点击显示 收起
1 From the Department of Internal Medicine, Hospital Clínic, Institut d’Investigacions Biomèdiques August Pi i Sunyer, University of Barcelona, Spain.
2 Supported by grants PI020533, PI020535, and G03/140 from the Fondo de Investigaciones Sanitarias and by grant 2001/SGR/00279 from the Generalitat de Catalunya. 3 Address reprint requests to JM Nicolás, Department of Internal Medicine, Hospital Clínic, Villarroel 170, 08036 Barcelona, Spain. E-mail: nicolas{at}medicina.ub.es.
ABSTRACT
Background: Muscle weakness and structural changes in striated skeletal muscle are common in persons with chronic alcoholism.
Objective: The objective of the study was to assess the role of malnutrition in the development of chronic alcoholic myopathy.
Design: We prospectively evaluated 146 men who reported an intake =" BORDER="0">100 g ethanol/d for the previous 5 y and 73 well-nourished control subjects. Alcohol consumption, energy and protein nutritional status, and deltoid muscle strength were determined. Deltoid muscle tissue specimens were taken from alcoholics and from 14 control subjects for histochemical studies and morphometric measurements of the fibers.
Results: Deltoid muscle strength was less in alcoholics than in control subjects (P < 0.001). Muscle strength correlated with lifetime consumption of ethanol (r = -0.56, P < 0.001), and a decrease in muscle strength was significantly greater in the presence of energy malnutrition. Using logistic regression analysis, we observed that alcoholics with muscle strength < 18 kg had the independent risk factors of an arm muscle area < 50 cm2 (odds ratio: 5.4; 95% CI: 2.3, 12.3), consumption of > 1600 kg ethanol throughout their lives (odds ratio: 4.5; 95% CI: 2.0, 10.1), and protein malnutrition (odds ratio: 4.2; 95% CI: 1.4, 12.7). Protein malnutrition was also associated with muscle inefficiency (P < 0.001). Histologic myopathy was present in 58% of alcoholics, was related to lifetime ethanol consumption (P = 0.001), and was more severe in the presence of protein malnutrition (P = 0.01).
Conclusion: Malnutrition is an additional developmental factor in the functional and structural muscle damage induced by chronic ethanol consumption.
Key Words: Nutrition • alcohol intake • skeletal myopathy
INTRODUCTION
A progressive decrease in muscle strength, predominantly involving proximal muscles, has been reported in about one-third of asymptomatic chronic alcoholics (1). Biopsy samples of muscle tissue from chronic alcoholics show type II fiber atrophy, variability in myocyte size, oxidative disturbances, and myocytolysis in 50% of the subjects (2,3). The pathogenesis of this chronic myopathy in alcoholics remains controversial, and it was suggested that it is multifactorial in its development. In the past, the presence of malnutrition in 5–40% of alcoholics (4–6) led to the view that skeletal myopathy is a disease of deficiency (7,8). In fact, malnutrition itself may induce skeletal muscle changes similar to those observed in chronic alcoholic myopathy (9). However, well-nourished chronic alcoholics also show muscle impairment and histologic damage that are related in a dose-dependent manner to ethanol consumption (1,10). Thus, the precise role of malnutrition in modifying alcohol-induced muscle changes has not yet been clarified. The present study was designed to compare the influence of energy and protein malnutrition with that of ethanol intake on the functional and structural involvement of skeletal muscle in chronic alcoholics.
SUBJECTS AND METHODS
Patients and control subjects
Over a 6-y period, we evaluated patients seen in the Alcoholism Unit of Hospital Clínic in Barcelona. This unit helps ambulatory patients who seek assistance in terminating their dependence on alcohol and who do not have other obvious disease. Patients with overt alcohol-related diseases (eg, uncompensated liver cirrhosis, heart failure, or peripheral neuropathy) or disorders other than alcoholism are referred to other clinics. Every Monday, we selected the first male patient aged < 65 y who came to the Alcoholism Unit for assistance and who reported a minimum daily ethanol consumption of 100 g over the previous 5 y. Of 185 alcoholics initially evaluated, 9 were being treated with agents that potentially have toxic effects on skeletal muscle (eg, cholesterol-lowering agents, neuroleptics, and diuretics), 4 reported the use of cocaine, and 3 were found to be HIV-positive; these 16 patients were excluded from the study. Three patients were excluded when the results of the muscle biopsy showed small-group atrophy and small-size angulated esterase-positive fibers, which are suggestive of peripheral neuropathy. Twenty patients refused to participate in the study. A final total of 146 chronic alcoholics were included in the study. None had causes of skeletal muscle damage other than alcoholism, and all patients gave written informed consent to undergo the various procedures.
The control subjects were drawn from a group of 79 healthy men who reported drinking < 20 g ethanol/wk without binge intake and who had no factors causing abnormal skeletal muscle function. They were hospital staff members (physicians, nurses, and technicians) who consecutively visited the medical ward, and all gave written informed consent to participate in the study. Four subjects with energy malnutrition, one subject with protein malnutrition, and one with elevated concentrations of serum creatine kinase were excluded. Finally, this group comprised 73 well-nourished subjects whose median age did not differ significantly from that of the alcoholics and who were studied in the same manner as the alcoholics, except that only 14 of the control subjects agreed to undergo muscle biopsy. The Institutional Review Board of Hospital Clínic approved the study protocol. All participants were white men of Spanish descent who lived with their families in or around Barcelona and had stable employment.
Clinical and nutritional studies
A detailed history of ethanol intake was obtained for each subject by using a structured questionnaire (11), and data were confirmed in consultation with family members, as previously reported (1). Anthropometric assessment of the nutritional status was performed by trained observers who used standardized methods, as previously reported (6). Subjects were considered to have energy malnutrition if their body weight was < 80% of their ideal weight or if the calculated lean body mass was < 47 kg (at least 10% below the control value). Protein malnutrition was diagnosed when the subjects had abnormal values for 3 of the following variables: hemoglobin, lymphocyte count, total protein, albumin, prealbumin, retinol-binding protein, and transferrin (6). Hepatic ultrasonography and percutaneous needle biopsy of liver tissue were performed in all patients who had hepatomegaly on physical examination or abnormally elevated concentrations of serum aminotransferase for > 2 mo after withdrawal of alcohol, or both.
Skeletal muscle studies
Muscle strength was evaluated at the deltoid level of the nondominant arm with the use of an electronic myometer (Penny & Giles, RingWood, United Kingdom) to measure force against resistance 5 times over a period of 20 min (12). Muscle efficiency was defined as the ratio of deltoid strength to the corresponding cross-sectional muscle area of the arm (13). An open-biopsy specimen was taken from the deltoid muscle of the nondominant arm in all alcoholics (n = 146) and 14 controls. The control subjects who underwent biopsy seemed to be representative of the overall control population, because their clinical and nutritional data did not differ significantly from those of the remaining 59 controls. Cryostat sections of deltoid-muscle biopsy specimens were stained with hematoxylin and eosin, modified Gomori one-step trichrome stain, the periodic acid-Schiff stain, Oil Red O (Sigma, St Louis), and nonspecific esterase activity, and stain reacted with ATPase (at pH 4.3, 4.6, and 9.4) and NADH diaphorase. Muscle fibers were typed by means of the ATPase reaction at pH 9.4. The specimens were randomly coded and evaluated by 2 observers with no knowledge of the clinical data. Histologic myopathy was established according to the modified criteria of Mastaglia and Walton and graded semiquantitatively (ie, absent, mild, and moderate-to-severe), as previously reported (14). Intraobserver and interobserver reliabilities were > 97% of cases evaluated, and 2 biopsies with initial minor discrepancies were resolved with agreement. Blind measurement of muscle fibers was performed by manual tracing with the use of the public-domain NIH IMAGE program (NIH, Bethesda, MD; available on the Internet: http://rsb.info.nih.gov/nih-image/).
Statistical analysis
Results are given as means ± SEMs. Differences between groups were evaluated with the use of analysis of variance and Student’s t test. Bonferroni correction was applied to adjust the type I error in multiple comparisons. When analysis of variance showed differences among the subgroups of alcoholics, a test for linear trend was applied. Pearson’s correlation coefficients between the variables were calculated. A multiple linear regression model (forward stepwise selection) was estimated by using those variables that had a significant correlation as independent variables. A logistic regression model was estimated by using the same variables, which were dichotomized on the basis of their median values. All tests were performed with a bilateral significance level of P < 0.05. Statistical analysis was performed with SPSS statistical software (version 10.0; SPSS, Chicago).
RESULTS
Epidemiologic and clinical data
The age range of the alcoholics was 25–64 y (
View this table:
TABLE 1 . Clinical, nutritional, and histologic data for the alcoholics and the control subjects1
Mean deltoid muscle strength was less in the alcoholics than in the control subjects (P < 0.001; Table 1). Deltoid muscle strength correlated (P < 0.001) with arm muscle area in both the alcoholics (r = 0.41) and the control subjects (r = 0.63). In addition, muscle efficiency (ie, muscle strength in relation to the muscle area of the arm) was less (P < 0.001) in the alcoholics than in the control subjects, ie, < 2 SD of the mean values of the control population in 55 (37%) of the alcoholics. Muscle strength did not correlate with age in either the alcoholics or the control subjects.
Histologic and nutritional data
Skeletal myopathy was diagnosed in 85 (58%) of the alcoholics: it was mild in 56, moderate in 26, and severe in the remaining 3, whereas none of the controls had skeletal myopathy (Figure 1). Morphometric evaluation of type II fibers showed a reduction in the cross-sectional area in the overall population of alcoholics compared with that in the controls (P < 0.001; Table 1 and Table 2). Among the subgroups of alcoholics, type II fiber area diminished according to the severity of myopathy (P < 0.001). Type I fibers tended to be thinner in the alcoholics with moderate-to-severe myopathy than in those without myopathy or with mild myopathy (P = 0.06). Deltoid muscle strength was reduced in concordance with the severity of myopathy (P < 0.01; Table 3), but muscle efficiency did not differ significantly among these subgroups of alcoholics.
FIGURE 1. . Biopsy specimens of the deltoid muscle from alcoholic patients (left) and nonalcoholic control subjects (right), which were studied histochemically. ATPase (pH 9.4) staining was used to identify type I and type II fibers, which appear light and dark, respectively. Measurement of the diameter and the cross-sectional area of the fibers was performed by using the NIH IMAGE software. This alcoholic patient had moderate type II fiber atrophy.
View this table:
TABLE 2 . Histologic data for the chronic alcoholics according to the degree of skeletal myopathy and for the control subjects1
View this table:
TABLE 3 . Epidemiologic and anthropometric data for the chronic alcoholics according to the degree of skeletal myopathy and for the control subjects1
Nutritional anthropometric data from the alcoholic and control populations are given in Table 1. Alcoholics had lower values than did well-nourished controls, with lower (P < 0.001) values for triceps skinfold thickness and cross-sectional muscle area of the arm. In addition, chronic alcoholics had lower plasma concentrations of total protein, albumin, and prealbumin (P < 0.01). As a result, 41 (28%) of the alcoholics had energy malnutrition, whereas 23 (16%) were deemed to have protein malnutrition. Nine of the alcoholics had mixed energy and protein malnutrition, and no relation was observed for the occurrence of the two forms of malnutrition. Energy and protein nutritional measurements were worse in the alcoholics, in parallel to their degree of histologic myopathy, than in the control subjects, although statistical significance was achieved only with respect to the arm muscle area (P < 0.001; Table 3) and the concentrations of prealbumin, retinol-binding protein, and transferrin (P < 0.05; Table 4).
View this table:
TABLE 4 . Laboratory and nutritional data for the chronic alcoholics according to the degree of skeletal myopathy and for the control subjects1
Biopsy was performed on liver tissue from 55 alcoholics who had elevated enzyme concentrations for 2 mo, abnormal results on ultrasonography, or both. The histologic examination found cirrhosis in 32 specimens, fatty liver in 12, alcoholic hepatitis in 6, and normal tissue in 5. Among those subjects with cirrhosis of the liver, 11 had antibodies to hepatitis C virus and 4 had antibodies to hepatitis B virus. None of these subjects presented with signs or symptoms of liver failure, and all were ranked as grade A in the Child-Pugh classification (15). Liver cirrhosis was more frequent in the alcoholics with moderate-to-severe histologic myopathy than in the other subgroups of alcoholics (P < 0.001; Table 4).
Influence of ethanol consumption
Muscle strength inversely correlated (P < 0.001) with the crude total (kg ethanol) and weight-proportionate (kg ethanol/kg body wt) lifetime amounts of ethanol consumed (r = -0.56 and -0.51, respectively). In addition, daily ethanol intake correlated with muscle efficiency in patients without skeletal myopathy (r = -0.36, P = 0.01). The reported crude total lifetime ethanol consumption paralleled the severity of histologic myopathy (P < 0.001; Table 3) and correlated with the area of the type II fibers (r = -0.33, P = 0.001; Figure 2).
FIGURE 2. . Cross-sectional area of type II fibers from chronic alcoholics correlated with the lifetime ethanol consumption (, solid line; r = -0.33, P = 0.001) (Pearson’s correlation coefficient). The values for the 14 tissue samples from control subjects () are shown at the right.
Influence of nutritional status
Alcoholics with either energy or protein malnutrition, despite a lack of difference in crude total ethanol consumption, had significantly (P < 0.001) less muscle strength than did well-nourished alcoholics (Table 5). Weakness was related to a decrease in muscle area and to muscle inefficiency in those with energy malnutrition and protein malnutrition, respectively (Table 5). It is interesting that the decrease in muscle strength with ethanol consumption was greater in energy-undernourished alcoholics than in well-nourished alcoholics (P = 0.001; Figure 3). On evaluating the influence of nutritional status on histologic myopathy, we observed that alcoholics with protein malnutrition were more prone (P = 0.03) to have moderate-to-severe skeletal myopathy (34%) than were protein-nourished alcoholics (17%), whereas those with energy malnutrition had a reduction (P = 0.04) in type II fiber area (Table 5). Even when only well-nourished alcoholics were analyzed (n = 91), their deltoid strength was less than that in well-nourished control subjects (P < 0.001).
View this table:
TABLE 5 . Epidemiologic and histopathologic data for the alcoholics in relation to their nutritional status
FIGURE 3. . Deltoid muscle strength in chronic alcoholics correlated with the lifetime ethanol consumption. The decrease in muscle strength in relation to ethanol consumption was significantly steeper in those who were energy undernourished (•, dashed line; r = -0.65, P < 0.001) than in those who were well nourished (, solid line; r = -0.31, P = 0.01) with slopes of 617 ± 116 and 178 ± 71, respectively (P = 0.001) (Pearson’s correlation coefficient and multiple linear regression analysis). The values for the control subjects () are shown at the right.
It is widely known that liver cirrhosis is a common finding in alcoholics, and it may be a confounding factor in evaluations of the influence of nutrition on disease in this population. In the present study, liver cirrhosis was more prevalent (P = 0.03) in the alcoholics with protein malnutrition than in those without protein malnutrition (39% compared with 18%; Table 5). Moreover, the alcoholics with liver cirrhosis had less (P = 0.002) muscle strength (16.4 ± 0.6 compared with 19.0 ± 0.4 kg) and a greater (P < 0.001) prevalence of histologic skeletal myopathy (28 of 32 subjects) than did the alcoholics without cirrhosis. However, there was no significant difference in the energy and protein nutritional status between the alcoholics with myopathy and liver cirrhosis and those with myopathy but without liver cirrhosis.
In the multivariate analysis, we observed that the muscle area of the arm (P < 0.001), the presence of protein malnutrition (P < 0.001), and the crude total lifetime ethanol consumption (P = 0.001) remained as independent variables influencing muscle strength in alcoholics (Table 6). By logistic regression analysis, we estimated that, in our sample, an arm muscle area < 50 cm2 increased the risk for deltoid strength < 18 kg (odds ratio: 5.4; 95% CI: 2.3, 12.3) after adjustment for reported ethanol consumption and protein malnutrition; a reported consumption of > 1600 kg ethanol throughout the subjects’ lives increased the risk for deltoid strength < 18 kg (odds ratio: 4.5; 95% CI: 2.0, 10.1) after adjustment for arm muscle area and protein malnutrition; and the presence of protein malnutrition increased the risk for deltoid strength < 18 kg (odds ratio: 4.2; 95% CI: 1.4, 12.7) after adjustment for arm muscle area and reported ethanol consumption. Protein malnutrition was associated (P < 0.001) with muscle inefficiency in the overall population of alcoholics, although the lifetime dose of ethanol was also related (P < 0.001) to muscle inefficiency when the regression analysis was restricted to well-nourished alcoholics. The diagnosis of histologic myopathy was related (P = 0.001) to the crude total lifetime ethanol consumption, with a threshold at 800 kg ethanol, and its severity was associated (P = 0.01) with the presence of protein malnutrition.
View this table:
TABLE 6 . Regression models of multivariate analyses for evaluation of the variables influencing deltoid muscle weakness in the chronic alcoholics1
DISCUSSION
The present study showed that malnutrition is an additional deleterious factor involved in the pathogenesis of functional impairment and structural muscle damage induced by chronic alcohol consumption. Energy malnutrition was present in 28% of our population of alcoholics and was found, by a decrease in arm muscle area, to increase the deltoid weakness that occurs in the setting of chronic ethanol intake. Protein malnutrition, seen in 16% of the cases, was an independent factor related to low muscle strength, a decrease in muscle efficiency, and the severity of histologic muscle damage.
Deltoid muscle strength correlated with the cross-sectional muscle area of the arm in alcoholics and control subjects. Both muscle strength as an absolute value and muscle efficiency as the ratio of deltoid muscle strength to the corresponding cross-sectional area of the arm were poorer in alcoholics than in control subjects. In previous studies evaluating well-nourished chronic alcoholics, we found that the decrease in muscle strength is consonant with the weight-proportionate lifetime consumption of ethanol (1,10,12). Therefore, those studies corrected to some degree for the energy nutrition of the patients. However, when evaluating ethanol intake in the present study, we found crude total ethanol consumption (kg ethanol) to be clearly related to the degree of muscle weakness. Because reports of the quantity of ethanol consumed by alcohol abusers are known to be somewhat unreliable, we assumed that alcohol intake data might not be totally accurate. However, because, in our experience, alcoholics tend not to exaggerate (and often underreport) the amount that they continue to drink, we considered the reports to be minimal estimates.
Muscle structural damage was observed in 58% of the chronic alcoholics in this study, a prevalence that does not differ significantly from that in previous studies (2,16,17). Nevertheless, the present study is limited to biopsies of deep deltoid muscle, which is a representative proximal muscle, with 50% of fibers being type I. Therefore, results may not be totally applicable to other distal muscle tissue with predominance of a different fiber. Type II fiber atrophy was the most relevant finding in muscle biopsies, even in some alcoholics without histologic criteria of myopathy. Type I fiber atrophy was evident only in advanced cases of skeletal myopathy. The diagnosis of histologic myopathy was related to the crude total lifetime amount of ethanol consumed, whereas protein malnutrition seemed to amplify the severity of structural muscle damage. As previously observed (18), alcoholics with liver cirrhosis had significantly less muscle strength and a greater prevalence of histologic myopathy than did alcoholics without liver cirrhosis. These findings were due to the greater ethanol consumption in alcoholics with liver cirrhosis and to a greater prevalence of energy and protein malnutrition. Although peripheral nerve studies were not performed in the present study, we did not include patients with clinical data or a histologic pattern suggestive of neuropathy. In addition, the atrophic pattern noted in our patients does not resemble that seen in neuropathic wasting, and hence it may be considered a predominantly myopathic pattern of wasting.
The incidence of malnutrition—whether of energy, protein, or mixed origin—in chronic alcoholics ranges from 5% to 40% (6). Potential conditions leading to malnutrition in these subjects are diverse, and they include deficient food intake (19), malabsorption (20), increased protein turnover (21,22), liver disease (23), and disruption of energy expenditure (24). Malnutrition has long been suspected to be an additional deleterious factor in the occurrence of alcohol-induced organ damage—specifically, alcohol-induced liver disease (23), central and peripheral neurologic damage (25,26), pancreatitis (27), and infectious complications (28). In evaluating skeletal muscle, Andersen et al (29) observed that muscle strength was substantially weaker in alcoholic patients with advanced liver cirrhosis than in control subjects, and that the weakness was related to the severity of malnutrition but not to the severity of liver disease, duration of alcohol abstinence, or neuropathy. Likewise, in a homogeneous population of asymptomatic chronic alcoholics, we found that either energy or protein malnutrition resulted in a deleterious effect on muscle power in addition to that of ethanol consumption. By decreasing the muscle area of the arm, energy malnutrition led to a decrease in deltoid muscle power. With respect to protein malnutrition, the deleterious effect was mainly due to muscle inefficiency and an increase in histologic myopathy. As we reported previously (12), no differences in muscle strength between the alcoholics and the healthy control subjects could be attributed to age or to the degree of physical activity. Nevertheless, the present study was conducted by comparing groups with homogeneous characteristics, not as a case-control study.
The deleterious influence of malnutrition, due to poor energy intake, on muscle performance has been recognized in persons with anorexia nervosa (9), the elderly (30), and critically ill patients (31). The effects of food intake on nutritional imbalance, and therefore on muscle damage, were not considered in the present study. However, we previously observed that the nutritional status of alcoholics was directly related to energy intake and inversely related to the amount of ethanol consumed (6). Then again, other nutritional factors, such as micronutrient shortages that may contribute to muscle weakness and potential atrophy, were not measured, and this should be noted as a limitation of the present study. However, in previous studies with similar alcohol populations, we observed no significant differences in serum and muscle vitamin concentrations between alcoholics with and without myopathy (32).
In conclusion, this study corroborates a dose-dependent effect of ethanol consumption on the development of skeletal myopathy in chronic alcoholics. This toxic effect is independent of the presence of malnutrition, but, if malnutrition is present, the effect is increased. Energy malnutrition and protein malnutrition were found to further increase the muscle weakness that is due to excessive ethanol consumption, and protein malnutrition also increased the severity of histologic myopathy. These findings show the need to detect and avoid malnutrition in the setting of chronic alcoholism to decrease malnutrition’s noxious influence on the development of alcohol-induced skeletal muscle damage. Further studies should evaluate the role of nutritional supplementation in reducing the functional and structural effects of ethanol on skeletal muscle.
ACKNOWLEDGMENTS
JMN contributed to the study design, the data analysis, and the writing of the manuscript; GG contributed to the morphometric analysis of muscle biopsies; FF evaluated the subjects’ nutritional status; ES recruited subjects; ET performed histologic processing of muscle biopsies; EB collected the data; RE recruited the alcoholic subjects and performed histologic assessment; and JF-S performed the surgical procedure for the deltoid muscle biopsies and the histologic assessment and contributed to the writing of the manuscript. None of the authors had any financial or personal conflict of interest with regard to any company or organization sponsoring the research, including advisory board affiliations.
REFERENCES