|
|
|
| September 2000, Volume 24, Number 9, Pages 1153-1157 |
| Table of contents Previous Article Next [PDF] |
 |
| Paper |
| Body composition and resting energy expenditure in humans: role of fat, fat-free mass and extracellular fluid |
 |
| S Nielsen, D D Hensrud, S Romanski, J A Levine, B Burguera and M D Jensen |
 |
Endocrine Research Unit, Mayo Clinic, Rochester, MN 55905, USA
|
 |
Correspondence to: M D Jensen, Endocrine Research Unit, 5-194 Joseph, Mayo Clinic, Rochester, MN 55905, USA. jensen.michael@mayo.edu
|
 |
| Abstract |
 | OBJECTIVE: The objective of this study was to determine whether there are independent effects of extracellular fluid volume (ECF) and fat mass (FM) on resting energy expenditure (REE) relative to fat-free mass (FFM) in adult men and women. METHODS: Multiple linear regression analysis was used to relate REE, as determined by indirect calorimetry, to FFM and FM (measured using dual energy X-ray absorptiometry) and ECF (measured using bromide space and/or the radiosulfate washout space) in 153 women and 100 men with varying amounts of body fat. RESULTS: REE correlated significantly with FFM and FM in women (r=0.65 and r=0.63, both P<0.001) and men (r=0.62 and r=0.48, both P<0.001, FFM and FM, respectively). In a multiple linear regression analysis FFM, FM and age significantly contributed to the ability to predict REE in both genders. The models that were derived were not significantly different between women and men. In women the contribution to REE from FM was easier to detect when FM was greater. Adjustment of FFM for ECF did not improve the relationship between FFM and REE. CONCLUSIONS: FFM, FM and age are significant, independent predictors of REE in both men and women. Adjustment of FFM for ECF does not improve the ability of FFM to predict REE, which suggests that ECF is a highly integrated component of FFM in healthy adults. Expressing REE relative to FFM alone will introduce errors when lean and obese populations are compared. International Journal of Obesity (2000) 24, 1153-1157 |
 |
| Keywords |
 | indirect calorimetry; dual energy X-ray absorptiometry; obesity |
 |  |
Introduction
Fat-free mass (FFM) is the single best predictor of resting energy expenditure (REE), explaining much of the inter-individual variability in REE.1,2,3 Some investigators report that body fat is an independent predictor of the inter-individual variation in REE,4 which is logical considering that adipose tissue consumes oxygen at a rate of 0.4 ml/kg/min.5 This rate of oxygen consumption is much less than that for lean tissue,6,7,8 and suggests that large numbers of subjects must be evaluated in order to detect an effect of body fat on REE in cross-sectional population studies. In addition, robust methods for measuring body composition and REE must be employed. A significant portion of the inter-individual variability in REE remains unexplained even when FFM and fat mass are known, however.
Family membership, age and gender can account for some, but not all of the remaining 15-40% variability in REE.9 One potential confounding factor in attempting to relate REE to FFM or fat mass is the extracellular fluid (ECF) compartment. ECF is a metabolically inert component of FFM and it is possible that inter-individual variability of the ECF/FFM relationship reduces the predictive value of FFM. Thus, adjustment of FFM for ECF might improve the relationship between REE and FFM. The primary goal of the present study was to determine whether subtracting ECF from FFM (an estimate of body cell mass) would improve the relationship between body composition and REE in a large sample of healthy subjects covering a broad range of body weight. Because there continues to be discrepant reports on the relationship between FFM, FM and REE we evaluated these aspects of our data as well.
|
 Subjects and methods
Subjects
For the present analysis data from several studies (153 women and 100 men) being conducted in our laboratory were compiled. All studies included healthy subjects participating in metabolic research studies of fatty acid metabolism that included measurements of body composition and REE. In 93 women (61% of all women volunteers) and in 77 men (77% of the volunteers) ECF was determined prospectively for the purpose of assessing its usefulness as a co-variate in measurements of energy metabolism. The studies were approved by the Mayo Clinic Institutional Review Board and informed written consent was obtained.
Resting energy expenditure
Resting energy expenditure was assessed by indirect calorimetry with a ventilated hood (Deltatrack Metabolic Monitor, SensorMedics, Yorba Linda, CA).10 The equipment was calibrated using standard oxygen and CO2 gases before each measurement. All volunteers had been weight stable for 2 months and consumed an isoenergetic diet for at least 3 days prior to the measurement of REE. Every volunteer was admitted to the Mayo Clinic General Clinical Research Center (GCRC) the evening before the study. The evening meal was consumed between 17:00 and 19:00 h and the volunteers remained non-per orum (NPO) (except water) until the REE measurement was completed before arising the next morning. The respiratory quotient (RQ) values for these volunteers were 0.81±0.04 (range 0.70-0.92).
Body composition
Total body fat mass and fat-free mass were measured using dual-energy X-ray absorptiometry (DXA; DPX-IQ, Lunar Radiation Corp., Madison, Wisconsin).11 Extracellular fluid volume was measured using bromide ion chromatography12 and/or the radiosulfate washout technique13 as previously described. Body mass index (BMI) was calculated as body weight (kg)/height (m)2. Adjusted FFM was calculated as FFM (kg)-ECF (kg), assuming that 1 litre of ECF equals 1 kg.
Statistics
Values are given as mean±s.d. or median (interquartile range) (if the data were skewed) unless otherwise indicated. Comparisons between genders were performed using Student's t-test or the Mann-Whitney Two Sample Test. Correlations were evaluated by Pearson's or Spearman's test as appropriate. Predictors of REE were evaluated using multiple linear regression analysis with REE as the dependent variable and FFM, FM, age and gender as independent variables. The analysis was initially performed in the total study group (ie women and men combined). However, multicollinearity among FFM and gender and heteroscedacity (ie lack of constant variance of residuals) prompted gender specific analysis. P-values of <0.05 were considered statistically significant.
|
 Results
Body composition and REE (Table 1)
A total of 153 women and 100 men (aged 33±8 and 32±8 y, respectively) were included in the analysis. The two groups were comparable with respect to age and BMI, but women had significantly more body fat and less FFM than men (Table 1) did. Both ECF and adjusted FFM were significantly less in women than men (Table 1). No differences in body composition or REE were present between subjects with and without ECF measures.
Predictors of REE (Tables 2 and 3, Figures 1 and 2)
REE correlated significantly with all constituents of body composition (Table 2). In men the correlation was strongest with FFM (Figure 1), whereas, in females the correlation coefficients with FFM and FM were similar (Figure 2, Table 2). As assessed by multiple linear regression analysis, REE was significantly determined by FFM, FM (or percentage fat), and age in women (Table 3). In men REE was significantly determined by FFM, FM and age (Table 3), however, when percentage body fat was substituted for FM, age was no longer a significant predictor of REE. The following formulas were derived:
Women
REE(kcal/day) = 16.2 ´ FFM (kg) + 8.0 ´ FM (kg)
4.7 ´ age (y) + 714
(SE of estimate = 150 (kcal/day); adjusted r2
=0.55 P<0.0001)
Men
REE (kcal/day) = 15.6´FFM (kg) + 7.8 ´ FM (kg)
5.2 ´ age (y) + 888
(SE of estimate = 180 (kcal/day); adjusted r2
= 0.45 P<0.0001)
The residuals of a multiple linear regression analysis of REE on FFM and age were plotted against FM in a linear regression analysis for each gender (Figure 3). In accordance with the multiple linear regression analysis, FM was significantly associated with the residuals despite a large degree of scatter around the regression line.
Effect of ECF (Tables 1 and 2)
ECF correlated significantly with FFM in women (r=0.57, P<0.01) and men (r=0.60, P<0.01). The correlation between REE and adjusted FFM was weaker than between REE and FFM in both genders (Table 2). Similarly, when adjusted FFM was substituted for FFM in a multiple linear regression analysis the prediction of REE was weaker than with FFM in both groups. The correlation between adjusted FFM and REE was not stronger than between ECF and REE (Table 2).
|
 Discussion
The major goal of this study was to determine whether knowledge of ECF volume allows a better understanding of the body composition determinants of REE. In accordance with previous studies,1,2,3 we found that FFM was a strong predictor of REE in both women and men. Adjustment of FFM by subtracting ECF did not improve the ability of FFM to predict REE. Moreover we found that FM was a significant predictor of REE in women and in men. In both genders age was inversely related to REE.
Fat-free mass or lean body mass1,2,3 is the single most important contributor to REE, explaining 60-85% of the inter-individual variability in REE. Consequently, REE is frequently expressed relative to FFM. Because low REE is considered a risk factor for future weight gain14 much attention has been directed towards characterizing the 15-40% of the variability that remains unaccounted for by FFM. Water comprises ~60% of body weight in adult humans, of which approximately one-third is ECF. In theory, adjustment of FFM for ECF might improve the predictive value of FFM in determining REE in as much as ECF is metabolically inert. Instead, we found that body cell mass, calculated by subtracting ECF from FFM, was actually of less value in predicting REE. The correlation between REE and ECF was similar to the correlation between REE and FFM before adjustment for ECF, indicating that ECF is a highly integrated component of FFM in healthy adults. Other explanations are insufficient precision of the methods used for assessing ECF which, consequently, would introduce additional scatter on the relationship between REE and adjusted FFM. However, the bromide space and radiosulfate washout techniques are the primary, readily available measures of ECF15 and both methods have been shown to produce accurate and reproducible estimates of ECF.16,17,18
Some investigators have concluded that fat mass contributes significantly to REE,3,4,19,20,21 whereas others have not.1,2,4,22,23 It is possible that less robust methods for assessment of body composition and/or REE have confounded the ability to find independent effects of FM on REE in some of these studies. In our study body composition was measured by DXA and REE was measured using indirect calorimetry under carefully controlled conditions. DXA is an accurate and reproducible method for assessing body fat (triglyceride) mass;11,24 however, it should be remembered that DXA FFM includes the 10-15% of adipose tissue that is not triglyceride. In the present analysis, both FFM and fat mass were predictors of REE. It has been suggested that a contribution from FM to REE only becomes appreciable when relative FM increases above normal,1,3,20 eg in obese women. It seems more likely that the effect, being relatively small, is more readily detected with the inclusion of individuals with a wide range of body fat. Heshka et al 25 reported a similar regression coefficient for the relationship between fat mass and adjusted resting energy expenditure. Interestingly, they detected this effect primarily because of a dramatic change in fat mass in response to a weight loss program, but could not detect an effect in their baseline measurements in this study of 35 obese adults. Other studies report that FM is a more important contributor to REE in men.21 In our final model including FFM, FM and age the percentage of the variability in REE explained by these parameters was moderate (56% and 45% in women and men, respectively) and in the low range of what has been reported previously by others.1 It should be noted that the range of FFM included in our population is less than that reported by other investigators,9,26 which may have reduced the correlation coefficient between FFM and REE.
Oxygen consumption by adipose tissue is ~0.4 ml/kg/min5 (~2.9 kcal/kg/day). Considering that adipose tissue is ~85% fat, we would predict that body fat (as measured by DXA) would increase REE by ~3.4 kcal/kg/day if only the metabolic needs of adipose tissue influence resting oxygen consumption. The factor associated with body fat in the regression formulae were at least double this value for both men and women in our study, suggesting there may be additional metabolic effects of adipose tissue. For example, effects of body fat on sympathetic nervous system to increase oxygen consumption in lean tissue have been proposed,21 and large amounts of abdominal fat might increase the work of diaphragmatic breathing in the supine position. One or both of these factors could explain the observations that REE relative to FFM appears to be greater in upper body obese women27 and in men with obesity.21
We conclude that FFM, FM and age contribute significantly to the inter-individual variability of REE. Adjustment of FFM by subtracting the metabolically inert ECF compartment does not improve the relationship between FFM and REE when currently available techniques are used. This suggests either that ECF is an integrated component of lean tissue in healthy adults and/or that ECF cannot be measured with sufficient accuracy to provide an accurate estimate of body cell mass. Expressing REE relative to FFM alone might introduce errors when lean and obese populations are compared because adipose tissue is an energy requiring tissue. However, we should point out that the relationships between resting energy expenditure and body composition variables reported in this manuscript are based solely on statistical correlations. Whether these relationships reflect the true physiological associations among these variables awaits further study.
|
 | Acknowledgements
We wish to thank Rita Nelson, Carol Siverling and the staff of the Mayo GCRC for excellent technical assistance. Supported by grants DK45353, DK40484 and RR00585 from the US Public Health Service, the Minnesota Obesity Center (DK50456) and the Mayo Foundation.
|  |
| References |
 |
1 Cunningham JJ. Body composition as a determinant of energy expenditure: a synthetic review and a proposed general prediction equation. Am J Clin Nutr 1991; 54: 963-969. MEDLINE
2 Mifflin MD, St Jeor ST, Hill LA, Scott BJ, Daugherty SA, Koh YO. A new predictive equation for resting energy expenditure in healthy individuals. Am J Clin Nutr 1990; 51: 241-247. MEDLINE
3 Nelson KM, Weinsier RL, Long CL, Schutz Y. Prediction of resting energy expenditure from fat-free mass and fat mass. Am J Clin Nutr 1992; 56: 848-856. MEDLINE
4 Ferraro R and Ravussin E. Fat mass in predicting resting metabolic rate. Am J Clin Nutr 1992; 56: 460-461. MEDLINE
5 Simonsen L, Bulow J, Madsen J. Adipose tissue metabolism in humans determined by vein catheterization and microdialysis techniques. Am J Physiol 1994; 266: E357-365. MEDLINE
6 Brundin T, Branstrom R, Wahren J. Effects of oral vs i.v. glucose administration on splanchnic and extrasplanchnic O2 uptake and blood flow. Am J Physiol 1996; 271: E496-E504. MEDLINE
7 Brundin T, Thorne A, Wahren J. Heat leakage across the abdominal wall and meal-induced thermogenesis in normal-weight and obese subjects. Metabolism 1992; 41: 49-55. MEDLINE
8 Jensen MD, Johnson CM, Cryer PE, Murray MJ. Thermogenesis after a mixed meal: role of leg and splanchnic tissues in men and women. Am J Physiol 1995; 268: E433-438. MEDLINE
9 Bogardus C, Lillioja S, Ravussin E. Familial dependence of the resting metabolic rate. New Engl J Med 1986; 315: 96-100. MEDLINE
10 Ferrannini E. The theoretical bases of indirect calorimetry: a review. Metabolism 1988; 37: 287-301. MEDLINE
11 Jensen MD, Kanaley JA, Roust LR. Assessment of body composition with use of dual-energy X-ray absorptiometry: evaluation and comparison with other methods. Mayo Clin Proc 1993; 68: 867-873. MEDLINE
12 Wong WW, Sheng HP, Morkeberg JC, Kosanovich JL, Clarke LL, Klein PD. Measurement of extracellular water volume by bromide ion chromatography. Am J Clin Nutr 1989; 50: 1290-1294. MEDLINE
13 Pierson RN, Wang J, Colt EW, Neumann P. Body composition measurements in normal man: the potassium, sodium, sulfate and tritium spaces in 58 adults. J Chron Dis 1982; 35: 419-428.
14 Ravussin E, Lillioja S, Knowler WC. Reduced rate of energy expenditure as a risk factor for body-weight gain. New Engl J Med 1988; 318: 467-472. MEDLINE
15 Omvik P, Tarazi RC, Bravo EL. Determination of extracellular fluid volume in uremic patients by oral administration of radiosulfate. Kidney Int 1979; 15: 71-79. MEDLINE
16 Ryan RJ, Pascal LR, Inoye T, Bernstein L. Experiences with radiosulfate in the estimation of physiologic extracellular water in healthy and abnormal man. J Clin Invest 1956; 35: 1119-1130.
17 Bauer JH, Willis LR, Burt RW, Grim CE. Volume studies. II. Simultaneous determination of plasma volume, red cell mass, extracellular fluid, and total body water before and after volume expansion in dog and man. J Lab Clin Med 1975; 86: 1009-1017. MEDLINE
18 Walser M, Seldin DW, Grollmann A. An evaluation of radiosulfate for the determination of the volume of extracellular fluid in man and dogs. J Clin Invest 1953; 32: 299-311.
19 Garrow JS and Webster J. Are pre-obese people energy thrifty? Lancet 1985; 1: 670-671. MEDLINE
20 Bernstein RS, Thornton JC, Yang MU. Prediction of the resting metabolic rate in obese patients. Am J Clin Nutr 1983; 37: 595-602. MEDLINE
21 Dionne I, Despres JP, Bouchard C, Tremblay A. Gender difference in the effect of body composition on energy metabolism. Int J Obes Relat Metab Disord 1999; 23: 312-319. MEDLINE
22 Cunningham JJ. A reanalysis of the factors influencing basal metabolic rate in normal adults. Am J Clin Nutr 1980; 33: 2372-2374. MEDLINE
23 Ravussin E, Lillioja S, Anderson TE, Christin L, Bogardus C. Determinants of 24-hour energy expenditure in man. Methods and results using a respiratory chamber. J Clin Invest 1986; 78: 1568-1578. MEDLINE
24 Roubenoff R, Kehayias JJ, Dawson-Hughes B, Heymsfield SB. Use of dual-energy X-ray absorptiometry in body composition studies: not yet a 'gold standard&rsquo. Am J Clin Nutr 1993; 58: 589-591. MEDLINE
25 Heshka S, Yang M, Wang J, Burt P, Pi-Sunyer FX. Weight loss and change in resting metabolic rate. Am J Clin Nutr 1990; 52: 981-986. MEDLINE
26 Ferraro R, Lillioja S, Fontvieille A, Rising R, Bogardus C, Ravussin E. Lower sedentary metabolic rate in women compared with men. J Clin Invest 1992; 90: 780-784. MEDLINE
27 Weststrate JA, Dekker J, Stoel M, Begheijn L, Deurenberg P, Hautvast JGAJ. Resting energy expenditure in women: impact of obesity and body-fat distribution. Metabolism 1990; 39: 11-17. MEDLINE
|
 |
| Figures |
 |
Figure 1 Resting energy expenditure as related to fat-free mass (left panel) and fat mass (right panel) in 153 healthy women. All correlations are significant (Table 2). |
Figure 2 Resting energy expenditure as related to the fat-free mass (left panel) and fat mass (right panel) in 100 healthy men. All correlations are significant (Table 2). |
Figure 3 Linear regression of FM on the residuals of a multiple linear regression analysis of REE on FFM and age in women (r2=0.13; P<0.001) and men (r2=0.07; P<0.01). |
 |
| Tables |
 |
Table 1 Body composition and resting enery expenditure in healthy women (n=153) and men (n=100) |
Table 2 Univariate correlations of resting energy expenditure and body composition in lean and obese women (n=153) and men (n=100) |
Table 3 Results of multiple linear regression analysis with resting expenditure as dependent variable and fat-free mass, fat mass, and age as independent variables |
 |
 |
 |
| Received 14 June 1999; revised 30 September 1999; accepted 10 March 2000 |
 |
| September 2000, Volume 24, Number 9, Pages 1153-1157 |
 |
| Table of contents Previous Article Next [PDF] |
|
|