| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Research Communications |
1 From the University of Rochester School of Medicine and Dentistry, Rochester, NY.
Supported by NIH grants HD18454 and RR00044.
Address reprint requests to GB Forbes, University of Rochester School of Medicine and Dentistry, 601 Elmwood Avenue, Box 777, Rochester, NY 14642.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Objective: The goal was to analyze results of assays done in the author's laboratory of the FFM of normal adults studied at intervals for 
2 decades to ascertain longitudinal changes.
Design: 40K counting was used to estimate FFM in adult university personnel (15 men, 5 women) over periods ranging from 21 to 38 y. No advice was given about diet or exercise.
Results: There was considerable variation in the change of FFM over time. Some subjects lost FFM as the years went by, whereas others actually gained FFM. Analysis of the data showed that change in body weight was a prime factor in determining the magnitude and direction of the FFM change (R2 = 0.54). Adults who maintained their weight lost an average of 1.5 kg FFM per decade and so gained an equal amount of fat; those who lost weight lost even more FFM, whereas those who gained weight either gained FFM or lost it more slowly than the others. Data from the literature confirmed this trend.
Conclusions: FFM loss is not inevitable during adulthood—at least up to age 81 y, the oldest age yet studied. The magnitude and direction of the FFM change, be it positive or negative, is strongly influenced by change in body weight.
Key Words: Aging • body composition • fat-free mass • body fat • longitudinal observation • humans
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
It has been my good fortune to have had access to a whole-body scintillation counter since its construction by John B Hursh in 1959. This instrument records and quantitates the amount of naturally occurring 40K in the body, from which an estimate of FFM can be made (2). I have used this instrument to make frequent 40K assays of 2 men, 1 for 27 y, the other for 38 y; the latter also had several measurements of body density and body water. In addition, I have assayed 13 men and 5 women on less frequent occasions for periods ranging from 21 to 38 y. Some of these data were reported previously (3, 4).
The literature contains several reports by investigators who have studied adults for periods ranging from 8 to 22 y (5–11). Albert Behnke had 14 assays of body density over a 31-y period (12). A variety of subjects are represented in these reports: young adults, men and women >70 y of age, sedentary individuals, those who exercised regularly, and trained athletes. Changes in body weight ranged from -25 to 15 kg per decade; changes in FFM ranged from -14 to 7 kg per decade.
In considering my own data together with those reported by others, I propose to show the relation of the variability in FFM change with age to the time course of change in body weight. A preliminary report has been published (13).
|
SUBJECTS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
3 times background. Repeated assays of normal subjects yield a CV of 2.3%. The instrument is standardized at frequent intervals by counting a bottle containing 1.1 kg KCl. A description of the instrument and of its calibration for human subjects was published previously (14, 15). The potassium content of the body is taken to be 68.1 mmol/kg FFM in males and 64.2 mmol/kg FFM in females. Body density was estimated by 3 different methods: underwater weighing, water displacement, and by using the Bod-Pod instrument (LMI, Inc, Concord, CA) (16, 17). FFM was calculated with the formula [1 - (4.95/D - 4.50)] x W, where D is density and W is weight. Total body water (TBW) was estimated by deuterium or tritium dilution. FFM was calculated as TBW divided by 0.73, and again by 1.05 (2). Body weight was measured in subjects dressed in cotton pajamas and paper slippers.
The time course of change in body weight and FFM was analyzed by linear regression and statistical analysis was done by using SAS (version 6.09; SAS Institute Inc, Cary, NC). For those reports in which the subjects were listed in groups, regressions were calculated from weighted data. All changes in body weight and FFM are presented as changes per decade of life.
Subjects
My own data
The first male subject had 25 40K assays between the ages of 53 and 80 y, a span of 27 y. During this time, he enjoyed good health while slowly gaining weight and making periodic (and only temporarily successful) attempts to lose weight by exercising. He is still in good health at 91 y of age.
The second male subject had 120 assays from the age of 44 y to the age of 82 y, a span of 38 y. His health, while generally good, was interrupted on several occasions: a partial colon resection for carcinoma at age 64 y, a gastric ulcer at age 69 y, and an episode of gout at age 80 y. Then, at age 76 y there was an episode of arthralgia, muscle aches, and weight loss (FFM was also lost) that was treated with salicylates, followed by a fracture of the femur. The fracture was repaired, the other symptoms disappeared, and both weight and FFM increased but failed to reach their previous status by age 80 y. Worthy of mention was his response to giving up a life-long smoking habit at age 75 y; the result was a 3.8-kg gain in weight and a 2.4-kg increase in FFM that was corrected subsequently by diet. The second subject also had 6 assays of TBW from the ages of 50 to 81 y, and 5 assays of body density from the ages of 67 to 80 y.
Additional subjects were 13 men and 5 women recruited from university personnel: they had 2–14 40K assays each over periods ranging from 21 to 38 y. All subjects were white. None engaged in strenuous exercise and none was advised about diet. Three of the women were nulliparous; 1 had had 8 children before the first assay, and had a ninth when she was aged 44 y, 18 mo after the first assay. This woman breast-fed all of her children for several months, the youngest infant for 18 mo. One man had a myocardial infarction, treated by angioplasty, at age 60 y, 2 y before his last assay. One man had a nephrectomy for renal carcinoma at age 70 y, 3 y before his final assay; he also has mild arthritis of his hands. One man had symptoms of Alzheimer disease at the time of his last assay at age 74 y and died 3 y later. One man had a diagnosis of Lyme disease at age 77 y, 8 y before his last assay; he was also being treated with digitalis for paroxysmal atrial flutter. One man had moderate hypertension and died suddenly 4 y after his last assay. One woman had a herniated lumbar disc at age 56 y, and fractured her wrist and ankle at age 67 y (her final assay was at age 71 y). One woman had an episode of tuberculous pleurisy at age 59 y, for which she was on bed rest for several months and lost 7 kg body weight; her last assay was at age 62 y, by which time she had regained all of her lost weight. One woman sustained a fracture of the femur 1 y before her last assay. Among the 5 women, menopause occurred from 11 to 21 y before the last assay. Plots of FFM against age gave no evidence of an acceleration of FFM loss after age 60 y, so the entire age span was included in the regression analysis.
Data from the literature
Data from the literature are listed in Table 1
, along with my own data. Some authors reported individual data whereas others recorded only group averages. Albert Behnke (12) had repeated assays of his body density over 31 y. He exercised regularly and, on one occasion, intentionally lost weight by dieting, only to regain it later. The subjects reported by Chien et al (6) were natives of Taiwan who had experienced an increase in the available food supply between the first and second densitometric assays. Some of these subjects gained weight during the 12-y interval whereas others did not. One subject <20 y of age was omitted. These subjects also had measurements of blood volume. The subjects studied by Flynn et al (5) were listed in their publication by age and sex to yield 4 groups of men and 4 of women. The men studied by Keys et al (7) were designated as one group; weight and FFM means were reported at the beginning and end of the 19-y period, but individual changes in weight and FFM were not given. These subjects also had measurements of basal metabolic rate.
|
Steen et al (8) kindly provided individual data on their 8 male and 11 female subjects, all of whom were >70 y of age. Trappe et al (10) divided their male subjects into several groups on the basis of recorded physical activity; these groups ranged from young, "highly trained" athletes to middle-aged, sedentary adults.
Parizkova and Eiselt (11) noted that about one-third of their original group of elderly subjects died during the 8–10-y follow-up period, and about half of the survivors failed to cooperate. Those who continued in the study were described as "healthy and fit," although most were not able to undergo hydrostatic weighing at the end of the study. None of the other authors had anything to say about the health of their subjects.
The data of Borkan and Norris (18), collected from 111 men of various ages, showed FFM changes ranging from -7.3 to 6.4 kg FFM per decade as determined by TBW. However, the fact that the subjects listed for body weight change differed in number from those listed for FFM change made it impossible to include their data.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. Linear regression analysis produced slopes of 4.4 kg body wt/decade and -0.244 ± 0.42 kg FFM/decade; the respective SEEs are 2.70 and 1.57 kg. There was no indication of a tendency for FFM to decline more rapidly after age 65 y; indeed, the slope for the later years (0.617 ± 0.601 kg/decade) did not differ significantly from zero. This subject enjoyed good general health, spending a sabbatical year abroad at age 66 y.
|
. Linear regression analysis produced slopes of 0.36 kg body wt/decade and -1.19 ± 0.13 kg FFM/decade; the respective SEEs were 1.89 and 1.77 kg. Various life events mentioned earlier are noted in Figure 2. This subject also had several estimates of FFM by TBW (6 assays with deuterium and 2 with tritium between the ages of 50 and 80 y) and 5 assays by densitometry between the ages of 67 and 81 y. The calculated regression line for FFM by TBW had a slope of -1.0 ± 0.71 kg FFM/decade and an SEE of 2.31 kg; for FFM by densitometry the slope was -0.94 ± 3.3 kg FFM/decade, with an SEE of 4.4 kg (the large variability in the results of the density assays may have been due to the fact that 3 different assay techniques were used). However, the slopes of the regression lines for FFM estimates by 40K, TBW, and densitometry were similar.
|
). For the earlier period, the slope for weight was 0.177 ± 0.46 kg/decade and for the latter period it was 3.02 ± 0.49 kg/decade. The respective values for FFM change were -1.91 ± 0.46 and -2.5 ± 0.49 kg/decade. Although the downhill course of FFM appeared to be more rapid in later years, these 2 FFM slopes did not differ significantly.
There was considerable variability in the change in FFM with time for my 20 subjects as well as variability in the change in weight. Change in FFM ranged from -4.2 to 2.8 (
± SEM: -0.57 ± 1.85) kg/decade, and change in body weight ranged from -1.8 to 6.3 ( ± SEM: 1.87 ± 2.23) kg/decade. When these data were graphed, there was a positive relation between the 2 variables: individuals who either maintained or lost weight usually lost FFM, whereas those who gained weight as the years went by often gained FFM (Figure 3). The calculated regression equation is as follows:
|
| (1) |
|
). The average regression slope for the 6 studies from the literature was 0.46 (range: 0.15–0.68) and the average y-axis intercept was -1.23 (range: 0.3 to -2.8) kg FFM/decade. The values recorded for my 20 subjects, namely a slope of 0.61 and an intercept of -1.85 kg FFM/decade, fall within these respective ranges.
|
is a plot of changes in FFM against changes in weight for my own data plus reports from the literature that dealt with individual subjects (upper portion of Table 2). The calculated regression equation for all 75 subjects is as follows:
|
| (2) |
).
|
The situation is quantitatively somewhat different for the grouped data (lower half of Table 2), though the general trend remains the same. The slopes of the regression lines are positive in each instance, and all but one of the intercepts are negative. Unfortunately, grouped data do not lend themselves to the usual statistical tests. The overall equation is as follows:
|
| (3) |
Because 
fat = W - FFM, longitudinal changes in body fat content per unit weight change are described by the following equations. The average for the individual data (upper portion of Table 2) is as follows:
|
| (4) |
) it is as follows:
|
| (5) |
The existence of the y-axis intercept in the equations shown in Table 2 means that the fraction of weight change due to fat (or due to lean) varies with the degree of weight change. For example, for those individuals who lost 5 kg/decade, the ratio of fat loss to FFM loss was 3:2, whereas those who gained 5 kg/decade put on 6 times more fat than lean, and, as noted earlier, those who maintained their weight actually gained fat and lost an equivalent amount of lean.
There were several groups of subjects who were described by the authors as being physically active; namely, 2 groups reported by Parizkova and Eiselt (11), 3 groups reported by Pollock et al (9), and 2 groups reported by Trappe et al (10). The remaining 13 groups were assumed to be relatively inactive because the authors made no mention of physical activity in their reports. Separate regressions for these 2 categories of subjects showed no difference in the regression slopes. The physically active groups had a higher intercept (-0.44 kg/decade) than the inactive groups (-2.54 kg/decade), but the difference was not significant. However, the observations of Pollock et al (9), who studied athletes for 20 y, showed that regular exercise may not prevent the long-term loss of FFM. Two of their groups of athletes who maintained their body weights lost 1.4 and 1.9 kg FFM/decade, values similar to those for sedentary individuals.
With regard to the effect of aging on the time course of change in FFM, it was noted earlier that the 2 subjects portrayed in Figures 1 and 2 showed no indication of an acceleration in FFM loss after age 65 y. There are 2 reports in the literature that deal with older individuals: Steen et al's (8) subjects were 70 y old at the time of the initial assay of body composition and those of Parizkova and Eiselt (11) were all >60 y (Tables 1 and 2). In both studies there was a positive relation between change in FFM and change in weight; 1 had a negative y-axis intercept and the other had an intercept close to zero. The fact that the influence of weight on change in FFM is less for Steen et al's data (R2 = 0.30) than for Chien et al's data (6) on younger subjects (R2 = 0.66) indicates that there was more scatter in the former, which suggests that factors other than weight change per se (eg, illness or diet) assume more importance in the elderly.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
), and hence, gain an equivalent amount of body fat. Those who lose weight tend to have greater losses of FFM, and those who gain weight tend to gain FFM or to lose it less rapidly than others while gaining a disproportionate amount of fat. Hence, recorded changes in FFM during aging must be viewed in the context of body weight change. The fact that these results were obtained with a variety of techniques for estimating FFM lends creditability to this conclusion.
It seems that the best way to avoid a loss of FFM during aging would be to gain 2.3 kg body weight each decade, which is the x-axis intercept of equation 2. The matter of influences other than body weight—such as sex, age, and physical activity—cannot be completely resolved by the available data. The regression slope for women in the upper portion of Table 2 is less than that for men (P = 0.0003), but the intercepts do not differ significantly. The existence of only 4 groups of women in the grouped data means that sex comparisons for such data were impossible.
Earlier mention was made of the fact that neither of my long-term subjects had a significantly greater loss of FFM after 65 y of age than before (Figures 1 and 2). Although Steen et al's (8) older subjects (aged 70 y) had an intercept lower than Chien et al's (6) younger subjects (aged 33–49 y), and the regression slope was less (Table 2, upper portion), the comparison may not be valid because there was relatively little overlap in the range of changes in weight for the 2 studies. Strangely, the 60-y-old subjects studied by Parizkova and Eiselt (11) had a positive y-axis intercept. Note that the data presented include no subjects older than 81 y. However, it can be said that body weight change is a prime determinant of FFM change in the elderly as well as in younger individuals.
Whether continual physical activity can forestall the age decline in FFM cannot be answered by the extant data. There were 7 groups of subjects who were said to be highly trained athletes or physically active [2 groups reported by Parizkova and Eiselt (11), 3 by Pollock et al (9), and 2 by Trappe et al (10)]. Regressions of change in FFM against change in weight were calculated for this group of 7 and also for the 13 groups for whom no comments were made about physical activity and who were therefore considered to be sedentary. Neither the intercepts nor the slopes differed significantly between the 2 groups. Of interest is the fact that the "high-intensity" and "moderate-intensity" activity groups reported by Pollock et al (9) lost 1.9 and 1.4 kg FFM/decade, respectively, without a significant change in body weight. These values are similar to those recorded for individuals who were not engaged in regular exercise. Thus, it seems that the finding that exercise can augment FFM in the short term, provided there is no loss in weight (19), cannot be used to predict what will happen in the long term.
It is known that the weight gained during deliberate overfeeding is composed of both lean and fat tissue (20), as is the excess weight of obese individuals (21). Weight that is lost by intentional underfeeding or subsequent to intestinal bypass surgery consists of both lean and fat tissue (2), and the same is true of the weight differences between pairs of twins (22). The data on aging presented here show that the relations between changes in FFM and changes in weight, and between changes in fat and changes in weight are both positive, in keeping with the above relations; however, the regression lines are displaced from the origin while retaining their respective positive slopes. If sufficient weight is gained as the years go by, FFM will increase in the face of an increase in body fat just as it does in response to deliberate overfeeding, so the positive energy balance that brings about this change serves to effectively counteract the effect of aging on FFM. As weight is lost, both FFM and fat decline, just as they do in response to an imposed negative energy balance. The findings of the present study provide further evidence for the concept that these 2 body components, lean and fat tissue, are not independent entities (23).
The present study provides no information on the cause or causes of the decline in FFM with age. Possibilities include the decline with age in the concentration of gonadal hormones, dehydroepiandrosterone, and insulin-like growth factor (24), and the progressive accrual of molecular oxidative damage (25). Tzankoff and Norris (26), Magnus-Levy (27), and Keys et al (7) documented a longitudinal decline in basal metabolic rate in men, and van Beresteijn et al (28) reported a longitudinal decline in bone mineral content of the radius in perimenopausal women.
It would be of great interest to know whether the time course of body weight change during adult life alters the degree to which various hormone concentrations and basal metabolic rate change with age. Whatever the ultimate causes of the adult decline in FFM turn out to be, body weight—and hence energy balance—must be numbered among them.
|
ACKNOWLEDGMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  F. Fantin, V. D. Francesco, G. Fontana, A. Zivelonghi, L. Bissoli, E. Zoico, A. Rossi, R. Micciolo, O. Bosello, and M. Zamboni Longitudinal Body Composition Changes in Old Men and Women: Interrelationships With Worsening Disability J. Gerontol. A Biol. Sci. Med. Sci., December 1, 2007; 62(12): 1375 - 1381. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. H Schmitz, P. J Hannan, S. D Stovitz, C. J Bryan, M. Warren, and M. D Jensen Strength training and adiposity in premenopausal women: Strong, Healthy, and Empowered study Am. J. Clinical Nutrition, September 1, 2007; 86(3): 566 - 572. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. P. Truesdale, J. Stevens, and J. Cai Nine-Year Changes in Cardiovascular Disease Risk Factors with Weight Maintenance in the Atherosclerosis Risk in Communities Cohort Am. J. Epidemiol., April 15, 2007; 165(8): 890 - 900. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 U. G. Kyle, K. Melzer, B. Kayser, M. Picard-Kossovsky, G. Gremion, and C. Pichard Eight-Year Longitudinal Changes in Body Composition in Healthy Swiss Adults J. Am. Coll. Nutr., December 1, 2006; 25(6): 493 - 501. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Chevalier, R. Gougeon, N. Choong, M. Lamarche, and J. A. Morais Influence of adiposity in the blunted whole-body protein anabolic response to insulin with aging. J. Gerontol. A Biol. Sci. Med. Sci., February 1, 2006; 61(2): 156 - 164. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. B Newman, J. S. Lee, M. Visser, B. H Goodpaster, S. B Kritchevsky, F. A Tylavsky, M. Nevitt, and T. B Harris Weight change and the conservation of lean mass in old age: the Health, Aging and Body Composition Study Am. J. Clinical Nutrition, October 1, 2005; 82(4): 872 - 878. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Guillet, A. Zangarelli, P. Gachon, B. Morio, C. Giraudet, P. Rousset, and Y. Boirie Whole Body Protein Breakdown Is Less Inhibited by Insulin, But Still Responsive to Amino Acid, in Nondiabetic Elderly Subjects J. Clin. Endocrinol. Metab., December 1, 2004; 89(12): 6017 - 6024. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. A. Schrager, S. M. Roth, R. E. Ferrell, E. J. Metter, E. Russek-Cohen, N. A. Lynch, R. S. Lindle, and B. F. Hurley Insulin-like growth factor-2 genotype, fat-free mass, and muscle performance across the adult life span J Appl Physiol, December 1, 2004; 97(6): 2176 - 2183. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. H. GARDE, A. M. HANSEN, J. KRISTIANSEN, and L. E. KNUDSEN Comparison of Uncertainties Related to Standardization of Urine Samples with Volume and Creatinine Concentration Ann. Hyg., March 1, 2004; 48(2): 171 - 179. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. J. Marcell Review Article: Sarcopenia: Causes, Consequences, and Preventions J. Gerontol. A Biol. Sci. Med. Sci., October 1, 2003; 58(10): M911 - 916. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Visser, M. Pahor, F. Tylavsky, S. B. Kritchevsky, J. A. Cauley, A. B. Newman, B. A. Blunt, and T. B. Harris One- and two-year change in body composition as measured by DXA in a population-based cohort of older men and women J Appl Physiol, June 1, 2003; 94(6): 2368 - 2374. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. A Hughes, W. R Frontera, R. Roubenoff, W. J Evans, and M. A F. Singh Longitudinal changes in body composition in older men and women: role of body weight change and physical activity Am. J. Clinical Nutrition, August 1, 2002; 76(2): 473 - 481. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. L. Heitmann and L. Garby Composition (lean and fat tissue) of weight changes in adult Danes Am. J. Clinical Nutrition, May 1, 2002; 75(5): 840 - 847. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Visser, M. Pahor, D. R. Taaffe, B. H. Goodpaster, E. M. Simonsick, A. B. Newman, M. Nevitt, and T. B. Harris Relationship of Interleukin-6 and Tumor Necrosis Factor-{alpha} With Muscle Mass and Muscle Strength in Elderly Men and Women: The Health ABC Study J. Gerontol. A Biol. Sci. Med. Sci., May 1, 2002; 57(5): M326 - 332. [Abstract] [Full Text]
|
|
|
|
|
|
M. Pirlich, H. Biering, H. Gerl, M. Ventz, B. Schmidt, S. Ertl, and H. Lochs Loss of Body Cell Mass in Cushing's Syndrome: Effect of Treatment J. Clin. Endocrinol. Metab., March 1, 2002; 87(3): 1078 - 1084. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. M. Matsumoto Andropause: Clinical Implications of the Decline in Serum Testosterone Levels With Aging in Men J. Gerontol. A Biol. Sci. Med. Sci., February 1, 2002; 57(2): M76 - 99. [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
