- Email: [email protected]
Low Serum Selenium Level Is Associated With Low Muscle Mass in the Community-Dwelling Elderly
JAMDA 15 (2014) 807e811
JAMDA journal homepage: www.jamda.com
Original Study
Low Serum Selenium Level Is Associated With Low Muscle Mass in the Community-Dwelling Elderly You-Ling Chen MD a, y, Kuen-Cheh Yang MD, MSc b, Hao-Hsiang Chang MD, MSc a, Long-Teng Lee MD, PhD a, y, Chia-Wen Lu MD, MSc a, Kuo-Chin Huang MD, PhD a, c, d, * a
Department of Family Medicine, National Taiwan University Hospital, Taipei, Taiwan Department of Community and Family Medicine, National Taiwan University Hospital, Hsin-Chu Branch, Hsinchu, Taiwan c Department of Family Medicine, College of Medicine, National Taiwan University, Taipei, Taiwan d Graduate Institute of Clinical Medical Science, China Medical University, Taichung, Taiwan b
a b s t r a c t Keywords: Skeletal muscle mass selenium bioelectrical impedance analysis elderly
Objectives: Elderly persons with low muscle mass (LMM) or sarcopenia are prone to frailty and functional decline. This study aimed to investigate the relationship between serum selenium level and skeletal muscle mass in community-dwelling elderly. Design: Cross-sectional observational study. Setting and participants: A total of 327 elderly Taipei citizens (mean age 71.5 4.7 years) were recruited from the community. Measurements: Skeletal muscle mass was measured by bioelectrical impedance analysis. LMM was defined by low skeletal muscle index (SMI: muscle mass (kg)/[height (m)]2). All participants were further divided into quartiles by serum selenium level and the risk for LMM among these quartiles was examined using multivariate logistic regression analyses. Estimated serum selenium levels for the LMM group vs the normal group and estimated SMI in the quartiles of serum selenium were computed by least square method in linear regression models. Results: The estimated mean (standard deviation) of serum selenium level was significantly lower in the LMM group compared with the normal group after adjusting for confounders (1.01 0.03 mmol/L vs 1.14 0.02 mmol/L, P < .001). After adjusting for age, sex, lifestyle, and physical and metabolic factors, the odds ratios (95% confidence interval, P value) of LMM in the bottom, second, and third selenium quartile groups were 4.62 (95% CI 2.11e10.10, P < .001), 2.30 (95% CI 1.05e5.03, P < .05) and 1.51 (95% CI 0.66 e3.46, P ¼ .327), respectively, compared with the top quartile group of serum selenium level. The least square mean of SMI increased with the quartiles of serum selenium (P < .001). Conclusions: This is the first study to demonstrate that low serum selenium is independently associated with low muscle mass in the elderly. The causality and underlying mechanism between selenium and low muscle mass or sarcopenia warrant further research. Ó 2014 AMDA e The Society for Post-Acute and Long-Term Care Medicine.
Sarcopenia is the gradual loss of muscle mass and strength with aging. Geriatric patients with sarcopenia usually suffer from frailty and disability.1 Previous studies have reported several risk factors for sarcopenia, including genetic factors, nutrition, exercise,
This study was funded by the National Health Institute of Taiwan (PH-101-PP42). The authors declare no conflicts of interest. * Address correspondence to Kuo-Chin Huang, MD, PhD, Department of Family Medicine, National Taiwan University Hospital, No. 7, Chung-Shan South Road, Taipei 100, Taiwan. E-mail address: [email protected] (K.-C. Huang). y Contributed equally to this work.
neuromuscular dysfunction, hormone reactivity, or inflammation.2 In addition, oxidative injuries during aging play a crucial role in muscle loss.3,4 It has been proposed that taking recommended dietary allowances of antioxidants may protect people from abnormal reductions of skeletal muscle mass.5 Selenium is one of the most important antioxidants in humans. The distribution and function of selenoenzymes such as cytosolic glutathione peroxidase and selenoprotein P modulate protection against oxidative damage, and mutations in selenoprotein genes can lead to muscular diseases.6,7 In lambs, selenium deficiency is also associated with white muscle disease characterized by weakness and calcification in skeletal muscles and myocardium.8 One study investigating the association between selenium and muscle function showed an
1525-8610/$ - see front matter Ó 2014 AMDA e The Society for Post-Acute and Long-Term Care Medicine. http://dx.doi.org/10.1016/j.jamda.2014.06.014
808
Y.-L. Chen et al. / JAMDA 15 (2014) 807e811
improvement in muscle strength after selenium supplementation in a patient with long-term parenteral nutrition,9 whereas another study reported no difference between normal persons and those with decreased muscle mass with regard to dietary selenium intake.5 The Invecchiare in Chianti (InCHIANTI) study enrolled randomly selected Italian persons older than 65 years and tested the association between serum selenium level and handgrip, hip flexion, and knee extension muscle strength of the persons.10 Beck et al11 also examined the association of serum selenium level and handgrip strength in elderly female persons. Both studies concluded that a low selenium level was associated with reduced muscle strength.10,11 Despite the attempt to elucidate the role of selenium in muscle function, these studies focused on muscle strength measures instead of direct evaluation of muscle mass. Whether or not selenium influences the maintenance of muscle mass, which provides clinicians with a greater insight into the pathogenic mechanism of sarcopenia, remains unclear. Therefore, we conducted this cross-sectional community study to investigate the relationship of serum selenium and skeletal muscle mass in the community-dwelling elderly. Methods Participants A total of 327 volunteers living in Taipei were recruited in 2007 by advertisements. The inclusion criteria were individuals aged 65 years or older and could stand steadily on the bioelectrical impedance analysis (BIA) system during measurements of bioelectrical impedance. Participants with cancer history, recent body weight change more than 5% in 3 months, and those implanted with cardiac pacemakers or metal implants were excluded. All participants provided written informed consent, and all protocols were approved by the Ethics Committee of the National Taiwan University Hospital. Data on age, sex, history of hypertension, diabetes mellitus, cardiovascular diseases, cigarette smoking, exercise habits, nutritional status, and functional status were obtained by individual interviews. Current and noncurrent smokers were defined according to their recent smoking habit.12 Exercise status was defined as whether the subject had exercise habit at least once a week. The nutritional status of the participants was evaluated by mini-nutritional assessment (MNA).13,14 Functional status was evaluated by Barthel index and Lawton-Brody instrumental activities of daily living scale.15,16 Height and weight were measured to obtain body mass index (BMI).12 Waist circumferences, and systolic and diastolic blood pressure were recorded for evaluating status of metabolic syndrome (MetS). Disease histories such as diabetes mellitus, hypertension, and hyperlipidemia were evaluated, and current prescriptions including antihypertensive, antihyperglycemic, or antihyperlipidemia agents were recorded based on self-reported history. MetS was diagnosed according to the America Heart Association and National Heart Lung and Blood Institute criteria modified for Taiwan population.17 Skeletal Muscle Mass Estimation and Low Muscle Mass Definition Skeletal muscle mass was estimated by bioelectrical impedance analysis.18 Skeletal muscle mass index (SMI) was calculated by dividing the skeletal muscle mass with the square of body height (kg/m2). Low muscle mass (LMM) was defined as a SMI value 2 standard deviations (SDs) below the mean, which has been reported to be less than 8.87 kg/m2 in males or 6.42 kg/m2 in females in Taiwan.19 There were 97 persons classified in the LMM group (31 males and 66 females) and 230 persons classified in the normal group (73 males and 157 females).
Measurement of Serum Selenium Level and Other Biomarkers Venous blood samples were taken after fasting for at least 8 hours. Serum glucose, total cholesterol, high-density lipoprotein cholesterol, low-density lipoprotein cholesterol (LDL-C), and triglycerides were assessed by automatic spectrophotometric assay. Selenium was measured using inductively coupled plasma mass spectroscopy.12 Statistical Analysis Statistical analyses were performed using SPSS v17.0 (SPSS Inc., Chicago, IL). A P value of less than .05 was considered to be statistically significant. The participants were divided into quartiles according to serum selenium level. Analysis of variance was used for continuous variables and the c2 test for categorical variables to analyze interquartile differences. The risk of LMM by selenium levels was analyzed using multivariate binary logistic regression models. Age, sex, and BMI were adjusted in model 1. Exercise, smoking, and MNA score were further adjusted in model 2. MetS, LDL-C, and uric acid level were added in model 3. To further explore the relationship between the serum selenium and SMI, the serum selenium levels of LMM and normal group were estimated by least square (LS) method in a multivariate linear regression model after adjusting confounding factors based on model 3 except sex. The SMI across serum selenium quartiles was also estimated by LS method in multivariate linear regression model after adjusting confounding factors in model 3. Results The demographics, personal history, and laboratory findings of the participants by serum selenium quartiles are shown in Table 1. The participant Barthel index was 100 and instrumental activities of daily living scale score was 7.81 0.51 (maximum is 8 points). The mean SMI values in the LMM group were 8.11 0.59 kg/m2 in males and 5.80 0.42 kg/m2 in females. The mean SMI values in the normal group were 10.12 0.82 kg/m2 in males and 7.81 0.93 kg/m2 in females. The percentages of LMM were 29.7% in all participants, 29.8% in men, and 29.6% in females, respectively. The mean selenium level was 1.10 0.25 mmol/L, and the interquartile cut-off values of selenium were 0.90, 1.08, and 1.29 mmol/L. There were no interquartile differences in lifestyle, functional status, and most metabolic or anthropometric measurements, however, there were significant differences in female BMI (P ¼ .016), serum LDL cholesterol level (P ¼ .005), and uric acid level (P ¼ .015). The SMI values increased with increasing quartile of serum selenium concentration (P ¼ .022), and the percentage of LMM decreased significantly across the quartiles of serum selenium (P ¼ .001). After adjustments for lifestyle, metabolic and anthropometric confounding factors, the LS mean (SD) of serum selenium level in the LMM group was significantly lower than in the normal group (1.01 0.03 mmol/L vs 1.14 0.02 mmol/L, respectively, P < .001, Figure 1). Table 2 showed the association of serum selenium and LMM by multivariate logistic regression analyses in models 1e3. The results showed that a lower selenium level was correlated with a higher risk of LMM after adjusting for confounding factors (P for trend were <.001, .001, and .001 in models 1e3; R2 were 0.530, 0.534 and 0.591 in models 1e3, respectively). The odds ratios of risk for LMM in the bottom, second, and third selenium quartile groups were 4.62 (95% CI 2.12e10.1, P < .001), 2.30 (95% CI 1.05e5.03, P < .05) and 1.51 (95% CI 0.66e3.46, P ¼ .327), respectively, compared with the top quartile in model 3. Figure 2 shows that the LS means (SDs) of the SMI increased with increasing serum selenium concentration in the linear multivariate regression models after adjusting for the confounding
Y.-L. Chen et al. / JAMDA 15 (2014) 807e811
809
Table 1 Demographics and Laboratory Data of the Study Participants by Serum Selenium Quartiles (N ¼ 327)* Selenium range (mmol/L)
Age (y) Male, n (%) BMI (kg/m2) Male Female Currently smoking, n (%) Exercise, n (%)z MNA score (points) MetS, n (%) LDL-C (mmol/L) (normal range: < 3.37 mmol/L) Uric acid (mmol/L) (normal range: < 0.42 mmol/L) Barthel Index (points) IADL score (points) Selenium (mmol/L) Male Female SMI (kg/m2) Male Female LMM, n (%)x Male Female
P Valuey
Quartiles
Total (n ¼ 327)
Q1 (n ¼ 81) <0.90
Q2 (n ¼ 81) 0.90w1.08
Q3 (n ¼ 82) 1.08w1.29
Q4 (n ¼ 83) >1.29
71.4 5.3 22 (27.2) 23.77 3.12 23.53 3.90 23.86 2.81 6 (7.4) 67 (82.7) 27.1 1.6 41 (50.6) 3.16 0.73
71.8 4.8 26 (32.1) 23.25 3.05 23.25 3.75 23.25 2.69 1 (1.2) 71 (87.7) 27.0 1.7 26 (32.1) 3.22 0.71
70.8 4.5 25 (30.5) 24.02 3.22 24.35 2.92 23.88 3.36 4 (4.9) 72 (88.9) 27.2 1.9 40 (48.8) 3.21 0.70
72.1 4.3 31 (37.3) 22.89 2.87 23.86 3.07 22.31 2.60 9 (10.8) 75 (90.4) 26.8 2.0 35 (42.2) 3.51 0.65
.333 .562 .08 .691 .016 .07 .485 .537 .074 .005
71.5 4.7 104 (31.8%) 23.48 3.08 23.76 3.38 23.35 2.94 20 (6) 285 (87.2%) 27.0 1.8 142 (43.4) 3.28 0.71
0.28 0.08
0.30 0.09
0.32 0.09
0.31 0.08
.015
0.30 0.08
100 7.86 0.38 0.79 0.07 0.77 0.07 0.79 0.07 7.49 1.66 9.22 1.11 6.85 1.34 37 (45.7) 10 (45.5) 27 (45.8)
100 7.85 0.39 1.00 0.05 1.00 0.05 1.00 0.06 7.97 1.56 9.37 1.39 7.31 1.16 25 (30.9) 9 (34.6) 16 (29.1)
100 7.71 0.71 1.17 0.06 1.18 0.06 1.16 0.05 8.17 1.63 9.84 0.98 7.43 1.27 18 (22) 4 (16.0) 24 (24.6)
100 7.82 0.47 1.43 0.14 1.42 0.09 1.44 0.16 8.17 1.57 9.60 1.22 7.32 1.06 17 (20.5) 8 (25.8) 9 (17.3)
N/A .184 <.001 <.001 <.001 .022 .285 .05 .001 .146 .008
100 7.81 0.51 1.10 0.25 1.12 0.25 1.09 0.25 7.95 1.62 9.52 1.19 7.22 1.23 97 (29.7) 31 (29.8) 66 (29.6)
IADL, instrumental activities of daily living (Lawton-Brody version); N/A, not applicable. Dichotomous items: currently smoking, exercise, metabolic syndrome (MetS), and LMM. *Continuous variable is presented by mean SD and categorical variables are presented as the percentage of participants (%). y P value according to the c2 test for categorical variables and ANOVA for continuous variables. z The variable of exercise is dichotomous. The number and percentage represented the participants who exercise at least once a week. x Low muscle mass is defined by a SMI<8.87 kg/m2 in males and <6.42 kg/m2 in females.
factors (quartile 1 SMI ¼ 7.53 0.12 kg/m2; quartile 2 SMI ¼ 7.94 0.12 kg/m2; quartile 3 SMI ¼ 8.08 0.12 kg/m2; quartile 4 SMI ¼ 8.27 0.12 kg/m2). Furthermore, the SMI adjusted for confounders in model 3 was 8.43 0.08 kg/m2 in normal group and 6.83 0.13 kg/m2 in the LMM group (P < .001). Discussion In our study, the participants in the lowest selenium quartile had 4.62-fold of risk of LMM compared with those in the highest quartile. The finding indicated a higher serum selenium level might be associated with higher skeletal muscle mass.
Fig. 1. The adjusted means SDs of serum selenium level in the LMM and normal groups. The means of serum selenium were calculated by LS method using a linear regression model after adjusting for age, BMI, currently smoking, exercise, MNA score, MetS, serum LDL-C, and uric acid level.
In our study, prevalence of LMM was 29.8% in males and 29.6% in females. Compared with previous Chien et al’s study in Taiwan using the same cutoff values of SMI for defining LMM and similar age group,19 our females had higher prevalence of LMM than in Chien et al’s (male 23.6% and female 18.6%). The difference of LMM prevalence might be related to different population characteristics, higher BMI, or exercise in Chien et al.’s population. Both our study and Chien et al.’s study used BIA method to evaluate SMI in participants. Chien et al.’s study showed high correlation (r2 ¼ 0.95) between skeletal mass estimated by using a BIA prediction equation and skeletal mass measured by magnetic resonance imaging. Compared with dualenergy X-ray absorptiometry and magnetic resonance imaging, BIA method has advantages of portability, convenience, and low cost. Therefore, BIA method was an appropriate measurement for skeletal mass in Taiwanese people. Only a few studies have investigated the relationship between selenium and muscles in the aging process. Previous 2 cross-sectional studies reported that a low selenium concentration is associated with poor muscle strength.10,11 One is from the Italian population; their plasma selenium level (0.95 0.15 mmol/L) was lower than ours (1.10 0.25 mmol/L). The other study is from the US; the serum selenium level is 1.49 0.23 mmol/L, which is much higher than the level in our population. There is no consensus about “normal selenium level” in human body currently and selenium status varies among regions and races. It might also be related to differences in body composition, dietary factors, smoking status, and disease status. In addition, these findings were either limited to female participants or lacked data on muscle mass, therefore, relationships among skeletal muscle mass, muscle strength, and selenium status in different populations need further research. Some studies investigate the association between selenium supplement and muscle strength or muscle mass. Chaput et al.’s study suggested that patients who took lower amounts of antioxidant
810
Y.-L. Chen et al. / JAMDA 15 (2014) 807e811
Table 2 Different Models of Multivariate Logistic Regression Analysis to Demonstrate Odds Ratios (95% CIs) of LMM Comparing the Bottom, Second, and Third Quartile Group With the Top Quartile of Selenium Concentration, Respectively Selenium (mmol/L)
Model 1 Model 2 Model 3
Quartile
P for Trend
Q1 (n ¼ 81) <0.90
Q2 (n ¼ 81) 0.90w1.08
Q3 (n ¼ 82) 1.08w1.29
Q4 (n ¼ 83) >1.29
3.98 (1.94e8.13)* 3.80 (1.83e7.87)* 4.62 (2.11e10.1)*
1.88 (0.91e3.89) 1.75 (0.83e3.69) 2.30 (1.05e5.03)y
1.23 (0.57e2.67) 1.14 (0.52e2.53) 1.51 (0.66e3.46)
Reference Reference Reference
<.001 .001 .001
Model 1: adjusted for age, sex, and BMI; R2 ¼ 0.530. Model 2: adjusted for age, sex, BMI, current smoking, exercise, and MNA; R2 ¼ 0.534. Model 3: adjusted for age, sex, BMI, current smoking, exercise, MNA, MetS, LDL-C, and uric acid; R2 ¼ 0.591. Odds ratio (95% CI, P value) of adjusted factors in Model 3: Age: 1.02 (0.96e1.08, P ¼ .506); sex: 1.61 (0.85e3.04, P ¼ .146); BMI: 0.86 (0.77e0.97, P ¼ .01); current smoking: 0.36 (0.09e1.41, P ¼ .14); exercise: 0.64 (0.31e1.33, P ¼ .23); MNA: 0.85 (0.72e1.00, P ¼ .045); MetS: 1.73 (0.92e3.26, P ¼ .087); LDL-C: 1.02 (1.01e1.03, P ¼ .002); uric acid: 0.76 (0.61e0.96, P ¼ .022). *P < .05. y P < .001.
supplements had lower levels of skeletal muscle mass, although the amount of selenium intake alone was not associated with differences in skeletal muscle mass.5 The study emphasizes the association between class 1 sarcopenia (SMI less than 1e2 SDs below the mean) and physical activity, the amount of antioxidant, and protein intake, therefore, the definition of LMM is different from that in our study. In addition, selenium intake may not reflect serum selenium level of participants. Another study reported improved muscle strength and fewer symptoms after selenium supplementation in patients with long-term parenteral nutrition9 and fibromuscular rheumatism.20 However, it is difficult to extrapolate the results to the general population because of the small sample size or confounding underlying illness. Various factors have been reported to be associated with the development of sarcopenia. These factors include hyperuricemia, insulin resistance, and MetS.2124 In our present study, we found that BMI, MNA score, serum uric acid level, and LDL-C level were associated with risks for LMM. Furthermore, serum glucose and lipids were found to be related to selenium in our previous study.12 In the current study, after adjusting for age, sex, BMI, MetS, LDL-C, and uric acid, a higher serum selenium level was still associated with a higher skeletal muscle index. Although the causal relationship was not conclusive in this cross-sectional study, the results suggest that selenium may be related to skeletal muscle mass and involved in the development of sarcopenia.
Fig. 2. The adjusted means SDs of SMI in the quartile groups of serum selenium. The means of SMI were calculated by LS method using linear regression model after adjusting for age, sex, BMI, currently smoking, exercise, MNA score, MetS, serum LDL-C, and uric acid level.
There are several possible mechanisms for the effects of selenium on muscle mass. Some selenoproteins in myocytes such as glutathione peroxidase and methionine sulfoxide reductase have been shown to exhibit antioxidative activity, while others have been shown to exhibit various roles in muscle function and development.6,7,25,26 Selenoprotein N has been shown to influence calcium homeostasis of muscles via action on ryanodine receptors.26,27 Selenoprotein W, which is highly expressed in skeletal muscles, hearts, and brains of mammals, has been reported to protect undifferentiated myoblasts from oxidative stress.28 Selenoprotein coactivator Peroxisome proliferator-activated receptor-gamma coactivator (PGC)1alpha has been reported to regulate selenoprotein P to prevent muscle wasting by reducing apoptosis and proteasome degradation during aging in mice.29,30 Mutations of selenoprotein genes or selenium deficiency have been reported to be related to certain diseases such as multiminicore disease, congenital muscular dystrophy, and Keshan disease.6 Mutations in selenoprotein N genes (eg, the SEPN1 gene) may lead to a number of early-onset muscular diseases.25,31,32 Taken together, selenium deficiency may lead to hypoactivity of selenoproteins and their co-factors, and then the proinflammatory condition and impairment of muscular mitochondrial function lead to an acceleration of myocyte apoptosis.10,33,34 Some studies have emphasized the therapeutic role of selenium. Mixed antioxidant supplements have been shown to reverse defects in leucine-stimulated muscle protein synthesis in rats, possibly indirectly through reducing systemic inflammation.35,36 Seleniumcontaining antioxidant supplements were also reported to potentially alleviate muscle damage in athletes by raising glutathione peroxidase activity and inhibiting muscle enzyme elevation.37 However, some evidence suggests that excessive intake of antioxidants may interfere with reactive oxygen species-mediated physiological adaptation, such as vasodilation and insulin signaling.38 Furthermore, long-term high dose selenium supplementations have been reported to potentially increase the risk of developing type 2 diabetes mellitus and cardiovascular disease.39,40 Therefore, the use of selenium supplementations to alleviate muscle loss or prevent sarcopenia warrants further research. There are some limitations to this study. First, according to the European consensus group, the definition of sarcopenia includes muscle function, which was not measured in this study.41 Second, information on the amount of selenium from dietary intake or supplementation was not available, and this may have influenced the selenium level. Third, selection bias may have existed because we recruited participants by advertisement and our participants were relatively independent elderly. Finally, we were not able to establish the causal relationship between serum selenium and sarcopenia because of the cross-sectional design of the study.
Y.-L. Chen et al. / JAMDA 15 (2014) 807e811
Conclusions In conclusion, this study is the first to directly investigate the relationship of serum selenium and skeletal muscle mass in the elderly. The results showed that after adjusting for age, sex, lifestyle, and physical and metabolic factors, lower serum selenium levels were associated with a higher risk of low skeletal muscle mass. However, recommending selenium supplements should be carefully evaluated. The underlying mechanism of selenium on sarcopenia warrants further research. Acknowledgments The authors would like to thank Miss Wen-Chao Weng and Yi-Ju Chen for their work. References 1. Morley JE. Anorexia, sarcopenia, and aging. Nutrition 2001;17:660e663. 2. Thomas DR. Sarcopenia. Clin Geriatr Med 2010;26:331e346. 3. Jackson MJ. Strategies for reducing oxidative damage in ageing skeletal muscle. Adv Drug Deliv Rev 2009;61:1363e1368. 4. Jones TE, Stephenson KW, King JG, et al. SarcopeniadMechanisms and treatments. J Geriatr Phys Ther 2009;32:39e45. 5. Chaput JP, Lord C, Cloutier M, et al. Relationship between antioxidant intakes and class I sarcopenia in elderly men and women. J Nutr Health Aging 2007;11: 363e369. 6. Rederstorff M, Krol A, Lescure A. Understanding the importance of selenium and selenoproteins in muscle function. Cell Mol Life Sci 2006;63:52e59. 7. Lescure A, Rederstorff M, Krol A, et al. Selenoprotein function and muscle disease. Biochim Biophys Acta 2009;1790:1569e1574. 8. Beytut E, Karatas F. Lambs with white muscle disease and selenium content of soil and meadow hay in the region of Kars, Turkey. Vet J 2002;163:214e217. 9. Brown MR, Cohen HJ, Lyons JM, et al. Proximal muscle weakness and selenium deficiency associated with long term parenteral nutrition. Am J Clin Nutr 1986; 43:549e554. 10. Lauretani F, Semba RD, Bandinelli S, et al. Association of low plasma selenium concentrations with poor muscle strength in older community-dwelling adults: The InCHIANTI Study. Am J Clin Nutr 2007;86:347e352. 11. Beck J, Ferrucci L, Sun K, et al. Low serum selenium concentrations are associated with poor grip strength among older women living in the community. Biofactors 2007;29:37e44. 12. Yang KC, Lee LT, Lee YS, et al. Serum selenium concentration is associated with metabolic factors in the elderly: A cross-sectional study. Nutr Metab (Lond) 2010;7:38. 13. Guigoz Y, Lauque S, Vellas BJ. Identifying the elderly at risk for malnutrition. The Mini Nutritional Assessment. Clin Geriatr Med 2002;18:737e757. 14. Tsai AC, Ho CS, Chang MC. Assessing the prevalence of malnutrition with the Mini Nutritional Assessment (MNA) in a nationally representative sample of elderly Taiwanese. J Nutr Health Aging 2008;12:239e243. 15. Mahoney FI, Barthel DW. Functional evaluation: The Barthel Index. Md State Med J 1965;14:61e65. 16. Lawton MP, Brody EM. Assessment of older people: Self-maintaining and instrumental activities of daily living. Gerontologist 1969;9:179e186. 17. Huang KC. Obesity and its related diseases in Taiwan. Obes Rev 2008;9:32e34. 18. Wu CH, Yang KC, Chang HH, et al. Sarcopenia is related to increased risk for low bone mineral density. J Clin Densitom 2013;16:98e103. 19. Chien MY, Huang TY, Wu YT. Prevalence of sarcopenia estimated using a bioelectrical impedance analysis prediction equation in community-dwelling elderly people in Taiwan. J Am Geriatr Soc 2008;56:1710e1715.
811
20. Robinson MF, Campbell DR, Stewart RD, et al. Effect of daily supplements of selenium on patients with muscular complaints in Otago and Canterbury. N Z Med J 1981;93:289e292. 21. Arnaud J, de Lorgeril M, Akbaraly T, et al. Gender differences in copper, zinc and selenium status in diabetic-free metabolic syndrome European populationdThe IMMIDIET study. Nutr Metab Cardiovasc Dis 2012;22: 517e524. 22. Beavers KM, Beavers DP, Serra MC, et al. Low relative skeletal muscle mass indicative of sarcopenia is associated with elevations in serum uric acid levels: Findings from NHANES III. J Nutr Health Aging 2009;13:177e182. 23. Chung JY, Kang HT, Lee DC, et al. Body composition and its association with cardiometabolic risk factors in the elderly: A focus on sarcopenic obesity. Arch Gerontol Geriatr 2013;56:270e278. 24. Seale LA, Hashimoto AC, Kurokawa S, et al. Disruption of the selenocysteine lyase-mediated selenium recycling pathway leads to metabolic syndrome in mice. Mol Cell Biol 2012;32:4141e4154. 25. Ferreiro A, Quijano-Roy S, Pichereau C, et al. Mutations of the selenoprotein N gene, which is implicated in rigid spine muscular dystrophy, cause the classical phenotype of multiminicore disease: Reassessing the nosology of early-onset myopathies. Am J Hum Genet 2002;71:739e749. 26. Castets P, Lescure A, Guicheney P, Allamand V. Selenoprotein N in skeletal muscle: From diseases to function. J Mol Med (Berl) 2012;90: 1095e1107. 27. Jurynec MJ, Xia R, Mackrill JJ, et al. Selenoprotein N is required for ryanodine receptor calcium release channel activity in human and zebrafish muscle. Proc Natl Acad Sci U S A 2008;105:12485e12490. 28. Loflin J, Lopez N, Whanger PD, Kioussi C. Selenoprotein W during development and oxidative stress. J Inorg Biochem 2006;100:1679e1684. 29. Speckmann B, Walter PL, Alili L, et al. Selenoprotein P expression is controlled through interaction of the coactivator PGC-1alpha with FoxO1a and hepatocyte nuclear factor 4alpha transcription factors. Hepatology 2008;48:1998e2006. 30. Wenz T, Rossi SG, Rotundo RL, et al. Increased muscle PGC-1alpha expression protects from sarcopenia and metabolic disease during aging. Proc Natl Acad Sci U S A 2009;106:20405e20410. 31. D’Amico A, Haliloglu G, Richard P, et al. Two patients with ’dropped head syndrome’ due to mutations in LMNA or SEPN1 genes. Neuromuscul Disord 2005;15:521e524. 32. Ferreiro A, Ceuterick-de Groote C, Marks JJ, et al. Desmin-related myopathy with Mallory body-like inclusions is caused by mutations of the selenoprotein N gene. Ann Neurol 2004;55:676e686. 33. Jackson MJ. Skeletal muscle aging: Role of reactive oxygen species. Crit Care Med 2009;37:S368eS371. 34. Walston J, Xue Q, Semba RD, et al. Serum antioxidants, inflammation, and total mortality in older women. Am J Epidemiol 2006;163:18e26. 35. Marzani B, Balage M, Venien A, et al. Antioxidant supplementation restores defective leucine stimulation of protein synthesis in skeletal muscle from old rats. J Nutr 2008;138:2205e2211. 36. Mosoni L, Balage M, Vazeille E, et al. Antioxidant supplementation had positive effects in old rat muscle, but through better oxidative status in other organs. Nutrition 2010;26:1157e1162. 37. Palazzetti S, Rousseau AS, Richard MJ, et al. Antioxidant supplementation preserves antioxidant response in physical training and low antioxidant intake. Br J Nutr 2004;91:91e100. 38. Peternelj TT, Coombes JS. Antioxidant supplementation during exercise training: Beneficial or detrimental? Sports Med 2011;41:1043e1069. 39. Stranges S, Marshall JR, Natarajan R, et al. Effects of long-term selenium supplementation on the incidence of type 2 diabetes: A randomized trial. Ann Intern Med 2007;147:217e223. 40. Stranges S, Navas-Acien A, Rayman MP, Guallar E. Selenium status and cardiometabolic health: State of the evidence. Nutr Metab Cardiovasc Dis 2010; 20:754e760. 41. Cruz-Jentoft AJ, Baeyens JP, Bauer JM, et al. Sarcopenia: European consensus on definition and diagnosis: Report of the European Working Group on sarcopenia in older people. Age Ageing 2010;39:412e423.
JAMDA journal homepage: www.jamda.com
Original Study
Low Serum Selenium Level Is Associated With Low Muscle Mass in the Community-Dwelling Elderly You-Ling Chen MD a, y, Kuen-Cheh Yang MD, MSc b, Hao-Hsiang Chang MD, MSc a, Long-Teng Lee MD, PhD a, y, Chia-Wen Lu MD, MSc a, Kuo-Chin Huang MD, PhD a, c, d, * a
Department of Family Medicine, National Taiwan University Hospital, Taipei, Taiwan Department of Community and Family Medicine, National Taiwan University Hospital, Hsin-Chu Branch, Hsinchu, Taiwan c Department of Family Medicine, College of Medicine, National Taiwan University, Taipei, Taiwan d Graduate Institute of Clinical Medical Science, China Medical University, Taichung, Taiwan b
a b s t r a c t Keywords: Skeletal muscle mass selenium bioelectrical impedance analysis elderly
Objectives: Elderly persons with low muscle mass (LMM) or sarcopenia are prone to frailty and functional decline. This study aimed to investigate the relationship between serum selenium level and skeletal muscle mass in community-dwelling elderly. Design: Cross-sectional observational study. Setting and participants: A total of 327 elderly Taipei citizens (mean age 71.5 4.7 years) were recruited from the community. Measurements: Skeletal muscle mass was measured by bioelectrical impedance analysis. LMM was defined by low skeletal muscle index (SMI: muscle mass (kg)/[height (m)]2). All participants were further divided into quartiles by serum selenium level and the risk for LMM among these quartiles was examined using multivariate logistic regression analyses. Estimated serum selenium levels for the LMM group vs the normal group and estimated SMI in the quartiles of serum selenium were computed by least square method in linear regression models. Results: The estimated mean (standard deviation) of serum selenium level was significantly lower in the LMM group compared with the normal group after adjusting for confounders (1.01 0.03 mmol/L vs 1.14 0.02 mmol/L, P < .001). After adjusting for age, sex, lifestyle, and physical and metabolic factors, the odds ratios (95% confidence interval, P value) of LMM in the bottom, second, and third selenium quartile groups were 4.62 (95% CI 2.11e10.10, P < .001), 2.30 (95% CI 1.05e5.03, P < .05) and 1.51 (95% CI 0.66 e3.46, P ¼ .327), respectively, compared with the top quartile group of serum selenium level. The least square mean of SMI increased with the quartiles of serum selenium (P < .001). Conclusions: This is the first study to demonstrate that low serum selenium is independently associated with low muscle mass in the elderly. The causality and underlying mechanism between selenium and low muscle mass or sarcopenia warrant further research. Ó 2014 AMDA e The Society for Post-Acute and Long-Term Care Medicine.
Sarcopenia is the gradual loss of muscle mass and strength with aging. Geriatric patients with sarcopenia usually suffer from frailty and disability.1 Previous studies have reported several risk factors for sarcopenia, including genetic factors, nutrition, exercise,
This study was funded by the National Health Institute of Taiwan (PH-101-PP42). The authors declare no conflicts of interest. * Address correspondence to Kuo-Chin Huang, MD, PhD, Department of Family Medicine, National Taiwan University Hospital, No. 7, Chung-Shan South Road, Taipei 100, Taiwan. E-mail address: [email protected] (K.-C. Huang). y Contributed equally to this work.
neuromuscular dysfunction, hormone reactivity, or inflammation.2 In addition, oxidative injuries during aging play a crucial role in muscle loss.3,4 It has been proposed that taking recommended dietary allowances of antioxidants may protect people from abnormal reductions of skeletal muscle mass.5 Selenium is one of the most important antioxidants in humans. The distribution and function of selenoenzymes such as cytosolic glutathione peroxidase and selenoprotein P modulate protection against oxidative damage, and mutations in selenoprotein genes can lead to muscular diseases.6,7 In lambs, selenium deficiency is also associated with white muscle disease characterized by weakness and calcification in skeletal muscles and myocardium.8 One study investigating the association between selenium and muscle function showed an
1525-8610/$ - see front matter Ó 2014 AMDA e The Society for Post-Acute and Long-Term Care Medicine. http://dx.doi.org/10.1016/j.jamda.2014.06.014
808
Y.-L. Chen et al. / JAMDA 15 (2014) 807e811
improvement in muscle strength after selenium supplementation in a patient with long-term parenteral nutrition,9 whereas another study reported no difference between normal persons and those with decreased muscle mass with regard to dietary selenium intake.5 The Invecchiare in Chianti (InCHIANTI) study enrolled randomly selected Italian persons older than 65 years and tested the association between serum selenium level and handgrip, hip flexion, and knee extension muscle strength of the persons.10 Beck et al11 also examined the association of serum selenium level and handgrip strength in elderly female persons. Both studies concluded that a low selenium level was associated with reduced muscle strength.10,11 Despite the attempt to elucidate the role of selenium in muscle function, these studies focused on muscle strength measures instead of direct evaluation of muscle mass. Whether or not selenium influences the maintenance of muscle mass, which provides clinicians with a greater insight into the pathogenic mechanism of sarcopenia, remains unclear. Therefore, we conducted this cross-sectional community study to investigate the relationship of serum selenium and skeletal muscle mass in the community-dwelling elderly. Methods Participants A total of 327 volunteers living in Taipei were recruited in 2007 by advertisements. The inclusion criteria were individuals aged 65 years or older and could stand steadily on the bioelectrical impedance analysis (BIA) system during measurements of bioelectrical impedance. Participants with cancer history, recent body weight change more than 5% in 3 months, and those implanted with cardiac pacemakers or metal implants were excluded. All participants provided written informed consent, and all protocols were approved by the Ethics Committee of the National Taiwan University Hospital. Data on age, sex, history of hypertension, diabetes mellitus, cardiovascular diseases, cigarette smoking, exercise habits, nutritional status, and functional status were obtained by individual interviews. Current and noncurrent smokers were defined according to their recent smoking habit.12 Exercise status was defined as whether the subject had exercise habit at least once a week. The nutritional status of the participants was evaluated by mini-nutritional assessment (MNA).13,14 Functional status was evaluated by Barthel index and Lawton-Brody instrumental activities of daily living scale.15,16 Height and weight were measured to obtain body mass index (BMI).12 Waist circumferences, and systolic and diastolic blood pressure were recorded for evaluating status of metabolic syndrome (MetS). Disease histories such as diabetes mellitus, hypertension, and hyperlipidemia were evaluated, and current prescriptions including antihypertensive, antihyperglycemic, or antihyperlipidemia agents were recorded based on self-reported history. MetS was diagnosed according to the America Heart Association and National Heart Lung and Blood Institute criteria modified for Taiwan population.17 Skeletal Muscle Mass Estimation and Low Muscle Mass Definition Skeletal muscle mass was estimated by bioelectrical impedance analysis.18 Skeletal muscle mass index (SMI) was calculated by dividing the skeletal muscle mass with the square of body height (kg/m2). Low muscle mass (LMM) was defined as a SMI value 2 standard deviations (SDs) below the mean, which has been reported to be less than 8.87 kg/m2 in males or 6.42 kg/m2 in females in Taiwan.19 There were 97 persons classified in the LMM group (31 males and 66 females) and 230 persons classified in the normal group (73 males and 157 females).
Measurement of Serum Selenium Level and Other Biomarkers Venous blood samples were taken after fasting for at least 8 hours. Serum glucose, total cholesterol, high-density lipoprotein cholesterol, low-density lipoprotein cholesterol (LDL-C), and triglycerides were assessed by automatic spectrophotometric assay. Selenium was measured using inductively coupled plasma mass spectroscopy.12 Statistical Analysis Statistical analyses were performed using SPSS v17.0 (SPSS Inc., Chicago, IL). A P value of less than .05 was considered to be statistically significant. The participants were divided into quartiles according to serum selenium level. Analysis of variance was used for continuous variables and the c2 test for categorical variables to analyze interquartile differences. The risk of LMM by selenium levels was analyzed using multivariate binary logistic regression models. Age, sex, and BMI were adjusted in model 1. Exercise, smoking, and MNA score were further adjusted in model 2. MetS, LDL-C, and uric acid level were added in model 3. To further explore the relationship between the serum selenium and SMI, the serum selenium levels of LMM and normal group were estimated by least square (LS) method in a multivariate linear regression model after adjusting confounding factors based on model 3 except sex. The SMI across serum selenium quartiles was also estimated by LS method in multivariate linear regression model after adjusting confounding factors in model 3. Results The demographics, personal history, and laboratory findings of the participants by serum selenium quartiles are shown in Table 1. The participant Barthel index was 100 and instrumental activities of daily living scale score was 7.81 0.51 (maximum is 8 points). The mean SMI values in the LMM group were 8.11 0.59 kg/m2 in males and 5.80 0.42 kg/m2 in females. The mean SMI values in the normal group were 10.12 0.82 kg/m2 in males and 7.81 0.93 kg/m2 in females. The percentages of LMM were 29.7% in all participants, 29.8% in men, and 29.6% in females, respectively. The mean selenium level was 1.10 0.25 mmol/L, and the interquartile cut-off values of selenium were 0.90, 1.08, and 1.29 mmol/L. There were no interquartile differences in lifestyle, functional status, and most metabolic or anthropometric measurements, however, there were significant differences in female BMI (P ¼ .016), serum LDL cholesterol level (P ¼ .005), and uric acid level (P ¼ .015). The SMI values increased with increasing quartile of serum selenium concentration (P ¼ .022), and the percentage of LMM decreased significantly across the quartiles of serum selenium (P ¼ .001). After adjustments for lifestyle, metabolic and anthropometric confounding factors, the LS mean (SD) of serum selenium level in the LMM group was significantly lower than in the normal group (1.01 0.03 mmol/L vs 1.14 0.02 mmol/L, respectively, P < .001, Figure 1). Table 2 showed the association of serum selenium and LMM by multivariate logistic regression analyses in models 1e3. The results showed that a lower selenium level was correlated with a higher risk of LMM after adjusting for confounding factors (P for trend were <.001, .001, and .001 in models 1e3; R2 were 0.530, 0.534 and 0.591 in models 1e3, respectively). The odds ratios of risk for LMM in the bottom, second, and third selenium quartile groups were 4.62 (95% CI 2.12e10.1, P < .001), 2.30 (95% CI 1.05e5.03, P < .05) and 1.51 (95% CI 0.66e3.46, P ¼ .327), respectively, compared with the top quartile in model 3. Figure 2 shows that the LS means (SDs) of the SMI increased with increasing serum selenium concentration in the linear multivariate regression models after adjusting for the confounding
Y.-L. Chen et al. / JAMDA 15 (2014) 807e811
809
Table 1 Demographics and Laboratory Data of the Study Participants by Serum Selenium Quartiles (N ¼ 327)* Selenium range (mmol/L)
Age (y) Male, n (%) BMI (kg/m2) Male Female Currently smoking, n (%) Exercise, n (%)z MNA score (points) MetS, n (%) LDL-C (mmol/L) (normal range: < 3.37 mmol/L) Uric acid (mmol/L) (normal range: < 0.42 mmol/L) Barthel Index (points) IADL score (points) Selenium (mmol/L) Male Female SMI (kg/m2) Male Female LMM, n (%)x Male Female
P Valuey
Quartiles
Total (n ¼ 327)
Q1 (n ¼ 81) <0.90
Q2 (n ¼ 81) 0.90w1.08
Q3 (n ¼ 82) 1.08w1.29
Q4 (n ¼ 83) >1.29
71.4 5.3 22 (27.2) 23.77 3.12 23.53 3.90 23.86 2.81 6 (7.4) 67 (82.7) 27.1 1.6 41 (50.6) 3.16 0.73
71.8 4.8 26 (32.1) 23.25 3.05 23.25 3.75 23.25 2.69 1 (1.2) 71 (87.7) 27.0 1.7 26 (32.1) 3.22 0.71
70.8 4.5 25 (30.5) 24.02 3.22 24.35 2.92 23.88 3.36 4 (4.9) 72 (88.9) 27.2 1.9 40 (48.8) 3.21 0.70
72.1 4.3 31 (37.3) 22.89 2.87 23.86 3.07 22.31 2.60 9 (10.8) 75 (90.4) 26.8 2.0 35 (42.2) 3.51 0.65
.333 .562 .08 .691 .016 .07 .485 .537 .074 .005
71.5 4.7 104 (31.8%) 23.48 3.08 23.76 3.38 23.35 2.94 20 (6) 285 (87.2%) 27.0 1.8 142 (43.4) 3.28 0.71
0.28 0.08
0.30 0.09
0.32 0.09
0.31 0.08
.015
0.30 0.08
100 7.86 0.38 0.79 0.07 0.77 0.07 0.79 0.07 7.49 1.66 9.22 1.11 6.85 1.34 37 (45.7) 10 (45.5) 27 (45.8)
100 7.85 0.39 1.00 0.05 1.00 0.05 1.00 0.06 7.97 1.56 9.37 1.39 7.31 1.16 25 (30.9) 9 (34.6) 16 (29.1)
100 7.71 0.71 1.17 0.06 1.18 0.06 1.16 0.05 8.17 1.63 9.84 0.98 7.43 1.27 18 (22) 4 (16.0) 24 (24.6)
100 7.82 0.47 1.43 0.14 1.42 0.09 1.44 0.16 8.17 1.57 9.60 1.22 7.32 1.06 17 (20.5) 8 (25.8) 9 (17.3)
N/A .184 <.001 <.001 <.001 .022 .285 .05 .001 .146 .008
100 7.81 0.51 1.10 0.25 1.12 0.25 1.09 0.25 7.95 1.62 9.52 1.19 7.22 1.23 97 (29.7) 31 (29.8) 66 (29.6)
IADL, instrumental activities of daily living (Lawton-Brody version); N/A, not applicable. Dichotomous items: currently smoking, exercise, metabolic syndrome (MetS), and LMM. *Continuous variable is presented by mean SD and categorical variables are presented as the percentage of participants (%). y P value according to the c2 test for categorical variables and ANOVA for continuous variables. z The variable of exercise is dichotomous. The number and percentage represented the participants who exercise at least once a week. x Low muscle mass is defined by a SMI<8.87 kg/m2 in males and <6.42 kg/m2 in females.
factors (quartile 1 SMI ¼ 7.53 0.12 kg/m2; quartile 2 SMI ¼ 7.94 0.12 kg/m2; quartile 3 SMI ¼ 8.08 0.12 kg/m2; quartile 4 SMI ¼ 8.27 0.12 kg/m2). Furthermore, the SMI adjusted for confounders in model 3 was 8.43 0.08 kg/m2 in normal group and 6.83 0.13 kg/m2 in the LMM group (P < .001). Discussion In our study, the participants in the lowest selenium quartile had 4.62-fold of risk of LMM compared with those in the highest quartile. The finding indicated a higher serum selenium level might be associated with higher skeletal muscle mass.
Fig. 1. The adjusted means SDs of serum selenium level in the LMM and normal groups. The means of serum selenium were calculated by LS method using a linear regression model after adjusting for age, BMI, currently smoking, exercise, MNA score, MetS, serum LDL-C, and uric acid level.
In our study, prevalence of LMM was 29.8% in males and 29.6% in females. Compared with previous Chien et al’s study in Taiwan using the same cutoff values of SMI for defining LMM and similar age group,19 our females had higher prevalence of LMM than in Chien et al’s (male 23.6% and female 18.6%). The difference of LMM prevalence might be related to different population characteristics, higher BMI, or exercise in Chien et al.’s population. Both our study and Chien et al.’s study used BIA method to evaluate SMI in participants. Chien et al.’s study showed high correlation (r2 ¼ 0.95) between skeletal mass estimated by using a BIA prediction equation and skeletal mass measured by magnetic resonance imaging. Compared with dualenergy X-ray absorptiometry and magnetic resonance imaging, BIA method has advantages of portability, convenience, and low cost. Therefore, BIA method was an appropriate measurement for skeletal mass in Taiwanese people. Only a few studies have investigated the relationship between selenium and muscles in the aging process. Previous 2 cross-sectional studies reported that a low selenium concentration is associated with poor muscle strength.10,11 One is from the Italian population; their plasma selenium level (0.95 0.15 mmol/L) was lower than ours (1.10 0.25 mmol/L). The other study is from the US; the serum selenium level is 1.49 0.23 mmol/L, which is much higher than the level in our population. There is no consensus about “normal selenium level” in human body currently and selenium status varies among regions and races. It might also be related to differences in body composition, dietary factors, smoking status, and disease status. In addition, these findings were either limited to female participants or lacked data on muscle mass, therefore, relationships among skeletal muscle mass, muscle strength, and selenium status in different populations need further research. Some studies investigate the association between selenium supplement and muscle strength or muscle mass. Chaput et al.’s study suggested that patients who took lower amounts of antioxidant
810
Y.-L. Chen et al. / JAMDA 15 (2014) 807e811
Table 2 Different Models of Multivariate Logistic Regression Analysis to Demonstrate Odds Ratios (95% CIs) of LMM Comparing the Bottom, Second, and Third Quartile Group With the Top Quartile of Selenium Concentration, Respectively Selenium (mmol/L)
Model 1 Model 2 Model 3
Quartile
P for Trend
Q1 (n ¼ 81) <0.90
Q2 (n ¼ 81) 0.90w1.08
Q3 (n ¼ 82) 1.08w1.29
Q4 (n ¼ 83) >1.29
3.98 (1.94e8.13)* 3.80 (1.83e7.87)* 4.62 (2.11e10.1)*
1.88 (0.91e3.89) 1.75 (0.83e3.69) 2.30 (1.05e5.03)y
1.23 (0.57e2.67) 1.14 (0.52e2.53) 1.51 (0.66e3.46)
Reference Reference Reference
<.001 .001 .001
Model 1: adjusted for age, sex, and BMI; R2 ¼ 0.530. Model 2: adjusted for age, sex, BMI, current smoking, exercise, and MNA; R2 ¼ 0.534. Model 3: adjusted for age, sex, BMI, current smoking, exercise, MNA, MetS, LDL-C, and uric acid; R2 ¼ 0.591. Odds ratio (95% CI, P value) of adjusted factors in Model 3: Age: 1.02 (0.96e1.08, P ¼ .506); sex: 1.61 (0.85e3.04, P ¼ .146); BMI: 0.86 (0.77e0.97, P ¼ .01); current smoking: 0.36 (0.09e1.41, P ¼ .14); exercise: 0.64 (0.31e1.33, P ¼ .23); MNA: 0.85 (0.72e1.00, P ¼ .045); MetS: 1.73 (0.92e3.26, P ¼ .087); LDL-C: 1.02 (1.01e1.03, P ¼ .002); uric acid: 0.76 (0.61e0.96, P ¼ .022). *P < .05. y P < .001.
supplements had lower levels of skeletal muscle mass, although the amount of selenium intake alone was not associated with differences in skeletal muscle mass.5 The study emphasizes the association between class 1 sarcopenia (SMI less than 1e2 SDs below the mean) and physical activity, the amount of antioxidant, and protein intake, therefore, the definition of LMM is different from that in our study. In addition, selenium intake may not reflect serum selenium level of participants. Another study reported improved muscle strength and fewer symptoms after selenium supplementation in patients with long-term parenteral nutrition9 and fibromuscular rheumatism.20 However, it is difficult to extrapolate the results to the general population because of the small sample size or confounding underlying illness. Various factors have been reported to be associated with the development of sarcopenia. These factors include hyperuricemia, insulin resistance, and MetS.2124 In our present study, we found that BMI, MNA score, serum uric acid level, and LDL-C level were associated with risks for LMM. Furthermore, serum glucose and lipids were found to be related to selenium in our previous study.12 In the current study, after adjusting for age, sex, BMI, MetS, LDL-C, and uric acid, a higher serum selenium level was still associated with a higher skeletal muscle index. Although the causal relationship was not conclusive in this cross-sectional study, the results suggest that selenium may be related to skeletal muscle mass and involved in the development of sarcopenia.
Fig. 2. The adjusted means SDs of SMI in the quartile groups of serum selenium. The means of SMI were calculated by LS method using linear regression model after adjusting for age, sex, BMI, currently smoking, exercise, MNA score, MetS, serum LDL-C, and uric acid level.
There are several possible mechanisms for the effects of selenium on muscle mass. Some selenoproteins in myocytes such as glutathione peroxidase and methionine sulfoxide reductase have been shown to exhibit antioxidative activity, while others have been shown to exhibit various roles in muscle function and development.6,7,25,26 Selenoprotein N has been shown to influence calcium homeostasis of muscles via action on ryanodine receptors.26,27 Selenoprotein W, which is highly expressed in skeletal muscles, hearts, and brains of mammals, has been reported to protect undifferentiated myoblasts from oxidative stress.28 Selenoprotein coactivator Peroxisome proliferator-activated receptor-gamma coactivator (PGC)1alpha has been reported to regulate selenoprotein P to prevent muscle wasting by reducing apoptosis and proteasome degradation during aging in mice.29,30 Mutations of selenoprotein genes or selenium deficiency have been reported to be related to certain diseases such as multiminicore disease, congenital muscular dystrophy, and Keshan disease.6 Mutations in selenoprotein N genes (eg, the SEPN1 gene) may lead to a number of early-onset muscular diseases.25,31,32 Taken together, selenium deficiency may lead to hypoactivity of selenoproteins and their co-factors, and then the proinflammatory condition and impairment of muscular mitochondrial function lead to an acceleration of myocyte apoptosis.10,33,34 Some studies have emphasized the therapeutic role of selenium. Mixed antioxidant supplements have been shown to reverse defects in leucine-stimulated muscle protein synthesis in rats, possibly indirectly through reducing systemic inflammation.35,36 Seleniumcontaining antioxidant supplements were also reported to potentially alleviate muscle damage in athletes by raising glutathione peroxidase activity and inhibiting muscle enzyme elevation.37 However, some evidence suggests that excessive intake of antioxidants may interfere with reactive oxygen species-mediated physiological adaptation, such as vasodilation and insulin signaling.38 Furthermore, long-term high dose selenium supplementations have been reported to potentially increase the risk of developing type 2 diabetes mellitus and cardiovascular disease.39,40 Therefore, the use of selenium supplementations to alleviate muscle loss or prevent sarcopenia warrants further research. There are some limitations to this study. First, according to the European consensus group, the definition of sarcopenia includes muscle function, which was not measured in this study.41 Second, information on the amount of selenium from dietary intake or supplementation was not available, and this may have influenced the selenium level. Third, selection bias may have existed because we recruited participants by advertisement and our participants were relatively independent elderly. Finally, we were not able to establish the causal relationship between serum selenium and sarcopenia because of the cross-sectional design of the study.
Y.-L. Chen et al. / JAMDA 15 (2014) 807e811
Conclusions In conclusion, this study is the first to directly investigate the relationship of serum selenium and skeletal muscle mass in the elderly. The results showed that after adjusting for age, sex, lifestyle, and physical and metabolic factors, lower serum selenium levels were associated with a higher risk of low skeletal muscle mass. However, recommending selenium supplements should be carefully evaluated. The underlying mechanism of selenium on sarcopenia warrants further research. Acknowledgments The authors would like to thank Miss Wen-Chao Weng and Yi-Ju Chen for their work. References 1. Morley JE. Anorexia, sarcopenia, and aging. Nutrition 2001;17:660e663. 2. Thomas DR. Sarcopenia. Clin Geriatr Med 2010;26:331e346. 3. Jackson MJ. Strategies for reducing oxidative damage in ageing skeletal muscle. Adv Drug Deliv Rev 2009;61:1363e1368. 4. Jones TE, Stephenson KW, King JG, et al. SarcopeniadMechanisms and treatments. J Geriatr Phys Ther 2009;32:39e45. 5. Chaput JP, Lord C, Cloutier M, et al. Relationship between antioxidant intakes and class I sarcopenia in elderly men and women. J Nutr Health Aging 2007;11: 363e369. 6. Rederstorff M, Krol A, Lescure A. Understanding the importance of selenium and selenoproteins in muscle function. Cell Mol Life Sci 2006;63:52e59. 7. Lescure A, Rederstorff M, Krol A, et al. Selenoprotein function and muscle disease. Biochim Biophys Acta 2009;1790:1569e1574. 8. Beytut E, Karatas F. Lambs with white muscle disease and selenium content of soil and meadow hay in the region of Kars, Turkey. Vet J 2002;163:214e217. 9. Brown MR, Cohen HJ, Lyons JM, et al. Proximal muscle weakness and selenium deficiency associated with long term parenteral nutrition. Am J Clin Nutr 1986; 43:549e554. 10. Lauretani F, Semba RD, Bandinelli S, et al. Association of low plasma selenium concentrations with poor muscle strength in older community-dwelling adults: The InCHIANTI Study. Am J Clin Nutr 2007;86:347e352. 11. Beck J, Ferrucci L, Sun K, et al. Low serum selenium concentrations are associated with poor grip strength among older women living in the community. Biofactors 2007;29:37e44. 12. Yang KC, Lee LT, Lee YS, et al. Serum selenium concentration is associated with metabolic factors in the elderly: A cross-sectional study. Nutr Metab (Lond) 2010;7:38. 13. Guigoz Y, Lauque S, Vellas BJ. Identifying the elderly at risk for malnutrition. The Mini Nutritional Assessment. Clin Geriatr Med 2002;18:737e757. 14. Tsai AC, Ho CS, Chang MC. Assessing the prevalence of malnutrition with the Mini Nutritional Assessment (MNA) in a nationally representative sample of elderly Taiwanese. J Nutr Health Aging 2008;12:239e243. 15. Mahoney FI, Barthel DW. Functional evaluation: The Barthel Index. Md State Med J 1965;14:61e65. 16. Lawton MP, Brody EM. Assessment of older people: Self-maintaining and instrumental activities of daily living. Gerontologist 1969;9:179e186. 17. Huang KC. Obesity and its related diseases in Taiwan. Obes Rev 2008;9:32e34. 18. Wu CH, Yang KC, Chang HH, et al. Sarcopenia is related to increased risk for low bone mineral density. J Clin Densitom 2013;16:98e103. 19. Chien MY, Huang TY, Wu YT. Prevalence of sarcopenia estimated using a bioelectrical impedance analysis prediction equation in community-dwelling elderly people in Taiwan. J Am Geriatr Soc 2008;56:1710e1715.
811
20. Robinson MF, Campbell DR, Stewart RD, et al. Effect of daily supplements of selenium on patients with muscular complaints in Otago and Canterbury. N Z Med J 1981;93:289e292. 21. Arnaud J, de Lorgeril M, Akbaraly T, et al. Gender differences in copper, zinc and selenium status in diabetic-free metabolic syndrome European populationdThe IMMIDIET study. Nutr Metab Cardiovasc Dis 2012;22: 517e524. 22. Beavers KM, Beavers DP, Serra MC, et al. Low relative skeletal muscle mass indicative of sarcopenia is associated with elevations in serum uric acid levels: Findings from NHANES III. J Nutr Health Aging 2009;13:177e182. 23. Chung JY, Kang HT, Lee DC, et al. Body composition and its association with cardiometabolic risk factors in the elderly: A focus on sarcopenic obesity. Arch Gerontol Geriatr 2013;56:270e278. 24. Seale LA, Hashimoto AC, Kurokawa S, et al. Disruption of the selenocysteine lyase-mediated selenium recycling pathway leads to metabolic syndrome in mice. Mol Cell Biol 2012;32:4141e4154. 25. Ferreiro A, Quijano-Roy S, Pichereau C, et al. Mutations of the selenoprotein N gene, which is implicated in rigid spine muscular dystrophy, cause the classical phenotype of multiminicore disease: Reassessing the nosology of early-onset myopathies. Am J Hum Genet 2002;71:739e749. 26. Castets P, Lescure A, Guicheney P, Allamand V. Selenoprotein N in skeletal muscle: From diseases to function. J Mol Med (Berl) 2012;90: 1095e1107. 27. Jurynec MJ, Xia R, Mackrill JJ, et al. Selenoprotein N is required for ryanodine receptor calcium release channel activity in human and zebrafish muscle. Proc Natl Acad Sci U S A 2008;105:12485e12490. 28. Loflin J, Lopez N, Whanger PD, Kioussi C. Selenoprotein W during development and oxidative stress. J Inorg Biochem 2006;100:1679e1684. 29. Speckmann B, Walter PL, Alili L, et al. Selenoprotein P expression is controlled through interaction of the coactivator PGC-1alpha with FoxO1a and hepatocyte nuclear factor 4alpha transcription factors. Hepatology 2008;48:1998e2006. 30. Wenz T, Rossi SG, Rotundo RL, et al. Increased muscle PGC-1alpha expression protects from sarcopenia and metabolic disease during aging. Proc Natl Acad Sci U S A 2009;106:20405e20410. 31. D’Amico A, Haliloglu G, Richard P, et al. Two patients with ’dropped head syndrome’ due to mutations in LMNA or SEPN1 genes. Neuromuscul Disord 2005;15:521e524. 32. Ferreiro A, Ceuterick-de Groote C, Marks JJ, et al. Desmin-related myopathy with Mallory body-like inclusions is caused by mutations of the selenoprotein N gene. Ann Neurol 2004;55:676e686. 33. Jackson MJ. Skeletal muscle aging: Role of reactive oxygen species. Crit Care Med 2009;37:S368eS371. 34. Walston J, Xue Q, Semba RD, et al. Serum antioxidants, inflammation, and total mortality in older women. Am J Epidemiol 2006;163:18e26. 35. Marzani B, Balage M, Venien A, et al. Antioxidant supplementation restores defective leucine stimulation of protein synthesis in skeletal muscle from old rats. J Nutr 2008;138:2205e2211. 36. Mosoni L, Balage M, Vazeille E, et al. Antioxidant supplementation had positive effects in old rat muscle, but through better oxidative status in other organs. Nutrition 2010;26:1157e1162. 37. Palazzetti S, Rousseau AS, Richard MJ, et al. Antioxidant supplementation preserves antioxidant response in physical training and low antioxidant intake. Br J Nutr 2004;91:91e100. 38. Peternelj TT, Coombes JS. Antioxidant supplementation during exercise training: Beneficial or detrimental? Sports Med 2011;41:1043e1069. 39. Stranges S, Marshall JR, Natarajan R, et al. Effects of long-term selenium supplementation on the incidence of type 2 diabetes: A randomized trial. Ann Intern Med 2007;147:217e223. 40. Stranges S, Navas-Acien A, Rayman MP, Guallar E. Selenium status and cardiometabolic health: State of the evidence. Nutr Metab Cardiovasc Dis 2010; 20:754e760. 41. Cruz-Jentoft AJ, Baeyens JP, Bauer JM, et al. Sarcopenia: European consensus on definition and diagnosis: Report of the European Working Group on sarcopenia in older people. Age Ageing 2010;39:412e423.