Previous - Next
Abstract
1. General concept of growth
2. Outcome variables2.1. Height
2.2. Biochemical and functional tests
2.3. Weight and nitrogen balance3. General principles relating nutrients to growth
4. Hierarchy of metabolic functions
5. Normal growth
6. Catch-up growth6.1. Nutritional determinants of catch-up growth
6.2. Use of weight/increment in body fat
6.3. Body composition during catch-up growth7. Factors affecting net energy accretion
7.1. Limiting specific nutrient
7.2. Effect of protein: Quantity and quality
7.3. Theoretical model for P:E ratio8.1. Colonic fermentation
8.2. Energy from SCFA
8.3. Factors influencing SCFA production
8.4. Gross versus metabolizable energy
8.5. Faecal energy and non-starch polysaccharide
8.6. Faecal energy in cystic fibrosis
A.A. JACKSON and S.A. WOOTTON *
* Department of Human Nutrition, University of Southampton, Bassett Crescent East Southampton S09 3TU, U.K.
One of the primary determinants of the rate of growth is the energy available to the organism. Energy is required for the synthesis of new tissue, and is also deposited in it. The density of the energy stored varies with the type of tissue, from 5.6 kJ/g for lean to 35 kJ/g for adipose tissue. The energy available for growth appears to be preferentially deposited as lean tissue, provided that all other nutrients (amino acids, minerals and vitamins) are available in sufficient amounts. The availability of the first limiting nutrient will determine the overall rate of utilisation of all other nutrients and hence the overall rate of growth. The absence of a single nutrient may lead to a grossly inefficient utilisation of the available energy, which may be dispersed as heat or deposited as adipose tissue.
Catch-up growth is a more complex process than simply an accentuation of normal weight gain. During catch-up weight gain, energy intakes may be as high as 4.5 times basal expenditure. At high rates of weight gain, there is a limitation in the deposition of lean tissue, with excessive deposition of adipose tissue. As the rate of lean tissue deposition increases, there is a disproportionate increase in the demand for nutrients relative to energy. To sustain very high rates of lean tissue deposition, requires a dietary intake with a high P:E ratio.
Understanding of the relative effect which the quality of dietary energy has upon the rate of weight gain is poor. Fat and carbohydrate exert different effects at the highest rates of weight gain. The effect of complex carbohydrates is virtually unknown. The metabolic activity of the microflora of the lower bowel is of much greater significance to growth than has been appreciated in the past.
Growth is a non-specific term that is used to include a constellation of changes associated with the elaboration of form and function. It is a process that has been characterised as canalised and target-seeking (genetically determined). Although growth is not necessarily a continuous process on a day-to-day basis, under optimal circumstances it appears to be continuous from week to week. There are a range of adverse factors, both dietary and environmental, which may impede, slow down, or even reverse some or all aspects of growth. Following the removal of the adverse circumstance, a process of repair and recovery takes place enabling the individual to regain the original growth channel. This process of recovery has been called catch-up growth (TANNER, 1978). Superficially, the process of catch-up growth may have the appearance of an intensification of the normal growth process. Although there may be aspects that can be perceived of in this way, it is not justified to consider that this is necessarily true for the entire process.
During growth, the changes in form are most readily identifiable as an increase in stature and mass, but also include more subtle variations in the composition of the body and changes in the relative size of the different organs and tissues (WIDDOWSON, 1970). The coordinated development and refinement of function may be more difficult to quantify than changes in size, and are not as clearly characterised.
Growth, as an increase in mass,
represents an increase in the net energy content of the body. The energy density of the
increased mass will vary with the nature of the tissue being deposited, from 35 kJ/g for
adipose tissue to 5.6 kJ/g for lean tissue (JACKSON, PICOU and REEDS, 1977). There is an
energy cost associated with the actual deposition of this tissue which is determined by
the nature of the initial substrate and the pattern of tissue to be deposited (PULLAR and
WEBSTER, 1977). To the extent that growth represents an elaboration and refinement of
form, there is an energy cost associated with the process of remodelling. The energy cost
of remodelling gives the appearance of inefficiency to the overall process of net
deposition. It is not altogether clear whether remodelling can be used to account for the
relative inefficiency of protein deposition during growth. In childhood, for example, 1.4
g of protein have to be synthesised for the net accretion of 1 g of protein (WATERLOW and
JACKSON, 1981). During the growth of normal children, the absolute requirement for energy
to satisfy all these components is small relative to the energy required for maintenance:
around 15 to 30% at birth and falling to only 5% at 1 year of age (FAO/WHO/UNU, 1985).
Although it is possible to adopt a factorial, biochemical approach to determining the
relative demands in energy for each of the processes, it is much more difficult to
determine the real physiological cost of all the processes associated with growth at the
level of the whole body. It is for this reason that we have only a limited appreciation in
detail of the real costs. One way to get around this problem is to study situations in
which the entire process has been intensified. The advantage in studying catch-up growth
is that there is an acceleration of the whole process with an intensification of the
changes in time. As a result, the proportion of the total energy intake devoted to tissue
deposition and remodelling may reach more than 50% of the total (ASHWORTH, 1974).
2.1. Height
2.2. Biochemical and functional tests
2.3. Weight and nitrogen balance
In practice, a number of outcome
variables have been used to define the interaction of nutrient intake with growth, and
each has its relative merits.
Following the introduction of
systems for the classification of malnutrition which differentiate between stunting and
wasting (WATERLOW, 1973), and the increased appreciation that stunting of itself may be
associated with long-term functional impairment (GRANTHAM-McGREGOR, POWELL and WALKER,
1989), there has been increased interest in factors that determine and control the
achievement of height potential (see WATERLOW, 1987). Attempts to explore the relationship
on a community-wide basis have produced some insights, but have not provided us with the
ability to focus intervention specifically (GOLDEN, 1985; KELLER, 1987; GRANTHAM-McGREGOR et
al., 1989). Similarly, specific associations have not been identified in analyses of
the extent to which children recovering from severe malnutrition are able to catch up in
height (ASHWORTH, 1975; WALKER and GOLDEN, 1988). Therefore, there are important factors
operating of which we have little understanding at the present. There are models of
producing substantial height with consistency and reliability that might be worthy of
exploration: following the use of human growth hormone in children of short stature; in
children with sickle cell disease following splenectomy for hypersplenism (EMOND, 1987);
and following the treatment of severe trichuriasis (COOPER and BUNDY, 1988). One important
feature of each of these conditions is that not only may the children experience
substantial gains in height, but the height gain appears to be of high priority, if
necessary at the expense of depositing adipose or lean tissue.
Biochemical and functional tests are
of great importance in defining the quality of growth, and detailed consideration of some
of these factors is provided by other speakers at this meeting.
The most widely used approach to
assessing growth is by the measurement of body weight, to the point where growth is often
used casually as a synonym for changes in weight. The great advantages of using body
weight are that the measurements are non-invasive and that they are relatively easy to
take accurately and reliably. The progressive changes in weight are sufficiently large for
a consistent pattern of change to be identified over a period of days to weeks. The major
disadvantage of the use of weight is that major alterations in weight may be a consequence
of changes in water, lean or adipose tissue content. Growth is generally perceived as a
net increase in energy content, with the energy in the form of functionally active, lean
tissue. The most usual way of attempting to attribute the increase in weight to an
increase in lean tissue is by measurement of nitrogen balance. Hence a substantial
literature on growth associates a net increase in the energy content of the body with a
net increase in the nitrogen content of the body.
If growth represents the net increase in the energy content of the body as lean tissue, then all the individual components that constitute lean tissue have to be available in the appropriate amounts for tissue formation to take place. This leads to the concept of the first limiting nutrient as that component of tissue which is available in least amounts relative to the demands for tissue formation. The rate of tissue growth is determined by the availability of the limiting component. All other nutrients can only be utilised at a rate determined by this rate of growth, and any individual components that are available in excess can not be used efficiently and will be wasted.
By application of this concept, it becomes clear that the basic principle that determines the overall rate at which the process of growth takes place will be the availability of the limiting nutrient. This principle can be applied equally to the entire process, or to individual parts of the process. One point which has attracted increased attention over the past decade is that, given the composition of most mixed diets, energy is more likely to be limiting for normal maintenance and growth than any specific nutrient. It is generally accepted that the requirement for protein increases as the demand for lean tissue synthesis goes up. For most nutrients it is accepted that the requirement for the nutrient increases with the rate of growth, although the basis upon which the magnitude of the increase is determined is not always clear.
During deliberations leading to the definition of requirements and safe levels of intake, it is accepted implicitly that the requirement for a nutrient can only be determined in a situation where the requirement for all other nutrients has already been satisfied. Sufficient care has not always been taken to ensure that this constraint is adhered to in experimental situations, even if this were possible. Of greater importance is the tendency to forget or to ignore the potential confounding effects that nutrient-nutrient interactions might have at the level of marginal or inadequate intakes of specific nutrients. In 1945, KLEIBER reviewed the data that related the efficiency of energy utilisation to the adequacy of the diet in respect of specific nutrients. He made the general observation that energy is used less efficiently in the face of a specific deficiency of many vitamins and nutrients:
"A diet is deficient in any nutrient whose addition decreases the calorigenic effect of the ration. A ration is deficient in any food constituent whose addition increases the total energy efficiency of energy utilisation."
Although there has not been a great deal of experimental work that seeks to test this observation formally, there is a substantial body of information in the literature to indicate that it represents a general rule of considerable importance to our appreciation of the availability and utilisation of energy within different population groups. There are many animal studies in which the effect of specific nutrient deficiencies has been explored. Almost invariably when a deficient diet is offered there is a consequent reduction in food intake. One necessary control for this experimental situation is to pair-feed a control group on a complete diet to the intake of the deficient group, to correct for any influence that total intake might have on the interpretation of the results. The general observation is that the pair-fed animals gain weight at a rate that is significantly greater than the deficient animals, although not as great as the group fed ad libitum. The details of the intensity of the response varies with the nutrient, but the overall consequence is the same (Table 1). This leads to an important, outstanding question: in the normal population, to what extent might the apparent variability in the energy requirements for maintenance or growth be accounted for by a marginal or inadequate availability of specific nutrients? Similar considerations apply to the intake of protein, with respect to the quality of the protein and the availability of individual amino acids. There is evidence that analogous considerations may be extended to the relative proportions and the quality of dietary carbohydrate and fat.
Table 1. Animals reared on a diet that is deficient in a specific nutrient fail to grow and lose their appetite. When a control group of animals is pair-fed to the intake of animals on a deficient diet the pair-fed animals gain weight at a rate that is significantly greater than the deficient animals
Specific deficiency |
Species |
Duration |
Body weight (g) |
Reference |
|
Deficient |
Pair fed |
||||
Vitamin A |
mouse |
12 wk |
39 |
47 |
AHMED et al. |
Copper |
rat |
8 wk |
259 |
317 |
BROWN et al. |
Iron |
rat |
5 wk |
219 |
241 |
BEARD (1987) |
Potassium |
rat |
2 wk |
50 |
95 |
DORUP and CLAUSEN (1989) |
We are not very clear about the extent to which growth in height or weight is a protected function. In the face of a limited availability of energy or nutrients, or in a situation in which there may be competing metabolic demands, to what extent is the body able to mount a flexible response determined by a 'perceived sense of a hierarchy of demands'? The extent to which available nutrients may be diverted to increments in height or weight has been touched on above. In the three examples given of marked gains in height, the predominant aspect of the catch-up growth is in height rather than in weight, and in relative terms at the expense of weight. This response is completely different from the catch-up seen in children recovering from severe malnutrition. It is not clear whether this is related to nutrition, is an age effect, or the result of some other factor.
It has become a standard requirement of experimental design that a series of treatments or diets be offered in a random order so that sequential effects of one diet upon another can be removed. The disadvantage of this approach is that an opportunity to observe important biological responses may be lost. TORUN and VITERI (1981) have shown that in young children provided with an adequate intake of protein and other nutrients, as the energy intake is reduced in a stepwise fashion from adequate to maintenance to inadequate, there is a progressive saving of energy by sacrificing function. In this study, physical activity appeared to have a lower priority than weight, being sacrificed at an earlier stage. KENNEDY, BADALOO and JACKSON (1990) have shown that, if children are presented with an adequate intake of energy, followed by an inadequate intake and then a marginal intake, a process of accommodation takes place on the inadequate intake which enables weight gain to take place on a marginal intake, at the same rate as on the adequate intake. The accommodation appeared to include a change in the metabolic behaviour of the colonic flora, but the detailed basis of the response is at the present unclear.
In animal experiments, it is evident that one of the factors that will determine the hierarchy of functions taking place during growth is determined by the relative availability of individual nutrients. For example, if iron is limiting in the diet, then iron containing compounds and functions will not develop or be repleted at the same rate as others. The same can be shown for a wide range of individual nutrients or nutrient combinations. As indicated above, one important response to a limiting nutrient is a decrease in appetite and a reduction in the overall intake of food, especially energy. If the energy intake is increased in the face of a limiting nutrient, then deleterious sequelae may ensue. As a general rule, an excessive intake of energy in the face of a limiting nutrient may be conceived as being toxic. There are good examples from animal studies as to how this affects function. LUNN and AUSTIN (1983), among others, have shown that rats fed a low-protein diet will limit their intake of food, and hence energy. If the animals are forced to ingest energy to the level of animals on an adequate protein intake there is a fall in the plasma albumin concentration and the animals either fail to grow, or even lose weight. At least a part of this response is determined at the level of the genome. A further example that has important practical implications is the extent to which immune responsiveness might be modulated. If animals are subjected to a reduced intake of a balanced diet, immune competence can be protected down to very low levels of intake, 50% of normal (AHMED, JONES and JACKSON, 1990b). However, if the intake of nutrients is reduced by dilution, but energy density preserved by the addition of sucrose to the ration, there is an impairment of immune function with a dilution of as little as 10% (NALDER et al., 1972).
Environmental effects exert
significant and important influences upon energy-nutrient interactions. Under
thermoneutral conditions, heat is generated by the body as a by-product of metabolism and
contributes to the maintenance of body temperature. In the cold, heat generation becomes
an objective in its own right, rather than simply a by-product of metabolism. Under these
circumstances, energy may be preferentially used to generate heat rather than for
synthetic activities or for growth. The modified use of dietary energy has secondary
consequences for the requirements of other nutrients (PAYNE and JACOB, 1965).
There is an extensive literature on normal growth, and other speakers will deal with specific aspects of the factors that are the subject of discussion and consideration. WHITEHEAD and PAUL (1988) have reviewed the literature which shows that there have been changes in the pattern of growth during the first year of life in infants in many countries of the world. The significance of these changes is not clear at the present time, nor is the extent to which they may be directly attributed to nutritional factors. Infants who are breast-fed grow well and, in general, by 6 months of age at a rate in excess of the NCHS standards.
Between 6 months and 1 year, there
is a falling off relative to the standards, with the result that by 1 year they are
shorter, lighter, with thinner skinfolds than the standards.
6.1. Nutritional determinants of catch-up growth
6.2. Use of weight/increment in body fat
6.3. Body composition during catch-up growth
If the concept of catch-up growth is that of accelerated weight gain, either to make up for faltering or to repair a deficit, then it is possible to define a spectrum of conditions for which the nutritional requirements must vary (Figure 1). In childhood, growth is the normal state, and a slowing of the rate of growth will result in a progressive falling away from the normal growth channel with time. As growth is canalised, a return to the original growth channel requires an acceleration in the rate of growth. By and large, the tissue deposited during the process of acceleration will have a composition that is essentially similar to that deposited during normal growth.
At the other extreme is the
repletion of tissue in an individual, adult or child, who has experienced a period of
actual weight loss. The weight that has been lost will comprise of tissue which is not
balanced. During the process of weight loss, there is a relative preservation of visceral
tissues at the expense of muscle and adipose tissue. Thus, the requirements for repletion
of tissues is unbalanced in relation to the requirements for normal growth or maintenance.
The greater the deficit that has to be repaired, the more unbalanced is the tissue that
needs to be regained. Therefore, one may conclude that if the losses are uneven between
individuals, then the desirable gains need to be of variable composition. Hence any
assessment of catch-up growth needs to take into consideration the composition of the
tissue gained. My comments will focus upon catch-up growth in humans as measured by a rate
of increase in weight that is substantially greater than the normal rate of weight gain at
the corresponding chronological or developmental age. Although the focus will concentrate
upon the energy requirements for catch-up in weight in young children recovering from
severe undernutrition, the general principles do not appear to be different in situations
such as catch-up in preterm infants, or in adults following stress or injury.
Up to 1961, the general impression was that the main determinant of the rate of catch-up weight gain should be the dietary intake of protein. However, based upon a series of balance studies, WATERLOW (1961) noted:
"The rate of weight gain depends more closely on the intake of calories than of protein within the range studied (100 to 200 kcal/kg per d, 2 to 7 g protein/kg per d). For 150 kcal/kg per d, a protein intake of 3 to 4 g/kg per d is adequate."
As growth represents an increment in the net energy content of the body, energy intake must exceed energy expenditure for tissue deposition to take place. Our interest is in the form in which the energy is deposited, the efficiency with which the deposition takes place and the effect of other specific nutrients upon the pattern of tissue deposition. Each consideration is intimately related to the other, and it is difficult to obtain a clear impression of the interrelationships, without considering them as isolated entities in the first instance.
A series of elegantly conducted
experiments enabled ASHWORTH (1975) to define with some precision the detailed
relationship between energy intake and the rate of catch-up weight. She was able to show
that during catch-up growth the rate of weight gain might be 15 times as fast as that of
normal children of the same age. The weight gain was associated with a high food intake.
As the weight of the children approached that appropriate for their height, there was an
abrupt and dramatic voluntary reduction in the food intake, which fell by 30%. In concert
with the decreased intake of food, the rate of weight gain dropped to levels comparable
with those of normal children of the same weight or length. Overall there was an increase
in the efficiency of food utilisation during the catch-up period. As weight-for-height was
reached, there was a tendency towards an increase in body fat. When the actual rates of
weight gain were compared with the rate of weight gain predicted from the equations of
MILLER and PAYNE (1963) there were major differences (ASHWORTH, 1969a). It could be shown
that there was a highly significant relationship between the rate of weight gain and the
increase in postprandial metabolic rate (BROOKE and ASHWORTH, 1972). The evidence appeared
to suggest that the children were eating to satisfy a need for energy. When they were
allowed ad libitum access to a high-energy milk preparation, they ingested 219
kcal/kg per day. With a lower energy density preparation offered ad libitum,
although there was a voluntary increase in the volume of milk consumed, they were unable
to achieve the same intake as on the denser diet, and the mean energy intake fell to 168
kcal/kg per day. After the children had achieved an appropriate weight for their height,
they no longer consumed an increased volume when the lower energy milk preparation was
offered. It has been suggested that the maximum intake of energy is of the order of four
to five times BMR (PAYNE and JACOB, 1965). ASHWORTH (1969b) has shown that the standard
metabolic rate of the children she studied was of the order of 50 kcal/kg per day;
therefore a maximum intake of 219 kcal/kg per day is about 4.3 times BMR. The conclusion
that children eat for calories during catch-up weight gain has its attractions, but has
never been critically put to test.
The advantage in exploring the protein requirements for weight gain is that the outcome indicator weight can be measured with considerable accuracy and reliability. However, the danger and disadvantage are that there is a tendency to assume that all weight gain has the same composition. This pitfall is often avoided for the relative proportions of lean and fat tissue gained, but the variability of lean tissue composition has not been as widely appreciated. Total body water has been used extensively as an index of lean tissue mass; the assumption being that the relationship between water and lean tissue is relatively invariant. PATRICK et al. (1978) have shown that, during the early phase of rapid weight gain, children recovering from severe malnutrition demonstrate a significant increase in the relative hydration of the body, which progressively tends towards the normal as recovery proceeds (Figure 2). Therefore, the use of total body water to derive values for lean body mass or fat mass is particularly liable to error during the early period of most rapid weight gain, although values derived from measurements of total body water may be of use when taken over the entire period of catch-up growth. This observation may explain why ASHWORTH found that the ratio of observed to theoretical weight gain was greater than 1 during early catch-up growth.
Assessment of changes in body composition requires an assessment of the absolute or relative contributions of water, lean and adipose tissue to weight gain, with some indication of the degree of cellular hyperplasia or hypertrophy. Unfortunately, there are very few direct measures of body composition which can be applied with confidence during catch-up growth. One of the predominant changes during weight loss is a loss of muscle mass, which justifies efforts to measure muscle mass as directly as possible. A useful, but difficult approach, has been to measure muscle mass by following the dilution and kinetics of a tracer dose of 15N-creatine (PICOU et al., 1976). The method has been validated in laboratory animals for whom in vivo measurements of muscle mass correlated closely with cadaver analysis (REEDS and LOBLEY, 1990). Muscle mass was measured in severely malnourished children over the course of recovery. With the attainment of an appropriate weight for their height, the muscle mass of the children was approximately doubled (Table 2; REEDS et al., 1978). As the muscle mass represented an increasing proportion of body weight with recovery, this provided some confirmation of the clinical impression of disproportionate loss of muscle in the malnourished state. In order to identify the cellular basis of these changes, the DNA and non-collagen protein content of a muscle biopsy were measured. Using this information it was possible to calculate the number of muscle nuclei and the non-collagen protein to DNA ratio, the effective 'cell size'. There was no significant difference in the total DNA between the malnourished and recovered children, and the values fell around the middle of the expected range for normal children of the same height. In children in whom paired studies were carried out, a small increase in DNA was appropriate for the increase in height over the same period of time. There was an increase in the protein/DNA ratio with recovery of about 20%. However, even at recovery, the protein/DNA ratio was only about 60% of the ratio that would have been expected for a child of the same height. Therefore, although recovery, defined as the attainment of an appropriate weight-for-height, was complete, the muscle mass had not shown complete recovery, even at the cellular level.
Table 2. Muscle mass was determined in malnourished children and again after recovery, from the dilution of a tracer dose of 15N-creatine. The composition of the muscle was determined from a muscle biopsy (REEDS et al., 1978)
Malnourished |
Recovered |
|
Muscle mass (kg) |
1.00 |
1.91 |
Muscle mass/body weight (%) |
16 |
22 |
Muscle DNA (g) |
2.05 |
2.38 |
Muscle non-collagen protein (g) |
153 |
265 |
Non-collagen protein/DNA (g) |
92 |
110 |
Similar conclusions have been reached using more indirect methods for assessing compositional changes. BROOKE and WHEELER (1976) found that skinfold thickness was up to 99% of normal by recovery, whereas arm muscle area and total body potassium, both indices of lean tissue mass, were 78% and 73% of normal, respectively. Therefore, in relation to a normal child of the same height, the children studied by BROOKE and WHEELER (1976) had a body composition with an excess of fat of up to 10% and a 10 to 15% deficit of lean tissue. On histological examination, the fibre size of clinically recovered subjects, age 13.8 months, was only 60% of that for a well-nourished 6-month-old (HANSEN-SMITH, PICOU and GOLDEN, 1979).
A somewhat different approach to assessing the nature of the tissue deposited during catch-up growth can be taken by a 'cost-of-growth' analysis (JACKSON, PICOU and REEDS, 1977). The energy available for deposition as new tissue is related to the weight of tissue gained. For any given quantity of energy deposited, adipose tissue with an energy density of 35 kJ/g will result in less weight being gained than lean tissue with an energy density of 5.6 kJ/g. The estimates of the energy cost of weight gain may vary from 35 to 14 for a range of different situations implying 0 to 53% of lean tissue accretion (FAO/WHO/UNU, 1985). An analysis of the energy content of the weight gained during catch-up would suggest that for many children the tissue deposited during catch-up growth tends to be relatively unbalanced, with a preponderance of adipose over lean tissue. This is the conclusion reached by Graham, having analysed the rate of recovery in a large series of children from Peru (MACLEEN and GRAHAM, 1980):
"The impressive gains in weight made by recovering malnourished infants are largely fat; reconstitution of lean tissue does not occur equally well at all rates of weight gain."
WHYTE, BAYLEY and SINCLAIR (1984) have reached a similar conclusion in their interpretation of the evidence relating weight gain in preterm infants to the intake of energy and protein: all studies show energy storage cost of growth (i.e., rate of accretion) much higher in low birthweight infants than in the fetus. Although KASHYAP et al. (1988) were able to demonstrate rates of weight gain and nitrogen retention equivalent to the intrauterine rate, the balance between energy and protein intake had to be very finely balanced. Protein tolerance appeared to be at its limit at optimal rates of accretion. Adults recovering from severe weight loss also appear to respond in a similar way, with a relative excess of adipose over lean tissue deposition, irrespective of the dietary protein intake (ELWYN et al., 1979). Similar observations have been made during recovery from anorexia nervosa (FORBES, KREIPE and LIPINSKI, 1984).