Current Treatment Options in Cardiovascular Medicine

, Volume 14, Issue 4, pp 342–355

Management of the Cardiorenal Syndrome in Acute Heart Failure


  • Valentina Lazzarini
    • Duke Clinical Research Institute
    • Section of Cardiovascular Diseases, Department of Experimental and Applied MedicineUniversity of Brescia
    • Duke Clinical Research Institute
Heart Failure (J Fang, Section Editor)

DOI: 10.1007/s11936-012-0186-5

Cite this article as:
Lazzarini, V. & Felker, G.M. Curr Treat Options Cardio Med (2012) 14: 342. doi:10.1007/s11936-012-0186-5

Opinion statement

Interactions between the heart and kidney in the setting of acute heart failure are complex and have a substantial impact on patient care and outcomes. Further research is needed to better distinguish the different causes of kidney injury, allow its early and accurate prediction and detection, and identify therapeutic targets. Novel renal biomarkers could potentially provide a useful tool for this purpose. Restoration of optimal fluid status and resolution of renal venous congestion are important goals of therapy. Changes in serum creatinine, although an important marker of renal function, may not be associated with adverse outcomes, especially if they are transient and a consequence of more aggressive decongestion, or the appropriate titration of drugs affecting the renin-angiotensin-aldosterone axis. In addition to loop diuretics, a variety of drugs and strategies have been investigated in acute heart failure. Use of mineralocorticoid receptor antagonists and vasopressin antagonists may have potential benefits and should be further investigated. Inotropic agents should be limited in those clinical settings suggesting hypoperfusion. Ultrafiltration seems to provide a safe and effective tool to overcome diuretic resistance and optimize fluid status avoiding detrimental effects of diuretic therapy.


Cardiorenal syndromeAcute heart failureWorsening renal function


Cardiorenal syndrome (CRS) has been most recently described as a disorder of the heart and kidneys whereby acute or chronic dysfunction in one organ may induce acute or chronic dysfunction of the other [1]. It is divided into five types, depending on the kind of damage and the organ primarily involved. In this paper, we will focus on the development of worsening renal function (WRF) in the setting of acute heart failure (AHF), ie, CRS type 1.

During an episode of AHF, the kidney can be affected by a wide range of mechanisms, many of which are common to the heart and the kidney. These mechanisms include hemodynamic abnormalities (hypoperfusion, vascular resistance dysregulation, fluid overload, and increased central venous pressure), neurohormonal hyperactivation (renin-angiotensin-aldosterone system [RAAS], sympathetic nervous system [SNS], and local mechanisms), ischemia, oxidative stress, and inflammation. The heart and kidneys also share several risk factors such as diabetes, hypertension, and age, which can affect both of them independently.

Most trials and observational studies define WRF as an increase in serum creatinine (sCr) levels of 0.3 mg/dL or above from baseline [2, 3]. Using this cut-off, in large registries and trials, up to 45 % of patients develop WRF during hospitalization for AHF. Notably, only 10 % of patients with AHF have normal renal function at baseline, and baseline renal impairment is a strong predictor of WRF [4, 5] and of poor outcome itself [6, 7].

However, the definition of WRF is not uniform, and methods used for its assessment are suboptimal. The most common surrogates of renal function are sCr and estimated glomerular filtration rate (eGFR), which have several limitations. Creatinine-based formulas tend to underestimate higher GFR and overestimate lower GFR [8]. SCr concentration also may be affected by extrarenal [8] as well as iatrogenic factors [9], without impact on real GFR. Moreover, a decline in real GFR does not always reflect kidney injury. This is the case of initiation of therapy with angiotensin-converting enzyme (ACE) inhibitors, whose own mechanism of action involves a decrease in GFR; still, they are demonstrated to improve prognosis and preserve renal function [10, 11]. Finally, renal function is not limited only to filtration activity, and glomerular capillary or tubulointerstitial damage can occur without affecting GFR, at least in the earliest phases [8].

WRF may therefore be thought of as a “final common pathway” with a variety of precipitants and mechanisms. This heterogeneity may partially explain why patients with WRF during AHF show such mixed outcome and response to treatment. In several registries and prospective studies, WRF is a predictor of poor prognosis [12, 13] with regard to length of in-hospital stay, mortality, and risk of readmission. However, other studies showed that WRF can be associated with a decrease in signs and markers of congestion [14••, 15•]. In this case, an increase in sCr does not necessarily worsen outcome, especially when transient [1618], whereas a decrease could indicate inadequate decongestion [19]. Conversely, venous congestion itself is a determinant of WRF [16, 20, 21] and an aggressive approach aimed to restoration of fluid balance may turn into a beneficial effect on renal function [22] and prognosis [16].

Because of these inconsistencies, it seems inappropriate to generalize risk assessment and treatment of patients depending only on sCr. Ongoing studies with novel renal biomarkers may provide greater insight into the pathophysiology and optimal treatment of the CRS. Compared to sCr, they promise to allow a more accurate estimation of GFR and earlier detection of kidney injury.

With regard to glomerular filtration function, cystatin C (CysC) is a 13 kDa protein, which is close to the “ideal” renal marker for estimation of GFR. It is produced at a constant rate in all nucleated cells, and is not significantly influenced by age, sex, diet, or muscle mass. It is completely filtered by the glomerulus, not secreted, and it is catabolized in the tubule. In fact, it has been demonstrated to be superior to sCr in various populations [23]. Preliminary data showed that CysC was an independent predictor for poor outcome in patients with chronic HF [24, 25] and AHF [26, 27].

Other molecules synthesized by tubular epithelial cells are emerging as markers of tubulointerstitial injury. Neutrophil gelatinase–associated lipocalin (NGAL) is a 21 kDa protein, freely filtered by the glomerulus and completely reabsorbed. It is secreted by the kidney and other organs, in various settings of local or systemic damage (Inflammation, cancer), and can be found in both serum and urine. Serum NGAL (sNGAL) is less specific and may rise as a consequence of systemic conditions despite normal tubular function. Increased urinary NGAL (uNGAL) levels are instead more suggestive of tubular injury, and may be due to both an incomplete reabsorption in the proximal tubule or its secretion from distal tubule [8]. Patients with chronic HF show increased uNGAL levels [28, 29], but its role as an independent prognostic factor is not clear [28, 30, 31]. In the setting of AHF, sNGAL seems to relate well with prognosis [32], although it appears to be less sensitive in detecting acute kidney injury [33, 34]. Urinary NGAL seems more specific in prediction of development of WRF [35, 36]. Kidney injury molecule 1 (KIM1) and N-acetyl-β-D-glucosaminidase (NAG) are a transmembrane glycoprotein and an enzyme, respectively, which are synthesized by proximal tubular cells. Their urinary levels increase as a consequence of proximal tubular damage. Both of them are confirmed to be strong markers of kidney injury in various clinical setting [8]. These markers have not been studied in AHF so far, but preliminary data show that their levels are increased in patients with chronic HF, and related to poorer outcomes [28, 31, 37].

Such promising findings need to be confirmed in larger trials to provide a definite validation of these biomarkers in the setting of AHF.


Pharmacologic treatment

Loop diuretics

  • The use of diuretics has not been extensively evaluated in carefully controlled clinical trials until recently, and level of evidence in current recommendations is based on expert opinion [11, 38]. Some concerns remain about the safety of diuretic therapy. Loop diuretics are known to have detrimental effects, like activation of the RAAS [39], and tubuloglomerular feedback [40], which possibly even worsen the vicious cardiorenal circle. Previous studies demonstrated an association between use of diuretics, and poor outcome [41] and WRF [42]. However, this association has been recently not confirmed [43], and it is still not clear whether this is a cause–effect association or just a marker for higher risk patients [44].

  • The recently published Diuretic Optimization Strategies Evaluation (DOSE) trial [45••] tried to answer these unresolved questions, providing the only available data in a sufficiently large, randomized, controlled cohort of patients. This study compared the effect on symptoms and renal function between high and low doses of furosemide, and continuous infusion versus bolus, in patients with AHF. With regard to the best approach for furosemide administration, no significant differences were found between bolus and infusion. This was in contrast with data from previous, smaller studies [4648], which suggested the superiority of continuous infusion in terms of effectiveness, safety profile in general, and minor impact on renal function.

  • However, high doses showed a trend toward a greater improvement of symptoms and greater effectiveness in secondary end points associated with fluid loss. Although this was associated with a significant higher rate of transient WRF, the mean change in creatinine levels from baseline to 72 h was not significantly different (Fig. 1). Moreover, although the study was underpowered to detect differences in clinical outcomes, patients in the high-dose cohort showed a numerically lower rate of death, rehospitalization, or emergency visits [45••]. These findings seem to confirm that a transient WRF may be acceptable if aimed to a better relief from congestion, which instead remains an ominous prognostic factor also with regard to renal function.
    Figure 1

    Mean change in serum creatinine level in the Diuretic Optimization Strategies Evaluation (DOSE) trial. The mean change in the serum creatinine level over the course of the 72-hour study-treatment period is shown for the group that received boluses every 12 h as compared with the group that received a continuous infusion and for the group that received a low dose of the diuretic (equivalent to the patients’ previous oral dose) as compared with the group that received a high dose (2.5 times the previous oral dose). (From Felker et al. [45••], with permission from the New England Journal of Medicine, 2011).

  • Patients with evidence of fluid retention should be treated with loop diuretics [49] administered intravenously (IV). This would prevent a delayed or insufficient effect due to intestinal edema or hypoperfusion, which can compromise oral absorption. Decreased GFR requires higher doses to reach therapeutic concentrations in the tubule and effective diuresis. Diuretics should be continued until optimization of fluid balance, based on careful assessment of volume status, blood pressure, perfusion, and renal function.

  • To avoid rebound effects on the RAAS and tubuloglomerular feedback, loop diuretics should always be associated with RAAS inhibitors. Careful monitoring of fluid balance, serum electrolytes and renal function is required.

Standard dosage

Usual parenteral single bolus dose (20 to 80 mg) may be not sufficient. 2X usual home oral dose should be given IV, with intermittent bolus or intravenous infusion, up to 40 mg/h.



Main drug interactions

Reduction of theophylline and gentamicin clearance, and therapeutic effect of warfarin. Increase in hypokalemic effect of amphotericin, antiepileptic effect of sodium valproate, risk from ototoxicity from aminoglycosides and vancomycin. Risk of hypotension with vasodilators, RAAS inhibitors. NSAIDs can increase the risk of WRF.

Main side effects

Excessive diuresis, volume depletion, and hypotension. Hypokalemia, hypocalcaemia, hypomagnesaemia. Increased risk for arrhythmias. Hyperuricemia. Ototoxicity.

Special points

Diuretic resistance is a common phenomenon, caused by several mechanisms: breaking effect, tubular cells hypertrophy, gut congestion. These mechanisms are exacerbated in the context of impaired renal function. Thiazide or thiazide-like diuretic (metolazone) may be added to overcome diuretic resistance, acting through a sequential nephron blockade. Thiazides are typically given as a single oral dose 1 h before loop diuretics.

Angiotensin-converting enzyme inhibitors and angiotensin receptor antagonists

  • In the setting of AHF and acute CRS, ACE inhibitors and angiotensin receptor antagonists (ARBs) are often discontinued for hypotension and/or fear of deteriorating in renal function. However, an early increase in sCr levels during ACE inhibitors therapy reflects adaptive hemodynamic mechanisms, namely post-glomerular vasoconstriction, aimed to maintain perfusion pressure and GFR, and there is increasing evidence that a rise in sCr due to ACE inhibitors does not necessarily portend a poor outcome [11, 42, 50]. ACE inhibitors and ARBs have well-known renoprotective action and favorable effects on cardiac outcome, arterial filling, diuresis and natriuresis, kidney perfusion, and long-term renal outcomes. They also balance detrimental effects of loop diuretics.

  • Current guidelines recommend initiation of these therapies before discharge, and continuing therapy with ACE inhibitors or ARBs in patients admitted with AHF already in treatment, in absence of hemodynamic instability or other contraindications.

Standard dosage

ACE inhibitors should be initiated at low doses before discharge and uptitrated to the maximum tolerated dose during the postdischarge follow-up, with serial assessment of blood pressure, electrolytes, and renal function.


Previous angioedema (ACE inhibitors), or anuric renal failure, bilateral stenosis of renal arteries. Signs of hypoperfusion, hemodynamic instability (systolic blood pressure [SBP] <80 mmHg), sCr above 3.0 mg/dl, potassium above 5.5 mEq/L.

Main drug interactions

NSAIDs. Combined use of ACE inhibitors, ARBs and mineralocorticoid receptor antagonists (MRAs) is not recommended.

Main side effects

Hypotension, hyperkalemia, angioedema, coughs (ACE inhibitors).


  • Use of inotropic or inodilator agents (dopamine, dobutamine, enoximone, milrinone, levosimendan) may be necessary in low cardiac output state to maintain acceptable organ perfusion. With regard to CRS, analysis from recent trials showed that optimization of hemodynamics and renal blood flow does not necessarily result in better kidney function [7, 51]. Therefore, the role of reduced cardiac output in the establishment of CRS has been reconsidered.

  • Use of inotropic agents should be limited to those settings in which WRF is closely related with global and renal hypoperfusion and used for a short time due to known possible adverse effects. Careful monitoring of electrolytes and continuous electrocardiography are required.

  • Low-dose dopamine (≤2 μg/kg/min) selectively acts on DA1 receptors without an actual inotropic action, but may have favorable effects on renal blood flow. At intermediate doses (2–5 μg/kg per min), dopamine interacts with the β1-receptors, producing positive inotropic effects There is no clear evidence of a benefit on major outcomes [5254]. However, it may be useful to improve diuresis and reduce the necessity of furosemide when used in association to diuretics. The ongoing ROSE trial will assess the efficacy and renal safety of low dose dopamine and low dose nesiritide in addition to furosemide [55].

Standard dosage

1 to 5 μg/kg/min. A predominant dopaminergic effect can vary significantly in any given individual.


Tachyarrhythmias, pheochromocytoma.

Main drug interactions

Monoamine oxidase inhibitors.

Main side effects

Tachyarrhythmias, myocardial ischemia.

Natriuretic peptides

  • Nesiritide is a recombinant B-type natriuretic peptide (NP) with vasodilating and mild diuretic properties. Previous studies demonstrated a reduction in pulmonary capillary wedge pressure and short-term improvement in dyspnea with nesiritide, and on this basis, this drug was approved by the U.S. Food and Drug Administration in 2001 for treatment of patients with AHF.

  • With subsequent studies, some concerns were raised about safety of nesiritide, including a higher risk for WRF [56, 57], thus leading to the design of a large trial on efficacy and safety of nesiritide (Acute Study of Clinical Effectiveness of Nesiritide in Decompensated Heart Failure [ASCEND-HF]) [58•]. Nesiritide showed neutral effects on cardiovascular events, dyspnea, and renal function compared to placebo but it was associated with a higher risk of hypotension. Based on the results of ASCEND, nesiritide cannot be recommended as a standard treatment in the general AHF population.

Standard dosage

Bolus 2 μg/kg (which can be omitted), followed by infusion from 0.005 to 0.03 μg/kg/min.


Cardiogenic shock, systolic blood pressure under 90 mmHg.

Main drug interactions

Antihypertensive drugs.

Main side effects

Hypotension, nausea, dizziness, headache, arrhythmias

Special points

Currently, new types of NPs are under development. In particular, CD-NP is a chimeric NP resulting from the combination of potent natriuretic and diuretic dendroaspis NP with C-type NP, which has venodilating and antiproliferating effects. CD-NP seems to have favorable cardiorenal properties, including preservation of glomerular filtration rate with minimal blood pressure–lowering effects [59, 60].

Mineralocorticoid receptor antagonists

  • Patients hospitalized for worsening HF often show inappropriately high levels of mineralocorticoids, which have several detrimental consequences. The MRAs spironolactone (nonselective competitive inhibitor of MR) and eplerenone (specific for the MR) have been extensively investigated in chronic systolic HF trials, and they are demonstrated to improve symptoms and prognosis of patients [6163]. They have not been investigated in AHF yet. However, most pathophysiologic mechanisms involved in AHF may be positively affected by MRAs.

  • With specific regard to renal function, MRAs may have several favorable effects: they may attenuate arterial vasoconstriction and blood flow decrease, counter sodium reabsorption, and prevent renal interstitial fibrosis, glomerulosclerosis, and proteinuria. As much important, they enhance diuresis and natriuresis, thus contributing to treat diuretic resistance and counter side effects of loop diuretics such as SNS, RAAS, and tubuloglomerular feedback hyperactivation, and potassium loss [64].

  • Previous small, nonrandomized, open-label trials have shown body weight decrease and diuretic response in AHF patients treated with high doses of MRAs, without significant effect on sCr [6567]. Untoward effects of MRAs include hyperkalemia and WRF, although in previous trials the rise in sCr in treatment group was clinically insignificant. It is uncertain whether the risk of adverse effects like hyperkalemia and WRF would be greater in an AHF population.

Standard dosage

Currently, there is no specific indication for the use of MRAs in AHF. Recommended initial dose of spironolactone is 12.5–25 mg daily, while natriuretic response starts from doses of spironolactone of 50 mg or more. Eplerenone is generally half as potent as spironolactone, with dosing beginning at 25–50 mg daily.


Hyperkalemia, sCr above 2.5 mg/dl.

Main drug interactions

Triple blockade with ACE inhibitors and ARBs should be avoided, as the use of other potassium-sparing diuretics. NSAIDs. Eplerenone has a higher potential for drug interactions being a substrate for the CYP3A4, which is inhibited by drugs as azole antifungals, macrolide antibiotics, and verapamil.

Main side effects

Hyperkalemia. Spironolactone nonselective action can induce gynecomastia, which is instead uncommon as a side effect of eplerenone.

Special points

A new potassium-binding molecule (RLY5016) recently demonstrated to reduce effectively potassium levels and in HF patients treated with spironolactone and could be a potential tool to avoid hyperkalemia associated to MRAs [68].

Other procedures


  • Ultrafiltration (UF) enables mechanical removal of isotonic fluid. It acts through convective transfer of plasma and small (<20 kDa) solutes through a semipermeable membrane, determined by hydrostatic pressure difference. The main target is removal of volume, and not solute clearance; this makes this technique particularly suitable for congestion in AHF patients in whom diuretic therapy is not sufficient [69]. In contrast to loop diuretics, electrolyte depletion is uncommon.

  • At the moment, indications for UF are based on several variables: urine output, electrolyte concentrations, renal function, response to diuretics, and metabolic parameters [70].

  • The Ultrafiltration versus IV Diuretics for Patients Hospitalized for Acute Decompensated CHF (UNLOAD) trial demonstrated safety of UF, and possibly, in comparison with patients treated with diuretics alone, less HF hospitalizations, and favorable effects on neurohormonal activation. Notably, sCr did not differ significantly between the two groups, as well as improvement in dyspnea and mortality [71•]. These preliminary data indicate UF as a promising tool to treat CRS, but larger trials are needed to demonstrate the impact of UF on prognosis and to assess cost-effectiveness. Notably, the ongoing prospective Cardiorenal Rescue Study in Acute Decompensated Heart Failure (CARESS-HF) trial will compare UF with stepped pharmacological therapy in patients with acute decompensated HF and CRS [72].

Standard procedure

Aquadex FlexFlow (Gambro, Lakewood, CO) allows UF at low flows (up to 40 mL/min) with a removal rate of 10 to 500 mL/h.


Inadequate (<85 mmHg) systolic blood pressure or end-stage renal disease.


Access-related issues including infections and bleeding (due to the need for aggressive concomitant anticoagulation of the system), hypovolemia, hyponatremia.


UF is expensive, and seems to increase the cost of hospitalization compared to diuretics, but decreased the overall costs of care through the lower rehospitalization rate [73].

Special points

Special attention should be given to blood volume and body fluid status [70]. Slow continuous UF (2–4 mL/min) does not generally require replacement fluid infusion [69].

Emerging therapies

Vasopressin antagonists

  • Arginine vasopressin (AVP) levels are often elevated in HF. This excess can lead to detrimental short-term effects such as water retention and hyponatremia (mediated by V2 receptor), vasoconstriction, and also long-term complications such as myocardial-fibrosis (V1a receptor) [74].

  • Currently available AVP antagonists are conivaptan (intravenous nonselective V1a and V2 receptor antagonist) and tolvaptan (oral, selective V2 receptor antagonist). They enhance free water clearance and a true aquaresis, without electrolyte loss. Currently they are approved for treatment of euvolemic and severe (serum Na <125) or symptomatic hypervolemic hyponatremia but not for the specific treatment of congestion of HF.

  • Conivaptan showed favorable effects on diuresis and natriuresis in comparison to placebo and in addition to furosemide in small randomized trials [75, 76], without adverse effects on neurohormonal activation and renal function.

  • Larger trials have been performed on the effect of V2 selective antagonist tolvaptan [77, 78], confirming its effectiveness in achieving fluid loss, improving hyponatremia, and without affecting renal function. However, it had no effect on morbidity and mortality [77]. On the basis of these results, AVP antagonists may be considered safe with regard to renal function in patients with volume overload and hyponatremia, but further investigations are needed in this direction.

Adenosine antagonists

  • Adenosine effects include reduction in renal blood flow and GFR through constriction of afferent arteriole, as well as enhancement in sodium and water reabsorption, thus contributing to diuretic resistance and renal dysfunction. Adenosine levels are demonstrated to be increased in HF. Therefore, promising results were expected from adenosine A1 receptor antagonists, supported by previous studies [7981].

  • Unfortunately, data from the recent, large, phase 3 Placebo-Controlled Randomized Study of the Selective A(1) Adenosine Receptor Antagonist Rolofylline for Patients Hospitalized With Acute Decompensated Heart Failure and Volume Overload to Assess Treatment Effect on Congestion and Renal Function (PROTECT) demonstrated no favorable effects of the adenosine antagonist, rolofylline, on outcomes and renal function [82, 83•].


  • Relaxin is a peptide hormone naturally secreted in pregnant women. Its action via nitric oxide pathways and endothelin B receptors leads to systemic and renal vasodilation, in particular increasing GFR during pregnancy. A more potent vasodilating effect has been observed in the setting of preconstricted vessels.

  • Reproducing this effect could be potentially useful in AHF, and previous nonclinical trials seemed to confirm this beneficial effect [84]. A preliminary double-blind, placebo-controlled, phase 2 trial (preRELAX) [19] was performed in AHF patients with high-normal blood pressure and mild to moderate renal dysfunction. Relaxin was associated with favorable relief of dyspnea and other clinical outcomes, and a tendency to greater weight loss with smaller doses of diuretics and nitrates. The most effective dose (30 μg/kg) was also associated with stable renal function. Higher dosages did not substantially improve other clinical outcomes and showed a greater but non-significant association with persistent renal impairment [19].

  • The ongoing Efficacy and Safety of Relaxin for the Treatment of Acute Heart Failure (RELAX-AHF) trial will provide more definitive information of the impact of relaxin on congestion, renal dysfunction, and outcomes in AHF [85•].


WRF in the setting of AHF remains a challenging clinical problem. At present, clinical tools are not sufficient to distinguish various causes and the clinical significance of WRF at the bedside. A greater insight into pathophysiology, and better distinction between changes in renal filtration and intrinsic kidney injury, are necessary to develop more targeted therapies. The current treatment options for acute CRS attempt to restore fluid balance with diuretics and concomitantly decrease neurohormonal activation. The development of new drugs that both treat AHF and provide renal protection remains an attractive, but as yet unrealized, therapeutic strategy.


Dr. V. Lazzarini: none. Dr. G. Michael Felker has served as a consultant for Amgen, Novartis, Trevana, Medtronic, and St. Jude’s, and has received grants from Roche Diagnostics, BG Medicine, Critical Diagnostics, Otsuka, and Amgen.

Copyright information

© Springer Science+Business Media, LLC 2012