CHAPTER 105: THE KIDNEY IN HEART DISEASE

Cardiac and renal physiology are intimately intertwined. Impaired cardiac function can lead to kidney injury and vice versa. Furthermore, many cardiovascular therapeutic interventions may cause renal injury. Impaired renal function is common among patients with heart disease and is one of the strongest predictors of worse outcome for these patients. Reports suggest that renal disease is underdiagnosed among patients with heart disease, which increases the risk of avoidable adverse outcomes.^1,2,3,4

Renal Physiology and Important Concepts

The main functions of the kidneys are to maintain water balance and body fluid homeostasis and to eliminate metabolic waste products. This is achieved by filtration of plasma in the glomerulus, as well as by secretion and reabsorption of water and solutes along the renal tubules. An average human filters 150 to 200 L of plasma across the glomeruli and produces 0.5 to 2.5 L of urine per day. Urine output greater than 2.5 L/d is termed polyuria; urine output less than 400 mL/d is termed oliguria; and urine output less than 100 mL/d is termed anuria.

Renal Clearance

The concept of renal clearance is based on Fick’s principle, predicated on the law of conservation of mass. A substance that is neither synthesized nor metabolized within the kidney must enter and leave the kidney in equal amounts. A substance can enter the kidney only via the renal artery but can leave the kidney either via the renal vein or via the urine. Renal clearance refers to the excretory function of the kidney and therefore considers only the rate at which a substance is excreted into the urine, not the rate at which it is returned to the circulation through the venous system. In terms of mass balance, the excretion rate of substance y (U_y*V), where U_y is the concentration of y in the urine and V is urine volume/time, is proportional to the plasma concentration of y (P_y). In a steady state, the rate at which y is moved from the plasma to the urine is the renal clearance (C_y):

Hence, renal clearance represents a volume of plasma from which the substance of interest has been removed and excreted into the urine per unit time. For substances that are eliminated from the body predominantly through the kidney, plasma concentrations in blood samples from peripheral venous blood will closely approximate arterial concentrations. Depending on the properties of the substance measured, renal clearance can be used to estimate glomerular filtration rate (GFR) and renal plasma flow (RPF). Substances such as inulin, which are only filtered by the glomerulus, are the gold standard for measurement of clearance, but because they need to be administered intravenously are not practical for routine clinical use. Creatinine, while also excreted and reabsorbed to some extent in the tubules, provides sufficient accuracy for routine estimation of renal function, especially with 24-hour urine collection.

Glomerular Filtration Rate and Creatinine Clearance

Glomerular ultrafiltration is the first step of urine formation. The glomerular filtration barrier restricts the filtration of molecules on the basis of size and molecular charge (ie, the reflection coefficient [σ] for these molecules across the glomerular membrane is ~ 1). As a consequence, the plasma ultra-filtrate is devoid of cells and has a very low concentration of plasma proteins in healthy individuals. Glomerular filtration is determined by Starling forces (hydrostatic and oncotic pressures) across the glomerular membrane and the ultrafiltration coefficient (Kf):

Where P_GC – P_BS and π_GC – π_BS are the differences in hydrostatic (P) and osmotic (π) pressures between the glomerulus (GC) and Bowman’s capsule (BS).

GFR is the sum of the glomerular filtration rate across all nephrons and is an important index of renal function.¹ A normal GFR ranges between 90 and 120 mL/min/1.73 m². GFR is commonly estimated in clinical practice from the renal clearance of creatinine. Creatinine, a degradation product of creatine phosphate metabolism that predominantly originates from skeletal muscles, is freely filtered across the glomeruli and is only minimally secreted and reabsorbed by the renal tubules. Hence, the clearance of creatinine is a reasonable estimate of the GFR. A reduction in GFR initially results in a drop in creatinine excretion. However, as plasma creatinine levels increase due to unchanged creatinine production, a new steady state is reached as production and excretion equilibrate at a higher plasma creatinine concentration. Hence, plasma creatinine levels correlate with GFR, but this relationship is nonlinear. Moreover, the relationship between GFR and creatinine is different across gender, age, and race. Therefore, the common clinical practice of defining impaired renal function by serum creatinine cutoffs may lead to significant errors. Laboratories should report results as estimated GFR.

Several formulas exist that estimate GFR from plasma creatinine levels, with the Cockcroft-Gault (GC), the Modification of Diet in Renal Disease (MDRD),² and the CKD-EPI³ equations. The CG equation was derived by measuring creatinine clearance in men, with an assumed correction factor of 0.85 in women⁴:

adjusted for body surface area (BSA) by 1.73 m²/BSA, where SCr = serum creatinine in mg/dL.

The CG equation tends to overestimate GFR, as creatinine is not only filtered by the glomerulus, but to some extent excreted by the tubules.⁵ In recent years the MDRD equation,^6,7 created by a stepwise regression analysis of measured GFR (mGFR) in 1628 chronic kidney disease, subjects of the MDRD study, has gained widespread use for estimating GFR from serum creatinine. Originally, a six-variable equation, the MDRD was further modified to a four-variable equation and then reexpressed for use with standardized creatinine assays⁸:

To account for differences across gender and race, the estimated GFR should be multiplied by 1.212 if the patient is black and/or by 0.742 for women.

The reexpressed MDRD equation, having been derived in patients with CKD, tends to underestimate kidney function at higher GFR. To address this limitation, a third equation [CKD-EPI(3)], which has now been adopted by most commercial labs and was endorsed by KDIGO (Kidney Disease Improving Global Outcomes; kdigo.org) in 2013, was developed using data from 12,150 subjects in diverse populations (8254 for development and 3896 for validation). The CKD-EPI equation, expressed as a single equation, is:

where SCr is serum creatinine (mg/dL), κ is 0.7 for females and 0.9 for males, α is –0.329 for females and –0.411 for males, min indicates the minimum of SCr/κ or 1, and max indicates the maximum of SCr/κ or 1.

Online calculators are available to calculate these formulas (http://www.niddk.nih.gov/health-information/health-communication-programs/nkdep/lab-evaluation/gfr-calculators/adults-conventional-unit-ckd-epi/Pages/default.aspx and http://www.niddk.nih.gov/health-information/health-communication-programs/nkdep/lab-evaluation/gfr-calculators/adults-conventional-unit/Pages/adults-conventional-unit.aspx).

Tubular Excretion and Reabsorption

Selective reabsorption of water and solutes and excretion of solutes occur along the tubular portion of the nephron. Transport of solutes across the tubular membrane is an energy-demanding process and may be impacted in various metabolic derangements.

Renal Blood Flow

Together, the kidneys constitute less than 0.5% of total body weight, but receive approximately 25% of the cardiac output during resting conditions (1.25 L/min). Renal blood flow (RBF) is an important determinant of GFR, solute and water reabsorption by the proximal tubule, concentration and dilution of urine, and excretion of substances in the urine. Another important function is to deliver oxygen, nutrients, and endocrine substances to the cells of the nephron and to return carbon dioxide and reabsorbed solutes to the circulation.

Like other organs, the kidneys regulate their blood flow by adjusting vascular resistance in response to arterial pressure. Two autoregulatory mechanisms that regulate the tone of the afferent arteriole contribute to RBF and GFR (of which RBF is a major determinant), remaining particularly constant over a wide range of systemic arterial pressures.⁹ The first is the myogenic mechanism, which is related to the intrinsic property of vascular smooth muscle cells to contract when it is stretched. The second mechanism is mediated by the juxtaglomerular apparatus at the macula densa and is termed tubuloglomerular feedback. When GFR increases and causes the NaCl concentration of the tubular fluid at the macula densa to increase, the afferent arteriole constricts. Nitric oxide produced by the macula densa attenuates, whereas angiotensin II enhances, tubuloglomerular feedback. In addition to these autoregulatory mechanisms, sympathetic nerves and angiotensin II play major roles in regulating RBF and GFR. Prostaglandins, nitric oxide, endothelin, bradykinin, adenosine triphosphate (ATP), and adenosine also affect RBF and GFR.

Regulation of Effective Circulatory Volume

Total body water accounts for 60% of total body weight and is divided into intracellular (40%) and extracellular (20%) fluid. The extracellular fluid volume (ECF) is largely influenced by Na⁺ content and is further divided into the interstitial fluid and blood plasma. The regulation of ECF volume involves a complex system of sensors and effector signals that act primarily on the kidneys to alter NaCl excretion.^10,11 The effective circulatory volume (ECV) refers to the portion of the blood plasma that effectively perfuses the tissues that contain the volume sensors that participate in ECF volume regulation. The kidneys adjust NaCl excretion in response to changes in the ECV. In a healthy person, any change in ECF volume results in a parallel change in plasma volume, ECV, blood pressure, and cardiac output. However, this relationship is disrupted in some pathological conditions, including heart failure.

Potassium and Acid–Base Balance

Excretion of K⁺ by the kidneys is determined predominantly by the distal tubule and collecting duct and is regulated by plasma [K⁺], aldosterone, and vasopressin. Renal K⁺ handling is linked to acid–base balance. Alkalosis increases whereas acidosis reduces K⁺ excretion.

Aging and Renal Function

Renal function gradually decreases with increasing age.¹² The mean estimated GFR among individuals over 70 years has been shown to be slightly below the threshold for CKD (< 60 mL/min/1.73 m²).¹³ It is difficult to separate the effects of “normal” aging from the effects of a cumulative disease burden (eg, diabetes, hypertension, heart disease) because many of the morphological features of CKD are observed in the senescent kidney.^14,15 Nevertheless, the senescent kidney is more susceptible to injurious stimuli (Table 105–1), and an age-related decline in renal function needs to be taken into consideration when administering drugs that are cleared through the kidney.^16,17

Biomarkers to Assess Renal Function

With the emergence of genomics, proteomics, and metabolomics in recent years, an increasing number of biomarkers of tissue injury and/or organ dysfunction have been detected. A myriad of potential biomarkers of renal disease have been studied and a few candidates are being tested in preclinical and clinical models. Table 105–2 lists some of these biomarkers. Unfortunately, the diagnostic and prognostic performances of these biomarkers have been heterogeneous across studies. For now, the traditional biomarkers (eg, serum creatinine and urine albumin) remain the most practical for diagnosis and risk stratification of patients with renal disease. Among the novel biomarkers, cystatin C and neutrophil gelatinase-associated lipocalin (NGAL) have emerged as the most promising and may be useful in clinical decision making.^18,19

Definition and Classification of Acute and Chronic Kidney Injury

Impaired renal function has traditionally been classified into two distinct syndromes, acute and chronic renal failure. However, these syndromes are closely connected and are not two inherently distinct entities. Acute kidney injury (AKI) is a risk factor for development of chronic renal disease, chronic renal disease is a risk factor for AKI, and both acute and chronic renal disease are risk factors for cardiovascular disease.²⁰

Definition of Acute Kidney Injury

AKI can be classified according to its etiology. The causes of AKI can be categorized as prerenal (upstream of the glomerulus), intrarenal (intrinsic renal disease), and postrenal (urinary tract obstruction) causes. Functional definitions, which classify patients into stages of AKI based on changes in serum creatinine concentration and urinary output, complement the anatomical approach to diagnosis (Table 105–3).²¹ These functional classifications have been shown to be strongly associated with outcomes, including progression of renal disease, ischemic coronary events, and mortality.^22,23 The most important of these classifications is the RIFLE (risk, injury, and heart failure, loss, and end-stage renal disease) criteria, a consensus-based classification formalized by the Acute Dialysis Quality Initiative.²⁴

Definition of Chronic Renal Disease

CKD is defined as the presence of persistent (> 3 months) functional or structural kidney abnormalities. A urine albumin-to-creatinine (ACR) ratio > 30 μg/mg from the first voided urine specimen and acquired under resting conditions is considered abnormal and an indicator of renal damage. An abnormal ACR is classified as either microalbuminuria (30–300 μg/mg) or macroalbuminuria (> 300 μg/mg). CKD is typically classified according to the GFR, supplanted by markers of kidney damage (eg, urinary ACR) (Table 105–4). CKD stages correlate with increased prevalence of hypertension and diabetes and an increased risk of end-stage renal failure, ischemic coronary events, and death.^25,26

Patients who are hospitalized due to heart disease are at increased risk of AKI regardless of baseline kidney function.^27,28 AKI is associated with adverse events, including an increased mortality.²⁹

Acute Kidney Injury as a Complication of Myocardial Infarction, Cardiogenic Shock, and Acute Decompensated Heart Failure

AKI may occur when renal hemodynamics become deranged secondary to acutely impaired cardiac function. Whereas renal arterial hypotension due to decreased cardiac output has traditionally been believed to be the principal cause of AKI, recent data imply that venous congestion is more important.^30,31,32,33 Increased central venous pressure and intra-abdominal pressure is transmitted to the renal veins, interstitium, and tubules. This causes a reduction in renal perfusion pressure and GFR. A reduced GFR and a lower effective circulating volume activate the renin-angiotensin-aldosterone system (RAAS), with resultant sodium retention, worsened congestion, and parenchymal ischemia. Inflammatory cytokines, which are consistently observed in decompensated acute heart failure, may also contribute to tubular damage.³⁰

AKI is common in patients with acutely and advanced decompensated heart failure, including patients with ST-segment elevation myocardial infarction (STEMI) who develop signs and symptoms of heart failure.²⁹ It occurs in 50% of patients who develop cardiogenic shock.^34,35 Preexisting CKD and/or heart failure, diabetes, hypertension, obesity, anemia, and old age all increase the risk of developing AKI.

Management of such patients is challenging and several currently ongoing trials are trying to identify more effective treatment alternatives. Reversible causes of cardiac decompensation need to be swiftly addressed. The therapeutic window for optimizing renal status is narrow because drugs that reduce congestion, including loop diuretics, typically cause arterial hypotension and worsened renal perfusion, with subsequent stimulation of the RAAS (Fig. 105–1).³⁶ Judicious administration of loop diuretics to relieve congestion is the cornerstone of acute heart failure complicated by AKI. Ultrafiltration achieves more effective fluid removal and decongestion than medical therapy, but does not influence prognosis.^37,38,39 Inotropes (eg, dobutamine, milrinone, and levosimendan) can be used to augment cardiac function, but may induce ventricular arrhythmias and have not been shown to improve prognosis of AKI.^40,41 In the worst cases where renal function continuous to deteriorate, continuous renal replacement therapy (CRRT) may be necessary.

Acute Kidney Injury in Coronary and Peripheral Procedures

Major technological advancements over the past decades allow for sophisticated percutaneous coronary (or peripheral) artery interventions. These interventions improve the quality of life for patients with coronary artery disease (CAD) and improve the prognosis for patients with acute coronary syndromes (ACS). However, they require imaging, often repeatedly, of the vasculature. This is achieved by intravascular injection of iodinated contrast material, which is potentially nephrotoxic. Despite a decreasing incidence over the past decade, contrast-induced acute kidney injury (CI-AKI) occurs in as many as 10% of all interventional procedures and as many as 20% of procedures in patients with acute myocardial infarction.^42,43 Patients who develop AKI during coronary and/or peripheral intervention have increased risk of serious complications, including death.

Rarely, periprocedural AKI may also be caused by embolization of cholesterol to the renal arteries due to catheter manipulation of atheromatous plaques in the aorta. AKI caused by atheroembolization has a poor prognosis, with a high risk of the patient requiring chronic dialysis.^44,45,46 Because of the poor prognosis related to atheroembolic AKI, it is important to differentiate between this and CI-AKI. Delayed (subacute) but sudden onset of renal insufficiency is common with atheroembolic AKI. Eosinophilia and eosinophiluria are common and concomitant skin findings (livedo reticularis, purpura, ulceration, or gangrene) should raise the level of suspicion. Renal biopsy may be necessary for establishing the diagnosis.⁴⁶

Contrast-Induced Acute Kidney Injury: Pathophysiology

Currently used iodinated contrast agents are highly water soluble iso-osmolar (~ 290 mOsm/kg) or low-osmolar (700-850 mOsm/kg) solutions. Higher osmolality increases the risk of vascular symptoms of warmth and pain as well as of AKI.⁴⁷ Injection of iodinated contrast causes transient vasodilation followed by vasoconstriction. In the renal vessels, this vasoconstriction can persist for hours, which may result in hypoperfusion of the renal parenchyma, including the renal medulla.⁴⁸

Contrast media, excreted in the glomerulus, are readily absorbed by the renal tubuli. The media therefore accumulate in the interstitial space and may cause direct cellular toxicity. Contrast material can reach very high concentrations in the renal medulla, where the osmolality of the interstitium is high. Contrast-mediated tubular cell death leads to sloughing of cellular constituents into the tubular space, which further impedes urine flow and contrast clearance. Hence, the iodinated contrast agents cause renal injury as a consequence of both hypoperfusion and direct cytotoxicity.

Glomerular function is intact in CI-AKI, but GFR decreases after 24 to 48 hours secondary to tubuloglomerular feedback triggered by the injury to the proximal tubule.⁴⁹ Hence, serum creatinine levels initially remain normal in CI-AKI, and an increase in serum creatinine may only be detected after 48 to 72 hours from exposure to iodinated contrast. Because the glomerulus remains uninjured with an intact filtration barrier to blood cells, there is typically no hematuria in CI-AKI.

Contrast-Induced Acute Kidney Injury: Prevention and Treatment

The most important measures to reduce the risk of CI-AKI are to use as little contrast material as possible and to identify patients at increased risk of AKI. Whereas low-osmolar agents can be used for most patients, only iso-osmolar agents should be administered to high-risk patients. Clearance of contrast can be facilitated by intravascular volume expansion by isotonic crystalloid solutions, which increase GFR and tubular flow. More aggressive volume expansion, driven either by the left ventricular end-diastolic pressure or by urine output, reduces the risk of CI-AKI.^50,51 Volume expansion may be even more effective at reducing AKI if combined with forced diuresis.³⁹

No established pharmacological treatment alternatives exist for complete prevention or treatment of CI-AKI. A number of promising substances (eg, N-acetylcysteine, ascorbic acid, and fenoldopam) have failed to demonstrate superiority to placebo when tested in randomized trials.^52,53 However, statins appear to reduce the risk of CI-AKI, whereas RAAS inhibitors (angiotensin-converting enzyme inhibitors [ACEIs] and angiotensin receptor blockers [ARBs]) may increase the risk.^54,55,56 Statins may mediate their protective effect through several mechanisms (including inhibition of contrast reabsorption, attenuation of inflammation, oxidative stress, and endothelial dysfunction, and protection of podocytes in the glomerulus), whereas ACEIs/ARBs are thought to interfere with protective autoregulatory mechanisms, including tubuloglomerular feedback.

Patients with end-stage renal disease (ESRD) should receive immediate hemodialysis after receiving contrast media if they develop pulmonary congestion or hyperkalemia, but the value of routine immediate hemodialysis for such patients is uncertain.⁵⁷

Risk Stratification of Patients Undergoing Coronary and/or Peripheral Procedures

Diabetes mellitus and preexisting CKD are important risk factors for CI-AKI. The risk increases gradually with increasing CKD severity.⁵⁸ STEMI and cardiogenic shock are also important risk factors, due at least in part to renal hypoperfusion. The need for acute intervention in STEMI precludes adequate volume expansion prior to the procedure, as does the concomitant acute left ventricular dysfunction. Other independent predictors of CI-AKI are increasing age, anemia, and congestive heart failure.^58,59 Several risk scores have been developed to predict the risk of CI-AKI, but none has been widely adopted in clinical practice. Many of these risk scores incorporate variables that are not known until during or after the procedure. They are, therefore, less helpful for preprocedural counseling.^58,59 As a general rule, the presence of diabetes and/or preexisting CKD warrants the preventive measures discussed above and special consideration of alternatives and conservative management without the invasive procedure.

Acute Kidney Injury in Patients Undergoing Transcatheter Aortic Valve Replacement

AKI is common (~ 20%) after transcatheter aortic valve replacement (TAVR) and is associated with increased mortality.^60,61,62 Similar to patients who undergo coronary angiography, patients who have TAVR receive periprocedural contrast. Furthermore, transient or sustained hemodynamic derangements are common during and/or after the TAVR procedure. Hence, all the mechanisms leading to AKI that were discussed in the previous sections may play a role in TAVR-related AKI.

The most important risk factors for AKI post-TAVR include preexisting CKD, hemodynamic instability and/or the need for circulatory support, blood transfusions, transapical access, amount of contrast used, diabetes, and age.^61,63

In order to prevent AKI among patients undergoing TAVR, it is important to optimize hydration before the procedure, use as little contrast as possible, withhold nephrotoxic drugs, restrict the use of unnecessary blood transfusions, and allow sufficient time between any preprocedural contrast-requiring imaging and the TAVR procedure.⁶⁴

Acute Kidney Injury in Patients Undergoing Cardiac Surgery

AKI is a common complication after cardiac surgery and is linked to reduced survival.^65,66,67 Impaired renal function (RIFLE risk or AKIN stage 1; see Table 105–3) is observed in 17% to 49% of patients who have cardiac surgery, and AKI (defined as RIFLE injury or AKIN stage 2) occurs in almost 10%.^{65,66,68,69,70} Approximately 2% will require dialysis due to new-onset renal failure.⁷¹ Among patients who require extracorporeal membraneous oxygenation (ECMO), the incidence of AKI may be as high as 80%.⁷²

The pathogenesis of AKI in the setting of cardiac surgery is incompletely understood. It is likely multifactorial, with the following factors playing a role:

• Nephrotoxins, both endogenous (eg, heme and iron pigments released secondary to blood trauma and hemolysis caused by blood pumps) and exogenous (eg, nonsteroidal anti-inflammatory drugs [impair autoregulation of renal blood flow] and intravenous contrast).^73,74,75

• Tissue hypoxia and ischemia-reperfusion injury. Intraoperative alterations in systemic or local renal hemodynamics and/or release of atherosclerotic emboli (eg, during aortic cross-clamping) may lead to tissue ischemia.^76,77 The renal medulla, which has a high metabolic demand and a low baseline PaO₂, is particularly susceptible to ischemia.⁷⁸

• The systemic inflammatory response syndrome (SIRS) is triggered by major surgery and artificial conduits, with activation of neutrophils, platelets, cytokines, coagulation factors, and fibrinolysis.⁷⁹

Risk factors for AKI in the setting of cardiac surgery are similar to those described earlier for AKI in other settings, and include CKD, reduced left ventricular ejection fraction, use of intravenous contrast agents, diabetes, and older age.^80,81,82 Chronic obstructive pulmonary disease, female gender, and emergent surgery have also been reported as risk factors for AKI in this setting.^80,81

No pharmacological treatment strategies that effectively prevent or treat AKI in the setting of cardiac surgery have been identified. Therefore, the clinician must try to prevent AKI by optimizing each patient’s pre-, intra-, and postoperative risk profile. Preoperative measures include optimization of glucose control for patients with diabetes, withholding of ACEIs and/or ARBs, and withholding surgery for up to 48 hours after contrast administration. Intra- and postoperative measures include minimizing cardiopulmonary bypass and cross-clamp times, favoring vasopressin over α-adrenoceptor agonists for blood pressure control, avoiding overly strict intra- and postoperative glucose control, basing decisions regarding blood transfusions on physiologic data (including lactate and central venous O₂ saturation) with restrictive transfusion thresholds (7.0-8.0 mg/dL), and hemodynamic goal-directed therapy (focus on cardiac output and central venous O₂ saturation, etc., rather than traditional hemodynamic indices such as arterial blood pressure) with judicious use of diuretics.^{83,84,85,86,87,88,89,90,91,92,93,94,95}

Drug Pharmacokinetics in Acute Kidney Injury

AKI influences the pharmacokinetics of water-soluble drugs that are eliminated by the kidneys. The plasma concentration of these drugs may increase in the setting of AKI. It is important to reduce or temporarily stop some drugs in patients with AKI. As an example, metformin-induced lactic acidosis is a rare, but potentially life-threatening, complication of AKI.⁹⁶ Metformin is therefore typically withheld for at least 48 hours following iodinated contrast administration, particularly in patients with an elevated serum creatinine.

Epidemiology and CHRONIC KIDNEY DISEASE–Specific Disease Mechanisms

CKD dramatically increases the risk of developing cardiovascular disease (CVD; 3-30 times depending on CKD stage and study).^97,98,99 Nearly half of all deaths in CKD patients are from cardiovascular events, and CKD patients are more likely to die from CVD than progress to ESRD during their lifetime.^100,101

Risk factors for the development of CVD in patients with CKD can be divided into traditional and nontraditional factors. Traditional factors include hyperlipidemia, hypertension, diabetes, and smoking and are important risk factors for CVD also among patients without CKD. These risk factors are also risk factors for CKD, and their prevalence has been reported to be twice as high among CKD patients compared to non-CKD patients.¹⁰² Nontraditional risk factors refer to the effects of chronic renal dysfunction on cardiac health. These factors are consequences of CKD and include volume overload, increased sympathetic tone, activation of the RAAS, oxidative stress, uremic toxins, abnormal mineral metabolism (calcium, phosphorous, vitamin D), anemia, and malnutrition. The importance of the nontraditional risk factors is highlighted by the poor performance of the Framingham Risk Criteria in CKD patients.¹⁰³ Three important mechanisms contribute to CVD progression in CKD: autonomic dysfunction, vascular pathology, and cardiac pathology.

Autonomic Dysfunction

Autonomic function is impaired in CKD patients, with a relative dominance of sympathetic over parasympathetic activity.^{104,105,106,107} The autonomic nervous system constitutes the efferent arm of the baro- and chemoreceptor reflex arcs. These reflexes are impaired in CKD. The detailed mechanisms through which CKD leads to autonomic dysfunction are unknown, but autonomic dysfunction in CKD is improved after transplantation, nephrectomy, or renal denervation. Hence, local pathological processes within the diseased kidney likely play a role.^108,109

Signs of autonomic dysfunction in CKD patients include reduced heart rate variability (a marker of vagal autonomic sensitivity and a poor prognostic sign) and greater low-frequency oscillations of systolic blood pressure variability (a marker of sympathetic tone).^110,111,112

Increased sympathetic tone promotes smooth muscle cell and fibroblast proliferation in blood vessels and activates the RAAS pathway.^113,114 Hence, autonomic dysfunction contributes to cardiovascular pathological remodeling. Through these mechanisms, autonomic dysfunction also causes worsening of renal function.¹⁰⁴

Vascular Pathology

CKD exacerbates atherosclerotic processes.¹¹⁵ CKD patients have more severe arterial hypertrophy and calcification and have stiffer arteries compared to non-CKD patients.¹¹⁶

The mechanisms leading to arterial pathology and endothelial dysfunction in CKD are incompletely understood, but increased oxidative stress, low-grade inflammation, uremic toxins, increased wall stress (associated with arterial hypertension), and impaired calcium/phosphorous homeostasis likely contribute.^117,118,119 Sympathetic hyperactivity promotes smooth muscle proliferation via norepinephrine, and indirectly via salt and fluid retention and arterial hypertension.

Cardiac Pathology

Left ventricular hypertrophy is present in 40% of CKD patients (mostly of the eccentric type), and the prevalence increases with declining GFR; for ESRD patients on hemodialysis, it is over 75%.¹²⁰ Left ventricular hypertrophy is likely promoted by chronic hypertension (increased afterload), anemia (reduced oxygen delivery), sympathetic and RAAS hyperactivity (increased cardiac fibrosis), and volume overload. Cardiomyocyte hypertrophy is not accompanied by a similar degree of neovascularization. Hence, the cardiomyocyte-to-capillary ratio increases, leading to relative ischemia of the myocardium. This is believed to explain why some CKD patients present with angina even in the absence of significant CAD.¹²¹

Almost 50% of CKD patients have either mitral or aortic calcification, and CKD is a strong predictor for valvular dysfunction (stenosis and/or regurgitation).¹²² Valvular disorders increase left ventricular afterload (aortic stenosis) and/or preload (aortic regurgitation, mitral regurgitation).

Heart failure is common in CKD, with a prevalence > 50% among ESRD patients. Heart failure with preserved ejection fraction accounts for approximately half of these cases and carries a worse prognosis than heart failure with reduced ejection fraction.¹²³ The etiologies of heart failure in CKD patients include pathological left ventricular remodeling secondary to sympathetic and RAAS overstimulation, left ventricular hypertrophy, valvular heart disease, and ischemic heart disease. Chronic anemia and uremic toxins may also contribute.

Management of Cardiovascular Risk Factors in the Patient with Chronic Kidney Disease

The majority of the pivotal large-scale multicenter trials on which most cardiovascular guidelines are based excluded patients with CKD. Therefore, the management of cardiovascular disease in the CKD patient requires judicious extrapolation of treatment effects in non-CKD patients, supplanted by knowledge derived from observational studies and with consideration of CKD pathophysiology.

Hypertension

Hypertension is prevalent among CKD patients. CKD-associated (nontraditional) risk factors include sodium retention and hypervolemia, increased RAAS activity, autonomic imbalance, and endothelial damage. Furthermore, traditional risk factors for hypertension are prevalent among patients with CKD and hypertension is itself an important risk factor for CKD.

Current guidelines¹²⁴ recommend a blood pressure goal of 140/90 mm Hg for CKD patients, which is the same as for non-CKD patients. Restriction of sodium intake reduces systolic blood pressure, extracellular fluid volume, and proteinuria compared to patients on a regular diet.¹²⁵ The National Institutes of Health (NIH)-funded SPRINT study randomized 9361 patients with hypertension, of whom 28% had preexisting CKD, to a systolic blood pressure goal of < 140 mm Hg versus < 120 mm Hg. The trial was stopped prematurely due to a significant 25% reduction in the composite primary end point of myocardial infarction, other ACS, stroke, heart failure, or death from cardiovascular causes and a 27% reduction in total mortality.¹²⁶ Guidelines are expected to be updated to more aggressive goals.

ACEIs and ARBs are considered first-line medical therapy, especially in cases of concomitant proteinuria or diabetes. These drugs lower blood pressure and slow the decline in renal function in CKD patients.^{127,128,129,130} ACEI/ARB treatment may result in hyperkalemia. Serum creatinine levels therefore need to be monitored in these patients. An increase in serum creatinine > 25% should prompt further evaluation. Inadequate volume status, too low blood pressure, drugs that may reduce GFR (nonsteroidal anti-inflammatory drugs), or excessive potassium intake should be corrected. If these reversible causes are absent, the ACEI/ARB should be stopped, which should result in the return of kidney function to baseline.^131,132 The combination of an ACEI and ARB can further lower blood pressure, but carries a considerable risk of hyperkalemia.¹³³ Dual therapy is, therefore, not recommended.

Calcium channel blockers and/or β-blockers are considered second-line agents.^134,135 Diuretics should be considered as second-line agents to be used in cases of volume overload, with loop diuretics being the preferred drugs in later stages of CKD.

Hyperlipidemia

Lipid abnormalities are common in nearly all stages of CKD. The most common abnormality is hypertriglyceridemia, as nearly 50% of CKD patients have fasting triglyceride levels above 200 mg/dL,¹³⁶ the result of impaired removal of triglycerides from blood, secondary to changes in the composition of triglycerides (enrichment in apolipoprotein C3), as well as reductions in hepatic lipoprotein lipase activity. CKD patients also frequently have decreased apo-A1 and high-density lipoprotein (HDL) levels. The typical association with greater low-density lipoprotein (LDL) and greater risk of cardiovascular events is not as robust in this population. In dialysis patients, in particular, those with both the highest and lowest levels of LDL are at greatest risk for CVD, likely related to confounding by malnutrition in those with significantly impaired renal function.^137,138 Patients with nephrotic syndrome typically have markedly increased LDL-cholesterol levels related to increased hepatic protein production.

In a post-hoc analysis of the JUPITER (Justification for the Use of Statins in Prevention—An Intervention Trial Evaluating Rosuvastatin)¹³⁹ primary prevention trial of rosuvastatin 20 mg compared with placebo, among those with moderate CKD (eGFR < 60 mL/min/1.73 m² at study entry, n = 3267), rosuvastatin was associated with a 44% reduction in all-cause mortality (hazard ratio [HR], 0.56; 95% CI, 0.37–0.85; P = .005) and a 45% reduction in risk of myocardial infarction, stroke, hospital stay for unstable angina, arterial revascularization, or confirmed cardiovascular death (HR, 0.55; 95% CI, 0.38–0.82; P = .002). LDL-cholesterol and high sensitivity C-reactive protein reductions as well as adverse events with rosuvastatin were similar among those with and without CKD.

By contrast, the German Diabetes and Dialysis Study¹⁴⁰ enrolled 1255 subjects with type 2 diabetes mellitus receiving maintenance hemodialysis who were randomized to receive 20 mg of atorvastatin per day versus placebo. The primary end point was a composite of death from cardiac causes, nonfatal myocardial infarction, and stroke. After a median follow-up of 4 years there was no significant reduction in the primary composite end point. Cardiac events were reduced with an HR of 0.82, while cerebrovascular events were increased twofold. The more recent SHARP (Study of Heart and Renal Protection)¹⁴¹ trial randomized 9270 patients with CKD (3023 on dialysis and 6247 not) without prior myocardial infarction or revascularization to simvastatin 20 mg plus ezetimibe 10 mg daily versus matching placebo. The primary end point—a composite of first major atherosclerotic event, nonfatal myocardial infarction or coronary death, nonhemorrhagic stroke, or any arterial revascularization procedure—was reduced by 17% without an increase in nonvascular mortality. Notably, results were consistent in the dialysis group. Given that there was no ezetimibe-only group in SHARP, it is not clear whether the results were driven by ezetimibe, simvastatin, or both.

Treatment guidelines generally reflect those found in non-CKD patients, with a particular emphasis on statin therapy for patients with moderate renal dysfunction. Indeed, as CKD is a coronary heart disease equivalent in ACC/AHA hyperlipidemia guidelines,¹⁴² most CKD patients are candidates for statin therapy. KDIGO guidelines recommend statin or statin/ezetimibe in patients older than 49 years of age, and statin therapy if there are other known risk factors in patients aged 18 to 49 years.¹⁴³ The KDIGO guidelines recommend against the use of LDL-cholesterol levels as a target of therapy in patients with CKD, focusing instead on treatment of high-risk patients irrespective of LDL-cholesterol levels. For patients with hypertriglyceridemia alone, the most recent KDIGO guidelines only recommend lifestyle modification therapy. Fibrates have only weak evidence of reducing pancreatitis risk or cardiovascular risk in CKD patients with elevated triglycerides, and are thus not recommended.¹⁴³

Anemia of Chronic Kidney Disease

Anemia is common in CKD and is almost uniformly present at late stages of CKD.¹⁴⁴ It is associated with increased risk for heart failure, stroke, and death.^145,146,147 CKD-associated anemia is normocytic and normochromic. It is caused by decreased production of erythropoietin by the kidney as well as by reduced red blood cell survival in the uremic environment. It contributes to the symptoms observed in patients with CKD (fatigue, reduced exercise tolerance, dyspnea) and may exacerbate angina symptoms in patients with concomitant ischemic heart disease. Anemia has also been shown to be a causative factor in left ventricular hypertrophy.

Therapeutic options include erythropoiesis-stimulating agents and red blood transfusion, with iron supplementation as appropriate. Alternative causes of anemia should be identified and corrected if possible. These include iron deficiency anemia and anemia of chronic disease, which are relatively common among patients with CKD. The optimal time to start treating CKD-associated anemia is unknown, but a target hemoglobin concentration between 10 and 12 g/dL is associated with improvements in physical activity, less fatigue, and less left ventricular hypertrophy. However, both blood transfusions and erythropoiesis-stimulating agents carry the risk of serious adverse events, including thrombosis and stroke. Therefore, current FDA and KDIGO guidelines emphasize a personalized approach to care.¹⁴⁸ No strict recommendations are given in regards to hemoglobin targets, but because three major randomized trials (CHOIR, CREATE, and NHCT) showed an increased mortality at Hb levels above 13 g/dL,^149,150,151 KDIGO guidelines recommend an upper limit of 11.5 g/dL (although the guidelines acknowledge that some patients may benefit from higher levels).

Acute Coronary Syndrome in the Patient with Chronic Kidney Disease

CKD is an important risk factor for CAD, including ACS.¹⁵² The difficulty in diagnosing ACS in CKD patients tends to delay treatment, and prognosis after ACS is worse for CKD patients than for non-CKD patients.¹⁵³ As is the case in other areas of cardiology, the majority of the pivotal multicenter randomized trials on which guidelines for ACS treatment are based have excluded CKD patients.¹⁵⁴

Diagnosis

CKD patients with ACS are less likely to present with chest pain (at least partly due to autonomic dysfunction). CKD patients are also more likely to have preexisting ST-segment depressions and/or Q waves on resting electrocardiogram (ECG; caused by cardiac hypertrophy and/or previous ischemic injury), which considerably decrease the specificity of the ECG for diagnosing ongoing ACS. Furthermore, plasma biomarkers of cardiac damage, including creatine kinase (CK) and troponins, are often elevated at baseline due to reduced clearance, low-grade inflammation and silent myocardial injury due to small vessel disease. This reduces the specificity of these tests.^155,156 The ACS diagnosis is therefore more likely to be missed in CKD patients than non-CKD patients.¹⁵⁵ It is recommended to focus on temporal trends in troponin levels and ST-segment depressions and on angina-equivalent symptoms such as worsening dyspnea.¹⁵⁷

Management

In the absence of dedicated trials for patients with CKD, ACS treatment guidelines are similar to those for patients without CKD. However, current guidelines recommend a judicious use of antithrombotic agents in patients with CKD (who have an increased bleeding risk) with consideration of CKD as well as other risk factors before initiating particularly potent antithrombotic therapy (eg, triple therapy).¹⁵⁸ Importantly, drugs that are cleared from the body predominantly through the kidneys need to be dose-adjusted (Table 105–5).

There are a paucity of high-quality data regarding the relative benefits of medical therapy (a conservative strategy) and revascularization by either percutaneous coronary intervention (PCI) or coronary artery bypass grafting (CABG) for patients with CKD and ACS. Retrospective analyses imply that revascularization is favorable compared to a conservative approach.^159,160,161 CABG is associated with particularly high risks of stroke and death for patients with CKD, but appears to offer a better long-term prognosis compared to PCI for patients with multi-vessel disease.^162,163,164 Hence, guideline recommendations relating to reperfusion strategies are similar for patients with and without CKD, including the importance of individualized decisions based on risk scores (eg, surgical risk, ability to tolerate dual antiplatelet therapy) as well as patient preference¹⁶⁵ and use of a heart team approach.

Cardiovascular Drug Pharmacokinetics in the Patient with Chronic Kidney Disease

Many cardiovascular drugs are eliminated from the body by the kidneys. Dose adjustments of some of these drugs are necessary in CKD. Unfortunately, failure to dose-adjust is a common cause of drug toxicity in CKD patients.

Renal clearance of drugs depends on GFR as well as tubular function (reabsorption and/or secretion) and the relative contributions of these mechanisms vary according to drug. Through alterations in body fluid composition and acid–base balance, CKD also affects drug absorption, distribution volume, protein binding, and affinity for drug target molecules.¹⁶⁶ Drug clearance in dialysis patients depend on water solubility, distribution volume, protein binding, and diffusion across the dialysis membrane. It can, therefore, be difficult to accurately predict plasma concentrations and therapeutic effect in patients with CKD. Target doses of cardiovascular drugs defined in current guidelines may be difficult to achieve. A general principle is to “start low and go slow” and consider the highest tolerable dose as the optimal dose.

Table 105–5 lists common cardiovascular drugs that need particular attention for patients with CKD. It is recommended to consult specific drug manufacturers’ dose adjustment recommendations when administering drugs to patients with CKD.

Heart Disease in Patients on Dialysis

More than 80% of dialysis patients have concomitant CVD, and CVD is the leading cause of death for dialysis patients (including heart failure, myocardial infarction, and sudden cardiac death [SCD]).^{167,168,169,170} Three major mechanisms contribute to worsening CVD in patients with ESRD: volume overload, pressure overload, and nonhemodynamic factors such as atherosclerosis, vascular calcification, oxidative stress, inflammation, and stimulation of profibrotic factors.^169,171 Managing a patient’s cardiovascular risk profile is, therefore, important in dialysis patients.

The New York Heart Association (NYHA) functional classification for heart failure assessment is less useful in dialysis patients because dyspnea can be related to factors other than heart failure. More importantly, the severity of heart failure–related symptoms changes considerably in relation to the timing of fluid removal. Alternative classification systems have been proposed that take into account the renal component (Fig. 105–2).¹⁷⁰

Important Cardiac Complications in the Patient with Chronic Kidney Disease

Pericarditis

Patients with CKD can develop pericarditis secondary to uremia or as a complication of dialysis.¹⁷² Rarely, these patients develop chronic constrictive pericarditis. The incidence of uremic pericarditis has declined considerably in the dialysis era, but it still occurs in 5% to 20% of uremic patients. Its exact cause is unknown, but appears related to the accumulation of uremic toxins.¹⁷² Uremic pericarditis is less responsive to anti-inflammatory therapy and should be treated by intensive hemodialysis (daily for at least 1 week). Dialysis-related pericarditis typically occurs after several months of dialysis therapy. Whether the etiology is different is debated, but dialysis-related pericarditis is less responsive to intensified dialysis.¹⁷³ Pericardiostomy and pericardiectomy may be necessary, with the latter the primary treatment for constrictive pericarditis.¹⁷² Pericardial effusions are common (volume overload is likely a contributing factor), but tamponade is rare.

Importantly, the differential workup must consider that patients with CKD can also develop viral, fungal, or bacterial pericarditis (particularly if on dialysis), drug-induced pericarditis, and pericarditis due to other underlying disease states.

Infective Endocarditis

Patients undergoing hemodialysis have an age-adjusted incidence of infective endocarditis (IE) almost 20 times higher than the general population.¹⁷⁴ Patients who are dialyzed through tunneled catheters are at particularly high risk.¹⁷⁵ The most common pathogen is Staphylococcus aureus (~ 50%) followed by coagulase-negative staphylococci, enterococci, and Streptococcus viridans.¹⁷⁵ The mitral valve is the most commonly involved valve, but ~ 20% of patients have involvement of more than one valve. Dialysis patients with IE have considerably higher mortality (30%-50%) than nondialysis patients with IE.¹⁷⁵ Mitral valve involvement and septic embolization are poor prognostic factors.¹⁷⁶

IE should be suspected in all hemodialysis patients with fever and/or bacteremia. Classical symptoms and signs include a new regurgitrant murmur, focal neurological deficits, or heart failure symptoms. Transesophageal echocardiography (sensitivity > 90%) is considerably superior to transthoracic echocardiography (sensitivity ~ 50%) at establishing the diagnosis.^177,178,179

As a general rule, current pathogen-specific treatment guidelines for IE apply to both dialysis and nondialysis patients with a required minimum of 4 to 6 weeks of parenteral therapy.^180,181,182 Antimicrobial therapy is typically delivered after the dialysis session due to the difficulty in maintaining long-term peripheral venous access in dialysis patients. Coordination between nephrologists and infectious disease staff to select optimal antibiotic agent and dosing schedule is recommended. Indications for surgery for dialysis patients with IE are similar to those for nondialysis patients, but operative mortality is considerably higher for dialysis patients.^{175,180,181,182} Whether dialysis patients who undergo valve replacement should receive mechanical (increased risk of valve thrombosis, requires anticoagulation) or bioprosthetic (accelerated degeneration and calcification in the dialysis patient) valves has not been sufficiently studied.¹⁸³ A case-to-case decision based on individual risk factors, suitability for long-term anticoagulant therapy, and life expectancy is recommended.

Cardiac Arrhythmias

SCD is estimated to account for almost 25% of all deaths among dialysis patients.¹⁰¹ Causes include ventricular tachycardia, Torsades de pointes, and ventricular fibrillation, as well as pronounced bradycardia. Ischemic heart disease, heart failure, vascular calcification, electrolyte and acid–base disturbances, and dialysis-induced electrolyte shifts are believed to contribute to the risk of ventricular arrhythmias in dialysis patients.^184,185

Predictors for SCD include atrioventricular or intraventricular conduction delays, left ventricular hypertrophy, QRS or QT-interval prolongation, decreased heart rate variability, documented episodes of nonsustained ventricular tachycardia, exercise-induced arrhythmia, multi-vessel CAD, and hypo- or hyperkalemia.¹⁸⁶

Primary and secondary prevention with β-blockers and ACEIs/ARBs is recommended for dialysis patients. Implantable cardioverter-defibrillators (ICDs) trials have systematically excluded patients on dialysis, and the majority of dialysis patients who experienced SCD did not fulfill current criteria for ICD implantation.¹⁸⁷ For those dialysis patients with ICDs implanted for ventricular fibrillation/SCD, however, observational data suggest that the benefit in terms of risk reduction is similar to patients without CKD, with a 42% risk reduction,¹⁸⁸ suggesting that this potentially lifesaving therapy is underutilized. However, long-term prognosis is still poor for dialysis patients irrespective of ICD implantation.^189,190

Cardiovascular Risk Stratification Prior to Renal Transplantation

CVD remains the most common cause of death for patients with CKD, even if they receive a kidney transplant.¹⁹¹ Cardiovascular events are common within the first 3 months following transplantation, but are thereafter less common compared to patients who remain on dialysis.¹⁹² Risk factors for posttransplant ischemic cardiac events include presence of CAD, heart failure, diabetes, male gender, age, an abnormal resting ECG, and smoking. Observational data suggest that asymptomatic patients with obstructive CAD benefit from revascularization before transplantation, but randomized trials are lacking.

Screening for CAD is considered warranted before transplantation, but the optimal risk stratification algorithm remains to be established. Whether noninvasive tests (pharmacologic stress echocardiography or nuclear imaging) should be preferred for initial screening, and coronary angiography reserved for patients who have objective evidence of ischemia or who are symptomatic, is debatable. Apart from screening for CAD, the pretransplant cardiac evaluation should seek to identify cardiovascular risk factors amenable to modification (eg, diabetes, hypertension, heart failure) and to identify patients who are unlikely to benefit from transplantation due to poor prognosis related to an extra-renal condition.¹⁹³

Worsening Renal Function in Patients with Chronic Heart Failure

Worsening renal function in chronic heart failure is associated with adverse outcomes and prolonged hospitalization. The prevalence of CKD among patients with chronic heart failure is at least 25%, and even relatively small reductions in GFR (> 9 mL/min) increase the risk of dying.

Cardiorenal Syndrome Type 2

The term “cardiorenal syndrome” (CRS) has been coined to describe the entity in which concomitant cardiac and renal dysfunction is present in the same patient. CRS is subdivided according to the temporal pattern of cardiac and renal disease (Table 105–6). Herein we focus on the most common CRS type 2 (CRS-2), in which chronic heart failure leads to CKD.

The pathophysiology behind the development of renal dysfunction in CRS-2 is incompletely understood. There does not appear to be a strong correlation between left ventricular ejection fraction and estimated GFR among patients with chronic heart failure.¹⁹⁴ Proposed mechanisms underlying the development of kidney dysfunction in chronic heart failure are outlined in Fig. 105–3.

Regardless of its exact mechanisms, declining renal function in the chronic heart failure patient has important therapeutic implications. Patients with chronic heart failure typically take a considerable number of medications. Many of these are cleared by the kidneys and need to be dose-adjusted when renal function declines.^195,196 Although most heart failure therapies are beneficial for patients with CKD, careful monitoring of renal function is necessary and changes in drug regimen may be appropriate.

An important principle in the treatment of CRS-2 is to optimize cardiac function in order to slow or reverse the pathophysiology outlined in Fig. 105–3. Cardiac resynchronization therapy, when indicated, appears to improve both ejection fraction and GFR in patients with CRS-2.¹⁹⁷ β-Blockers also appear to be beneficial.¹⁹⁸ Antagonism of the RAAS is an important component in both heart failure and CKD therapy. Both ACEIs/ARBs and aldosterone antagonists attenuate excessive RAAS activation and thereby prevent renal vasoconstriction and fibrosis. However, although ACEI/ARBs are beneficial in CRS-2, there is a risk of hyperkalemia as well as of inducing arterial hypotension resulting in worsening renal function. The risk of these complications is higher for patients with CKD than in patients without renal disease.^199,200 Hence, caution, frequent monitoring of electrolyte status, and involvement of a nephrologist is recommended.¹⁹⁹

Management of volume status is particularly challenging in patients with CRS-2. Restricting sodium intake is important. Diuretic therapy is recommended, but is complicated by the development of diuretic resistance (DR), which carries a poor prognosis.²⁰¹

Diuretic Resistance

The definition of DR is not straightforward and no consensus exists. One approach is to define diuretic efficacy as the change in weight, net fluid balance, or urine output per a defined diuretic dose (eg, 40 mg of intravenous furosemide-equivalent dose). Patients with DR typically have lower GFR than patients without DR, and DR is a predictor of worse outcomes independent of GFR.²⁰² Although intimately intertwined, DR and GFR probably represent different aspects of renal function. Whereas GFR is reflective of the clearance of metabolites (renal ultrafiltration), DR may be a better index of volume and electrolyte homeostasis (tubular function). Even at a severely depressed GFR of 20 mL/min, almost 30 L of fluid (~ 4000 mEq of sodium) is filtered through the glomeruli each day. Volume overload and organ congestion are the main drivers of readmissions and adverse outcomes in CHF.²⁰³ This provides a pathophysiological rationale for the strong independent association between DR and poor outcome.

Pharmacokinetic and Pharmacodynamic Considerations

Loop diuretics inhibit the sodium-potassium-chloride cotransporter in the luminal surface of tubular cells in the thick ascending loop of Henle and are typically first-line diuretics used in chronic heart failure.²⁰⁴ Several characteristics of the loop diuretics, which are not related to tubular damage and/or dysfunction, may explain diuretic resistance. Orally administered loop diuretic absorption may be reduced in congestive heart failure due to gut edema and poor intestinal perfusion. In the plasma, > 90% of the loop diuretics will be bound to plasma protein and are therefore not filtered across the glomerulus. In order to reach the tubular lumen they must be secreted by anion transporters in the proximal tubules. Organic anions (eg, urate) compete with the loop diuretics and their transport is inhibited by acidosis. Significant proteinuria increases protein-binding of loop diuretics in the tubular lumen and prevents their pharmacological action.^{201,205,206,207,208} All these mechanisms lead to a shift to the right in the dose–response curve for loop diuretics. Lastly, sufficient Na⁺ must be present in the tubule for loop diuretics to work. If proximal tubular Na⁺ reabsorption is sufficiently increased (for example, due to poor renal perfusion and increased neurohormonal activity), too little NaCl may be delivered to the distal nephron for loop diuretics to work efficiently. Hence a multitude of factors can cause diuretic resistance.²⁰⁴

Managing Diuretic Resistance

After verifying compliance with therapy, the next step in the management of patients with apparent DR is to establish whether intravascular volume overload is present. When intravascular volume overload is not present, volume reduction by diuretics will not unload the heart, but may instead worsen neurohormonal activation. Clinical signs and symptoms (eg, orthopnea, jugular venous distension, rales, hepatomegaly, and edema) are helpful, but lack sufficient sensitivity. Body weight (unless an accurate estimate of “dry weight”) and natriuretic peptides (which track better with wall stress than volume overload per se) are also helpful, but imperfect, means of assessing volume status. Hence, establishing whether a lack of response to diuretics is due to diuretic resistance or relative intravascular depletion remains to some extent an art of medicine.

In the absence of high-quality data from randomized clinical trials, treatment approaches in diuretic resistance are based on knowledge of heart failure pathophysiology and pharmacology (Fig. 105–4). First, the treatment should aim to optimize renal perfusion. This includes optimizing cardiac function, which may include use of inotropes and mechanical assist devices.²⁰⁹ Second, addition of non-loop diuretics should be tried. Thiazide-type diuretics are the most commonly used diuretics in patients who develop resistance to loop diuretics. Thiazide diuretics are a good choice because they act on the distal tubule NaCl symporter, which largely mediates the intrinsic renal adaptations to loop diuretics called the “braking phenomenon.” In other words, with prolonged loop diuretic use, the kidney adapts by upregulating thiazide-sensitive NaCl reabsorption. Mineralocorticoid receptor antagonists and acetazolamide may also be considered, the latter being particularly useful in the setting of diuretic-induced alkalosis. Ultrafiltration may also be effective in DR, but unfortunately the only large-scale randomized controlled trial was terminated early.^204,210,211

Decreased diuretic absorption due to gut edema is particularly common in the patient with right-sided congestion (increased central venous pressure, a surrogate of which can be detected at the bedside by examining for jugular venous distention). Torsemide has been shown to have better oral bioavailability than furosemide and may be useful in the patient with DR due to poor absorption.

An important function of the kidney is to maintain electrolyte homeostasis. Electrolyte homeostasis may be disturbed in patients with chronic heart failure and may be acutely impaired during acute decompensated heart failure. An important contributing factor is often the fact that many cardiovascular drugs affect renal handling of electrolytes. The pathophysiological consequences of heart failure per se are hyponatremia (activation of vasopressin) and hypokalemia (increased secretion of K⁺ driven by the RAAS). However, because of the use of RAAS inhibitors in clinical practice and concomitant CKD, hyperkalemia is a greater clinical problem in these patients than hypokalemia. Worsening renal function in the patient with heart failure may lead to dangerous electrolyte abnormalities.

Hyponatremia

Hyponatremia is defined as serum [Na⁺] < 135 mEq/L and is the most common electrolyte disorder among hospitalized patients. It is common in chronic as well as in acute decompensated heart failure, and is associated with increased mortality.^212,213 Hyponatremia can lead to neurological dysfunction, decreased mental function, cerebral edema, gait disturbances resulting in falls, and osteoporosis/fractures.

Under normal conditions, osmotic homeostasis is largely mediated by arginine vasopressin (AVP) and is controlled predominantly by osmo-receptors in the hypothalamus. Volume status and blood pressure (arterial pressure sensors) affect AVP release to a lesser degree. AVP increases free water reabsorption in the distal tubules by increasing the number of aquaporins (water channels). Water is then moved passively down its chemical gradient, driven by the hyperosmotic renal medullary interstitium. AVP levels are elevated and aquaporin expression is increased in heart failure.^214,215,216

The first step in the approach to a patient with suspected hyponatremia is to confirm plasma hypotonicity. Plasma tonicity may be normal or even increased in hyponatremia if unmeasured osmoles are present (eg, glucose in uncontrolled diabetes).²¹⁷ Once hyponatremia has been established, it is important to differentiate between dilutional hyponatremia (primary increase in water reabsorption) and depletional hyponatremia (net loss of sodium), as these will necessitate different treatment approaches.²¹⁸

Dilutional hyponatremia occurs when effective circulatory volume is reduced and is mediated by baroreceptor activation as well as angiotensin II and sympathetic overdrive (the latter also increases nonosmotic AVP release).²¹⁸ Dilutional hyponatremia should be corrected by limiting water intake and by promoting free water excretion, eg, by administering an AVP antagonist such as tolvaptan or by loop diuretics. Loop diuretics increase fluid delivery to the distal nephron and reduce interstitial hypertonicity, thereby decreasing water reabsorption. Hypertonic or isotonic saline may be co-administered with loop diuretics to increase efficacy. Acute treatment of dilutional hyponatremia should focus on free water excretion, whereas prevention of a positive free water balance is the focus in the chronic setting. Vasodilators and/or inotropes may in theory improve effective circulatory volume and reduce the nonosmotic release of AVP. Tolvaptan, a selective vasopressin V2 receptor antagonist, is approved for up to 30-day treatment of hyponatremia (Na < 125 mEq/L) in the setting of hyper- or euvolemia. The risk of hypernatremia, too rapid correction of the hyponatremia, and hepatic toxicity have limited the use of this drug. Of note, the EVEREST²¹⁹ trial failed to demonstrate a benefit of the drug for heart failure outcomes.

Depletional hyponatremia may occur secondary to loss of sodium into third spaces (diarrhea, ascites), excessive diuresis (diuretic use or osmotic diuresis [eg, diabetes]), and/or potassium or magnesium deficiency. Depletional hyponatremia should be corrected with isotonic saline, which provides the necessary sodium. Diluted urine can then be produced to reestablish eutonicity.²¹⁸ Any potassium and magnesium deficiencies should also be corrected. Chronic hyponatremia should be corrected slowly (< 5 mEq/L/d) due to the risk of central pontine myelinolysis with too rapid correction of plasma osmolality.²²⁰

Hyperkalemia

Hyperkalemia is defined as serum potassium > 5.0 mEq/L (severe hyperkalemia is defined as > 6.5 mEq/L) and is associated with increased mortality. Elevation of plasma potassium concentration reduces the electrochemical gradient for potassium across the cell membrane, leading to partial depolarization of the cardiac cells. This may lead to conduction delay (reflected on the ECG by widening of the QRS complex), which increases the risk of ventricular arrhythmias including ventricular fibrillation. Severe hyperkalemia can alter the resting membrane potential sufficiently to reduce automaticity of cardiac pacemaker cells, leading to severe bradycardia or even asystole.²²¹

Potassium homeostasis is controlled by potassium secretion at the distal renal tubule (distal convulated tubule and proximal collecting duct). In CKD, the kidneys can compensate for reductions in GFR and loss of functioning nephrons (tubulointerstitial injury) by increasing K⁺ secretion. However, they become vulnerable to acute increases in K⁺ load, which increases the risk of hyperkalemic episodes.²²² Importantly, hyperkalemia is an important potential side effect of many drugs used for treatment of CVD. The clinically most important are the RAAS inhibitors, which directly inhibit K⁺ secretion. Patients with CKD who receive RAAS inhibitors are at particularly high risk of developing hyperkalemia.²²³

Interventions used to reduce serum potassium or the effects of hyperkalemia are outlined in Table 105–7. Urgent management is typically indicated at [K⁺] > 6.0 mEq/L, but additional factors should be considered (eg, electrocardiographic changes [peaked T waves and prolonged P-R interval or QRS complex, ventricular arrhythmias, or pronounced bradycardia]). Acute management should focus on swiftly lowering [K⁺] and stabilize membrane potential, whereas a cornerstone of chronic management is to identify and correct the causes of hyperkalemia (eg, dietary intake, diuretic regimen, hemodynamics).