Methods

Top Methods Statistical Analysis Results Discussion Author & Article Info References

Patient Selection

Group A

Twenty-four men with chronic alcoholism who had dilated cardiomyopathy were included in group A. During a period of 1 year, 30 male patients with chronic alcoholism were admitted to the Hospital Clinic of Barcelona because of heart disease. All had been drinking more than 100 g of ethanol daily in the previous 2 years and had maintained this intake until the day before admission. They fulfilled the diagnostic criteria for alcoholism as defined by the Diagnostic and Statistical Manual (DSM-III-R) of the American Psychiatric Association [17]. One patient had severe hypertension, 1 had rheumatic mitral valve disease, and 4 had evidence of ischemic heart disease. These 6 patients were excluded from the study. The remaining 24 patients did not present with other causes of cardiomyopathy except for their alcoholism.

Of these 24 patients, 14 presented to the emergency department because of left heart failure. The other 10 patients were selected from among 192 alcoholic patients who presented consecutively to the Alcoholism Unit of Hospital Clinic during a period of 6 months because they wanted to rid themselves of alcohol dependence. After careful questioning, they admitted having dyspnea on exertion, palpitations, or other cardiac symptoms and were subsequently admitted to the hospital for study.

Group B

Of the 192 consecutive patients seen in the Alcoholism Unit, 154 had maintained their ethanol intake until the day before the visit, and of these, 100 were randomly assigned to receive radionuclide cardiac angiography. Those patients who had normal left ventricular ejection fractions ( 65%) and who did not complain of any cardiac symptoms were consecutively selected for the study (a total of 24 patients).

Group C

During a period of 6 months, 17 nonalcoholic patients with dilated cardiomyopathy because of coronary heart disease (which was documented by coronary angiography) were selected for study. All were younger than 60 years of age and were admitted to the hospital because of heart failure. Although they were clearly not alcoholic patients, 3 patients were excluded because they reported consuming more than 20 g of ethanol daily during the last 2 years. Two additional patients were removed from the study because they appeared malnourished (body weight less than 90% of ideal weight).

Group D

We identified 12 patients with idiopathic dilated cardiomyopathy [4] who did not report any ethanol intake. All had presented to the emergency department, during a 6-month period because of left heart failure.

Group E

Twenty-four normal male volunteers who were employed at the hospital and who did not drink more than 20 g of ethanol daily were included in group E. They were matched for age and sex with group A.

Group F

We studied the hearts of five young men who died in traffic accidents; their families reported the young men had no cardiac complaints and had a daily ethanol intake less than 20 g. The mean age of these patients was 36.6 ±7.0 years.

All patients were white men of Spanish origin. They lived with their families in or around Barcelona, and most had histories of stable employment. None were indigent. The study protocol was approved by the Institutional Review Board of the Hospital Clinic of Barcelona. No patient objected to being in the study, and all gave informed consent for the various procedures.

Clinical Data

A detailed history of ethanol intake, dietary habits, and cardiac and neuromuscular symptoms was elicited. These data were obtained by one physician, using structured forms, and were confirmed with family members. The type of alcoholic beverage, the average quantity, the frequency of ethanol intake, and the average amount consumed were recorded. The current daily ethanol intake was considered to be the average amount of ethanol consumed per day during the last month. Life events such as marriage, military service, and work posts were used as "anchor points" to assist in recollection (time-line follow-back method) [18]. The total lifetime dose of ethanol was estimated by multiplying the amount of ethanol consumed per day by the number of years of each alcohol intake period multiplied by 365 and by adding the total amounts ingested during these different periods. The cardiac status of the patients was classified according to the New York Heart Association criteria [19].

Laboratory and Nutritional Studies

Within 24 hours of admission to the study, the following laboratory analyses of the blood were done: hematocrit; hemoglobin levels; counts of erythrocytes, leukocytes, and platelets; prothrombin time; total protein levels; albumin levels; prealbumin levels; retinol-binding protein; glucose levels; creatinine levels; electrolyte levels; aspartate and alanine aminotransferase levels; -glutamyl transpeptidase levels; lactic dehydrogenase levels; creatine kinase levels; and aldolase levels. Ethanol levels in blood and urine were also determined.

Nutritional status was assessed according to the proportion of actual weight to ideal weight; the lean body mass and the muscular and fatty areas of the arm were assessed by means of previously described methods [20, 21]. Nutritional protein levels were estimated according to the values obtained for total lymphocytes, total protein, albumin, prealbumin, and retinol-binding protein. Patients were considered malnourished if their weight was less than 90% of the ideal weight or if the calculated fat-free body mass was more than 10% below the normal value.

Cardiac Studies

A chest radiograph, with assessment of the cardiothoracic ratio, and a conventional electrocardiogram were obtained. Monodimensional and bidimensional echocardiography was done using a Toshiba SS 10 instrument, with assessment of end-diastolic and end-systolic diameters, of the shortening fraction and the thickness of the septum, and of the posterior wall of the left ventricle. The mass of the left ventricle was also calculated [22]. The end-diastolic and end-systolic diameters and the mass of the left ventricle were indexed by dividing by the body surface area [23, 24]. All these measurements were done according to the recommended standards of the American Society of Echocardiography [25]. The left ventricular ejection fraction was also measured by technetium-99 radionuclide angiocardiography, using a Picker-Dyna 4-15 -camera. In this facility, the variation in the measurement of the ejection fraction in the same patient by radionuclide angiography is 2% and by echocardiography is 3%.

Left and right cardiac catheterization, with coronary angiography, contrast ventriculography, and endomyocardial biopsy of the left ventricle, was done in all patients with chronic alcoholism who had an ejection fraction less than 50% (group A). This is standard procedure in the Hospital Clinic in all patients in whom there is reason to suspect a cardiomyopathic abnormality. Three to five biopsy specimens were obtained from the left ventricle (using Cordis biopsy forceps) and were embedded in methacrylate resin. Sections were stained with toluidine blue, Congo red, malachite green, and Perl stain for iron and were examined by light microscopy. The specimens were coded to prevent the identification of specific patients. Myocardial hypertrophy, myocytolysis (defined by the presence of myofiber disarray or cell vacuolization), and fibrosis were evaluated by two independent observers. Interobserver disagreement occurred in only two specimens and was resolved by discussion before the code was broken.

The amounts of interstitial fibrosis (volume fraction of fibrosis) and of cardiac muscle cells (volume fraction of the myocytes) were determined morphometrically by means of a point counting test, according to Weibel [26]. The size of the grid used for point counting was 24 x 13 cm, comprising 1248 points. Four µgraphs, taken one from each corner of each specimen (final magnification, x 400), were examined. Each µgraph included 20% to 25% of the surface of the section, without overlap between the µgraphs. The volume fractions of fibrosis and myocytes contained in the unit volume of myocardium equal the number of points falling on profiles of connective or myocardial tissue, respectively, divided by the total number of test points [26]. To assess the degree of cardiac hypertrophy, the mean fiber area and the transverse diameter were measured in each µgraph using a Hewlett-Packard digitized 9111A graphics table. The fiber area was assessed by measuring the maximum transverse diameter and the perimeter of each fiber. The mean volume fractions of fibrosis and myocytes, the mean fiber area, and the mean fiber diameter were calculated by averaging the data from the µgraphs of each patient. Three to five endomyocardial specimens were obtained with a Cordis forceps from five young men who died in traffic accidents (group F). These specimens were studied in the same manner as those from the alcoholic patients with dilated cardiomyopathy.

Studies of Skeletal Muscle

The day before discharge, the strength of the deltoid muscle of the nondominant arm was measured five times in 20 minutes with an electronic myometer (Penny and Giles) that measures muscular force against a fixed resistance. Repeated measurements in the same participant show a variation of 4%. In accordance with previous studies [16], muscle weakness was defined as an inability to exert a force greater than 196 newtons (20 kg), a value that was 2 SDs below the mean of the control group [16]. Conventional electrophysiologic examinations were done in all patients to rule out a peripheral neuropathic abnormality. An open biopsy of the deltoid muscle of the nondominant arm was taken from all patients with chronic alcoholism (groups A and B) and from 10 normal participants (group E). These specimens were studied by light microscopy.

Cryostat sections of muscle were examined for adenosine triphosphatase at pH 9.4, for the reduced form of nicotinamide-adenine dinucleotide diaphorase, and for nonspecific esterase activity. The specimens were coded and examined in the same way as the myocardial specimens. A diagnosis of myopathy was made according to the criteria proposed by Mastaglia and Walton [27] and Dubowitz and Brooke [28]. Muscle changes were evaluated on a scale of 0 to 3+ (absent, mild, moderate, severe). Interobserver disagreement between a classification of mild or moderate myopathy occurred in five specimens, whereas in three specimens, the differences were between a classification of normal or mild myopathy. The classification for these specimens was resolved by discussion before the code was broken.

Statistical Analysis

Top Methods Statistical Analysis Results Discussion Author & Article Info References

The data were analyzed using a statistical software package [29]. Differences between groups were analyzed using analysis of variance, the two-tailed t-test (unpaired), and the Fisher test. Correlation studies were obtained by the Pearson correlation coefficient and by regression analysis. All variables are expressed as mean ±SD.

Results

Top Methods Statistical Analysis Results Discussion Author & Article Info References

Clinical Data

Characteristics of all patients and controls are summarized in Table 1.

Group A

The age of the patients in group A ranged from 30 to 59 years (mean, 44.5 ±8 years), and their reported daily intake of alcohol varied from 112 to 380 g (mean, 200 ±62 g) for a period of 10 to 42 years (mean, 24.1 ±7.3 years). In this respect, 10 g of ethanol is equivalent to about 1 ounce (30 mL) of 86 proof (43%) spirits or 3.5 ounces (105 mL) of 12% wine. Ethanol was consumed mainly as wine, beer, and brandy and less frequently as anisette or gin. All but three patients smoked 1 to 2 packs of cigarettes daily. No other drugs were used. Of the 24 alcoholic patients included in group A, 10 were diagnosed as New York Heart Association functional class I or II and 14 were diagnosed as class III or IV. Ten class III or IV patients also presented with clinical signs of right heart failure, a symptom that is characteristic of advanced cardiomyopathy. Clinical symptoms became apparent 4 to 6 months before admission.

Group B

The 24 patients in group B were admitted solely because of alcohol dependence and had no signs or symptoms of other diseases. The age of these patients ranged from 26 to 45 years (mean, 34.9 ±5 years), and their mean daily ethanol intake was 251 ±83 g (range, 120 to 390 g) Table 1 for a period of 16.6 ±4.0 years (range, 10 to 25 years). All but four of the patients in group B smoked 1 to 2 packs of cigarettes daily, and none used other drugs.

Group C

These 12 patients with ischemic heart disease, with a mean age of 51.8 ±6.0 years (range, 43 to 59 years), were hospitalized because of left heart failure (New York Heart Association functional class III or IV) for 2.5 ±1.2 years (range, 1 to 4 years). Four of these patients had advanced disease and showed signs of right heart failure. Seven patients from group C smoked 1 to 2 packs of cigarettes daily, and none used other drugs.

Group D

The 12 patients with idiopathic dilated cardiomyopathy were admitted because of heart failure (New York Heart Association functional class III or IV) for 3.4 ±1.2 years (range, 1 to 5 years). The mean age was 51.1 ±9.1 years (range, 27 to 59 years). In this group, five patients also presented with clinical evidence of right heart failure. Six of them smoked 1 to 2 packs of cigarettes daily, and none used other drugs. Patients with dilated cardiomyopathy because of coronary heart disease (group C) and those with idiopathic dilated cardiomyopathy (group D) had a longer period of heart failure than did alcoholic patients with cardiomyopathy (group A) (both P < 0.001) (Table 1).

Laboratory Analyses and Nutritional Status

Except for a mild increase of hepatic enzymes found in 15 patients from group A and in 16 patients from group B, no statistically significant biochemical abnormalities were noted among the patients studied. No abnormal nutritional variables were observed in these patients, except for a slightly smaller thickness of the tricipital skin fold in alcoholic patients from groups A and B. Blood and urine levels of ethanol determined the day after admission were negative in all alcoholic patients.

Cardiac Studies

The results of cardiac functional studies are presented in Table 1. Alcoholic patients with cardiomyopathy and patients with nonalcoholic dilated cardiomyopathy had lower ejection and shortening fractions and had higher end-diastolic indices and left ventricular masses than did alcoholic patients with normal left ventricular ejection fractions and control participants. No statistically significant differences in the echocardiographic indices or in the ejection fractions were observed between patients with nonalcoholic dilated cardiomyopathy (groups C and D) and alcoholic patients with cardiomyopathy (group A).

Myocardial specimens from all group A patients showed evidence of histologic damage, including interstitial fibrosis in all specimens (Figure 1). Heart biopsy specimens from 20 patients (83%) had some degree of myocytolysis and 23 (93%) had myofiber and nuclear hypertrophy. Endocardial thickening was found in 14 of 18 patients (77%). The myocardial specimens from group F were entirely normal.

View larger version (122K): [in this window] [in a new window]   Figure 1. Endomyocardial biopsy specimen of the left ventricle from an alcoholic patient with dilated cardiomyopathy. The cardiac myocytes show atrophy (arrows) and hypertrophy, and prominent fibrosis is evident. (Hematoxylin and eosin; original magnification, x 200).

Using morphometric analysis, myocardial specimens from alcoholic patients with cardiomyopathy (group A) showed a higher mean volume fraction of fibrosis than did the specimens from control participants in group F (P < 0.01) (Table 2). The volume fraction of the myocytes was also less than in control participants (P < 0.01). Although alcoholic patients with cardiomyopathy had a lower fraction of myocytes than did control participants, their mean myofiber areas and diameters were higher (P < 0.05). These findings are consistent with a loss of myocardial fibers and with hypertrophy of the remaining cells.

Studies of Skeletal Muscle

We defined clinical weakness as a progressive impairment of muscle strength that interfered with doing daily tasks. On specific questioning, 13 alcoholic patients from group A reported proximal muscle weakness for the last 6 months compared with only 1 alcoholic patient in group B (P < 0.001), 2 patients in group C (P < 0.03), and none in group D (P < 0.001). Acute episodes of weakness related to binges of ethanol intake were reported by 3 of the alcoholic patients from group A and 1 of the alcoholic patients from group B.

Alcoholic patients with dilated cardiomyopathy had less strength than did patients with chronic alcoholism who had normal left ventricular ejection fractions, patients with dilated cardiomyopathy because of coronary heart disease, patients with idiopathic dilated cardiomyopathy, and control participants (analysis of variance, P < 0.001) (Figure 2). Fourteen of the 24 alcoholic patients with dilated cardiomyopathy (group A) were unable to exert a force greater than 196 newtons (20 kg). In contrast, muscle weakness was found in only one alcoholic patient with a normal left ventricular ejection fraction (group B) and in two patients with ischemic cardiomyopathy (group C). None of the patients with idiopathic dilated cardiomyopathy had a force less than 196 newtons. No significant differences were observed between the strength of alcoholic patients with normal left ventricular ejection fractions (group B) and that of patients with ischemic cardiomyopathy (group C) or those with idiopathic dilated cardiomyopathy (group D) (both P > 0.02). However, patients from these three groups (B, C, and D) had less strength than did control participants (group E) (all P < 0.01) (Figure 2). None of the patients had electrophysiologic signs of peripheral neuropathy.

View larger version (17K): [in this window] [in a new window]   Figure 2. Measurements of muscular strength in five groups of patients. The groups included patients with dilated cardiomyopathy (group A), patients with chronic alcoholism who had normal left ventricle ejection fractions (group B), patients with dilated cardiomyopathy because of coronary heart disease (group C), patients with idiopathic dilated cardiomyopathy (group D), and normal participants (group E). Muscle strength was measured in the nondominant deltoid muscle (see Methods). The statistical differences between the groups were as follows: group A compared with groups B, C, D, and E (P < 0.001, P < 0.007, P < 0.002, and P < 0.001, respectively); group B compared with groups C, D, and E (P > 0.2, P > 0.2, P < 0.01, respectively); group C compared with groups D and E (P > 0.2 and P < 0.01); and group D compared with group E (P < 0.005).

Histologic changes characteristic of myopathy were found in muscle biopsy specimens of 20 of the 24 patients with alcoholic cardiomyopathy (group A) compared with only 1 of the 24 patients with chronic alcoholism who had normal left ventricular ejection fractions (P < 0.001). In alcoholic patients from group A, myopathy was classified as mild in 10, moderate in 8 Figure 3, and severe in 2. The deltoid muscle appeared histologically normal in the remaining 4 patients. The severity of the myopathy was not related to the age of the patients. In alcoholic patients with normal ejection fractions (group B), mild myopathy was discovered in one patient. The muscle specimens from group E (control participants) appeared normal.

View larger version (146K): [in this window] [in a new window]   Figure 3. Section of deltoid muscle from an alcoholic patient with moderate skeletal myopathy who was admitted to the hospital because of heart failure. An area of focal necrosis of the muscle is shown in the center.

Relation between Myopathy and Cardiomyopathy

In patients with chronic alcoholism who had cardiomyopathy (group A), a statistically significant relation between cardiomyopathy and myopathy was observed. Patients who were New York Heart Association functional class III or IV complained more frequently of clinical weakness than did those who were class I or II (P = 0.024). In addition, a statistically significant correlation between New York Heart Association functional class and muscle strength was observed. Thus, 11 of the 14 patients with chronic alcoholism who presented with functional class III or IV showed muscle weakness (muscle strength <196 newtons) compared with only 3 of the 10 alcoholic patients who presented with class I or II (P = 0.028). Conversely, patients with moderate and severe myopathy had greater myocardial dysfunction, as measured by echocardiography and radionuclide cardiac angiography (Table 3). Compared with alcoholic patients who had histologically normal muscle or mild skeletal myopathy, patients with more severe muscle changes (2+ or 3+) showed a higher end-diastolic index (P = 0.024), end-systolic index (P = 0.016), and left ventricular mass (P = 0.048), as well as lower shortening and ejection fractions (P = 0.048 and 0.043, respectively).

No correlation between muscular strength and the ejection fraction was observed in patients with ischemic cardiomyopathy because of coronary heart disease (group C) (r = 0.24, P > 0.2) or in those with idiopathic dilated cardiomyopathy (group D) (r = 0.11, P > 0.2). However, in alcoholic patients with dilated cardiomyopathy (group A), a trend toward a correlation between muscular strength and ejection fraction was observed, although it was not statistically significant (r = 0.34; P = 0.1).

Patients with New York Heart Association functional class III or IV had greater degrees of cardiac myocyte hypertrophy (P = 0.008), myocytolysis (P = 0.02), and interstitial fibrosis (P < 0.05) than did those who were class I or II. Conversely, patients with more extensive myocytolysis, cellular hypertrophy, and interstitial fibrosis of the myocardium showed a greater decrease in deltoid muscle strength (all histologic parameters; P = 0.028). Using histologic data, patients with more conspicuous myopathic damage had more severe myocardial injury (see Table 2). Further, alcoholic patients with more severe grades of skeletal myopathy had a lower mean volume fraction of myocytes and a higher mean volume fraction of fibrosis than did those with normal histologic findings in skeletal muscle or mild skeletal myopathy (see Table 2). Patients with more severe grades of skeletal myopathy also had larger mean fiber areas and larger diameters of their cardiac myocytes compared with alcoholic patients with lesser grades of myopathy, although this relation did not attain statistical significance.

Discussion

Top Methods Statistical Analysis Results Discussion Author & Article Info References

In this study of patients with alcoholic cardiomyopathy, we found a relation between cardiac dysfunction and myopathy at the clinical level. Patients with more severe cardiac disease, according to New York Heart Association criteria, complained more frequently of clinical weakness and had less deltoid muscle strength. Conversely, patients with more severe grades of histologic myopathy had greater cardiac dysfunction, as assessed by echocardiography and radionuclide angiography. A statistically significant correlation was also observed between the histologic changes in cardiac and skeletal muscle. Patients with more prominent damage in the myocardial biopsy specimens showed more conspicuous histologic changes in skeletal muscle. Among patients with chronic alcoholism with normal left ventricular ejection fractions, only one (4%) had a muscular force less than 196 newtons and had histologic evidence of skeletal myopathy, whereas the prevalence of functional and histologic myopathy in an unselected group of patients with chronic alcoholism was 42% and 46%, respectively [16].

Chronic ethanol abuse has a dose-related effect on skeletal and cardiac striated muscle, independent of the presence of malnutrition or of vitamin deficiencies [16]. Thus, a correlation between cardiac damage and skeletal muscle injury is not unexpected among patients with chronic alcoholism. Even alcoholic patients with normal ejection fractions (group B) had less muscular strength than did control participants (group E). This finding is consistent with the observation that in alcoholic patients, evidence of skeletal myopathy tends to precede that of cardiomyopathy, and the threshold dose for the former is less than that for the latter [30]. Further evidence for the deleterious effect of alcohol abuse on striated muscle is the demonstration that as many as one third of asymptomatic patients with chronic alcoholism have cardiac dysfunction related to cardiomyopathy [16] compared with a prevalence of idiopathic dilated cardiomyopathy of about 1 in 3000 in the general population [31].

In patients with congestive heart failure, the reported poor correlation between exercise performance and cardiac function [32-34] raises the possibility that skeletal muscle function may be impaired and may limit exercise performance. Recent studies have also shown that patients with congestive heart failure have abnormalities in the structure and biochemical properties of skeletal muscle [35] and that during exercise, peripheral blood flow [36] and skeletal muscle metabolism [37, 38] are abnormal. Consistent with these reports, patients with nonalcoholic dilated cardiomyopathy and heart failure in the current study had less muscle strength than a control group. However, despite the fact that patients with nonalcoholic dilated cardiomyopathy were older, had lower left ventricular ejection fractions, and had longer periods of left heart failure, alcoholic patients with dilated cardiomyopathy were still substantially weaker. Muscle changes detected in alcoholic patients with dilated cardiomyopathy should, therefore, not be attributed simply to left heart failure. Although the pathogenesis of muscle damage mediated by ethanol has not been defined [39, 40], the correlation between cardiac and skeletal muscle damage reinforces the hypothesis that ethanol exerts similar toxic effects on both types of striated muscle. However, not all persons who consume excessive amounts of alcohol develop cardiomyopathy or myopathy, and other environmental and genetic [41] factors may also play a role.

Because the high incidence of sudden death among alcoholic patients [6, 42-46] could be related to a subclinical cardiomyopathy, it would be useful to develop new methods to detect alcoholic patients at high risk for cardiomyopathy. Because it appears that skeletal muscle weakness may serve as a surrogate for myocardial damage, patients with chronic alcoholism might be screened to detect those with a high risk for subclinical cardiomyopathy.