FOOD AND AGRICULTURE ORGANIZATION OF THE UNITED NATIONS	ESN: FAO/WHO/UNU EPR/81/33 August 1981
	WORLD HEALTH ORGANIZATION
	THE UNITED NATIONS UNIVERSITY

Item 3.2.5 of the Provisional Agenda

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

Rome, 5 to 17 October 1981

ENVIRONMENTAL STRESSORS AND PROTEIN UTILIZATION

by

R.G. Whitehead
MRC Dunn Nutrition Unit, Cambridge UK

1. Introduction

1.1 Infections together with physical and psychological trauma undoubtedly do affect nutrient utilization by the individual: this fact is not in dispute. The issue is whether any special increments need to be made to energy and nutrient allowances for communities, or groups within communities, where infection prevalence is high. An extension to this consideration is how such changes could be effected in practice, especially the protection of vulnerable groups.

1.2 Most infections alter energy and nutrient balance leading to weight loss and, in the immature person, growth retardation. There are a number of ways in which the effects of infection interact directly with diet, but of most importance in the present context are the anorexia with which infection is frequently associated, followed by the need to provide sufficient extra dietary energy and nutrients to permit recovery and catch-up growth after the infection is over.

1.3 It has become customary to reason that the recommended allowance should be sufficient to cover the effects of commonly occurring mild infections and thus no extra dietary protein or energy need ever be considered for communities in the more wealthy countries of the world, where general levels of health and hygiene are high. FAO/WHO committees prior to 1971 did, however, conclude that many developing countries, where the incidence of relatively severe infections such as infantile diarrhoeas and malaria is high, represent a spacial case. Thus, for example, a special allowance was made for protein to cover such situations, but the increase was largely arbitrary and the 1971 committee reasoned that it was illogical to make allowance on such a basis. A later Ad Hoc Committee (1975) set up to deal with practical issues arising from the 1973 Report confirmed this view, and furthermore considered that additional allowances could actually be harmful in that they might convey to governmental planners that simple dietary measures alone were capable of overcoming the worst effects of infection. It was feared this might inhibit the introduction of more direct public health measures to combat infection.

1.4 It is inevitable that this issue will have to be considered once again by the present committee. Any decision cannot be a simple one. Quite apart from the socio-political consideration just mentioned, the general level of scientific knowledge in this area has advanced little since 1971. Only a valued judgment is possible. In coming to any conclusion it should be borne in mind that one needs to weigh the issues according to age: young children could well represent a special case.

2. Effect of infection and other adverse environmental factors on food intake

2.1 This clearly depends on a number of variables, including the nature and severity of the insult, the type of food available and the age of the individual. Unfortunately there have been all too few studies of this important topic, but it is obvious that with certain illnesses the effect can be large. Martorell et al¹ have estimated that during a diarrhoeal episode there was an average reduction in daily intake of nearly 20%, equivalent to 175 kcal and 4.8 g protein/d, but during respiratory infections the effect was less - namely 61 kcal and 1 g protein. The authors pointed out that common illnesses were widespread and thus an important cause of the low dietary intakes that had been found in Guatemala. Working in Bangladesh Hoyle et al² observed that in hospitalized children aged 6–35 months with diarrhoea, mean intake varied between 61 and 75 kcal/kg, in contrast to 130 kcal/kg observed in non-diarrhoeal cases. More worryingly, this anorectic effect could not be overcome even by intensive educational efforts (see also 7.1).

2.2 Hoyle et al ² did point out, however, that breast feeding was particularly valuable in this respect since the consumption of breast milk appeared to be much less affected and thus breast-fed children were better protected against reduced intake during diarrhoea than were non-breast fed children. This finding was in accord with previous findings by Rutishauser³ in Uganda (Table 1).

^a Significance of difference between children with good and poor appetite.

NS, mean values for good and poor appetite groups not significantly different by student's 't' test;

+++ mean values significantly different p <0.001.

The amount of bulky and stodgy plantain consumed was reduced by more than 50% when young children were ill, but the intake of breast milk was virtually unaffected. This meant that in the second and third years of life the effect of infection on food energy intake was marked but during early infancy - when breast milk was the predominant source of food - the difference in food intake was non-existent.

2.3 The recognition by a mother that liquid foods are better tolerated than solid foods during infection can also be a direct cause of food deprivation unless foods of high nutritional value, like cow's milk, are available. Mothers in the developing world frequently replace more solid food by ‘paps’ during illness, which are nutritionally totally inadequate, especially when continued for long periods. In The Gambia, for example, cereal gruels, known locally as monos, can have an energy content as low as 0.3 kcal/g⁴. These foods are also very imbalanced in their nutritional content, since the more protein and nutrient rich components of the adult diet, from legumes and animal products, are often not included.

3. Infection and nutrient loss via vomiting, diarrhoea and gut losses

3.1 There is little need to point out that quite apart from illness-induced apathy and anorexia, considerable nutrient losses can occur as a result of diarrhoea and vomiting. Once more the relative contribution of this cause of malnutrition has not been quantified, but it is bound to be a very variable component.

3.2 Other intestinal infections contribute to malnutrition by adversely affecting nutrient absorption. Gardia, for example, coats the intestinal villi. It has been postulated that this seriously affects their absorptive capacity. 22% of children aged 1–3 years in The Gambia have been shown to be thus affected⁵. Upper bowel colonization is also important in this respect. Studies by my group based in The Gambia⁵ have revealed an abnormality of gut flora distribution and consequent bile salt metabolism in approximately three-quarters of the children examined, irrespective of whether or not they had diarrhoea on the day of examination. Rowland et al⁵ concluded that the associated metabolic changes were of sufficient severity to cause significant malabsorption of nutrients.

3.3 Nutrient loss is not the only way in which diseases affecting the intestinal tract can change nutritional status. There is a growing interest in the relationship between infection and gut albumin loss. Dosseter et al⁶, working in Nigeria, have conclusively demonstrated that the acute drop in plasma albumin concentration during measles is at least partly due to protein - losing enteropathy. Other studies would indicate that measles is not the only cause of such albumin loss. An increased loss of plasma albumin into the gut has been demonstrated during hookworm infection (Blackman et al⁷ and Neilson⁸) and inferred in diarrhoeal disease (Lunn et al ⁹).

4. Infection and metabolic imbalance

4.1 The hormonal pattern on to which nutrient deficiencies are superimposed can markedly affect the efficiency with which the body handles dietary energy and nutrients. Many infections result in high plasma cortisol concentrations and correspondingly low insulin ones. This metabolic balance is decidedly anti-anabolic and has an especially severe effect on the efficiency of protein utilization from the diet.

4.2 Wannermacher¹⁰ has demonstrated that infection is not accompanied by the nitrogen conservation encountered in simple starvation. In an individual adapted to a low dietary intake the utilization of amino acids for gluconeogenesis is significantly decreased below that found in a normally fed subject. Lipolysis of adipose tissue and other metabolic adaptations are able to protect the body against the worst effects of energy deficit. The acutely infected individual, however, is frequently unable to mobilize his fat stores fast enough to meet energy needs. This results in a marked increase in gluconeogenesis from amino acids and an increase in the breakdown of skeletal muscle and skin proteins to supply the missing metabolites needed to protect general homoeostasis. In addition, some of these amino acids have to be utilized for the synthesis of acute phase globulins and other proteins involved in the immune defense system. Wannermacher¹⁰ has estimated that in the septic patient N loss from the body can be 4 times that of the steady-state adapted individual. The full significance of this hypothesis cannot be fully evaluated in nutritional terms but it could be that the normal adaptive processes encountered in people living under poor conditions in the developing world may be adversely affected by infection.

4.3 Scrimshaw¹¹ has reasoned that the metabolic effects of infection are more crucial to protein than energy needs. He points out that illness is usually associated with decreased physical activity which would tend to reduce energy needs. In contrast, the metabolic stress response to infection increases N losses. He also emphasizes the loss of the protein sparing effect of carbohydrate and the fact that a deficient energy intake during infection will exacerbate nitrogen losses from the body and the total nitrogen deficit is thus likely to be disproportionately large.

5. Length of effect of infection on dietary intake

5.1 It is probable that both the anorexia as well as the metabolic consequences of each infection last for considerably longer than the duration of the acute clinical event. One investigator who has studied the first of these is Pereira¹², from Bellore in South India. She showed, from studies in a metabolic ward, that while the average attack of fever may last for only 4 days, the mean time taken to ‘average out’ food intake to make up for the anorectic phase was 13 days, but intervals up to 25 d had been measured. Likewise with diarrhoea and dysentery the recognisable clinical event lasted on average 6 days but the measured effect on dietary intake could be as long as 40 d (average 20 d).

5.2 Unfortunately there have been virtually no studies on the prevalence of different types of infection and the total time during which an individual's metabolism is adversely affected. Most of the quantitative data we possess are confined to the duration of the clinically identifiable episode. This has recently been reviewed by Professor Doris Calloway and presented to a meeting of the UNU held in Bellagio, Italy. She examined diarrhoeal data compiled by Reddy from India, Mata from Central America and Cole and Parkin from Uganda and The Gambia, and understandably it was found to vary widely. Estimates of days affected by diarrhoea varied from 15 in India compared with 141 in The Gambia between 6 months and 3 years. Prevalence of all disease categories recorded in The Gambia was 0.456¹³, which means that the average child had something clinically wrong for almost half of each month. In Uganda disease incidence measured by the same team of investigators was much lower, only 2 days per month. This emphasizes the problem of making a global recommendation; clearly each country would need to interpret any dietary guidance given in the Report in the light of its own specific problems.

6. Effect of infection on infant and pre-school child growth

6.1 A number of investigators have attempted to quantify the effect of infection on growth in order to assess the relative contributions of diet and disease to malnutrition and also to assess subsequent catch-up needs. The acquisition of this knowledge is highly time-consuming and the problems are compounded by both cause and effect being such variable and interacting factors. Some diseases, like malaria and diarrhoea, occur frequently and even if the duration of each attack is short, the accumulative effect of a succession of attacks can be considerable. In The Gambia¹³ it was estimated that malaria caused, on average, around 779 g weight loss for each 30 days the child had an elevated temperature. It should be pointed out, however, this probably underestimates the ‘normal’ situation because of the instant treatment the children received. A very similar weight deficit/30 d was obtained in the case of dairrhoea (773 g), but again easily accessible treatment may have minimized the effect. Since prevalence data showed that the average child was ill with diarrhoeal disease for 0.131 of the time and with malaria for 0.010, it can be calculated that these two diseases alone were on average responsible for 3.3 kg weight deficit between 6 months and 3 years, even in children who had available to them an atypically good medical service within their village. The Gambia, although probably similar to much of the African tropical dry savanna, has been said to represent the extreme for a non-crisis situation and it must be admitted that overall growth data for that country do stand out by international comparison. It is unfortunate, however, that there are all too few carefully and systematically collected data of this type to arrive at a meaningful overall global picture. One thing is certain, however, some countries will need to consider the nutritional effects of infection more seriously than others.

7. Energy and nutrient requirements to counter the effects of infection

7.1 While it is clearly sound advice to maintain as good a food intake as possible during infection, it seems that this will at best only minimize the worst effects of the disease. For example, Poskitt¹⁴ demonstrated that children recovering from kwashiorkor, and rigorously maintained on 4 g/protein per kg body wt plus 120 kcal/kg, still lost substantial amounts of weight during an episode of measles, and their plasma albumin concentration dropped precipitously as well. A similar experience is described in section 2.1. From the point of view of recommended dietary allowances it is probably more profitable to concentrate on the period of time when the child is well and physiologically capable of replacing depleted tissues and catching up in growth.

7.2 It has already been pointed out that this time might be less than the duration of the clinical event might indicate, owing to the anorexic and catabolic influences of infection lasting much longer than the acute episode. To take the allegedly extreme case of The Gambia, the average child showed clinical symptoms of some illness, severe or mild, for 45% of the time between 6 months and 3 years. To catch up 3.3 kg of weight loss from diarrhoea and malaria, plus say 0.7 kg from all other disease causes, the child would during each anabolic phase have to grow at a rate considerably in excess of that normally needed to remain on the 50th centile. Even if the non-clinically discernable ‘anorexic overhang’ were as little as 15%, the mean growth rate over the 40% ‘well’ period would have to increase up to 3 times the 50th centile incremental values. The fact that variations in monthly growth velocity much greater than is customarily accounted for in the derivation of recommended dietary allowances, do occur in individual pre-school children is shown in Table 2, which gives 3rd, 50th and 97th centile values for rural Jamaica, Uganda and The Gambia.

^a Increments at the Boston 50th percentile used by WHO/FAO (1973) were: 9–11 months, 330 g/month; 1–2 years, 208 g/month; and 2–3 years, 170 g/month.

Growth velocities of this magnitude are not out of the ordinary, however. In children recovering from severe malnutrition in special nutritional wards, monthly catch-up weights 10–20 times the normal can be achieved.

7.3 Large though disease induced weight deficits may seem, Calloway has pointed out that the extra dietary energy needed to achieve the actual catch-up is relatively small. If the energy cost of growth is 5 kcal/g body weight, only 20, 000 kcal would be required to restore 4 kg weight deficit, or 100 kcal/d extended over 200 days. This is less than a 10% increase over the RDA for a 2 year old child growing normally. A 4 kg weight deficit represents a rather extreme situation and the net dietary energy increment for the actual catch-up process by less affected children would be very small indeed. It should be borne in mind, however, that the energy required for the actual catch-up growth process is not the whole story. An increased weight also creates raised needs for maintenance. Thus the total increase in energy requirement at the end of a period during which body weight has caught up by 4 kg from 7–11 kg, would be an extra 400 kcal/d or around 57% more than if he had not caught up.

7.4 The situation with protein is complex. Two groups of investigators - Whitehead¹⁶, and Waterlow and Payne¹⁷ - have calculated on theoretical grounds the relative requirements for dietary energy and protein for different rates of catch-up growth in underweight children of 7 kg. This weight was chosen as it is typical of the sort of child most in need of dietary improvement in the developing world. Both concluded that catch-up growth might require relatively more dietary protein than energy. Table 3 shows the more recent calculations of Waterlow and Payne¹⁷. Rates of growth 3 times normal would theoretically require an increase in the minimum P:E ratio over those needed for 50th centile rates of growth by 11%, and for 5 times 21%. Such variations in monthly growth rates are encountered in practice, as shown by Table 2. It should be pointed out, however, that this effect would be much less marked with older children of higher body weight. This emphasizes the need to consider the separate requirements of different age groups in coming to a decision as to whether a special increment to cover repeated infection is necessary.

^a Data are based on requirements for milk or egg protein. Protein increment for normal growth was taken as 0.15 g/kg/d. From Waterlow and Payne¹⁷.

7.5 Scrimshaw¹¹ has examined protein requirements during recovery from a different perspective. On the basis of data published by Powanda¹⁸ that the net loss of nitrogen during infection varied from 0.6–1.2 kg/d, and assuming the anabolic phase of recovery would last 3 times longer than the catabolic period, he concluded that an average extra daily allowance of 0.2–0.4 g protein/kg/d would be necessary to replace that lost during the acute infectious episode. Scrimshaw pointed out the difficulty in calculating a similar increment for energy as this would depend greatly on the degree of activity during recovery, but assuming a person may well be less active at this time a higher minimum P:E ratio than that calculated for the maintenance of normal health would be conducive to rapid recovery.

8. Conclusions

8.1 The hypothesis that increasing the P:E ratio of a child's diet does permit a faster rate of catch-up growth and chances for recovery still needs to be substantiated by direct investigation. One variable which it was impossible to take into account in the above calculations was a change in metabolic efficiency to compensate for the increased metabolic activity.

8.2 At the present state of knowledge, however, it would seem that in young children, especially in countries where the dietary staples are starch root crops or plantains, and an adequate P:E ratio depends on added food stuffs like legumes or meat, minimal levels of protein in the food prepared for the whole family could well represent one factor limiting catch-up and recovery from endemic diseases. Recognising the improbability that such children will receive anything other than the adult diet, caution should be exercised in setting adult recommendations for protein relative to energy at too low a level.

8.3 The fact that complete catch-up growth following infection is rare in the developing world but commonplace in affluent countries, is well known. The precise cause is not so obvious. Some would argue that in the worst affected children it is the constant onslaught of infection, others that a low food energy density is the basic problem. Extra protein plus increased requirements for various vitamins and trace elements during catch-up have also been implicated. While it is illogical to attempt to limit the effects of infection by dietary means rather than through direct public health measures directed at the disease itself, commonsense dictates that the unhygienic circumstances prevailing in many parts of the world will continue for some time to come. Nutritionists should be encouraged to direct their research to provide answers to some of the enigmas which have been raised, in the hope of breaking the nutrition-infection vicious circle which can all too frequently lead to frank malnutrition. In the meantime, caution needs to be exercised so that the recommended allowance for protein is set at a realistic level. This consideration should be kept in mind throughout the deliberations of the Committee.

References

1. Martorell, R., Yarbrough, C., Yarbrough, S. and Klein, R.E. (1980) The impact of ordinary illnesses on the dietary intakes of malnourished children. Am. J. Clin. Nutr. 33, 345–350.

2. Hoyle, B., Yunus, Md. and Chen, L.C. (1980) Breast feeding and food intake among children with acute diarrhoeal disease. Am. J. Clin. Nutr. 33, 2365–2371.

3. Rutishauser, I. H. E. (1974) Factors affecting the intake of energy and protein by Ugandan preschool children. Ecology of Food & Nutrition. 3, 213–222.

4. Hudson, G. J., John, P. M. V. and Paul, A. A. (1980) Variations in the composition of Gambian foods: the importance of water in relation to energy and protein content. Ecology of Food & Nutrition 10, 9–17.

5. Rowland, M. G. M. and McCollum, J. P. K. (1977) Malnutrition and gastroenteritis in The Gambia. Trans. Roy. Soc. trop. Med. Hyg. 71, 199–203.

6. Dosseter, J. F. B. and Whittle, H. C. (1975) Protein losing enteropathy and malabsorption in acute measles enteritis. Br. Med. J. ii, 592–593.

7. Blackman, V., Marsden, P. O., Banwell, J. and Hall-Craggs, M. (1965) Albumin metabolism in hookworm anaemia. Trans. Roy. Soc. trop. Med. Hyg. 59, 472–482.

8. Neilson, K. (1976) Plasma protein metabolism. In: Pathophysiology of Parasitic Infection. Soulsby, E. J. L. (ed.). London and New York: Academic Press, p. 23–40.

9. Lunn, P. G., Whitehead, R. G. and Coward, W. A. (1979) Two pathways to kwashiorkor? Trans. Roy. Soc. trop. Med. Hyg. 73, 438–444.

10. Wannermacher, R. W. (1977) Key role of various individual amino acids in host response to infection. Am. J. Clin. Nutr. 30, 1269–1280.

11. Scrimshaw, N. S. (1977) Effect of infection on nutrient requirements. Am. J. Clin. Nutr. 30, 1536–1544.

12. Pereira, D. S. M. (1979) Quoted in: Protein-Energy requirements under conditions prevailing in Developing Countries: Current Knowledge and Research Needs. UNU: Tokyo.

13. Cole, T. J. and Parkin, J. M. (1977) Infection and its effect on the growth of young children: a comparison of The Gambia and Uganda. Trans. Roy. Soc. trop. Med. Hyg. 71, 196–198.

14. Poskitt, E. M. E. (1971) Effect of measles on plasma albumin levels in Ugandan village children. Lancet, ii, 68.

15. Whitehead, R. G. (1979) Dietary allowances of Energy and Nutrients.In: International Review of Biochemistry, Biochemistry of Nutrition 1A, 27, 281–325.

16. Whitehead, R. G. (1973) The Protein Needs of Malnourished Children. In: Proteins in Human Nutrition, chapter 7. Porter, J. W. G. and Rols, B. A. (eds.). Academic Press: New York and London.

17. Waterlow, J. C. and Payne, P. R. (1975) The protein gap. Nature, Lond. 258, 113–116.

18. Powanda, M. C. (1977) Changes in body balances of nitrogen and other key nutrients: descriptions and underlying mechanisms. Am. J. Clin. Nutr. 30, 1254–1268.