Portal de Periódicos da CAPES

Cardiorenal Syndrome

2010; Lippincott Williams & Wilkins; Volume: 121; Issue: 23 Linguagem: Inglês

10.1161/circulationaha.109.886473

1524-4539

Jeremy S. Bock, Stephen S. Gottlieb,

Cardiovascular Function and Risk Factors

HomeCirculationVol. 121, No. 23Cardiorenal Syndrome Free AccessReview ArticlePDF/EPUBAboutView PDFView EPUBSections ToolsAdd to favoritesDownload citationsTrack citationsPermissionsDownload Articles + Supplements ShareShare onFacebookTwitterLinked InMendeleyReddit Jump toSupplementary MaterialsFree AccessReview ArticlePDF/EPUBCardiorenal SyndromeNew Perspectives Jeremy S. Bock, MD and Stephen S. Gottlieb, MD Jeremy S. BockJeremy S. Bock From the Division of Cardiology, Department of Medicine, University of Maryland School of Medicine, Baltimore, Md. and Stephen S. GottliebStephen S. Gottlieb From the Division of Cardiology, Department of Medicine, University of Maryland School of Medicine, Baltimore, Md. Originally published15 Jun 2010https://doi.org/10.1161/CIRCULATIONAHA.109.886473Circulation. 2010;121:2592–2600Maintenance of blood volume, vascular tone, and hemodynamic stability depends on a set of elegant interactions between the heart and kidney. For some time, physicians have recognized that severe dysfunction in either of these organs seldom occurs in isolation. However, only recently have we attempted to define and apply the widespread concept of the cardiorenal syndrome (CRS). Despite our growing use of the term, there is still some debate as to its true definition. More important, the process itself remains enigmatic; our understanding of the complex physiological, biochemical, and hormonal derangements that encompass the CRS is woefully deficient and may lead to improper medical management of patients.Because renal dysfunction portends such a dismal prognosis in cardiac failure and vice versa, there has been a recent surge of interest in identifying precise pathophysiological connections between the failing heart and kidneys. Understanding the mechanisms involved in the CRS will allow us to target therapies that interrupt this dangerous feedback cycle.In 2004, a working group of investigators at the National Heart, Lung, and Blood Institute defined the CRS as a state in which therapy to relieve heart failure (HF) symptoms is limited by further worsening renal function.1 Although this definition is succinct and understandable and probably reflects the most common use of the term, some authors argue that it is simplistic to the point of being inaccurate.2 Several groups have recently proposed that the definition of CRS be broadened in an attempt to stress the complex and bidirectional nature of pathophysiological interactions between the failing heart and kidneys. That is, each dysfunctional organ has the ability to initiate and perpetuate disease in the other organ through common hemodynamic, neurohormonal, and immunologic/biochemical feedback pathways.Proper use of the term CRS should correct a common misunderstanding: that kidney dysfunction in HF is a direct consequence of impaired renal blood flow in the setting of depressed left ventricular systolic function. Recent investigations do not support this as the sole derangement in CRS. Increasing evidence supports the roles of central venous congestion, neurohormonal elaboration, anemia, oxidative stress, and renal sympathetic activity as other potential contributors to this complex syndrome. This review stresses the ways in which the heart and kidney interact, often in a deleterious manner.Epidemiology and Outcomes in Combined Cardiorenal Disease: The Scope of the ProblemPrevalence of Renal Disease in Patients With HFIn the Acute Decompensated Heart Failure National Registry (ADHERE) of >105 000 individuals admitted for acute decompensated HF, 30% had a history of renal insufficiency, 21% had serum creatinine concentrations >2.0 mg/dL, and 9% had creatinine concentrations >3.0 mg/dL.3 McAlister et al4 found that only 17% of 754 outpatients with HF had creatinine clearances >90 mL/min. In their cohort, 39% with New York Heart Association (NYHA) class IV symptoms and 31% with NYHA class III symptoms had creatinine clearance 1 million hospital admissions for decompensated HF in the United States annually.Impact of Renal Disease on Clinical Outcomes in Patients With HFRenal dysfunction is one of the most important independent risk factors for poor outcomes and all-cause mortality in patients with HF. Baseline glomerular filtration rate (GFR) appears to be a stronger predictor of mortality in patients with HF than left ventricular ejection fraction or NYHA functional class. Both elevated serum creatinine on admission and worsening creatinine during hospitalization predict prolonged hospitalization, rehospitalization, and death.5,6 Even small changes in creatinine <0.3 mg/dL are common (Figure 1) and have been associated with increased mortality and prolonged hospitalization.7Download figureDownload PowerPointFigure 1. The frequency and time course of developing an increase in creatinine in patients hospitalized with HF. The percent of patients with an increase (by that time in the hospitalization) in creatinine of at least the value indicated is shown. Worsening renal function is common in patients with HF. Reprinted from Gottlieb et al.7HF Outcomes in Patients With Renal DiseaseOn the basis of estimates provided by the Third National Health and Nutrition Examination Survey (NHANES III), almost 8 million individuals living in the United States have a GFR <60 mL/min.8 Patients with chronic renal insufficiency are at strikingly higher risk for myocardial infarction, HF with systolic dysfunction, HF with preserved left ventricular ejection fraction, and death resulting from cardiac causes compared with individuals with normal GFR.9 A recent meta-analysis suggests that individuals with primary renal disease are more likely to eventually die of cardiovascular causes than renal failure itself.10 This is not just secondary to atherosclerotic disease; in a multicenter cohort study of 432 patients, 31% planning to initiate hemodialysis had HF symptoms, and 33% of such patients had estimated left ventricular ejection fraction <40%.11 Patients with HF and new hemodialysis had a median survival of only 36 months compared with 62 months in patients without HF. Furthermore, 25% who did not have HF symptoms on initiation of dialysis developed these symptoms after a median follow-up of 15 months. Conversely, reversal of renal dysfunction can improve cardiac function. In a study of 103 hemodialysis patients with HF and left ventricular ejection fraction <40%, the mean ejection fraction increased from 32% to 52% after renal transplantation, and 70% had normalization of cardiac function.12Hypertensive heart disease and HF with a normal ejection fraction are common among individuals with advanced and end-stage renal disease. One study showed that there is echocardiographic evidence of left ventricular hypertrophy in 45% of individuals with creatinine clearance 100 years ago.19,20 In one such early experiment, Winton20 observed that urine formation by isolated canine kidney was markedly reduced at renal venous pressures of 20 mm Hg and abolished at pressures >25 mm Hg. Renal blood flow was also diminished in proportion to the decrease in pressure gradient across the afferent and efferent renal circulations, probably caused by the increased efferent arterial pressure. Rising renal venous pressure limited urine formation and renal blood flow more than a reduction in arterial pressure. Elevation of renal venous pressure from extrinsic compression of the veins has also been shown to compromise renal function.21 More than 60 years ago, Bradley and Bradley22 showed that abdominal compression to produce IAP of 20 mm Hg in normal individuals markedly reduced GFR and renal plasma flow. These relationships are supported by modern in vivo animal models.23 In recent years, there has also been increasing recognition that oliguric acute renal dysfunction frequently accompanies abdominal compartment syndrome in surgical and trauma patients.24 These changes are promptly reversed by abdominal decompression and may be associated with subsequent polyuria.An international panel recently defined elevated IAP as pressure ≥8 mm Hg and intraabdominal hypertension as pressure ≥12 mm Hg.25 In a recent study, 24 of 40 consecutive patients admitted for acute decompensated HF (mean left ventricular ejection fraction, 19%) had an IAP ≥8 mm Hg.17 None of the 40 patients in the cohort complained of abdominal symptoms at study entry. Patients with elevated IAP had significantly lower baseline GFR compared with those with normal IAP, and the degree of reduction in IAP after diuresis predicted an improvement in renal function (Figure 2). Other initial hemodynamic parameters such as pulmonary capillary wedge pressure and cardiac index were not different between patients with elevated IAP and those with normal IAP. The concept that venous congestion, not arterial blood flow, is an important mediator of cardiorenal failure is supported by the findings of the Evaluation Study of Congestive Heart Failure and Pulmonary Artery Catheterization Effectiveness trial, in which only baseline right atrial pressure, not arterial blood flow, correlated with baseline serum creatinine.15Download figureDownload PowerPointFigure 2. The relationship between changes in IAP with diuresis and the change in serum creatinine. The close relationship suggests that increased IAP may cause renal dysfunction. Reprinted with permission from Mullens et al.17In considering whether elevated IAP in congestive heart failure is a true culprit in the development of progressive renal dysfunction or an innocent bystander, several mechanisms by which abdominal pressure might contribute to CRS have been explored. Elevation of renal parenchymal pressure does not appear to have significant effects on GFR or renal blood flow. This was shown in studies of isolated porcine kidneys subjected to increasing amounts of extrinsic pressure.26 In contrast, elevated central and renal venous pressures offer a stronger explanation for the relationship between elevated IAP and renal dysfunction. Elevating renal venous pressure by 30 mm Hg for 2 hours in intact porcine kidneys resulted in a substantial reduction in renal blood flow and GFR.23 Furthermore, patients with HF with impaired renal function at baseline or worsening renal function during hospitalization have significantly elevated central venous pressure relative to those with less renal impairment18,27 (Figure 3). In one study of intensive medical therapy directed at volume reduction, hemodynamic profiles were monitored in all patients with pulmonary artery catheters, and only elevated central venous pressure correlated with worsening versus preserved renal function.18 The role of elevated central and renal venous pressures is further supported by the association of elevated jugular venous pulsations on physical examination with higher baseline serum creatinine and increased risk for hospitalization and death caused by pump failure.28 Finally, the association of tricuspid regurgitation with renal dysfunction was recently examined in 196 consecutive patients with HF.29 The authors found that patients with at least moderate tricuspid regurgitation by transthoracic echocardiography had lower estimated GFR and that a linear relationship existed between severity of tricuspid regurgitation and degree of GFR impairment. Download figureDownload PowerPointFigure 3. Distribution of central venous pressure (CVP) and the relationship between CVP and estimated GFR in 2557 patients. CVP has repeatedly been shown to correlate well with renal dysfunction in patients with HF. Reprinted with permission from Damman et al.27Sympathetic OveractivityThe adverse consequences of sympathetic nervous system activity are well known. Sustained elevated adrenergic tone causes a reduction in β-adrenergic receptor density, particularly β1, within the ventricular myocardium, as well as uncoupling of the receptor from intracellular signaling mechanisms. Less well appreciated are the systemic effects of renal sympathetic stimulation. As left ventricular systolic failure progresses, diminished renal blood flow and perfusion pressure (whether from arterial underfilling or renal venous congestion) lead to baroreceptor-mediated renal vasoconstriction, activation of the renal sympathetic nerves, and release of catecholaminergic hormones. This problem is compounded in patients with HF with advanced renal insufficiency because there is reduced clearance of catecholamines by the kidneys.30There are now good data to suggest that the renal sympathetic activation leads to direct vascular effects. A recent pilot study of catheter-based renal sympathetic denervation in patients with resistant hypertension found significant improvements in GFR in 24% of patients undergoing the procedure.31 Bilateral renal nerve ablation has also been shown to reduce renal norepinephrine spillover, renin activity, and systemic blood pressure 12 months later32 (Figure 4). Although this intervention has not been tested specifically in an HF population, denervation could possibly affect renal function and halt renal sympathetic nerve-mediated progression of cardiac failure related to elaboration of catecholamines and the RAAS. Further investigation into this exciting concept is needed to determine whether it is clinically relevant. Download figureDownload PowerPointFigure 4. The change in blood pressure after radiofrequency ablation of renal sympathetic nerves. The decrease in blood pressure suggests systemic effects from renal sympathetic nerve activity. Reprinted with permission from Krum et al.31Renin-Angiotensin-Aldosterone Axis and Renal DysfunctionThe extreme sodium avidity and ventricular remodeling conferred by RAAS elaboration in HF are a maladaptive response to altered hemodynamics, sympathetic signaling, and progressive renal dysfunction. The benefits of angiotensin-converting enzyme (ACE) inhibition and aldosterone antagonism through blockade of the intracardiac RAAS, reduction in adrenergic tone, improvement in endothelial function, and prevention of myocardial fibrosis are well described in cardiac failure; RAAS inhibition has been a main focus of therapy in HF for the last 2 decades and has led to improved outcomes for many patients. Unfortunately, little is known about the long-term benefits or adverse effects of RAAS inhibition on kidney function in HF.ACE inhibitors and angiotensin receptor blockers have important renoprotective effects in hypertensive patients with nondiabetic renal disease and individuals with diabetic nephropathy.33 In contrast, whether there is a renoprotective role of ACE inhibitors and angiotensin receptor blockers in systolic HF that is independent of direct preservation of ventricular function has not been established. ACE inhibitors and angiotensin receptor blockers cause dose-dependent increases in angiotensin II (AT-II).34 This may contribute to the phenomenon described as escape from ACE inhibition.35 Significantly, AT-II directly contributes to kidney damage. AT-II upregulates the cytokines transforming growth factor-β, tumor necrosis factor-α, nuclear factor-κB, and interleukin-6 and stimulates fibroblasts, resulting in cell growth, inflammation, and fibrotic damage in the renal parenchyma.36,37Oxidative Injury and Endothelial DysfunctionNeurohormones are strong precipitants and mediators of an oxidative injury cascade that leads to widespread endothelial dysfunction, inflammation, and cell death in the CRS. AT-II seems to be particularly important in this process, exerting many deleterious effects through the activation of NADPH oxidase and NADH oxidase. AT-II activates these 2 enzymes within vascular smooth muscle cells, cardiac myocytes, and renal tubular epithelial cells, generating superoxide, a reactive oxygen species.38–40 Reactive oxygen species have many unfavorable effects in living tissues and likely contribute to the processes of aging, inflammation, and progressive organ dysfunction. Growing evidence supports oxidative injury as a common link between progressive cardiac and renal dysfunction. Because both primary cardiac failure and primary renal failure lead to elaboration of the RAAS, activation of oxidases by AT-II in one organ has the potential to lead to progressive dysfunction in the secondary organ through reactive oxygen species generation.Inactivation of nitric oxide is a particularly important effect of superoxide and other reactive oxygen species. Decreased bioavailability of nitric oxide may partially explain the endothelial dysfunction observed in vascular smooth muscle and abnormal contractile properties of cardiac myocytes in HF. There is heightened NADPH oxidase activity in explanted failing hearts compared with healthy hearts awaiting implantation,39 and high-dose antioxidant agents attenuate left ventricular remodeling after experimental ligation of the left anterior descending coronary artery.41 Dahl salt-sensitive rats with systolic HF have substantial elevations in AT-II and NADPH oxidase expression and reduced nitric oxide production in kidney tissue compared with control animals without experimental HF.42 Interestingly, these changes were prevented with the ACE inhibitor imidapril. Other groups have shown that both ACE inhibitors and angiotensin receptor blockers increased the availability of nitric oxide through upregulation of superoxide dismutase.43 These observations provide a good example of dysfunction in a secondary organ, in this case kidney, associated with primary disease in another organ.Erythropoietin and the Cardiorenal-Anemia SyndromeAnemia is common in individuals with chronic kidney disease and HF and may contribute to the abnormal renal oxidative state; hemoglobin is an antioxidant. Although anemia should induce increased erythropoietin, there is evidence that decreased concentrations in patients with CRS may directly exacerbate the renal abnormalities. Therefore, the combination of anemia and decreased erythropoietin may exacerbate the underlying factors causing CRS.The high frequency of anemia in CRS and HF has repeatedly been demonstrated.44 In the Organized Program to Initiate Lifesaving Treatment in Hospitalized Patients With Heart Failure (OPTIMIZE-HF) registry, 51% of the nearly 50 000 patients with HF had hemoglobin ≤12 g/dL and 25% had hemoglobin between 5 and 10.7 g/dL.45 Patients with HF with anemia had increased mortality, length of hospital stay, and hospital readmission rates compared with nonanemic patients with HF. It should be noted that anemia in advanced kidney diseases is due to an absolute deficiency in erythropoietin production. HF alone, on the other hand, may be marked by insensitivity to elevated erythropoietin concentrations secondary to sustained inflammation.46 Patients with both HF and kidney disease, however, may have low erythropoietin concentrations.The lack of erythropoietin could exacerbate HF in multiple ways.47 In cardiac cells, erythropoietin can prevent apoptosis and increase the number of cardiomyocytes.48 Similar observations have been made in renal cells.49 Although it is unclear what effect erythropoietin has on nitric oxide synthesis, it does appear to decrease oxidative stress.47 Small studies suggest that these actions might exert clinical benefit. In a single-center prospective trial, 32 anemic NYHA class II to IV patients were randomized to receive erythropoietin and intravenous iron or routine management. After a mean follow-up of 8 months, patients with active treatment demonstrated improved ejection fraction by multigated acquisition, decreased diuretic requirements, unchanged serum creatinine, and improvements in NYHA functional class. Control patients had worsened ejection fraction, worsening serum creatinine, and deterioration in NYHA functional class.50 Unfortunately, this study was not placebo controlled or blinded. Other studies have focused on clinical benefits and have not carefully evaluated possible mechanisms.At this point, it is therefore unclear whether anemia is a marker of progressive heart and renal failure or a true mediator of the CRS. Further long-term study is needed to address the interesting possibility that treatment of anemia in HF may improve renal function. Studies of patients with advanced renal disease suggest that partial correction of anemia leads to improved quality of life, reduced progression to end-stage renal disease, and reduced mortality.51 Because aggressive correction of anemia in this population has been associated with high rates of adverse events,52 exploring the utility of correcting anemia in patients with HF should be done with caution.Other Renal TargetsArginine vasopressin is a nonapeptide that is released by oncotic stimuli but also by blood pressure and cardiac factors. Concentrations are increased in HF and could lead to water retention and hyponatremia.53 Furthermore, it has vasoactive effects (mainly through V1 receptors) that could be important. More clearly relevant to patients with HF is that activation of the V2 receptor increases the permeability to water of the renal collecting tubular cells, resulting in water retention. Vasopressin antagonists have been shown to lead to more aquaresis and resolution of hyponatremia, with some weight loss and improvement in overall fluid balance. However, these effects have not resulted in clear demonstrable clinical benefit or improvement in renal function.54 At present, vasopressin appears important as a cause of water retention in some patients but does not appear integral to renal function in these patients.The importance of adenosine as a mediator of the CRS is also not known. Adenosine-A1 receptors are found in afferent arterioles, juxtaglomerular cells, the proximal tubule, and thin limbs of Henle, and GFR and urine output could improve by countering the effects of adenosine. Indeed, adenosine concentrations are increased in patients with HF.55 Initial studies suggested that this mechanism was important. An adenosine-A1 antagonist, BG9719, maintained creatinine clearance while permitting diuresis.56 In a crossover study of another adenosine-A1 antagonist, rolofylline, GFR increased by 32% with active drug, and renal plasma flow increased by 48%.57 Unfortunately, the pivotal Prophylaxis of Thromboembolism in Critical Care Trial (PROTECT), recently presented at the European Society of Cardiology (2009), showed no beneficial effects in patients with acute decompensated HF. The reason for the very different results between the early studies and PROTECT is unknown. It could reflect the lack of importance of adenosine as a mediator of the CRS or could indicate problems with the drug or the particular patient population studied.In contrast to most of the neurohormones discussed, there is no suggestion that endogenous natriuretic peptides worsen cardiac or renal function. However, the lack of response of many patients with HF raises questions as to why endogenous natriuretic peptides are not effective. With stimulation of cGMP, these substances (both endogenous and pharmacological) would be expected to increase urine output and to improve renal blood flow. However, it is unclear what their effects are in patients with HF. Possible reasons for different actions include the consequences of lowered blood pressure, altered degradation leaving inactive peptides that are nonetheless assayed, and changes in the target.58Therapeutic ImplicationsFactors Influencing Medication UseCardiac failure and renal failure have synergistic effects that magnify the poor outcomes associated with either disease alone. However, physicians may also have a role in these poor outcomes in that we are often reluctant to prescribe or titrate valuable medications.16,59 Slight elevation in creatinine concentration during diuretic treatment of decompensated congestive HF may be seen as depletion of intraarterial volume or "overdiuresis" and limit more aggressive diuresis. Such patients are frequently discharged from the hospital with inadequate resolution of symptoms and thus have high short-term rehospitalization rates.Inpatients with acute decompensated HF are often not started on ACE inhibitor therapy at discharge for fear of worsening serum creatinine.60 Recognition that elevated serum creatinine portends worse outcomes in HF prompts physicians to be concerned about the renal effects of these agents. However, the benefits of ACE inhibitor use are clear and outcomes are extremely poor in individuals with HF in whom ACE inhibitors are held. Although it is possible that the prognostic importance of the lack of ACE inhibitors is partly reflective of the severity of disease in these patients, it must also be recognized that ACE inhibitors and angiotensin receptor blockers can lead to decreased renal function even in patients who benefit from their use; mean serum creatinine increased even though outcomes were better in the Cooperative North Scandinavian Enalapril Survival Study (CONSENSUS).61 With diuresis, serum creatinine is more likely to increase in patients receiving ACE inhibition and in those with the lowest blood pressures.61 These data suggest that some increase in creatinine should be tolerated with the use of ACE inhibition, and other interventions (such as decreased diuresis) might be needed to accomplish this. The advantage of ACE inhibitors in delaying progression and death in HF is undeniable, and their use should be encouraged unless detrimental effects are clearly proven.Fluid Removal and Renal EffectsDiuretics are commonly used in HF and appear necessary, but there are suggestions that they might be detrimental. Furosemide decreases GFR in many patients.56 Higher doses of loop diuretics are also associated with elevated serum creatinine and reduced survival in the HF population, but this might just reflect the need to use higher doses in the sickest patients.62 More disturbing are their effects on neurohormone