Original Research
Postprandial Thermogenesis Is Increased 100% on a
High-Protein, Low-Fat Diet versus a High-Carbohydrate,
Low-Fat Diet in Healthy, Young Women
Carol S. Johnston, PhD, FACN, Carol S. Day, MS, and Pamela D. Swan, PhD
Departments of Nutrition (C.S.J., C.S.D.) and Exercise Wellness (P.D.S.) Arizona State University East, Mesa, Arizona
Key words: high-protein diets, postprandial thermogenesis
Objective: The recent literature suggests that high-protein, low-fat diets promote a greater degree of weight
loss compared to high-carbohydrate, low-fat diets, but the mechanism of this enhanced weight loss is unclear.
This study compared the acute, energy-cost of meal-induced thermogenesis on a high-protein, low-fat diet versus
a high-carbohydrate, low-fat diet.
Methods: Ten healthy, normal weight, non-smoking female volunteers aged 19-22 years were recruited from
a campus population. Using a randomized, cross-over design, subjects consumed the high-protein and the
high-carbohydrate diets for one day each, and testing was separated by a 28- or 56-day interval. Control diets
were consumed for two days prior to each test day. On test day, the resting energy expenditure, the non-protein
respiratory quotient and body temperature were measured following a 10-hour fast and at 2.5-hour post breakfast,
lunch and dinner. Fasting blood samples were collected test day and the next morning, and complete 24-hour
urine samples were collected the day of testing.
Results: Postprandial thermogenesis at 2.5 hours post-meal averaged about twofold higher on the high
protein diet versus the high carbohydrate diet, and differences were significant after the breakfast and the dinner
meals (p ⬍ 0.05). Body temperature was slightly higher on the high protein diet (p ⫽ 0.08 after the dinner meal).
Changes in the respiratory quotient post-meals did not differ by diet, and there was no difference in 24-hour
glomerular filtration rates by diet. Nitrogen balance was significantly greater on the high-protein diet compared
to the high-carbohydrate diet (7.6 ⫾ 0.9 and ⫺0.4 ⫾ 0.5 gN/day, p ⬍ 0.05), and at 24-hour post-intervention,
fasting plasma urea nitrogen concentrations were raised on the high protein diet versus the high-carbohydrate diet
(13.9 ⫾ 0.9 and 11.2 ⫾ 1.0 mg/dL respectively, p ⬍ 0.05).
Conclusions: These data indicate an added energy-cost associated with high-protein, low-fat diets and may
help explain the efficacy of such diets for weight loss.
⬎10 kg, whereas only 9% of the HC subjects attained this
degree of weight loss. In obese, hyperinsulinemic subjects,
hypoenergetic (80% of calculated energy expenditure) HP diets
(45% energy from protein, 30% energy from fat) induced an
even more pronounced weight loss compared to an isocaloric,
HC diet (58% energy from carbohydrate, 30% energy from fat),
8.3 kg vs. 6.0 kg after four weeks [3]. The frequency of a
weight loss ⬎7 kg was 71% in the HP subjects vs. 16% in HC
subjects. Fasting serum insulin, triglycerides and total cholesterol concentrations decreased significantly on both diets but
reductions were 20% to 50% greater on the HP diet compared
to the HC diet [3].
The mechanism of enhanced weight loss on HP vs. HC diets
Counter to the current U.S. Dietary Guidelines which promote diets high in complex carbohydrates (58% of total daily
energy) [1], recent clinical investigations support the efficacy
of high-protein, reduced fat diets for weight loss, as well as for
improved insulin sensitivity and blood lipid profiles. In a randomized trial, 65 healthy, overweight and obese subjects consumed ad libitum a high-protein (HP), reduced fat diet (24%
energy from protein, 29% energy from fat) or a high-carbohydrate (HC), reduced fat diet (59% energy from carbohydrate,
29% energy from fat) for six months [2]. Subjects consuming
the HP diet lost more weight compared to subjects consuming
the HC diet (7.5 kg vs. 5.0 kg after three months and 8.7 kg vs.
5.0 kg after six months). Moreover, 35% of the HP subjects lost
Address reprint requests to: Carol Johnston, PhD, Department of Nutrition, Arizona State University East, 7001 E. Williams Field Rd, Mesa, AZ 85212. E-mail:
[email protected]
Presented in abstract form at Experimental Biology, April 2001, Orlando, FL.
Journal of the American College of Nutrition, Vol. 21, No. 1, 55–61 (2002)
Published by the American College of Nutrition
55
Metabolic Cost of High-Protein, Low-Fat Diets
may be attributed to a reduced energy intake, a lesser reduction
in resting energy expenditure (REE) and/or greater food-derived thermogenesis. Subjects consuming an ad libitum low fat
HP diet for six months averaged 450 kcal/day less than control
subjects ingesting a low fat HC diet ad libitum [2]. Under
experimental conditions, subjects consumed less energy when
given HP meals vs. HC meals [4]; moreover, they consumed
about 20% less energy at the subsequent meal [5,6]. Various
physiologic consequences of protein ingestion likely impact
satiety; in particular, proteins, unlike fats, starches or glucose,
are potent stimulators of cholecystokinin, the major gastrointestinal hormone inducing satiety [7].
Modest energy restriction significantly lowers REE, but
relative to comparable HC diets, hypoenergetic HP diets appear
to spare REE. In obese subjects adhering to a reduced-energy
HC diet for four weeks (diets provided 80% of maintenance
energy), REE fell 17% (⫺380 ⫾ 80 kcal/day); whereas, in
subjects adhering to a reduced-energy HP diet, REE fell only
6% (⫺130 ⫾ 50 kcal/day)[3]. The difference in REE by diet
composition appears as early as six days after the initiation
of the energy restriction and may relate to improved nitrogen
balance, hence reduced losses in muscle mass, on the HP
diet [8].
The thermic response to protein ingestion is 50% to 100%
higher than that for carbohydrate [9,10], an effect generally
attributed to the metabolic costs of peptide-bond synthesis,
ureogenesis and gluconeogenesis. Using 24-hour energy expenditure data, Westerterp et al. [11] estimated that the difference
in diet-induced thermogenesis between a combined HP/HC diet
vs. a high-fat, low protein/carbohydrate diet amounted to an
extra 90 kcal over a 24-hour period. However, the metabolic
cost of a HP, low-fat diet relative to the currently recommended
HC, low-fat diet is not known.
Using commonly consumed foods and meal plans, we designed a low-fat, HP (30% total energy) diet based on the
popular HP diet book The Zone [12] and compared postprandial
thermogenesis on this diet versus a HC, low-fat diet in young,
healthy, normal weight women. To characterize the daylong
effect of diet composition on thermogenesis, postprandial REE
was measured at 2.5 hours after the breakfast, lunch and dinner
meals. Identifying the magnitude of the extra thermic effect of
HP diets may help explain the demonstrated success of these
diets for weight loss.
measures were in normal ranges (Table 1). Participants gave
informed consent, and the study was conducted in accordance
with the guidelines of the Human Subjects Institutional Review
Board at Arizona State University.
Subjects consumed each of the two experimental diets for
one day in a randomized crossover manner. Testing was separated by a 28 or 56 day interval to control for possible
confounding effects of menstrual cycle on energy expenditure.
To assure similar baseline measures, on the two days prior to all
testing, subjects consumed the HC (control) diet providing 50%
of energy as complex carbohydrate, 10% as simple sugar, 15%
as protein and 25% as fat. This diet was formulated based on
the U.S. Dietary Guidelines and the U.S.D.A. Food Guide
Pyramid. Control diet meals were prepared using scales and
liquid measures at the test site, packaged and taken home by
subjects. The energy content of the diets were determined for
individual subjects using the Harris-Benedict equation to estimate basal metabolic rate (BMR), and total energy was calculated as BMR ⫻ 1.3, an energy intake appropriate for weight
maintenance in lightly active adults. Subjects were instructed to
consume all the foods provided, and only these foods, on these
two days.
The third day, subjects reported to the test site at 0600 hours
in a rested fasted state (no food or drink other than water after
2000 hours) and weight and height were recorded. Body composition was determined in the fasting state using whole body
air displacement plethysmography (Bod Pod® Body Composition Systems, Life Measurement Instruments, Concord, CA).
The mean of two body composition assessments was used to
calculate percent body fat using the Lohman equation [13].
Metabolic measurements were recorded using a respiratory
mask and two-way, non-rebreathing valve (Hans-Rudolph,
Inc., Kansas City, MO) interfaced with a MAX-1 metabolic
cart (Physiodyne Instrument Corporation, Quoque, NY). Upon
arrival at the laboratory, subjects were positioned in a reclining
chair and habituated to the open circuit spirometry metabolic
analysis apparatus for 30 minutes in a temperature controlled
(25–27°C) quiet room. The respiratory mask was placed over
the subject’s face and carefully checked and sealed to prevent
air leakage. Subjects were instructed to remain awake and not
to move, fidget or talk once the mask was in place. Body
temperature was recorded during this interval. Following the
30-minute habituation period, resting REE was estimated from
METHODS
Table 1. Physical Characteristics of the Female Subjects
(n ⫽ 10)
Subjects and Experimental Design
Ten healthy female volunteers were recruited from a campus population. None had renal or hepatic disease, diabetes
mellitus, heart disease, hypertension or took prescription medications. All were nonsmokers and had regular menstrual cycles. Ages ranged from 19 to 22 years, and body composition
56
Age (years)
Height (m)
Weight (kg)
BMI1
Body fat (%)
1
Mean ⫾ SEM
Range
19.0 ⫾ 0.4
1.66 ⫾ 0.01
64.4 ⫾ 2.4
23.4 ⫾ 0.9
25.2 ⫾ 2.3
18–22
1.59–1.74
55.0–77.3
18.6–28.4
13.9–36.1
Body mass index, kg/m2.
VOL. 21, NO. 1
Metabolic Cost of High-Protein, Low-Fat Diets
a mean of 20 minutes of continuous gas sampling via indirect
calorimetry using the Weir formula [14]. The coefficient of
variation for this procedure was 3.08%, and the between and
within day correlations were 0.70 and 0.90, respectively. The
non-protein respiratory quotient (RQ) was calculated (VCO2/
VO2) to estimate fuel utilization [15]. Gas analyzers were
calibrated before and after each test by nitrogen and two
primary standard gases accurate to 0.01%. The pneumotachometer was calibrated using a 3L syringe to deliver fixed volumes
at variable flow rates.
Following testing, a baseline fasting blood sample was
collected, and subjects then ingested the breakfast meal within
a 15 minute period under observation. At two hours post-meal,
subjects were again positioned in the reclining chair and habituated to the metabolic cart for 30 minutes, and energy expenditure was measured for 20 minutes as described above. The
lunch meal was provided four hours after the breakfast meal
and consumed within a 15-minute interval. Energy expenditure
was again determined at 2.5 hours post-meal. Similarly, the
dinner meal was provided four hours after the lunch meal, and
energy expenditure determined at 2.5 hours post-meal. Subjects
remained at the test site until all measurements were completed; their activity was restricted to reading or watching TV
while sitting quietly. Subjects returned to the test site the
following morning in the rested, fasted state and blood collected. On the day of testing, subjects provided complete 24hout urine collections, defined as all urine excreted following
the first morning void through the initial next morning void.
Two experimental diets were tested: the control HC diet
described above and the HP diet (30% of energy as complex
carbohydrate, 10% as simple sugar, 30% as protein and 30% as
fat). Diets were devised using the Food Processor® for Windows Nutrition Analysis Software (Version 6.11, Esha Research, Salem, OR), and menus are shown in Table 2. Only
common foods and food combinations were used, and the diet
plans reflected typical American meal patterns. In the HP diet,
egg whites, cottage cheese, turkey, and tuna were substituted
for grains. The macronutrient compositions of the test meals are
shown in Table 3. Test meals were prepared using scales and
liquid measures at the test site immediately prior to their
consumption. Water and an artificially sweetened beverage
(Crystal Light®) were allowed freely during the feeding period.
Subjects were provided a small evening snack to consume after
testing was completed but prior to 2000 hours. Subjects consumed self-selected diets for the 25 days between feeding
periods.
Table 2. Experimental Diets1
Meal
Breakfast
Lunch
High-Carbohydrate
Diet
Soy milk (133 ml)
Total cereal (40)
Pink grapefruit (123)
English muffin (66)
Butter (10)
Turkey slices (57)
Swiss cheese (21)
Looseleaf lettuce (10)
Whole wheat bread
(56)
Tomatoe slices (20)
Mayonnaise (20)
Pretzels (57)
Apple (136)
Diet caffeine-free cola
(400 mL)
Dinner
Tomato sauce (123)
English muffin (52)
Mozzarella cheese
(28)
Mild cheddar cheese
(28)
Green bell peppers
(75)
Mushrooms (69)
Onions (40)
Broccoli (88)
Ranch dressing (35)
Pretzels (28)
Soy milk (200 mL)
High-Protein Diet
Egg substitute (166)
Nonfat cheese (28)
Whole wheat bread
(50)
Peanut butter (5)
2% milk (133 mL)
Cottage cheese (68)
Whole wheat bread
(56)
Mayonnaise (9)
Turkey (85)
Iceberg lettuce (30)
Egg white (67)
Cheese (28)
Carrots (50)
Pretzels (28)
Raisins (8)
Diet caffeine free cola
(400 mL)
Celery (60)
Fat-free cream cheese
(29)
Tuna (172)
Mayonnaise (16)
Salad greens (83)
Roll (56)
Butter (10)
Animal crackers (43)
Diet caffeine free cola
(400 mL)
1
The small evening snack consumed after testing was completed is not included
in these diets; portion size (g) in parentheses unless otherwise noted.
(ICN Pharmaceuticals, Costa Mesa, CA). Plasma and urine
urea nitrogen and plasma and urinary creatinine were measured
colorimetrically (Procedures #555 and #640, Sigma-Aldrich,
St. Louis, MO). Apparent nitrogen balance was calculated as
the difference between total dietary nitrogen and total urine
nitrogen (urine urea nitrogen ⫻ 1.25) with a correction of 1.8 g
nitrogen loss daily to account for obligatory and fecal losses
[16]. Glomerular filtration rates (GFR) were calculated as
[urine creatinine (mg/dL) ⫻ urine volume (mL)]/[plasma creatinine (mg/dL) ⫻ min].
Blood and Urine Measures
Blood samples collected in EDTA-treated Vacutainer®
tubes (Becton Dickinson and Co., Franklin Lakes, NJ) were
immediately centrifuged for 15 minutes at 1000 ⫻ g. Aliquots
of plasma were stored separately at ⫺45° until analyses.
Plasma insulin was measured by immunoradiometric assay
JOURNAL OF THE AMERICAN COLLEGE OF NUTRITION
Statistical Analyses
Data are reported as the mean ⫾ SEM. Differences between
blood and urine indices were analyzed using one-tailed, paired
t tests. Differences between mean post-meal values for thermogenesis, non-protein RQ and body temperatures were evaluated
57
Metabolic Cost of High-Protein, Low-Fat Diets
Table 3. Composition of High-Carbohydrate and High-Protein Diets
High-Carbohydrate Diet
Energy (kcals)
Protein (g)
Carbohydrates (g)
Fat (g)
Fiber (g)
High-Protein Diet
Breakfast
Lunch
Dinner
Total
Breakfast
Lunch
Dinner
Total
455
14
76
12
7
643
27
96
20
10
668
34
92
19
9
1766
75
264
51
26
422
37
38
13
4
623
45
71
20
7
709
54
68
24
5
1760
136
177
57
16
using multiple analyses of variance for repeated measures and
appropriate post-hoc tests. Relationships between variables
were assessed by the Pearson correlation measure. The level of
significance was set at p ⬍ 0.05. The Statistical Package for the
Social Sciences (SPSS Base 7.5 for Windows, Chicago, IL)
was used for all statistical calculations.
RESULTS
Body weights did not differ on the mornings of the HP and
HC test days. Fasting REE was similar prior to the diet intervention, 1396 ⫾ 53 and 1359 ⫾ 59 kcals/24 hours for HC and
HP, respectively. Postprandial REE was 8 kcals/hour higher at
2.5 hours following the breakfast meal (p ⬍ 0.05) and 8
kcals/hour higher at 2.5 hours following the lunch meal on the
HP diet versus the HC diet (Fig. 1a). At 2.5 hours after the
dinner meal, postprandial REE was 14 kcals/hour higher on the
HP diet compared to the HC diet (p ⬍ 0.05).
Body temperature rose steadily throughout the day on both
diets (Fig. 1b). The change in body temperature from the
fasting baseline value was ⫹0.1 and ⫹0.4 °F at 2.5 hours after
the breakfast meal and ⫹0.4 and ⫹0.6° F at 2.5 hours after the
lunch meal for the HC and HP diets respectively (p ⬎ 0.05). At
2.5 hours after the dinner meal, the change in body temperature
for the HP diet was nearly 40% higher than that for the HC diet
(⫹0.8 and ⫹0.5 °F respectively, p ⫽ 0.08). Body temperature
was related to REE in the HP group only (HP: r ⫽ 0.66, p ⬍
0.05; HC: r ⫽ 0.29, p ⬎ 0.05).
The fasting non-protein RQ, an indicator of substrate oxidation, was similar prior to diet intervention (0.81 ⫾ 0.01 and
0.79 ⫾ 0.02 for HC and HP, respectively), and the change in
post-meal non-protein RQ did not differ by diet (Fig. 1c).
Fasting plasma insulin concentrations before and after the diet
intervention did not vary by diet (20.8 ⫾ 2.7 and 19.8 ⫾ 2.2
␮U/mL prior to and 22.1 ⫾ 2.7 and 22.7 ⫾ 3.7 ␮U/mL at 24
hours post-intervention, HC and HP respectively).
Fasting plasma urea nitrogen concentrations were similar
prior to diet intervention, 11.2 ⫾ 0.8 and 11.0 ⫾ 0.8 mg/dL for
HC and HP, respectively. At 24 hours post-intervention, fasting
plasma urea nitrogen concentrations were raised (p ⬍ 0.05) on
the HP diet versus the HC diet, 13.9 ⫾ 0.9 and 11.2 ⫾ 1.0
mg/dL, respectively (Table 4). GFR did not vary by diet treatment (131.6 ⫾ 15.2 and 129.9 ⫾ 19.5 mL/min, HC and HP,
58
Fig. 1. (a) Postprandial thermogenesis (calculated as the difference
between post-meal REE and the baseline, fasting REE), (b) postprandial change in body temperature, and (c) postprandial change in respiratory quotient in young, healthy women after ingestion of high-protein
meals vs. high-carbohydrate meals. Values are reported as the mean ⫾
SEM; * denotes significant difference between diets at the meal indicated (p ⬍ 0.05); **denotes a trend for difference between diets at the
meal indicated (p ⫽ 0.082).
VOL. 21, NO. 1
Metabolic Cost of High-Protein, Low-Fat Diets
Table 4. Nitrogen Balance Measurements and Urine Characteristics of Young Women Fed a High-Carbohydrate (60% Total
Energy), Low-Fat Diet or a High-Protein (30% Total Energy), Low-Fat Diet for One Day1
High-Carbohydrate Diet
High-Protein Diet
12.0
11.2 ⫾ 1.0
8.5 ⫾ 0.4
⫺0.4 ⫾ 0.5
2299 ⫾ 240
1085.9 ⫾ 88.9
0.52 ⫾ 0.04
131.6 ⫾ 15.2
21.8
13.9 ⫾ 0.9*
9.9 ⫾ 0.7
7.6 ⫾ 0.9*
2658 ⫾ 306
1049.7 ⫾ 70.2
0.50 ⫾ 0.07
129.9 ⫾ 19.5
N intake (g)
Plasma urea N (mg/dL)
Urine urea N (g)
Apparent N balance2 (gN/day)
Urine (mL)
Urine creatinine (mg/day)
Plasma creatinine (mg/dL)
Creatinine clearance3
Mean ⫾ SEM, n ⫽ 10; asterisk denotes significant difference, p ⬍ 0.05.
Corrections of 1.8 g N/day were made for obligatory and fecal losses [16], and urine urea nitrogen was extrapolated to total urine nitrogen (⫻ 1.25).
3
Calculated as [urine creatinine (mg/dL) ⫻ urine volume (ml)]/[plasma creatinine (mg/dL) ⫻ min].
1
2
respectively, Table 4), and urinary urea nitrogen values postintervention were also similar between groups (Table 4). Apparent
nitrogen balance was greater for the HP diet than the HC diet,
⫹7.6 ⫾ 0.9 and ⫺0.4 ⫾ 0.5 gN/day, respectively (Table 4).
DISCUSSION
These data demonstrate that meal-induced thermogenesis at
2.5 hours post-meal averages about twofold higher on a HP,
low fat diet versus a HC, low-fat diet. Generally, postprandial
thermogenesis has been associated with the protein content of
a meal [9,10,17–19], and our data confirm this relationship.
However, the difference in the energy cost of HP versus HC
diets, particularly in the context of weight loss promotion, has
not been addressed by healthcare professionals. Increased dietinduced thermogenesis, in association with the preservation of
REE [3,8], may contribute to the reported weight loss success
of diets high in protein with moderate levels of carbohydrate
and lends credence to the observation that weight loss on HP
diets is predominately body fat, not body water [2,3].
Following protein consumption, postprandial REE rises rapidly and is sustained for as long as four to five hours post meal
[18,20,21]. Carbohydrate consumption, however, induces a
more modest rise in REE relative to protein, and REE falls
rapidly one to two hours post-meal [18,21]. In the present
report, we measured REE at 2.5 hours post-meal to examine the
magnitude of the difference in postprandial thermogenesis as a
function of diet. Over the course of the day, postprandial
thermogenesis on the HP diet totaled 30 more kcals at 2.5 hours
post-meals. If this same energy differential was maintained for
two-to-three-hour intervals post-meals, the added postprandial
thermogenesis associated with the HP diet may have been as
high as 90 kcals. This is the first study to examine the difference in diet-induced energy expenditure in subjects consuming
HC versus HP diets which contain common foods offered in
typical meal plans.
Westerterp et al. [11] estimated that the difference in dietinduced thermogenesis for a combined HP/HC diet versus a
JOURNAL OF THE AMERICAN COLLEGE OF NUTRITION
high-fat diet was 90 kcal. Crovetti et al. [21] reported that
during the seven-hour period post-meal consumption, postprandial energy expenditure was 40 kcals greater for a HP meal
versus a HC meal (195 g bresaola versus 120 g pasta with 80 g
tomato sauce). Using liquid formulas, Robinson et al. [22]
reported that the thermic response to HP feeding (35 g protein,
15 g carbohydrate, 6 g fat each hour for nine hours) was 60
kcals higher than that for HC feeding (15 g protein, 35 g
carbohydrate, 6 g fat each hour for nine hours). It should be
noted that food palatability enhances meal thermogenesis [23].
The ingestion of a palatable meal (parmesan fondue, spaghetti
with meatballs and chocolate éclair) increased diet-induced
thermogenesis nearly 50% more than if the same meal ingredients were blended, desiccated and consumed as a tasteless
biscuit.
The absence of a thermic response to the HC diet at 2.5
hours after the breakfast meal seems contradicted, but this
phenomenon has been noted by others [18]. The metabolic cost
of glycogen synthesis and lipogenesis is believed to account for
55% to 65% of the thermic effect of carbohydrate ingestion
[22], but only for 10% to 30% of the thermic effect of protein
ingestion. The breakfast meal was consumed after an overnight,
12-hour fast; hence, it is conceivable that the ingested glucose
was readily utilized by body tissues and not available for
glycogen synthesis. At the lunch and dinner meals, nutrient
storage would be enhanced as nutrient availability increased
with repeated food ingestion.
Concomitant with the thermic response to the test diets was
a slight rise in body temperatures. Although the changes in
body temperature were not significantly different by diet, HP
feeding was associated with a greater degree of body temperature change versus HC feeding, and at the dinner meal this
change was nearly significant (⫹0.8 °F, p ⫽ 0.082). Furthermore, body temperature was related to REE over the 12-hour
testing period for the HP diet only (r ⫽ 0.66, p ⬍ 0.05),
supporting the contention of Brundin and Wahren [24] that
protein ingestion elicits a pyrogen-like effect.
Postabsorptive protein synthesis increases 10% to 25% for
high vs. low protein meals [16,25], and the metabolic cost of
59
Metabolic Cost of High-Protein, Low-Fat Diets
this enhanced protein synthesis likely accounts for the added
thermic effect of dietary protein [22, 26]. Also, there is a net
retention of amino acids in the body when dietary protein is
increased, and nitrogen balance is positive, as high as ⫹7 to 13
gN/day [27,28], but more typically ⫹3 gN/day, even after the
HP diet is consumed for an extended period [29]. The high
positive nitrogen balance noted in the present trial likely represents a transient retention of nitrogen, either as urea or free
amino acids such as glutamine in muscle tissue. Pannemans et
al. [16] fed ten young women (27 ⫾ 4 years) a HP diet (21%
total energy as protein) for two weeks at which time nitrogen
balance was positive, ⫹2.1 gN/day.
The metabolic consequences of HP diets are controversial,
but most experts agree that protein intakes should not exceed
2 g 䡠 kg⫺1 䡠 day⫺1, the level utilized in the present report
[30,31]. Americans typically consume 15% of dietary energy as
protein, corresponding to about 1 g 䡠 kg⫺1 䡠 day⫺1 [32]. An
often cited, adverse effect of diets high in protein is a potential
effect on renal function, and individuals with impaired kidney
function are advised to reduce levels of dietary protein. Experimental data, however, indicate that GFR varies little when
dietary protein ranges from 10% to 30% of total energy [33–
35]. The data presented here further demonstrated that an acute
change in dietary protein, 15% to 30% of dietary energy, had
little effect on renal function in healthy individuals. Furthermore, both plasma urea nitrogen and urine urea nitrogen concentrations remained within normal ranges following the HP
diet intervention.
The popularity of HP diets for weight loss is unquestionable. Although this research did not assess weight loss or the
long-term effects of a HP diet, results indicated that the increased thermogenesis of a HP diet may contribute to its
efficacy. The recent literature suggests that diets high in protein, but with a moderate carbohydrate and low fat content, do
promote a greater degree of weight loss compared to the
currently recommended high-carbohydrate, low-fat diets.
When considering other health issues, HP diets should be low
in saturated fat and rely on low-fat milk products, egg whites,
poultry and fish as protein sources. Changes in postprandial
thermogenesis induced by HP diets based on non-animal products versus HC diets awaits investigation.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
ACKNOWLEDGMENTS
This research was funded by the Lloyd S. Hubbard Nutrition
Research Fund of the Arizona State University Foundation at
Arizona State University.
17.
18.
REFERENCES
19.
1. US Dietary Guideline Committee, US Department of Agriculture.
US Department of Health and Human Services. Nutrition and Your
60
Health: Dietary Guidelines for Americans. 4th ed. Home & Garden
Bulletin 232, 1995.
Skov AR, Toubro S, Ronn B, Holm L, Astrup A: Randomized trial
on protein vs carbohydrate in ad libitum fat reduced diet for the
treatment of obesity. Int J Obesity 23:528–536, 1999
Hwalla Baba N, Sawaya S, Norbay N, Habbal Z, Azar S, Hashim
SA: High protein vs. high carbohydrate hypoenergetic diet for the
treatment of obese hyperinsulinemic subjects. Int J Obesity 1999;
23:1202–1206.
Araya H, Hills J, Alvina M, Vera G: Short-term satiety in preschool children: a comparison between high protein meal and a
high complex carbohydrate meal. Int J Food Sci Nutr 51:119–124,
2000.
Barkeling B, Rossner S, Bjorvell H: Effects of a high-protein meal
(meat) and a high-carbohydrate meal (vegetarian) on satiety measured by automated computerized monitoring of subsequent food
intake, motivation to eat and food preferences. Int J Obes 14:743–
751, 1990.
Latner JD, Schwartz M: The effects of a high-carbohydrate, highprotein or balanced lunch upon later food intake and hunger
ratings. Appetite 33:119–128, 1999.
Liddle RA, Green GM, Conrad CK, Williams JA: Proteins but not
amino acids, carbohydrates, or fats stimulate cholecystokinin secretion in the rat. Am J Physiol 251:G243–248, 1986.
Agus MSD, Swain JF, Larson CL, Eckert EA, Ludwig DS: Dietary
composition and physiologic adaptations to energy restriction.
Am J Clin Nutr 71:901–907, 2000.
Zed C, James WP: Dietary thermogenesis in obesity. Response to
carbohydrate and protein meals: the effect of beta-adrenergic
blockade and semistarvation. Int J Obes 10:391–405, 1986.
Schutz Y, Bray G, Margen S: Postprandial thermogenesis at rest
and during exercise in elderly men ingesting two levels of protein.
J Am Coll Nutr 6:497–506, 1987.
Westerterp KR, Wilson SAJ, Rolland V: Diet induced thermogenesis measured over 24 h in a respiration chamber: effect of diet
composition. Int J Obes 23:287–292, 1999.
Sears B: “The Zone.” New York: Harper Collins, 1995.
Heyward VH, Stolrezyk LM: “Applied Body Composition Assessment.” Champaign, IL: Human Kinetics, 1996.
Weir JB: New methods for calculating metabolic rate with special
reference to protein. J Physiol 109:1–49, 1949.
Brooks G, Fahey T, White T: “Exercise Physiology: Human
Bioenergetics and Its Applications,” 2nd ed. Mountain View, CA:
Mayfield Publishing Co, pp 44–45, 1996.
Pannemans DLE, Halliday D, Westerterp KR, Kester ADM: Effect
of variable protein intake on whole-body protein turnover in young
men and women. Am J Clin Nutr 61:69–74, 1995.
Dauncey MJ, Bingham SA: Dependence of 24 h energy expenditure in man on the composition of the nutrient intake. Br J Nutr
50:1–13, 1983.
Karst H, Steiniger J, Noack R, Steglich HD: Diet-induced thermogenesis in man: thermic effects of single proteins, carbohydrates
and fats depending on their energy amount. Ann Nutr Metab
28:245–252, 1984.
Jequier E: Thermic responses induced by nutrients in man: their
importance in energy balance regulation. Experientia 44:26–44,
1983.
VOL. 21, NO. 1
Metabolic Cost of High-Protein, Low-Fat Diets
20. Soucy J, LeBlanc J: Protein meals and postprandial thermogenesis.
Physiol Behav 65:705–709, 1999.
21. Crovetti R, Santangelo A, Testolin G: The influence of thermic
effect of food on satiety. Eur J Clin Nutr 52:482–488, 1997.
22. Robinson SM, Jaccard C, Persaud C, Jackson AA, Jequier E,
Schutz Y: Protein turnover and thermogenesis in response to
high-protein and high-carbohydrate feeding in men. Am J Clin
Nutr 52:72–80, 1990.
23. LeBlanc J, Brondel L: Role of palatability on meal-induced thermogenesis in human subjects. Am J Physiol 248:E333–E336,
1985.
24. Brundin T, Wahren J: Influence of protein ingestion on human
splanchnic and whole-body oxygen consumption, blood flow, and
blood temperature. Metabolism 43:626–632, 1994.
25. Pannemans DLE, Wagenmakers AJM, Westerterp KR, Schaafsma
G, Halliday D: Effect of protein source and quantity on protein
metabolism in elderly women. Am J Clin Nutr 68:1228– 1235,
1998.
26. Coulson RA, Hernandez T: Increase in metabolic rate of the
alligator fed proteins or amino acids. J Nutr 109:538–550, 1979.
27. Poortmans JR, Dellalieux O: Do regular high protein diets have
potential health risks on kidney function in athletes? Int J Sport
Nutr Exerc Metab 10:28–38, 2000.
28. Tarnopolsky MA, MacDougall JD, Atkinson SA: Influences of
protein intake and training status on nitrogen balance and lean
body mass. J Appl Physiol 64:187–193, 1988.
29. Garlick PJ, McNurlan MA, Patlak CS: Adaptation of protein
JOURNAL OF THE AMERICAN COLLEGE OF NUTRITION
30.
31.
32.
33.
34.
35.
metabolism in relation to limits to high dietary protein intake. Eur
J Clin Nutr 53(Suppl 1):S34–S43, 1999.
Metges CC, Barth CA: Metabolic consequences of a high dietaryprotein intake in adulthood: assessment of the available evidence.
J Nutr 130:886–889, 2000.
Durnin JV, Garlick P, Jackson AA, Schurch B, Shetty PS, Waterlow: Report of the IDECG Working Group on lower limits of
energy and protein and upper limits of protein intakes. International Dietary Energy Consultative Group. Eur J Clin Nutr
53(Suppl 1):S174–S176, 1999.
Life Sciences Research Office, Federation of American Societies
for Experimental Biology: “Third Report on Nutrition Monitoring
in the United States,” vol 1. Washington DC: United States Government Printing Office, pp 148–149, 1995.
Kerstetter JE, Caseria DM, Mitnick ME, Ellison AF, Gay LF,
Liskov TA, Carpenter TO, Insogna KL: Increased circulating concentrations of parathyroid hormone in healthy, young women
consuming a protein-restricted diet. Am J Clin Nutr 66:1188–1196,
1997.
Skov AR, Toubro S, Bulow J, Krabbe K, Parving HH, Astrup A:
Changes in renal function during weight loss induced by high vs.
low-protein low-fat diets in overweight subjects. Int J Obes 23:
1170–1177, 1999.
Brändle E, Sieberth HG, Hautmann RE: Effect of chronic dietary
protein intake on the renal function in healthy subjects. Eur J Clin
Nutr 50:734–740, 1996.
Received March 30, 2001; accepted October 25, 2001.
61