This prospective cohort study with convenience sampling enrolled patients aged 18 years or older presenting with all of the following: signs and symptoms of acute decompensated heart failure (ADHF), left ventricular ejection fraction of 40% or below, New York Heart Association (NYHA) functional class II to IV symptoms, and clinical evidence of venous congestion necessitating intravenous diuretics. Patients with hepatic cirrhosis, those on dialysis at discharge, those who declined participation, and those with planned rehospitalization for elective procedures such as revascularization or arrhythmia ablation were excluded.

The study took place at 2 tertiary hospitals in the South Region of Brazil from September 2021 to November 2022 and was approved by the Research Ethics Committees of both institutions. Written informed consent was obtained from all the patients or their legal representatives prior to inclusion.

Patients received treatment from cardiologists and cardiology fellows who were not involved in the study, and researchers were notified upon discharge. Just prior to discharge, after consent was obtained, a single operator (PR) conducted ultrasound evaluations (VExUS score and LUS) and clinical examinations to assess the EVEREST score. The ultrasound findings were not disclosed to the attending physician and did not influence the patient’s treatment plan.

Baseline and discharge data were extracted from electronic medical records. We collected demographic data (age, sex), medical history (comorbidities, history of ADHF admission), NYHA class at admission, need for noninvasive ventilation, intravenous vasodilators, and inotropes during hospitalization. Biochemical markers (hemoglobin, creatinine, and sodium) and medications used at admission and prescribed at discharge were recorded, as well as blood pressure and heart rate at discharge. Last available echocardiography evaluation measures were also collected, including left ventricular ejection fraction and presence of right ventricular systolic dysfunction, defined as one of the following: tricuspid annular plane systolic excursion < 16 mm, Doppler tissue imaging-derived tricuspid lateral annular systolic velocity wave (S’) < 9.5 cm/s, or fractional area change < 35%.

The primary outcome was a composite endpoint of readmission or emergency visits due to ADHF (defined as admission for inpatient intravenous administration of diuretics to manage volume status) within 90 days following hospital discharge. Follow-up data were collected through a review of the patient’s hospital records and a telephone interview. To avoid bias, patients who died during follow-up were also excluded, as death was a competing event for readmission.

The EVEREST score was calculated as the sum of scores ranging from 0 to 3 assigned to 6 clinical parameters: dyspnea, orthopnea, fatigue, jugular venous pressure, rales, and peripheral edema, resulting in a final score ranging from 0 (no congestion) to 18 (maximal clinical congestion).^10,11 Upon discharge, an EVEREST score of ≥ 2 was interpreted as indicative of clinical residual congestion.

Ultrasound assessment was performed using M-TURBO, Fujifilm Sonosite Inc. All patients were placed in a dorsal decubitus position with the head of the bed at 30°. Images obtained were recorded for subsequent evaluation.

The VExUS score was assessed using a phased-array probe and all Doppler findings were obtained during the end-expiratory phase of the respiratory cycle. The diameter of the intrahepatic portion of the IVC was initially measured using a longitudinal view from a subxiphoid position, 2 cm from the junction with the hepatic vein.¹⁵ If an adequate view was not obtained, the probe was moved laterally to the right side of the body over the liver. The maximal diameter during the respiratory cycle was recorded, and patients with an IVC diameter less than 2 cm were classified as having VExUS 0. For patients with a plethoric IVC (diameter ≥ 2 cm), hepatic vein and portal vein Doppler were evaluated using pulsed-wave Doppler. Hepatic vein Doppler was interpreted as normal (S > D), mildly abnormal (D > S), or severely abnormal (reverse S) based on the A, S, and D waves identified using pulsed-wave Doppler. Portal vein Doppler was evaluated based on peak (Vmax) and nadir velocities (Vmin) during the cardiac cycle, and the pulsatility fraction was subsequently calculated ([Vmax − Vmin] / Vmax) and interpreted as follows: normal (pulsatility fraction < 0.3), mildly abnormal (pulsatility fraction 0.3 to < 0.5), or severely abnormal (pulsatility fraction ≥ 0.5). The VExUS score interpretation used in this study did not include an assessment of intrarenal veins, similar to the approach used by Bhardwaj et al.: 0 (IVC < 2 cm), 1 (IVC ≥ 2 cm and normal Doppler patterns), 2 (IVC ≥ 2 cm and at least 1 mild Doppler abnormality), or 3 (IVC ≥ 2 cm and at least 1 severe Doppler abnormality).¹⁶

LUS assessment was performed using a curvilinear probe with an 8-zone examination approach of 15 cm depth. Each lung was divided into 4 zones, and a zone was considered positive if ≥ 3 B-lines were identified.¹⁷ The number of positive lung zones was categorized into the following 3 groups: 0 to 1 positive zone, 2 to 3 positive zones, and 4 to 8 positive zones.

PR and MMB independently assessed the recorded images. They were blinded to the clinical data and outcomes, and any disagreements were adjudicated through discussion.

Continuous variables are reported as mean ± standard deviation or median with interquartile range according to data normality, while categorical variables are presented as frequencies and proportions. Normality of the data was tested using the Shapiro-Wilk test. Differences between groups at baseline were analyzed using the unpaired Student’s t-test or Wilcoxon-Mann-Whitney U test, depending on normality assumptions. The chi-squared test was used to analyze categorical variables. The sensitivity, specificity, positive predictive value, and negative predictive value were calculated for VExUS score 2 or 3 as a post hoc analysis. All statistical analyses were conducted using IBM SPSS Statistics, version 20.0 (IBM Corp., Armonk, NY, USA). Statistical significance was set at p < 0.05.

We initially enrolled 58 patients; however 9 individuals were excluded. Two patients were excluded due to planned rehospitalization, specifically, 1 for atrial flutter ablation and 1 for coronary artery bypass graft surgery. One patient required dialysis at discharge; 1 declined to provide consent, and 5 participants died during follow-up period. The final cohort comprised 49 individuals, among whom 11 (22.4%) experienced the primary outcome.

The baseline characteristics of the study sample are described in Table 1. The study population predominantly consisted of male patients with a median age of 60 years. Nearly half of the patients had ischemic cardiomyopathy; 30% had a history of HF hospitalization during the last year; 55% had associated right ventricular dysfunction, and two thirds were in NYHA functional class IV upon admission. Seventy percent of patients were receiving at least 2 of the 4 pillars of guideline-directed medical therapy for HFrEF at admission, which increased to 90% at discharge.

Table 1

Variables	All patients (n=49)	Without primary outcome (n=38)	With primary outcome (n=11)	p
	n (%)*	n (%)*	n (%)*
Age, years†	59.5 (47.3 - 69.0)	56.0 (46.5 - 70.0)	64.0 (57.0 - 69.0)	0.259
Male sex	35 (71.4)	25 (65.8)	10 (90.9)	0.104
Preexisting conditions
Hypertension	32 (65.3)	23 (60.5)	9 (81.8)	0.191
Diabetes	15 (30.6)	10 (26.3)	5 (45.5)	0.225
Ischemic cardiomyopathy	21 (42.9)	16 (42.1)	5 (45.5)	0.843
Atrial fibrillation	14 (28.6)	12 (31.6)	2 (18.2)	0.386
CKD	4 (8.2)	2 (5.3)	2 (18.2)	0.168
COPD	7 (14.3)	6 (15.8)	1 (9.1)	0.576
Admission for ADHF in the last 12 months	16 (32.7)	10 (26.3)	6 (54.5)	0.079
NYHA functional classification				0.439
II	5 (10.2)	3 (7.9)	2 (18.2)
III	15 (30.6)	13 (34.2)	2 (18.2)
IV	29 (59.2)	22 (57.9)	7 (63.6)
Baseline medications
Beta blocker	36 (73.5)	26 (68.4)	10 (90.9)	0.137
CE inhibitor/ARB	36 (73.5)	27 (71.1)	9 (81.8)	0.476
Mineralocorticoid antagonist	22 (44.9)	15 (39.5)	7 (63.6)	0.156
ARNi	3 (6.1)	3 (7.9)	-	0.336
Nitrate	8 (16.3)	6 (15.2)	2 (18.2)	0.850
SGLT2 inhibitor	4 (8.2)	4 (10.5)	-	0.261
Digitalis	9 (18.4)	4 (10.5)	5 (45.5)	0.008
Loop diuretic	37 (75.5)	28 (73.7)	9 (81.8)	0.581
Thiazide	2 (4.1)	-	2 (18.2)	0.047
Laboratory data at baseline
Hemoglobin (g/dL)‡	13.4 ± 2.0	13.6 ± 2.0	12.8 ± 2.1	0.291
Creatinine (mg/dL)†	1.2 (0.9-1.4)	1.1 (0.9-1.3)	1.3 (1.1-1.6)	0.122
Sodium (mmol/L)†	139.0 (136.0-141.0)	140.0 (136.8-142.0)	136.0 (132.0-139.0)	0.018
Hospitalization data
Ejection fraction‡	24.2 ± 6.4	24.3 ± 6.2	24.0 ± 7.0	0.896
Right ventricular dysfunction	27 (55.1)	21 (55.3)	6 (54.5)	0.966
Noninvasive ventilation	6 (12.2)	6 (15.8)	-	0.159
Intravenous vasodilator	15 (30.6)	11 (28.9)	4 (36.4)	0.638
Inotrope use	5 (10.2)	3 (7.9)	2 (18.2)	0.321
Hospital LOS (days)†	15.0 (9.0-21.0)	14.5 (8.8-21.0)	15.0 (10.0-18.0)	0.548
Discharge medications
Beta blocker	47 (95.9)	37 (97.4)	10 (90.9)	0.340
ACE inhibitor/ARB	34 (69.4)	28 (73.7)	6 (54.5)	0.225
Mineralocorticoid antagonist	39 (79.6)	28 (73.7)	11 (100.0)	0.057
ARNi	7 (14.3)	5 (13.2)	2 (18.2)	0.675
Nitrate	13 (26.5)	10 (26.3)	3 (27.3)	0.950
SGLT2 inhibitor	10 (20.4)	8 (21.1)	2 (18.2)	0.835
Digitalis	19 (38.8)	15 (39.5)	4 (36.4)	0.852
Loop diuretic	46 (93.9)	35 (92.1)	11 (100.0)	0.336
Thiazide	10 (20.4)	5 (13.2)	5 (45.5)	0.019
Vital signs at discharge
Systolic blood pressure†	109.0 (94.5-119.0)	109.5 (94.8-118.5)	105.0 (93.0-120.0)	0.914
Heart rate‡	72.4 ± 12.9	74.1 ± 13.6	66.5 ± 8.0	0.027
Laboratory findings at discharge
Creatinine (mg/dL)†	1.1 (1.0-1.6)	1.1 (1.0 - 1.3)	1.6 (1.1-2.1)	0.028
Sodium (mmol/L)†	138.0 (134.0-139.0)	138.0 (135.8-139.0)	134.0 (133.0-138.0)	0.047

ACE: angiotensin-converting enzyme; ADHF: acute decompensated heart failure; ARB: angiotensin II receptor blocker; ARNi: angiotensin receptor-neprilysin inhibitor; CKD: chronic kidney disease; COPD: chronic obstructive pulmonary disease; LOS: length of stay; NYHA: New York Heart Association; SGLT2: sodium-glucose co-transporter-2. * Unless otherwise stated. ^† Values are expressed as the median (interquartile range). ^‡ Values are expressed as the mean ± standard deviation.

Patients who experienced the primary outcome exhibited a significantly higher proportion of digoxin and thiazide use at admission, along with more restricted heart rate control and lower plasma sodium concentrations. Upon discharge, they also showed higher plasma creatinine concentrations and continued to have lower plasma sodium levels. Furthermore, 45.5% of patients (5 out of 11) were prescribed sequential nephron blockade, involving the administration of 3 classes of diuretics, in contrast to only 10.5% (4 out of 38) of individuals without the outcome (p = 0.08).

At discharge, the majority of patients (53.1%) had an EVEREST score ≥ 2; 38.7% had at least 2 positive quadrants on LUS, and approximately one third had a VExUS score of 2 or 3. Table 2 displays the EVEREST score, LUS, VExUS score, and primary outcome.

Table 2

Variable	All patients (n=49)	Without primary outcome (n=38)	With primary outcome (n=11)	p
	n (%)*	n (%)*	n (%)*
EVEREST score†	2.0 (1.0 - 3.0)	1.5 (1.0 - 3.0)	2.0 (1.0 - 3.0)	0.759
EVEREST score ≥ 2	26 (53.1)	19 (50.0)	7 (63.6)	0.425
Number of positive quadrants on the LUS				0.692
0-1	30 (61.2)	24 (63.2)	6 (54.5)
2-3	13 (26.5)	9 (23.7)	4 (36.4)
4-8	6 (12.2)	5 (13.2)	1 (9.1)
VExUS score
0	22 (44.9)	20 (52.6)	2 (18.2)
1	10 (20.4)	7 (18.4)	3 (27.3)	0.131‡
2 or 3	17 (34.7)	11 (28.9)	6 (54.5)	0.044‡

EVEREST: Efficacy of Vasopressin Antagonism in Heart Failure: Outcome Study with Tolvaptan; LUS: Lung ultrasound; VExUS: venous excess ultrasonography. * Unless otherwise stated. ^† Values are expressed as the median (interquartile range). ^‡ Using VExUS score 0 as reference.

While there was no difference in the EVEREST score and number of positive lung zones between patients with and without the primary outcome, participants with VExUS score of 2 and 3 had a significantly higher proportion of the primary outcome (6 out of 17, 35.3%) when comparing with patients with VExUS score of 0 with the outcome (2 out of 22, 9%, p = 0,044), as demonstrated in the Central Illustration.

When each individual component of VExUS was assessed in isolation, only IVC > 2 cm showed a statistically significant difference, as it was present in 81.8% of those with the primary outcome (9 out of 11) and in 47.4% of those without the outcome (18 out of 38, p = 0.043). The isolated assessment of the hepatic vein Doppler and the pulsatility of the portal vein showed no association with the primary outcome. However, among all participants with IVC > 2 cm (27 out of 49, 55.1%), only 33.3% experienced the outcome, and, of these, 66.7% had at least 1 Doppler abnormality (VExUS 2 or 3). In contrast, among those with IVC > 2 cm but without any altered Doppler parameters (VExUS 1), 70% did not experience the outcome.

In predicting HF-related readmission or emergency visits, the VExUS score of 2 or 3 exhibited a sensitivity of 54.5% and a specificity of 71%, yielding a positive predictive value of 35.3% and a negative predictive value of 84.4%.