FOOD AND AGRICULTURE ORGANIZATION OF THE UNITED NATIONS	ESN: FAO/WHO/UNU EPR/81/7 August 1981
	WORLD HEALTH ORGANIZATION
	THE UNITED NATIONS UNIVERSITY

Provisional Agenda Item 2.2.3

Joint FAO/WHO/UNU Expert Consultation on
Energy and Protein Requirements

Rome, 5 to 17 October 1981

THE ENERGY COST OF GROWTH

by

J.C. Waterlow
London School of Hygiene
and Tropical Medicine
University of London
London, U.K.

Except just before and after birth, the proportion of energy supply used for growth is much less in man than in farm animals (table 1). It might be expected, therefore, that if the intake is marginal, the small amount of energy needed for growth could easily be made available by reductions in other routes of expenditure. The work of Torun (1980) shows that this is only true to a limited extent. When the intake of pre-school children was reduced from 100 to 90 kcal/kg/day, growth was maintained, but there was evidence of a fall in physical activity; when the intake was further reduced to 80 kcal/kg/day, weight gain ceased. Thus the small size of the energy requirement for growth does not make it less important; it only makes it more difficult to measure.

1. Theoretical estimates of the energy cost of growth

As is obvious, the energy cost of tissue deposition (E_g) can be divided into two parts : the energy stored (E_s) and the energy used in the synthesis of tissue components (E_m).

1.1. Energy stored The value of E_s will depend on the composition of the tissue gained. One can look at this either in terms of tissues - adipose and lean - or in terms of substances - principally triglycerides and protein. Schutz (1979) points out that adipose tissue contains some 15% of water, so that its energy value will be less than that of pure fat. Lean tissue contains about 5% of lipid in cell membranes and some 80% of water. If the heat of combustion of protein is taken as 5.7 kcal/g, and that of fat as 9.3 kcal/g, the energy content of lean tissue would be 1.6 kcal/g, and that of adipose 7.9 kcal/g.

According to the composition of the tissue gained, the observed energy cost of growth may be expected to vary over a wide range, from a minimum with pure lean to a maximum with pure adipose tissue. Jackson et al. (1977), in studies on children recovering from malnutrition, estimated the amount of lean tissue gained from isotopic measurements of muscle mass; the difference between this and the amount of weight gained represented deposition of fat. Thus, knowing the relative amounts of fat and lean, they were able to calculate the theoretical cost of weight gain. Good agreement was found between calculated and observed cost.

While accepting that there will be a range in the energy cost of growth, according to the composition of the tissue laid down, we should obviously aim at an appropriate balance between fat and lean. If the proportions laid down are 25% fat and 75% lean, the average energy stored per g will be (7.9 × .25)+(1.6 × .75) = 3.2 kcal.

1.2. The energy cost of synthesis (E_m) If triglycerides are provided in the diet, the energy cost of their hydrolysis in the gut and resynthesis in tissues is negligible. The conversion of carbohydrate to fat, and of amino acids to protein, on the basis of the ATP used and the stoichiometry of the reactions, require similar amounts of energy - 0.15 kcal per kcal deposited (Millward, Garlick & Reeds, 1976). On this basis the metabolic cost (E_m) of storing 3.2 kcal would be 0.48 kcal. Thus theoretically the total energy cost for balanced tissue gained would amount to about 3.7 kcal/g.

The estimate of E_m is likely to be too low, because no account is taken of the energy cost of synthesizing compounds other than fat, such as RNA. Moreover, there is evidence from studies in human infants and in young pigs that during rapid growth, for every g of protein deposited, 0.5–1 g of extra protein has to be synthesized, over and above the basal rate of protein synthesis in the non-growing state (Waterlow & Jackson 1981; Reeds et al. 1980). However, these sources of error are of little practical importance, since E_m is small compared with E_s.

2. Measured estimates of the energy cost of growth

Two methods are available for estimating E_g: from the regression of weight gain on energy intake, and by comparison of energy expenditure with intake.

2.1. Estimates from regression of weight gain on intake Kielanowski, Thorbeck and others have determined the energy cost of growth in farm animals from the regression of weight gain on metabolizable energy intake, and have attempted to calculate separately the energy costs of depositing protein and fat. Some of their results are discussed by Millward et al. (1976). From five studies on pigs and lambs, summarized by Spady et al. (1976) the estimated energy cost (E_g) of depositing 1 g fat is 12.0 kcal, and 1 g protein 8.7 kcal. This is the total cost. If we subtract the values for energy stored, the metabolic cost (E_m) by difference is 12 - 9.7 = 2.3 kcal/g for fat and 8.7 - 5.7 = 3.0 kcal/g for protein. As might be expected these estimates of metabolic cost, based on observed rates of tissue deposition, are higher than the theoretical costs based on the biochemical reactions. From these values the energy cost (E_g) of producing 1 g balanced tissue, as defined above, would be 4.9 kcal.

In humans this type of regression analysis has almost without exception been done in children recovering from malnutrition, because only in this way can growth rates be obtained which are large enough for analysis. This method, of course, gives the total cost E_g; it is not possible to separate the two components E_s and E_m without additional information. Schutz (1979) has suggested that it may not be legitimate to apply such results to normal children. I do not think this objection is valid; the recovering malnourished child certainly differs from the normal in its partition of energy: it spends more on growth, probably less on physical activity, and possibly more, rather than less, on maintenance, because of differences in body composition. However, the chemical reactions involved in the synthesis of protein and fat are presumably the same, and produce the same amount of heat.

Some of the results obtained in children are summarized in table 2. The range of variation is not surprising, considering differences in methods and subjects, and no doubt in the average composition of the tissue laid down. The mean of 5.0 kcal/g weight gain is in good agreement with the calculated value quoted above of 4.9 kcal/g for the deposition of balanced tissue in farm animals.

2.2. Estimates from measurements of energy expenditure The most detailed study that I know of by this method is that of Spady et al. (1976)*. They determined energy expenditure by the calibrated heart rate method. The difference between net intake (after allowing for faecal and urinary losses) represents stored energy. The energy stored per g weight gain (E_s) was 3.3±0.4 (SEM) kcal. From the regression on intake the total cost of growth (E_g) came to 4.4 kcal/g. Therefore by difference E_m = E_g - E_s = 4.4 - 3.3, = 1.1 kcal/g.

As a check on the reliability of these results, it is of interest that the average value for the maintenance energy requirement (zero weight gain) obtained from the regression was 85 kcal/kg/day, which is very close to the rule-of-thumb estimate suggested by Payne that maintenance requirement = 1.5 × BMR (see Waterlow et al. 1976). It should be noted that the value for E_m of 1.1 kcal/g, having been obtained by difference, must include the energy cost of any physical activity over and above that allowed for by the maintenance requirement.

*Footnote
Similar measurements have been made by Torun and co-workers but I am not aware of their findings, except as summarized by Schutz (1979).

Finally, if we know the amount of energy stored, it is simple to calculate from their energy values the relative amounts of fat and protein laid down. In this case, if E_s = 3.3 kcal/g, the proportion of fat must be 26.5%, and of protein 14.7%. It is then possible to calculate E_m by applying the previously quoted estimates obtained in animals for the cost of protein and fat deposition (Millward et al. 1976). Table 3 shows that the result agrees closely with that observed.

3. Conclusion

I see no reason to recommend any change in the rounded-off value previously used, of 5 kcal/g as an allowance for the energy cost of growth. This should support the deposition of tissue with balanced proportions offat and lean.

(1) Data from Widdowson, 1981.Total energy intake (column B) taken as energy equivalent ofO₂ utilized + energy content of weight gained
(2) Assuming ‘balanced’ tissue - see text
(3) From Spady et al. 1976
(4) From Millward et al. 1976

(1) Recalculated from author's figures,after allowing for faecal losses

REFERENCES

Ashworth A (1969) Br J Nutr 23, 835

Fomon S J (1971) Acta Pediatr Scand (Suppl) 223, 22

Jackson A A, Picou D & Reeds P J (1977) Am J Clin Nutr 30, 1514

Kerr D, Ashworth A, Picou D, Poulter N, Seakins A, Spady D & Wheeler E (1973) In : Endocrine Aspects of Malnutrition. Eds: L Gardner & P Amacher. Kroc Foundation, Santa Ynez, California. p467

Krieger L & Whitten C F (1976) Am J Clin Nutr 29, 38

Millward D J, Garlick P J & Reeds P J (1976) Proc Nutr Soc 35, 339

Reeds P J, Cadenhead A, Fuller M F, Lobley G E & McDonald J D (1980) Br J Nutr 43, 445

Schutz Y (1979) In: Wissenschaftliche Arbeiten und Beitrage aus dem Kreise der Stiftung zur Forderung der Ernahrungsforschung in der Schweiz 1974–79. Eds: H E Aebi & R M Kunz. Hans Huber, Bern. p113 Beiheft zur Internationalen Zeitschrift fur Vitamin- und Ernahrungs-forschung Nr.20

Spady D W, Payne P R, Picou D & Waterlow J C (1976) Am J Clin Nutr 29, 1073

Torun B (1980) In: Report on the informal gathering of investigators to review the collaborative research programme on protein requirements and energy intake. ESN/MISC/80/3 FAO, Rome

Waterlow J C, Ashworth Hill A & Spady D W (1976) In: Early Nutrition and Later Development. Ed: A. W. Wilkinson. Pitman Medical Publishing Co Ltd, Tunbridge Wells. p175

Waterlow J C & Jackson A A (1981) Br Med Bull. 37, 5

Whitehead R G (1973) In: Proteins in Human Nutrition. Eds: J W Porter & B A Rolls. Academic Press, London. p103

Widdowson E M (1981) In: Maternal Nutrition in Pregnancy - eating for two? Ed: J Dobbing. Academic Press, London. p1