E. JÉQUIER*
* Institute of Physiology, University of Lausanne, 7, rue du Bugnon, 1005 Lausanne, Switzerland.
Abstract
1. Introduction
2. Influence of nutrient intake on nutrient oxidation
3. Effect of energy intake on nitrogen retention
4. Effect of protein intake on nitrogen retention
5. The role of glucose and lipid in nitrogen sparing
6. Mechanisms of the sparing effect of dietary carbohydrate and fat
7. Effect of amino acid plasma levels on protein synthesis
8. Practical considerations: Role of the thermic effect of nutrients
9. Conclusions
References
The maintenance of
nitrogen balance depends on both protein and energy intake. The
nitrogen balance is very sensitive to energy intake. In adult
men, the N deficit induced by an acute fast (12 g N/d) is reduced
to about 7 g N/d when the subjects receive 700 kcal of
non-protein energy per day. When protein intake is close to the
minimum requirements (as estimated for individuals in energy
balance), N balance is negative as long as energy balance remains
negative; increasing the energy intake from 900 to 1600 kcal,
improves the N balance by 4 mg N per kcal of energy intake. The
sensitivity of N balance to energy intake is lower when energy
intake is close to maintenance levels (1-2 mg N/kcal of energy
intake).
The minimal daily nitrogen requirement of adult men is about 100 mg N/kg, when proteins of high biological value are ingested. In normal men, intakes of protein exceeding requirements induce a rise in protein oxidation with a N balance close to zero. In contrast, depleted patients show a high degree of N retention, when given N supplements together with non-protein energy in amounts that are slightly above requirements.
The role of glucose and lipids in N sparing has been much studied recently. During caloric restriction (i.e., hypocaloric diets for obesity therapy), glucose has a larger protein-sparing effect than lipids. However, when the dietary energy requirements are met, changing the proportions of glucose and fat markedly (with a constant administration of amino acids) has little effect on N retention; both glucose and lipids have a protein-sparing effect. The protein-sparing action of glucose is mediated in part by increased insulin secretion; insulin inhibits muscle proteolysis, hepatic gluconeogenesis and renal ammoniogenesis. Dietary lipids could exert their protein-sparing action through FFA oxidation in the liver, where the resulting increased NADH/NAD ratio inhibits the activity of branched-chain µ-keto-acid dehydrogenase.
The
efficiency with which the nutrients are utilized is also
important to provide optimal amounts of energy for anabolic
processes. The thermogenic response to fat administration is
smaller than that to glucose (3% vs 7%, respectively). The
administration of high doses of glucose leads to de novo
lipogenesis. Lipogenesis is an energy-requiring process, and the
thermogenic response is enhanced. Under these conditions, some
energy is wasted and less energy, as ATP, is available for
anabolic processes, such as protein synthesis. A 'physiologic'
balance of dietary carbohydrate and fat results in optimal
nitrogen retention in growing infants or in depleted patients
undergoing nutritional rehabilitation.
The energy-yielding
substrates have different fates depending on whether they pass
through their respective biochemical pathways for oxidation or
storage. There are mechanisms of competition between substrates
for oxidation in tissues such as muscle. Free fatty acids (FFA),
for instance, compete with glucose for utilization by
insulin-dependent tissues (RANDLE et al. 1963). Conditions
leading to an increased plasma FFA level therefore impede glucose
uptake and oxidation in vivo (FERRANNINI et al., 1983),
thereby conserving plasma glucose and sparing glycogen stores.
The quantity and source of macro-nutrients in the diet has a
major influence on protein metabolism. The nutrient composition
of the non-protein energy supply may influence N balance.
Carbohydrate has been reported to have a much greater
protein-sparing effect than fat (MUNRO, 1964). However, recent
studies in adults and children have challenged this classical
assumption (McCARGAR et al., 1989; BRESSON et al., 1989).
The goal of this paper is therefore to review recent data on
substrate balance and macronutrient metabolism in man in order to
clarify our understanding of the relationships between energy
metabolism and N balance in man, with a particular interest in
linking the composition of non-protein energy utilization to N
retention.
There are important
differences in the control of nutrient metabolism and oxidation
following food ingestion. The relationships between glucose and
fat metabolism have been much studied and a clear concept emerges
from recent studies. Shortly after meal ingestion, the
carbohydrate content of the meal induces a stimulation of glucose
oxidation, whereas fat oxidation is inhibited. Three to four
hours later, glucose oxidation decreases with a concomitant rise
in fat oxidation. Overall, the intake of dietary carbohydrate has
the effect of reducing the rate of fat oxidation (ACHESON, FLATT
and JÉQUIER, 1982; ACHESON et al., 1988). De novo
lipogenesis from carbohydrate is not an important pathway in man,
and dietary carbohydrates do not increase an individual's fat
mass by de novo lipogenesis (ACHESON et al. 1988). Since
carbohydrate stores are small compared to fat stores and usually
do not vary from day to day, carbohydrate oxidation is adjusted
to carbohydrate intake over a 24-hour period (FLATT, 1988).
After a meal, the increase in plasma insulin levels stimulates glucose uptake in muscle and other tissues, which induces a rise in glucose oxidation. The inhibitory effect of insulin on lipolysis in adipose tissue, and the subsequent decline in plasma free fatty acid levels leads to the postprandial inhibition of fat oxidation. This is a consequence of the glucose-fatty acid cycle described by RANDLE et al. (1963).
From our laboratory, FLATT et al. (1985) reported that fat intake did not promote fat oxidation over a period of nine hours following a meal. Neither did a fat supplement (106 ± 6 g fat) stimulate fat oxidation over 24 hours (SCHUTZ, FLATT and JÉQUIER, 1989). Thus, there is no metabolic response increasing fat oxidation to correct a surfeit of fat. It can be concluded that fat balance differs markedly from the regulation of carbohydrate balance: the latter is adjusted by an increased oxidation of glucose after meal ingestion, whereas no stimulation of fat oxidation occurs, even after a meal with a high fat content. Carbohydrate intake is therefore the main regulator of non-protein energy expenditure (FLATT, 1988). By contrast, fat oxidation is regulated by energy needs rather than by fat intake (FLATT et al., 1985).
The maintenance of nitrogen balance depends on both protein and energy intakes (MUNRO, 1964; CALLOWAY and SPECTOR, 1954; SCRIMSHAW et al., 1972; INOUE, FUJITA and NIIYAMA, 1973; GARZA, SCRIMSHAW and YOUNG, 1976; 1978). In the adult human subject who maintains energy balance for prolonged periods, increasing N intake above requirements only causes a transient positive N balance. The ability to achieve protein balance over a wide range of intake is well documented in humans, although the mechanisms, which account for an increased rate of protein oxidation when protein intake is increased above requirements, are complex and poorly understood.
The relationships between energy and protein intakes on protein metabolism depend on the nutritional and clinical state of the patient. The efficiency with which protein retention occurs is greater in depleted or starved patients than in patients suffering from accidental trauma, acute infection or burns (SHENKIN et al., 1980). The latter patients are often hypermetabolic with a decrease in lean tissues and a net loss of proteins from the body. It is therefore important to identify the nutritional and clinical conditions of the patients when one deals with the complex problem of energy-protein relationships (ELIA, 1982).
In studies
of N balance, it would be desirable to know the independent
effects of changes in N or in energy intakes. To do so is
difficult, since the minimal requirement of protein intake to
obtain N balance depends on the state of the subject's energy
balance.
3.1. Fasting and very low caloric intake
3.2. Moderately hypocaloric diets
3.3. Maintenance diets
3.4. Energy intake in excess of maintenance
Complete fasting, in normal
men, induces N losses amounting to 12 g/d (CALLOWAY and SPECTOR,
1954; CAHILL et al., 1966; CAHILL, 1970). This N deficit
was reduced to about 7 g N/d by supplying approximately 700
kcal/d of non-protein energy (CALLOWAY and SPECTOR, 1954), and
increasing the caloric intake to 2800 kcal/d had no additional
protein-sparing effect (Figure 1). This means that a small
amount of dietary non-protein energy, corresponding to about 25%
of energy requirements, reduces the net protein utilization by
about 5 g N/d. These data were obtained in acute conditions, and
do not take into account the adaptive responses to a prolonged
fast (CAHILL, 1970) that will not be discussed here.
It is interesting to mention that normal men receiving 400 to 600 kcal/d (i.e., less than 25% of energy requirements) and 7 g N/d were found to be in a similar negative N balance (-7 g N/d) as the subjects receiving a comparable energy intake with no protein (Figures 1 and 2). In normal men receiving a very-low-calorie diet, the protein in the diet is mainly burned to provide energy, producing a concomitant rise in N excretion (CALLOWAY and SPECTOR, 1954).
The question whether it is possible to maintain N balance on a markedly hypocaloric intake has been mainly studied in obese subjects. Nitrogen balances close to zero have been reported in obese patients who were fed 14 g N/d and non-protein energy amounting to about 100 kcal/d (APFELBAUM et al., 1970; APFELBAUM, 1981; MARLISS, MURRAY and NAKHOODA, 1978). However, non-obese subjects receiving about 14 g N/d as intravenous amino acids for six days were in negative N balance by approximately 6 g N/d (TWEEDLE et al., 1977; SIM et al., 1979). Thus, it is likely that short-term N balance can be maintained in obese subjects ingesting a very-low-calorie, high-protein diet whereas normal-weight individuals go into negative nitrogen balance when they are on similar energy and protein intakes. Although short-term studies show that obese subjects can maintain N balance when they ingest very-low-calorie, high-protein diet, most long-term studies on the composition of the body mass loss indicate that lean body mass is a component of weight reduction (GRANDE, 1968; BUSKIRK et al., 1963). It is well established that a long-term hypocaloric therapy of obesity induces not only a mobilization of fat from adipose tissue, but also a reduction in the lean body mass, even if the hypocaloric diet has a high protein content. These data support the general concept that prolonged hypocaloric diets induce a negative nitrogen balance and a loss of lean tissues which accompany the fat loss.
Increasing the energy
intake of active men, requiring 2800 to 3000 kcal/d, from 900 to
1600 kcal/d (CALLOWAY and SPECTOR, 1954) showed that the N
balance was very sensitive to the energy intake when 6 g N or
more were provided (Figure 1). For each kcal of energy
intake, there was an improvement of 4 mg in the N balance. In
other words, an increase of the energy intake by 100 kcal induced
a 400 mg improvement of the N balance, or a sparing of 13 g of
lean tissue. This emphasizes the importance of energy intake for
the protein economy of the body. In depleted patients, who
require nutritional rehabilitation, it is clear that any plan of
therapy should aim at supplying a sufficient amount of
non-protein energy in order to provide the best conditions for a
positive N balance and the restauration of the lean body mass.
The sensitivity of N
balance to energy intake is smaller when energy intake is close
to maintenance levels (Figure 1). In normal men, ingesting a
maintenance amount of energy with 100 to 300 mg N/kg, there is a
retention of 1-2 mg N for every extra kcal added (CALLOWAY and
SPECTOR, 1954; KISHI, MIYATANI and INOUE, 1973), which
corresponds to a gain of 30 to 60 mg of lean tissues. Thus, an
excess of 100 kcal/d over maintenance leads to a gain of
approximately 4 g of lean tissues and 10 g of fat. It is
important to mention that these results have been obtained in
short-term studies. In everyday life, a chronic excess of energy
intake leads to weight gain, due to an increase in both fat mass
and lean body mass (the latter representing about 25% of the
weight gain). The increase in lean body mass is accompanied by an
increase in the basal metabolic rate, which tends to offset the
influence of the surfeit of energy intake on energy balance
(JÉQUIER and SCHUTZ, 1988).
It is to be
emphasized that the energy-dependent changes in protein retention
were observed in subjects ingesting at least 40 mg N/kg (i.e., 15
g of proteins with a high biological value for a man weighing 60
kg). With low nitrogen intakes (<40 mg N/kg), the
protein-sparing action of energy supplements is less efficient or
even completely lost (INOUE, FUJITA and NIIYAMA, 1973).
INOUE, FUJITA and NIIYAMA
(1973) showed that in young men the supply of an excess of energy
resulted in a reduction of protein requirements for maintaining N
balance. With maintenance energy (45 kcal/kg), the egg protein
requirement was estimated at 0.65 g/kg, whereas with excess
energy (57 kcal/kg), the minimum requirement was 0.46 g/kg. This
protein-sparing effect of extra energy continued for as long as
three weeks. This observation is of practical importance in many
developing countries with a seasonal food energy and protein
shortage inducing weight loss; periods of increased food
availability allow a seasonal increase in body weight of most
inhabitants (PRENTICE et al., 1981). In these situations
it is important that an increased energy intake has a sparing
effect on protein utilization in order to favor restauration of
lean body mass.
4.1. Normal and obese subjects
4.2. Severely depleted subjects
4.3. Moderately depleted subjects
At energy intake levels
below 25% of energy requirements, CALLOWAY and SPECTOR (1954)
showed that dietary nitrogen was without appreciable benefit on
nitrogen balance in normal young men (Figure 2). However,
as previously discussed, dietary nitrogen improves nitrogen
balance in obese patients treated with hypocaloric diets. In
healthy young men, however, at energy intakes covering 25 to 50%
of requirements, increasing N intake above 3 g/d was without
further benefit on N balance (Figure 2). When energy
intakes reached 50 to 75% of requirements, 7 to 9 g N/d resulted
in the same sparing of body nitrogen as higher N intake. At
energy intakes reaching 100% of requirements, nitrogen balance
was attained at about 8 g N/d, and a substantially higher N
intake (15 g N/d) was without further effect on N balance (Figure
2).
The minimal
daily nitrogen requirement for a normal man is about 100 mg N/kg,
when proteins of high biological value are ingested. With an
intake of N below 100 mg/kg, the subject is in negative N
balance. ODDOYE and MARGEN (1979) fed subjects two maintenance
diets (40 kcal/kg) providing either 160 or 480 mg N/kg per day
for 50 days each. The mean N balance was close to zero with the
diet providing 160 mg N/kg, whereas on the 480 mg N/kg diet there
was a relatively small positive N balance of about 20 mg N/kg per
day. This study clearly shows that an excessive intake of protein
induces an increased protein oxidation which reduces the N
balance close to zero. The mechanisms which stimulate protein
oxidation, so that N balance is only slightly positive when
excessive amounts of proteins are ingested, are poorly
understood.
In contrast to normal
subjects, patients suffering from depletion deposit protein when
fed with large amounts of proteins during nutritional
rehabilitation. BARAC-NIETO et al. (1979) carried out
nutritional repletion in undernourished adult males living in
rural areas of Colombia providing them with an adequate energy
intake and 320 mg N/kg per day. Nitrogen balance was + 140 mg
N/kg per day, indicating that the overall efficiency of N
retention was 44%. Similar values have been reported in premature
infants who can retain more than 50% of dietary N.
Patients with moderate
degrees of depletion exhibit a lower efficiency of N retention
than patients with severe undernutrition. WARNOLD, EDEN and
LUNDHOLM (1988) showed that the infusion of 200 mg N/kg/d in
moderately malnourished patients receiving total parenteral
nutrition (TPN) (non-protein calories at 120% of resting energy
expenditure) was insufficient to promote protein synthesis.
Increasing the infusion of N to 330 mg/kg/d stimulated whole-body
protein synthesis, but skeletal muscles remained in negative N
balance. These studies reflect the difficulties of expanding lean
body mass by TPN in moderately malnourished patients.
When
depleted patients are studied long after trauma, positive N
balance can be obtained with high nitrogen intake. SHAW et al.
(1983) reported nitrogen retention of 21% of the increment in
intake when patients received intravenous nutrition with a high
(364 mg N/kg/d) in comparison with a low (180 mg N/kg/d) nitrogen
content.
5.1. Healthy young subjects
5.2. Patients receiving total parenteral nutrition
It is well established that
protein metabolism is highly sensitive to energy intake. The
estimated minimum requirement for dietary protein is influenced
by the level of energy intake (INOUE, FUJITA and NIIYAMA, 1973).
Dietary carbohydrate and fat provide the major sources of energy
for support of body protein metabolism. According to the
classical studies of MUNRO (1964), administration of carbohydrate
has a protein-sparing effect in the fasting subject, whereas fat
does not have this effect. Isocaloric substitution of fat for
carbohydrate results in a transient increase in N output (MUNRO,
1964). These studies, however, have been carried out at intakes
of protein that exceed those considered to be sufficient for
maintenance of N balance. Therefore it is of interest to know
what is the best ratio of carbohydrate/fat calories for the
utilization of dietary protein in healthy young men given a
protein intake that corresponds to the minimum requirements for
achieving N balance.
RICHARDSON et al. (1979) compared the effect of two maintenance diets on protein utilization in healthy young men; the first diet supplied an equal proportion of energy from carbohydrate (47%) and from fat (47%) with 6% energy from milk protein. The second diet supplied twice as much energy from carbohydrate (62%) as from fat (31%), with the same protein content. Nitrogen balance and dietary protein utilization were improved on the high-carbohydrate diet as compared to the diet with a ratio of carbohydrate to fat of 1/1. The protein-sparing action of dietary carbohydrate is probably mediated in part by increased insulin secretion, since the release of insulin is less stimulated by dietary fat than carbohydrate. There is substantial evidence that insulin inhibits muscle proteolysis (JEFFERSON, LI and RANNELS, 1977).
The effect
of altering the ratio of carbohydrate to fat in the diet (2/1 vs
1/1) on N retention was studied by McCARGAR et al. (1989)
at maintenance and submaintenance energy intake levels in young
healthy men. The purpose of the study was to assess the influence
of energy balance on N balance when changing the ratio of
carbohydrate to fat in the diet. The diets were fed at a level
estimated to provide either maintenance or 75% of maintenance
energy requirements, respectively. Surprisingly, the diet with a
carbohydrate-to-fat ratio of 1/1 increased N retention more than
the high-carbohydrate diet, particularly at maintenance energy
intake. The high-fat diet resulted in lower weight loss at 75% of
maintenance energy requirements than the high-carbohydrate diet.
The reason for the discrepancy between these results and those of
RICHARDSON et al. (1979) has not been explained.
In patients requiring total
parenteral nutrition (TPN), it is important that the non-protein
energy infused promote optimal nitrogen retention. JEEJEEBHOY et
al. (1976) gave sequential infusions of non-protein calories
either as glucose, or 83% as lipid (+17% glucose), with a
constant administration of amino acids (1 g/kg/d). Despite these
two very different metabolic situations, the nitrogen balance
with both types of treatment was positive to a comparable degree.
This study clearly shows that not only glucose, but also
exogenous lipid can spare protein and induce nitrogen retention.
The finding that nitrogen retention during lipid infusion occurs
with low insulin plasma levels indicates that mechanisms distinct
from insulin action are responsible for the protein-sparing
action. The efficacy of intravenous infusions of fat emulsion in
improving nitrogen economy has also been reported by WOLFE et
al. (1977) and BARK et al. (1976).
Newborn infants requiring TPN have a lower energy expenditure when they are infused with a mixture of fat (25% of non-protein energy) and glucose, than when they receive glucose alone as the non-protein energy source (VAN AERDE et al., 1989). This probably reflects a reduction in de novo lipogenesis in the lipid-supplemented infants. VAN AERDE et al. (1989) also showed that the fat-glucose infusion and the glucose infusion resulted in a similar nitrogen balance.
In infants
requiring TPN for gastrointestinal anomalies, it is important to
know the influences of the level and source of infused energy on
N balance. PINEAULT et al. (1988) studied 16
appropriate-for-gestational-age newborn infants who received two
isocaloric and isonitrogenous (450 mg N/kg/d) infusions,
differing in the source of calories (high or low fat intake).
They showed that glucose and fat have an equivalent
nitrogen-sparing effect in the newborn infant. NOSE et al. (1987)
carried out similar studies in infants and children (ages 2
months to 9 years), and reported that a physiologic balance of
fat and carbohydrate (8% of energy as amino acids, 60% as
carbohydrate and 32% as fat) resulted in the greatest nitrogen
retention.
During caloric restriction
of obese individuals, VASQUEZ, MORSE and ADIBI (1985) reported
that the rate of leucine oxidation was increased, and an
isocaloric amount (500 kcal/d) of dietary carbohydrate reduced
the urinary nitrogen excretion by 12 g/wk as compared with a 500
kcal/d fat diet. The carbohydrate diet significantly decreased
the rate of leucine oxidation, whereas the fat diet had the
opposite effect. Thus, during brief caloric restriction, dietary
carbohydrate decreases the catabolism of branched-chain amino
acids, whereas dietary fat has no effect. This decreased
branched-chain amino acid oxidation is associated with protein
sparing. Plasma leucine concentration was increased during
consumption of the fat diet whereas it was not significantly
affected by the carbohydrate diet. A direct relationship between
changes in oxidation rates and changes in plasma concentration of
leucine was observed. Leucine oxidation is catalyzed by
branched-chain µ-ketoacid (BCKA) dehydrogenase,
and leucine has been shown to activate BCKA dehydrogenase by
inhibiting its phosphorylation (HUGHES and HALESTRAP, 1981;
PAXTON and HARRIS, 1984).
It is interesting to know whether the differential effects of carbohydrate and fat on leucine oxidation and nitrogen balance persist if their respective intake meets the total non-protein energy requirement. VASQUEZ, PAUL and ADIBI (1988) showed in rats that glucose inhibits the whole-body rate of leucine oxidation more effectively than fat does. The mechanism of this difference in metabolic effect appears to include a greater inhibition of muscle proteolysis by glucose than by lipid. With glucose feeding, the higher plasma insulin level suggests that insulin played a role since this hormone is known to inhibit muscle proteolysis (JEFFERSON, LI and RANNELS, 1977). However, with fat feeding, proteolysis was partially inhibited with no change in plasma insulin levels as compared to the starved animals. Therefore, other factors beside insulin are involved. TESSARI et al. (1986) showed that in dogs the availability of free fatty acids (FFA) is inversely related to whole-body proteolysis and to the oxidation of leucine. A possible mechanism for explaining this phenomenon is that an increased FFA availability enhances the hepatic NADH/NAD ratio, creating a metabolic condition which inhibits the activity of BCKA dehydrogenase activity (TESSARI et al., 1987; BUSE, HERLONG and WEIGAND, 1976; HARRIS and PAXTON, 1985; TISCHLER and FAGAN, 1982).
Another possible mechanism by which fat infusion might contribute to nitrogen sparing is through an effect of FFA on the binding to albumin of µ-ketoisocaproate (KIC), the µ-ketoacid produced by leucine desamination. During triglyceride infusion with heparin in humans, NISSEN et al. (1982) showed that the elevation of plasma FFA levels significantly increased the circulating free KIC concentration, by displacing KIC which is bound to albumin. An increase in the circulating free KIC pool may affect the concentration gradient between plasma and tissues favoring the uptake of free KIC by the liver, where reamination of KIC to leucine occurs. In addition, KIC decreases proteolysis (BUSE et al., 1972; BUSE and WEIGAND, 1977; FULKS, LI and GOLDBERG, 1975) and stimulates protein synthesis (KIRSCH, FRITH and SAUNDERS, 1976) in vitro.
The greater
effectiveness of glucose in sparing body nitrogen as compared to
lipid is probably due to the fact that it inhibits not only amino
acid oxidation but also hepatic gluconeogenesis and renal
ammoniagenesis (VASQUEZ, PAUL and ADIBI, 1988). By contrast,
lipid administration induces sparing of body nitrogen only by a
reduction of amino acid oxidation; in addition, lipid is not as
effective as glucose in inhibiting muscle proteolysis. This was
shown by a twofold greater intracellular tyrosine and leucine
concentration in muscle of lipid-than of glucose-infused rats.
Neither glucose nor lipid infusion stimulated amino acid
incorporation in muscle protein (VASQUEZ, PAUL and ADIBI, 1988).
The role of insulin and
amino acid plasma concentration on protein metabolism was
investigated in normal men (TESSARI et al., 1987). On the
basis of leucine kinetic data, it was shown that experimental
hyperinsulinemia decreased proteolysis but did not stimulate the
incorporation of leucine into protein. By contrast, elevation of
plasma levels of amino acids, by infusion of an amino acid
mixture, stimulated the incorporation of leucine into protein,
but did not suppress endogenous proteolysis. Similar results were
reported by CASTELLINO et al. (1987) showing that
hyperinsulinemia inhibits protein breakdown, and
hyperaminoacidemia stimulates protein synthesis; in contrast to
TESSARI et al. (1987), these authors reported that
hyperaminoacidemia also inhibited protein breakdown. Thus,
substrates and hormone appear to exert different and
complementary effects in stimulating protein anabolism.
The studies on the effects
of dietary glucose and lipid on protein metabolism have provided
evidence that both energy substrates have a protein-sparing
effect when given with a mixed diet. For patients who need
nutritional rehabilitation, or for growing infants in whom
protein retention is high, it may be relevant to take into
account not only the protein-sparing effect of nutrients but also
the efficiency with which the dietary energy is utilized.
After ingestion of a meal or during intravenous infusion of nutrients, the resting energy expenditure of the subject increases above the premeal baseline. This thermogenic response is often called the thermic effect of nutrients; it depends on the absorption, processing and storing of the nutrients. The carbohydrate-induced thermogenic response is about 7% of the glucose energy administered (SCHUTZ et al., 1983), whereas the thermic effect of fat is approximately 3% (THIÉBAUD et al., 1983). By contrast, the stimulation of energy expenditure following protein ingestion or amino acid infusion is close to 30% of the energy administered (FLATT, 1978), which is mainly due to the stimulation of protein synthesis, gluconeogenesis and ureogenesis. These data suggest that, to provide the non-protein energy, a mixture of carbohydrate and lipid has a smaller thermic effect than an isocaloric amount of carbohydrate. Studies in adults (NORDENSTROM et al., 1982; THIÉBAUD et al., 1983) have indicated that a combined infusion of glucose and lipids might confer a metabolic advantage over a glucose infusion alone, because it increases the efficiency of energy utilization. In full-term newborn infants, a combination of intravenous glucose and fat results in lower energy expenditure than glucose alone as non-protein energy source, leaving more energy for storage and growth (SAUER et al., 1986).
With high
glucose administration under conditions of TPN, there is
extensive stimulation of lipogenesis from glucose, a process
requiring a great deal of energy. Thus, when carbohydrates are
given in large quantities, their thermic effect is increased
because of an enhanced lipogenesis; in addition, stimulation of
the sympathetic nervous system has been reported, which results
in a 'facultative' component of thermogenesis and contributes to
further enhance energy expenditure (WELLE, LILAVIVAT and
CAMPBELL, 1981). When large amounts of glucose are given to
infants and children, one approaches a physiologic limit of
glucose oxidation. This physiologic maximum of glucose
utilization is clearly exceeded when glucose is the only
non-protein energy source given to patients requiring TPN. It
follows that when glucose alone is given, as compared with a
balanced glucose and fat regimen, some energy is wasted through
lipogenesis and stimulation of sympathetic activity, and less
energy, as ATP, is available for anabolic processes such as
protein synthesis. NOSE et al. (1987) showed in infants and
children requiring TPN that a solution providing 8% of energy as
amino acids, 60% as carbohydrate, and 32% as fat resulted in a
greater nitrogen retention than an isoenergetic solution
providing 8% of energy as amino acid, 87% as carbohydrate, and 5%
as fat. These results support the conclusion that a 'physiologic
balance' of fat and carbohydrate results in optimal nitrogen
retention.
The dietary conditions to
get the best protein-sparing effect require that energy balance
be achieved in providing carbohydrate and fat in sufficient
amounts. When severely depleted patients undergo nutritional
rehabilitation, a positive energy balance (not exceeding + 20%)
is advisable, since an energy excess may improve N retention. The
proportion of carbohydrate and fat energy depends upon the age
and clinical condition of the patients. Most studies show that a
proportion of 25 to 50% of fat as non-protein energy is to be
recommended for patients who need nutritional rehabilitation or
for growing children. It is well established that glucose and
free fatty acids have a protein-sparing action. When the clinical
conditions do not allow (or do not require) to reach energy
balance, dietary glucose has a greater protein-sparing action
than dietary lipids. This is probably mainly due to the increased
glucose-induced secretion of insulin, which inhibits muscle
proteolysis and hepatic gluconeogenesis. When energy balance is
reached, or under conditions of positive energy balance, glucose
and lipids appear to have a similar protein-sparing action. For
patients requiring TPN, the advantage of a balanced mixture of
glucose and lipid is a smaller thermogenic response to nutrients,
than when glucose is the only non-protein energy source.
When energy balance cannot be obtained, it is not reasonable to increase N intake to improve N balance, since protein oxidation will be increased as a result and the dietary protein will be used mainly as energy source with the disadvantage of an increased ureogenesis and an enhanced thermogenic response.