Full text of DOI:10.1093/gerona/56.6.M373

Journal of Gerontology: MEDICAL SCIENCES
Copyright 2001 by The Gerontological Society of America
2
001, Vol. 56A, No. 6, M373–M380
The Recommended Dietary Allowance for Protein
May Not Be Adequate for Older People to Maintain
Skeletal Muscle
Wayne W. Campbell,1,2 Todd A. Trappe,1,2 Robert R. Wolfe, and William J. Evans
3
1,2
1The Donald W. Reynolds Department of Geriatrics, University of Arkansas for Medical Sciences, Little Rock.
2The Geriatric Research, Education, and Clinical Center, Central Arkansas Veterans Healthcare System, Little Rock.
3Metabolism Unit, Shriners Burns Institute and the University of Texas Medical Branch, Galveston.
Background. Inadequate dietary protein intake results in loss of skeletal muscle mass. Some shorter-term nitrogen
balance studies suggest that the Recommended Dietary Allowance (RDA) of protein may not be adequate for older peo-
ple. The aim of this study was to assess the adequacy of the RDA of protein for older people by examining longer-term
responses in urinary nitrogen excretion, whole-body protein metabolism, whole-body composition, and mid-thigh mus-
cle area.
Methods. This was a 14-week precisely controlled diet study. Ten healthy, ambulatory men and women, aged 55 to
7 years, were provided eucaloric diets that contained 0.8 g protein·kg ·day . The study was conducted at a General
ꢀ1
ꢀ1
7
Clinical Research Center using an outpatient setting for 11 weeks and an inpatient setting for 3 weeks. The main out-
come measures included urinary nitrogen excretion, postabsorptive and postprandial whole-body leucine kinetics via in-
13
fusion of L-[1- C]-leucine, whole-body density via hydrostatic weighing, total body water via deuterium oxide dilution,
and mid-thigh muscle area via computed tomography scans.
Results. Mean urinary nitrogen excretion decreased over time from Weeks 2 to 8 to 14 (p ꢁ .025). At Week 14,
compared with Week 2, there were no changes in postabsorptive or postprandial leucine kinetics (turnover, oxidation,
incorporation into protein via synthesis, release via breakdown, or balance). Whole-body composition (% body fat, fat-
free mass, and protein ꢂ mineral mass) did not change over time in these weight-stable subjects. Mid-thigh muscle area
2
was decreased by ꢀ1.7 ꢃ 0.6 cm (p ꢁ .019) at Week 14 compared with Week 2. The loss of mid-thigh muscle area
was associated with the decrease in urinary nitrogen excretion (Spearman r ꢁ .83, p ꢁ .010).
Conclusions. The maintenance of whole-body leucine metabolism and whole-body composition is generally consis-
tent with a successful adaptation to the RDA for protein. However, the decrease in mid-thigh muscle area and the asso-
ciation with decreased urinary nitrogen excretion are consistent with a metabolic accommodation. These results suggest
that the RDA for protein may not be adequate to completely meet the metabolic and physiological needs of virtually all
older people.
HE current Recommended Dietary Allowance (RDA)
The current consensus (1,6) is that adult protein allow-
ances should be established from shorter-term (2–3-week)
nitrogen balance studies. However, longer-term (several
weeks to several months) nitrogen balance studies should be
highly valuable in assessing dietary protein adequacy by ad-
ditional criteria, such as changes in protein metabolism,
body composition, physical status, and functional status.
Results of the limited number of shorter-term nitrogen bal-
ance studies in older adults are conflicting, with some sup-
porting (4,5) and others questioning (2,3,7) the adequacy of
for protein, 0.8 g protein·kgꢀ ·day , is set as the safe
1
ꢀ1
T
and adequate intake for virtually all healthy men and
women aged 19 years and older (1). Specific to the protein
RDA for older adults (age ꢄ50 years), several important
considerations were recognized. First, very little data were
available to help set an RDA for protein in older adults, and
the data that were available (2–5) were contradictory. Thus,
the protein RDA for older men and women was basically an
extrapolation from nitrogen balance studies in young men.
Second, whereas significant changes in body composition,
food intake, physical activity, and the frequency of disease
occur with aging, how such changes affect protein metabo-
lism and dietary protein needs was largely unknown. The
consensus was that, because of age-associated changes in
body composition, primarily the loss of muscle mass (sar-
0.8 g protein·kgꢀ ·day . Retrospective re-analyses (8) of
these shorter-term nitrogen balance data, on the basis of cal-
culations recommended by the 1985 Joint FAO/WHO/UNU
Expert Consultation (6), support the conclusion that the pro-
tein RDA may not be adequate for many older and elderly
adults. These conclusions (8) have been questioned (9,10).
To date, the longest-term nitrogen balance study in older
1
ꢀ1
copenia), an RDA of 0.8 g protein·kgꢀ ·day would be
higher per unit of fat-free mass in older people than in
younger people and should allow for any decrease in the ef-
ficiency of protein utilization (1).
1
ꢀ1
men and women fed the RDA of 0.8 g protein·kgꢀ ·day
1
ꢀ1
was 4 weeks (2). Three out of seven men and four out of
eight women were in negative nitrogen balance during Days
M373

M374
CAMPBELL ET AL.
2
6 to 30 of the study, and the authors concluded that 0.8 g
scribed by Campbell and colleagues (13). Whereas animal
striated tissues (i.e., beef, pork, poultry, fish) were not used,
animal-based proteins were provided in the forms of milk-
based proteins (28.5% ꢃ 1.2% of total protein) and egg-
based proteins (10.2% ꢃ 0.8% of total protein). The non-
protein portion of each of the three menus contained 60% of
energy from carbohydrate and 40% of energy from fat. Wa-
ter, decaffeinated coffee, and decaffeinated tea were al-
lowed ad libitum.
The total energy intake of each subject was initially set
on the basis of the sex-specific Harris-Benedict equation
(14) of resting energy expenditure plus an allowance of 0.7
times this predicted resting energy expenditure to account
for the energy expenditure of physical activity. Total energy
intake was adjusted during the nonresidency periods as nec-
essary to maintain body weight within ꢃ0.5 kg of each sub-
ject’s mean body weight during Days 4 to 15. The adjust-
ments to total energy intake were done either by adding or
subtracting protein-free or very-low–protein foods and bev-
erages from each subject’s daily menus while maintaining
the nonprotein energy ratio at 60% carbohydrate to 40% fat.
The energy and macronutrient contents of the daily menus
were calculated using Nutritionist IV software (version 4.0;
N-squared Computing, First Data Band, San Bruno, CA) as-
suming that the metabolizable energy content of protein,
carbohydrate, and fat were 16.7, 16.7, and 37.7 kJ/g, respec-
tively.
ꢀ
1
ꢀ1
protein·kg ·day was not adequate for a majority of men
and women aged 70 years and older (2). More recently,
Castaneda and colleagues (11,12) assessed the longer-term
metabolic and physiological changes of 66- to 79-year-
old women who consumed either 0.45 or 0.92 g pro-
tein·kg ·day (56% or 115% of the RDA) for 9 weeks.
The marked negative nitrogen balance, loss of body cell
mass and muscle mass, reduced muscle strength, and im-
paired immune responses all indicated that 0.45 g pro-
tein·kg ·day was inadequate and compromised the phys-
ical and functional capacities of these elderly women. In
contrast, the apparent adequacy of 115% of the protein
RDA for elderly women was supported by the maintenance
of body composition, functional capacity, and immune re-
sponses.
The present study assesses the adequacy of the RDA for
protein in older people in a longer-term nitrogen balance
study. Older men and women were fed diets providing 0.8 g
ꢀ
1
ꢀ1
ꢀ
1
ꢀ1
ꢀ1
ꢀ1
protein·kg ·day and sufficient energy to maintain body
weight continuously for 14 weeks. The criteria of adequacy
included measurements of urinary nitrogen excretion, whole-
body protein metabolism, body composition, energy metab-
olism, and muscle strength and function. It is hypothesized
that the protein RDA is not adequate to maintain these indi-
cators of metabolic, physical, and functional status.
METHODS
Resting Energy Expenditure
Postabsorptive resting energy expenditure was measured
at Weeks 2 and 14 by indirect calorimetry, as described by
Campbell and colleagues (13).
Subjects
Four men and six women, aged 55 to 77 years, partici-
pated in this metabolic study. The study was completed at
the General Clinical Research Center (GCRC) located at the
University Park campus of The Pennsylvania State Univer-
sity. Prestudy medical evaluations, which included a medi-
cal history questionnaire, a physician-administered physical
examination, routine blood and urine chemistries, and a
resting and resistance exercise electrocardiogram, were used
to exclude people with clinically abnormal thyroid, liver,
kidney, or heart function. All six of the women studied were
postmenopausal and were not taking any estrogen replace-
ment medications. Each subject signed an informed consent
agreement. The study protocol and informed consent agree-
ment were reviewed and approved by the Institutional Re-
view Board, The Pennsylvania State University, University
Park, PA and by the Clinical Investigation Committee, The
Milton S. Hershey Medical Center, Hershey, PA.
Body Composition
Postabsorptive, nude body weight was measured (model
181; Toledo Scale, Toledo, OH) to the nearest 0.1 kg each
2
weekday during the study, as described previously (13).
Body height was measured to the nearest 0.1 cm with a
wall-mounted stadiometer one morning during Week 1.
Whole-body density was measured by hydrostatic weigh-
ing (15), with lung residual volume measured by nitrogen
dilution (16) during the underwater procedure. Total body
water was measured by the deuterium oxide dilution tech-
nique (17) using a 20.0 g dose of deuterium oxide and a
fixed-filter single-beam infrared spectrophotometer (Miran
1
FF; Foxboro Analytical, South Norwalk, CT) for analyses
of prepared urine samples, as described by Campbell and
colleagues (18). Percent body fat was calculated from body
density and total body water using the 3-compartment equa-
tion of Siri (19), as described previously (18). Fat-free mass
(FFM) was calculated as body mass minus fat mass, and
protein plus mineral mass was calculated as FFM minus
body water mass.
Experimental Design
The study design was a 14-week precisely controlled di-
etary intake trial. Testing and evaluations were completed at
study Weeks 2, 8, and 14, as indicated. During these weeks,
each subject was in residency at the GCRC. The remainder
of the study was conducted on an outpatient basis.
Mid-thigh muscle and fat areas were measured by image
analyses (IMAGE, version 1.60; National Institutes of Health,
Bethesda, MD) of computed tomography (CT) scans (Picker
2000 operating at 130 kVp; Picker International Inc., Cleve-
land, OH). The CT scans were taken of the dominant leg at
Weeks 2 and 14; a 10-mm CT slice was taken midway be-
tween the inguinal crease and the lower pole of the patella.
Diet
Each subject’s diet was designed to provide 0.8 g pro-
ꢀ
1
ꢀ1
tein·kg ·day and sufficient dietary energy to maintain
body weight. The diet was provided as a rotating cycle of
three daily menus of lacto-ovo-vegetarian foods, as de-

THE RDA FOR PROTEIN AND OLDER ADULTS
M375
At Week 2, the exact location of the mid-thigh CT slices
were determined from bony landmarks on the femur and
were used to identify the identical slice location for the
Week-14 scans. Digital Imaging and Communication in
Medicine (DIACOM) software (version 3.0) was used to
maintain spatial and density calibration during the transfer
of the images from the CT scanner to a SUN (SPARK) sta-
tion for imaging analysis. Digitized images, analyzed in a
blinded fashion, were used to calculate the total muscle and
subcutaneous fat areas. The Hounsfield units used to detect
muscle and fat were ꢀ30 to ꢂ150 and ꢀ250 to ꢀ40, re-
spectively. Computed tomography data from two subjects
were inadvertently lost during processing. Therefore, results
are presented for eight subjects (3 men and 5 women).
The constant infusion was continued for 8 hours, with all
subjects in the postabsorptive state during the first 4 hours
and in the postprandial state during the last 4 hours. During
the postprandial period, each subject consumed a formula
beverage hourly (minutes 240, 300, 360, and 420 of the
infusion procedure) providing one-twelfth of their daily
protein and energy intakes at baseline. The dietary leucine
concentration of each formula beverage was estimated as-
suming factors of 6.38 g protein/g nitrogen and 95 mg leu-
cine/g protein (6). All blood and breath samples were ob-
tained and processed using standard methods (21,22). The
1
3
C enrichment of plasma ꢈ-ketoisocaproic acid (KIC) was
determined by gas chromatography mass spectrometry
(model HP 5989; Hewlett-Packard, Palo Alto, CA) using
the silylquinoxalinol derivative (22). The expired air sam-
ples were analyzed for 13CO enrichment by isotope ratio
2
Muscle Strength and Power
mass spectrometry (model cira 2; VG Isogas, VG Instru-
ments, Middlewich, UK). Prior to and during the fourth and
Dynamic concentric muscle strength was measured as the
maximum force that each subject could move through a full
range of motion one time only for the seated chest press,
seated arm pull, seated unilateral knee extension, and seated
bilateral leg curl exercises (Keiser Sports Health Equip-
ment, Fresno, CA). Upper-body strength was calculated as
the sum of the linear forces (newtons) of the chest press and
arm pull exercises. Lower-body strength was calculated as
the sum of the angular forces (newton ꢅ meter) of the knee
extension and leg curl exercises. Peak leg power was mea-
sured using a NUMAZ (University of Nottingham, UK)
power rig (20). Subjects were asked to exert a maximal push
of both legs to a large foot pedal while in a seated position
with folded arms. The highest value (watts) from 10 trials
was recorded, with each trail separated by a minimum of 60
seconds of rest.
eighth hours of the infusion, the rate of CO production was
2
measured by indirect calorimetry using a ventilated hood
system.
Calculations
All kinetics parameters are expressed as ꢇmol leu-
cine·kgꢀ ·hour . The parameters included leucine turn-
1
ꢀ1
over, leucine oxidation, leucine incorporation into protein
(
synthesis), leucine release from protein (breakdown), and
leucine intake from the diet. Leucine turnover was calcu-
lated as described by Matthews and colleagues (23), with
1
3
the intracellular leucine pool estimated by the plasma [ C]
KlC enrichment at plateau (24). The rate of leucine oxida-
tion was calculated as described previously (23), assuming
fractional bicarbonate retention factors in the postabsorptive
and postprandial states of 0.70 and 0.82, respectively (25).
Synthesis was calculated as turnover minus oxidation.
Breakdown was calculated as turnover minus intake. In the
postabsorptive state intake from diet was zero, and break-
down equaled turnover. In the postprandial state intake was
corrected for the amount of leucine estimated to be removed
during the first pass through the splanchnic tissues (50%)
Nitrogen Intake and Excretion
At Weeks 2, 8, and 14, samples from duplicate menu
composites, four consecutive 24-hour urine collections, and
four-day fecal collections (made between two dye markers)
were collected, processed, and aliquots stored frozen at
ꢀ20ꢆC. All food, urine, and feces samples were analyzed
for total nitrogen using an Elementar Macro N (Elementar
Analysensytene GmBh, Hanau, Germany) nitrogen ana-
lyzer. The National Institute of Standards and Technologies
Total Diet standard reference material and in-house pooled
food and urine standards were used as quality controls. Di-
(
26). Predicted 24-hour leucine balance (input ꢀ output,
ꢀ1
ꢇmol·kg ) was calculated from leucine intake (dietary)
and leucine oxidation, as described by El-Khoury and col-
leagues (27).
etary protein intake was calculated from the food nitrogen
Serum Albumin
data assuming a factor of 6.25 g protein·g nitrogenꢀ
1
.
A postabsorptive-state serum sample collected at baseline
of the infusion procedure was analyzed for albumin concen-
tration using an automated analyzer technique at a clinical
medical laboratory (American Medical Laboratory, Chan-
tilly, VA).
Infusion Procedure
At the end of Weeks 2 and 14, a primed constant infusion
1
3
of L-[1- C]leucine was performed, as described previously
21), to determine postabsorptive and postprandial leucine
(
kinetics. After baseline blood and expired breath samples
were collected, the infusion of the isotope was begun with
Statistical Methods
Values are reported as means ꢃ SEM. The effect of time
was assessed either by paired t test (Week 2 vs Week 14)
or one-factor repeated measures analysis of variance
(ANOVA) (Weeks 2, 8, and 14). For these paired t tests,
statistical significance was assigned if p ꢉ .05 (two-sided).
For the one-factor repeated measures ANOVA, when a sta-
tistically significant time effect was established, separate
13
the intravenous administration of priming doses of NaH CO
3
ꢀ
1
13
ꢀ1
(
2.35 ꢇmol·kg ) and L-[1- C]leucine (7.6 ꢇmol·kg ).
This was followed immediately by a continuous infusion
1
3
ꢀ1
ꢀ1
of L-[1- C]leucine (7.6 ꢇmol·kg ·hour ) using a cali-
brated syringe pump (model 55-2222; Harvard Apparatus,
Natick, MA).

M376
CAMPBELL ET AL.
comparisons (paired t tests) were done among time points
i.e., Weeks 2 vs 8, Weeks 2 vs 14, and Weeks 8 vs 14). For
Table 2. Dietary Energy, Macronutrient Intakes, and
Nitrogen Excretions
(
these comparisons, the Bonferroni correction was applied,
and statistical significance was assigned if p ꢉ .017 (two-
sided). For the leucine kinetics data, a two-factor repeated
measures ANOVA was used to assess the main effects of
metabolic state (postabsorptive vs postprandial) and time
Parameter
Week 2
Week 8
Week 14
Dietary Intakes
Energy
MJ/d
kcal/d
9.99 ꢃ 0.75 10.99 ꢃ 1.08 10.95 ꢃ 0.99
2388 ꢃ 180 2626 ꢃ 259 2617 ꢃ 236
(
Weeks 2 vs 14) and the metabolic state-by-time interaction.
For these tests, statistical significance was assigned if p ꢉ
05. The degree of linear association between variables was
Protein, g/d
Carbohydrate, g/d
Fat, g/d
52.9 ꢃ 3.0
52.9 ꢃ 2.9
53.0 ꢃ 2.9
347.2 ꢃ 43.3 364.6 ꢃ 39.7 362.2 ꢃ 35.4
95.5 ꢃ 7.4 105.3 ꢃ 10.2 106.3 ꢃ 9.5
.
Nitrogen Intakes and Excretions (mg Nꢊkgꢀ1ꢊdꢀ1
Nitrogen intake
Urinary nitrogen excretion* 102.9 ꢃ 4.4
Fecal nitrogen excretion 23.7 ꢃ 2.0
)
established by using the Spearman Rho nonparametric
ranked correlation (Prob ꢄ |Rho|, p ꢉ .05). All data were
processed using Microsoft Excel 5.0 (Microsoft, Redmond,
WA). Statistical evaluations were done using JMP software
131.3 ꢃ 1.3 131.0 ꢃ 1.5 130.1 ꢃ 1.5
87.7 ꢃ 4.1
25.2 ꢃ 2.3
80.2 ꢃ 4.3**
22.1 ꢃ 2.4
Note: Data are presented as mean ꢃ SEM.
(
version 3.2.2; SAS Institute, Inc., Cary, NC).
*
*
Time effect; p ꢁ .025.
*Different from Week 2; p ꢁ .005.
RESULTS
Table 1 presents the group mean values for the descrip-
tein·kgꢀ1·dayꢀ1. These levels of protein intake represent
tive characteristics, body composition, mid-thigh muscle
and fat areas, resting metabolic rate, and serum albumin
concentration, at Week 2 and Week 14. Body weight, BMI,
body density, total body water, % body fat, FFM, protein-
mineral mass, mid-thigh circumference and fat area, resting
metabolic rate, and serum albumin were not different at
Week 14 compared with Week 2. Mid-thigh muscle area
was decreased by ꢀ1.7 ꢃ 0.6 cm (p ꢁ .019; n ꢁ 8) at
Week 14 compared with Week 2.
Upper body strength (672 ꢃ 98 N vs 657 ꢃ 89 N; n ꢁ 8),
lower body strength (297 ꢃ 42 Nm vs 288 ꢃ 33 Nm; n ꢁ
about 102% of the current RDA of protein (1).
Dietary nitrogen intake was 131.3 ꢃ 1.3 g N·kgꢀ1·dayꢀ1
at Week 2 and was not different over time (i.e., at Week 8
and Week 14; Table 2). At Week 2, urinary nitrogen excre-
tion of each subject did not change over the four days of
collection (i.e., no significant slope was observed when re-
gression analysis was performed), consistent with each sub-
ject being in steady state. Over time, urinary nitrogen excre-
tion decreased (p ꢁ .025), and fecal nitrogen excretion was
not different. Subsequent paired t test analyses (with Bon-
ferroni correction) showed that urinary nitrogen excretion
was different between Week 2 and Week 14 (p ꢁ .005).
From Week 2 to Week 14, the change in urinary nitrogen
excretion was associated with the change in mid-thigh mus-
cle area (r ꢁ .83, p ꢁ .010; Table 3 and Figure 1).
Table 4 presents the whole-body leucine kinetics data.
Leucine turnover and oxidation were higher, and synthesis
and breakdown were not different in the postprandial state
versus the postabsorptive state. Postabsorptive leucine bal-
ance was negative, and postprandial leucine balance was
positive. Net leucine balance was negative. There were no
significant changes over time or metabolic state- (postab-
sorptive vs postprandial) by-time interactions for any of the
leucine kinetics parameters.
2
8
), and leg peak power (222 ꢃ 47 W vs 216 ꢃ 44 W; n ꢁ 7)
were not different at Week 14 compared with Week 2.
From computer-based diet analyses, the group mean di-
etary intakes at Week 2 were 9.99 ꢃ 0.75 MJ energy/d, 52.9 ꢃ
3
7
.0 g protein/d, 347 ꢃ 43.3 g carbohydrate/d, and 95.5 ꢃ
.4 g fat/d and were not significantly different at Week 8
and Week 14 (Table 2). On the basis of nitrogen analysis,
the group mean protein intakes at Weeks 2, 8, and 14 were
0
.821 ꢃ 0.008, 0.819 ꢃ 0.009, and 0.813 ꢃ 0.009 g pro-
Table 1. Subject Characteristics
Parameter
Week 2
Week 14
Age, y†
Height, cm
66 ꢃ 3 (range 55–77)
167 ꢃ 3
DISCUSSION
When longer-term metabolic balance studies are used to
assess the adequacy of protein intake, it is important to dis-
tinguish between adaptation and accommodation (28–31).
Adaptation, an appropriate and desired response, refers to
metabolic changes that occur in response to changes in pro-
tein intake and result in the establishment of a new steady
state without a compromise or loss in physiological func-
tion. Accommodation, a survival response, refers to further
metabolic changes in response to the decreased protein in-
take that the body undergoes to establish steady state, but
only with a compromise or loss in physiological function.
Waterlow (30,31) and Young and colleagues (28,29) have
proposed that it might be possible to distinguish between
adaptation and accommodation by measuring longer-term
changes in urinary nitrogen excretion in response to mar-
Weight, kg
67.7 ꢃ 4.1
24.2 ꢃ 1.0
1.020 ꢃ 0.006
28.9 ꢃ 2.3
38.7 ꢃ 2.8
42.1 ꢃ 3.1
12.5 ꢃ 0.9
50.9 ꢃ 1.6
100.4 ꢃ 8.0
76.8 ꢃ 8.0
5.67 ꢃ 0.33
39 ꢃ 3
67.7 ꢃ 4.0
24.2 ꢃ 0.9
1.019 ꢃ 0.006
27.8 ꢃ 3.0
40.1 ꢃ 3.0
41.0 ꢃ 3.0
12.6 ꢃ 0.9
50.9 ꢃ 1.6
98.7 ꢃ 7.5*
78.3 ꢃ 14.1
5.65 ꢃ 0.38
40 ꢃ 3
Body mass index, kg/m2
Body density, kg/l
Total body water, l
Body fat, %
Fat-free mass, kg
Protein-mineral mass, kg
Thigh circumference, cm
Thigh muscle area, cm
Thigh fat area, cm2
2
Resting metabolic rate, MJ/d
Albumin (serum), g/l
Note: Data are presented as mean ꢃ SEM, n ꢁ 10 subjects, except for thigh
circumference, muscle area, and fat area (n ꢁ 8 subjects).
*
Significantly different from Week 2; p ꢁ .019.
†
The ages of the subjects were 55, 56, 58, 62, 63, 68, 71, 72, 76, and 77.

THE RDA FOR PROTEIN AND OLDER ADULTS
M377
Table 3. Changes in Steady-State Urinary Nitrogen Excretion and Leg Muscle Area
Urinary Nitrogen Excretion
Leg Muscle Area
Week 2
(g N/d)
Change at Week
14 (g N/d)
Week 2
(cm )
Change at Week
2
2
Subject
%
14 (cm )
%
1
2
3
4
5
6
7
8
7.36 ꢃ 0.28
5.01 ꢃ 0.49
4.78 ꢃ 0.22
5.13 ꢃ 0.12
7.04 ꢃ 0.73
6.91 ꢃ 1.19
7.07 ꢃ 0.81
9.98 ꢃ 0.80
6.66 ꢃ 0.60
ꢀ3.07
ꢀ0.48
ꢂ0.77
ꢀ0.13
ꢀ1.37
ꢀ2.25
ꢀ1.67
ꢀ4.93
ꢀ42
ꢀ10
ꢂ16
ꢀ3
ꢀ19
ꢀ33
ꢀ24
ꢀ49
ꢀ21 ꢃ 8
104.5
89.9
69.5
82.1
82.0
130.4
120.8
123.9
100.4 ꢃ 8.1
ꢀ2.82
ꢀ1.42
ꢂ0.22
ꢀ1.13
0.79
ꢀ3.31
ꢀ3.02
ꢀ3.41
ꢀ2.7
ꢀ1.6
ꢂ0.3
ꢀ1.4
ꢂ1.0
ꢀ2.4
ꢀ2.5
ꢀ2.8
ꢀ1.5 ꢃ 0.5
Group
ꢀ1.84 ꢃ 0.53
ꢀ1.74 ꢃ 0.57
Note: Data are presented as mean ꢃ SEM.
ginal or inadequate protein intakes. Upon changing from a
higher (completely adequate) to a lower protein intake, uri-
nary nitrogen excretion decreases within approximately 1
week to a new steady state. Rand and colleagues (32)
showed that a new steady state was achieved in both young
men and elderly women in an average of 4.5 days, with 95%
of all subjects achieving steady state within 8 days. For sub-
jects who metabolically adapt, this new steady state would
be maintained, and urinary nitrogen excretion would remain
fairly constant over longer periods of time. In contrast, for
subjects who metabolically accommodate, urinary nitrogen
excretion would gradually decrease over time, presumably
due to decreased total body nitrogen, until a new lower
steady state was achieved. This accommodation is subtle and
may require several years to achieve a new steady state (31).
To distinguish between adaptation and accommodation,
three criteria must be met. First, was steady state achieved
and was urinary nitrogen excretion stable after the subjects
consumed the lower-protein diet? Second, was urinary ni-
trogen excretion maintained (consistent with adaptation) or
decreased (consistent with accommodation) in the subjects
over longer periods of time? Third, was it possible to show
that the change in urinary nitrogen excretion over time was
associated with an adverse change in physiological function
consistent with accommodation? In the present study, all
subjects presumably achieved steady state at Week 2, as ev-
idenced by the absence of significant changes in urinary ni-
trogen excretion over time from Day 8 to Day 11. Second,
from Week 2 to Week 14, the group mean urinary nitrogen
excretion decreased by ꢀ1.84 ꢃ 0.53 g N/d (ꢀ21% ꢃ 8%),
but with a wide within-group variability in response (ꢂ16%
to ꢀ49%) (Table 3). Third, the changes over time in urinary
nitrogen excretion were associated with changes in leg mus-
cle area (Figure 1). The subjects who showed greater reduc-
tions in urinary nitrogen excretion over time also experi-
enced greater losses of mid-thigh skeletal muscle area.
Thus, these data may be interpreted as accommodation.
This interpretation is made with all due caution. For ex-
ample, one would expect that subjects experiencing accom-
modation due to an inadequate protein intake, reflected by a
decrease over time in urinary nitrogen excretion (28,29,31),
might also show adverse changes in whole-body composi-
tion, serum albumin concentration, or muscle strength and
power. The apparent maintenance of these parameters is
generally consistent with the interpretation that the subjects
Table 4. Whole-Body Leucine Kinetics
Parameter
Week 2
Week 14
Leucine Kinetics, ꢇmol Leucineꢊkgꢀ1ꢊhꢀ1
Turnover*
Fasting
Fed
98.9 ꢃ 3.1
116.0 ꢃ 4.3
94.3 ꢃ 3.4
114.7 ꢃ 4.1
Oxidation**
Fasting
Fed
19.6 ꢃ 1.4
33.4 ꢃ 1.2
17.9 ꢃ 1.4
31.1 ꢃ 1.4
Synthesis
Fasting
Fed
79.3 ꢃ 2.7
82.6 ꢃ 4.5
76.4 ꢃ 2.6
83.6 ꢃ 3.6
Breakdown
Fasting
Fed
98.9 ꢃ 3.1
94.9 ꢃ 4.0
94.3 ꢃ 3.4
93.7 ꢃ 4.0
†
Balance **
Fasting
Fed
Total
ꢀ19.6 ꢃ 1.4
8.9 ꢃ 1.8
ꢀ10.7 ꢃ 2.2
ꢀ17.9 ꢃ 1.4
10.8 ꢃ 1.4
ꢀ7.1 ꢃ 2.0
Note: Data are presented as mean ꢃ SEM.
†
Balance ꢁ leucine intake (dietary) ꢀ leucine oxidation; total balance ꢁ
fasting balance ꢂ fed-state balance.
Figure 1. Correlation between changes in urinary nitrogen excre-
tion and leg muscle area in older people who consumed the Recom-
mended Dietary Allowance for protein (0.8 g protein·kgꢀ ·day ) for
1
ꢀ1
*p ꢁ .002; **p ꢉ .001; Fed state different from fasting state (i.e., main ef-
fect of metabolic state).
1
4 weeks.

M378
CAMPBELL ET AL.
successfully adapted to this protein intake by increasing the
efficiency of nitrogen retention and nitrogen availability
without resorting to the breakdown of body proteins. It is
generally recognized that nitrogen balance data are inher-
ently biased toward showing a more positive balance (33),
associated with either incomplete dietary protein intake or
incomplete collection of nitrogen excretions. However, it
seems unlikely that the decrease over time in urinary nitro-
gen excretion was due to a systematic shift in food con-
sumption or urine collection during the study. It is noted,
also, that computed tomography scanning is a more sensi-
tive and accurate method of detecting subtle body composi-
tion changes than measures of whole-body density or total
body water (34).
It is important to point out the limitations of the current
study. Future studies that aim to assess protein intake ade-
quacy of older people might include a larger number of sub-
jects, higher and lower protein control groups, and a younger
age control group. Another consideration might be to extend
the baseline period to achieve an initial steady state for uri-
nary nitrogen excretion beyond 8 to 10 days. Whereas Rand
and colleagues (32) reported that 95% of young men and el-
derly women achieved steady state within 8 days, their study
was conducted for only 10 days. There is some suggestion
that it might take longer than 10 days to adjust to a given
protein intake and to achieve steady state (2). In the present
study, we did not determine the pattern of change in urinary
nitrogen excretion between Week 2 and Week 8.
The lack of a substantial change over time in whole-body
leucine kinetics in response to the diet is consistent with
previous research (12). The accommodation to marginal
protein intake is subtle and takes place over many months
(30,31). The increase in nitrogen loss that led to a measur-
able decrement in leg muscle over 14 weeks resulted from
an imbalance between muscle protein synthesis and break-
down. However, the imbalance between synthesis and
breakdown over any period of a few hours that would even-
tually lead to a loss of 1 kg of muscle over 14 weeks (1.8 g
nitrogen·dꢀ ꢅ 98 days ꢅ 6.25 g protein·g nitrogen
1
ꢀ1
)
would not be expected to be detectable with the isotopic
technique. Thus, a loss of 1.84 g nitrogen·dꢀ would corre-
1
spond to approximately 5.8 ꢇmol leucine·kgꢀ ·hour
1
ꢀ1
.
Sarcopenia, the age-associated loss of skeletal muscle
mass, is indeed a multi-factorial syndrome (35–37). We do
not presume that inadequate dietary protein intake is the
only cause of sarcopenia. However, inadequate protein in-
take causes considerable losses of muscle mass in elderly
people (11), and the present data suggest that more subtle,
but significant, losses of muscle occur with longer-term
consumption of the protein RDA in older people. Thus, ha-
bitual consumption of marginal amounts of protein by older
people may contribute to sarcopenia. Research showing no
association between protein intake and skeletal muscle size
in elderly people were conducted using healthy, free-living,
predominantly Caucasian subjects, the majority of whom
consumed diets providing protein in excess of the RDA
(38,39).
We (8,40) and other researchers (9,10,41) agree that there
are insufficient data to establish a mean protein requirement
and a suggested safe and adequate protein allowance for
older persons, and the present study was not designed to ac-
complish these tasks. Tentatively, however, retrospective
reassessments of published shorter-term nitrogen balance
studies (8), using the currently standard nitrogen balance
formula (6), estimate a mean protein requirement of 0.9 g
Prior to the current study, a nitrogen balance study in 15
men and women aged 70 years or older who consumed 0.8 g
ꢀ
1
ꢀ1
protein·kg ·day for 30 days provided the longest-term
assessment of the RDA for protein in older people (2). Ger-
sovitz and colleagues (2) showed that nitrogen balance, neg-
ative for 13 of the 15 subjects at Days 1 through 10, in-
creased over time due to a reduction in urinary nitrogen
excretion but remained negative in 7 subjects at Days 21
through 30. Body weight trended down over time, but no
significant changes in muscle mass (assessed via urinary
creatinine excretion) or body cell mass (assessed via 40K-
potassium scans) were detected. From these data, the au-
protein·kgꢀ ·day and suggest a protein allowance of at
1
ꢀ1
ꢀ1
ꢀ1
least 1.0 g protein·kg ·day for older people. An intake of
1.0 g protein·kgꢀ ·day was shown to be adequate to main-
tain protein status in healthy, free-living older people in the
United States (39) and in western European countries
(42,43). A 10-year longitudinal study in initially healthy el-
derly women showed that women who habitually consumed
1
ꢀ1
thors concluded that the RDA of 0.8 g protein·kgꢀ ·day
1
ꢀ1
was inadequate for some older people but that the shift in
nitrogen balance over time and no measurable changes in
body composition indicated adaptation to the diet. The au-
thors cautioned, however, that the relatively short-term
study period limited the ability to quantify subtle changes in
muscle or body cell mass.
greater than 1.2 g protein·kgꢀ ·day
1
ꢀ1
developed fewer
health problems than those who consumed less than 0.8 g
protein·kgꢀ ·day (44).
1
ꢀ1
Castaneda and colleagues (11) reported that postmeno-
pausal women who consumed 0.45 g protein·kgꢀ ·day
1
ꢀ1
(
56% of the RDA) for 9 weeks experienced 8% declines in
Conclusions
body cell mass and skeletal muscle mass along with de-
clines in muscle strength and function. These results clearly
Many of the results of this 14-week assessment of 0.8 g
protein·kgꢀ ·day suggest that this protein intake is ade-
quate for older people. However, a gradual decline in uri-
nary nitrogen excretion and loss of body cell mass with
longer-term consumption of inadequate protein (0.45 g pro-
1
ꢀ1
showed 0.45 g protein·kgꢀ ·day to be an inadequate pro-
tein intake and to cause physiological accommodation in
older women. In contrast, postmenopausal women who con-
1
ꢀ1
sumed 0.92 g protein·kgꢀ ·day (115% of the RDA) main-
1
ꢀ1
tein·kg ·day ), reported by Castaneda and colleagues
(11), is consistent with the concept of metabolic and physio-
logical accommodation (28,29,31). The gradual decline in
urinary nitrogen excretion over time in older people who
ꢀ1
ꢀ1
tained body cell mass and skeletal muscle mass over the
9
-week period. These results indicated that a protein intake
greater than the RDA was adequate for these older women
to adapt and maintain physiological status.
consumed 0.80 g protein·kgꢀ ·day reported by Gersovitz
1
ꢀ1

THE RDA FOR PROTEIN AND OLDER ADULTS
M379
and colleagues (2) suggests that the RDA might be margin-
ally inadequate and result in longer-term accommodation in
skeletal muscle. The loss of skeletal muscle reported in the
present study supports this suggestion. The RDAs are de-
fined as “the levels of intake of essential nutrients that, on
the basis of scientific knowledge, are judged . . . to be ade-
quate to meet the known nutrient needs of practically all
healthy persons” (1). Research should continue to question
whether the RDA for protein is indeed adequate to meet the
dietary needs of older people.
16. Wilmore JH. A simplified method for determination of residual lung
volumes. J Appl Physiol. 1969;27:96–100.
1
7. Schloerb PR, Friis-Hansen BJ, Edelman IS, Solomon AK, Moore FD.
The measurement of total body water in the human subject by deute-
rium oxide dilution. J Clin Invest. 1950;29:1296–1310.
18. Campbell WW, Crim MC, Young VR, Evans WJ. Increased energy re-
quirements and body composition changes with resistance training in
older adults. Am J Clin Nutr. 1994;60:167–175.
1
9. Siri WE. Body composition from fluid spaces and density: analysis of
methods. In: Brozek J, Henschel A, eds. Techniques for Measuring
Body Composition. Washington, DC: National Academy of Sciences;
1961:223–244.
2
2
2
0. Bassey EJ, Short AH. A new method for measuring power output in a
single leg extension: feasibility, reliability, and validity. E J Appl
Physiol Occup Physiol. 1990;60:385–390.
1. Campbell WW, Crim MC, Young VR, Joseph LJ, Evans WJ. Effects
of resistance training and dietary protein intake on protein metabolism
in older adults. Am J Physiol. 1995;268:E1143–E1153.
2. Wolfe RR. Radioactive and Stable Isotope Tracers in Biomedicine:
Principles and Practice of Kinetic Analysis. New York: Wiley-Liss;
1992.
Acknowledgments
This study was supported by National Institutes of Health Grants RO1
AG11811 and 1 R29 AG13409 and General Clinical Research Center Grant
MO1 RR10732.
We are grateful for the cooperation of the volunteers in this study. Spe-
cial thanks go to the staff members of the GCRC nursing and metabolic
kitchen for providing expert support for this study. We especially thank
Deanna Cyr-Campbell, MS, RD, for developing the diets and coordinating
the production and distribution of the menus.
23. Matthews DE, Motil KJ, Rohrbaugh DK, Burke JF, Young VR, Bier
DM. Measurement of leucine metabolism in man from a primed, con-
13
tinuous infusion of L-[1- C] leucine. Am J Physiol. 1980;238:E473–
E479.
Address correspondence to W.J. Evans, Nutrition, Metabolism and Exer-
cise Division, Renolds Department of Geriatrics and GRECC, 4301 West
Markham Street, Slot 806, Little Rock, AR 72205. E-mail: evanswilliamj@
exchange.uams.edu
2
4. Matthews D, Schwarz H, Yang R, Motil K, Young V, Bier D. Rela-
13
tionship of plasma leucine and a-ketoisocaproate during a L-[1- C]
leucine infusion in man: a method for measuring human intracellular
leucine tracer enrichment. Metabolism. 1982;31:1105–1112.
2
5. Hoerr RA, Yu YM, Wagner DA, Burke JF, Young VR. Recovery of
1
3C in breath from NaH13CO3 infused by gut and vein: effect of feed-
References
ing. Am J Physiol. 1989;257(Endocrinol. Metab. 20):E426–E438.
1
. National Research Council. Recommended Dietary Allowances. 10th
ed. Washington, DC: National Academy Press; 1989.
26. Boirie Y, Gachon P, Beaufrere B. Splanchnic and whole-body leucine
kinetics in young and elderly men. Am J Clin Nutr. 1997;65:489–495.
27. El-Khoury AE, Fukagawa NK, Sanchez M, et al. Validation of the
tracer-balance concept with reference to leucine: 24-h intravenous
tracer studies with L-[1–13C]leucine and [15N–15N]urea. Am J Clin
Nutr. 1994;59:1000–1011.
2
. Gersovitz M, Munro H, Scrimshaw N, Young V. Human protein re-
quirements: assessment of the adequacy of the current recommended
dietary allowance for dietary protein in elderly men and women. Am J
Clin Nutr. 1982;35:6–14.
3
. Uauy R, Scrimshaw N, Young V. Human protein requirements: nitro-
gen balance response to graded levels of egg protein in elderly men
and women. Am J Clin Nutr. 1978;31:779–785.
. Zanni E, Calloway D, Zezulka A. Protein requirements of elderly men.
J Nutr. 1979;109:513–524.
. Cheng A, Gomez A, Bergan J, Tung-Ching L, Monckeberg F, Chi-
chester C. Comparative nitrogen balance study between young and
aged adults using three levels of protein intake from a combination
wheat-soy-milk mixture. Am J Clin Nutr. 1978;31:12–22.
. FAO/WHO/UNU. Energy and Protein Requirements. Geneva, Swit-
zerland: World Health Organization; 1985. (Technical Report Series
28. Young VR. Kinetics of human amino acid metabolism: nutritional im-
plications and some lessons. Am J Clin Nutr. 1987;46:709–725.
29. Young VR, Marchini JS. Mechanisms and nutritional significance of
metabolic responses to altered intakes of protein and amino acids, with
reference to nutritional adaptation in humans. Am J Clin Nutr. 1990;
51:270–289.
30. Waterlow JC. Protein turnover with special reference to man. Q J Exp
Physiol. 1984;69:405–438.
31. Waterlow JC. Nutritional adaptation in man: general introduction and
concepts. Am J Clin Nutr. 1990;51:259–263.
32. Rand WM, Young VR, Scrimshaw NS. Change of urinary nitrogen ex-
cretion in response to low-protein diets in adults. Am J Clin Nutr.
1976;29:639–644.
33. Kopple JD. Uses and limitations of the balance technique. JPEN.
1987;11(suppl 5):S79–S85.
34. Nelson ME, Fiatarone MA, Layne JE, et al. Analysis of body-compo-
sition techniques and models for detecting change in soft tissue with
strength training. Am J Clin Nutr. 1996;63:678–686.
35. Evans WJ, Campbell WW. Sarcopenia and age-related changes in
body composition and functional capacity. J Nutr. 1993;123:465–468.
36. Evans WJ. What is sarcopenia? J Gerontol Biol Sci Med Sci. 1995;
50A(special issue):5–8.
37. Dutta C, Hadley EC. The significance of sarcopenia in old age. J Ger-
ontol Biol Sci Med Sci. 1995;50A(special issue):1–4.
38. Baumgartner RN, Koehler KM, Romero L, Garry PJ. Serum albumin
is associated with skeletal muscle in elderly men and women. Am J
Clin Nutr. 1996;64:552–558.
39. Munro HN, McGandy RB, Hartz SC, Russell RM, Jacob RA, Otra-
dovec CL. Protein nutriture of a group of free-living elderly. Am J Clin
Nutr. 1987;46:586–592.
40. Campbell WW, Evans WJ. Protein requirements of elderly people.
Eur J Clin Nutr. 1996;50(suppl 1):S180–S185.
4
5
6
7
8
9
7
24).
. Bunker VW, Lawson MS, Stanfield MF, Clayton BE. Nitrogen bal-
ance studies in apparently healthy elderly people and those who are
housebound. Br J Nutr. 1987;57:211–221.
. Campbell WW, Crim MC, Dallal GE, Young VR, Evans WJ. In-
creased protein requirements in elderly people: new data and
retrospective reassessments. Am J Clin Nutr. 1994;60:501–509.
. Millward DJ, Roberts SB. Protein requirements of older individuals.
Nutr Res Rev. 1996;9:67–87.
1
1
0. Millward DJ, Fereday A, Gibson N, Pacy PJ. Aging, protein require-
ments, and protein turnover. Am J Clin Nutr. 1997;66:774–786.
1. Castaneda C, Charnley JM, Evans WJ, Crim MC. Elderly women ac-
commodate to a low-protein diet with losses of body cell mass, muscle
function, and immune response. Am J Clin Nutr. 1995;62:30–39.
2. Castaneda C, Dolnikowski GG, Dallal GE, Evans WJ, Crim M. Pro-
tein turnover and energy metabolism of elderly women fed a low-
protein diet. Am J Clin Nutr. 1995;62:40–48.
3. Campbell WW, Cyr-Campbell D, Weaver JA, Evans WJ. Energy re-
quirement for long-term body weight maintenance in older women.
Metabolism. 1997;46:884–889.
4. Harris JA, Benedict FG. A Biometric Study of Basal Metabolism in
Man. Washington, DC: Carnegie Institute of Washington; 1919.
5. Akers R, Buskirk ER. An underwater weighing system utilizing “force
cube” transducers. J Appl Physiol. 1969;26:649–652.
1
1
1
1
41. Fereday A, Gibson NR, Cox M, Pacy PJ, Millward DJ. Protein re-
quirements and ageing: metabolic demand and efficiency of
utilization. Br J Nutr. 1997;77:685–702.

M380
CAMPBELL ET AL.
4
4
4
2. Moreiras O, van Staveren WA, Amorim-Cruz JA, et al. Longitudinal
among [free-living elderly persons:] a 10-year longitudinal study. Nu-
trition. 1997;13:515–519.
changes in the intake of energy and macronutrients of elderly Europe-
ans. SENECA investigators. Eur J Clin Nutr. 1996;50(suppl 2):S67–S76.
3. Lesourd B, Decarli B, Dirren H. Longitudinal changes in iron and pro-
tein status of elderly Europeans. SENECA investigators. Eur J Clin
Nutr. 1996;50(suppl 2):S16–S24.
Received April 27, 2000
Accepted May 2, 2000
Decision Editor: John E. Morley, MB, BCh
4. Vellas BJ, Hunt WC, Romero LJ, Koehler KM, Baumgartner RN,
Garry PJ. Changes in nutritional status and patterns of morbidity