European Journal of Clinical Nutrition (2001) 55, 10±18
ß 2001 Macmillan Publishers Ltd All rights reserved 0954±3007/01 $15.00
www.nature.com/ejcn
Dose-response relationship between fat ingestion and oxidation:
quantitative estimation using whole-body calorimetry and 13C
isotope ratio mass spectrometry
BJ Sonko1*{, AM Prentice2, WA Coward3, PR Murgatroyd4 and GR Goldberg5
1
University of Colorado Health Sciences Center, Department of Pediatrics, Denver, Colorado, USA; 2MRC International Nutrition Group,
Public Health Nutrition Unit, London School of Hygiene and Tropical Medicine, London, UK; 3MRC Human Nutrition Unit, Cambridge,
UK; 4University of Cambridge, Department of Clinical Biochemistry, Addenbrooke's Hospital, Cambridge, UK; and 5British Nutrition
Foundation, London, UK
Objective: To determine dose-dependent relationship between ingested fat and its oxidation in the immediate
post-prandial period in humans.
Design: Subjects were randomly selected for the study at the Dunn Clinical Nutrition Centre, Cambridge, UK.
Subjects ingested naturally enriched 13C corn-oil doses (range 20 ± 140 g) in a whole-body indirect calorimeter,
and were studied for 8 h. Ingested fat oxidation was estimated from the subject's breath 13C enrichment and total
carbon dioxide production. Total fat and carbohydrate oxidation were estimated from non-protein oxygen and
carbon dioxide exchanges. Endogenous fat oxidation was estimated as the difference between total fat and ingested
fat oxidation.
Results: The amount of fat dose oxidized was nonlinearly related to the amount ingested. On average, 25.6 2.7%
of the mean fat dose was oxidized. A signi®cant (r ˆ 7 0.72, P < 0.001) inverse correlation was found between
the amount of fat dose and the proportion oxidized. Endogenous carbohydrate oxidation was negatively and
signi®cantly correlated to fat dose oxidized (r ˆ 7 0.61, P < 0.01), but it was not correlated to endogenous fat
oxidation.
Conclusions: There was a nonlinear relationship between amount of fat dose and its quantity that was oxidized in
the immediate post-prandial period. The inverse relationship between the size of the fat load and the proportion
that was oxidized post-prandially implies increased dietary fat storage beyond about 50 g in a normal resting adult.
This has important implications for 13CO2-based studies.
Sponsorship: This study was funded by the Nestle Foundation, Switzerland and the British Medical Research
Council, UK.
Descriptors: fat; dose±response; oxidation; indirect calorimetry; carbon-13 mass spectrometry
European Journal of Clinical Nutrition (2001) 55, 10±18
*Correspondence: BJ Sonko, University of Colorado Health Sciences
Center, Department of Pediatrics, Campus Box C232, 4200 E. Ninth Ave,
Denver, CO 80262, USA.
E-mail: [email protected]
Guarantor: Dr BJ Sonko & Dr AM Prentice.
Contributors: Dr Sonko co-conceived the idea for the study and codeveloped it into a study project. He was involved in subject recruitment,
data collection and data analysis. He produced the ®rst draft and co-re®ned
it into the ®nal draft. Dr Prentice co-conceived the idea and co-developed
it into a study project. He was involved in data collection as well as data
analysis. He commented on the ®rst draft and also helped intellectually in
the re®nement of the ®nal draft. Dr Coward co-developed the idea into a
study project. He was involved in the mass spectrometry data analysis of
the study. He commented on the draft and was involved in the re®nement
of the ®nal draft. Dr Murgatroyd was involved in data collection,
particularly in the calorimeter component of the study. He commented on
the ®rst draft as well as being involved in the re®nement of the ®nal draft.
Dr Goldberg was involved in subject recruitment, data collection,
commented on the ®rst draft and was involved in the re®nement of the ®nal
draft. All the contributors participated intellectually in the development of
the paper.
{
Formerly of Dunn Nutrition Group, c=o MRC Laboratories, Fajara,
Banjul, The Gambia. Supported by the Nestle Foundation, Switzerland.
Received 16 December 1999; revised 29 August 2000; accepted
29 August 2000
Introduction
The continuing increase in the incidence of obesity in
Western industrialized societies in recent years, has
helped focus attention on the factors which predispose
individuals to develop this disorder (Bray & Popkin,
1998). Genetic predisposition, physical inactivity and the
composition of the habitual diet are some of the factors that
have been identi®ed (Prentice & Poppitt, 1996). Among
these the amount of dietary fat in the usual diet is considered by many to be particularly important in the development of obesity (Bray & Popkin, 1998). The case for
high-fat diet as a predisposing agent for obesity is strengthened by a number of factors which are characteristic of
high-fat diets. Firstly, high-fat diets have high energy
density and secondly, although the exact mechanism is
still debated, it has been suggested by some investigators
that consumption of high-fat diets leads to hyperphagia
(Lissner et al, 1987; Stubbs et al, 1995). A possible third
factor could be that dietary fat is mainly stored and not
Fat ingestion and oxidation
BJ Sonko et al
oxidized in the post-prandial period (Flatt et al, 1985;
Schutz et al, 1989).
However, there are some differences of opinion about
the metabolic fate of dietary fat. Grif®ths et al, (1994)
suggested that 10% of dietary fat in a mixed meal was
oxidized in the immediate post-prandial period. In the
literature, there is indirect evidence in support of this
®nding. For example, Hill et al (1991) and Westerterp
(1993) have shown that ingestion of high-fat meals result in
a signi®cant decrease in respiratory quotient (RQ) values,
which is indicative of increased fat oxidation (Lusk, 1906).
To determine whether the observed increases in fat oxidation in the studies quoted above were a result of ingested or
endogenous fat oxidation, both indirect calorimetry and 13C
tracer methodologies must be used. Some studies have
previously used 14C breath test methodology to determine
the extent of malabsorption of some fatty acids in humans
(Duncan et al, 1992; Benini et al, 1984; Newcomer et al,
1979). However, these studies did not address questions
regarding the contribution of dietary fat to total energy
expenditure nor its interaction with other endogenous
macronutrients in the post-prandial state.
The objectives of the present study were to use
both whole-body indirect calorimetry and 13C tracer
methodology to investigate: (a) whether there is a linear
dose±response relationship between ingested fat and its
subsequent in vivo oxidation in the short-term post-prandial
period; and (b) to examine the interaction between ingested
fat oxidation and the oxidation of both endogenous fat and
carbohydrate in humans.
Table 1 Quantities of vegetable ingredients used in preparing the 20, 35
and 50 g corn-oil test meals
Subjects
Nine volunteers (four females and ®ve males) took part in
18 study sessions conducted at the Dunn Clinical Nutrition
Centre, Cambridge, UK. The subjects were normal healthy
adults who were recruited through advertisement. Their
mean s.d. physical characteristics for age, weight, height
and body mass index (BMI) were: 29 11.1 y;
67.1 10.5 kg; 1.71 0.1 m, and 22.9 2.1 kg=m2, respectively. For subjects who participated in more than one
session, the respective study sessions were separated by
at least one week. Nine subjects started the study and none
dropped out.
Materials and methods
Corn-oil dosea
Vegetable ingredients
50 g
35 g
20 g
Aubergine (egg plant)
Tomato
Green pepper
Mushroom
Celery
90
160
62
90
77
63
112
43
63
54
36
64
25
36
31
11
a
Each corn-oil dose was prepared as a vegetarian meal (`ratatouille') using
the amounts of vegetable ingredients listed in its column.
Table 2 Quantity of fat dose ingested, exogenous fat and other components of substrate oxidation
Test no.a
Dose
given
(g)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
20
20
20
35
35
35
45
50
50
50
50
80
85
130
137
138
140
140
Mean
s.e.
70
10.9
Dose
oxidized b
(g)
Endogenous
fat oxidized
(g)
Endogenous
carbohydrate oxidized
(g)
Endogenous
protein oxidized
(g)
5.4
6.3
11.2
6.9
12.3
14.9
10.2
14.2
8.0
15.6
16.4
22.2
17.6
13.4
26.3
22.0
18.4
16.4
21.2
37.6
34.3
24.7
41.9
11.3
30.6
25.4
20.6
40.7
29.7
19.4
32.4
31.8
32.3
28.0
21.4
26.1
86.3
91.9
45.7
90.8
93.2
67.8
46.8
57.1
89.5
84.6
28.9
58.8
84.7
57.7
37.3
43.5
51.7
46.3
35.6
30.5
8.1
34.4
19.0
13.7
29.1
20.5
34.3
30.6
11.8
26.6
18.9
28.9
19.0
17.1
20.5
14.3
14.3
1.4
28.3
1.9
64.6
5.1
22.9
2.0
a
Some subjects participated in more than one study session.
Values have been corrected by applying the acetate correction factor.
b
European Journal of Clinical Nutrition
Fat ingestion and oxidation
BJ Sonko et al
12
Test meals
The test meals (fat doses) consisted of corn-oil fat. The
natural 13C in corn-oil served as the tracer. In addition, a
small quantity (150 mg 1-13C-palmitic acid=49 g corn-oil)
of arti®cially enriched (99%) 1-13C-palmitic acid (Isotec
Inc., Ohio, USA) was added to the meals. This increased
the 13C enrichment of the fat added from 715.6d to about
78.3d relative to PDB and was done to provide a large 13C
breath response than would otherwise have been obtained.
The corn-oil doses (g) ingested by the subjects ranged from
20 to 140 g (Table 2). The low dose (20±50 g) meals were
prepared as vegetable stew (Table 1) and the high dose (80±
140 g) meals were prepared as mayonnaise salad. The
vegetable ingredients used in preparing the meals had
only trace amounts of fat (0.2 g=100 g) and very little
protein and carbohydrate (1.0 g=100 g and 1.7 g=100 g
vegetable ingredients, respectively; Paul & Southgate,
1978). The natural 13C content of the vegetable ingredients
was comparable to the subjects' baseline breath 13C enrichment (Schmidt & Metges, 1986).
Protocol
On the day prior to the study, the subjects reported to the
Dunn Clinical Nutrition Centre between 09:00 and 09:30
and were immediately put on a 24 h maintenance diet. One
of the main purposes of this diet was to ensure that the
background level of 13C in the subject's expired air was
stabilized prior to starting the study. The subject's maintenance diet was estimated to be equivalent to 1.4 times the
basal metabolic rate (BMR), which was estimated from
Scho®eld's equations (Scho®eld et al, 1985). The last
maintenance meal was eaten at 19:00 the night before the
study, and subjects entered the whole-body calorimeter at
20:00. The calorimeter used in this study has been
described elsewhere (Murgatroyd et al, 1993). In the
calorimeter, subjects went to bed between 22:30 and
23:00. At 07:00 and 09:00 on the study day, subjects
emptied the bladder and the urine collected was discarded.
The subjects' BMR was measured between 08:00 and
09:00. Four breath samples were collected shortly before
09:30, and these served as baseline breath. The test meal
was given at 09:30 and all was consumed within 30 min.
Subjects were not allowed any other foods during the study,
but water was allowed ad libitum. At 30 min intervals postmeal, duplicate breath samples were collected in vacutainers (Becton Dickson Inc., USA) until the end of the
protocol at 18:30. Urine samples were collected from
subjects at 13:30, 15:30 and at 18:30. At each void, total
urine volume was recorded and duplicate samples were
stored for later analysis of total urinary nitrogen.
Calorimetry techniques and precision
Oxygen consumption was measured by a Servomex OA184
paramagnetic analyser and carbon dioxide (CO2) production
European Journal of Clinical Nutrition
was measured by Servomex 121 infrared analyser (Servomex, Crowborough, UK). The consumption of oxygen and
CO2 production was estimated using the equation of Brown et
al (1984). This provided an instantaneous measure of changes
in the subject's rates of gas exchange. The net oxidation of fat,
F, and carbohydrate, C may be expressed in terms of oxygen
consumption, O2, carbon dioxide production, CO2, and nitrogen excretion, N, as:
F ˆ ‰CO2 ÿ Rc O2 ‡ …Rc ÿ Rp †Vp kNŠ=Vf …Rf ÿ Rc †
and
C ˆ ‰CO2 ÿ Rf O2 ‡ …Rf ÿ Rp †Vp kNŠ=Vc …Rc ÿ Rf †;
where Rf, Rc and Rp are the ratios of CO2 production to O2
consumption and Vf, Vc and Vp are the volumes of O2
consumed per gram oxidized for fat, carbohydrate and
protein, respectively. The constant, k relates protein oxidation to nitrogen excretion, with the protein oxidation rate
being equal to kN. The precision (s) values of 30 min O2
consumption and CO2 production measurements are typically 3.8 and 2.4 ml min71, respectively. The major potential sources of systematic error in the system, which could
bias the results of a whole experiment, are CO2 analyser
span calibration and chamber ventilation rate calibration.
The systematic error in CO2 analyser span calibration was
less than 1%. After taking into account the correlated
effects of the ventilation rate calibration errors on O2 and
CO2 exchange estimates, the systematic errors in substrate
oxidation measurements for a 30 min measurement
period were not worse than 0.96 g for fat and 2.04 g for
carbohydrate.
Estimation of total substrate oxidation and energy balance
Total urinary nitrogen was determined by the Kjeldahl
method (Kjeltec System 1002 distilling unit and Metrohm
Herisau multi-burette E485; Tecator). Total carbohydrate
and fat oxidation were estimated from non-protein O2 and
CO2 exchanges using the coef®cients of Elia et al (1988).
Endogenous fat oxidation was estimated as the difference
between total fat oxidation and ingested (exogenous) fat
oxidation. Endogenous carbohydrate oxidation was estimated as total carbohydrate oxidation since very little CHO
was ingested with the test meals. Energy balance was
calculated as the difference between energy intake and
energy expenditure.
Breath 13CO2 determination and estimation of ingested fat
oxidation
Carbon-13 enrichment of the breath samples was measured
by isotope-ratio mass spectrometry (SIRA-10, VG Isogas,
UK). Results were expressed in per mille (delta) units
relative to an internal standard which was standardized
against Pee Dee Belemnite (PDB lime stone). The external
precision of breath CO2 measurements (n ˆ 5) for the mass
spectrometer used was 0.120d. Exogenous (ingested) fat
Fat ingestion and oxidation
BJ Sonko et al
13
Figure 1
Time course of breath
13
C enrichment (d=30 min) following ingestion of fat doses.
oxidation was estimated according to Mosora et al (1976)
using the equation shown below. For a fuller description of
the assumptions and derivation of the equation see Sonko
(1992).
Qsub ˆ
…dt ÿ db † ‡ …dtÿ1 ÿ db † QCO2 0 Avg0 Mwt
2…dsub ÿ db † n 0:56
where Qsub ˆ exogenous substrate oxidized, dt and
dt71 ˆ breath 13C enrichment at time t, and t 7 1,
db ˆ mean baseline breath 13C enrichment, and dsub ˆ 13C
enrichment of a representative sample of the enriched cornoil which was obtained by combusting duplicate samples at
high temperature (1000 C). QCO2 is moles of CO2 exhaled
by the subject in the calorimeter over a given time interval.
`Avg' Mwt is the weighed molecular weight (869 g) of
corn-oil fat. The number of carbon atoms in a corn-oil
molecule ˆ 55 (n), and 0.56 is the acetate correction factor
(Sidossis et al, 1995).
Statistics
Values are expressed as mean s.e. Repeated measures
analysis was used (Crowder & Hand, 1990) to analyse the
data. GraphPad Prizm statistical software (GraphPad Prizm
2.01, GraphPad Software, San Diego, CA, USA) was used
to determine the correlation and linear regression relationship between ingested fat oxidation and some of the
covariances.
Ethical approval
The study was approved by the Dunn Clinical Nutrition
Centre's Ethical Committee and informed written consent
was obtained from all participants.
Results
Exogenous fat oxidation
In Figure 1 mean breath 13C enrichments following dose
ingestion are presented for only the 20, 35 and 50 g doses.
Time zero on the graph corresponds to 09:30 on the day of
the study. There was a slight decrease in breath 13C
enrichment from 0±30 min after test meal ingestion
which, was followed by a rapid increase until 270 min.
There was very little change in the 13C-elimination rate
between 270 min and the end of the study at 480 min.
Table 2 shows ingested fat and endogenous substrate
oxidation during the study. The relationship between the
European Journal of Clinical Nutrition
Fat ingestion and oxidation
BJ Sonko et al
14
Figure 2 Fat dose oxidized (g) as a function of fat dose ingested (g) (data was not adjusted for body weight, amount of fat dose ingested or gender;
F ˆ females; M ˆ males).
Figure 3 Fat dose oxidized (g) as a function of fat dose ingested (g) (data was adjusted for body weight, amount of fat dose ingested and gender;
F ˆ females; M ˆ males).
European Journal of Clinical Nutrition
Fat ingestion and oxidation
BJ Sonko et al
15
Figure 4
Correlation between proportion (%) of fat dose oxidized and fat dose ingested (g).
Figure 5
Correlation between endogenous carbohydrate oxidation and quantity of ingested fat dose oxidized.
amount of fat dose ingested and the quantity which was
oxidized is given in Figure 2. The graph shows a nonlinear
relationship, with female subjects showing greater capacity
to oxidize the ingested fat than men. After the data was
adjusted for sex and body weight effects, the equation that
best describes the relationship between ingested fat dose
and the amount of it oxidized is a quadratic function, given as:
Fat dose oxidized …g† ˆ ÿ1:482 ‡ 0:2796 …Dose†
‡ 3:896 …Sex† ÿ 0:001357
…Dose†2
(Dose is measured in grams; Sex: male ˆ 0, female
ˆ 1). When this equation was applied to the dose range
European Journal of Clinical Nutrition
Fat ingestion and oxidation
BJ Sonko et al
16
administered to the subjects, the graph in Figure 3 was
obtained. There was a distinct constant sex difference of
approximately 4.0 g (P < 0.02). Body weight was not signi®cantly (P < 0.49) correlated to ingested fat oxidation
when sex was included in the model. In this model,
maximal ingested fat oxidation for both sexes was reached
at approximately 100 g dose with males oxidizing 12.9 g
and females 16.8 g, respectively. The correlation between
fractional fat dose oxidized and the amount ingested was
negative and signi®cant (P < 0.001) (Figure 4). The fractional fat dose oxidized expressed as a proportion of total
fat oxidation was nonlinear (graph not shown) and is
similar to Figure 2.
Endogenous substrate oxidation and energy expenditure
Endogenous carbohydrate. Figure 5 shows the relationship between oxidation of endogenous carbohydrate (CHO)
and ingested fat before adjusting for any variables. The data
shows a negative and signi®cant (P < 0.01) correlation.
After adjusting for both sex and body weight, endogenous
carbohydrate oxidation was still negatively and signi®cantly (P < 0.02) correlated to ingested fat oxidation.
There was no sex effect between the oxidation of these
substrates, but body weight was signi®cantly (P < 0.02) and
positively correlated to endogenous carbohydrate oxidation. The equation which best describes the relationship
between endogenous carbohydrate oxidation and the quantity of fat dose ingested is given as:
Endogenous CHO oxidized …g† ˆ
ÿ22:78 ÿ 0:1658 …Dose† ‡ …1:428† …Weight†
(Dose is measured in grams, and Weight is measured in
kilograms). There was a signi®cant (P < 0.01) and negative
correlation between endogenous carbohydrate oxidation
and quantity of fat ingested. There was no signi®cant
relationship between ingested fat oxidation and the oxidation of either endogenous fat or protein. There was also no
body weight effect on the oxidation of these substrates.
Energy expenditure. Mean total energy expenditure over
the 8 h study period was 3.15 0.3 MJ. Exogenous fat,
endogenous fat, CHO and protein contributed 17.5, 33.6,
36.1 and 12.9%, respectively, to total energy expenditure.
There was a signi®cant (P < 0.001) positive correlation
between energy balance and fat balance. This was not
surprising since fat was the only signi®cant source of
energy in the test meals. Given that most of the ingested
fat was probably stored (Table 2), one would expect such a
correlation.
Discussion
In the present study we investigated in humans a dose±
response relationship between fat intake and its oxidation in
the immediate post-prandial period. In addition, we also
investigated the interaction between ingested fat metabolism
European Journal of Clinical Nutrition
and the oxidation of both endogenous fat and carbohydrate.
The data showed a nonlinear relationship between fat intake
and its subsequent oxidation in the immediate post-prandial
period. Approximately 25% of the mean fat load was oxidized
during the protocol, providing 17.5% of total energy expenditure. Endogenous fat oxidation was not signi®cantly related
to the amount of fat dose oxidized, but endogenous carbohydrate oxidation was negatively and signi®cantly correlated.
Endogenous carbohydrate oxidation was also positively and
signi®cantly correlated to body weight.
In addition, the data also showed sex differences. Female
subjects oxidized signi®cantly (P < 0.02) greater amounts of
the fat load than male subjects. The reasons for this gender
difference are not clear. However, due to the small sample
size in the current study it will be dif®cult to speculate about
possible reasons for the observed sex effect.
The inverse correlation between endogenous carbohydrate oxidation and ingested fat oxidation was in the
expected trend, since carbohydrate oxidation has an inhibitory effect on fat oxidation. The positive correlation
between endogenous carbohydrate and body weight may
be related to body muscle content, although body composition was not measured. A bigger body mass will have
greater muscle mass, which in turn will have more capacity
for carbohydrate storage. In energy de®cit conditions as in
the present study, more of the stored carbohydrate in
muscle could be mobilized for fuel metabolism.
In studies involving 13C breath test techniques for
assessing substrate oxidation, two major issues should be
addressed. Firstly, it is important to prevent perturbing the
natural background 13C enrichment in breath. In the present
study, we controlled for this by using low 13C vegetable
ingredients to prepare the test meals. A subject was
recruited to eat the low 13C vegetable ingredients and his
breath CO2 was collected at 30 min intervals for 360 min.
The mean (n ˆ 4) baseline breath 13CO2 enrichment of this
subject at time zero was 7 25.9d, while his 13CO2 enrichment after eating vegetable ingredients was 7 26.0 0.1d.
This suggested that the non-labelled vegetable component
of the test meal had negligible effect on the subject's
baseline 13C enrichment.
The second issue concerns the underestimation of 13CO2
excretion from the body. When breath test methodologies
are applied it is assumed that the appearance of the label in
breath is re¯ective of the extent of ingested=infused substrate oxidation in vivo. This is not necessarily true. In
studies where 13C-bicarbonate had been infused in human
subjects the percentage of 13CO2 reported as recovered
varied widely from approximately 50 to almost 100%
(Irving et al, 1983; Allsop et al, 1978; Elia et al, 1988;
Halliday, 1991). Furthermore, there is evidence to suggest
that feeding or underfeeding could affect the amount of
13
CO2 recovered (Wenham et al, 1991) in labelled experiments. The bicarbonate correction factor has been used to
account for this discrepancy (Allsop et al, 1978). More
recently, Sidossis et al (1995) suggested that the use of the
bicarbonate correction factor is invalid because it does not
account for labelled ®xation, which occurs in many tissues
Fat ingestion and oxidation
BJ Sonko et al
through tricarboxylic acid cycle (TCA) exchange reactions.
This could lead to signi®cant underestimation, for example,
of plasma free fatty acid oxidation. These authors proposed
an acetate correction factor of 56% for resting conditions
(Sidossis et al, 1995). This factor is supposed to account for
all labelled carbon atoms from the time of their entry into
the TCA cycle to their excretion in breath. However,
Etienne et al (1998) argued that the acetate correction
factor would overestimate substrate oxidation since it
does not account for the fact that some of the acetate
generated during substrate oxidation is disposed through
non-oxidative pathways. In the present study we assumed a
56% recovery and adjusted the data accordingly.
The value estimated for ingested fat oxidation in the
current study is within the range reported in the literature.
Binnert et al (1998) compared the oxidation of ingested
long chain triglycerides (LCT) to medium chain triglycerides (MCT), and found that 15% of the LCT was oxidized
against 57% of the MCT over 360 min. In an earlier study,
the same authors (Binnert et al, 1996) had also demonstrated that 19% of an oral fat load was oxidized over the
same study period. In the two studies mentioned above the
authors gave 30 g of olive oil (LCT study) as the fat load. In
the present study the comparable fat dose given to our
subjects was 35 g (n ˆ 3). The mean proportion of the 35 g
fat dose oxidized in our study was 33%. The difference
between the two studies in the proportion of fat load
oxidized could in part be explained by use of the different
correction factors (acetate vs bicarbonate correction factors) and the study duration. As in our study, Binnert et al
(1996) also observed a positive fat balance in their subjects,
even though energy balance remained negative. In another
study, Metges and Wolfram (1991) compared the oxidation
of orally and parenterally administered MCTs and LCTs.
They concluded that the MCTs were more oxidized than
the LCTs in either mode of administration. The average
proportion of the LCTs oxidized in either mode was about
25% after a 7.5 h protocol. Jones et al (1985) investigated
whole body dietary fatty acid (1-13C stearic, oleic and
linoleic acids) oxidation in humans, and concluded that
over a 9 h protocol, stearic acid was less oxidized compared
to both oleic and linoleic acids. The proportions of the latter
two excreted in breath were 16 and 10%, respectively.
The data presented in this study, as well as the others
mentioned above, suggest that dietary fat is oxidized in the
immediate post-prandial period. The inability of Flatt et al
(1985) and Schutz et al (1989) to show any increase in fat
oxidation following ingestion of the mixed meals given to
their subjects could perhaps be explained in light of the
present ®ndings. Their studies used only RQ values from
indirect calorimetry to estimate ingested fat oxidation. If a
small amount of the fat load is oxidized during the study
compared to total substrate oxidation as demonstrated in
this study, its in¯uence on the over all RQ value is also
likely to be small. Under the usual conditions of indirect
calorimetry, it could be dif®cult to detect such minor
changes, since RQ measurements are not very sensitive to
minor changes in substrate ¯ux.
It is of signi®cance to note that in the present study only
25% of the ingested fat load was oxidized, even though the
subjects were in negative energy balance. In contrast to the
autoregulation usually observed in carbohydrate intake and
oxidation (Jebb et al, 1996), the data in this study shows
that the proportion of ingested fat dose oxidized decreases
signi®cantly with the quantity of fat ingested. This clearly
demonstrates the potential obesity-promoting effect of dietary fat. Whether signi®cantly more ingested fat would be
oxidized over a longer study period can only be answered
by performing longer-term studies. The quadratic nature of
the relationship between the amount of fat dose ingested
and the amount oxidized suggests an upper limit for fat
intake beyond which oxidation is not linear.
In conclusion, the present study demonstrated that about
25% of the ingested fat was oxidized in the immediate postprandial period. The relationship between ingested fat and
its short-term oxidation was nonlinear. The amount of fat
dose ingested was signi®cantly and inversely correlated to
the proportion of it that was oxidized. Endogenous fat and
protein oxidation were not related to ingested fat oxidation,
but endogenous carbohydrate oxidation was signi®cantly
and negatively correlated. It can also be deduced from the
data that studies of inter-individual differences in oxidative
responses to a fat test meal would require two factors to be
valid: ®rst, a carefully adjusted dose which was equivalent
for all subjects; and second, low doses which are in the
linear range in Figure 2, ie not on the plateau.
17
Acknowledgements ÐThe authors wish to thank Dr Richard Jones of the
Department of Biometrics and Preventative Medicine, University of
Colorado Health Sciences Center, Denver, USA, for his help in data
analysis.
References
Allsop JR, Wolfe RR & Burke JF (1978): Tracer priming the bicarbonate
pool. J. Appl. Physiol. 45, 137±139.
Benini L, Scuro LA, Menini E, Manfrini C, Vantini I, Brocco G, Talamini
G & Cavallini (1984): Is the 14C-triolein breath test useful in the
assessment malabsorption in clinical practice? Digestion 29, 91±97.
Binnert C, Pachiaudi C, Beylot M, Croset R, Cohen R, Riou JP & Laville
M (1996): Metabolic fate of an oral long-chain triglyceride load in
humans. Am. J. Physiol. (Endocrinol. Metab. 33) 270, E445±E450.
Binnert C, Pachiaudi C, Beylot M, Hans D, Vandermander J, Chantre P,
Riou JP & Laville M (1998): In¯uence of obesity on the metabolic fate
of dietary long- and medium-chain triacylglycerides. Am. J. Clin. Nutr.
67, 595±601.
Bray GA & Popkin BM (1998): Dietary fat intake does affect obesity! Am.
J. Clin. Nutr. 68, 1157±1173.
Brown D, Cole TJ, Dauncey MJ, Marks RW & Murgatroyd PR (1984):
Analysis of gaseous exchange in open-circuit indirect calorimetry. Med
& Biol Eng & Comput 22, 333±338.
Crowder MJ & Hand DJ (1990): Analysis of Repeated Measures. London:
Chapman & Hall.
Duncan A, Cameron A, Stewart MJ and Russel RI (1992): Limitations of
the tinolein breath test. Clinica Chemica Acta 205, 51±64.
Elia M & Livesey G (1988): Theory and validity of indirect calorimetry
during net lipid synthesis. Am. J. Clin. Nutr. 47, 591±607.
European Journal of Clinical Nutrition
Fat ingestion and oxidation
BJ Sonko et al
18
Elia M, Fuller NJ & Murgatroyd PR (1988): The potential use of the
labelled bicarbonate method for energy expenditure in man. Proc. Nutr.
Soc. 47, 247±258.
Etienne P, Pascale M, Dominique D, J Sergio M, Hugues P, Henri D,
Patrick N & Michel K (1998): Role of glucose and glutamine in the
differential recovery of 13CO2 from infused [2-13C] versus [1-13C]
acetate. Metabolism 47, 549±554.
Flatt JP, Ravussin E, Acheson KJ & JeÂquier E (1985): Effects of dietary fat
on postprandial substrate oxidation and on carbohydrate and fat balances. J. Clin. Invest. 76, 1019±1024.
Grif®ths AJ, Humphery SM, Clark ML, Fielding BA & Frayn KN (1994):
Immediate metabolic availability of dietary fat in combination with
carbohydrate. Am. J. Clin. Nutr. 59, 53±59.
Halliday D (1991): Nutritional applications of stable isotopes. In New
Techniques in Nutritional Research, ed. RG Whitehead & A Prentice,
pp 131±136. London: Academic Press.
Hill JO, Peters JC, Reed GW, Schlundt DG, Sharp T & Greene HL (1991):
Nutrient balance in humans: effects of diet composition. Am. J. Clin.
Nutr. 54, 10±17.
Irving C, William WW, Shulman RJ, Smith EO & Klein PD (1983): [13C]
bicarbonate kinetics in humans: intra-vs-inter-individual variations. Am.
J. Physiol. 245, R190±R202.
Jebb AS, Prentice AM, Goldberg GR, Murgatroyd PR & Black AE (1996):
Changes in macronutrient balance during over- and underfeeding
assessed by 12-d continuous whole-body calorimetry. Am. J. Clin.
Nutr. 64, 259±266.
Jones PJH, Pencharz PB & Clandinin MT (1985): Whole body oxidation of
dietary fatty acids: implications for energy utilization. Am. J. Clin. Nutr.
42, 769±777.
Lissner L, Levitsky DA, Strupp BJ, Kalkwarf HJ & Roe DA (1987):
Dietary fat and the regulation of energy intake in human subjects. Am. J.
Clin. Nutr. 46, 866±892.
Lusk G (1906): The Elements of the Science of Nutrition. Philadelphia,
PA: W.B. Saunders.
Metges CC & Wolfram G (1991): Medium- and long-chain triglycerides
labeled with 13C: a comparison of oxidation after oral or parenteral
administration in humans. J. Nutr. 121, 31±36.
Mosora F, Lefebvre P, Pirnay F, Lacroix M, Luyckx A & Duchesne J (1976):
Quantitative evaluation of the oxidation of an exogenous glucose load
using naturally labelled 13C-glucose. Metabolism 25, 1575±1582.
European Journal of Clinical Nutrition
Murgatroyd PR, Sonko BJ, Wittekind A, Goldberg GR, Ceesay SM &
Prentice AM (1993): Non-invasive techniques for assessing carbohydrate ¯ux: 1. Measurement of depletion by indirect calorimetry. Acta
Physiol. Scand. 147, 91±98.
Newcomer AD, Hofmann AF, DiMagro EP, Thamos PJ & Calson GL
(1979): Triolein breath test. A sensitive and speci®c test for fat
malabsorption. Gastroentrology 76, 6±13.
Paul AA & Southgate DAT (1978): McCance and Widdowson's The
Composition of Foods, 4th edn. London: HMSO.
Prentice AM & Poppitt SD (1996): Importance of energy density and
macronutrients in the regulation of energy intake. Int. J. Obes. Relat.
Metab. Disord. 20(Suppl 2), S18±23.
Schmidt HL & Metges C (1986): Variations of the natural isotope
abundance in diet Ð causes of artifacts or the basis of new possibilities
in stable isotope work? In Clinical Nutrition and Metabolic Research.
Proceedings of the 5th Congress ESPEN, ed. Dietz, Grunhert, Kleinberger & Wolfram. pp 163±168. Munich.
Scho®eld WN, Scho®eld C & James WPT (1985): Predicting basal
metabolic rate, new standards and review of previous work. Hum.
Nutr. Clin. Nutr. 39C, 5±41.
Schutz Y, Flatt JP & JeÂquier E (1989): Failure of dietary fat intake to
promote fat oxidation: a factory favoring the development of obesity.
Am. J. Clin. Nutr. 50, 307±314.
Sidossis LS, Coggan AR, Gastaldelli A & Wolfe RR (1995): A
new correction factor for use in tracer estimations of plasma fatty
acid oxidation. Am. J. Physiol. (Endocrinol. Metab. 32) 269, E649±
E656.
Sonko BJ (1992) PhD thesis, Chap 3, pp 77±79. University of Cambridge,
UK.
Stubbs RJ, Ritz P, Coward WA & Prentice AM (1995): Covert manipulation of the ratio of dietary fat to carbohydrate and energy density: effect
on food intake and energy balance in free-living men eating ad libitum.
Am. J. Clin. Nutr. 62, 330±337.
Wenham D, Pacy PJ, Price G & Halliday D (1991): Bicarbonate recovery:
feeding vs time. Proc. Nutr. Soc. 50, 47A.
Westerterp KR (1993): Food quotient, respiratory quotient, and energy
balance. Am. J. Clin. Nutr. 57(Suppl), 759S±765S.
Wolfe RR & Jahoor F (1990): Recovery of labeled CO2 during the infusion
of C-1- vs C-2-labeled acetate: implications for tracer studies of
substrate oxidation. Am. J. Clin. Nutr. 51, 248±252.