Protein requirements for adults
The working group on protein requirements for adults was asked to address the
following major questions:
1. An important factor in the utilization of food proteins, especially of
plant compared with animal sources, is their digestibility. What differences in
faecal nitrogen exist between populations with low protein intakes corresponding
to those used to measure obligatory losses at requirement levels, and
populations with habitual levels of protein intake from different food sources?
2. What variation in nitrogen digestibility occurs with different diets and
levels of dietary fibre in relation to various hygienic factors?
3 How should requirements, expressed as egg or milk protein, be adjusted for
other protein sources and diets (i.e., of different nutritional quality)?
4 What is the significance of zero-balance intercepts for different diets and
populations? What effect does inclusion of integumentary losses have on
estimates of nitrogen balance and of protein requirements?
5. What are the effects of acute and chronic infections and of other stress
6. What long-term indicators of dietary protein and energy adequacy are
7. What reference protein source is appropriate for establishing requirements
The task force responded to these questions as follows.
1. Faecal Nitrogen
Mean values for endogenous faecal nitrogen loss reported by various groups
range from 8 to 16 mg N/kg/day. It was thought that two factors might explain
First is the time at which measurements were made after initiation of the
experimental, low-protein or protein-free diet, although the available data are
not sufficiently extensive to assess the quantitative importance of this
Second, the ingredient composition of the experimental diet may also
influence the output of endogenous faecal nitrogen. Thus, in order to resolve
the issue as to whether differences exist between specific population groups, a
highly standardized diet would have to be used. However, a resolution of this
problem would not be expected to contribute importantly to the further
refinement of estimations of adult protein needs.
The limited data available do not suggest marked differences in faecal
nitrogen between the groups of adults studied. There appears to be little need,
therefore, to generate additional data on obligatory faecal nitrogen losses for
the purpose of refining estimates of the protein requirement.
It has been customary to consider faecal nitrogen losses of individuals after
brief adaptation to a nitrogen-free, but otherwise adequate, diet as
representing obligatory losses; the latter are subtracted from faecal losses in
calculating "true digestibility." There is no reason, however, for
making such a correction when the object of the study is to determine the total
amount of ingested nitrogen necessary to reach a defined level of body nitrogen
balance. Moreover, there is no evidence to support the assumption that
obligatory faecal losses measured on a nitrogen-free diet are representative of
metabolic faecal nitrogen losses when the protein content of the diet
approximates an intake in the requirement range. On the contrary, there are
reasons to believe that this assumption is invalid.
Measuring the nitrogen balance at various levels of intake of usual diets
avoids the problem. In other words, when attempts are made to determine the
amount of dietary protein required for long-term nitrogen equilibrium, separate
estimation of endogenous faecal nitrogen losses is unnecessary.
At requirement levels of intake with highly digestible protein
sources and refined diets, faecal nitrogen output approximates 20 per cent of
total nitrogen output and increases to about 35 per cent for diets containing
predominantly vegetable protein sources. The available data from the current UNU
sponsored studies, together with previous data from the Institute of Nutrition
of Central America and Panama (INCAP) and other data from developed countries,
suggest that there are minimal differences in mean faecal nitrogen losses
between different groups of adults when highly digestible animal protein foods
serve as the major source of protein intake. However, for diets based on
vegetable protein sources, there appear to be larger differences in faecal
nitrogen output both within and among the experimental populations comprising
subjects from developed and developing regions. Among the possible dietary
factors are kinds of protein (animal, mixed, or predominantly plant sources),
levels of dietary fibre, presence of inhibitors of digestive enzyme activity,
such as polyphenolic compounds, and the structural organization of the foods. In
addition, the population differences observed may be due to the higher faecal
nitrogen outputs arising from chronic, subclinical pathophysiological changes in
gastrointestinal function and metabolism.
At habitual levels of protein and food intake that may provide
protein at levels in excess of requirements, differences among populations may
not be readily distinguishable, but there are no adequate data to evaluate this
2. Variation in Nitrogen Digestibility
As to variation in nitrogen digestibility with different types of diet,
comparison among apparently healthy populations consuming the same diet
indicates that differences in digestibility are due largely to differences in
health, especially the reduced absorption due to what has been described as
"tropical jejunitis" 11*). There is little evidence to indicate
long-term and favourable adaptive responses of the gastrointestinal tract to
poorly digestible diets by individuals or different ethnic groups. Furthermore,
there is no good evidence that favourable adaptations in nitrogen digestion and
absorption occur in response to conditions that initially result in poor
nitrogen digestibility. Improvements over time can be achieved by appropriate
health interventions or by upgrading the sanitary conditions of the environment.
3. Adjustment of Requirements for Various Protein Sources
The adjustment of requirements expressed as egg or milk protein for other
protein sources and diets is necessary, according to data presented at this
meeting. With diets based on mixed protein sources that include small amounts
(10-25 per cent) of high quality protein, differences in protein requirements,
relative to milk or egg as a reference, appear to be due mainly to differences
in digestibility and absorption of ingested proteins. This observation
underscores the importance of including digestibility, as proposed by the 1975
Consultative Group (2), in adjustments of protein quality based on a chemical
The data presented at this meeting indicate that the mean requirement for a
protein source of nutritional value equal to that of egg or milk protein is
approximately 0.6 g/kg/day, or a value 30 per cent greater than that proposed in
the 1973 FAD/WHO report (3).
An observation made in a number of these studies was that the efficiency of
nitrogen utilization is low at very low protein intakes, in contrast to the
experience reported in the literature (4) based on individual protein sources,
such as egg or milk, and on studies in healthy Caucasian and Oriental subjects.
This difference in the observations may reflect various factors: (a) an
experimental design problem associated with rates of adaptation of nitrogen
metabolism to different nitrogen intakes, and (b) biological and metabolic
factors related to balance and utilization of amino acids. If this observation
is biologically significant it will have important implications for protein
nutritional status where people consume diets low in protein either on occasion
or for more frequent or extended periods.
The problem of further exploring the rate of adjustment in nitrogen
utilization to changes in source and level of nitrogen intake is not only a
matter of significance in experimental design, but also possibly a problem that
has importance in relation to the meal and daily pattern of protein intake to
maintain an adequate protein nutritional status.
In reference to adults, an overview of the data at the meeting supports the
appropriateness of correcting requirements for egg or milk to other protein
sources of lower quality; this correction then applies to both adults and
healthy, normally growing children. Protein quality adjustment factors must be
developed and validated under test protein levels that approach maintenance
4. Zero-Balance Intercepts
For the measurement of zero nitrogen balance intercepts, three test levels
are inadequate and four test levels represent a minimally acceptable design. In
reference to zero nitrogen balance intercepts, estimates of balance should
always include a specified level of integumentary loss, the level depending upon
the particular environment under which the nitrogen balances were obtained. No
data were presented at the meeting to substantiate the concept that there are
quantitatively significant differences in protein requirements among genetically
distinct groups of healthy individuals of comparable age and body composition.
5. Effects of Infections
Mild to moderate parasitic infestations of the gastrointestinal tract with
helminths did not reveal unfavourable effects on nitrogen balance and
utilization for generous protein intakes. The group felt that there is a
continuing need to examine this issue in further studies involving levels of
protein intakes that approximate requirements and to include other infestations,
such as Giardia.
6. Long-term Indicators
The group agreed strongly that nitrogen-balance measurements do not provide a
definitive evaluation of long-term protein requirements. There is an urgent need
to examine the metabolic and health significance of nitrogen-balance
data in critical detail.
In order to achieve this goal the most critical need is to define
better the physiology of human protein and amino acid metabolism and its
relationship to current measures of requirements, physiological function, and
health status. This requires further basic research, specifically directed
toward improving the definition of human protein and amino acid requirements and
the factors that affect them.
This question must include exploration of the design of experimental
protocols for assessing protein and amino acid requirements. The group
considered that multiple approaches are necessary to arrive at a definite
statement on these requirements, and that highly controlled metabolic studies
involving both shorter- and longer-term diet periods play an important role in
reaching this goal. However, the ultimate definition will require studies of
populations living under natural environmental conditions, and these studies
must incorporate improved measurements of food intake, as well as of the
nutritional and health status of the members of the population.
Choice of Reference Protein
The question was raised by the drafting group as to whether there was
evidence indicating a more adequate reference protein than milk or egg. This
question was addressed, in part, because various investigators at the meeting
have observed a relatively large variation in nitrogen-balance data when using
egg, compared with other sources of protein tested. Although the group did not
re-examine the data in specific detail, it was their tentative conclusion that
there is no justification for recommending high quality protein sources other
than egg, milk, or lean beef as substitute reference proteins. A good quality
soy-protein product might be considered as an alternative reference protein, but
more comparative data from other investigators and laboratories would be
required to examine this issue.
It was the recommendation of the group that a description of the reference
protein used in metabolic studies should be given and based, in part, on results
obtained with standard chemical and animal bioassay procedures. The importance
of a reference protein in human metabolic studies is analagous to the use of a
standard in analytical chemistry. However, this does not imply that concurrent
studies with a reference protein are always necessary in each series of human