ± 0.2 | 1.9 ± 0.3 | 1.9 ± 0.3 | 0.9 |
Smoker | 27 (18) | 16 (20) | 18 (23) | 0.3 |
NYHA class | 0.001 | |||
I | 28 (19) | 12 (15) | 7 (9) | |
II | 97 (65) | 46 (56) | 37 (47) | |
III | 24 (16) | 24 (29) | 35 (44) | |
KCCQ score | 53 ± 22 | 50 ± 27 | 48 ± 29 | 0.1 |
Arterial hypertension | 112 (75) | 65 (79) | 57 (72) | 0.4 |
Stroke/TIA | 18 (12) | 8 (10) | 8 (10) | 0.5 |
Diabetes mellitus | 43 (29) | 25 (30) | 25 (32) | 0.2 |
COPD | 24 (16) | 16 (20) | 15 (19) | 0.3 |
CAD | 48 (32) | 29 (35) | 26 (33) | 0.3 |
Previous MI | 34 (23) | 21 (26) | 18 (23) | 0.2 |
Previous PCI/CABG | 42 (28) | 25 (30) | 22 (28) | 0.5 |
Pacemaker | 31 (21) | 18 (22) | 18 (23) | 0.3 |
ICD | 26 (17) | 12 (15) | 14 (18) | 0.6 |
CRT | 19 (13) | 9 (11) | 4 (5) | 0.1 |
Atrial fibrillation | 39 (26) | 38 (46)* | 45 (57)*,** | 0.01 |
Clinical evaluation (0 missing) | ||||
Brachial systolic BP, mmHg | 131 ± 22 | 130 ± 22 | 121 ± 19*,** | 0.001 |
Brachial diastolic BP, mmHg | 77 ± 14 | 76 ± 14 | 76 ± 13 | 0.8 |
Heart rate, b.p.m. | 72 ± 12 | 73 ± 16 | 73 ± 15 | 0.9 |
No clinical signs of congestion | 129 (87) | 61 (74)* | 34 (43)*,** | <0.0001 |
Pitting oedema (any degree) | 23 (15) | 21 (25) | 35 (44)*,** | 0.001 |
Lung crackles (any degree) | 4 (3) | 7 (9) | 23 (28)*,** | <0.0001 |
Jugular vein distension (any degree) | 0 | 6 (7) | 14 (18)*,** | <0.0001 |
Blood tests (0 missing) | ||||
Haemoglobin, g/dL (men) | 13.5 ± 1.7 | 13.5 ± 1.9 | 13.4 ± 1.6 | 0.1 |
Haemoglobin, g/dL (women) | 12.9 ± 1.8 | 12.8 ± 1.8 | 12.7 ± 2.1 | 0.1 |
Na+, mEq/L | 140 ± 3 | 141 ± 3 | 142 ± 3* | 0.005 |
K+, mEq/L | 4.3 ± 0.5 | 4.3 ± 0.4 | 4.3 ± 0.5 | 0.9 |
Total cholesterol, mg/dL | 163 ± 41 | 163 ± 39 | 164 ± 40 | 0.3 |
Fasting glucose, mg/dL | 98 ± 31 | 99 ± 28 | 102 ± 26 | 0.6 |
HbA1c, mmol/mol | 41 ± 10 | 41 ± 10 | 41 ± 9 | 0.9 |
Uric acid, mg/dL | 7.3 ± 1.9 | 7.7 ± 2.4 | 8.3 ± 2.9*,** | <0.0001 |
hs-CRP, mg/dL | 0.42 (0.24–0.60) | 0.44 (0.25–0.63) | 0.48 (0.27–0.69)* | 0.006 |
Creatinine, mg/dL | 1.01 (0.82–1.23) | 1.05 (0.91–1.30) | 1.08 (0.95–1.32) | 0.1 |
eGFR, mL/min/1.73 m2 | 65 (54–82) | 64 (52–77) | 61 (50–72) | 0.1 |
Urea, mg/dL | 47 (36–57) | 48 (37–58) | 54 (43–72)*,** | 0.001 |
ePVS, mL/g | 4.7 (4.1–5.3) | 4.8 (4.3–5.5) | 5.3 (4.7–6.2)*,** | <0.0001 |
Osmolality, mOsm/kg | 294 (289–298) | 294 (290–299) | 298 (292–301)*,** | <0.0001 |
Norepinephrine, pg/mL | 318 (232–384) | 332 (287–483) | 398 (313–592)*,** | <0.0001 |
Renin, mIU/L | 21.1 (6.1–51.4) | 23.7 (8.2–75.4) | 28.3 (9.1–123.1) | 0.1 |
Aldosterone, ng/dL | 9.5 (6.5–12.7) | 10.3 (7.1–13.8) | 15.1 (9.8 −23.1)*,** | <0.0001 |
NT-proBNP, pg/mL | 409 (308–862) | 936 (627–1671)* | 1862 (1379–2886)*,** | <0.0001 |
NT-proBNP, pg/mL (SR only) | 211 (160–543) | 336 (185–749) | 847 (561–1590)*,** | <0.0001 |
hs-Troponin T, pg/mL | 15 (11–24) | 18 (12–27) | 25 (15–46)*,** | <0.0001 |
Urine test (12 missing) | ||||
Osmolality, mOsm/kg | 540 (417–680) | 507 (408–630) | 456 (334–571)* | 0.001 |
UACR, mg/g | 21 (8–63) | 23 (9–65) | 57 (21–99)*,** | <0.0001 |
Micro-albuminuriaa | 38 (25) | 26 (32) | 51 (65) | <0.0001 |
Macro-albuminuriaa | 7 (5) | 5 (6) | 11 (14) | 0.04 |
Spot urinary sodium, mEq/L | 69 (48–118) | 63 (44–117) | 45 (30–76)*,** | <0.0001 |
FENa, % | 0.58 (0.40–0.91) | 0.53 (0.37–0.89) | 0.40 (0.20–0.66)*,** | <0.0001 |
Therapy (0 missing) | ||||
Beta-Blocker | 118 (79) | 66 (80) | 64 (81) | 0.5 |
DHP CCB | 34 (23) | 18 (22) | 9 (11) | 0.1 |
Non-DHP CCB | 0 | 2 (2) | 0 | 0.2 |
ACEi or ARB | 94 (63) | 50 (61) | 52 (66) | 0.3 |
MRA | 69 (46) | 43 (52) | 45 (57) | 0.2 |
ARNI | 37 (25) | 14 (17) | 15 (19) | 0.3 |
ASA | 67 (45) | 34 (41) | 37 (47) | 0.7 |
Statins | 83 (56) | 42 (51) | 41 (52) | 0.3 |
Thiazides/thiazide-like diuretics | 16 (11) | 11 (13) | 5 (6) | 0.4 |
Loop diuretics | 106 (71) | 62 (75) | 62 (78) | 0.1 |
Furosemide equivalent dose | 0.4 | |||
1–50 mg | 28 (19) | 14 (17) | 12 (15) | |
51–100 mg | 70 (47) | 39 (48) | 42 (53) | |
>100 mg | 8 (6) | 7 (8) | 10 (13) | |
SGLT2i | 19 (13) | 9 (11) | 8 (10) | 0.3 |
Insulin | 15 (10) | 6 (7) | 5 (6) | 0.2 |
Oral anticoagulants | 40 (27) | 39 (48)* | 46 (58)* | 0.01 |
Values are mean ± standard deviation, n (%), or median (25th quartile, 75th quartile).
ACEi, angiotensin-converting enzyme inhibitor; ARB, angiotensin receptor blocker; ARNI, angiotensin receptor neprilysin inhibitor; ASA, acetylsalicylic acid; BMI, body mass index; BSA, body surface area; CABG, coronary artery bypass graft; CAD, coronary artery disease; COPD, chronic obstructive pulmonary disease; CRT, cardiac resynchronization therapy; DHP CCB, dihydropyridine calcium channel blocker; eGFR, estimated glomerular filtration rate; ePVS, estimated plasma volume status; FENa, fractional excretion of sodium; HbA1c, glycated haemoglobin (available only in patients with diabetes mellitus); HFpEF, heart failure with preserved ejection fraction; HFrEF, heart failure with reduced ejection fraction; hs-CRP, high sensitivity C-reactive protein; ICD, implantable cardioverter defibrillator; KCCQ, Kansas City Cardiomyopathy Questionnaire; MI, myocardial infarction; MRA, mineralocorticoid receptor antagonist; NT-proBNP, N-terminal prohormone of brain natriuretic peptide; PCI, percutaneous coronary intervention; SR, sinus rhythm; SGLT2i, sodium glucose co-transporter 2 inhibitors; TIA, transient ischaemic attack; UACR, albumin-to-creatinine ratio.
Micro-albuminuria and macro-albuminuria were defined as UACR >30 mg/g and >300 mg/g, respectively.
*P < 0.01 vs No US congestion.
**P < 0.01 vs 1 US sign of congestion.
Variable . | Missing . | HF w/o US signs of congestion (n = 149) . | HF with 1 US sign of congestion (n = 82) . | HF with ≥2 US signs of congestion (n = 79) . | P-value . |
---|---|---|---|---|---|
Left ventricle size and function | |||||
LVMi, g/m2 | 0 | 131 ± 38 | 137 ± 35 | 140 ± 40 | 0.3 |
RWT | 0 | 0.36 ± 0.10 | 0.36 ± 0.10 | 0.38 ± 0.13 | 0.5 |
LVEDV, mL | 0 | 171 ± 58 | 174 ± 53 | 182 ± 61 | 0.6 |
LV Ejection fraction, % | 0 | 50 ± 14 | 52 ± 14 | 47 ± 14 | 0.2 |
3D-LV Ejection fraction, % | 15 | 48 ± 11 | 49 ± 13 | 44 ± 12 | 0.1 |
LVGLS, % | 5 | −14.1 ± 4.2 | −13.9 ± 4.9 | −11.6 ± 4.7*,** | 0.001 |
Stroke volume, mL/beat | 0 | 61 ± 32 | 59 ± 28 | 54 ± 19 | 0.4 |
Cardiac output, L/min | 0 | 4.6 ± 2.2 | 4.4 ± 1.6 | 3.9 ± 1.3 | 0.1 |
Mitral E wave, cm/s | 0 | 93 ± 44 | 111 ± 48 | 127 ± 43* | <0.0001 |
Average e′, cm/s | 0 | 7.3 ± 2.3 | 7.3 ± 2.2 | 8.9 ± 3.1*,** | 0.002 |
Average E/e′ | 0 | 13.6 ± 5.6 | 16.1 ± 6.2 | 16.9 ± 6.6* | 0.004 |
Mitral regurgitation (≥ moderate) | 0 | 14 (9) | 8 (10) | 10 (13) | 0.2 |
Left atrium size and function | |||||
LAVi, mL/m2 | 0 | 45 ± 21 | 49 ± 18 | 62 ± 18*,** | <0.0001 |
LA reservoir strain, % | 0 | −20 ± 8 | −17 ± 10 | −12 ± 6*,** | <0.0001 |
Right ventricle and pulmonary circulation | |||||
TAPSE, mm | 0 | 19 ± 4 | 19 ± 4 | 18 ± 4 | 0.1 |
RVFAC, % | 0 | 50 ± 10 | 49 ± 11 | 42 ± 9*,** | <0.0001 |
RV free wall longitudinal strain, % | 19 | −28 ± 6 | −27 ± 6 | −24 ± 6 | 0.03 |
3D-RVEDV | 19 | 137 ± 34 | 151 ± 39 | 165 ± 50*,** | 0.001 |
3D-RV Ejection fraction, % | 19 | 56 ± 10 | 55 ± 13 | 46 ± 10*,** | <0.0001 |
Tricuspid regurgitation (≥ moderate) | 0 | 11 (7) | 8 (10) | 8 (10) | 0.3 |
Systolic PAP, mmHg | 8 | 37 ± 13 | 45 ± 15* | 53 ± 14*,** | <0.0001 |
Diastolic PAP, mmHg | 10 | 9 ± 4 | 12 ± 7* | 17 ± 7*,** | <0.0001 |
Mean PAP, mmHg | 10 | 19 ± 6 | 24 ± 8* | 30 ± 9*,** | <0.0001 |
ePVR, WU | 10 | 1.5 ± 0.7 | 1.9 ± 0.9* | 2.1 ± 0.7* | <0.0001 |
ePAWP, mmHg | 10 | 12 ± 4 | 15 ± 6* | 20 ± 5*,** | <0.0001 |
Congestion assessment | 0 | ||||
IVC, mm | 17 (15–18) | 18 (15–21)* | 23 (21–27)*,** | <0.0001 | |
IVC ≥21 mm | 0 | 7 (8) | 52 (66)*,** | <0.0001 | |
IVC collapse <50% | 3 (2) | 15 (18)* | 21 (27)*,** | <0.0001 | |
B-lines | 0 (0–2) | 7 (4–14)* | 13 (8–28)*,** | <0.0001 | |
B-lines ≥4 | 0 | 43 (52)* | 62 (78)*,** | <0.0001 | |
RVF pattern | <0.0001 | ||||
Continuous | 149 (100) | 50 (61)* | 17 (21)*,** | ||
Discontinuous: pulsatile | 0 | 29 (35)* | 11 (14)*,** | ||
Discontinuous: biphasic | 0 | 3 (4) | 32 (41)*,** | ||
Discontinuous: monophasic | 0 | 0 | 19 (24)*,** | ||
Renal venous impedance index | 0.2 (0.1–0.3) | 0.3 (0.2–1)* | 1 (1–1)*,** | <0.0001 | |
Renal venous discontinuity index, %a | — | 12 (10–25) | 29 (14–66) | 0.01 |
Variable . | Missing . | HF w/o US signs of congestion (n = 149) . | HF with 1 US sign of congestion (n = 82) . | HF with ≥2 US signs of congestion (n = 79) . | P-value . |
---|---|---|---|---|---|
Left ventricle size and function | |||||
LVMi, g/m2 | 0 | 131 ± 38 | 137 ± 35 | 140 ± 40 | 0.3 |
RWT | 0 | 0.36 ± 0.10 | 0.36 ± 0.10 | 0.38 ± 0.13 | 0.5 |
LVEDV, mL | 0 | 171 ± 58 | 174 ± 53 | 182 ± 61 | 0.6 |
LV Ejection fraction, % | 0 | 50 ± 14 | 52 ± 14 | 47 ± 14 | 0.2 |
3D-LV Ejection fraction, % | 15 | 48 ± 11 | 49 ± 13 | 44 ± 12 | 0.1 |
LVGLS, % | 5 | −14.1 ± 4.2 | −13.9 ± 4.9 | −11.6 ± 4.7*,** | 0.001 |
Stroke volume, mL/beat | 0 | 61 ± 32 | 59 ± 28 | 54 ± 19 | 0.4 |
Cardiac output, L/min | 0 | 4.6 ± 2.2 | 4.4 ± 1.6 | 3.9 ± 1.3 | 0.1 |
Mitral E wave, cm/s | 0 | 93 ± 44 | 111 ± 48 | 127 ± 43* | <0.0001 |
Average e′, cm/s | 0 | 7.3 ± 2.3 | 7.3 ± 2.2 | 8.9 ± 3.1*,** | 0.002 |
Average E/e′ | 0 | 13.6 ± 5.6 | 16.1 ± 6.2 | 16.9 ± 6.6* | 0.004 |
Mitral regurgitation (≥ moderate) | 0 | 14 (9) | 8 (10) | 10 (13) | 0.2 |
Left atrium size and function | |||||
LAVi, mL/m2 | 0 | 45 ± 21 | 49 ± 18 | 62 ± 18*,** | <0.0001 |
LA reservoir strain, % | 0 | −20 ± 8 | −17 ± 10 | −12 ± 6*,** | <0.0001 |
Right ventricle and pulmonary circulation | |||||
TAPSE, mm | 0 | 19 ± 4 | 19 ± 4 | 18 ± 4 | 0.1 |
RVFAC, % | 0 | 50 ± 10 | 49 ± 11 | 42 ± 9*,** | <0.0001 |
RV free wall longitudinal strain, % | 19 | −28 ± 6 | −27 ± 6 | −24 ± 6 | 0.03 |
3D-RVEDV | 19 | 137 ± 34 | 151 ± 39 | 165 ± 50*,** | 0.001 |
3D-RV Ejection fraction, % | 19 | 56 ± 10 | 55 ± 13 | 46 ± 10*,** | <0.0001 |
Tricuspid regurgitation (≥ moderate) | 0 | 11 (7) | 8 (10) | 8 (10) | 0.3 |
Systolic PAP, mmHg | 8 | 37 ± 13 | 45 ± 15* | 53 ± 14*,** | <0.0001 |
Diastolic PAP, mmHg | 10 | 9 ± 4 | 12 ± 7* | 17 ± 7*,** | <0.0001 |
Mean PAP, mmHg | 10 | 19 ± 6 | 24 ± 8* | 30 ± 9*,** | <0.0001 |
ePVR, WU | 10 | 1.5 ± 0.7 | 1.9 ± 0.9* | 2.1 ± 0.7* | <0.0001 |
ePAWP, mmHg | 10 | 12 ± 4 | 15 ± 6* | 20 ± 5*,** | <0.0001 |
Congestion assessment | 0 | ||||
IVC, mm | 17 (15–18) | 18 (15–21)* | 23 (21–27)*,** | <0.0001 | |
IVC ≥21 mm | 0 | 7 (8) | 52 (66)*,** | <0.0001 | |
IVC collapse <50% | 3 (2) | 15 (18)* | 21 (27)*,** | <0.0001 | |
B-lines | 0 (0–2) | 7 (4–14)* | 13 (8–28)*,** | <0.0001 | |
B-lines ≥4 | 0 | 43 (52)* | 62 (78)*,** | <0.0001 | |
RVF pattern | <0.0001 | ||||
Continuous | 149 (100) | 50 (61)* | 17 (21)*,** | ||
Discontinuous: pulsatile | 0 | 29 (35)* | 11 (14)*,** | ||
Discontinuous: biphasic | 0 | 3 (4) | 32 (41)*,** | ||
Discontinuous: monophasic | 0 | 0 | 19 (24)*,** | ||
Renal venous impedance index | 0.2 (0.1–0.3) | 0.3 (0.2–1)* | 1 (1–1)*,** | <0.0001 | |
Renal venous discontinuity index, %a | — | 12 (10–25) | 29 (14–66) | 0.01 |
Values are mean ± standard deviation, n (%), or median (25th quartile, 75th quartile).
EDV, end-diastolic volume; ePAWP, echo-derived pulmonary artery wedge pressure; ePVR, echo-derived pulmonary vascular resistance; GLS, global longitudinal strain; HFpEF, heart failure with preserved ejection fraction; HFrEF, heart failure with reduced ejection fraction; IVC, inferior vena cava; LA, left atrium; LAVi, left atrial volume; LV, left ventricle; LVMi, left ventricle mass index; PAP, pulmonary artery pressure; RVF, renal venous flow; RWT, relative wall thickness; TAPSE, tricuspid annular plane systolic excursion.
Measured only in patients with the discontinuous renal venous flow (n = 94).
*P < 0.01 vs No US congestion.
**P < 0.01 vs 1 US sign of congestion.
Variable . | Missing . | HF w/o US signs of congestion (n = 149) . | HF with 1 US sign of congestion (n = 82) . | HF with ≥2 US signs of congestion (n = 79) . | P-value . |
---|---|---|---|---|---|
Left ventricle size and function | |||||
LVMi, g/m2 | 0 | 131 ± 38 | 137 ± 35 | 140 ± 40 | 0.3 |
RWT | 0 | 0.36 ± 0.10 | 0.36 ± 0.10 | 0.38 ± 0.13 | 0.5 |
LVEDV, mL | 0 | 171 ± 58 | 174 ± 53 | 182 ± 61 | 0.6 |
LV Ejection fraction, % | 0 | 50 ± 14 | 52 ± 14 | 47 ± 14 | 0.2 |
3D-LV Ejection fraction, % | 15 | 48 ± 11 | 49 ± 13 | 44 ± 12 | 0.1 |
LVGLS, % | 5 | −14.1 ± 4.2 | −13.9 ± 4.9 | −11.6 ± 4.7*,** | 0.001 |
Stroke volume, mL/beat | 0 | 61 ± 32 | 59 ± 28 | 54 ± 19 | 0.4 |
Cardiac output, L/min | 0 | 4.6 ± 2.2 | 4.4 ± 1.6 | 3.9 ± 1.3 | 0.1 |
Mitral E wave, cm/s | 0 | 93 ± 44 | 111 ± 48 | 127 ± 43* | <0.0001 |
Average e′, cm/s | 0 | 7.3 ± 2.3 | 7.3 ± 2.2 | 8.9 ± 3.1*,** | 0.002 |
Average E/e′ | 0 | 13.6 ± 5.6 | 16.1 ± 6.2 | 16.9 ± 6.6* | 0.004 |
Mitral regurgitation (≥ moderate) | 0 | 14 (9) | 8 (10) | 10 (13) | 0.2 |
Left atrium size and function | |||||
LAVi, mL/m2 | 0 | 45 ± 21 | 49 ± 18 | 62 ± 18*,** | <0.0001 |
LA reservoir strain, % | 0 | −20 ± 8 | −17 ± 10 | −12 ± 6*,** | <0.0001 |
Right ventricle and pulmonary circulation | |||||
TAPSE, mm | 0 | 19 ± 4 | 19 ± 4 | 18 ± 4 | 0.1 |
RVFAC, % | 0 | 50 ± 10 | 49 ± 11 | 42 ± 9*,** | <0.0001 |
RV free wall longitudinal strain, % | 19 | −28 ± 6 | −27 ± 6 | −24 ± 6 | 0.03 |
3D-RVEDV | 19 | 137 ± 34 | 151 ± 39 | 165 ± 50*,** | 0.001 |
3D-RV Ejection fraction, % | 19 | 56 ± 10 | 55 ± 13 | 46 ± 10*,** | <0.0001 |
Tricuspid regurgitation (≥ moderate) | 0 | 11 (7) | 8 (10) | 8 (10) | 0.3 |
Systolic PAP, mmHg | 8 | 37 ± 13 | 45 ± 15* | 53 ± 14*,** | <0.0001 |
Diastolic PAP, mmHg | 10 | 9 ± 4 | 12 ± 7* | 17 ± 7*,** | <0.0001 |
Mean PAP, mmHg | 10 | 19 ± 6 | 24 ± 8* | 30 ± 9*,** | <0.0001 |
ePVR, WU | 10 | 1.5 ± 0.7 | 1.9 ± 0.9* | 2.1 ± 0.7* | <0.0001 |
ePAWP, mmHg | 10 | 12 ± 4 | 15 ± 6* | 20 ± 5*,** | <0.0001 |
Congestion assessment | 0 | ||||
IVC, mm | 17 (15–18) | 18 (15–21)* | 23 (21–27)*,** | <0.0001 | |
IVC ≥21 mm | 0 | 7 (8) | 52 (66)*,** | <0.0001 | |
IVC collapse <50% | 3 (2) | 15 (18)* | 21 (27)*,** | <0.0001 | |
B-lines | 0 (0–2) | 7 (4–14)* | 13 (8–28)*,** | <0.0001 | |
B-lines ≥4 | 0 | 43 (52)* | 62 (78)*,** | <0.0001 | |
RVF pattern | <0.0001 | ||||
Continuous | 149 (100) | 50 (61)* | 17 (21)*,** | ||
Discontinuous: pulsatile | 0 | 29 (35)* | 11 (14)*,** | ||
Discontinuous: biphasic | 0 | 3 (4) | 32 (41)*,** | ||
Discontinuous: monophasic | 0 | 0 | 19 (24)*,** | ||
Renal venous impedance index | 0.2 (0.1–0.3) | 0.3 (0.2–1)* | 1 (1–1)*,** | <0.0001 | |
Renal venous discontinuity index, %a | — | 12 (10–25) | 29 (14–66) | 0.01 |
Variable . | Missing . | HF w/o US signs of congestion (n = 149) . | HF with 1 US sign of congestion (n = 82) . | HF with ≥2 US signs of congestion (n = 79) . | P-value . |
---|---|---|---|---|---|
Left ventricle size and function | |||||
LVMi, g/m2 | 0 | 131 ± 38 | 137 ± 35 | 140 ± 40 | 0.3 |
RWT | 0 | 0.36 ± 0.10 | 0.36 ± 0.10 | 0.38 ± 0.13 | 0.5 |
LVEDV, mL | 0 | 171 ± 58 | 174 ± 53 | 182 ± 61 | 0.6 |
LV Ejection fraction, % | 0 | 50 ± 14 | 52 ± 14 | 47 ± 14 | 0.2 |
3D-LV Ejection fraction, % | 15 | 48 ± 11 | 49 ± 13 | 44 ± 12 | 0.1 |
LVGLS, % | 5 | −14.1 ± 4.2 | −13.9 ± 4.9 | −11.6 ± 4.7*,** | 0.001 |
Stroke volume, mL/beat | 0 | 61 ± 32 | 59 ± 28 | 54 ± 19 | 0.4 |
Cardiac output, L/min | 0 | 4.6 ± 2.2 | 4.4 ± 1.6 | 3.9 ± 1.3 | 0.1 |
Mitral E wave, cm/s | 0 | 93 ± 44 | 111 ± 48 | 127 ± 43* | <0.0001 |
Average e′, cm/s | 0 | 7.3 ± 2.3 | 7.3 ± 2.2 | 8.9 ± 3.1*,** | 0.002 |
Average E/e′ | 0 | 13.6 ± 5.6 | 16.1 ± 6.2 | 16.9 ± 6.6* | 0.004 |
Mitral regurgitation (≥ moderate) | 0 | 14 (9) | 8 (10) | 10 (13) | 0.2 |
Left atrium size and function | |||||
LAVi, mL/m2 | 0 | 45 ± 21 | 49 ± 18 | 62 ± 18*,** | <0.0001 |
LA reservoir strain, % | 0 | −20 ± 8 | −17 ± 10 | −12 ± 6*,** | <0.0001 |
Right ventricle and pulmonary circulation | |||||
TAPSE, mm | 0 | 19 ± 4 | 19 ± 4 | 18 ± 4 | 0.1 |
RVFAC, % | 0 | 50 ± 10 | 49 ± 11 | 42 ± 9*,** | <0.0001 |
RV free wall longitudinal strain, % | 19 | −28 ± 6 | −27 ± 6 | −24 ± 6 | 0.03 |
3D-RVEDV | 19 | 137 ± 34 | 151 ± 39 | 165 ± 50*,** | 0.001 |
3D-RV Ejection fraction, % | 19 | 56 ± 10 | 55 ± 13 | 46 ± 10*,** | <0.0001 |
Tricuspid regurgitation (≥ moderate) | 0 | 11 (7) | 8 (10) | 8 (10) | 0.3 |
Systolic PAP, mmHg | 8 | 37 ± 13 | 45 ± 15* | 53 ± 14*,** | <0.0001 |
Diastolic PAP, mmHg | 10 | 9 ± 4 | 12 ± 7* | 17 ± 7*,** | <0.0001 |
Mean PAP, mmHg | 10 | 19 ± 6 | 24 ± 8* | 30 ± 9*,** | <0.0001 |
ePVR, WU | 10 | 1.5 ± 0.7 | 1.9 ± 0.9* | 2.1 ± 0.7* | <0.0001 |
ePAWP, mmHg | 10 | 12 ± 4 | 15 ± 6* | 20 ± 5*,** | <0.0001 |
Congestion assessment | 0 | ||||
IVC, mm | 17 (15–18) | 18 (15–21)* | 23 (21–27)*,** | <0.0001 | |
IVC ≥21 mm | 0 | 7 (8) | 52 (66)*,** | <0.0001 | |
IVC collapse <50% | 3 (2) | 15 (18)* | 21 (27)*,** | <0.0001 | |
B-lines | 0 (0–2) | 7 (4–14)* | 13 (8–28)*,** | <0.0001 | |
B-lines ≥4 | 0 | 43 (52)* | 62 (78)*,** | <0.0001 | |
RVF pattern | <0.0001 | ||||
Continuous | 149 (100) | 50 (61)* | 17 (21)*,** | ||
Discontinuous: pulsatile | 0 | 29 (35)* | 11 (14)*,** | ||
Discontinuous: biphasic | 0 | 3 (4) | 32 (41)*,** | ||
Discontinuous: monophasic | 0 | 0 | 19 (24)*,** | ||
Renal venous impedance index | 0.2 (0.1–0.3) | 0.3 (0.2–1)* | 1 (1–1)*,** | <0.0001 | |
Renal venous discontinuity index, %a | — | 12 (10–25) | 29 (14–66) | 0.01 |
Values are mean ± standard deviation, n (%), or median (25th quartile, 75th quartile).
EDV, end-diastolic volume; ePAWP, echo-derived pulmonary artery wedge pressure; ePVR, echo-derived pulmonary vascular resistance; GLS, global longitudinal strain; HFpEF, heart failure with preserved ejection fraction; HFrEF, heart failure with reduced ejection fraction; IVC, inferior vena cava; LA, left atrium; LAVi, left atrial volume; LV, left ventricle; LVMi, left ventricle mass index; PAP, pulmonary artery pressure; RVF, renal venous flow; RWT, relative wall thickness; TAPSE, tricuspid annular plane systolic excursion.
Measured only in patients with the discontinuous renal venous flow (n = 94).
*P < 0.01 vs No US congestion.
**P < 0.01 vs 1 US sign of congestion.
Of the 224 (72%) patients with HF free of clinical congestion, 95 (42%) had at least one US marker of congestion, which was associated with higher plasma concentrations of NT-proBNP [1525, interquartile range (IQR) 810–2205 pg/mL vs. 409, IQR: 308–862 pg/mL in those without US evidence of congestion; P < 0.0001]. Of the 161 (52%) patients with HF having at least one sign of congestion, a similar percentage had one (n = 82, 51%) or ≥2 (n = 79, 49%) US signs of congestion (Figure 1). Although patients without HF had no clinical signs of congestion and normal NT-proBNP, US assessment revealed the presence of mild congestion in 9 cases (9%), mostly B-lines ≥4 (n = 7).
Patients with one or more US markers of congestion were older and more likely to have clinical congestion. They had lower urine osmolality, spot urinary sodium, and FENa, with a higher prevalence of micro- and macro-albuminuria; they also showed greater ePVS and neurohormonal activation. Pharmacological therapy was similar, apart from greater use of anticoagulants in those with more signs of congestion, reflecting a higher prevalence of atrial fibrillation (AF). LVEF, measured by 2D or 3D echocardiography, was also similar, regardless of the presence of congestion. However, LV global longitudinal strain (LVGLS) was higher (less negative), while LV diastolic function, left atrial function, right ventricular systolic function, and pulmonary haemodynamics were worse for those with more severe congestion by US (Table 2). The distribution of the characteristics of HF patients by each US measure of congestion is shown in Supplementary material online, Table S4.
Patients with HFpEF were older than those with HFrEF and more likely to be women and to have hypertension and AF. Clinical and US signs of congestion were similarly common in patients with HFrEF or HFpEF (see Supplementary material online, Tables S1 and S2). However, patients with HFrEF had higher NT-proBNP and were more likely to be treated with and receive a higher dose of loop diuretics.
Ultrasound markers of congestion correlated modestly with ePAWP and Log(NT-proBNP), but correlations amongst different US measures of congestion were generally much weaker, except for IVC and renal VII (Figure 2). The latter also showed positive correlations with serum urea and norepinephrine and an inverse correlation with FENa, LVGLS and 3D-RVEF.

Univariate correlation matrix for different markers of congestion using Spearman’s rank correlation. *P < 0.05, **P < 0.001, ***P < 0.0001. Red shading indicates positive correlations, and blue shading indicates inverse correlations. White boxes are non-significant (P > 0.05). Aldo, aldosterone; ePAWP, echo-derived pulmonary artery wedge pressure; ePVS, estimated plasma volume status; FENa, fractional excretion of sodium; hs-CRP, high sensitive C-reactive protein; IVC, inferior vena cava; LVGLS, left ventricle global longitudinal strain; LVEF, left ventricle ejection fraction; Nora, norepinephrine; RVEF, right ventricle ejection fraction; UNa+, urinary sodium; VII, venous impedance index.
Survival analysis
During a median follow-up of 13 months (IQR 6–15), 23 HF patients died, and 49 were hospitalized with worsening HF; 5 patients without HF at the enrolment developed HF requiring hospitalization (all of them had at least one US sign of congestion at baseline assessment). No patients were hospitalized immediately after the enrolment visit because of findings at the clinic visit.
Clinical and US signs of congestion were associated with adverse events (Figure 3). Patients with HF with ≥2 US signs had the worst outcome, while those free of congestion by US had a prognosis similar to patients without HF at enrolment (Graphical Abstract). In models adjusted for age, sex, AF, clinical congestion, eGFR, UACR, LVEF, and left atrial volume, only increasing Log(NT-proBNP) and worsening congestion by US were associated with an adverse outcome. HF patients with ≥2 US markers of congestion were 14 times more likely to experience the composite endpoint than those without US signs of congestion, and no interaction was noted with LVEF (Table 3).

Multi-panel Kaplan–Meier for the primary outcome of all-cause death (ACM) and heart failure hospitalizations (HFHs) in patients with heart failure (n = 310) according to clinical or ultrasound signs of congestion. (A) Signs of clinical congestion (pitting oedema, lung crackles, jugular vein distension). (B) Inferior vena cava (IVC) diameter. (C) B-lines (Tertile 1: < 1 B-line, Tertile 2: 1–3 B-lines, Tertile 3: ≥ 4 B-lines). (D) Renal venous flow (RVF) patterns.
Cox proportional-hazards multivariable regression analysis for predicting the composite endpoint (all-cause death and hospitalization for heart failure) in the HF population (n = 310)
Variable . | HR (95% CI) . | P-value . |
---|---|---|
Model 1 | ||
Log NT-proBNP | 5.73 (1.67–19.64) | 0.006 |
US signs of congestion | ||
None | Reference | |
1 US signs of congestion | 1.45 (0.52–4.33) | 0.4 |
≥2 US signs of congestion | 13.9 (4.9–39.3) | <0.0001 |
Model 2 | ||
Log(NT-proBNP) | 3.94 (2.38–6.05) | <0.0001 |
IVC, mm | ||
<21 mm | Reference | |
≥21 mm | 3.35 (1.67–6.75) | 0.001 |
B-lines | ||
Tertile 1: < 1 | Reference | |
Tertile 2: 1–3 | 0.99 (0.39–2.52) | 0.9 |
Tertile 3: ≥ 4 | 1.66 (1.15–2.88) | 0.01 |
RVF pattern | ||
Continuous | Reference | |
Discontinuous: pulsatile | 1.83 (0.62–5.39) | 0.2 |
Discontinuous: biphasic | 8.14 (3.37–19.67) | <0.0001 |
Discontinuous: monophasic | 10.12 (2.84–26.16) | <0.0001 |
Model 3 | ||
Log(NT-proBNP) | 3.85 (2.13–6.95) | <0.0001 |
IVC, mm | 1.13 (1.06–1.21) | <0.0001 |
B-lines | 1.02 (1.01–1.05) | 0.03 |
Log(Renal venous impedance index) | 5.23 (2.24–22.31) | <0.0001 |
Variable . | HR (95% CI) . | P-value . |
---|---|---|
Model 1 | ||
Log NT-proBNP | 5.73 (1.67–19.64) | 0.006 |
US signs of congestion | ||
None | Reference | |
1 US signs of congestion | 1.45 (0.52–4.33) | 0.4 |
≥2 US signs of congestion | 13.9 (4.9–39.3) | <0.0001 |
Model 2 | ||
Log(NT-proBNP) | 3.94 (2.38–6.05) | <0.0001 |
IVC, mm | ||
<21 mm | Reference | |
≥21 mm | 3.35 (1.67–6.75) | 0.001 |
B-lines | ||
Tertile 1: < 1 | Reference | |
Tertile 2: 1–3 | 0.99 (0.39–2.52) | 0.9 |
Tertile 3: ≥ 4 | 1.66 (1.15–2.88) | 0.01 |
RVF pattern | ||
Continuous | Reference | |
Discontinuous: pulsatile | 1.83 (0.62–5.39) | 0.2 |
Discontinuous: biphasic | 8.14 (3.37–19.67) | <0.0001 |
Discontinuous: monophasic | 10.12 (2.84–26.16) | <0.0001 |
Model 3 | ||
Log(NT-proBNP) | 3.85 (2.13–6.95) | <0.0001 |
IVC, mm | 1.13 (1.06–1.21) | <0.0001 |
B-lines | 1.02 (1.01–1.05) | 0.03 |
Log(Renal venous impedance index) | 5.23 (2.24–22.31) | <0.0001 |
All the models were adjusted for age (years), sex (male), atrial fibrillation, any signs of clinical congestion, eGFR (mL/min/1.73 m2), UACR (mg/g), LV ejection fraction (%), LAVi (mL/m2).
Legend as in the previous tables.
Cox proportional-hazards multivariable regression analysis for predicting the composite endpoint (all-cause death and hospitalization for heart failure) in the HF population (n = 310)
Variable . | HR (95% CI) . | P-value . |
---|---|---|
Model 1 | ||
Log NT-proBNP | 5.73 (1.67–19.64) | 0.006 |
US signs of congestion | ||
None | Reference | |
1 US signs of congestion | 1.45 (0.52–4.33) | 0.4 |
≥2 US signs of congestion | 13.9 (4.9–39.3) | <0.0001 |
Model 2 | ||
Log(NT-proBNP) | 3.94 (2.38–6.05) | <0.0001 |
IVC, mm | ||
<21 mm | Reference | |
≥21 mm | 3.35 (1.67–6.75) | 0.001 |
B-lines | ||
Tertile 1: < 1 | Reference | |
Tertile 2: 1–3 | 0.99 (0.39–2.52) | 0.9 |
Tertile 3: ≥ 4 | 1.66 (1.15–2.88) | 0.01 |
RVF pattern | ||
Continuous | Reference | |
Discontinuous: pulsatile | 1.83 (0.62–5.39) | 0.2 |
Discontinuous: biphasic | 8.14 (3.37–19.67) | <0.0001 |
Discontinuous: monophasic | 10.12 (2.84–26.16) | <0.0001 |
Model 3 | ||
Log(NT-proBNP) | 3.85 (2.13–6.95) | <0.0001 |
IVC, mm | 1.13 (1.06–1.21) | <0.0001 |
B-lines | 1.02 (1.01–1.05) | 0.03 |
Log(Renal venous impedance index) | 5.23 (2.24–22.31) | <0.0001 |
Variable . | HR (95% CI) . | P-value . |
---|---|---|
Model 1 | ||
Log NT-proBNP | 5.73 (1.67–19.64) | 0.006 |
US signs of congestion | ||
None | Reference | |
1 US signs of congestion | 1.45 (0.52–4.33) | 0.4 |
≥2 US signs of congestion | 13.9 (4.9–39.3) | <0.0001 |
Model 2 | ||
Log(NT-proBNP) | 3.94 (2.38–6.05) | <0.0001 |
IVC, mm | ||
<21 mm | Reference | |
≥21 mm | 3.35 (1.67–6.75) | 0.001 |
B-lines | ||
Tertile 1: < 1 | Reference | |
Tertile 2: 1–3 | 0.99 (0.39–2.52) | 0.9 |
Tertile 3: ≥ 4 | 1.66 (1.15–2.88) | 0.01 |
RVF pattern | ||
Continuous | Reference | |
Discontinuous: pulsatile | 1.83 (0.62–5.39) | 0.2 |
Discontinuous: biphasic | 8.14 (3.37–19.67) | <0.0001 |
Discontinuous: monophasic | 10.12 (2.84–26.16) | <0.0001 |
Model 3 | ||
Log(NT-proBNP) | 3.85 (2.13–6.95) | <0.0001 |
IVC, mm | 1.13 (1.06–1.21) | <0.0001 |
B-lines | 1.02 (1.01–1.05) | 0.03 |
Log(Renal venous impedance index) | 5.23 (2.24–22.31) | <0.0001 |
All the models were adjusted for age (years), sex (male), atrial fibrillation, any signs of clinical congestion, eGFR (mL/min/1.73 m2), UACR (mg/g), LV ejection fraction (%), LAVi (mL/m2).
Legend as in the previous tables.
For the entire cohort of HF patients, the base clinical model yielded a C-index of 0.75, which did not increase significantly when Log(NT-proBNP) or US signs of congestion were added singly. In contrast, adding Log(NT-proBNP) with discontinuous RVF and another measure of congestion by US significantly improved prediction. The highest C-index was observed when Log(NT-proBNP) and all US measures of congestion were added simultaneously to the base model (0.86, P < 0.001; Supplementary material online, Table S5). Adding Log(NT-proBNP) or US measures of congestion to the base model improved reclassification, increasing the integrated discrimination improvement (IDI) and continuous net reclassification improvement (cNRI; Supplementary material online, Table S5). Adding discontinuous RVF pattern alone or combined with other US signs of congestion to a model that included Log(NT-proBNP) further improved reclassification, primarily by IDI. The greatest improvement by IDI was when all the US measures were simultaneously added (P = 0.004) to this model. Adding a multi-organ assessment of congestion by US to a clinical model that included Log(NT-proBNP) provided a cNRI of 28% (P = 0.03; Supplementary material online, Table S5).
Discussion
In this prospective study, we performed an integrated evaluation of venous, pulmonary, and renal congestion in outpatients with HF, together with a detailed clinical, echocardiographic, and biomarker assessment. Previous reports focusing prevalently on one district showed that US signs of congestion are associated with higher plasma concentrations of NT-proBNP and a greater risk of an adverse outcome in patients with or at risk of HF, regardless of LVEF.2,3,6,12–14 The present study confirms and expands those findings by demonstrating the clinical and prognostic utility of a multi-organ assessment of congestion by US in patients with chronic HF, independently of indices of cardiac contractility—including LVEF—and renal function.
Under normal conditions, hydrostatic and oncotic pressures in the capillaries and interstitium are in equilibrium. In HF patients, sympathetic activation due to elevated filling pressures and ventricular dysfunction promotes renal sodium and water retention, with subsequent intravascular volume expansion and decline in oncotic pressure.15 When the capacity of lymphatics to drain interstitial fluids is exceeded, tissue oedema develops,1,16 which is rarely clinically evident unless severe. During months or even years, circulatory congestion may worsen, ventricular and atrial stretch increases silently, left and right ventricle longitudinal systolic function deteriorate, and arrhythmias (e.g. AF) may develop, eventually leading to pulmonary and systemic venous hypertension. For many patients, the diagnosis is delayed until symptoms and signs are so severe that they require hospitalization and intensive diuretic therapy17,18; more than 20% of these patients will die within a year.19,20 Earlier identification of congestion before the onset of symptoms and signs might allow intervention to prevent clinically overt congestion and potentially improve outcomes.21,22
Currently, the gold standard for assessing congestion is to measure atrial pressures invasively, which incurs substantial cost, requires specialized equipment and expertise, and carries small risks7; therefore, it is neither attractive nor practical for routine clinical use. US is widely available and affordable, does not expose patients to radiation, and images can be interpreted in real-time to assess congestion and identify its cardiac origin and recorded for later discussion at team meetings.6 Pellicori et al. showed that half of the patients attending an HF outpatient clinic who did not have clinical signs of congestion had US evidence of pulmonary or intravascular congestion and greater clinical risk.14 Many patients with risk factors for HF but who do not fulfil the current guideline definition of HF probably have pre-clinical congestion when investigated adequately, and there is clear evidence that this is associated with a worse prognosis.23–26 Our results are consistent with these observations and show that the absence of congestion by US is associated with an excellent prognosis.
RVF is a novel tool to help HF specialists manage congestion.6 We assessed RVF with the same transducer used for echocardiography within about 5 min, with good reproducibility and without dedicated training in renal Doppler ultrasound.6 The shift from a continuous to a discontinuous RVF pattern in patients with HF has been associated with higher right atrial pressure and a poorer prognosis in acute and chronic settings.13 Conversely, the normalization of RVF is associated with an excellent response to treatments for congestion.27,28 Interestingly, the transition from euvolemic status to volume overload is accompanied by a worsening RVF pattern (i.e. from a continuous to a discontinuous pattern) without any significant changes in CVP estimated from the analysis of the IVC.28 Changes in the RVF pattern could be an early marker of congestion, salt-avid state, and neurohormonal activation.
In the present study, we observed a moderate relation between US markers of congestion and plasma concentrations of NT-proBNP. NT-proBNP is a robust marker of cardio-renal dysfunction and congestion and is one of the most reliable prognosticators in patients with or at risk of HF.7 However, NT-proBNP concentrations are influenced by several other clinical factors (e.g. age, sex, body mass index, and AF)7 and are lower in patients with HFpEF than HFrEF for the same LV end-diastolic pressure.29 Notably, the results of blood tests are not immediately available, and natriuretic peptide-guided treatment has not been convincingly shown to improve outcomes in HF patients.30 Conversely, US is widely available, and our experience indicates that this proposed protocol is easy to implement in a heart failure clinic for the real-time assessment of congestion. Whether the initial measurement of NT-proBNP followed by serial US evaluation with a particular focus on multi-organ congestion improves HF diagnosis and management is a concept that requires further investigation, particularly now that miniaturized devices are available. Indeed, there are no guideline recommendations for US markers of congestion to guide HF treatment. Three small randomized clinical trials suggest that LUS-guided treatment significantly reduces HF rehospitalizations and/or the number of urgent visits for worsening HF both in acute and chronic setting.31–33 These data have been pooled in a meta-analysis, supporting the role of LUS in evaluating the risk of adverse outcomes in patients with HF regardless of the clinical setting, and supporting the presence of a beneficial effect of LUS-guided management, particularly in HF outpatients.34 The existing evidence, the present findings, and a robust pathophysiological background support the concept that to master HF we need to identify and master congestion efficiently and with more precision, irrespective of LVEF.35 Hopefully, this concept will pave the way for designing a large randomized clinical trial to demonstrate if and how addressing congestion more precisely with US can improve the prognosis of patients with HF.
Limitations
This is a single-centre study from a tertiary referral centre: it has inherent flaws related to selection and referral bias and the absence of a validation cohort. As an observational study, causality cannot be deduced based on these data. We classified all patients with LVEF <50% as HFrEF because only 50 cases had mildly reduced LVEF (41–49%). A sensitivity analysis revealed a similar distribution of signs of congestion by US for patients with LVEF ≤40%. We used spot samples to assess urinary sodium, as 24-h urine collection was impractical; spot samples are easy to obtain and clinically valid.36 Different protocols have been proposed for performing LUS,37 and our findings may have been different if we had used another method. Some patient characteristics, such as a high body mass index, might decrease the number of identifiable B-lines37 and make RVF and IVC assessment difficult.6 B-lines are not specific for pulmonary congestion and may also reflect parenchymal lung disease, albeit we excluded moderate-to-severe lung disease with spirometry. The study could not determine how concomitant factors such as intrinsic renal pathology contributed to RVF patterns; however, we adjusted the models for UACR. Guidelines make no recommendation to guide treatment according to markers of congestion on ultrasound, but it is possible that knowledge of the results influenced subsequent management for our patients. We did not study additional US measurements of congestion, e.g. jugular, portal, and hepatic vein US.6
Conclusions
Simultaneous assessment of pulmonary, venous, and kidney congestion by US is feasible and identifies a high prevalence of sub-clinical congestion associated with poor outcomes, irrespective of LVEF. The feasibility and speed of investigation, along with the widespread availability and relatively low cost of ultrasound, make the proposed protocol easy to implement in an HF clinic for the real-time assessment of congestion, especially when point-of-care testing (e.g. NT-proBNP) is not readily available. Whether treatment of congestion guided by US improves outcomes requires further investigation.
Supplementary material
Supplementary materials are available at European Heart Journal - Cardiovascular Imaging online.
Funding: None declared
Data availability
Data available on request.
References
Author notes
Nicola Riccardo Pugliese and Pierpaolo Pellicori contributed equally to this work.
Conflict of interest: P.P. has received consultancy honoraria and/or sponsorship support from Boehringer Ingelheim, Pharmacosmos, Novartis, Vifor, AstraZeneca, and Caption Health and research support from Bristol Myers Squibb. L.G. has received consultancy honoraria from GE Healthcare, Philips Healthcare, Caption Health. Other authors have no relationships relevant to the contents of this paper to disclose.