The Journal of Nutrition Vol. 128 No. 6 June 1998, pp. 1008-1010

Fraction of Carbon-Free Body Mass as Oxygen Is a Constant Body Composition Ratio in Men^1,2

ZiMian Wang³, Paul Deurenberg^*, Wei Wang, Richard N. Pierson Jr., and Steven B. Heymsfield

Department of Medicine, St. Luke's-Roosevelt Hospital, Columbia University College of Physicians and Surgeons, New York, NY and ^* Department of Human Nutrition and Epidemiology, Wageningen Agricultural University, The Netherlands

	ABSTRACT

Abstract Introduction Methods Results Discussion References

Although elements are the foundation of the human body, information concerning the atomic level of body composition is still limited. The aim of this study was to explore potentially constant relationships among elements found in vivo. Based on the known stoichiometries of relevant chemical components, a theoretical model was derived, suggesting the existence of a relatively constant ratio of total body oxygen to carbon-free body mass (TBO/CFM) in men. Eight elements (C, H, N, Ca, P, K, Na and Cl ) were measured in 22 healthy male subjects by using in vivo neutron activation-⁴⁰K whole-body counting, and TBO was calculated as the difference between body mass and the sum of the eight measured elements. TBO (in kg) was significantly correlated with CFM (in kg): TBO = 0.829 × CFM  1.8; r = 0.998, P < 0.001, standard error of estimate = 0.4 kg. The ratio of TBO to CFM was relatively constant, mean ± SD at 0.800 ± 0.009 with a CV of 1.1%. Oxygen and carbon are the two most abundant elements in the human body. The discovery of a constant relationship between oxygen and carbon is not only helpful for understanding the atomic level of body composition, but also provides the possibility of estimating the content of specific elements in vivo.

	INTRODUCTION

Abstract Introduction Methods Results Discussion References

One of the primary aims of body composition research is to discover body composition "rules" or quantitative relationships among components that are constant during steady-state periods (Wang et al. 1992). Of the known body composition rules, the best established is the relatively stable ratio of total body water to fat-free body mass (TBW/FFM).⁴ At the molecular level, body mass (BM) can be expressed as the sum of fat and FFM. The fat-free compartment is not homogeneous but contains water, protein, minerals and glycogen. Water, the largest component within FFM, occupies a relatively constant fraction (~0.73) of FFM in humans and many mammals (Sheng and Huggins 1979). The constancy of FFM hydration is determined by both total body water and fat contents

(1)

An interesting and reasonable question thus arises: are there corresponding constant ratios at the atomic body composition level? Oxygen is the most abundant element of water (88.9% of water is oxygen) and carbon is the most abundant element of fat (75.9% of fat is carbon). This observation stimulated us to explore the potentially constant TBO/CFM ratio at the atomic level

(2)

where TBO and TBC are total body oxygen and carbon, respectively; and CFM is carbon-free body mass that is equal to the sum of all elements except carbon. Hence

(3)

and

We hypothesize that the TBO/CFM ratio is constant in healthy men. The aim of this investigation is to explore the constancy of the TBO to CFM ratio.

	SUBJECTS AND METHODS

Abstract Introduction Methods Results Discussion References

TBO/CFM model derivation.  At the molecular level body mass can be expressed as The chemical stoichiometry of these components are C₅₁H₉₈O₆ for fat, H₂O for water, C₁₀₀H₁₅₉N₂₆O₃₂S_0.7 for protein, and (C₆H₁₀O₅)_x for glycogen (Heymsfield et al. 1991). There are constant fractions of oxygen in these components: 0.119 for fat, 0.889 for water, 0.227 for protein, 0.428 for mineral and 0.494 for glycogen. Hence, TBO can be derived as Both in vitro and in vivo studies showed that the FFM fractions as water, protein, mineral and glycogen are relatively constant at 0.732, 0.187, 0.073 and 0.008, respectively (Snyder et al. 1975). Thus, the above TBO equation may be rewritten as

Similarly, carbon fractions are constant at 0.759 in fat, 0.532 in protein, 0.020 in mineral and 0.444 in glycogen (Heymsfield et al. 1991). Thus the total body carbon equation is CFM is equal to the difference between body mass and total body carbon Therefore, where f_fat and f_FFM are the fractions of BM as fat and FFM, respectively. Because f_FFM = 1  f_fat, According to Equation 9, when f_fat increases from 0.05 to 0.40, the TBO/CFM ratio decreases slightly from 0.809 to 0.765. This theoretical model therefore suggests that the TBO/CFM ratio is relatively constant in the human body over a wide range of fatness.

Subjects.  The subject pool consisted of 22 healthy male adults with body fat from 6.9% to 38.6%. The rational for inclusion of subjects with low or high fatness was that possible limitations of the derived TBO/CFM model (Equation 9) may be observed that are not evident in subjects with normal fatness. Subjects were recruited from the hospital staff and the local university. Each subject completed a medical history, physical examination and routine screening blood studies to exclude the presence of underlying disease. The subjects of this investigation participated in other body composition studies and signed an informed consent that was approved by the hospital's Institutional Review Board. The observations described in this report were not included in our early investigations.

Body composition measurements.  Consenting subjects were studied after an overnight fast. Body mass was measured to the nearest 0.1 kg and height to 0.5 cm.

Elements. Total body content of eight elements was quantified using the in vivo neutron activation facility at Brookhaven National Laboratory with precisions from 1.2 to 4.5% (Ma et al. 1993). Total body carbon was measured using an inelastic neutron scattering system. Total body calcium, phosphorus, sodium and chlorine were measured using a delayed- neutron activation system. Total body nitrogen and hydrogen were evaluated using a prompt- neutron activation system. Total body potassium was estimated using the St. Luke's 4 -whole body ⁴⁰K counter with a precision of 3.2% (Pierson et al. 1990).

Nine main elements (O, C, H, N, Ca, P, K, Na and Cl ) occupy 99.3% of body mass (Snyder et al. 1975). Total body oxygen was thus calculated as the difference between 99.3% of body mass and the sum of the eight elements (C, H, N, Ca, P, K, Na and Cl ).

Fat. Body fat was measured by dual energy X-ray absorptiometry (DXA). A whole-body DXA scanner (Lunar DPX with software version 3.6, Madison, WI) was used to measure total body fat mass with a precision of 3% (Russel-Aulet et al. 1991).

Statistical analysis.  Results are expressed as group mean and SD. Paired Student's t tests were applied to compare the measured and calculated TBO/CFM ratios, and P < 0.05 was considered significant. Simple linear regression analysis was used to describe the relationship between TBO and CFM. The difference between measured and calculated TBO/CFM ratios was then related to the mean value of the two TBO/CFM ratios, as described by Bland and Altman (1986). Data were analyzed using SAS version 5 (SAS Institute, Cary, NC).

	RESULTS

Abstract Introduction Methods Results Discussion References

The baseline characteristics and total body element content of the subject group are presented in Table 1. Body mass was 79.8 ± 12.9 kg and total body oxygen and carbon-free body mass were 49.9 ± 9.6 and 62.3 ± 11.6 kg, respectively. There was a significant correlation between TBO and CFM (Fig. 1),

(10)

where r = 0.998, P < 0.001 and standard error of estimate (SEE) = 0.4 kg.

View larger version (13K): [in this window] [in a new window]   Fig 1. Simple linear regression between total body oxygen content (TBO, in kg) and carbon-free body mass (CFM, in kg), in men. TBO = 0.829 × CFM  1.8; r = 0.998, P < 0.001, standard error of estimate (SEE) = 0.4 kg, n = 22.

The TBO/CFM ratio measured by in vivo neutron activation analysis was 0.800 ± 0.009. This value is very close to the TBO/CFM ratio (0.793 ± 0.010) calculated from fatness (fat %) by the theoretical model (Equation 9), although there is a significant difference between the measured and calculated ratios (P < 0.001). The correlation between measured and calculated TBO/CFM ratios was strong (r = 0.76, P < 0.001). The Bland and Altman analysis (1986) indicated that the difference between measured and calculated TBO/CFM ratios was not significantly related to the mean value of the two TBO/CFM ratios (r = 0.25, P > 0.05).

The theoretical model (Equation 9) suggests that fatness is the major factor influencing the TBO/CFM ratio. Simple linear regression analysis confirmed that the measured TBO/CFM ratio is significantly correlated with percentage of body weight as fat (r = 0.76, P < 0.001), indicating that 58% of the variability in the TBO/CFM ratio can be explained by fatness.

	DISCUSSION

Abstract Introduction Methods Results Discussion References

Predicted and measured TBO/CFM ratios.  In this study, we derived a theoretical model suggesting a relatively constant TBO/CFM ratio in healthy male subjects. The predicted TBO/CFM ratio was very close to the measured value (0.793 vs. 0.800), thus supporting our proposed model. The small but significant difference between the predicted and measured TBO/CFM ratios may be caused by both measurement errors for the eight elements and model assumptions. In deriving our model, we assumed that the FFM fractions as water, protein and mineral are constant at 0.732, 0.187 and 0.073, respectively. These fractions actually vary within a narrow range among subjects. This may cause individual differences between predicted and measured TBO/CFM ratios.

Our proposed model revealed that the TBO/CFM ratio is a function of body fat fraction (Equation 9), and this function is close to a straight line (TBO/CFM = 0.80). When fat fraction increases within a large range (e.g., 0.05-0.40), the TBO/CFM decreases only slightly from 0.81 to 0.77. We were thus able to propose a relatively constant TBO/CFM ratio (0.80) for healthy male subjects. Although not well studied at present, a similar small variation with body mass may be present in the classical TBW/FFM ratio. Increasing adiposity is associated with a relative increase in hydration which may in turn have a small influence on the TBW/FFM ratio.

Application in elemental measurement.  Although elements are the foundation of the human body, information on the atomic level of body composition is still limited. Neutron activation analysis at present is the only means of measuring total body element content in vivo except for potassium and hydrogen. However, neutron activation is associated with moderate radiation, so that some subject groups such as children and young women cannot be tested. Moreover, there are only a few in vivo neutron activation facilities in the world. Therefore, the measurement of total body elemental content is to a great extent limited. An alternative is to estimate elements by establishing relationships to measurable components. Exploring body composition rules at the atomic level provides the possibility of revealing potentially constant relationships among components. Thus an unknown element might be calculated from a measurable element. Previous studies reported that total body hydrogen occupies a constant 10% of body weight (Vartsky et al. 1984). This study suggests a new rule: total body oxygen occupies 80% of carbon-free body mass. Oxygen, carbon and hydrogen are the three most abundant elements in the human body. If carbon is known, the elemental composition of ~94% of body mass (i.e., the sum of O, C and H) can thus be determined.

Both theoretical analysis and experimental measurement lead us to a relatively constant TBO/CFM ratio for healthy male subjects. Further studies are required to confirm the TBO/CFM ratio in females and in the case of different diseases such as AIDS. This investigation indicated that although there are ~50 elements in the human body, they are well organized according to definable quantitative relations.

	FOOTNOTES

Manuscript received 24 October 1997. Initial reviews completed 2 December 1997. Revision accepted 11 February 1998.

	LITERATURE CITED

Abstract Introduction Methods Results Discussion References

• Bland J. M., Altman D. G. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986; 8:307-310
• Heymsfield S. B., Waki M., Kehayias J., Lichtman S., Dilmanian F. A., Kamen Y., Wang J., Pierson R. N. Jr. Chemical and elemental analysis of humans in vivo using improved body composition models. Am. J. Physiol. 1991; 261:E190-E198[Abstract/Free Full Text]
• Ma, R., Dilmanian, F. A., Rarback, H., Stamatelatos, I. E., Meron, M., Kamen, Y., Yasumura, S., Weber, D. A., Lidofsky, L. J. & Pierson, R. N., Jr. (1993) Resent upgrade of the IVNA facility at BNL. In: Human Body Composition, In Vivo Methods, Models, and Assessment (Ellis, K.J. & Eastman, J.D., eds.), pp. 345-350. Plenum Press, New York, NY.
• Pierson, R. N., Jr., Wang, J., Heymsfield, S. B., Dimanian, F. A. & Weber, D. A.. (1990) High precision in vivo neutron activation analysis: a new era for compartmental analysis in body composition. In: In vivo Body Composition Studies (Yasumura, S., Harrison, J.E., McNeil, K.G., Woodhead, A.D. & Dilmanian, F.A., eds. ), pp. 317-325. Plenum Press, New York, NY.
• Russel-Aulet M., Wang J., Thornton J., Colt E.W.D., Pierson R. N. Jr. Bone mineral density and mass by total body dual photon absorptiometry in normal Caucasian and Asian men. J. Bone Miner. Res. 1991; 10:1009-1113
• Sheng H. P., Huggins R. A. A review of body composition studies with emphasis on total body water and fat. Am. J. Clin. Nutr. 1979; 32:630-647[Abstract/Free Full Text]
• Snyder, W. S., Cook, M. J., Nasset, E. S., Karhausen, L. R., Howells, G. P. & Tipton, I. H. (1975) Report of the Task Group on Reference Man. Pergamon Press, Oxford, UK.
• Vartsky D., Ellis K. J., Vaswani A. N., Yasumura S., Cohn S. H. An improved calibration for in vivo determination of body nitrogen, hydrogen and fat. Phys. Med. Biol. 1984; 29:209-218[Medline]
• Wang Z. M., Pierson R. N. Jr., Heymsfield S. B. The five-level model: a new approach to organizing body composition research. Am. J. Clin. Nutr. 1992; 56:19-28[Abstract/Free Full Text]

0022-3166/98 $3.00 ©1998 American Society for Nutritional Sciences