Determinants of energy density with conventional foods and artificial feeds.

The energy densities of conventional foods (Paul & Southgate, 1978) and artificial feeds (Livesey & Elia, 1985~) are usually calculated from their compositions, i.e. protein, fat and carbohydrate contents, and the energy conversion factors for these components. Most often the factors are metabolizable energy (ME) values; that is an amount of energy thought useful for doing physical or metabolic work. This quantity has been formally equated to the difference in the heat of combustion of food and excreta, faeces and urine, for subjects in nitrogen balance. In animal nutrition (Agricultural Research Council, 1980; Blaxter, 1989) excreta also includes the combustible gases hydrogen and methane, but in human nutrition these losses are usually small and ignored. Energy conversion factors for substrates which undergo fermentation in the human large intestine, the sugar alcohols (Dutch Nutrition Council, 1987) and certain complex carbohydrates (British Nutrition Foundation, 1990) required a different approach. In these cases net energy (NE) values for physical and metabolic work were derived. These values exclude: losses of combustible energy in gases, losses of energy to heat of fermentation (spent on the growth and maintenance of colonic micro-organisms) and losses of energy, relative to glucose or sucrose, on producing ATP during the oxidation of short-chain fatty acids (the products of carbohydrate fermentation absorbed from the colon). To prevent confusion it is important to distinguish between values for NE, ME, heat of combustion or gross energy (GE) and digestible energy (DE), in human nutrition the last usually discounts losses to faeces only. This distinction is made here by including abbreviations with units. When the values (NE, ME, DE, GE) are thought to be similar more than one abbreviation may appear with the units. Here NE always refers to energy useful for physical and metabolic work and should not be confused with values used in animal production such as NE for growth, reproduction, lactation etc. (Blaxter, 1989). Further, in modern scientific works it is preferable to use kJ rather than kcal, an exception is made here because of the historical perspective on which the review is built. Presently, there remains only one well-established energy loss not embodied in an officially recognized food energy assessment system; it is the loss due to urea synthesis