16 Cytokines and inflammatory markers

The expression of classic neurohormones, such as angiotensin II and noradrenaline, plays an important role in disease progression in chronic heart failure (HF). This so-called neurohormonal activation seems to be involved in the cardiomyopathic process of adverse left ventricular remodelling and dysfunction, via both direct and indirect effects.^1,2 Therapies blocking the excessive activation of the renin–angiotensin system and the adrenergic system have become the mainstay of pharmacological treatment of chronic HF.³

Another important pathway in chronic HF progression is inflammatory activation.^4,5 Experimental studies have shown that proinflammatory cytokines may induce many aspects of the syndrome of chronic HF, such as left ventricular dysfunction, pulmonary oedema, and the process of left ventricular remodelling, including myocyte hypertrophy, progressive myocyte loss through apoptosis, and endothelial dysfunction. Although the cause of the inflammation is unknown, both infectious (e.g.endotoxins) and noninfectious (e.g. oxidative stress, haemodynamic overload) events could be operating, including interaction with the neurohormone system. Inflammatory markers have emerged as potential indicators of the evolution of HF, ranging from their use for screening, diagnosis, determining prognosis, and guiding treatment.⁶

The emerging association of inflammatory mediators with the pathogenesis and progression of chronic HF has already resulted in the development of new anti-inflammatory strategies, which might be used as adjunctive therapy in patients with chronic HF.^7,8 Moreover, there is accumulating evidence that a critical network of interactions is formed by inflammatory and the classic neurohormonal mediators, and that many of the conventional therapies for HF may, at least partially, modulate the proinflammatory cytokine milieu. However, therapies tested so far have been largely disappointing.

In the early 1980s, cytokines were first characterized as a new group of peptides. They are secreted by different cell types and mediate cell-to-cell interactions via specific cell-surface receptors. They regulate key aspects of various cellular functions, such as activation, expansion, differentiation, and death.^9,10 The best-studied of these cytokines is tumour necrosis factor (TNF). Comparatively less is known about the interleukins (IL) IL-1, IL-2, IL-6, and interferon (IFN)-γ in the setting of chronic HF.

The chronic HF syndrome seems to progress, at least in part, as a result of the toxic effects exerted by endogenous cytokine cascades on the cardiac and skeletal muscle and on the peripheral circulation. So far, the origin of inflammatory mediators remains unclear and has been the subject of controversy. Several hypotheses have been described with respect to the source of proinflammatory cytokines in HF. One is endotoxin-induced immune activation secondary to bowel oedema (Fig. 16.1). Persistent immunological stimulation by microbial antigens, which translocate into the body from the oedematous gastrointestinal tract, may lead to cytokine production by monocytes in the bloodstream and possibly other tissues.¹¹ Lipopolysaccharide is a bacterial endotoxin which strongly induces the production of TNFα and other proinflammatory mediators.^12,13 However, this ‘infectious hypothesis’ fails to account for several clinical and experimental observations. The beneficial effects of antimicrobial therapy are not clear so far. Patients with chronic HF and no peripheral oedema also have elevated plasma cytokines, whereas patients with right heart failure, although oedematous, do not have elevations of plasma cytokines.

Nevertheless, the ‘cytokine hypothesis’ for disease progression in chronic HF (Fig. 16.2) does not depend only on the infectious hypothesis. Other factors may lead to the the tissue injury (‘tissue injury hypothesis’) and the myocardial cytokine production. Mechanical overload and shear stress may induce the myocardial production of cytokines, growth factors and stress proteins.^14,15 Hypoxia and ischaemia result in the expression of inflammatory cytokines such as TNFα, monocyte chemoattractant protein (MCP)-1, and IL-8 via activation of the transcription nuclear factor NF-κB.¹⁴ Oxidized low-density lipoprotein cholesterol is a potent inducer of cytokine expression.

Recent studies have suggested that a group of receptors named Toll-like receptors (TLRs) may be involved in immunological and inflammatory activation within the myocardium not only in response to microbes but also to molecules released from injured and stressed cells. Ligand binding leads to the activation of several kinases and NF-κB.¹⁵ Enhanced monocyte and macrophage

expression of stimulatory molecules, including proinflammatory cytokines, follow as downstream effects. In this way, a powerful immune response is possible even in the absence of infection.

The chronic HF syndrome may cause tissue hypoxia and free-radical production, in turn leading to NF-κB-mediated production of cytokines. Elevated plasma cytokines further reduce impaired vasodilator reserve, triggering a vicious circle of more severe tissue underperfusion.

Chronic β-adrenergic stimulation induces myocardial, but not systemic, elaboration of the major proinflammatory cytokines TNFα, IL-1β, and IL-6. There is thus a biological ‘crosstalk’ between the two cardinal neurohumoral systems in the myocardium.¹⁶

The deleterious effects of cytokines are mediated by reduced protein synthesis and increased protein degradation leading to muscle atrophy of the central (cardiac) and, perhaps more importantly, peripheral (skeletal) muscles. Proinflammatory cytokines and oxidative stress cause insulin resistance and downregulation of gene expression for the anabolic peptide IGF-I and reduce phosphorylation of the phosphatidyinositol-3-OH kinase (PI-3K), which in turn lowers the activation of the protein kinase B (Akt).^17,18 Reduced Akt activation in both skeletal muscle and heart (1) decreases protein synthesis via reduced phosphorylation of the ‘mammalian target of rapamycin’ (known as mTOR) and glucogen synthase kinase (GSK)¹⁹ and (2) up-regulates activity of forkhead box O (FOXO) transcription factors, which in turn activates the ubiquitin–proteasome pathway, resulting in protein degradation.^20,21 In addition, both reduced tissue IGF-1 and insulin resistance activate caspase 3, resulting in further myofibrillar protein breakdown and degradation through the ubiquitin-proteasome pathway.²²

Whether activated caspase 3 results in muscle apoptosis remains controversial.²³ ‘Skeletal myopathy’, resulting from the imbalance between increased muscle catabolism and attenuated muscle anabolism, plays a leading role in the genesis of symptoms of exercise limitation and, through the exaggerated muscle ergoreflex,²⁴ dyspnoea.

TNFα is one of the best-characterized proinflammatory mediators in chronic HF. It was first described in 1975 and named cachectin. TNFα is now recognized as a pleiotropic cytokine which can be expressed by almost all nucleated cells and has multiple actions on both local and systemic inflammation.

TNFα is increased in relation to the severity of chronic HF and correlates with the degree of sympathetic and renin–angiotensin system activation.²⁵ TNF is elevated not only in the circulation, but also in the myocardium of patients with chronic HF.²⁶ Recent research has shown that while TNFα rises in the serum of deteriorating patients who require left ventricular assist devices (LVADs), myocardioal expression of TNFα falls after mechanical circulatory support.^27,28

TNF levels are particularly elevated in cachectic chronic HF patients and these levels have been found to be the strongest predictors of the degree of previous weight loss.²⁹

TNFα may contribute directly to the evolution and progression of HF (Fig. 16.3) and is a predictor of worse outcome. In a substudy of the SOLVD study, patients with TNFα plasma levels above 6.5 pg/mL had worse survival.³⁰ In a large population of advanced chronic HF patients, circulating TNFα was an independent predictor of mortality.³¹

Although TNF is usually thought of as being harmful, some studies have suggested that TNF has a protective, inotropic action on the failing heart as a stress response protein.³² Further, injection of TNF improves survival of TNF knockout mice with viral myocarditis in a dose-dependent manner by increasing viral clearance.³³

TNFα has been implicated in the development of left ventricular dysfunction and remodelling. It increases cardiac myocyte apoptosis, via activation of cytokine-induced nitric oxide (NO) synthase, ceramidase and sphingomyelin pathways, NF-κB activation, and the eventual uncoupling of β-adrenergic signalling³⁴ as well as via mitochondrial DNA damage. TNFα can mediate cardiac myocyte mitochondrial DNA damage and, therefore, dysfunction in cardiac myocytes via enhanced oxidative stress (overproduction of reactive oxygen species).³⁵

TNFα is capable of inducing endothelial dysfunction and skeletal muscle wasting, leading to ‘skeletal myopathy’ through chronic tissue underperfusion, enhanced muscle catabolism, and possibly myocyte apoptosis. Serum TNFα is inversely correlated with skeletal muscle blood flow and exercise capacity, in both stable and decompensated patients with chronic HF.³⁶ It may also contribute to elevated insulin and leptin levels and the development of anorexia and cachexia.³⁴

The effects of TNFα on the heart are initiated by two specific receptors on myocytes: a lower affinity, 55-kDa receptor (TNFR1) and a higher affinity, 75-kDa receptor (TNFR2).¹⁰ TNFR1 is more abundant and appears to mediate the deleterious effects of TNFα, whereas TNFR2 appears to have a more protective role. Both receptors are cleaved from the cell membrane and subsequently converted to their soluble forms, sTNFR1 and sTNFR2. The soluble receptors may not only neutralize TNFα, inhibiting its highly cytotoxic activities, but perhaps also stabilize TNFα, potentiating its detrimental long-term actions in lower concentrations.³⁷

Like TNFα itself, sTNF receptors are highest in patients with severe decompensated chronic HF and in cachectic chronic HF patients, but are also increased in stable patients with mild chronic HF.³⁸ Plasma concentrations of sTNFR vary less than those of TNFα and are thought to indicate the history of inflammatory immune activation; therefore, they may be better markers of HF than serum levels of TNFα.

In patients with advanced chronic HF, sTNFR2 is a strong predictor of mortality, which may reflect its ability to act as a ‘slow-release reservoir’ of bioactive TNF into the circulation.³⁰ In the setting of acute myocardial infarction, which is a powerful trigger for cytokine activation, sTNFR1 is a better short- and long- term predictor of death and chronic HF.³⁸ The reasons for the discrepancy may relate to differences in clinical settings, study population sizes, and demographics. It is not yet clear whether circulating sTNFR is an epiphenomenon, simply indicative of a generalized inflammatory state and not a true causal mediator of disease progression. Recent evidence suggests that signalling through both receptors is required to induce the inflammatory and remodelling responses to TNF and the overall balance between opposing receptor-specific effects determines the ultimate impact of TNF.³⁹

Circulating IL-6 is elevated in patients with chronic HF in relation to disease severity.⁴⁰ Like TNF, IL-6 is a maladaptive protein, which participates in the development and progression of HF by exerting direct toxic effects on the heart and peripheral circulation. Its role is complex, as it has both proinflammatory and anti-inflammatory effects.⁴¹ IL-6 may induce a hypertrophic response from myocytes and cause myocardial dysfunction by NO generation, but also appears to block cardiac myocyte apoptosis.⁴² In the peripheral circulation, IL-6 production may contribute to abnormalities of endothelium-dependent vasodilation, vascular resistance, increased vascular permeability, or muscle wasting. IL-6 spillover in the peripheral circulation increases with the severity of chronic HF and is associated with sympathetic nervous system activity.⁴³

Serum IL-6 is produced by many cell types, including leucocytes, endothelial cells, vascular smooth muscle cells, cardiomyocytes, and fibroblasts.⁴⁴ IL-6 may be more important in the development of chronic HF than other inflammatory markers, as it has effects on platelets, endothelium, the coagulation cascade, and metabolism factors. IL-6 is a central mediator of the acute-phase response and a primary determinant of the hepatic production of C-reactive protein (CRP) and TNF. IL-6 concentrations, not surprisingly, correlate with TNFα and CRP levels.

IL-6 is a strong predictor of new-onset heart failure in healthy populations^45,46 and in older patients with acute ischaemic heart disease.⁴⁷ Its prognostic power in patients with established chronic HF is not clear. Increased concentrations of IL-6 are independently associated with a poorer prognosis in chronic HF patients, but concentrations of its soluble receptor (IL-6R) are not.^10,43 Other reports have not found significant prognostic value of IL-6.⁴⁸ Because of a relatively high short-term variability and non-normal distribution of IL-6 concentrations, interpretation of results may vary between studies.⁴⁹ In addition, episodes of myocardial ischaemia may trigger IL-6 production, making interpretation of prognostic data more difficult. Notably, a small subunit within the IL-6 receptor, named gp130, is a potent inducer of cardiomyocyte hypertrophy.^50,51 The gp130-signalling pathway seems to mediate the expression and activation of IL-6, IL-6/IL-6R complex, and other IL-6- related cytokines, such as leukaemia inhibitory factor and cardiotrophin-1, playing a critical role in both adaptive and maladaptive responses within the myocardium. Ventricular-restricted gp130 knockout mice develop dilated cardiomyopathy and profound myocyte apoptosis.⁵¹ However, the clinical significance of these findings in chronic HF remains uncertain.

The IL-1 cytokine family has four main members: IL-1α, IL-1β, IL-1 receptor antagonist (IL-1Ra), and IL-18. IL-1β is the major extracellular form in humans and a major proinflammatory cytokine.^52,53 The expression of IL-1β is increased in the coronary arteries and myocardium of patients with dilated cardiomyopathy when compared to those with ischaemic HF.⁵⁴ It has negative inotropic effects on the myocardium by uncoupling β-adrenergic signalling in a dose-dependent fashion, and it depresses myocardial contractility by stimulating NO synthase and ceramidase pathways.⁵⁵ IL-1 β may also suppress cardiac function by increasing cyclooxygenase-2 and phospholipase A2 gene expression, and by phosphatidylinositol-3′ kinase activation, which results in NF-κB activation.^52,53 In addition, IL-1 is involved in myocyte apoptosis, hypertrophy, and arrhythmogenesis. IL-1β is elevated in the myocardium of patients with HF and is increased in deteriorating patients.⁵⁶

There is still little information about endogenous IL-1Ra, a naturally occurring cytokine, which blocks the action of IL-1, attenuating its effects. IL-1Ra is often considered a more sensitive marker of IL-1 system activation than IL-1 levels.⁵⁷ However, given that IL-1Ra is a specific antagonist of IL-1, elevated levels of IL-1Ra could represent an appropriate response to counteract the inflammatory process caused by IL-1.⁵⁷

IL-18 is a more recently identified member of the IL-1 family, initially identified for its role in inducing IFN-γ production.⁵⁸ IL-18 is produced by vascular endothelial cells and macrophages in the human heart. It is a proinflammatory cytokine with multiple biological functions and, like many cytokines, it acts synergistically with other similar proteins and mediators.⁵⁹ IL-18 is a strong predictor of future cardiovascular risk in stable and unstable angina⁶⁰ and is up-regulated in the myocardium of patients with chronic HF. Although ischaemic insult is a major trigger of IL-18 expression, enhanced IL-18 processing is seen in patients with chronic HF of either ischaemic or nonischaemic origin.⁶¹ IL-18 may aggravate the inflammatory response via increased expression of endothelial cell adhesion molecules and secretion of proinflammatory mediators. It is a potent antiangiogenic cytokine and its inhibition might have beneficial effects on tissue remodelling.⁵⁸ IL-18 up-regulates membrane Fas ligand expression and may therefore contribute to Fas-mediated apoptosis of Fas-expressing cardiomyocytes.

The glycoprotein GM-CSF stimulates the proliferation, differentiation, and activity of multiple myeloid cells including neutrophils, monocytes/macrophages, eosinophils, and dendritic cells. It belongs to the large family of haemopoietic cell colony-stimulating factors and stimulates a range of activities, including leucocyte adhesion, free-radical generation, and cytokine production.⁴⁴ It is implicated in myelosuppressive disorders, drug-induced agranulocytosis, and immunodeficiency syndromes. In atherosclerosis, GM-CSF has angiogenic properties and confers some protective effects.⁶² In human tissue from endstage HF, GM-CSF is highly expressed. Elevated GM-CSF levels have been demonstrated in chronic HF, which were associated with both the neurohormonal activation and haemodynamic deterioration.⁶³

IL-10 was initially described as a cytokine synthesis inhibitory factor. It is produced by a variety of inflammatory cells, especially macrophages and T-cells, and is found in advanced atherosclerotic plaques, where it confers a protective effect: IL-10 levels predict outcome in acute coronary syndromes.⁶⁴ IL-10 inhibits monocyte adherence to human aortic endothelial cells in vitro. The ability of IL-10 to suppress certain CD40/CD40L ligand-mediated monocyte responses may account for some of its antiatherogenic effects.⁶⁵

IL-10 is involved in the production of matrix metalloproteinases (MMP) and cytokines, activation of NF-κB, and apoptosis and cell death.⁶⁶ IL-10 down-regulates the secretion of TNFα, IL-1, and IL-6 but enhances the release of sTNFR, contributing to the reduction of TNFα activity. IL-10 also attenuates the production of macrophage-derived NO and oxygen free radicals. Circulating IL-10 levels can be elevated in patients with dilated cardiomyopathy and IL-10 mRNA can be detected in the failing myocardium, possibly as a counter-regulatory response.

One study indicated a differentiation in cytokine patterns with respect to HF aetiology; IL-10 was much lower in patients with dilated cardiomyopathy as compared to ischaemic cardiomyopathy.⁶⁷ On the other hand, other studies have shown decreased plasma levels of IL-10 with the lowest concentrations observed in advanced chronic HF. Administration of immunoglobulin to chronic HF patients increased plasma concentrations of IL-10 and improved left ventricular ejection fraction (LVEF).⁶⁸

TGFβ deactivates macrophages by suppressing inducible NO synthase protein expression. It is a potent negative regulator of inflammation in vascular cells by down-regulating cytokine-induced expression of adhesion molecules.⁶⁹ Members of the TGFβ superfamily are able to promote the differentiation of embryonic stem cells into cardiomyocytes.⁷⁰ Patients with idiopathic dilated cardiomyopathy have increased TGFβ₁ gene expression in macrophages associated with increased plasma concentrations. Excessive production of TGFβ₁ may reflect either an adaptive role of macrophages in tissue repair or impaired ventricular compliance with increased collagen deposition.⁷¹

Chemokines are a family of chemotactic cytokines and are important factors in the control and regulation of leucocyte trafficking into inflamed tissues.^72,73 The attraction of leucocytes is essential for inflammation and the host response to infection but may also play a critical role in the pathogenesis of chronic HF. Chemokines may promote myocardial failure both directly (e.g. modulation of apoptosis, fibrosis, and angiogenesis) and indirectly (e.g. recruitment and activation of infiltrating leucocytes). Chemokines are classified into three distinct families on the basis of structure and function: C-C (e.g. monocyte chemoattractant protein-1), CXC (e.g. IL-8), and CX3C (e.g. fractalkine).

MCP-1 belongs to the C-C subfamily, which lack an amino acid between the first two N-terminal cysteine residues.⁷³ It is produced by a variety of leucocytes, endothelial cells, and fibroblasts. MCP-1 is mainly characterized by its ability to induce directional migration of leucocytes, with a crucial role in controlling inflammation and immune responses. MCP-1 possesses chemotactic and activating effects for both monocytes and lymphocytes, and in particular, MCP-1 is a major signal for the accumulation of mononuclear leucocytes in disease.

The pathogenic role of MCP-1 (and its receptor CCR2) in atherosclerosis and its complications is via monocyte and neutrophil interactions with endothelium.⁷⁴ Raised levels of MCP-1 have been found in cardiac lymph and in the endothelium of small veins from ischaemic canine myocardium. Pressure overload induces myocardial expression of MCP-1, which attracts and activates monocytes and macrophages, and the recruited cells produce proinflammatory cytokines.^75,76 Hypoxia and ischaemia are also potent inducers of MCP-1, involving activation of NF-κB.

Myocardial overexpression of MCP-1 is associated with monocyte infiltration of the myocardium and cardiac hypertrophy, ventricular dilatation, and depressed contractile function. Serum MCP-1 correlates with the degree of left ventricular dysfunction and may also be involved in cardiomyocyte apoptosis in severe HF. Chronic exposure to MCP-1 favours myocardial apoptosis and ventricular dysfunction by inducing transcriptional factors: gene therapy directed against MCP-1 may slow the progression of HF.⁷⁷

MIP-1 is a C-C chemokine produced by various types of inflammatory cells, exerting chemotactic activity for both monocytes and lymphocytes. MIP-1 is high in patients with chronic HF, with particularly high levels in patients with the most severe HF.⁷⁷ Abundant expression of MIP-1 may be an important factor in mediating the infiltration and activation of mononuclear leucocytes into the myocardium of chronic HF patients but may also have other functions, such as generating reactive oxygen species and cytokine production.

RANTES is a member of the C-C chemokine group, produced by a variety of cell types including platelets. It is a potent chemoattractant for T cells and monocytes, and is implicated in inflammatory diseases including atherosclerosis. RANTES may also modulate free-radical generation and the production of other cytokines.⁷⁸ It is highly expressed within atheroma and is up-regulated (and has prognostic significance) in acute coronary syndromes.⁴⁴ RANTES is elevated in patients with advanced chronic HF and may have a role in disease progression via its effects on platelet-inflammatory cell interactions.⁷⁷ However, data on RANTES in chronic HF are sparse and further studies are needed.

IL-8 is probably one of the best characterized neutrophil chemoattractants and degranulating agents. It is consistently found in macrophage-rich atherosclerotic plaques and it is thus implicated in early atherosclerotic progression.⁷⁹

IL-8 is increased in patients with chronic HF with particularly high concentrations in those with the most both severe HF. Activated monocytes and platelets may contribute to increased levels of IL-8 in chronic HF.⁷² IL-8 may be an important participant in both the systemic inflammatory response and the procoagulant activity in chronic HF. High IL-8 serum levels fall to near normal after haemodynamic recovery following ventricular assist device placement, suggesting that IL-8 may be marker of tissue damage.⁸⁰

Cell adhesion molecules (CAMs) are involved in the interactions between endothelial cells, leucocytes, and platelets. Thus, they have been implicated in a vast range of conditions including atherosclerosis, thrombosis, allograft rejection post-transplantation, and restenosis following coronary angioplasty.⁵⁴ Three families of proteins have been described so far. The intracellular cell adhesion molecule-1 (ICAM-1) and vascular cell adhesion molecule-1 (VCAM-1) belong to the immunoglobulin superfamily. Integrins form the second subfamily. The selectins cause a typical ‘rolling’ of leucocytes on the endothelial surface, which is mainly mediated by leucocyte (L)-selectin and platelet (P)-selectin.

The significance of CAMs in chronic HF is unclear. The failing myocardium, and in particular the microvascular endothelium, gives signals to assist in leucocyte infiltration, via the up-regulation and/or secretion of CAMs including P-selectin, e-selectin, L-selectin, ICAM-1 and VCAM-1. These molecules are important mediators of both endothelial–leucocyte adhesion and inflammatory responses.^81,82 Damage induced by oxygen free radicals and cytokine activation are stimulators of CAMs. The soluble forms of the adhesion molecules, generated by proteolytic cleavage of cell membrane-bound molecules, act as systemic activation signals for circulating cells.

There is increased endothelial production of adhesion molecules in chronic HF, increased expression of sICAM-1 and integrin CD11a/CD18 (lymphocyte function-associated antigen-1), and increased levels of soluble adhesion molecules. VCAM-1, E-selectin, and P-selectin are all also raised.^83,84 Soluble adhesion molecules (sVCAM-1 and sL-selectin) decrease after the implantation of mechanical circulatory support devices in patients with decompensated HF.

sICAM-1 increases with increasing severity of failure, which suggests that ICAM-1 may be associated with an adverse prognosis. High levels of sP-selectin or VCAM-1 are also independent predictors of outcome in patients with endstage HF.⁸⁴ However, data regarding the prognostic value of CAMs are inconsistent. The limited sample size and the different immunological actions of the studied CAMs are all possible explanations for the inconsistency. Furthermore, the different sCAMs vary differently with time, HF treatment, and heart transplantation.⁸³

NF-κB is a transcription factor mainly involved in stress-induced, immune, and inflammatory responses. Activation of NF-κB can be triggered by multiple stimuli, such as angiotensin II, TLR, IgG and reactive oxidant species (Fig. 16.4). Functional NF-κB requires formation of heterodimers of the p50 and p65 subunits.⁸⁵ Activation of NF-κB involves the degradation of its inhibitory proteins IκBs by specific kinases. The free NF-κB passes into the nucleus, where it binds to sites in the promoter regions of genes for inflammatory proteins such as TNFα, IL-1β, inducible NO synthase and adhesion molecules.⁸⁶ The activation of NF-κB leads to a coordinated increase in the expression of many genes, whose products are important mediators in the pathogenesis of chronic HF.

NF-κB is activated in myocardial tissue of the failing human heart.⁵ Activation of NF-κB ameliorates myocardial hypertrophy and is involved in pro and antiapoptic pathways in HF. Strategies targeting NF-κB improve the long-term prognosis in HF. In mice with targeted disruption of the NF-κB subunit p50, early survival after myocardial infarction is increased and ventricular dilatation is prevented, which is linked to decreased collagen production and deposition.⁸⁷

The free-radical gas nitric oxide (NO) is enzymatically formed from L-arginine by three isoforms of NO synthase, which are all present in the heart:

NO inhibits platelet adherence and aggregation, induces vasodilatation, reduces the adherence of leucocytes to the endothelium, and suppresses the proliferation of vascular smooth muscle cells.⁸⁹

iNOS expression is increased with increasing severity of HF. It is unclear whether wall stress or cytokine activation is the predominant stimulus for iNOS. TNFα is a potent inducer of iNOS expression in both endothelial and vascular smooth muscle cells resulting in enhanced NO production. NO can exert both negative inotropic and apoptotic effects on cardiac myocytes. Therapy with LVADs was found to normalize iNOS expression in association with diminished cardiomyocyte apoptosis in the failing heart.⁸⁹

CRP is a simple downstream marker of inflammation. Il-6 is the primary stimulus for the hepatic production of CRP within 6 h of stimulus.⁹⁰ CRP can also be produced from vascular walls, particularly in the atherosclerotic intima of human coronary arteries. Left ventricular dysfunction, systemic underperfusion by low cardiac output, hypoxia, and venous congestion may all be sources of increased IL-6 and hence, CRP production.

CRP might worsen HF through multiple mechanisms (see Box 16.1). CRP is raised in chronic HF and higher plasma levels of CRP are associated with a worse haemodynamic and clinical profile; however, it is unclear whether the finding is related to active atherosclerosis.^91,92 Raised CRP is a predictor of future heart fai

lure and adverse events in patients with vascular disease,⁹³ but its prognostic value in patients with established HF is less clear. There are no data in HF patients on the effects of treatment on CRP.

Leptin is a major regulator of body mass and appetite.^94,95 In animal studies, it induces weight loss and anorexia and suppresses cardiac contractility through an NO-dependent pathway. Leptin is primarily produced by adipocytes, whereas its receptors are expressed in a variety of tissues, including the heart. Leptin is also related to fat mass in HF patients and therefore, it should be interpreted once corrected for fat mass.⁹⁶. Leptin can modulate other cytokines including interference with NF-κB effects. Leptin is increased in noncachectic HF patients normal or even innapropriately low in cachectic patients—a paradox which may be related to sympathetic nervous activation.

Activin A is a member of the TGFβ superfamily involved in growth, differentiation, and survival. Activin A is raised in patients with chronic HF. Activin A is involved in ventricular remodelling by enhancing MCP-1 production, the generation of TGFβ₁ and MMP, and specific gene expression associated with myocardial hypertrophy.⁹⁷

Fas and Fas ligand belong to the TNF receptor superfamily and might contribute to inflammation, apoptosis, and matrix degradation within the failing myocardium.⁹⁸

The cross-linking of Fas with Fas ligand mediates apoptosis by triggering caspase activation. Soluble forms of Fas are raised in autoimmune diseases, myocarditis and severe chronic HF. Fas/Fas ligand are associated with left ventricular remodelling and may have prognostic value in chronic HF.

The osteoprotegerin receptor activator for NF-κB (RANK)/RANK ligand (RANKL) axis is another member of the TNF receptor superfamily, which is a mediator in both experimental and clinical HF.⁹⁹

The beneficial effects of the traditional cardiovascular medications cannot be explained solely by their haemodynamic effects. Some drugs used in the treatment of chronic HF may also influence the persistent immune activation and inflammatory pathways.

Treatment with high doses of angiotensin converting enzyme (ACE) inhibitors reduces circulating levels of IL-6.¹⁰⁰ ACE inhibitors may affect TNFα and can reduce insulin-like growth factor-1 (IGF-1). Interestingly, ACE inhibitors may prevent NF-κB activation and MCP-1 expression, and reduce macrophage infiltration. Angiotensin II receptor antagonists down-regulate inflammation by reducing plasma levels of TNFα, IL-6, and brain natriuretic peptides in mild to moderate HF.¹⁰¹

Amlodipine reduces IL-6 levels, but with no effect on TNFα.⁴ β-Adrenergic stimulation may modulate cytokine production from lymphocytes and monocytes. Carvedilol reduces IL-6.¹⁰² On the other hand, long-term treatment with metoprolol has no significant effect on cytokine levels. The differential effects of α- and β-blockade on the cytokine network still need to be determined in more detail.

Statins may attenuate inflammatory responses and promote plaque stability independent of their cholesterol-lowering effects.^4,104 Statins reduce CRP levels and may be effective in preventing coronary events in patients with relatively low lipid levels but with elevated CRP.¹⁰³

Apart from their short-term haemodynamic benefits, phosphodiesterase inhibitors can inhibit the production of TNFα and other cytokines in failing myocardium.¹⁰⁴ However, phosphodiesterase inhibitors are also related to an adverse outcome in chronic HF. Cardiac glycosides can also reduce levels of IL-6 and TNFα in vivo. Amiodarone causes a significant decrease in TNFα production by human mononuclear cells, suggesting a possible mechanism for its effect in HF.

However, some of the effects on the immune system may be secondary to improved left ventricular function and not a direct effect of the drugs. A nonpharmacological approach, physical training, induces beneficial changes in exercise capacity which are correlated with a reduction in inflammatory markers (Fig. 16.5).¹⁰⁵ Exercise training restores, at least partially, abnormal immuno-inflammatory responses by depressing systemic inflammation with reduced proinflammatory cytokine (TNFα and IL-6) expression at both circulation¹⁰⁶ and tissue¹⁰⁷ level. In consequence, there is a fall in oxidative stress as demonstrated by the reduced expression of iNOS and the increased activity of radical scavenger enzymes in skeletal muscle.¹⁰⁸ There is an inverse relation between iNOS expression and cytochrome c oxidase activity, suggesting that local anti-inflammatory effects may contribute to improved skeletal muscle oxidative metabolism with physical training.¹⁰⁹ The local anti-inflammatory effects of exercise may slow down or even reverse the catabolic wasting process associated with the progression of HF.

Given the potential central role of TNFα, etanercept, a recombinant sTNFR type 2 protein, which functionally inactivates TNFα, has been tried as a treatment for chronic HF. An initial study showed beneficial effects on cardiac function and left ventricular remodelling in a small population with severe chronic HF. Subsequently, the long-term effects of etanercept were assessed by three large placebo-controlled trials: RENEWAL, RENAISSANCE, and RECOVER. The trials were terminated prematurely because of lack of evidence of beneficial effects.¹¹⁰

Infiximab is a chimeric monoclonal antibody which directly binds to the transmembrane form of TNF. Its use was associated with time- and dose-related increase in death and HF hospitalization in patients with moderate/severe HF.¹¹¹ The xanthine derivative pentoxifylline, which exerts peripheral vasodilatory effects and improves blood flow, also reduces TNFα. Treatment with pentoxifylline was associated with a significant improvement in functional class and LVEF, along with a decrease in circulating TNFα.¹¹² The observed changes did not correlate to each other, suggesting that the beneficial effects of pentoxifylline were independent of TNFα.

The failure of anti-TNFα therapy has led to much discussion. The intervention on a single cytokine may not be sufficient to have an impact on the progression of HF. TNFα is a pleiotropic cytokine, also implicated in cardioprotective pathways, at least early on in the disease process. It may be that immunomodulation is only or particularly valuable for those patients with strongly up-regulated inflammatory status. Etanercept and infliximab may even have increased the biological half-life of TNFα, or it may be that the dose used was not sufficient to inhibit TNFα function. The future of anti-TNF therapy is very uncertain. Other agents that modulate the production of TNFα, such as inhibitors of lysophosphatidic acid acyl transferase, p38 MAP kinases, NF-κB, and TNFα converting enzyme may be treatments of the future.

IVIG administration improves LVEF, haemodynamics, and exercise capacity in patients with chronic HF, independent of aetiology.¹¹³ Others have found no impact of IVIG on recent-onset idiopathic dilated cardiomyopathy.¹⁰⁴ To add to the controversory, long-term therapy results in improved LVEF, associated with a marked elevation of IL-10, IL-1 receptor antagonist, and sTNFR. IVIG also reduced chemokines and their receptors in peripheral blood mononuclear cells, suggesting that direct blockade of the chemokine network may be an approach for future intervention.

A pilot study of INF-1β for the treatment of virus-related dilated cardiomyopathy showed improvement in symptoms and quality of life but no change in objective variables.¹⁰⁴ It is being investigated in a large phase III trial.

Celacade is an immunomodulation therapy designed to target chronic inflammation by activating the immune system’s physiological anti-inflammatory response. A blood sample is rapidly exposed to a combination of physiochemical stressors ex vivo and then reinjected intramuscularly in an attempt to evoke beneficial immune responses. The physiological response of the immune system to the reinjected apoptotic cells is likely to increase inflammatory cytokine production, rather than impair it. Phase I and II clinical trials of celacade showed a low risk of side effects and improved quality of life in HF patients.^114,115 ACCLAIM, a phase III trial, did not show any significant reduction in mortality or cardiovascular hospitalization,¹¹⁶ but there was a benefit in some subgroups, particularly those with NYHA II class HF and those without a history of previous myocardial infarction.

Immunoadsorption allows the removal of circulating autoantibodies, such as those against the β₁-adrenergic and muscarinergic receptors or troponin I. It improves cardiac structure and function and decreases oxidative stress and myocardial inflammation.^99,104 It may only be effective in patients with cardiodepressant autoantibodies or only in combination with IVIG acting additively or synergistically.

Thalidomide has both anti-inflammatory and antioncogenic properties and has recently been evaluated on a limited number of patients. Other immunomodulatory treatments have the potential to improve myocardial function. Pentraxins are cytokine-inducible genes, expressed in specific tissues, which reduce early myocardial damage after myocardial infarction. Drugs targeting the kinase activity of PI3Kγ—a major component of signal transduction controlling leucocyte migration—can reduce cardiac inflammation. The MMP system is involved in ventricular remodelling: its inhibition may have beneficial effects, particularly in acute HF where MMP expression is related to acute dilatation and failure. Gene therapy with MCP-1 blocker attenuates the development of cardiac remodelling after myocardial infarction. IL-10 and IL-1R antagonist exert cardioprotective effects in viral myocarditis and against ischaemia reperfusion injury in mice.

Activators of peroxisome proliferator-activated receptors reduce endotoxin-stimulated TNFα expression and cardiac hypertrophy by inhibiting NF-κB activation, thereby decreasing inflammatory response. Tranilast is a mast-cell stabilizing agent, which has been found to modulate compensated hypertrophy, ventricular remodelling, and the production of anti-inflammatory cytokines like IL-10.

Other emerging therapeutic targets including mannose-binding lectin, IL-18 and IL-6 antagonists, and T-cell and caspase inhibitors, which can tackle inflammation in the heart. More knowledge on inflammatory cytokines in HF and larger placebo-controlled randomized studies will allow the development of more effective therapeutic options. Finally, a promising immunomodulatory approach with respect to the efficacy of immunosuppressive therapy in patients with chronic inflammatory cardiomyopathy was examined in the randomized TIMIC Study, underlining the importance of endomyocardial biopsy in dilated cardiomyopathy. Eighty-five patients with biopsy-proven myocarditis and no evidence of myocardial viral genomes who received 6 months prednisone and azathioprine, in addition to conventional therapy for heart failure, showed a significant improvement of left-ventricular ejection fraction and a significant decrease in left-ventricular dimensions and volumes compared with baseline (anti-remodelling effect).¹¹⁷