The primary myocardial defect in patients with dilated cardiomyopathy is an abnormality in contraction. As a result of reduction in myocardial contractility, the heart dilates out of proportion to the degree of hypertrophy, resulting in four-chamber enlargement with relatively thin cardiac walls (6). This increase in radius relative to wall thickness increases myocardial wall stress (7) and decreases subendocardial coronary flow (8), leading to further cardiac dysfunction. Figure 18-1 demonstrates a typical echocar-diogram from a patient with dilated cardiomyopathy. The M-mode echocardiogram shows the left ventricle (LV) to be markedly enlarged and systolic function reduced. The two-dimensional echocardiogram from the same patient demonstrates that all four chambers are enlarged.

In most cases of dilated cardiomyopathy, the mech-anism(s) for the initial myocardial insult are unknown. Furthermore, it is unclear whether the defect is one of primary individual myocardial cell loss or cell dysfunction followed ultimately by myocardial cell loss and
replacement with fibrous connective tissue. A host of molecular, biochemical, and cellular abnormalities have been described in patients with dilated cardiomyopathy. These include alterations in calcium handling by the sarcoplasmic reticulum (9), a reduction β₁-receptor density (10,11), alterations in Gs and Gi proteins (12,13), alterations in cholinergic receptor density and function (14), and expression of altered or fetal myosin subtypes (15). It is not clear, however, whether any of these noted abnormalities is specific to dilated cardiomyopathy or is a more general manifestation of cardiac dilatation and failure.

As previously noted, as early as the late nineteenth century it was hypothesized that cardiomyopathy is the result of chronic inflammatory disease (1). About 80 years later, the concept that dilated cardiomyopathy might be the result of an acute, subacute, or chronic myocarditis again received attention (16). The hypothesis put forth was that a myocardial insult, be it an acute viral infection or toxin, somehow altered the antigenicity of the myocardium so that it was mistaken for foreign tissue and an autoimmune attack was elicited (17). The problem with this hypothesis is that the pieces of the puzzle have been difficult to prove in a convincing way despite major research efforts to support them.

In terms of the initial insult, it has long been appreciated that the coxsackie group B virus is an important pathogen in acute viral myocarditis, and this has provided an excellent model for the study of acute myocarditis in animals. Furthermore, in some murine strains, acute cox-sackievirus myocarditis leads to a more chronic myocarditis and cardiac dilation suggestive of dilated cardiomyopathy (18). The difficulty in linking coxsackievirus to human dilated cardiomyopathy has been the inability to identify the virus in cardiac tissue. Serology is nonspecific (19), viral cultures are rarely positive (20), and electron microscopy for viral particles is difficult to interpret. With the recent growth of molecular biological techniques, new methods for viral diagnosis have become available. Bowles et al. (21) utilized in vitro nucleic acid hybridization to study whether enterovirus RNA is present in the myocardium of patients with dilated cardiomyopathy (21). Of the 40 patients studied, 21 had dilated cardiomyopathy and 19 were controls who had heart failure secondary to either coronary disease or congenital heart disease. Enterovirus RNA was detected in the myocardium of 6 of the 21 patients with dilated cardiomyopathy and only one of the controls.

A number of studies have investigated the immunological milieu in dilated cardiomyopathy. Limas and Limas (22) found that HLA-DR4 was represented in 40% of patients with dilated cardiomyopathy, but in only 24% of controls. Koike (23) investigated cell-mediated immune function in 18 patients with dilated cardiomyopathy. In these patients, phytohemagglutinin (PHA)-mediated lymphocyte blastogenesis was reduced; the ratio of CD4⁺ (helper) to CD8⁺ (suppressor) lymphocytes was higher; the PHA-mediated release of interleukin-2 (IL-2), a T-cell growth factor, was enhanced; and the number of activated lymphocytes (positive for IL-2 receptor) was reduced in comparison with controls. These data would suggest an inherited predisposition to immunological disease.

Antimyocardial antibodies have been extensively studied during the past several years. However, a link between the presence of antibody and an alteration in cardiac function or metabolism has not been established. In previous studies by Schultheiss et al. (24), an antibody directed against the adenosine diphosphate-adenosine triphos-phate (ADP-ATP) carrier protein of the mitochondrial membrane was described in patients with dilated cardiomyopathy. The regulation of β₁-receptors on myocardium in patients with heart failure has been well-described (10). The mechanism for this downregulation is felt to be a negative feedback related to chronically elevated levels of circulating catecholamines, presumably a nonspecific mechanism related to the degree of heart failure. Limas et al. (25) demonstrated that a high percentage of patients with dilated cardiomyopathy, especially those in whom HLA-DR4 was represented, had circulating anti-β₁-receptor autoantibodies. An additional mechanism for modulating cell surface β-receptors specific to dilated cardiomyopathy could therefore be hypothesized.

Mechanisms for the development of dilated cardiomy-opathy other than those involving immunological abnormalities have been investigated. Based on their experimental data derived from the Syrian hamster, Factor et al. (27) postulated that some forms of dilated cardiomyopathy are secondary
to small-vessel disease and specifically to microvascular spasm. This mechanism is presumed to play an important role in the dilated cardiomyopathy associated with pheochromocytoma (28). A study by Treasure et al. (29) demonstrated that the endothelium-dependent dilation of the coronary microvasculature is impaired in patients who have dilated cardiomyopathy in comparison with controls. Van Hoeven and Factor (30) reviewed endomyocardial biopsy specimens from 145 patients with dilated cardiomyopathy and demonstrated small-vessel thickening in 56% of these patients. This was not a specific finding, however, as a significant proportion of patients with hypertension, diabetes, unexplained arrhythmias, and chest pain syndromes had similar changes.

Direct myocardial toxicity is suggested for alcohol-induced cardiomyopathy (31) and is clearly present in anthracycline-induced dilated cardiomyopathy (31,32,33). Although alcohol is a potent depressant of myocardial function in the muscle bath, the variable effects of alcohol on the myocardium in vivo are poorly understood. Although several mechanisms (to be discussed later) have been suggested for anthracydine cardiotoxicity, there is at least a dose-related effect of anthracyclines on cardiac function.

What is becoming increasingly clear is that many of the idiopathic dilated cardiomyopathies have a genetic basis and, with more extensive screening, appear to be familial. This is a fairly new concept as, historically, most considered dilated cardiomyopathy to be sporadic in nature, unlike well-defined genetic diseases such as hypertrophic cardiomyopathy, which will be discussed later in this chapter. In fact, familial disease was found in as high as 48% of patients with apparent idiopathic cardiomyopathy when cardiac enlargement by echocardiography was used as a marker for early cardiomyopathy (34). Other studies using echo and other screening have suggested prevalences in the 25% to 35% range depending on the criteria used (35,36).

A recent review of the subject of familial dilated cardiomyopathy discusses these prior investigations as well as the specific genetic abnormalities identified to date. Ninety percent of familial dilated cardiomyopathy appear to be inherited with autosomal dominance, with 16 gene mutations identified. Defects in the lamin A/C or β-myosin heavy chain account for between 5% to 10%, with abnormalities in actin and desmin being less common (37). X-linked familial cardiomyopathy usually secondary to defects in the dystrophin gene may account for between 5% and 10% of familial dilated cardiomyopathy and is usually also associated with skeletal muscle involvement (muscular dystrophies). Rarely, autosomal recessive inheritance patterns have been described, as in one family with a mutation in the cardiac troponin I gene (37).

In most cases, proof of a genetic cause of a cardiomyopathy has a limited impact on the treatment of the index patient. Gene therapy, referring to the introduction of altered or foreign genes into human tissues for the purpose of correcting a genetic defect, is not now feasible for most genetic cardiomyopathies, and there are serious logistic reasons to question whether such treatment will ever be possible or desirable. However, the demonstration that a cardiomyopathy is genetic in origin has important implications beyond those of clinical management, such as the need for family screening to identify other affected persons, or for genetic counseling of affected patients who wish to have children.

Table 18-1 provides a partial listing of the causes of and conditions associated with dilated cardiomyopathy. Although the list is extensive, in the vast majority of patients there will be no obvious cause of disease

Duchenne muscular dystrophy, described in 1868 as a “hypertrophic paraplegia of infancy” (38), is an X-linked recessive neuromuscular disorder that affects primarily boys. Clinical manifestations are usually evident by age 5. Progressive muscle weakness is usually first evident in the lower extremities, with involvement of proximal musdes being more marked than that of distal musdes. Calf musdes, and sometimes other muscles as well, become enlarged, partly owing to true muscle hypertrophy and partly because of infiltration of the musde with fat. Joint contractures develop and walking becomes more and more difficult. Respiratory musde weakness leads to pulmonary infection and respiratory failure. Intellectual impairment may also occur. Laboratory findings include elevation of creatine kiriase and; electromyographic abnormalities (39).

Genetic mapping techniques led in 1986 to the doning and identification of the gene for Duchenne muscular dystrophy (40,41). As previously noted, the product of this gene, called dystrophin (42), is a rod-shaped cytoskeletal protein that is normally found on the cytoplasmic surface of the sarcolemma (43,44). It is known to bind on its amino-terminal end to actin (45) and on its carboxy-terminal end to a membrane-spanning group of proteins called the dystrophin-assodated glycoprotein complex (46,47). One
of the extracellular components of this complex named α-dystroglycan, binds to the extracellular matrix protein laminin (48). The result is that dystrophin acts as a link between the cytoskeleton and the extracellular matrix (49). It is therefore theorized that the function of dystrophin is to stabilize the sarcolemma and to protect it from the mechanical stresses of muscle contraction (50).

In patients with Duchenne muscular dystrophy, dystrophin is almost entirely absent from the muscle cell membrane (51,52). The absence of dystrophin from the sarcolemma is thought to expose the membrane to mechanical stress, which leads to abnormal ion fluxes, loss of force transduction across the membrane, and perhaps disruption of the membrane itself (53,54,55). Genetic studies have demonstrated that a variety of mutations, including deletions, duplications, and point mutations (56,57,58), can all lead to dystrophin deficiency.

Cardiac involvement has long been recognized to be a feature of Duchenne muscular dystrophy (59,60). Electrocardiograms are often abnormal in childhood (61) and echocardiography demonstrates that LV dysfunction is the rule by age 18 (62,63). Symptomatic congestive heart failure is less common, partly because the patient’s progressive motor debility places decreasing demands on cardiac function (64).

The extent of cardiac involvement does not correlate well with the severity of the skeletal muscle disorder (65). The myocardial dysfunction is most marked in the poster-obasal LV free wall (66,67); not only is this region of the heart hypokinetic but also its echodensity often appears to be increased, suggestive of fibrosis (68). These observations have led to the hypothesis that myofiber orientation and force distribution in the posterior LV differ from those of the anterior wall, leading to greater and earlier damage to the sarcolemma in the posterior wall in the absence of dystrophin (69). The focal nature of the ventricular dysfunction seen on noninvasive imaging corresponds to the histopaihological findings, which include myofibrillar loss and fibrosis most pronounced in the posterobasal LV (70,71). These observations may also explain the typical electrocardiographic features, which include tall R waves in leads V₁ and V₂ and deep, narrow Q waves in the lateral and sometimes inferior leads (72). Although dilated car-diomyopathy is the most common form of disease seen in Duchenne muscular dystrophy, hypertrophic cardiomy-opathy is seen in some patients (73). Arrhythmias and conduction disturbances have also been described (74). Because cardiac involvement may contribute to morbidity and may at times be the cause of death, regular screening for cardiac manifestations is recommended (75).

Becker muscular dystrophy is a less-severe form of X-linked muscular dystrophy. The clinical manifestations are similar to those of Duchenne muscular dystrophy but they occur later in life and progress more slowly (76,77). With the identification of the Duchenne muscular dystrophy gene and its protein product, it has been demonstrated that Becker muscular dystrophy is also a consequence of mutations in dystrophin (78,79). Typically, mutations causing Becker muscular dystrophy result in the production of a structurally abnormal dystrophin protein; in contrast, mutations causing Duchenne muscular dystrophy result in a severe quantitative deficiency of the protein (80,81). The clinical Collrse and degree of cardiac involve-ment in Becker muscular dystrophy are much more van-able than in Duchenne muscular dystrophy (82,83). Data suggest that even among sibling pairs with identical dystrophin mutations, the onset of cardiomyopathy is variable and unpredictable (84), and the cardiac dysfunction may precede the evidence of skeletal muscle weakness (85). Cardiac failure is actually a more common cause of death in Becker muscular dystrophy than in Duchenne muscular dystrophy. This may be because the skeletal muscle function of patients with Becker muscular dystrophy is relatively well-preserved, so that they remain more active and place a heavier work load on the myocardium, which leads to a more rapid progression of cardiomyopathy (86,87).

In 1993, Towbin et al. (88) reported that the family originally described by Berko and Swift, as well as another family with X-linked dilated cardiomyopathy, showed evidence of genetic linkage to the 5′ portion of the dystrophin gene locus. This report provided the first evidence that a dilated cardiomyopathy without any skeletal muscle manifestations Collld be caused by an abnormality of dystrophin. Since then, numerous other families with dystrophin mutations and isolated dilated cardiomyopathy have been described (89,90,91,92). Some of these reports have also investigated the abundance and structure of dystrophin in the heart and skeletal muscle of the subject families; both qualitative (93) and quantitative (94) dystrophin abnormalities have been detected. Absence of cardiac membrane-associated dystrophin can be seen with normal skeletal muscle findings (95,96).

A family history in which only the male relatives have dilated cardiomyopathy may be the first clue that a patient has a dystrophinopathy; however, some women may be affected as manifesting carriers, as previously noted (97). Furthermore, as many as a third of cases of Duchenne muscular dystrophy represent de novo mutations with no family history (98). Therefore, a diagnosis may be difficult to make based on kindred analysis alone. Creatine kinase levels are almost always elevated in patients with a dystrophin disorder; such elevations should prompt further investigation of skeletal muscle function, possibly a cardiac and skeletal muscle biopsy, and consideration of dystrophin gene or protein analysis (99).

Several other muscular dystrophies are known to have a significantly frequent association with cardiac involvement. Emery-Dreifuss muscular dystrophy is an X-liriked disorder that, in contrast to Duchenne muscular dystrophy, is characterized by an absence of calf hypertrophy and a distinct pattern of muscle wasting referred to as humeroperoneal (proximal muscles of the upper extremities and distal muscles of the lower extremities are the most affected) (100,101). Joint contractures are a prominent feature of this condition, especially at the elbows, Achilles tendons, and posterior cervical muscles (102). The most common forms of cardiac involvement are rhythm disturbances, including atrial fibrillation or flutter and conduction block (103). However, dilated cardiomyopathy has also been described (104,105) and cardiac transplantation for cardiac failure has been reported (106). Genetic mapping of the
relevant region of the X chromosome has led to the identification of the responsible gene (107). The protein product of this gene has been named emerin. It has been demonstrated to localize to the inner membrane of the nucleus (108) but its function has not yet been determined. A disorder that is clinically indistinguishable from Emery-Dreifuss muscular dystrophy, but with an autosomal dominant mode of inheritance, has also been described (109).

Limb-girdle muscular dystrophy refers to a group of disorders, some with autosomal dominant and some with autosomal recessive transmission, that typically are characterized by proximal muscle weakness and wasting that progresses to involve all extremiuies; the face is spared (110). Conduction system disease has been described in limbgirdle muscular dystrophy type I, and dilated cardiomy-opathy has been reported (111,112,113).

The mitochondria are the site of oxidative phosphorylation, the metabolic process that generates most of the ATP produced by eukaryotic cells. Pyruvic acid, the product of glycolysis, is taken up into the mitochondrial matrix and reacts with coenzyme A (CoA) to form acetyl CoA. The enzymes of the citric acid cycle, mostly in solution in the mitochondrial matrix, oxidize the acetyl group of acetyl CoA to carbon dioxide and, in the process, reduce nicoti-namide adenine dinucleotide (NAD) to NADH and flavin adenine dinucleotide (FAD) to FADH₂. Protein complexes on the inner mitochondrial membrane then transport electrons from NADH and FADH₂ through the cytochrome system to oxygen. The energy accumulated during this process is used to create high-energy phosphate bonds, which can then be used for all the metabolic requirements of the cell (114). Muscle and nerve cells, which rely heavily on oxidative metabolism to supply their high energy demands, typically contain large numbers of mitochondria (115). Genetic abnormalities of mitochondrial function, especially of oxidative phosphorylation, therefore tend to affect the nervous system, skeletal muscle, and the heart, although they may affect a wide variety of other organ systems as well (116). Mitochondria have their own genome, a circular molecule of DNA that is distinct from the nuclear chromosome. Mitochondrial DNA encodes several of the component polypeptides of the enzymatic pathway for oxidative phosphorylation; in addition, it encodes a set of distinct mitochondrial ribosomal RNAs and transfer RNAs, which are necessary for translation of the mitochondrial genes (117). The replication, transcription, and translation of mitochondrial DNA occur independently of the analogous nuclear processes. Each mitochondrion typically contains several copies of the mitochondrial genome, and each cell usually contains numerous mitochondria. Thus, the stoichiometry of mitochondrial genetic information is quite different from the strict diploidy of nuclear genes. Mutations may occur in one or several copies of the mitochondrial DNA but not in others. When cells divide, the daughter cells may inherit unequal numbers of mitochondria, some bearing mutations and others identical to the germline (118).

Mitochondrial DNA is maternally inherited—children inherit virtually all their mitochondrial genes from their mother, not their father (119). This is a consequence of the fact that the great majority of the mitochondria in a developing embryo are acquired from the ovum; the sperm contains very few mitochondria. As a result, mutations of mitochondrial DNA are transmitted from mother to child, whereas paternal mitochondrial mutations are not transmitted (120). This maternal pattern of inheritance is distinct from the usual mendelian categories and may be the first evidence leading to clinical recognition of a mitochondrial disorder (121,122). However, because some of the enzymatic components involved in oxidative phosphorylation are encoded in the nucleus, genetic defects of mitochondrial function may also exhibit mendelian inheritance (123). Furthermore, abnormalities of mitochondrial DNA have been identified that exhibit autosomal dominant inheritance, which suggests that nuclear genes responsible for the regulation of mitochondrial DNA synthesis are affected (124,125). Finally, because mitochondrial DNA has an exceptionally high mutation rate, many cases of mitochondrial disease occur in the absence of any family history (126).

The diseases of oxidative phosphorylation are clinically diverse, with heterogenous manifestations. There is a wide variety of possible mutations, induding point mutations, deletions, insertions, and multiple sequence anomalies. These genetic alterations may be transmitted in varying copy numbers to different tissues and organs, and the requirements of the tissues and organs themselves for the energy produced by oxidative metabolism may vary. Over time, additional mutations may accumulate in the mitochondria of some tissues but not others (127,128). As a result, it is difficult to establish a useful dassification system for mitochondrial diseases (129,130). Nonetheless, several distinct phenotypes have been identified and are used in describing individual patients with disorders of oxidative phosphorylation (131,132,133).

An example of this is the Keams-Sayre syndrome, which was first described in 1958 (134). The diagnosis is defined by onset before the age of 20, external ophthalmoplegia, ptosis, and pigmentary retinopathy. In addition, a number of other clinical features may be present including heart block, skeletal musde weakness, cerebellar dysfunction, deafness, and elevated cerebrospinal fluid protein (135,136,137). Conduction disturbances in Kearns-Sayre syndrome include left anterior hemiblock, right bundle-branch block, Mobitz type II second-degree atrioventricular block, and complete heart block (138,139). Permanent pacemakers have prolonged survival significantly in Kearns-Sayre syndrome type II. Dilated cardiomyopathy has been reported in Kearns-Sayre syndrome (140,141) and patients have undergone successful cardiac transplantation (142,143).

On skeletal muscle biopsy in Kearns-Sayre syndrome, myocytes stained with the Gomori trichrome stain show a characteristic pattern, with the eosinophilic mitochondria dumped in shaggy clusters around the periphery of the cell. This appearance has given rise to the term ragged red fibers to describe the histology of Kearns-Sayre syndrome and other mitochondrial disorders. Under the electron microscope, the mitochondria themselves are seen to be abnormal; they are
often enlarged, elongated, or fused, with loss of the usual folding pattern of the internal cristae, and they may contain crystalline inclusion bodies sometimes described as having a so-called railway track or parking lot appearance, in reference to their highly regular, blocklike array. Mitochondria in the myocardium have also been shown to be abnormal in patients with cardiac involvement although the inclusion bodies have typically been more amorphous and unlike the rectilinear crystals noted in skeletal muscle.

Arrythmogeneic right ventricular dysplasia (ARVD) is a primary myocardial disorder characterized by fatty infiltration of the right ventricle, progressive loss of myocytes, inflammatory infiltrate, thinning and dilatation of the ventricular wall, and replacement fibrosis. The right ventricle is affected predominantly; however, both interventricular septal and left ventricular involvement have been reported. Usually manifested by ventricular arrhthymias with left bundle configuration, electrocardiographic abnormalities (including an epsilon wave and sudden-death heart failure) are a less-common feature. The fibrofatty infiltration progresses to yield systolic dysfunction of the right ventricle and, in advanced cases, the left ventricle as well. A negative endomyocardial biopsy does not exclude the diagnosis, and consensus diagnostic criteria include endomyocardial biopsy, echocardiography, electrocardiography, and magnetic resonance imaging (MRI). ARVD is a genetic disorder. Eight loci have been mapped and three genes identified. Autosomal dominant transmission is the most common transmission, but an autosomal recessive has been reported (144,145,146,147,148).

The major difficulty with this diagnosis is that cardiomy-opathy is a relatively uncommon event and alcohol abuse is a common condition. Clinically and histologically, the overt response to general medical therapy is not significantly different in these patients than in those with idiopathic cardiomyopathy. Alcohol abuse has been identified as a significant risk factor in cardiomyopathy with a frequency of between 20% and 30% (149,150).

It is clear that high blood levels of alcohol produce mildly negative inotropism (151,152). Furthermore, several animal studies have investigated the effects of prolonged alcohol administration. In primates, long-term alcohol administration results in myocytolysis and fibrosis (153). In dogs, cardiac fibrosis is produced, leading mainly to measurable myocardial dysfunction (154,155). In a hamster model, long-term alcohol administration resulted in a reduction in myocardial high-energy phosphates (156).

The specific mechanisms by which alcohol might produce myocardial dysfunction have also been investigated. Cellular events that have been described with long-term alcohol exposure include impaired sarcoplasmic reticular uptake of calcium (157), inhibition of myosin ATPase (158), elevation of intracellular Na⁺ and water (159), inhibition of the Na⁺,K⁺-ATPase (160), and alterations in the incorporation of membrane fatty acid and phospholipids (155,161).

The expression of alcohol-induced cardiomyopathy is obviously quite variable. There is evidence to suggest that excessive alcohol consumption is likely to be present for a minimum of 10 years before the onset of heart failure (162,163,164), that men are more susceptible than women (165), and that concurrent smoking, hypertension, and malnutrition may be contributing factors. Furthermore, there appear to be differences between persons in organ susceptibility to alcohol, as chronic cirrhosis is unlikely to develop in those with cardiomyopathy (166).

An interesting, likely unrelated phenomenon occurred in the 1960s when cobalt was added to beer as a foam-stabilizing agent in a few areas of Europe and Canada. An acute and aggressive form of cardiomyopathy developed in heavy beer drinkers that was attributable to the cobalt (167,168). When this element was no longer used, no further cases of this form of cardiomyopathy were reported.

There is some evidence to suggest that abstaining from alcohol has a beneficial effect on prognosis in alcohol-induced cardiomyopathy. In one study of 64 patients, approximately one-third discontinued excessive alcohol use. This subgroup had a 9% mortality rate during the subsequent 4 years, contrasted with a 57% mortality rate in the remainder who continued drinking (31). A second but smaller study (31 patients) supported these findings (169).

Long-term cocaine abuse may result in the development of a dilated cardiomyopathy. Termed cocaine-induced cardiomyopathy, cocaine’s direct toxicity to the myocardium may be responsible. These direct effects include provocation of coronary spasm, increased catecholamine release, enhanced platelet aggregation, decreased fibrinolysis, and upregulation of tissue plasminogen activator inhibitors (predisposing the abuser to coronary thrombosis) (170,171,172). Hypersensitivity myocarditis due to cocaine or associated contaminants may be another mechanism (173). Finally, cocaine may induce lethal arrhythmias. Similar to alcoholic cardiomyopathy, abstinence is crucial in therapy and cessation-related reversal of dysfunction has been reported (174).

Treatment with β-blockade was popular in the 1980s; however, reports of cocaine-induced hypertension or myocardial ischemia were thought to be due to use of β-blockers and the resulting unopposed α-receptors. Thus, treatment with β-blockers decreased. While no significant data exist, utilization of labetalol or carvedilol (both of which have α- and β-receptor antagonist activities) may warrant further investigation (175).