Author + information
- Received March 16, 2017
- Revision received May 4, 2017
- Accepted May 12, 2017
- Published online August 28, 2017.
- Petra Nijst, MDa,b,
- Pieter Martens, MDa,b,
- Matthias Dupont, MDa,
- W.H. Wilson Tang, MDc and
- Wilfried Mullens, MD, PhDa,d,∗ ()
- aDepartment of Cardiology, Ziekenhuis Oost-Limburg, Genk, Belgium
- bDoctoral School for Medicine and Life Sciences, Hasselt University, Diepenbeek, Belgium
- cDepartment of Cardiovascular Medicine, Heart and Vascular Institute, Cleveland Clinic, Cleveland, Ohio
- dBiomedical Research Institute, Faculty of Medicine and Life Sciences, Hasselt University, Diepenbeek, Belgium
- ↵∗Address for correspondence:
Dr. Wilfried Mullens, Department of Cardiology, Ziekenhuis Oost-Limburg, Schiepse Bos 6, 3600 Genk, Belgium.
Objectives The goal of this study was to assess: 1) the intrarenal flow in heart failure (HF) patients during the transition from euvolemia to intravascular volume overload; and 2) the relationship between intrarenal flow and diuretic efficiency.
Background Intrarenal blood flow alterations may help to better understand impaired volume handling in HF.
Methods Resistance index (RI) and venous impedance index (VII) were assessed in 6 healthy subjects, 40 euvolemic HF patients with reduced ejection fraction (HFrEF), and 10 HF patients with preserved ejection fraction (HFpEF). Assessments were performed by using Doppler ultrasonography at baseline, during 3 h of intravascular volume expansion with 1 l of hydroxyethyl starch 6%, and 1 h after the administration of a loop diuretic. Clinical parameters, echocardiography, and biochemistry were assessed. Urine output was collected after 3 and 24 h.
Results In response to volume expansion, VII increased significantly in HFrEF patients (0.4 ± 0.3 to 0.7 ± 0.2; p < 0.001) and in HFpEF patients (0.4 ± 0.3 to 0.7 ± 0.2; p = 0.002) but not in healthy subjects (0.2 ± 0.2 to 0.3 ± 0.1; p = 0.622). This outcome was reversed after loop diuretic administration. In contrast, RI did not change significantly after volume expansion. Echocardiographic-estimated filling pressures did not change significantly. VII during volume expansion was significantly correlated with diuretic response in HF patients independent of baseline renal function (R2 = 0.35; p < 0.001).
Conclusions In HF patients, intravascular volume expansion resulted in significant blunting of venous flow before a significant increase in cardiac filling pressures could be demonstrated. The observed impaired renal venous flow is correlated with less diuretic efficiency. Intrarenal venous flow patterns may be of interest for evaluating renal congestion.
- Doppler ultrasonography
- renal resistance index
- venous impedance index
Congestive heart failure is characterized by signs and symptoms of volume overload, contributing to a high morbidity and mortality burden. There is increasing recognition that the capacity of the kidneys to compensate for fluid overload relates not only to the underlying intrinsic renal function but also to renal blood flow, in part influenced by increased venous pressure (1). Currently, we have very limited insight at the bedside to distinguish these major factors influencing diuretic efficiency.
Renal vascular ultrasound has been used to assess the degree of severity of renal artery stenosis or to assess vascular and endothelial dysfunction. The intrarenal resistance index (RI) (Figure 1) reflects renal arterial flow and changes in response to alterations in arterial resistance and capacitance. A normal RI is approximately 0.60; however, an RI >0.70 is seen in renal artery stenosis, hypotension, renal vein thrombosis, and arterial stiffness (2). In contrast, renal venous flow is normally continuous and the venous impedance index (VII) low. The VII changes in response to alterations in venous compliance, which are determined by central venous pressure and/or renal interstitial pressure (3,4). Interestingly, recent reports have shown that renal flow indices strongly correlate with clinical outcomes in HF independent of estimated glomerular filtration rate (eGFR) and other conventional prognostic factors of HF (5,6). Whether such alterations are directly associated with intravascular volume expansion or removal has not been shown.
Therefore, the objectives of the present study were to: 1) study intrarenal flow in HF patients compared with healthy subjects during the transition from euvolemia to intravascular volume overload; and 2) investigate the relationship between intrarenal flow and the ability to alleviate intravascular volume excess.
This study was conducted in a single tertiary care center (Ziekenhuis Oost-Limburg, Genk, Belgium) between September 2014 and October 2015. The study complies with the principles of the Declaration of Helsinki, and the institutional review board approved the study protocol. Written informed consent was obtained from every subject before any study-specific action was performed.
Subjects were eligible for study inclusion if they were ≥18 years of age and able to give informed consent. Healthy volunteers had the following: 1) no history of cardiac or renal disease except for adequately treated hypertension with guideline-recommended therapy; 2) a normal clinical examination; and 3) a normal transthoracic echocardiography.
Patients with chronic heart failure with reduced ejection fraction (HFrEF) had the following: 1) a clinical diagnosis of heart failure with evidence of impaired left ventricular ejection fraction (LVEF) ≤40% diagnosed at least 6 months before inclusion; and 2) stable doses of medical therapy according to current guideline recommendations during ≥3 months. Exclusion criteria were as follows: 1) renal replacement therapy or severe renal dysfunction with an eGFR ≤15 ml/min/1.73 m2 determined by using the Chronic Kidney Disease Epidemiology Collaboration equation; and 2) any clinical sign or symptom of volume overload (i.e., pulmonary rales, orthopnea, jugular venous distention or ≥1 peripheral edema).
Patients with chronic heart failure with preserved ejection fraction (HFpEF) had a clinical diagnosis of HFpEF diagnosed at least 6 months before inclusion. The diagnosis of HFpEF was based on the presence of clinical signs or symptoms of HF in combination with an LVEF ≥50%, elevated levels of natriuretic peptides, and the presence of left ventricular hypertrophy, left atrial enlargement, or diastolic dysfunction (7). Exclusion criteria were similar to HFrEF patients.
All patients were admitted to the cardiology intensive care unit for research purposes. Each patient took his or her usual dose of medication at 8:00 am except for the maintenance dose of loop diuretics. Subjects were placed in the semi-supine position, and a venous catheter was placed in the forearm. After a 60-min equilibration period, all subjects were instructed to empty their bladder, and baseline vital parameters (blood pressure, heart rate, and weight), transthoracic echocardiography, intrarenal Doppler ultrasonography, and a venous blood sample were obtained. After baseline measurements, 0.5 l of isotonic hydroxyethyl starch (HES) 6% was infused over 10 min followed by an infusion of 0.5 l over a period of 3 h to maintain a stable intravascular volume expansion of 3 h. Start of infusion was appointed as time point zero, and every hour afterward appointed, for example, as +1 h, +2 h, and so forth. At +3 h, 1 mg of bumetanide was intravenously administered as a bolus infusion in all subjects. Clinical assessment, vital parameters, a venous blood sample, and urine output were collected hourly up to +3 h. Subsequently, urine output was collected until the time point of +24 h. Transthoracic echocardiography and intrarenal Doppler ultrasonography were obtained at baseline, +1 h (during intravascular volume expansion), and +4 h (1 h after administration of the intravenous loop diuretic). Subjects were discharged from the hospital at approximately +5 h. In-hospital intake of oral fluid was 100 ml in all subjects. After hospital discharge, patients were instructed to maintain their usual low-salt diet and maximum intake of 1.5 l over 24 h.
Laboratory measurements and urine sampling
Plasma N-terminal pro–B-type natriuretic peptide (NT-proBNP) levels were measured by using the Roche Diagnostics Assay (Roche, Rotkreuz, Switzerland). Values for eGFR, a measure of glomerular filtration function, were calculated by using the Chronic Kidney Disease Epidemiology Collaboration Formula. Spontaneous diuresis at +3 h and at +24 h was calculated as the cumulative collection of urine starting at 0 h until 3 h or until 24 h after the start of infusion.
Two-dimensional echocardiographic examination was performed with the use of a commercially available system (iE33W, Philips Healthcare, Andover, Massachusetts). Images were acquired in the left lateral decubitus position. All reported echocardiography measurements were averaged from 3 consecutive cycles and assessed as recommended by the American Society of Echocardiography (Online Appendix). Central venous pressure (CVP) was estimated based on the guidelines provided by the American Society of Echocardiography (8). An inferior vena cava diameter >2.1 cm that collapses <50% with a sniff was defined as an elevated CVP corresponding with a CVP of 15 mm Hg. An inferior vena cava dimeter ≤2.1 cm that collapses >50% with a sniff suggests a normal CVP pressure of 3 mm Hg. Indeterminate cases in which the inferior vena cava diameter and collapse do not fit this definition were given an intermediate value of 8 mm Hg.
Invasive pressure measurement with a pulmonary artery catheter
In the first 10 HFrEF patients, central hemodynamics with a pulmonary artery catheter were measured. This catheter was inserted through the right jugular approach with correct positioning confirmed through fluoroscopic guidance in the catheterization laboratory the day before.
Renal Doppler ultrasonography
Intrarenal Doppler ultrasonography was performed with the use of a commercially available system (Philips Healthcare) with a sector transducer frequency range of 2.5 to 5 MHz. Doppler echography was recorded of the right kidney with each subject in the left semi-lateral decubitus position or, in case of unsatisfactory image quality, the left kidney. The electrocardiogram (ECG) signal was simultaneously recorded by the ultrasound system. Color Doppler images were used to determine interlobar vessels. Pulsed Doppler waveforms of the interlobar arteries and veins were recorded simultaneously. The renal RI at an interlobar artery was calculated as the maximum flow velocity minus diastolic flow velocity, divided by maximum flow velocity (9) (Figure 1). The VII was calculated as the peak maximum flow velocity minus the maximum flow velocity at nadir, divided by peak maximum flow velocity (10). In addition, Doppler waveforms were divided into 2 flow patterns: continuous and discontinuous. Discontinuous flow was defined as a pattern in which velocity at the nadir was close to zero. If the nadir was zero, VII was 1.0; therefore, VII ranged from 0 to 1. All measurements were averaged over 3 cardiac cycles during sinus rhythm.
Two observers (P.N. and P.M.) independently assessed RI, VII, and intrarenal Doppler flow patterns in 15 patients. To test intraobserver variability, a single observer analyzed the data twice on occasions separated by a 1-month interval. Reproducibility was assessed as the mean percentage of error (absolute difference divided by the mean of the 2 observations).
Statistical analyses were performed with commercially available software SAS JMP Pro version 11.2 for Windows (SAS Institute, Inc., Cary, North Carolina). Continuous variables are expressed as mean ± SD in tables and as mean and 95% confidence intervals in figures if normally distributed, or otherwise as median (interquartile range). Normality was assessed by using the Shapiro-Wilk statistic. Values between groups were compared with the Wilcoxon test. Categorical data were expressed as percentages and compared by using the Pearson chi-square test. For comparison of repeated measures, the Wilcoxon signed rank test was used. Correlations were calculated by using Pearson’s coefficient or Spearman’s ρ as appropriate. Independence between renal function and VII was confirmed by using multiple regression modeling. Variables with a p value <0.10 in univariate regression analyses were included in a standard multivariate regression model. Variables in the univariate regression model were chosen based on clinical relevance. Statistical significance was always set at a 2-tailed probability level of <0.05.
Six healthy subjects, 40 HFrEF patients, and 10 HFpEF patients were included in the study. Baseline characteristics are presented in Table 1. The arterial RI was significantly lower in healthy subjects (0.5 ± 0.1) compared with HFrEF patients (0.6 ± 0.1; p = 0.013) and HFpEF patients (0.7 ± 0.1; p = 0.004). VII was nonsignificantly lower in healthy subjects (0.2 ± 0.2) compared with HF patients (both 0.4 ± 0.3). All healthy subjects exhibited a continuous venous flow pattern versus 70% of HFrEF patients and 60% of HFpEF patients. Age and right ventricular systolic pressure (RVSP) were significantly correlated with baseline VII. After correction for age, RVSP remained the only significant factor (R2 = 0.27; p = 0.002) (Online Methods, Online Table 1). There was no significant relation between baseline VII and eGFR, NT-proBNP, estimated CVP, LVEF, stroke volume, mean arterial pressure, E/A and E/E′, history of hypertension or diabetes, or medication use.
Response to intravascular volume expansion
At +1 h, the average infused amount of HES 6% was 0.6 ± 0.1 l. There was no significant change in clinical status in any subject in response to intravascular volume expansion, except for an increase in total body weight. NT-proBNP levels did not significantly change (Online Table 2).
Based on echocardiography, there was a trend to higher E/E′ values but no significant change in any other echocardiographic value in healthy subjects. In HFrEF patients, the grade of tricuspid regurgitation and RVSP significantly increased (from 30 ± 10 mm Hg to 33 ± 11 mm Hg; p = 0.005), and there was a trend toward higher values of E/A and E/E′. In the 10 patients with a pulmonary artery catheter in place, findings were not different from echocardiography, and they showed no significant increase in cardiac filling pressures from baseline during intravascular volume expansion (Online Table 2). Similarly, in HFpEF patients, there was a trend toward higher E/E′ and E/A values.
Based on renal Doppler ultrasound, there was no significant change in RI in any of the groups during intravascular volume expansion (Figure 2). VII increased nonsignificantly in healthy subjects but significantly in HFrEF and HFpEF patients from 0.4 ± 0.3 at baseline to 0.7 ± 0.2 during intravascular volume expansion (p < 0.001 and p = 0.002, respectively) (Figure 2). During intravascular volume expansion, all healthy subjects demonstrated a continuous flow pattern versus 20% of HF patients. Based on multivariate regression analysis, there was no significant relation between VII and age, eGFR, estimated CVP, LVEF, mean arterial pressure, E/A and E/E′, NT-proBNP, and RVSP at this time point (Online Table 3).
Spontaneous cumulative diuresis after 3 h of volume expansion was 0.9 ± 0.3 l in healthy subjects, 0.5 ± 0.3 l in HFrEF patients, and 0.7 ± 0.4 l in HFpEF patients (Table 2).
Loop diuretic efficiency
At 3 h, 1 mg of bumetanide was intravenously administered. At that point, the average infused amount of HES 6% was 0.9 ± 0.1 l. Compared with intravascular volume expansion at +1 h, VII decreased significantly in HFrEF and HFpEF patients (in HFrEF patients from 0.7 ± 0.2 to 0.4 ± 0.2 [p < 0.001] and in HFpEF patients from 0.7 ± 0.2 to 0.5 ± 0.3 [p = 0.017]) (Figure 2). One h after administration of intravenous diuretics, 73% of HFrEF patients and 70% of HFpEF patients had a continuous flow pattern. NT-proBNP levels were significantly higher in healthy subjects and HFrEF patients and nearly significantly higher in HFpEF patients compared with baseline (Online Table 2). Echocardiographic parameters at +4 h were comparable with baseline. After 24 h, cumulative diuresis was 2.6 ± 0.4 l in healthy subjects, 2.1 ± 0.7 l in HFrEF patients, and 2.4 ± 0.6 l in HFpEF patients.
HF patients with high versus low VII during intravascular volume expansion
The median VII during intravascular volume expansion in HFrEF patients was 0.739 and 0.756 in HFpEF patients. Both HFrEF and HFpEF patients were divided into 2 groups based on VII above (“high VII”) or below (“low VII”) the median.
HFrEF patients with high VII were characterized by a higher prevalence of hypertension, higher baseline levels of NT-proBNP, E/A, grade of tricuspid regurgitation, and baseline VII value (0.5 ± 0.3 vs. 0.3 ± 0.2; p = 0.012) (Online Table 4). HFpEF patients with high VII were significantly older (74 ± 2 years vs. 67 ± 8 years; p = 0.27) and had a significantly higher baseline VII value (0.6 ± 0.3 vs. 0.2 ± 0.1; p = 0.016) (Online Table 5). Among HF patients with high VII during intravascular volume expansion, one-half had a discontinuous flow pattern at baseline versus 1 of 10 patients in the group with a low VII during expansion.
HFrEF patients with a low VII during intravascular volume expansion had significantly better cumulative spontaneous diuresis after 3 and 24 h compared with HF patients with high VII. In HFpEF patients, cumulative diuresis after 3 h was significantly higher and diuresis after 24 h was nonsignificantly higher in the group of patients with low versus high VII (Figure 3, Table 3). In addition, VII remained significantly associated with cumulative diuresis at +3 and +24 h (p = 0.002 and p = 0.014, respectively) after correction for eGFR in all HF patients.
Intraobserver and interobserver variability of RI and VII measurements were as follows: RI, 0 ± 9% and 3 ± 12%, respectively; and VII, 4 ± 13% and 5 ± 12%. Classifications of intrarenal Doppler flow patterns were consistent between intraobserver and interobserver assessments.
The present study was an assessment of renal arterial and venous flow index alterations in optimal treated HF patients as well as healthy subjects undergoing the transition from euvolemia to intravascular volume overload. Our main findings are that intravascular volume expansion with 0.6 l led to a significant blunting of venous (not arterial) flow in HF patients before a significant increase in cardiac filling pressures could be demonstrated, which can be reversed after the administration of intravenous diuretics. In addition, blunting of renal venous flow in HFrEF and HFpEF patients during intravascular volume expansion was related to a lower diuretic response, independent of the underlying renal function. Intrarenal Doppler flow patterns provide real-time insight into renal hemodynamics and might help to better understand factors contributing to impaired volume handling.
Physiology of intrarenal flow and effects of increased CVP
Under normal conditions, blood flow in the renal arterial circulation is antegrade and maintained during diastole. When renal arterial vascular resistance increases or compliance lessens, a decrease in renal diastolic blood flow occurs, which is more pronounced than the decrease in the systolic component (11). This action results in an increased RI. Importantly, the Doppler waveform is altered not by vascular resistance alone but by the interaction of vascular resistance and compliance (e.g., large arterial distensibility, pulse pressure). Therefore, aging, atherosclerosis, and stiffening of the large arteries will result in an increased RI (12). This scenario likely explains the significantly lower baseline RI value in the cohort of healthy young subjects compared with HF patients.
Baseline levels of VII were not different between groups. After multivariate analyses, there was a poor but significant correlation between RVSP and baseline VII, but no significant correlation remained between clinical, echocardiographic, or biochemical variables and VII during intravascular volume expansion. Previously, an increased intrarenal VII was associated with increased CVP in HF and tricuspid regurgitation (6). Importantly, CVP is transmitted directly to the renal veins, and, consequently, elevated CVP leads to increased renal capillary pressure and decreased venous vessel compliance. Because the kidneys are encapsulated organs, renal congestion will be accompanied by increased renal interstitial pressure and decreased interstitial compliance, which can eventually reduce glomerular filtration pressure (13). This outcome is extensively studied in the context of acute decompensated HF in which elevated CVP (more than cardiac output) is identified as one of the main drivers of worsening renal function (1,14,15). The lack of (or only weak) correlation between VII and different variables in our study suggests that VII is rather an independent, or additional, parameter different from currently used clinical and biochemical parameters in HF patients.
Intravascular volume expansion leads to a change in intrarenal venous flow
RI in response to intravascular volume expansion did not significantly change, indicating that resistance and compliance of the arterial bed are relatively unchanged, which might be explained by the fact that approximately three-quarters of the intravascular volume resides in the venous circulation (16). In contrast, a significant change in renal venous Doppler flow index and pattern was observed after intravascular volume expansion with 0.6 l in both HFrEF and HFpEF patients but not in healthy subjects. Importantly, no significant change in echocardiographic estimates of CVP could be shown.
In addition to venous and interstitial pressure, vessel compliance can influence venous flow alterations (13,17). Most HF patients demonstrated a continuous flow pattern at baseline, meaning venous flow was continuous or with subtle variations during a cardiac cycle. In contrast, after intravascular volume expansion, marked fluctuations in flow occurred with values near zero at the nadir of flow in the vast majority of HF patients. Interestingly, the increased pulsatility or discontinuation of the venous flow signal was typically observed a few milliseconds after the P-wave and T-wave of the simultaneous recorded ECG. We hypothesize that small increases during atrial and ventricular contraction, in combination with a decreased venous compliance secondary to intravascular volume expansion, can induce larger pressure waves in the inferior vena cava and intermittently blunt (decrease or interrupt) forward flow in the interlobar veins (18) (Figure 4).
A significant increase in NT-proBNP levels at +4 h and a significant increase in RVSP and tricuspid regurgitation were observed in HFrEF patients. These outcomes suggest that intravascular volume expansion of 0.6 l caused subtle cardiovascular hemodynamic alterations, which influence renal venous flow before large increases in cardiac filling pressures are present. The venous system of healthy subjects can likely better handle increases in intravascular volume that might relate to intrinsic compliance of the vascular bed and baseline intravenous blood volume.
Higher VII relates to worse renal response
HF patients with the highest VII during intravascular volume expansion demonstrated a lower diuretic response after 3 and 24 h. Importantly, the relation between VII and renal response remained significant after correction for the underlying glomerular filtration rate. A negative correlation between sodium and water excretion and renal venous congestion has been shown previously (13). Renal venous congestion can influence diuresis in many ways. First, increased interstitial renal pressure as a result of venous congestion can reduce the hydrostatic driving force for filtration in Bowman’s space (19). Second, interstitial pressure rises and lymph flow massively increases, washing out interstitial proteins and decreasing interstitial colloid osmotic pressure, which facilitates sodium and water reabsorption, especially in the proximal tubules (19–21). Third, renal venous congestion stimulates the sympathetic nervous system and renin-angiotensin-aldosterone system, which is responsible for increased reabsorption of sodium and water in mainly the distal tubules (19,22–24).
One hour after administration of an intravenous diuretic, the increase in VII was reversed and nonsignificantly different from baseline values. VII seems to be a parameter that can rapidly reflect changes in intravascular volume or renal congestion, which are both negatively associated with HF prognosis (25,26). Indeed, it has been shown that renal flow indices strongly correlate with clinical outcomes in HF, independent of eGFR and other conventional prognostic factors of HF (5,6,27).
First, this trial was a small, single-center pilot study. The mechanistic approach and statistical methods used in this study were aimed at putting forward a new hypothesis regarding renal handling of volume in HF. Future research is necessary to confirm our results. Second, intrarenal arterial and venous flow was not directly measured but estimated based on ultrasound-derived flow indices. Third, the study was not able to determine how concomitant factors such as intrinsic renal pathology and hypertension contributed to the intrarenal Doppler flow patterns. Fourth, the healthy subjects were significantly younger than the HF patients, and age may be (partly) responsible for differences in venous compliance in the kidneys and other organs. Moreover, the mean age of HFrEF patients was slightly younger than the average age of HF patients in the general population, which is likely explained by the fact that older patients were less willing to participate in this invasive experiment and to comply with inclusion and exclusion criteria. Despite these limitations, our data warrant future studies to further clarify the role of intrarenal flow patterns in the pathophysiology of sodium avidity and the clinical value in HF.
In HFrEF patients, intravascular volume expansion leads to significant changes in renal venous (but not arterial) flow before a significant increase in cardiac filling pressures is present. HF patients with higher VII had a significantly lower natriuretic and diuretic response to intravascular volume expansion. Intrarenal Doppler flow patterns, which are readily clinically obtainable, may be of additional interest to the classic evaluation of HFrEF patients regarding interpretation of renal congestion.
COMPETENCEY IN MEDICAL KNOWLEGE: Currently, rehospitalization rates for signs and symptoms of volume overload and congestion remain unacceptably high. The clinical assessment of congestion is difficult, and reliable techniques are lacking. Insights based on this pilot study might contribute to a greater understanding of renal congestion and renal hemodynamic responses in HF. Renal Doppler ultrasound may allow detection of subclinical intravascular volume overload and help to explain differences in renal response to decongestive therapies beyond eGFR. Important for clinical implementation, intrarenal flow indices can readily, noninvasively, inexpensively, and in a short period of time be assessed with a conventional echocardiographic system. It has already been shown that intrarenal Doppler flow indices are related to prognosis in HF patients beyond conventional factors in HF.
TRANSLATIONAL OUTLOOK: Future studies might assess intrarenal Doppler ultrasonography–guided strategies for the prevention and treatment of volume overload in HF.
For an expanded Methods section and tables, please see the online version of this paper.
Drs. Nijst, Martens, and Mullens are researchers for the Limburg Clinical Research Program (LCRP) UHasselt–ZOL–Jessa; and supported by the foundation Limburg Sterk Merk, Hasselt University, Ziekenhuis Oost-Limburg, and Jessa Hospital. Dr. Martens is supported by a doctoral fellowship by the Research Foundation-Flanders (FWO). Drs. Nijst and Dupont are supported by a research grant provided by Vision4Life-Sciences. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- central venous pressure
- estimated glomerular filtration rate
- hydroxyethyl starch
- heart failure
- heart failure with preserved ejection fraction
- heart failure with reduced ejection fraction
- left ventricular ejection fraction
- N-terminal pro–B-type natriuretic peptide
- resistance index
- right ventricular systolic pressure
- venous impedance index
- Received March 16, 2017.
- Revision received May 4, 2017.
- Accepted May 12, 2017.
- 2017 American College of Cardiology Foundation
- Mullens W.,
- Abrahams Z.,
- Francis G.S.,
- et al.
- Di Nicolo P.,
- Granata A.
- Iida N.,
- Seo Y.,
- Sai S.,
- et al.
- Ponikowski P.,
- Voors A.A.,
- Anker S.D.,
- et al.
- Rudski L.G.,
- Lai W.W.,
- Afilalo J.,
- et al.
- Burnett J.C. Jr..,
- Knox F.G.
- Damman K.,
- van Deursen V.M.,
- Navis G.,
- Voors A.A.,
- van Veldhuisen D.J.,
- Hillege H.L.
- Mullens W.,
- Nijst P.
- Boron W.F.,
- Boulpaep E.
- Tang W.H.,
- Kitai T.
- Tomsin K.,
- Vriens A.,
- Mesens T.,
- Gyselaers W.
- Haddy F.J.,
- Scott J.,
- Fleishman M.,
- Emanuel D.
- Lebrie S.J.,
- Mayerson H.S.
- Colombo P.C.,
- Doran A.C.,
- Onat D.,
- et al.