Introduction
The cardiorenal syndrome describes the interaction between heart and kidney and was classified into five subtypes in 2008 [
1]. In acute decompensated heart failure (ADHF), classified as cardiorenal syndrome type 1, deterioration of renal function is very common and is associated with adverse clinical outcomes [
1‐
3]. Elevation of right atrial and central venous pressure (CVP) is transmitted to the renal veins leading to an increased interstitial and tubular hydrostatic pressure within the encapsulated kidney, which consequently decreases the glomerular filtration rate (GFR) [
4]. Whereas renal function is relatively robust against reduced cardiac output, increased CVP leading to renal congestion has been one of the main pathophysiologic findings in cardiorenal syndrome [
5]. Hence, it is obvious that an adequate control of the congestion state combined with preservation of renal function is a central goal in the management of heart failure (HF). Beyond CVP as a surrogate for congestion, there is upcoming evidence that parameters of renal hemodynamics are superior in determination of renal venous congestion and renal function. [
6‐
8] However, the well-established resistance index (RI) which correlates with renal function, pathophysiology, and prognosis in both renal and cardiac disease was inferior in the prediction of outcome and diuretic response in HF compared to intrarenal venous flow (IRVF) pattern or venous impedance index (VII), which represent parameters of renal venous flow [
6,
7,
9]. Thereby, in patients with heart failure with reduced ejection fraction (HFrEF) volume expansion leads to a significant change in renal venous flow, even before changes in cardiac filling pressure becomes evident [
6]. In ADHF congestion contributes to renal dysfunction, neurohormonal activation, increased intra-abdominal pressure, excessive tubular sodium reabsorption, fluid overload and diuretic resistance [
10]. In some refractory cases, it is necessary to treat patients with renal replacement therapy (RRT) to overcome diuretic resistance and to treat acute kidney injury [
11]. Effective decongestion strategies might preserve renal function and relief symptoms of acute decompensation. Though there is growing evidence that Doppler sonographic parameters might be associated with the prognosis in HF, response to volume expansion as well as diuretic therapy [
6,
7,
12], no data are available for hospitalized patients with HFrEF undergoing standard decongestion therapy. Therefore, the objective of the present study was to analyze VII and IRVF pattern as Doppler sonographic parameters of renal venous congestion in patients with HFrEF and ADHF and to determine whether recompensation has a meaningful impact on these renal venous congestion parameters.
Discussion
The present study has three major findings: (1) Patients with ADHF show distinct Doppler sonographic signs of renal venous congestion at admission with high VII and altered IRVF pattern as well as signs of renal functional and/or structural impairment. (2) Decongestive strategies based on standard ADHF therapy contribute to an improvement of VII and IRVF pattern as well as albuminuria, whereas renal RI and plasma creatinine remained unchanged. (3) Patients with higher degree of renal venous congestion at the time of decompensation determined by IRD are more likely to experience clinically meaningful complications such as need for RRT or dead during hospitalization.
There are a few previous trials investigating renal venous congestion in patients with HF using IRD. In the first study by Iida et al., 217 patients with HF were analysed for renal venous congestion [
7]. It is remarkable that despite the investigated collective consisted of stable HF patients (N = 71) and patients before discharge after recompensation (N = 151), a large proportion of patients with distinct signs of renal venous congestion was identified (100 patients (46%) with abnormal IRVF pattern—biphasic and monophasic) [
7]. Herein, the study could impressively demonstrate that IRVF pattern is able to predict clinical outcome [
7]. The second notable study by Nijst et al. investigated renal venous congestion parameters in stable HF patients without signs of volume overload and differentiated between HFrEF and HFpEF [
6]. Thereby, volume expansion and consecutive diuretic treatment was performed resulting in an initial increase of VII during volume expansion followed by a decrease of VII after administration of intravenous diuretic therapy [
6]. The observed acute decrease of VII 1 h after diuretic administration is in line with the present data showing an effect of standard ADHF therapy on renal venous congestion parameters during hospitalization. A recent study confirmed the prognostic value of discharge IRVF pattern in patients with ADHF [
12]. Moreover, a small study showed an association between sonographic parameter of renal venous congestion with creatinine level rise in ADHF [
18]. Though there are several similarities with the aforementioned studies such as determined high levels of renal venous congestion parameters in patients with HF, the key differences of the present study is the longitudinal observation analysing the effect of ADHF treatment on IRVF. The present study recruited patients within the first 48 h after admission and thus much earlier in the time continuum of decompensation as compared to previous studies. Subsequently, the effect of decongestion strategies on renal venous congestion in ADHF patients was observed in the present study, showing a clear impact. As previously reported, an eGFR decline during aggressive decongestion in ADHF might paradoxically be associated with a beneficial outcome [
19,
20]. Aggressive recompensation strategies do not lead to tubular injury in ADHF, so that moderate impairment of renal function through decongestion strategies must be distinguished from traditional causes of acute kidney injury [
21]. Thus, the lack of an improvement of plasma creatinine or eGFR despite an improvement in Doppler sonographic parameters of renal venous congestion in the present study should not automatically lead to a negative interpretation. Moreover, a high proportion of patients in the present study showed WRF during hospitalisation (49%) which is described previously in a comparable way in a prior ADHF trial [
21]. In fact, there was a significant reduction in albuminuria as well as an increase in the proportion of patients with normoalbuminuria, indicating a beneficial effect of decongestion therapy on the kidneys.
In the present study, 66% of patients had moderate to severe impairment of renal function and 91% showed abnormal eGFR (< 90 ml/min/1.73 m
2) at inclusion which is comparable to previous data of the ADHERE database which showed that deterioration of renal function is very common in patients with ADHF [
2]. However, there was no significant change in eGFR between admission and discharge in the present study. This might be consequences of the mutually cancelling effects of improving renal venous congestion on the one side and decongestive therapy associated hemodynamic changes leading to a reduction of effective blood volume on the other side.
In line with a previous study that analyzed patients after recompensation of ADHF in which prevalence of WRF was similar in patients with different IRVF patterns at discharge, the present study revealed no significant differences in VII or distribution of IRVF pattern between patients with or without WRF [
7]. However, there is even evidence that aggressive decongestion is associated with improved clinical outcomes despite WRF during therapy for ADHF [
19,
22]. In accordance to a study that showed volume depending changes in NT-pro BNP levels in stable patients with HFrEF, the present study demonstrated significant reduction of NT-pro BNP levels from inclusion to discharge with higher levels in patients with discontinuous IRVF pattern [
6].When interpreting the IRVF pattern at discharge, with 14 patients (47%) showing discontinuous IRVF pattern, which may either indicate incomplete recompensation or insufficient capability of IRD to detect improvement of renal venous congestion, it should be taken into account that patients with HF, even in an apparently stable state, frequently exhibit Doppler sonographic signs of renal venous congestion (30–40%) [
6]. Thereby, it must be considered, up to approximately 40% of patients who have been admitted for ADHF have clinical signs of residual congestion at the time of discharge and that this is associated with a higher risk for mortality and rehospitalization [
23]. If IRD is sensitive enough to reliably detect improvement of renal venous congestion in ADHF and is suitable for guiding individual treatment of renal congestion, however, remains to be explored. Of note, renal congestion is not present in all patients with ADHF as shown in our study where 6 patients (20%) had continuous IRVF pattern at inclusion. It remains speculative whether those patients exhibited an isolated left-sided cardiac decompensation or whether the sensitivity of the IRD is restricted in ADHF.
In contrast to the observed changes of parameters of the renal venous system, RI as a parameter of the renal arterial system remained unchanged during recompensation, which is consistent with previous data highlighting the predominance of the venous system in cardiorenal syndrome. [
5] The present study has several limitations. Due to the observational character of the study, there was a lack of blinding and randomization. Moreover, there was no control group and the sample size was relatively small. The VII can differentiate renal venous congestion in defined spectrum from 0 to 1, but fails to differentiate congestion state in patients with severe renal venous congestion as predominantly seen in the present study at inclusion, where a great proportion of patients exhibited a VII of 1. In this condition, determination of IRVF pattern is apparently superior to differentiate the severity of renal venous congestion. Interpretation of the provided reliability parameters needs some caution, as the investigated ultrasound parameters are considered as markers of congestion for which stability during recompensation and over several weeks cannot be assumed. It must be kept in mind, that renal ultrasound remains a dynamic study with substantial observer-dependency. By interpreting eGFR values it must be kept in mind, that in acute kidney injury, which often occur in ADHF, GFR estimations are not valid. During acute kidney injury and especially with accompanying ADHF, patients are usually not in a stable state and there may be a marked variation in blood pressure and volume status. Moreover, there is a delay between the acute decrease in GFR and increase of serum creatinine and consecutive decrease in eGFR in the acute phase of acute kidney injury. [
24,
25] However, baseline creatinine and eGFR values are of clinical interest to characterize the study cohort and to get an approximate impression of the development of renal function during recompensation. A substantial number of HF patients with CKD exhibit renal artery stenosis. [
26] In the present study, no patients with known renal artery stenosis were included, however, there was no formal exclusion. In accordance with relevant previous reports investigating IRD in heart diseases [
6,
7], in the present study IRD was predominantly performed in the right renal vessels. However, a mean RI of 0.76 ± 0.07 with a lower range of 0.57 in the present cohort, makes the presence of severe, hemodynamic relevant renal artery stenosis rather unlikely. Because of the aforementioned limitations, the present results should therefore be interpreted with caution.