Sarcopenia and Its Implications for the Elderly{

Sarcopenia and its implications for the elderly{

R Roubenoff1*

1Nutrition, Exercise Physiology, and Sarcopenia (NEPS) Laboratory, Jean Mayer USDA Human Nutrition Research Center on Aging at Tuffs University, Boston, MA, USA

Sarcopenia is the loss of muscle mass and strength with age. Sarcopenia is a part of normal aging, and occurs even in master athletes, although it is clearly accelerated by physical inactivity. Sarcopenia contributes to disability, reduced ability to cope with the stress of a major illness, and to mortality in the elderly. The etiology of sarcopenia is unclear, but several important factors have been identi®ed. These include loss of alpha motor neurons, decline in muscle cell contractility, and several potential humoral factors, such as androgen and estrogen withdrawal and increase in production of catabolic cytokines. Treatment of sarcopenia with progressive resistance training is safe and effective, but dissemination of this technique to the general population has yet to occur. As the number of elderly persons increases exponentially in the new century, a public health approach to prevention and treatment of sarcopenia, based on increasing physical activity at all ages, will be crucial to avoiding an epidemic of disability in the future. Descriptors: aging; sarcopenia European Journal of Clinical Nutrition (2000) 54, S40±S47

De®ning sarcopenia hypertrophy, at least in the ®rst few months of training. For example, Figure 2 shows the strength and vastus lateralis Sarcopenia is the loss of muscle mass and strength caused area (measured by CT scan) lost over a 12 y follow-up by aging (Rosenberg, 1989). It is distinct from muscle loss period in seven healthy men, and the amount gained during (cachexia) caused by in¯ammatory disease, or from the a 12 week period of intensive resistance training (Frontera weight loss and attendant muscle wasting caused by starva- et al, 2000). It is clear that there is hysteresis in this system: tion or advanced disease (Roubenoff et al, 1997a). As be®ts the decline and the regain occur at vastly different rates. a truly age-driven phenomenon, it is universal with Thus, de®ning sarcopenia solely on compositional terms advanced age. That is, reduced muscle mass and strength may be useful and attractive for research purposes, but it are evident in all elderly persons compared to young, may be simplistic in explaining functional changes caused healthy, physically active young adults. If the sarcopenia by treatment of sarcopenia. progresses beyond a threshold of functional requirements, it Second, there is reasonable evidence that there is a limit leads to disability and frailty, and this can occur indepen- on how much lean body mass can be lost before death dently of any disease-induced frailty (Figure 1). Of course, supervenes. The available data, based on starvation superimposed illness will accelerate the loss of muscle (Winick, 1979), AIDS patients (Kotler et al, 1989), and mass, and thus increase the risk of disability, frailty, and critical illness (Tellado et al, 1989), suggest that loss of death. more than about 40% of baseline lean mass is fatal. There is no absolute level of lean mass, body cell mass, `Baseline' is a slippery concept here, because again abso- or muscle mass at which one can de®nitely say that lute mass is not explanatory Ð basketball players do not sarcopenia is present. Such a de®nition would be an necessarily outlive jockeys Ð but rather the amount of loss important advance. In reaching such a de®nition, one as a function of the baseline mass that the individual started should consider two important and generally agreed-upon with. Reference Man and Woman are one benchmark, concepts in relation to lean body mass. First, there is a based on a few cadaver studies in generally healthy persons direct structure ± function link between muscle mass and (Ellis, 1990). Kehayias et al (1997) de®ned baseline as the strength, in that more muscle generally equals greater mean for adults aged 20 ± 30 y; no healthy subjects were strength and vice versa. However, the function de®ning found below approximately 70% of that standard, and there the relationship between muscle loss and strength loss is was a steady decline in body cell mass for both men and not the same as that applying to muscle and strength gain. women across age groups between 30 and 100 y (Figure 3). Pharmacologic interventions such as growth hormone or The latter point also raises the issue of the importance of testosterone, which increase lean mass (and evidently sarcopenia as an indicator of reduced protein stores for muscle mass as well) do not alter strength much, while times of stress. It is well accepted that during illness, progressive resistance training, which causes large gluconeogenesis increases in importance, while ketogenesis increases in strength, can do so with little evident muscle is relatively suppressed, so that protein is burned for energy in excess of the levels seen in starvation adaptation. Given *Correspondence: R Roubenoff, Nutrition, NEPS Laboratory, Human the anorexia caused by acute illness, and by the iatrogenic Nutrition Research Center on Aging, 711 Washington Street, Boston, MA limitation on dietary intake that often obtains in hospitals, 02111, USA. endogenous protein stores are crucial in determining the E-mail: [email protected] availability of metabolic substrate to cope with the illness, {The contents of this publication do not necessarily re¯ect the views or and thus the ability to survive it. Therefore, it is no wonder policies of the US Department of Agriculture nor does mention of trade names, commercial products, or organizations imply endorsement by the that elderly, sarcopenic patients fare worse than young, US Government. healthy adults for almost all diseases. Tellado et al (1989) Sarcopenia R Roubenoff S41

Figure 1 Relationship between sarcopenia, frailty, and medical outcomes. (From Roubenoff & Harris, 1997c). have shown that measurement of body cell mass was the only independent determinant of survival in intensive care unit patients in multivariate analysis, removing the signi®- cance of univariate predictors such as albumin, age, and even diagnosis. Thus, the metabolic signi®cance of sarcopenia in illness should be considered independently of its Figure 3 Body cell mass, measured as total body potassium (TBK) functional impact during times of better health, as both are per kg body weight, as a function of age in a cross-sectional study (from important to the survival and well-being of elderly persons. Kehayias et al, 1997). Data are expressed as a percentage of the reference 20 ± 30-y-old groups for men (s) and women (6) separately. (From Kehayias et al, 1997).

Epidemiology of sarcopenia muscle mass 2 or more standard deviations below the There in one population-based study of the prevalence of mean for young healthy participants in the Rosetta Study, sarcopenia with advancing age. Data are available from the a large cross-sectional study of body composition in New New Mexico Elder Health Survey, by Baumgartner et al York. The prevalence of sarcopenia by this de®nition (1998), who measured appendicular muscle mass by dual- increased from 13 ± 24% of persons under age 70 to over energy X-ray absorptiometry (DXA) in 883 elderly Hispa- 50% of those over age 80 y. nic and non-Hispanic White men and women. The subjects Kehayias et al (1997) found that the quality of the lean were selected randomly from the Health Care Financing body mass, de®ned as the ratio of cell mass (measured by Administration (HCFA) Medicare listing for Bernalillo whole body potassium counting) to lean mass (measured by County, New Mexico. A total of 2200 subjects were hydrodensitometry or neutron activation), declined with sampled; 534 had died, moved, could not be contacted, or age in a cross sectional analysis. These data suggest that were ineligible. Of the 1666 eligible subjects contacted, sarcopenia is universal, and indeed this would be consistent 1130 (67.8%) completed the home interview and 883 with an age-related phenomenon, and complement the data (53%) underwent DXA. Sarcopenia was de®ned as a of Baumgartner et al, where a cutoff was used to de®ne

Figure 2 Change in strength and vastus lateralis area (measured by CT scan) over a 12 y follow-up period in 7 healthy men, and the amount gained during a 12 week period of intensive resistance training in the same men. (From Frontera et al, 2000).

European Journal of Clinical Nutrition Sarcopenia R Roubenoff S42 sarcopenia. Gallagher et al (1997) have shown that the ®bers. The capillary=®ber ratio was also signi®cantly cross-sectional change in appendicular muscle mass mea- reduced (1.39 vs 1.08, P < 0.05). These data indicate that sured by DXA parallels the decline in total body cell mass longitudinal change in the amount of muscle is strongly measured by total body potassium (TBK), the reference linked to change in strength in healthy elderly men. method for this compartment. Cross-sectional studies of muscle ®ber distribution have suggested that the loss of muscle is largely con®ned to type II ®bers, the fast twitch Impact of sarcopenia cells that are most responsible for strength and anaerobic, short-term movements (Lexell, 1995). There is a strong direct relationship between muscle mass However, one cannot easily come to conclusions about and strength. Thus, sarcopenic persons are weaker than the phenomenology of sarcopenia based on cross-sectional persons with normal muscle mass. In reality, there may be a data. Longitudinal data are harder to come by, but a few feedback loop between muscle mass and function that can studies do exist. Forbes and Reina (1970) reported long- be driven in either a positive (healthy) or a negative itudinal change in total body potassium in 18 men and two (disabling) direction. In the positive direction, people women aged 22 ± 53 (mean 33 y), followed for a mean of who are ®t tend to be physically active, and people with 7.8 y. There was a decline in lean mass of 0.25 kg=y over chronic diseases who enter an exercise program demon- this time span. Flynn et al (1989) measured body cell mass strate an increase in their physical activity outside of the by TBK at 2 y intervals over 18 y in a group of 564 male training program. Studies from our laboratory have pre- and 61 female faculty employees of the University of viously shown this for adults with rheumatoid arthritis (Rall Missouri. Lean mass (calculated from TBK) declined in et al, 1996a, b), HIV infection (Roubenoff et al, 1997b), as the men who were ®rst measured after age 40, with a well as for frail elderly nursing home residents (Fiatarone median loss of 0.3 kg=y. No signi®cant change was seen in et al, 1994). In the negative direction, as people become women, although this is likely to be due to their smaller weaker, either because of disease or because of age-related sample size, as the point estimates for change were nega- sarcopenia, the proportion of maximal effort required to tive in this group after age 50. These observations raise the perform daily tasks increases, so that it becomes progres- question of whether menopause accelerates sarcopenia as it sively uncomfortable to perform these tasks, and they are does osteopenia. A small study by Poehlman et al suggests abandoned. While both cardiopulmonary ®tness and muscle that this is indeed the case, as women who experienced strength are important determinants of functional capacity, menopause had a much greater decline in lean body mass, in frail, elderly persons with advanced sarcopenia but gain in fat mass, and drop in metabolic rate than did women without heart failure or emphysema, muscle weakness of similar ages who continued to have menses (Poehlman may be more limiting than aerobic ®tness (Shephard, et al, 1995). 1987). Weakness in turn leads to further disuse, as people As noted earlier, these data all de®ne sarcopenia in terms avoid activities that are uncomfortable. As shown in Figure of its compositional, rather than functional (strength) 4, the level of muscular effort required to rise from a chair aspect. However, there are data suggesting that the decline that is well within the capacity of a young, ®t person, can in strength with age exceeds the decline in lean mass. be at the maximal capacity of an elderly, sarcopenic person. Kallman et al (1990), using the Baltimore Longitudinal Thus, reduced physical activity follows loss of muscle Study on Aging data, showed that older men have a weaker mass, and then accelerates it by removing the trophic grip strength, and young men have a stronger grip, than stimulus of the activity. The improved survival and reduced would be predicted by arm muscle mass alone. This ®nding disability of elderly athletes who remain physically active may be partly limited by reliance on anthropometric mea- suggest that such a vicious cycle is avoidable under some sures of muscle mass, which are relatively imprecise; data circumstances (Paffenbarger et al, 1986, 1993). More using CT scans are discussed below. Nevertheless, 71% of importantly, perhaps, the ability to reverse these changes men aged 40 ± 59 and 85% of men over age 60 had a with PRT suggests that they are modi®able effects of aging. decline in strength over a 9 y follow-up, again suggesting The New Mexico study (Baumgartner et al, 1998) gives the universal nature of sarcopenia at the functional as well an idea of the relationship between sarcopenia and func- as structural level. tional status. Sarcopenic women had 3.6 times higher rates Other studies of change in muscle strength over time have shown declines (Bassey & Harries, 1993; Rantanen et al, 1997), no change (Greig et al, 1993), or even gains (Bassey & Harries, 1993; Rantanen et al, 1997) in strength over 4 ± 25 y. A reduction in type II muscle ®bers has been shown in some studies, but no change was found in others. Recently Frontera et al (1999) examined changes in thigh skeletal muscle mass by CT scanning and strength by isokinetic dynamometry in nine healthy men studied on two occasions 12 y apart. Thigh extensor and ¯exor cross- sectional area declined by 16.1% and 14.9%, respectively, while strength fell by 20 ± 30%. However, speci®c force (strength=cross sectional area) did not change, and loss of muscle explained 50% of the variance in strength loss in the knee extensors. There was also a decline in the Figure 4 Relationship between maximal voluntary strength and ability to perform a task such as rising from a chair, for a healthy young person and a percentage of type I ®bers on histologic examination of sarcopenic elderly person of the same weight. The force required is the biopsies of the vastus lateralis muscle (from 60% to 42%), same for both persons, but the dif®culty is much greater or impossible for but no change in the mean area of either type I or type II the older one (from Frontera & Meredith, 1995).

European Journal of Clinical Nutrition Sarcopenia R Roubenoff S43 of disability, and men 4.1 times higher rates, compared to respond to training by increasing neuronal ®ring rates, study participants with normal muscle mass. There were improving the recruitment of motor units in response to signi®cantly greater risks of use of cane or walker, and a the signal to contract a muscle, and by increasing the history of falling, in the sarcopenic subjects as well. These innervation of muscle over time. Additional evidence for odds ratios were signi®cant after adjustment for age, race, the importance of the CNS in determining strength (If not obesity, income, alcohol intake, physical activity, current muscle mass) comes from the cross-education effect seen in smoking, and comorbidity. Thus, sarcopenia is indepen- contralateral muscles when only one leg or arm is trained. dently associated with important health outcomes and The increase in strength on the untrained side is often as disabilities in this relatively healthy, ambulatory much as half of that seen on the exercised side (Sale, 1988). population. Even imagining exercise can signi®cantly increase strength (Herbert et al, 1998). Finally, the speci®city of training is such that the greatest strength gains occur at or near the Mechanisms underlying sarcopenia angular velocity and joint angle at which training is done: It is simplistic to consider sarcopenia to have a single strength gains fall off markedly above and below this target cause. We have approached the problem as a multifactorial (Fleck & Kraemer, 1997). one with a range of possible explanations, in the expecta- tion that several, if not many, factors will turn out to be Intrinsic changes in muscle contractility and muscle cell important when sarcopenia is better understood. Figure 5 biology shows our current understanding of important etiologic Data collected in the NEPS laboratory in collaboration with factors. Several have been addressed over the past few Dr. Walter Frontera at Spaulding Rehabilitation Hospital years by studies and have thus risen or fallen in terms of indicate that there is a decline in the speci®c force produced their relative importance. by single muscle cells from elderly adults (`muscle quality'), in addition to the decline in number of muscle cells with age (`muscle mass'). In addition, data published Central nervous system by the laboratory over the past two years suggest that Our current concept is that loss of alpha motor units from growth hormone (GH) de®ciency is not a major feature the spinal cord is a critical feature of sarcopenia that of sarcopenia, since 24 h secretion of GH is higher in differentiates it from cachexia or wasting. Loss of neurons sarcopenic women than in women with normal muscle is a continuous process throughout life, and is currently mass. However, the other potential etiologic factors have considered irreversible. Older adults have larger motor not been studied in detail until recently. units than young adults, as the dropout of neurons and The skinned ®ber preparation technique allows the muscle cells is partially compensated for by the `adoption' investigation of the function of muscle proteins in a cell of muscle ®bers by surviving neurons. These motor units with an intact myo®lament lattice. It avoids the confound- are less ef®cient as a result, and in the extreme case can ing effects of the intercellular connective tissue, protein cause tremor as well as weakness (Enoka, 1997). Clearly heterogeneity between cells found in multicellular prepara- neuronal death can and does lead to muscle atrophy, as tions (Wood et al, 1975), and incomplete activation of noted in strokes and peripheral neuropathic conditions. It is motor units. Since the ®bers are permeabilized (`chemically highly likely that this change also occurs in sarcopenia, skinned') and the composition of the solutions bathing the although it is also possible that the primary event is loss of contractile proteins are carefully controlled, any differences muscle ®bers, with subsequent `dying back' of the neuron. in contractile performance should be attributable to differ- The change in strength in response to exercise which ences in the contractile apparatus of the cell. The usefulness occurs in the ®rst few weeks after beginning a training of this technique for studying the relationships between the program is thought to occur via changes in CNS, output to contractile properties of human muscle ®ber segments the motor units, as there is no change in muscle mass obtained with the biopsy technique and the molecular during this time (Akima et al, 1999). The CNS may protein composition of the ®ber segment in normal human muscle tissue has been documented (Fitts et al, 1989; Larsson & Moss, 1993). A handful of studies of the contractile properties and biochemical composition of human single muscle ®ber have been published (Fitts et al, 1989; Larsson & Moss, 1993; Harridge et al, 1995; Lankford et al, 1995; Larsson et al, 1995, 1996; Larsson & Frontera, 1997; Widrick et al, 1996, 1997; Galler et al, 1997). However, as noted above, the cellular mechanisms underlying muscle weakness and atrophy in aging humans have not been elucidated. Further, none of the above-mentioned studies examined the relationships among functional performance, muscle structure and function, and contractile properties and biochemical composition of single muscle ®bers in the same subjects. Thus, it is not known to what extent structural and physiological properties of single cells correlate with whole muscle measurements and activities of dally living. Only two studies (Larsson & Moss, 1993; Larsson et al, 1995) have included women (one in one study and three in Figure 5 Potentially important factors in the etiology of sarcopenia. another) and, to our knowledge, only one group of

European Journal of Clinical Nutrition Sarcopenia R Roubenoff S44 researchers included any elderly subjects, a total of only strength training can counteract the age-related changes in four men (Larsson et al, 1997). Four of the studies men- function and morphology of skeletal muscle. tioned above (Larsson & Moss, 1993; Harridge et al, 1995; Larsson et al, 1996; Larsson & Frontera, 1997) have Humoral factors examined single ®bers from the vastus lateralis muscle, Many hormones and cytokines have metabolic effects that but only one included women and none elderly women. can alter muscle mass and function. Anabolic hormones This is particularly relevant since most of the power needed such as growth hormone (GH), testosterone, and estrogen to perform important activities of daily living such as rising are reduced with age, and the withdrawal of their trophic from a chair is generated by the knee extensors (Wreten- effects may lead to muscle atrophy and sarcopenia. As berg & Arborelius, 1994). Three studies have examined the mentioned earlier, there is evidence of accelerated loss of effects of physical activity levels or exercise training on lean mass (and gain in fat) around menopause, suggesting force production, maximal shortening velocity, and the that estrogen could have a role in supporting muscle mass expression of myosin heavy and light chains in single (Poehlman et al, 1995). In addition, estrogen and testoster- muscle ®bers but again only one (Larsson & Frontera, one have been shown to suppress catabolic cytokine 1997) included elderly subjects. In that study, however, expression in animal models, and may have an indirect the sample size (two physically active and two sedentary role in modulating subclinical in¯ammation in aging (see men) was small and the results must be interpreted with below). caution. The type of exercise in two of the three studies Because the decline in GH begins in the fourth decade of (Fitts et al, 1989; Widrick et al, 1996) was aerobic (swim- life and parallels the decline in lean mass, it is intuitively ming or running) and of the two active subjects in the third appealing that loss of GH could act as a central coordinat- study (Larsson & Frontera 1997), only one was involved in ing mechanism driving the loss of muscle in aging. If this is regular strength training sessions. These studies demon- so, it follows that people with lower GH secretion should strate the response of muscle ®bers to various types of have lower lean body mass, and that this relationship aerobic exercise training by the adaptability of shortening should extend into middle and old age. However, we velocity (Fitts et al, 1989; Widrick et al, 1996) and have recently shown that this is not the case: postmeno- maximal force (Widrick et al, 1996). Further studies are pausal women with the lowest 24-hour GH levels had the needed to de®ne the extent to which single muscle ®bers highest body cell mass, not the lowest (Roubenoff et al, adapt to strength training, and whether this adaptation is 1998b). In fact, the relationship between GH and lean mass modi®ed by age. Finally, the relationship between these is highly confounded by fat mass, and there is some adaptations and changes in whole muscles and functional evidence that leptin suppresses GH production (see tasks important for independent living must be evaluated. Figure 6). In addition, we have now shown that patients Several researchers have examined the myosin heavy with rheumatoid arthritis, a model for in¯ammatory chain composition of human single muscle ®bers in healthy cachexia and muscle loss, do not have lower GH levels sedentary and trained young and elderly subjects without than age- and BMI-matched healthy controls (Rall LC & including measurements of contractile properties (Klitgaard Roubenoff R, unpublished observations). As a result, it et al, 1990a ± c; Andersen et al, 1994). In general, the appears that GH is not central to the development of results have been con¯icting. Changes with aging in sarcopenia. myosin heavy chain composition determined by gel elec- Activation of the immune system leads to production of trophoresis are controversial. Very sedentary young men in¯ammatory cytokines such as interleukin-1b (IL-1), have been reported to have a high proportion of ®bers co- tumor necrosis factor (TNF), and interleukin-6 (IL-6), expressing type IIa and IIb MHC. This is similar to ®ndings which can cause amino acid export from muscle (Baracos in the m. vastus lateralis of older (68 ± 70 y) subjects et al, 1983; Clowes et al, 1983; Tsujinaka et al, 1996). Such (Klitgaard et al, 1990c) suggesting that low physical a situation has been clearly demonstrated for acute illness activity contributes to the alterations in the biochemical and for chronic in¯ammatory conditions such as rheuma- composition of human single ®bers. However, a high toid arthritis and HIV infection (Roubenoff et al, 1994). proportion of ®bers co-expressing type I and IIa myosin There is moderately strong evidence that aging is asso- heavy chain isoforms in older subjects was also observed ciated with immune senescence that includes loosening of and others have shown an increase in type I myosin heavy the tolerance which allows a healthy immune system to chain in older men (Klitgaard et al, 1990c). The reasons for correctly identify `self' from `non-self' antigens and avoid these con¯icting observations are not clear. No studies have targeting self antigens. For example, autoimmune phenom- examined the effect of strength training concurrently on ena such as production of rheumatoid factors (antibodies both the contractile properties and biochemical composi- against other antibodies) and anti-nuclear antibodies tion of single muscle ®bers in older subjects. increase with age, as does the prevalence of most auto- Klitgaard et al (1990c) used a cross-sectional design to immune disease (Vaughan, 1980). study protein expression of aging skeletal muscle in elderly These changes suggest, but do not prove, the theory that men with different training backgrounds including seven aging is associated with a subclinical in¯ammatory state, strength-trained subjects (68 Æ 0.8 y). They examined which leads to Increased production of catabolic cytokines muscle biopsy homogenates and reported a myosin heavy and with it an increased rate of muscle catabolism and thus chain isoform composition in the strength-trained older sarcopenia. We have shown (Figure 7) that, consistent with men similar to young controls, with a lower content of this idea, there is an increase in IL-6 and indirect evidence type I and a higher content of type II myosin heavy chain of increased IL-1 (measured as increased production of the than older controls and older runners. Since whole-muscle natural neutralizer of IL-1, IL-1 receptor antagonist) pro- cross-sectional area, speci®c tension, and maximal iso- duced by peripheral blood mononuclear cells from elderly metric torque were similar in the young controls and participants in the Framingham Heart Study compared with strength-trained older men, these authors concluded that young controls (Roubenoff et al, 1998a).

European Journal of Clinical Nutrition Sarcopenia R Roubenoff S45

Figure 6 Relationship between 24-hour GH secretion and lean body mass (a), and between 24 h GH and leptin secretion (b) (from Roubenoff et al, 1998b).

Lifestyle factors Figure 7 Production of interleukin-6 (IL-6) and interleukin-1b receptor Finally, there is clear evidence that sarcopenia is worsened antagonist (IL-1Ra) in young adults and in older adults further subdivided by their serum C-reactive protein (CRP), a marker of systemic in¯amma- with disuse, and that physical inactivity leads to faster and tion. IL-6 increases with age and in¯ammation, while IL-1Ra, an indirect greater muscle loss than seen in physically active elders. measure of IL-1 production, is affected by age only. Black bars indicate However, even master athletes develop sarcopenia, so it is unstimulated cells; shaded bars indicate stimulation in vitro with phyto- not completely prevented by exercise, and remains a result hemagglutinin for IL-6 and lipopolysaccharide for IL-1Ra (from Rouben- of aging and not purely of disuse. Nevertheless, the over- off et al, 1998a). whelming trend in industrial and post-industrial societies is toward a decline in physical activity with advancing age, and there is no doubt that this sedentary lifestyle promotes loss of muscle and gain in fat, with their attendant morbid- ity and mortality. tarone et al, 1990), persons with heart disease (Beniamini et al, 1997; Pu et al, 1999), arthritis (Rall et al, 1996a), and renal disease (Castaneda et al, 1998). The ability of the muscle to adapt to training is not lost with age. The key is to induce volitional failure in the muscle, to work it until it Treatment of sarcopenia fatigues. There is no longer any doubt that progressive resistance Pharmacologic treatment of sarcopenia has not been training (PRT) can increase muscle mass and strength at studied in as much detail as PRT, and remains controversial any age, and probably even in the face of rather advanced because of its expense and potential side effects. There is chronic disease. Data from the NEPS laboratory have also the concern about medicalization of sarcopenia as a shown that PRT can be performed in nonagenarians (Fia- disease requiring treatment. Recombinant GH has been

European Journal of Clinical Nutrition Sarcopenia R Roubenoff S46 shown to increase lean mass in elderly men with low- References normal insulin-like growth factor-1 levels (Rudman et al, Akima H, Takahashi H, Kuno S-Y, Masuda K, Masuda T, Shimojo H, 1990), but the cost is over US $20,000 per year. Similarly, Anno I, Itai y & Katsuta S (1999): Early phase adaptations of muscle testosterone and other anabolic steroids have been shown to use and strength to isokinetic training. Med. Sci. Sports Exerc. 31, increase lean mass in healthy young men and in patients 588 ± 594. with HIV infection (Grinspoon et al, 1998), but the poten- Andersen JL, Klitgaard H, Bangsbo J & Saltin B (1994): Myosin heavy chain isoforms in single ®bres from m. vastus lateralis of soccer players: tial side effects include liver damage, testicular atrophy, effects of strength-training. Acta Physiol. Scand. 150, 21 ± 26. and hyperlipidemia. With GH, there is also the concern that Baracos V, Rodemann HP, Dinarello CA & Goldberg AL (1983): the increase in mass is not accompanied by an increase in Stimulation of muscle protein degradation and prostaglandin E2 release strength (Yarasheski et al, 1995), so it is not clear whether by leukocytic pyrogen (interieukin-1). N. Engl. J. Med. 308, 553 ± 558. Bassey EJ & Harries UJ (1993): Normal value for handgrip strength in 920 GH actually increases muscle protein or whether much of men and women aged over 65 y, and longitudinal changes over 4 y in the lean gain is in the visceral compartment. 620 survivors. Clin. Sci. 84, 331 ± 337. Baumgartner RN, Koehler KM, Gallagher D, Romero L, Heyms®eld SB, Ross RR, Garry PJ & Lindeman RD (1998): Epidemiology of sarcopenia among the elderly in New Mexico. Am. J. Epidemiol. 147, Muscle adaptation to resistance training 755 ± 763. Klitgaard et al (1990b) used gel electrophoresis to examine Beniamini Y, Rubenstein JJ, Zaichkowsky LD & Crim MC (1997): Effects of high-intensity strength training on quality-of-life parameters in the myosin heavy chain composition of single ®bers from cardiac rehabilitation patients. Am. J. Cardiol. 80, 841 ± 846. the biceps brachii muscle in four young sedentary men and Castaneda C, Grossi L & Dwyer J (1998): Potential bene®ts of resistance four male body builders. Compared with sedentary men, exercise training on nutritional status in renal failure. J. Ren. Nutr. 8, body builders showed a higher proportion of ®bers expres- 2 ± 10. Clowes GHA, George BC, Villee CA & Saravis CA (1983): Muscle sing type IIa myosin heavy chain isoform and a lower proteolysis induced by a circulating peptide in patients with sepsis or proportion of ®bers expressing type IIb and ®bers co- trauma. N. Engl. J. Med. 308, 545 ± 552. expressing type IIa and IIb myosin heavy chain isoforms. Ellis KJ (1990): Reference man and woman more fully characterized: Their results suggest an altered expression of myosin heavy variations on the basis of body size, age, sex, and race. Biol. Trace chain isoforms after body building. It is important to note Elem. Res. 26 ± 27, 385 ± 400. Enoka R (1997): Neural strategies in the control of muscle force. Muscle that in this study traditional histochemistry methods did not Nerve 5(Suppl), S66 ± S69. show the differences detected with the more sensitive Fiatarone MA, Marks EC, Ryan ND, Meredith CN, Lipsitz LA & Evans electrophoretic techniques. On the other hand, Andersen WJ (1990): High-intensity strength training in nonagenarians. Effects et al (1994) looked at the effects of strength training on the on skeletal muscle. JAMA 263, 3029 ± 3034. Fiatarone MA, O'Neill EF, Ryan ND, Clements KM, Solares GR, Nelson expression of myosin heavy chain isoforms in single ®bers ME, Roberts SB, Kehayias JJ, Lipsitz LA & Evans WJ (1994): Exercise from young soccer players. A modest increase in strength training and nutritional supplementation for physical frailty in very was not accompanied by changes in the myosin heavy elderly people. N. Engl. J. Med. 330, 1769 ± 1775. chain composition of single ®bers. The type of exercise Fitts RH, Costill DL & Gardetto PR (1989): Effect of swim training (strengthening vs endurance) may be a determinant exercise training on human muscle ®ber function. J. Appl. Physiol. 66, 465 ± 475. of the response since endurance athletes have a higher Fleck SJ & Kraemer WJ (1997): Designing Resistance Training Programs, proportion of ®bers co-expressing type I and IIa myosin 2nd edn. Champaign, IL: Human Kinetics. heavy chain isoforms with a major amount of type I Flynn MA, Nolph GB, Baker AS, Martin WM & Krause G (1989): Total isoform (Klitgaard et al, 1990a). body potassium in aging humans: a longitudinal study. Am. J. Clin. Nutr. 50, 713 ± 717. Forbes GB & Reina JC (1970): Adult lean body mass declines with age: some longitudinal observations. Metabolism 19, 653 ± 663. Frontera WR, Meredith CN (1995): Exercise rehabilitation of the elderly. Conclusion In: Rehabilitation of the Aging and Elderly Patient, ed. G Felsenthal, S Garrison & FU Steinberg, pp 35 ± 45. Baltimore, MD: Williams & Sarcopenia is an important consequence of aging, which is Wilkins. associated with loss of strength, decreased protein reserves, Frontera WR, Hughes VA, Fielding RA, Fiatarone MA, Evans WJ & Roubenoff R (2000): Aging of skeletal muscle: a 12-year longitudinal and increased disability. There are many potential mechan- study. J. Appl. Physiol. (in press). isms leading to sarcopenia, but the most important ones are Gallagher D, Visser M, De Meersman RE, Sepulveda D, Baumgartner RN, probably central nervous system decline, intrinsic loss of Pierson RN, Harris T & Heyms®eld SB (1997): Appendicular skeletal muscle contractile function, and possibly humoral effects of muscle mass: effects of age, gender, and ethnicity. J. Appl. Physiol. 83, 229 ± 239. loss of gonadal steroids and increase in catabolic mediators Galler S, Hilber K & Pette D (1997): Stretch activation and myosin heavy produced by the immune system. As is often the case in chain isoforms of rat, rabbit, and human skeletal muscle ®bers. biomedicine, we know more about the treatment of sarco- J. Muscle Res. Cell Motil. 18, 441 ± 448. penia than about its etiology. Thus, it is clear that PRT can Greig CA, Botella J & Young A (1993): The quadriceps strength reverse and at least partially prevent sarcopenia. As the of healthy elderly people remeasured after 8 y. Muscle Nerve 16, 6 ± 10. global population of elderly persons grows rapidly in the Grinspoon S, Corcoran C, Askari H, Schoenfeld D, Wolf L, Burrows B, next few decades, there is an urgent need to translate the Walsh M, Hayden D, Parlman K, Anderson E, Basgoz N & Klibanski A current research knowledge into public exercise and phy- (1998): Effects of androgen administration in men with the AIDS sical activity programs that can prevent an epidemic of wasting syndrome: a randomized, double-blind, placebo-controlled trial. Ann. Intern. Med. 129, 18 ± 26. sarcopenia-related disability from causing catastrophic Harridge SDR, Bottinelli R, Reggiani C, Pellegrino MA, Canepari M & health and societal costs in the 21st century. Saltin B (1995): Similar maximum velocity of shortening in single ®bres expressing the same myosin heavy chain, but from different Acknowledgements ÐThis work was supported by USDA Cooperative human skeletal muscles. J. Physiol. 487, 152 ± 158. Agreement 58-1950-9-001 and NIH grants AG15797 and DK45734. The Herbert RD, Dean C & Gandevia SC (1998): Effect of real and imagined training on voluntary muscle activation during maximal isometric author thanks Walter Frontera, and Virginia Hughes, for their thoughtful contractions. Acta Physiol. Scand. 163, 361 ± 368. contributions.

European Journal of Clinical Nutrition Sarcopenia R Roubenoff S47 Kallman DA, Plato CC & Tobin JD (1990): The role of muscle loss in the Rosenberg IH (1989): Summary comments. Am. J. Clin. Nutr. 50(Suppl), age-related decline of grip strength: cross-sectional and longitudinal 1231 ± 1233. perspectives. J. Gerontol. 45, M82 ± M88. Roubenoff R, Roubenoff RA, Ward L, Holland S, Stevens M & Hellmann Kehayias JJ, Fiatarone MF, Zhuang H & Roubenoff R (1997): Total body D (1990): Rheumatoid cachexia II: body composition and tumor potassium and body fat: relevance to aging. Am. J. Clin. Nutr. 66, necrosis factor in rheumatoid arthritis. Clin. Res. 38, 757A (Abstract). 904 ± 910. Roubenoff R, Roubenoff RA, Cannon JG, Kehayias JJ, Zhuang H, Klitgaard H, Bergman O, Betto R, Salviati G, Schiaf®no S, Clausen T & Dawson-Hughes B, Dinarello CA & Rosenberg IH (1994): Rheumatoid Saltin B (1990a): Co-existence of myosin heavy chain I and IIa cachexia: cytokine-driven hypermetabolism and loss of lean body mass isoforms in human skeletal muscle ®bres with endurance training. in chronic in¯ammation. J. Clin. Invest. 93, 2379 ± 2386. P¯ugers Arch.=Eur. J. Physiol. 416, 470 ± 472. Roubenoff R, Heyms®eld SB, Kehayias JJ, Cannon JG & Rosenberg IH Klitgaard H, Zhou M & Richter EA (1990b): Myosin heavy chain (1997a): Standardization of nomenclature of body composition in composition of single ®bres from m. biceps brachii of male body weight loss. Am. J. Clin. Nutr. 66, 192 ± 196. builders. Acta Physiol. Scand. 140, 175 ± 180. Roubenoff R, Suri J, Raymond J, Fauntleroy J & Gorbach S (1997b): Klitgaard H, Zhou M, Schiaf®no S, Betto R, Salviati G & Saltin B (1990c): Feasibility of Increasing lean body mass in HIV-infected adults using Ageing alters the myosin heavy chain composition of single ®bres from progressive resistance exercise [abstract]. Nutrition 13, 271. human skeletal muscle. Acta Physiol. Scand. 140, 55 ± 62. Roubenoff R, Harris TB (1997c). Failure to thrive, sercopenia, and Kotler DP, Tierney AR & Pierson RN (1989): Magnitude of body cell functional decline in the elderly. In: Failure to Thrive in Older mass depletion and the timing of death from wasting in AIDS. Am. J. People. Clinics in Geriatric Medicine 13, 1 ± 10. Clin. Nutr. 50, 444 ± 447. Roubenoff R, Harris TB, Abad LW, Wilson PWF, Dallal GE & Dinarello Lankford EB, Epstein ND, Fananapazir L & Sweeney HL (1995): CA (1998a). Monocyte cytokine production in an elderly population: Abnormal contractile properties of muscle ®bers expressing beta- effect of age and in¯ammation. J. Gerontol. 53A, M20 ± M26. myosin heavy chain gene mutations in patients with hypertrophic Roubenoff R, Rall LC, Veldhuis JD, Kehayias JJ, Rosen C, Nicolson M, cardiomyopathy. J. Clin. Invest. 95, 1409 ± 1414. Lundgren N & Reichlin S (1998b): The relationship between growth Larsson L, Li X & Frontera WR (1997): Effects of aging on shortening hormone kinetics and sarcopenia in postmenopausal women: the role of velocity and myosin isoform composition in single human skeletal fat mass and leptin. J. Clin. Endocrinol. Metab. 83, 1502 ± 1506. muscle cells. Am. J. Physiol. 272, C638 ± C649. Rudman D, Feller AG, Nagraj HS, Gergans GA, Lalitha PY, Goldberg AF, Larsson L & Moss RL (1993): Maximum velocity of shortening in relation Schlenker RA, Cohn L, Rudman IW & Matison DE (1990): Effects of to myosin isoform composition in single ®bres from human skeletal human growth hormone in men over 60 y old. N. Engl. J. Med. 323, muscles. J. Physiol. 472, 595 ± 614. 1±6. Larsson L, Li X, Tollback A & Grimby L (1995): Contractile properties in Sale DG (1988): Neural adaptation to resistance training. Med. Sci. Sports single muscle ®bers from chronically overused motor units in relation to Exerc. 20(Suppl 5), S135 ± S145. motoneuron ®ring properties in prior polio patients. J. Neurol. Sci. 132, Shephard RJ (1987): Physical Activity and Aging, 2nd edn. Rockville, MD: 182 ± 192. Aspen. Larsson L, Li X, Berg HE & Frontera WR (1996): Effects of removal of Tellado JM, Garcia-Sabrido JL, Hanley JA, Shizgal HM & Christou NV weight-bearing function on contracility and myosin isoform composi- (1989): Predicting mortality based on body composition analysis. Ann. tion in single human skeletal muscle cells. P¯ugers Arch.=Eur. J. Surg. 208, 81 ± 87. Physiol. 462, 320 ± 328. Tsujinaka T, Fujita J, Ebisui C, Yano M, Kominami E, Suzuki K, Tanaka Lexell J (1995): Human aging, muscle mass, and ®ber type composition. K, Katsume A, Ohsugi Y, Shiozaki H & Monden M (1996): Interleukin J. Gerontol. 50A(Special issue), 11 ± 16. 6 receptor antibody inhibits muscle atrophy and modulates proteolytic Paffenbarger RS, Hyde RT, Wing AL & Hsieh C-C (1986): Physical systems in interleukin 6 transgenic mice. J. Clin. Invest. 97, 244 ± 249. activity, all-cause mortality, and longevity of college alumni. N. Engl. J. Vaughan JH (1980): Aging, immunity, and rheumatoid arthritis. In: Aging, Med. 314, 605 ± 613. Immunity, and Arthritis Diseases, ed. MMB Kay, J Galpin & T Paffenbarger RS, Hyde RT, Wing AL, Lee I-M, Jung DL & Kampert JP Makinodan. New York: Raven Press. (1993): The association of changes in physical-activity level and other Widrick JJ, Trappe SW, Blaser CA, Costill DL & Fitts RH (1996): lifestyle characteristics with mortality among men. N. Engl. J. Med. Isometric force and maximal shortening velocity of single muscle 328, 538 ± 545. ®bers from elite master runners. Am. J. Physiol. 271, C666 ± C675. Poehlman ET, Toth MJ & Gardner AW (1995): Changes in energy balance Widrick JJ, Romatowski JG, Karhanek M & Fitts RH (1997): Contractile and body composition at menopause: a controlled longitudinal study. properties of rat, rhesus monkey, and human type I muscle ®bers. Am. J. Ann. Intern. Med. 123, 673 ± 675. Physiol. 41, R34 ± R42. Pu CT, Johnson MT, Forman DE, Hausdorff JM, Fielding RA, Fiatarone- Winick M, ed. (1979): Hunger Disease Ð Studies by Jewish Physicians in Singh MA (1999): The effects of high-intensity strength training on the Warsaw Ghetto. New York: John Wiley & Sons. skeletal muscle and exercise performance in older women with heart Wood DS, Zollman J & Reuben JP (1975): Human skeletal muscle: failure: a randomized controlled trial (in press). properties of the ``chemically skinned'' ®ber. Science 187, 1075 ± 1076. Rall LC, Meydani SN, Kehayias JJ, Dawson-Hughes B & Roubenoff R Wretenberg P & Arborelius UP (1994): Power and work produced in (1996a): The effect of progressive resistance training in rheumatoid different leg muscle groups when rising from a chair. Eur. J. Appl. arthritis: increased strength without changes in energy balance or body Physiol. 68, 413 ± 417. composition. Arthr. Rheum. 39, 415 ± 426. Yarasheski KE, Zachwieja JJ, Campbell JA & Bier DA (1995): Effect of Rall LC, Rosen CJ, Dolnikowski G, Hartman WJ, Lundgren NT, Abad growth hormone and resistance exercise on muscle growth and strength LW, Dinarello CA & Roubenoff R (1996b): Protein metabolism and its in older men. Am. J. Physiol. 268, E268 ± 276. mediators before and after strength training in aging and chronic in¯ammation. Arthr. Rheum. 39, 1115 ± 1124. Rantanen T, Era P & Heikkinen E (1997): Physical activity and changes in maximal isometric strength in men and women from the age of 75 to 80 y. J. Am. Geriatr. Soc. 45, 1439 ± 1445.

European Journal of Clinical Nutrition