Proceedings zyxwvutsrqpo of zyxwvutsrq tile Nufrition zyxwvutsrqpo Society zyxwvutsrq (1990) zyxwvutsr 49,423436 423 Body composition: the precision and accuracy of new methods and their suitability for longitudinal studies BY M. F. FULLER', P. A. FOWLER2, G. McNEILLl AND M. A. FOSTER2 'Rowett Research Institute, Bucksburn, Aberdeen AB2 9SB 2University of Aberdeen, Department of Biomedical Physics and Bioengineering, Foresterhill, Aberdeen AB9 2 2 0 The ability to estimate body composition is of central importance in studies of growth and nutrition, in both animals and man. However, what we mean by body composition and what we wish to estimate differ according to how we view the body. Anatomists may think in terms of the sizes of organs and tissues, physiologists in terms of cellular and extracellular components, nutritionists in terms of nitrogen, lipid or energy contents, animal breeders or butchers in terms of meat, fat and bone. The methods we choose to estimate body composition reflect these different viewpoints. Our aims also dictate the precision we require of our techniques and, since it is generally true that precision is expensive (Kempster, 1984), we need at the outset to define our aims rather carefully, in the form of two questions: what information is needed? how accurate and how precise must it be? APPROACHES TO DESCRIBING BODY COMPOSITION Methods of estimating body composition, whether in living subjects or destructively, have been of two kinds, those which describe the chemical composition of the body and those which describe the anatomical distribution of tissues. In the first, the body is treated as if it contained a homogeneous mixture of substances, and, for the purposes of estimation, it would be far better if it did, for inhomogeneity is a major source of difficulty in many of these methods. Destructive techniques of this kind are exemplified by homogenizing the whole carcass and determining the water, N and lipid contents. In the second category, interest centres on determining the sizes and weights of individual bones, muscles and so on, classically by dissection. For much nutritional work methods in the first category have been of most interest, but with greater understanding of the factors controlling tissue growth it becomes increasingly important to have information of both kinds. Recent findings on the effects of adipose tissue distribution on the metabolic complications of obesity (Kissebah et al. 1982; Krotkiewski et al. 1983) and on regional differences in lipid storage and mobilization (Rebuffe-Scrive et al. 1985) emphasize this need. For studies with small and eyen large animals the destructive technique of comparative slaughter with dissection or homogenization of the carcass is very widely used and is often the most cost-effective method. The need for non- destructive techniques arises: (1) in human studies, (2) with valuable animals, (3) where the sequential study of individuals is necessary or desirable.