LIPID CONTENT AND COMPOSITION OF ANIMAL 563 



Brozek 166 suggest that sulfanilamide might be a satisfactory test solute for 

 the estimation of body water if an empirical correction factor were applied. 



Tritium-labeled water has been recommended as having some advantages 

 over deuterium oxide as a test substance. 202 - 239 Prentice et al. 240 found that 

 the exchange of tritium with labile organic hydrogen is small or undetect- 

 able in fat, somewhat greater in muscle and gut, and considerable in liver, 

 kidney, and plasma solids. Tritium appears to measure the same water 

 volume as deuterium, averaging 2 to 4% higher than with antipyrine, and 

 undergoes a similar degree of exchange with organic molecules as does 

 deuterium. The exchange error can be eliminated by a constant correc- 

 tion factor. Other disadvantages of tritium-labeled water include its ex- 

 pense, the complication of its analysis, and the fact that some question 

 arises as to the health hazard to man entailed by its long half-life (twelve 

 years). 214 Presumably no danger factor is involved by its use in a single 

 test. 240 Moreover, no extensive data are as yet available as to the reli- 

 ability and uniformity of the results obtained by the employment of this 

 radioisotope. 



Brodie and co-workers 241 proposed the use of A r -acetyl-4-aminopyrine 

 (NAAP) instead of antipyrine for the estimation of body water. In con- 

 tradistinction to antipyrine, NAAP does not combine with proteins to an 

 appreciable extent. 



b'. Formula for the Calculation of Fat Content from Body Water Data: 

 The general equation for the calculation of total fat from the calculation of 

 body water is as follows 166 : 



F = 1 - kA 



in which F is body fat as a proportion of body weight, A is total body water 

 as a proportion of body weight, and A; is a constant. The value of k has 

 been assumed to be 1.393, which is the ratio of 1.000/0.718; the denomina- 

 tor is the assumed percentage of water in fat-free tissue. 196 The figures ob- 

 tained by the use of this equation gave a standard error of ±3.8% of the 

 body weight as fat, when compared with the results obtained from densio- 

 metry. Keys and Brozek 166 proposed a value of 1.4 for k based upon a 

 water content of 71% in fat-free tissue. 



However, according to Keys and Brozek, 166 data are still unavailable to 

 support a final conclusion as to the absolute validity of the estimation of 



239 E. A. Pinson and E. C. Anderson, Am. J. Physiol, 163, 741 (1950). 



24 ° T. C. Prentice, W. Siri, N. I. Berlin, G. M. Hyde, R. J. Parsons, E. E. Joiner, and 

 J. H. Lawrence, J. Clin. Invest, 31, 412-418 (1952). 



241 B. B. Brodie, E. Y. Berger, J. Axelrod, M. F. Dunning, Y. Porosowska, and J. M. 

 Steele, Proc. Soc. Exptl. Biol. Med., 77, 794-798 (1951). 



