Multicomponent methods: evaluation of new and traditional soft tissue mineral models by in vivo neutron activation analysis

¹ From the Department of Medicine, Obesity Research Center, St Luke’s–Roosevelt Hospital, Columbia University, College of Physicians and Surgeons, New York (ZMW, FXP-S, DPK, RNP, and SBH); the Department of Applied Science, Brookhaven National Laboratory, Upton, NY (LW); and the Exercise Physiology Laboratory, Flinders University, Adelaide, Australia (RTW).

² Supported by National Institutes of Health grant NIDDK 42618.

³ Reprints not available. Address correspondence to ZM Wang, Weight Control Unit, 1090 Amsterdam Avenue, 14th Floor, New York, NY 10025. E-mail: zw28{at}columbia.edu.

ABSTRACT  
Background: Practical and accurate methods for quantifying the soft tissue mineral component of multicomponent fat-estimation models are needed.

Objectives: The aims were to develop a new complete model for estimating soft tissue minerals based on measured total body water (TBW) and extracellular water (ECW) and a simplified new model based on TBW measurements only and to compare these estimates with those determined with 2 traditional models (ie, the Broek and Selinger models) and with criterion estimates based on in vivo neutron activation (IVNA) analysis.

Design: The subjects were 156 healthy adults and 50 patients with AIDS. Total body potassium, sodium, chlorine, and calcium were measured by IVNA; TBW by ³H₂O or D₂O dilution; ECW by bromide dilution; and bone mineral by dual-energy X-ray absorptiometry.

Results: The mean (± SD) mass of total-body soft tissue minerals in healthy adults was 467 ± 62 g with the IVNA model, 492 ± 62 g with the new model, and 487 ± 59 g with the simplified new model. Compared with the IVNA model, the complete and simplified new models overestimated soft tissue minerals by 5.4% and 4.6% (both P < 0.001), respectively. In contrast, the Broek and Selinger models overestimated overall mean soft tissue minerals by 35% and 99% (both P < 0.001), respectively. Overall results for soft tissue mineral prediction with the 2 new models were less satisfactory for the patients with AIDS, although the results were better than those with the traditional models.

Conclusions: The physiologically formulated complete new model for estimating soft tissue minerals provides the opportunity to upgrade the accuracy of current multicomponent models for estimating total body fat.

Key Words: Extracellular water • intracellular water • body composition • multicomponent methods • neutron activation analysis • soft tissue mineral • AIDS

INTRODUCTION  
Early body-composition investigators introduced the 2-component model in which body mass is divided into fat and fat-free mass (ie, body mass = fat + fat-free mass) (1, 2). An important recent advance is the introduction of multicomponent models that partition body mass into 3 components (3). Because the addition of measured components usually reduces the number of applied assumptions, multicomponent models are often considered the criterion against which other methods of estimating total body fat are validated.

One important group of multicomponent models is based on body-volume measurements that are usually estimated by hydrodensitometry or air plethysmography (4, 5). Body-volume estimates are used in one term of the classical 2-component model (4, 5) that serves as the basis for this group of multicomponent models. The addition of an estimate of total body water (TBW) by isotope dilution allows the development of a 3-component model (1, 2). The 3-component model can then be extended to a 4-component model by adding an estimate of bone mineral by dual-energy X-ray absorptiometry (DXA) (3, 6). Three- and 4-component models are now widely applied in body-composition laboratories throughout the world. However, both the 3- and 4-component models do not include a discrete estimate of soft tissue mineral, a small but important molecular level component (7).

Soft tissue minerals consist largely of soluble minerals and electrolytes found in the extracellular and intracellular compartments of soft tissue. Although the mass of soft tissue minerals (400 g) is relatively small in adults, its contribution to body density should be considered because soft tissue minerals collectively have a higher density (3.317 g/cm³) at normal body temperature than do each of the other components, including fat (0.900 g/cm³), water (0.994 g/cm³), protein (1.34 g/cm³), and bone mineral (2.982 g/cm³) (8).

There are currently only 3 in vitro cadaver studies that report estimates of total-body soft tissue minerals (9–11). On the basis of these in vitro studies, Broek and Selinger developed 2 traditional models to predict total-body soft tissue mineral mass (1, 12). No previous studies have critically evaluated or challenged the 2 prevailing models of soft tissue minerals as applied in conventional multicomponent-model methods (7, 13). Although total-body soft tissue mineral mass can be measured in vivo by delayed- in vivo neutron activation (IVNA) analysis, this method cannot be applied in most laboratories because of the complex instrumentation required and the radiation exposure that precludes use in some subject groups (14).

The aim of the present study was to advance a new physiologically formulated method of quantifying total-body soft tissue mineral mass. Our intent in developing this approach was to provide a practical means of estimating soft tissue mineral mass in vivo. We examined results provided by the new model and a simplified version of the new model along with the previously reported traditional models of Broek and Selinger with the use of IVNA estimates as the criterion.

METHODS  
Soft tissue mineral models
Three models are currently used to estimate soft tissue minerals: the IVNA, Broek, and Selinger models. Each of these models is described below, as is a new model for estimating soft tissue minerals.

IVNA model
The established criterion method is to estimate soft tissue minerals separately by IVNA and whole-body counting. Total-body sodium (TBNa) and total-body calcium (TBCa) are measured with delayed- IVNA. Sodium is found as a cation in soft tissues and is also bound to the crystalline matrix of bone mineral (15, 16). Because the ratio of sodium to calcium in bone mineral is known (15), bone mineral sodium can be estimated from TBCa as 0.038 x TBCa. Sodium in soft tissues can then be calculated as the difference between TBNa and sodium in bone mineral. With the use of similar approaches, 6 main soft tissue minerals electrolytes (K⁺, Na⁺, Mg²⁺, Cl^- , H₂PO₄^- , and HCO₃^- ) can be calculated from 4 measurable elements (potassium, sodium, chlorine, and calcium) and then summed for total-body soft tissue minerals (Ms) mass (17, 18):

RESULTS  
Physical characteristics and body composition
A total of 206 subjects were evaluated in 2 groups: healthy adults and patients with AIDS (Table 2). The 156 healthy adults ranged in age from 25 to 74 y, in body mass from 55.0 to 116.1 kg, and in body mass index (BMI; in kg/m²) from 21.7 to 50.1. The 50 patients with AIDS ranged in age from 22 to 66 y, in body mass from 44.7 to 76.3 kg, and in BMI from 17.6 to 28.8. The healthy adults were older, were heavier, and had a greater BMI and percentage body fat than did the AIDS patients (all P < 0.01–0.001). The body-composition results for the 2 groups are presented in Table 2. TBK, TBNa, TBCl, bone mineral, TBW, and ECW were greater in the healthy adults than in the AIDS patients (all P 0.05), although there was no significant difference in TBCa between the 2 groups.

View this table:
TABLE 2 . Physical characteristics and body-composition results for the 2 subject groups¹  
Total-body soft tissue mineral measurements
IVNA model
Total-body soft tissue minerals determined with the IVNA model were 467 ± 62 g in the healthy subjects and 402 ± 93 g in the AIDS patients (Table 3).

View this table:
TABLE 3 . Total-body soft tissue minerals assessed by the five models¹  
New model
The total-body soft tissue mineral component predicted by the new model (ie, Equation 6) was 492 ± 62 g in the healthy subjects (Table 3), an average overestimate of 25 g, or 5.4% (P < 0.001). The estimates of soft tissue minerals by the new model were highly correlated with estimates of soft tissue minerals by the IVNA model in the healthy subjects:

DISCUSSION  
The relatively small and chemically diverse soft tissue mineral component is not easily quantified in vivo, even at specialized laboratories that measure body composition. In the present study we derived a physiologically based soft tissue mineral model along with a simplified form and we showed good agreement with the IVNA criterion approach in healthy subjects. These supportive findings thus provide the opportunity to upgrade current body volume (BV)–based 4-component models. Accordingly, simultaneous body mass (in kg) and body volume (in L) models can be written as follows:

REFERENCES  

1. Broek J, Grande F, Anderson JT, Keys A. Densitometric analysis of body composition: revision of some quantitative assumptions. Ann N Y Acad Sci 1963;110:113–40.
2. Siri WE. Body composition from fluid spaces and density: analysis of methods. In: Broek J, Henschel A, eds. Techniques for measuring body composition. Washington, DC: National Academy of Sciences, National Research Council, 1961:223–44.
3. Withers RT, Laforgia J, Heymsfield SB, Wang ZM, Pillans RK. Two, three and four-compartment chemical models of body composition analysis. In: Norton K, Olds T, eds. Anthropometrica. Australia: UNSW Press, 1996:199–231.
4. Going SB. Densitometry. In: Roche AF, Heymsfield SB, Lohman TG, eds. Human body composition. Champaign, IL: Human Kinetics, 1996:3–23.
5. Nunez C, Kovera AJ, Pietrobelli A, et al. Body composition in children and adults by air displacement plethysmography. Eur J Clin Nutr 1999;53:382–7.
6. Fuller NJ, Jebb SA, Laskey MA, Coward WA. Four compartment model for the assessment of body composition in humans: comparison with alternative methods, and evaluation of the density and hydration of the fat-free mass. Clin Sci 1992;82:687–93.
7. Heymsfield SB, Wang ZM, Withers RT. Multicomponent molecular level models of body composition analysis. In: Roche AF, Heymsfield SB, Lohman TG, eds. Human body composition. Champaign, IL: Human Kinetics, 1996:129–47.
8. Heymsfield SB, Waki M, Kehayias J, et al. Chemical and elemental analysis of humans in vivo using improved body composition models. Am J Physiol 1991;261:E190–8.
9. Forbes RM, Cooper AR, Mitchell HH. The composition of the adult human body as determined by chemical analysis. J Biol Chem 1953;203:359–66.
10. Mitchell HH, Hamilton TS, Steggerda FR, Bean HW. The chemical composition of the adult human body and its bearing on the biochemistry of growth. J Biol Chem 1963;158:625–37.
11. Widdowson EM, McCance RA, Spray CM. The chemical composition of the human body. Clin Sci 1951;10:113–25.
12. Selinger A. The body as a three component system. PhD thesis. University of Illinois, Urbana, 1977.
13. Baumgartner RN, Heymsfield SB, Lichtman S, Wang J, Pierson RN Jr. Body composition in elderly people: effect of criterion predictive estimates on equations. Am J Clin Nutr 1991;53:1345–53.
14. Cohn SH, Vaswani AN, Yasumura S, Yuen K, Ellis KJ. Improved models for determination of body fat by in vivo neutron activation. Am J Clin Nutr 1984;40:255–9.
15. Woodard HQ. The elementary composition of human cortical bone. Health Physics 1962;8:513–7.
16. Woodard HQ. The composition of human cortical bone. Clin Orthop 1962;37:187–93.
17. Heymsfield SB, Lichtman S, Baumgartner RN, et al. Body composition of humans: comparison of two improved four-compartment models that differ in expense, technical complexity, and radiation exposure. Am J Clin Nutr 1990;52:52–8.
18. Wang ZM, Ma R, Pierson RN Jr, Heymsfield SB. Five-level model: reconstruction of body weight at atomic, molecular, cellular, and tissue-system levels from neutron activation analysis. In: Ellis KJ, Eastman JD, eds. Human body composition, in vivo methods, models, and assessment. New York: Plenum Press, 1993:125–8.
19. Rhoades R, Pflanzer R. Human physiology. Philadelphia: Saunders, 1989:754–79.
20. Maffy RH. The body fluids: volume, composition, and physical chemistry. In: Brenner BM, Rector FC, eds. The kidney. Vol 1. Philadelphia: WB Saunders, 1976:65–103.
21. Dick DAT. Cell water. Washington, DC: Butterworths, 1966.
22. Olmstead ED. Mammalian cell water. Philadelphia: Lea & Febiger, 1966.
23. Wang ZM, Deurenberg P, Wang W, Pietrobelli A, Baumgartner RN, Heymsfield SB. Hydration of fat-free body mass: new physiological modeling approach. Am J Physiol 1999;276:E995–1003.
24. Dilmanian FA, Weber DA, Yasumura S, et al. Performance of the delayed- and prompt-gamma neutron activation systems at Brookhaven National Laboratory. In: Yasumura S, Harrison JE, McNeill KG, Woodhead AD, Dilmanian FA, eds. Advances in in vivo body composition studies. New York: Plenum Press, 1990:309–15.
25. Pierson RN Jr, Wang J, Heymsfield SB, Dilmanian FA, Weber DA. High precision in vivo neutron activation analysis: a new era for compartmental analysis in body composition. In: Yasumura S, Harrison JE, McNeill KG, Woodhead AD, Dilmanian FA, eds. Advances in in vivo body composition studies. New York: Plenum, 1990:317–25.
26. Schoeller DA. Hydrometry. In: Roche AF, Heymsfield SB, Lohman TG, eds. Human body composition. Champaign, IL: Human Kinetics, 1996:25–43.
27. Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986;8:307–10.
28. Lohman TG. Advances in body composition assessment. Champaign, IL: Human Kinetics, 1992:22.
29. Forbes GB. Human body composition: growth, aging, nutrition, and activity. New York: Springer-Verlag, 1987.
30. Ma K, Kotler DP, Wang J, Thornton JC, Ma R, Pierson RN Jr. Reliability of in vivo neutron activation analysis for measuring body composition: comparisons with tracer dilution and dual-energy X-ray absorptiometry. J Lab Clin Med 1996;127:420–7.