Exercise
and body composition
GILBERT B. FORBES Departments of Pediatrics and Biophysics, University of Rochester School of Medicine and Dentistry, Rochester, New York 14642
FORBES, GILBERT B. Exercise and body composition. J. Appl. Physiol. 70(3): 994-997, 1991.-We assessed changes in body composition in 41 young adults who engaged in various exercise and/or training programs on ad libitum diets. Most of those who gained weight sustained an increase in lean body mass (LBM), and most of those who lost weight lost LBM as well as fat. The change in LBM was directly related to the change in weight, with a regression slope of 0.500. An analysis of published data confirms these findings and, in concert with our data, provides the additional information that the magnitude of the change in body composition in exercising individuals is influenced by body fat content, just as it is for nonexercising individuals. physical
activity; weight
change; lean body mass; body fat
LEAN BODY MASS (LBM)
usually changes in response to significant changes in body weight: energy deficit leads to a fall in LBM as well as fat, and energy surfeit leads to an increase in both body components (7-10). Moreover, the relative contribution of LBM to the total weight change is a function of body fat content: obese individuals lose proportionately less LBM during energy deficit and gain proportionately less LBM during energy surfeit than do thin individuals (89). The question is whether these relationships can be altered by exercise. It is generally held that exercise and/or training programs can augment LBM, albeit to a modest degree (7, l7), as well as cause a decrease in body fat. Many athletes of both sexestend to have a larger LBM than their sedentary peers (8), but the relative contributions of heredity, training, and nutrition to this difference have not been defined. The purpose of the present study is to examine the changes in body composition, i.e., lean weight and fat, associated with exercise programs, some of which also involved low energy diets, and to relate these changes to changes in body weight. The entire LBM, as determined by body composition assays, is the focus of this inquiry rather than specific muscle groups. Data from my own laboratory will be used for this purpose, together with those of other investigators who have provided individual data on the subjects they have studied. Individual data are necessary for the simple reason that most studies reveal considerable individual variability in the degree of weight change for a given program of diet and exercise.
young adults who engaged in exercise and/or training programs for 1 to 10 mo. The programs were self-selected: 19 subjects participated in running and varsity sports, 1 subject rode his bicycle across the United States in 61 days, 1 male and 1 female were “body builders,” and 19 were varsity football players studied before and after 6 wk of spring practice combined with thrice-weekly sessions on the Nautilus machine. Use of anabolic steroids was denied by all. Diet was not prescribed. The subjects ranged in age from 18 to 40 yr, in height from 152 to 196 cm, in weight from 50 to 116 kg, in LBM from 42 to 93 kg, in body fat from 5 to 19 kg (save one with 28 kg fat, weight 116 kg, height 196 cm) and in body mass index (Wt/Ht2) from 19 to 25 (1 was 30). LBM was estimated by potassium-40 counting at the beginning and end of each program and, in some instances, several times during the course of the program. The observed change in LBM was regressed against change in body weight with the use of the least-squares technique; the statistical analysis was done on the Clinfo Data Analysis System (BBN Software Products, Cambridge, MA). Data from the literature. Published reports of the effects of various exercise programs on body composition were evaluated. Some investigators allowed their subjects to eat ad libitum, whereas others prescribed a lowenergy, protein-adequate diet in addition to exercise. In perusing individual subjects’ data, which were kindly provided by the authors, it became clear that there was considerable variation in the body weight response within each group. For example, Hill et al. (13) asked their subjects to consume a 1,200-kcal diet for 12 wk while exercising regularly: the recorded loss of weight varied from 2 to 15 kg, so it is possible that some of their subjects were not compliant with the diet and/or the exercise program. It is because of this variability that the analysis of individual responses was so important. Only those reports of groups of 210 subjects were analyzed. Estimates of LBM were made by potassium-40 counting in one group; hydrodensitometry wa.s used for the others, and in one of these the total blady water method was also used. With the individual data at hand, we proceeded to regress the observed change in LBM against the change in body weight for each group of subjects. RESULTS
METHODS
Own observations. During the past several years we have studied 31 male and 10 female college students and 994
0161-7567/91
$1.50 Copyright
Own data. Figure 1 is a plot of change in LBM against change in weight for our 41 su.bjects. It is to be seen that of the 1.Osubjects who gained 21 kg during the exercise
0 1991 the American
Physiological
Society
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on August 31, 2018. Copyright © 1991 American Physiological Society. All rights reserved.
EXERCISE
ol
1 Females Males
I
n
o
1
BODY
of Exercise
Effect +4
AND
X
0
Y = 0.08 + 0.50x
0
'I
0
technical error
a
I
0 0
0 -6
!
I -6
I
1
I
r
1
1
-4
-2
0
+2
+4
+6
A Weight,
995
COMPOSITION
regression equation are affected by the initial body fat status of the subjects. This effect is made clear when the seven groups of subjects who have smaller body fat burdens are compared with the seven groups with larger values for body fat (Table 2). Although there is a fair amount of variation among the groups, on average the thinner groups have a steeper regression slope (ALBM vs. AWt) than those with larger fat burdens. In addition, the y-axis intercept is higher for the latter group, and the x-axis intercept is more negative. Hence weight change during exercise in thin individuals is associated with a relatively greater LBM change than in those who are fatter; the latter can lose more weight without a loss of LBM, and they can gain a modest amount of LBM in the absence of a weight gain. Looked at in another way, weight change in fatter individuals involves a larger proportion of fat than in those who are thin.
kg
1. Change in LBM vs. change in weight. Calculated regression equation is y = +0.08 + 0.34 + 0.500 + 0.167x (SE); r = 0.433. Horizontal dotted lines define technical error of 2%. FIG.
period, 8 also gained LBM; of the 11 who lost >1 kg, 8 lost LBM; and of the 20 subjects whose weights changed by 1.5 kg of weight will be associated with a decline in LBM. The data of Table 2 also suggest that, on average, weight gains of more than - 1.5 kg during exercise will be associated with some gain in fat content as well as LBM. It should be remembered that the data presented here pertain for the most part to exercise programs of only
(a/b) - 1.44kO.64 -3.85kO.85
’ The derived equation entiating and rearranging, (Refs. 8, 9).
is LBM (kg) = 23.9 log fat (kg) + 14.2; differwe have d(LBM)/dwt = 10.4/(fat + 10.4)
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on August 31, 2018. Copyright © 1991 American Physiological Society. All rights reserved.
EXERCISE
AND
BODY
moderate intensity, and they do not address the problems that may occur in prolonged semistarvation. It should be remembered that the data presented here pertain for the most part to exercise programs of only moderate intensity, and they do not address the problems that may occur in prolonged semistarvation. The author thanks Cheryl Porta for technical assistance and R. A. Boileau, E. R. Buskirk, J. P. Despres, H. C. Lukaski, R. Ross, P. D. Wood, D. L. Ballor, and J. 0. Hill, who sent me individual data for their subjects. This study was supported by National Institute of Child Health and Human Development Grant HD-18454 and Division of Research Resources Grant RR-00044. Address for reprint requests: G. B. Forbes, Department of Pediatrics, Box 777, University of Rochester Medical Center, Rochester, NY 14642. Received
17 May
1990; accepted
in final
form
20 August
1990.
REFERENCES 1. BALLOR, D. L., V. L. KATCH, M. D. BECQUE, AND C. R. MARKS. Resistance training during calorie restriction enhances lean body weight maintenance. Am. J. Clin. Nutr. 47: 19-25, 1988. 2. BALLOR, D. L., J. P. MCCARTHY, AND E. J. WILTERDINK. Exercise intensity does not affect the composition of diet- and exercise-induced body mass loss. Am. J. CLin. Nutr. 51: 142-146, 1990. 3. BODDY, K,, R. HUME, P. C. KING, E. WEYERS, AND T. ROWAN. Total body, plasma and erythrocyte potassium and leucocyte ascorbic acid in ‘ultra-fit’ subjects. Clin. Sci. Land. 46: 449-456, 1974. 4. BOILEAU, R. A., E. R. BUSKIRK, D. H. HORSTMAN, J. MENDEZ, AND W. C. NICHOLAS. Body composition changes in obese and lean men during physical conditioning. Med. Sci. Sports 3: 183-189, 1971. 5. BROWN, M. R., W. J. KLISH, J. HOLLANDER, M. A. CAMPBELL, AND G. B. FORBES. A high protein, low calorie liquid diet in the treatment of very obese adolescents: long-term effect on lean body mass. Am. J. Clin. Nutr. 38: 20-31, 1983.
COMPOSITION
997
6. DESPRBS, J. P., C. BOUCHARD, A. TREMBLAY, R. SAVARD, AND M. MARCOTTE. Effects of aerobic training on fat distribution in male subjects. Med. Sci. Sports Exercise 17: 113-118, 1985. 7. FORBES, G. B. Body composition as affected by physical activity and nutrition. Federation Proc. 44: 343-347, 1985. Body Composition: Growth, Aging, Nutrition 8. FORBES, G. B. Human and Activity. New York: Springer-Verlag, 1987. 9. FORBES, G. B. Lean body mass-body fat relationships in man: dietary changes induce changes in both body components. Nutr. Reu. 45: 225-231, 1987. 10. FORBES, G. B., M. R. BROWN, S. L. WELLE, AND B. A. LIPINSKI. Deliberate overfeeding in women and men: energy cost and composition of the weight gain. Br. J. Nutr. 56: l-9, 1986. 11. FRANKLIN, B., E. BUSKIRK, J. HODGSON, H. GAHAGAN, J. KOLLIAS, AND J. MENDEZ. Effects of physical conditioning on cardiorespiratory function, body composition and serum lipids in relatively normal weight and obese middle-aged women. Int. J. Obesity 3: 97-109, 1979. 12. HAMMER, R. L., C. A. BARRIER, E. S. ROUNDY, J. M. BRADFORD, AND A. G. FISHER. Calorie-restricted low-fat diet and exercise in obese women. Am. J. Clin. Nutr. 49: 77-85, 1989. 13. HILL, J. O., D. G. SCHLUNDT, T. SBROCCO, T. SHARP, J. POPECORDLE, B. STETSON, M. KALER, AND C. HEIM. Evaluation of an alternating-calorie diet with and without exercise in the treatment of obesity. Am. J. Clin. Nutr. 50: 248-254, 1989. 14. KEYS, A., J. BR~ZEK, A. HENSCHEL, 0. MICHELSEN, AND H. L. TAYLOR. The Biology of Human Staruation. Minneapolis: Univ. of Minnesota Press, 1971. 15. LUKASKI, H. C., B. S. HOVERSON, D. B. MILNE, AND W. W. BOLONCHUK. Copper, zinc, and iron status of female swimmers. Nutr. Res. 9: 493-502, 1989. 16. Ross, R., L. LEGER, P. MARTIN, AND R. ROY. Sensitivity of bioelectrical impedance to detect changes in human body composition. J. Appl. Physiol. 67: 1643-1648, 1987. 17. WILMORE, J. H. Body composition in sport and exercise: directions for future research. Med. Sci. Sports Exercise 15: 21-31, 1983. 18. WOOD, P. D., M. L. STEFANICK, D. M. DREON, B. FREY-HEWITT, S. C. GARAY, P. T. WILLIAMS, H. R. SUPERKO, S. P. FORTMANN, J. J. ALBERS, K. M. VRANIZAN, N. M. ELLWORTH, R. B. TERRY, AND W. L. HASKELL. Changes in plasma lipids and lipoproteins in overweight men during weight loss through dieting as compared with exercise. N. Engl. J. Med. 319: 1173-1179, 1988.
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on August 31, 2018. Copyright © 1991 American Physiological Society. All rights reserved.
and body composition
GILBERT B. FORBES Departments of Pediatrics and Biophysics, University of Rochester School of Medicine and Dentistry, Rochester, New York 14642
FORBES, GILBERT B. Exercise and body composition. J. Appl. Physiol. 70(3): 994-997, 1991.-We assessed changes in body composition in 41 young adults who engaged in various exercise and/or training programs on ad libitum diets. Most of those who gained weight sustained an increase in lean body mass (LBM), and most of those who lost weight lost LBM as well as fat. The change in LBM was directly related to the change in weight, with a regression slope of 0.500. An analysis of published data confirms these findings and, in concert with our data, provides the additional information that the magnitude of the change in body composition in exercising individuals is influenced by body fat content, just as it is for nonexercising individuals. physical
activity; weight
change; lean body mass; body fat
LEAN BODY MASS (LBM)
usually changes in response to significant changes in body weight: energy deficit leads to a fall in LBM as well as fat, and energy surfeit leads to an increase in both body components (7-10). Moreover, the relative contribution of LBM to the total weight change is a function of body fat content: obese individuals lose proportionately less LBM during energy deficit and gain proportionately less LBM during energy surfeit than do thin individuals (89). The question is whether these relationships can be altered by exercise. It is generally held that exercise and/or training programs can augment LBM, albeit to a modest degree (7, l7), as well as cause a decrease in body fat. Many athletes of both sexestend to have a larger LBM than their sedentary peers (8), but the relative contributions of heredity, training, and nutrition to this difference have not been defined. The purpose of the present study is to examine the changes in body composition, i.e., lean weight and fat, associated with exercise programs, some of which also involved low energy diets, and to relate these changes to changes in body weight. The entire LBM, as determined by body composition assays, is the focus of this inquiry rather than specific muscle groups. Data from my own laboratory will be used for this purpose, together with those of other investigators who have provided individual data on the subjects they have studied. Individual data are necessary for the simple reason that most studies reveal considerable individual variability in the degree of weight change for a given program of diet and exercise.
young adults who engaged in exercise and/or training programs for 1 to 10 mo. The programs were self-selected: 19 subjects participated in running and varsity sports, 1 subject rode his bicycle across the United States in 61 days, 1 male and 1 female were “body builders,” and 19 were varsity football players studied before and after 6 wk of spring practice combined with thrice-weekly sessions on the Nautilus machine. Use of anabolic steroids was denied by all. Diet was not prescribed. The subjects ranged in age from 18 to 40 yr, in height from 152 to 196 cm, in weight from 50 to 116 kg, in LBM from 42 to 93 kg, in body fat from 5 to 19 kg (save one with 28 kg fat, weight 116 kg, height 196 cm) and in body mass index (Wt/Ht2) from 19 to 25 (1 was 30). LBM was estimated by potassium-40 counting at the beginning and end of each program and, in some instances, several times during the course of the program. The observed change in LBM was regressed against change in body weight with the use of the least-squares technique; the statistical analysis was done on the Clinfo Data Analysis System (BBN Software Products, Cambridge, MA). Data from the literature. Published reports of the effects of various exercise programs on body composition were evaluated. Some investigators allowed their subjects to eat ad libitum, whereas others prescribed a lowenergy, protein-adequate diet in addition to exercise. In perusing individual subjects’ data, which were kindly provided by the authors, it became clear that there was considerable variation in the body weight response within each group. For example, Hill et al. (13) asked their subjects to consume a 1,200-kcal diet for 12 wk while exercising regularly: the recorded loss of weight varied from 2 to 15 kg, so it is possible that some of their subjects were not compliant with the diet and/or the exercise program. It is because of this variability that the analysis of individual responses was so important. Only those reports of groups of 210 subjects were analyzed. Estimates of LBM were made by potassium-40 counting in one group; hydrodensitometry wa.s used for the others, and in one of these the total blady water method was also used. With the individual data at hand, we proceeded to regress the observed change in LBM against the change in body weight for each group of subjects. RESULTS
METHODS
Own observations. During the past several years we have studied 31 male and 10 female college students and 994
0161-7567/91
$1.50 Copyright
Own data. Figure 1 is a plot of change in LBM against change in weight for our 41 su.bjects. It is to be seen that of the 1.Osubjects who gained 21 kg during the exercise
0 1991 the American
Physiological
Society
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on August 31, 2018. Copyright © 1991 American Physiological Society. All rights reserved.
EXERCISE
ol
1 Females Males
I
n
o
1
BODY
of Exercise
Effect +4
AND
X
0
Y = 0.08 + 0.50x
0
'I
0
technical error
a
I
0 0
0 -6
!
I -6
I
1
I
r
1
1
-4
-2
0
+2
+4
+6
A Weight,
995
COMPOSITION
regression equation are affected by the initial body fat status of the subjects. This effect is made clear when the seven groups of subjects who have smaller body fat burdens are compared with the seven groups with larger values for body fat (Table 2). Although there is a fair amount of variation among the groups, on average the thinner groups have a steeper regression slope (ALBM vs. AWt) than those with larger fat burdens. In addition, the y-axis intercept is higher for the latter group, and the x-axis intercept is more negative. Hence weight change during exercise in thin individuals is associated with a relatively greater LBM change than in those who are fatter; the latter can lose more weight without a loss of LBM, and they can gain a modest amount of LBM in the absence of a weight gain. Looked at in another way, weight change in fatter individuals involves a larger proportion of fat than in those who are thin.
kg
1. Change in LBM vs. change in weight. Calculated regression equation is y = +0.08 + 0.34 + 0.500 + 0.167x (SE); r = 0.433. Horizontal dotted lines define technical error of 2%. FIG.
period, 8 also gained LBM; of the 11 who lost >1 kg, 8 lost LBM; and of the 20 subjects whose weights changed by 1.5 kg of weight will be associated with a decline in LBM. The data of Table 2 also suggest that, on average, weight gains of more than - 1.5 kg during exercise will be associated with some gain in fat content as well as LBM. It should be remembered that the data presented here pertain for the most part to exercise programs of only
(a/b) - 1.44kO.64 -3.85kO.85
’ The derived equation entiating and rearranging, (Refs. 8, 9).
is LBM (kg) = 23.9 log fat (kg) + 14.2; differwe have d(LBM)/dwt = 10.4/(fat + 10.4)
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on August 31, 2018. Copyright © 1991 American Physiological Society. All rights reserved.
EXERCISE
AND
BODY
moderate intensity, and they do not address the problems that may occur in prolonged semistarvation. It should be remembered that the data presented here pertain for the most part to exercise programs of only moderate intensity, and they do not address the problems that may occur in prolonged semistarvation. The author thanks Cheryl Porta for technical assistance and R. A. Boileau, E. R. Buskirk, J. P. Despres, H. C. Lukaski, R. Ross, P. D. Wood, D. L. Ballor, and J. 0. Hill, who sent me individual data for their subjects. This study was supported by National Institute of Child Health and Human Development Grant HD-18454 and Division of Research Resources Grant RR-00044. Address for reprint requests: G. B. Forbes, Department of Pediatrics, Box 777, University of Rochester Medical Center, Rochester, NY 14642. Received
17 May
1990; accepted
in final
form
20 August
1990.
REFERENCES 1. BALLOR, D. L., V. L. KATCH, M. D. BECQUE, AND C. R. MARKS. Resistance training during calorie restriction enhances lean body weight maintenance. Am. J. Clin. Nutr. 47: 19-25, 1988. 2. BALLOR, D. L., J. P. MCCARTHY, AND E. J. WILTERDINK. Exercise intensity does not affect the composition of diet- and exercise-induced body mass loss. Am. J. CLin. Nutr. 51: 142-146, 1990. 3. BODDY, K,, R. HUME, P. C. KING, E. WEYERS, AND T. ROWAN. Total body, plasma and erythrocyte potassium and leucocyte ascorbic acid in ‘ultra-fit’ subjects. Clin. Sci. Land. 46: 449-456, 1974. 4. BOILEAU, R. A., E. R. BUSKIRK, D. H. HORSTMAN, J. MENDEZ, AND W. C. NICHOLAS. Body composition changes in obese and lean men during physical conditioning. Med. Sci. Sports 3: 183-189, 1971. 5. BROWN, M. R., W. J. KLISH, J. HOLLANDER, M. A. CAMPBELL, AND G. B. FORBES. A high protein, low calorie liquid diet in the treatment of very obese adolescents: long-term effect on lean body mass. Am. J. Clin. Nutr. 38: 20-31, 1983.
COMPOSITION
997
6. DESPRBS, J. P., C. BOUCHARD, A. TREMBLAY, R. SAVARD, AND M. MARCOTTE. Effects of aerobic training on fat distribution in male subjects. Med. Sci. Sports Exercise 17: 113-118, 1985. 7. FORBES, G. B. Body composition as affected by physical activity and nutrition. Federation Proc. 44: 343-347, 1985. Body Composition: Growth, Aging, Nutrition 8. FORBES, G. B. Human and Activity. New York: Springer-Verlag, 1987. 9. FORBES, G. B. Lean body mass-body fat relationships in man: dietary changes induce changes in both body components. Nutr. Reu. 45: 225-231, 1987. 10. FORBES, G. B., M. R. BROWN, S. L. WELLE, AND B. A. LIPINSKI. Deliberate overfeeding in women and men: energy cost and composition of the weight gain. Br. J. Nutr. 56: l-9, 1986. 11. FRANKLIN, B., E. BUSKIRK, J. HODGSON, H. GAHAGAN, J. KOLLIAS, AND J. MENDEZ. Effects of physical conditioning on cardiorespiratory function, body composition and serum lipids in relatively normal weight and obese middle-aged women. Int. J. Obesity 3: 97-109, 1979. 12. HAMMER, R. L., C. A. BARRIER, E. S. ROUNDY, J. M. BRADFORD, AND A. G. FISHER. Calorie-restricted low-fat diet and exercise in obese women. Am. J. Clin. Nutr. 49: 77-85, 1989. 13. HILL, J. O., D. G. SCHLUNDT, T. SBROCCO, T. SHARP, J. POPECORDLE, B. STETSON, M. KALER, AND C. HEIM. Evaluation of an alternating-calorie diet with and without exercise in the treatment of obesity. Am. J. Clin. Nutr. 50: 248-254, 1989. 14. KEYS, A., J. BR~ZEK, A. HENSCHEL, 0. MICHELSEN, AND H. L. TAYLOR. The Biology of Human Staruation. Minneapolis: Univ. of Minnesota Press, 1971. 15. LUKASKI, H. C., B. S. HOVERSON, D. B. MILNE, AND W. W. BOLONCHUK. Copper, zinc, and iron status of female swimmers. Nutr. Res. 9: 493-502, 1989. 16. Ross, R., L. LEGER, P. MARTIN, AND R. ROY. Sensitivity of bioelectrical impedance to detect changes in human body composition. J. Appl. Physiol. 67: 1643-1648, 1987. 17. WILMORE, J. H. Body composition in sport and exercise: directions for future research. Med. Sci. Sports Exercise 15: 21-31, 1983. 18. WOOD, P. D., M. L. STEFANICK, D. M. DREON, B. FREY-HEWITT, S. C. GARAY, P. T. WILLIAMS, H. R. SUPERKO, S. P. FORTMANN, J. J. ALBERS, K. M. VRANIZAN, N. M. ELLWORTH, R. B. TERRY, AND W. L. HASKELL. Changes in plasma lipids and lipoproteins in overweight men during weight loss through dieting as compared with exercise. N. Engl. J. Med. 319: 1173-1179, 1988.
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on August 31, 2018. Copyright © 1991 American Physiological Society. All rights reserved.
