**2. Definitions and sub-types of the cardiorenal syndrome**

Historically, the CRS is thought to have been due to impaired renal perfusion secondary to low cardiac output states or the result of HF therapies negatively impacting renal function. In 2004, the National Heart, Lung and Blood Institute defined CRS as a "state in which therapy to relieve heart failure symptoms is limited by further worsening in renal function" (National Heart, Lung and Blood Institute, 2004). By this paradigm, the heart was considered to be the central driving force behind impaired renal function in patients with HF.

Our understanding of the pathophysiology behind the CRS has evolved in the last number of years and there is increasing recognition of the complexity of interactions which exist between the heart and the kidneys, particularly when either or both organs are diseased. This organ cross talk is bidirectional in nature and the resultant dialogue is dependent on whether the heart or the kidney is the primary affected organ as well as the time course over which the associated pathophysiological changes may occur.

It is within this context, that newer definitions for the CRS have been proposed which recognize that either the heart or kidney may be the primary site of organ injury. A more comprehensive definition and classification schema for the CRS has the advantage of allowing clinicians to make a more accurate diagnosis which in turn informs our understanding of a given patient's natural history, prognosis and optimal treatment strategy.

The definition and classification system for CRS introduced by Ronco and colleagues in 2008 (Ronco et al., 2008) is now widely considered to be the preferred mechanism for describing patients and the pathophysiological processes associated with CRS. Ronco and colleagues broadly define CRS as "a pathophysiological disorder of the heart and kidneys whereby acute or chronic dysfunction of one organ may induce acute or chronic dysfunction of the other." Additionally, they characterize five sub-types of the CRS based on this definition. These are described and discussed below. It should be noted that CRS types 1-5 may frequently co-exist in a given patient, underscoring the complexity of interaction between the heart and kidney and the importance of appointing chronology to these processes.

#### **2.1 Cardiorenal syndrome type 1 (acute cardiorenal syndrome)**

Type 1 CRS is distinguished by an acute deterioration in cardiac function or acute cardiac injury, from any cause, that secondarily results in acute kidney injury (AKI).

of standard HF medications, such as angiotensin converting enzyme inhibitors, angiotensin receptor blockers and beta blockers in patients with HF and CKD, regardless of glomerular

Many novel therapies for HF have been introduced over recent years, several of which were appealing for treatment of the CRS, given the pathophysiological processes towards which they were directed. Unfortunately, natriuretic peptides, vasopressin antagonists, and adenosine antagonists have all failed to show meaningful clinical benefits in patients with HF (Hernandez, 2010; Konstam et al., 2007; Massie et al., 2010). Other approaches, particularly peripheral ultrafiltration, have shown more promise in this patient population

Historically, the CRS is thought to have been due to impaired renal perfusion secondary to low cardiac output states or the result of HF therapies negatively impacting renal function. In 2004, the National Heart, Lung and Blood Institute defined CRS as a "state in which therapy to relieve heart failure symptoms is limited by further worsening in renal function" (National Heart, Lung and Blood Institute, 2004). By this paradigm, the heart was considered to be the central driving force behind impaired renal function in patients with

Our understanding of the pathophysiology behind the CRS has evolved in the last number of years and there is increasing recognition of the complexity of interactions which exist between the heart and the kidneys, particularly when either or both organs are diseased. This organ cross talk is bidirectional in nature and the resultant dialogue is dependent on whether the heart or the kidney is the primary affected organ as well as the time course over

It is within this context, that newer definitions for the CRS have been proposed which recognize that either the heart or kidney may be the primary site of organ injury. A more comprehensive definition and classification schema for the CRS has the advantage of allowing clinicians to make a more accurate diagnosis which in turn informs our understanding of a given patient's natural history, prognosis and optimal treatment

The definition and classification system for CRS introduced by Ronco and colleagues in 2008 (Ronco et al., 2008) is now widely considered to be the preferred mechanism for describing patients and the pathophysiological processes associated with CRS. Ronco and colleagues broadly define CRS as "a pathophysiological disorder of the heart and kidneys whereby acute or chronic dysfunction of one organ may induce acute or chronic dysfunction of the other." Additionally, they characterize five sub-types of the CRS based on this definition. These are described and discussed below. It should be noted that CRS types 1-5 may frequently co-exist in a given patient, underscoring the complexity of interaction between the heart and kidney and the importance of appointing chronology to these processes.

Type 1 CRS is distinguished by an acute deterioration in cardiac function or acute cardiac injury, from any cause, that secondarily results in acute kidney injury (AKI).

filtration rate (GFR) (Berger et al., 2007; Cice et al., 2003).

**2. Definitions and sub-types of the cardiorenal syndrome** 

which the associated pathophysiological changes may occur.

**2.1 Cardiorenal syndrome type 1 (acute cardiorenal syndrome)** 

(Costanzo et al., 2005).

HF.

strategy.

Pathophysiologically, Type 1 CRS is characterized by decreased cardiac output with impaired renal perfusion as well as elevated central venous pressures and acute renal edema. Renal ischemia may be mediated by decreased oxygen delivery due to impaired myocardial contractile performance, elevated interstitial pressures in the renal medulla and by peripheral/systemic vasoconstriction which occurs as a compensatory mechanism in the face of low cardiac output.

Historically, decreased forward cardiac flow was thought to be the primary determinant for AKI in this context, however recent clinical trials have suggested this mechanism may not be as important in the development of CRS Type I as previously hypothesized. Specifically, data from ADHERE (Acute Decompensated Heart Failure National Registry) which included over 100,000 patients admitted to hospital in the United States with acute decompensated heart failure (ADHF) showed that <2% of patients had systemic hypotension, a surrogate for low cardiac output, while the vast majority of patients had symptoms/signs of volume overload (Adams et al., 2005). This is corroborated by the findings of the ESCAPE (Evaluation Study of Congestive Heart Failure and Pulmonary Artery Catheterization Effectiveness) Trial in which 433 patients admitted to hospital with ADHF were randomized to pulmonary artery catheterization versus standard care to assess the efficacy of tailored haemodynamic therapy (Binanay et al., 2005). In the ESCAPE Trial, cardiac index was not associated with baseline renal function or deterioration in renal function, however right atrial pressure was weakly correlated with baseline creatinine and GFR (Nohria et al., 2008).

The impact of central venous pressures (CVP) on worsening renal function in the setting of ADHF has been receiving greater attention in recent years. Elevated CVP is more predictive of a decline in renal function than other relevant haemodynamic variables such as cardiac index, blood pressure and pulmonary capillary wedge pressure (Mullens et al., 2009). Moreover, elevated CVP predicts risk of re-hospitalization for HF and death suggesting that it is a potent prognosticator for poor outcomes and a potential target for therapy (Uthoff et al., 2011). Elevated intra-abdominal venous pressures have also been shown to have a similar relationship with GFR at baseline and changes in GFR with therapy (Bock & Gottlieb, 2010; S. E. Bradley & G. P. Bradley). This may be the result of a direct mechanical effect on renal blood flow or simply a reflection of elevated CVP.

Among patients with ADHF, activation of the sympathetic nervous system (SNS) and the renin-angiotensin-aldosterone system (RAAS) is a homeostatic mechanism intended to maintain intraglomerular perfusion pressures and preserve GFR. Paradoxically however, systemic vasoconstriction by these mechanisms increases cardiac afterload leading to further decline in cardiac output and renal blood flow. Additionally, these neurohormones have a maladaptive effect on the myocardium resulting in fibrosis and ventricular remodeling. Treatment with β-blockers is relatively contra-indicated in the face of an acute decompensation due to their negative inotropic effects and the relative dependence of cardiac output on heart rate in this patient population; therefore, SNS activation in CRS Type 1 may remain unchecked leading to ischemia of both renal and cardiac tissue beds.

Acute administration of RAAS inhibition may exacerbate renal injury in CRS Type 1 by reducing pressure in Bowman's capsule; this effect may be magnified in the presence of volume shifts associated with diuretics, which remains the mainstay of therapy. Moreover, diuretics may directly result in additional neurohormonal activation and there is now an

Sub-Types and Therapeutic Management of the Cardiorenal Syndrome 127

hospitalization solely to CRS Type 1 thereby underestimating the presence of concomitant CRS Type 2 in this cohort of patients. Novel biomarkers of AKI may help clinicians to decipher the relative contributions of CRS Type 1 versus CRS Type 2 in patients hospitalized for HF who have poor renal function upon presentation (Siew et al., 2011; Coca et al., 2008). Regardless of cause, renal insufficiency in hospitalized HF patients appears to be relatively common; among those enrolled in ADHERE, the prevalence of at least moderate renal impairment, as determined by GFR, was greater than 60% at baseline (Heywood et al., 2007). This is in sharp contrast to initial reports from the same registry which suggested a prevalence rate of only 20% when a serum creatinine of 2.0 mg/dL was employed as a cut off (Adams et al., 2005). Calculation of GFR, therefore, is paramount to accurately

The true burden of pre-existing renal dysfunction among patients with HF was best characterized in a meta-analysis performed by Smith and colleagues. In their systematic review of the literature, approximately 80,000 hospitalized and non-hospitalized patients with HF were identified across 16 clinical trials. While 29% of patients were found to have moderate to severe renal impairment (GFR <53 mL/min or cystatin C of >1.56 mg/dL), 63% were found to have at least some degree of impaired kidney function. Moreover, these findings are likely to underestimate the true prevalence of renal insufficiency in HF populations given that 8 of the clinical trials included in the meta-analysis excluded patients

In the meta-analysis performed by Smith and colleagues, renal impairment at baseline conferred an increased risk of mortality at one year follow-up compared to patients with normal kidney function (Smith et al., 2006). The adjusted hazard ratio for patients with any renal impairment or moderate to severe renal impairment was 1.56 and 2.31 respectively. Excess risk was conferred in an incremental fashion with each 10 mL/min reduction in GFR correlating to a 7% increase in the risk of death. This observation is strengthened by similar findings across a spectrum of clinical trials in both hospitalized and ambulatory HF populations (Adams et al., 2005; Dries et al., 2000; Fonarow et al., 2005; Heywood et al., 2007;

Many of the pathophysiological mechanisms which characterize CRS Type 1 are also implicated in the development of CRS Type 2, although many of these processes may occur slowly and over longer periods of time. For example, elevated central venous pressure is strongly associated with a decline in eGFR among patients with chronic HF (Damman et al., 2009; Firth et al., 1988); as described above, the same is true for patients with ADHF and CRS Type 1. Elevated CVP and secondarily an elevation in renal venous pressure may trigger a number of downstream events, including interstitial ischemia, neurohormonal activation and decreased responsiveness to natriuretic peptides which all combine to reduce GFR directly or indirectly (Damman et al., 2007; Bock & Gottlieb, 2010) in the setting of chronic HF. Chronically low cardiac output, particularly in combination with micro and macrovascular renal disease, may also contribute to fibrosis and structural changes in the

RAAS activation occurs in both HF and CKD with an associated increase in Angiotensin II levels (AII). AII mediates oxidative injury and endothelial dysfunction through both the formation of reactive oxygen species and a decrease in nitric oxide bioavailability. Each of

on the basis of age or an elevated serum creatinine at baseline (Smith et al., 2006).

identifying the burden of renal disease in all forms of CRS.

Hillege et al., 2006).

kidney which result in impaired renal function.

increasing body of literature suggesting that diuretics, in and of themselves, may be associated with worse outcomes in patients with ADHF independent of other relevant clinical variables. In a single centre retrospective analysis of 1354 patients admitted with ADHF, Eshaghian and colleagues (Eshaghian et al., 2006) demonstrated that patients requiring the highest doses of diuretics, stratified by quartiles, had higher rates of sudden death, death due to progressive pump failure and all cause mortality compared to patients in the lowest quartile of diuretic dose. This type of observation has fueled a growing interest in identifying alternate strategies for fluid management in the acute setting, independent of diuretic administration (see section 3.8).

Of particular concern among patients who present with the features of CRS Type 1 is the impact of diagnostic imaging and invasive cardiac procedures which may have an additional and direct toxic effect on the kidneys through a variety of mechanisms. Individuals who present with an acute deterioration in cardiac function will frequently require imaging or investigation to identify a precipitant or cause for their symptoms. Independently, percutaneous interventions and cardiac surgery impart a risk of AKI which is higher in patients who have pre-existing or concomitant acute renal insufficiency (Anderson et al., 1999; Best et al., 2002).

Upwards of 70% of patients admitted to hospital with ADHF will experience a rise in serum creatinine over the course of their admission (Gottleib et al., 2002); this may be the result of therapies administered, either medical or invasive, or a consequence of the various pathophysiological processes which characterize CRS Type 1. Regardless of mechanism, worsening renal function portends a poor prognosis and is associated with higher mortality rates. (Gottleib et al., 2002; Damman et al., 2007).
