See

discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/5841686

Effect of CHO Loading Patterns on Running

Performance

Article in International Journal of Sports Medicine · August 2008

DOI: 10.1055/s-2007-989265 · Source: PubMed

CITATIONS READS

9 384

6 authors, including:

Chen Yajun Stephen H Wong

Sun Yat-Sen University The Chinese University of Hong Kong
58 PUBLICATIONS 228 CITATIONS 91 PUBLICATIONS 690 CITATIONS

SEE PROFILE SEE PROFILE

Chun Kwok Wong

The Chinese University of Hong Kong
228 PUBLICATIONS 6,037 CITATIONS

SEE PROFILE

All in-text references underlined in blue are linked to publications on ResearchGate, Available from: Chun Kwok Wong
letting you access and read them immediately. Retrieved on: 24 July 2016
598 Nutrition

Effect of CHO Loading Patterns on Running

Performance

Authors Y. Chen 1, S. H. S. Wong 1, X. XU 2, X. Hao 2, C. K. Wong 3, C. W. Lam 3

1
Affiliations Department of Sports Science and Physical Education, The Chinese University of Hong Kong, Hong Kong
2
College of Sports Science and Physical Education, South China Normal University, Guangzhou, China
3
Department of Chemical Pathology, The Chinese University of Hong Kong, The Prince of Wales Hospital, Hong Kong

Key words Abstract
" glycemic index
l !
" glycemic load
l This study examined the influence of 3-day iso-
" run performance
l
" metabolism
energetic carbohydrate (CHO) loading with dif-
l
ferent glycemic index (GI) and glycemic load
(GL) meals on running performance and meta-
bolic responses. With a randomized crossover
design, nine male runners performed a 1-h run
at 70% V̇O2max followed by a 10-km performance
run after a 3-day diet adaptation, which involved
different GI and GL meals [CHO intake (%), GI, and
GL per day were 73%, 80, and 553 for the high GI
and high GL (HH); 73%, 36, and 249 for the low GI
and low GL (LL); and 31%, 79, and 227 for the
high GI and low GL (HL), respectively]. There
were no differences in the time to complete the
10-km run between the two high-CHO trials;
however, the performance in the LL trial was im-
proved as compared to that in the HL trial (mean
± SEM: HH vs. LL vs. HL: 51.3 ± 5.3 vs. 48.6 ± 1.3 vs.
55.3 ± 6.9 min). It appears that the amount,
rather than the nature, of the CHO consumed
during the 3-day isoenergetic CHO loading may
be the most overriding factor on subsequent me-
tabolism and endurance run performance.

Introduction ability in the absolute glycemic response to foods

accepted after revision
August 25, 2007
Bibliography
DOI 10.1055/s-2007-989265
Published online Nov. 14, 2007
Int J Sports Med 2008; 29:
598 – 606 © Georg Thieme
Verlag KG Stuttgart • New York •
ISSN 0172-4622
Correspondence
Prof. Stephen H. S. Wong
The Chinese University of
Hong Kong
Department of Sports Science
and Physical Education
G02, Kwok Sports Building,
Shatin, Hong Kong
852 Hong Kong
China
Phone: + 85 2 26 09 60 95
Fax: + 85 2 26 03 57 81
hsswong@cuhk.edu.hk
! and also reflects the rate of digestion and absorp-
The onset of fatigue during prolonged exercise is
associated with the depletion of the body’s
stored muscle glycogen [11]. For this reason,
there has been great interest over the years in
the nutritional means of increasing the supply of
exogenous carbohydrate (CHO) both before and
during exercise. Carbohydrate loading is known
to produce an increase in stored muscle glycogen,
and this practice has been found to correlate pos-
itively with performance during submaximal ex-
ercise events lasting over 90 min [10], even
though the factors affecting glycogen kinetics
and the underlying mechanism are not com-
pletely understood. During the last decade, the
most commonly used method for CHO loading
was from either the 3-day classical or the modi-
fied regimen [7, 23].
Carbohydrate can be functionally classified ac-
cording to the extent to which it increases blood
glucose levels. This leads to the concept of glyce-
mic index (GI), which is a ranking of foods based
on their actual postprandial blood glucose re-
sponse as compared to a reference food, either
glucose or white bread [16]. This index counters
the problem of considerable interindividual vari-
tion of a CHO-rich food. In addition, the glycemic
load (GL) was first proposed in 1997 by Salmeron
and co-workers as the measure which incorpo-
rates both the quantity and quality of the dietary
CHO consumed [21].
Some studies have indicated that a low GI (LGI)
meal may have potential benefits over a high GI
(HGI) meal when considering a high CHO intake
prior to exercise because of the promotion of sus-
tained CHO availability during exercise [25, 30].
In comparison to a number of studies that have
examined the application of a pre-exercise CHO
GI meal [25, 27, 30], the concept of GL is less
studied directly in sports nutrition and is primar-
ily applied in epidemiological research [16, 21,
26]. Furthermore, GL is derived by multiplying
the amount of available CHO consumed in the
diet by its GI value. Therefore, GL can be reduced
by lessening either the CHO intake or the dietary
GI. Although both of these approaches reduce
acute plasma glucose and insulin responses [15],
their effect on metabolic and physiological re-
sponses during exercise is still unclear. A latest
study in our laboratory [8] observed that the in-
gestion of mixed isoenergetic meals with an

Chen Y et al. Effect of CHO … Int J Sports Med 2008; 29: 598 – 606

equal GL 2 h before exercise, regardless of their differences in GI
and CHO content, produces similar glycemic and insulin re-
sponses and other metabolic changes during the postprandial
period and during exercise. However, there is little information,
if any, on the short-term adaptation to the effects of the meals
with different GI and GL on subsequent metabolic responses.
Although the above-mentioned CHO loading regimens can ele-
vate muscle glycogen content to levels of 160 – 200 mmol • kg–1
wet mass, it is important to note that none of the studies from
the original/modified regimen has attempted to establish
whether or not the CHO type (GI) ingested in the CHO loading
period might influence the supercompensation status and sub-
sequent endurance performance [7]. Thus, the purpose of this
study was to investigate the influence of 3-day CHO loading with
different GI and GL meals on running performance and subse-
quent metabolic responses. To prevent consuming more time,
some modified CHO-loading protocols started the regimen with
a high-CHO diet while ceasing or tapering-down exercise [7].
However, muscle biopsy studies showed that when glycogen
concentrations are decreased to 66 – 70 mmol • kg–1 wet wt, gly-
cogen synthesis activity increases rapidly [3]. Upon CHO loading,
competitive athletes generally prefer moderate training to rest
out of concern that detraining may occur and offset the potential
benefits of supercompensated muscle glycogen [19]. Conversely,
a high-intensity, exhaustive prolonged exercise and CHO loading
will prevent muscle glycogen supercompensation [24]. It has
been reported that 20 min of moderate exercise could be per-
formed daily throughout the CHO loading protocol without neg-
atively affecting muscle glycogen supercompensation [13]. Thus,
the present study designed a modified CHO loading protocol
with a 1-h exhaustive exercise before the CHO loading for glyco-
gen depletion and a 20-min light training during the CHO load-
ing for the maintenance of metabolic adaptations acquired
through endurance training.
It was hypothesized that (1) a 3-day CHO loading with an equal
GL diet would produce similar glycemic and metabolic re-
Nutrition 599
Fig. 1 Schematic representation of the experi-
mental procedures.
sponses regardless of their difference in GI and CHO content,
and (2) that as compared to a HGI diet with an equal amount of
CHO, a 3-day CHO loading with a LGI meal would result in an at-
tenuated metabolic response to exercise and an improved run-
ning performance.
Methods
!
Subjects
Nine healthy male runners (age: 20.1 ± 0.8 years; body mass:
64.8 ± 5.0 kg; V̇O2max: 61.9 ± 2.9 ml • kg–1 • min–1) were recruited
for this study. Ethical approval was obtained from the Clinical
Research Ethical Committee of the University. Written informed
consents were collected from all the subjects. They were also re-
quired to complete medical history questionnaires and provide
details of their running ability. None of the subjects had a history
of diabetes mellitus.
Experimental design
With a counterbalanced crossover design, subjects completed
three different standardized preloaded treadmill performance
runs. The procedure involved a 3-day CHO loading with different
GI and GL meals after a 1-h exhaustive run, followed by a stan-
dardized preloaded exercise protocol, i.e., 1-h preloaded con-
stant run at 70% V̇O2max (T1), and a 10-km time trial (TT)
(l
" Fig. 1). The pretrial preload allows for achievement of a
“steady state”, which may be advantageous to researchers mea-
suring gas exchanges. Furthermore, this 1-h run also allows for a
lowering of endogenous substrate stores, i.e., muscle glycogen,
before a TT performance test, and this type of testing has been
suggested to be a relatively reproducible method when mea-
sured in endurance-trained runners, which is important for re-
search assessing the effect of diet on endurance running per-
formance [9, 25]. After performing a glycogen-depletion exer-
cise, the subjects were randomly assigned to one of the three tri-
Chen Y et al. Effect of CHO … Int J Sports Med 2008; 29: 598 – 606
600 Nutrition
Table 1 Nutritional protocol (for a 70-kg subject)
Trials High GI, Low GI,
High GL Low GL
(HH) (LL)
Fat (g/day) (%) 60 (14) 62 (14)
Protein (g/day) (%) 126 (13) 124 (13)
CHO (g/day) (%) 701 (73) 699 (73)
GI 80 36
GL (GI × CHO [g]/100) 553 249
Calorie (kcal/day) 3838 3830
Note: Because the GL is a calculated indicator that incorporates the quantity and
quality of the CHO [GI × CHO (g)/100], an isocaloric diet with a low GI (< 45) and high
GL (LH) is practically unfeasible.
als, namely, high GI, high GL (HH), low GI, low GL (LL), and high
GI, low GL (HL), in the following three days (l
" Table 1). Each trial
was separated by at least seven days.
Preliminary test
After familiarizing the subjects with treadmill running and the
experimental procedures, they were required to undertake a se-
ries of preliminary tests before the main trials, which included a
speed-V̇O2 test, a V̇O2max test, and a speed-lactate test, through
which the speeds of corresponding %V̇O2max were determined.
All the air samples were analyzed by a metabolic cart system
(MAX-1 Physio-Dyne, Quogue, NY, USA). Before the main trials,
the postprandial glycemic responses of the subjects were as-
sessed during a screening session to ensure that they responded
normally to the prescribed test meals [12, 25].
Dietary analyses and training control
Each subject was required to have a 3-day food diary record be-
fore the 3-day adaptation to the diets with different GI/GL and to
repeat the same diet before each main trial in order to minimize
the variation in muscle and liver glycogen concentrations. In or-
der to maintain a euhydration condition before each experimen-
tal test, the subjects were instructed to drink approximately
500 ml of water the night before they reported to the laboratory.
In order to exclude any residual effects of previous exercise on
the experimental treatment, they were also required to refrain
from strenuous exercise and consumption of alcohol or caffeine
24 h prior to each main trial.
Prescribed GI/GL meals
All meals of the three trials were planned by a registered dieti-
cian from the Hong Kong Sports Institute and were prepared by
the same person. The HH and LL trials provided 10.0 g • kg–1 BM of
available CHO per day, whereas the HL trial provided only
4.0 g • kg–1 BM of available CHO per day. The water content of all
meals was standardized so that each provided equal fluid. In or-
der to avoid the occurrence of bias, the specific purpose related
to the GI/GL foods and the weight of the food was not disclosed
to the subjects. The detailed contents of the meals used in the 3-
day CHO loading are shown in l " Table 2.
Experimental procedures
On the first day of the CHO loading protocol, the subjects com-
pleted a 1-h exhaustive exercise, which consisted of a 30-min
run at 80 % V̇O2max, followed by a further 30-min run at 70%
V̇O2max, in order to reduce the muscle glycogen [2]. On the sub-
Chen Y et al. Effect of CHO … Int J Sports Med 2008; 29: 598 – 606
sequent two days in the CHO loading period, all the subjects
were required to perform light exercise training, i.e., 20-min
High GI, run at 70% V̇O2max on a treadmill. On day 4, the subjects began
Low GL the standardized exercise protocol after the test breakfast. All
(HL) the last breakfast foods (each regimen started from the lunch
205 (49) on day 1 and ended at the breakfast on day 4) in the main trial
198 (21) had to be consumed within 15 to 20 min. The subjects remained
284 (30) seated in a quiet area of the laboratory, and their activity level
79 was minimal. Blood glucose concentration and the perceived
227
rating of gut fullness were recorded throughout the 2-h post-
3822
prandial period at specific time points (i.e., 15, 30, 45, 60, 90,
and 120 min after the test meal). These blood glucose assess-
ments were made using finger-prick sampling and were a sepa-
rate measurement from the venous blood sampling measure-
ments indicated in l " Fig. 1.
A standardized 5-min warm-up at 60 % V̇O2max was performed
after the 2-h resting period. The treadmill speed was increased
to that required to elicit 70% V̇O2max immediately following the
warm-up. T1 required the subjects to run at a fixed 70% V̇O2max
for 1 h [32]. During the TT, the subjects could change their
speeds ad libitum and were required to run as quickly as possible
to complete the 10-km distance. To ensure maximal effort dur-
ing the 10-km performance run, strong verbal encouragement
was given to the participants throughout the run solely by the
chief experimenter, who was unaware of the treatments. Dis-
tilled water intake was not restricted throughout the exercise
and the 2-h recovery period.
Throughout each exercise test, expired air samples were col-
lected and analyzed at 20-min intervals during the 1-h run at
70% V̇O2max (T1), at 2.5-km intervals during the 10-km TT, and
at 1-h intervals during the 2-h recovery. From the V̇O2 and
V̇CO2 values collected, the substrate oxidation rates and the total
CHO and fat oxidized were estimated using the stoichiometric
equations [18]. Heart rate (HR) was monitored continuously us-
ing a HR monitor (Sport Tester PE 4000, Polar Electro, Kempele,
Finland) and recorded throughout the experimental trials. Rat-
ings of perceived exertion (RPE), ratings of abdominal discom-
fort (AD), and perceived thirst (PT) were also obtained [25].
At each time point in the premeal, 1-h pre-exercise (– 60 min),
immediately before exercise (Pre-Ex: 0 min), 20-min intervals
during T1, 5-km interval during the 10-km TT, and the first
(Post-60 min) and second (Post-120 min) hour in the recovery,
5-ml venous blood samples were drawn from an antecubital
vein into evacuated collection tubes (Vacutainer; Becton Dickin-
son, Mountain View, CA, USA) and then were stored at – 70 8C for
later measurement of serum insulin (Insulin ELISA, Mercodia,
Sweden), serum free fatty acid (FA 115 kit, Randox Laboratories
Ltd., Crumlin, UK), and serum glycerol (GY 105 kit, Randox Labo-
ratories Ltd., Crumlin, UK) concentrations by using commercially
available kits.
The data were examined using a two-factor (trial × time of mea-
surement) ANOVA with repeated measures design. Any signifi-
cant F ratios were assessed using the post hoc Tukey test, where
appropriate, to locate any significant difference. All data were
presented as a mean ± standard error of the mean (SEM), with
the significance set at p < 0.05. The effect size (ES) was also esti-
mated for the strength of meaningfulness of the treatment ef-
fects.
 Nutrition 601

Table 2 Three meals used in the 3-day CHO loading (for a 70-kg subject)

HH HL LL
D1 252 g white bread + 84 g white bread + 350 g macaroni (cooked) +
10 g margarine + 20 g margarine + 10 g lean ham +
350 ml soy milk + 50 g raw whole egg + 74 g tomato sauce +
65 g cornflakes cereal + 134 g egg white + 114 g sponge cake +
500 ml orange juice + 45 g lean ham + 400 g spaghetti +
10 g white sugar + 180 g beef sirloin + 500 ml Fanta orange soda +
900 g white rice (cooked) + 142 g iceberg lettuce + 71 g ground beef +
100 g beef sirloin (lean) + 110 g banana + 200 g apple +
113 g bok choy + 90 g cashew nuts + 74 g raw egg white +
500 ml apple juice (canned) + cornflakes cereal + 250 g bok choy +
25 g strawberry jam + 500 ml soy milk + 550 ml apple juice (canned) +
250 ml lemon tea + 425 g white rice (cooked) + 330 g rice noodle +
4 tsp sugar (20 g) + 180 g pork loin chop with lean and fat + 300 g low-fat fruit yoghurt +
100 g chicken breast without skin + 142 g choy sum + 60 g chicken breast without skin +
113 g choy sum + 150 g watermelon + 150 g grapes
250 g grapes 500 ml lemon tea with total 7 tsp sugar (35 g)
D2 252 g white bread + 84 g white bread + 60 g Kellogg’s all bran cereal +
90 g salmon + 33 g margarine + 300 ml soy milk+
300 ml soy milk + 200 g salmon + 20 g sugar +
60 g cornflakes cereal + 100 g egg white + 120 g banana +
500 ml pineapple juice + 70 g lean ham + 250 ml orange juice +
15 g white sugar + 142 g iceberg lettuce + 400 g rice noodle +
800 g white rice (cooked) + 100 g banana + 85 g salmon +
300 g watermelon + 95 g cashew nuts + 200 g choy sum +
114 g spinach (raw) + 60 g tofu + 250 ml lemon tea +
550 ml Coca-Cola classic + 300 ml soy milk + 3 tsp sugar (15 g) +
15 g strawberry jam + 460 g white rice (cooked) + 131 g orange +
20 g peanut butter + 180 g chicken meat without skin + 57 g blueberry muffin +
100 g chicken breast without skin + 142 g choy sum + 500 ml apple juice +
113 g choy sum + 100 g watermelon + 400 g spaghetti, +
150 g banana 250 ml lemon tea with 3 tsp sugar (15 g)+ 90 g lean pork loin chop +
100 g low fat fruit yoghurt 113 g lettuce +
152 g kiwi fruit +
700 ml Coca-Cola classic +
250 g low-fat fruit yoghurt
D3 252 g white bread + 126 g white bread + 350 g macaroni (cooked) +
10 g margarine + 35 g margarine + 30 g lean ham +
90 g beef sirloin (lean) + 50 g raw whole egg + 60 g lean beef sirloin +
350 ml soy milk + 67 g egg white + 76 g sponge cake +
60 g cornflakes cereal + 60 g corned beef + 425 g spaghetti +
500 ml orange juice + 180 g beef sirloin + 500 ml orange juice +
5 g white sugar + 142 g bok choy + 113 g iceberg lettuce +
800 g white rice (cooked) + 200 g apple + 200 g apple +
250 g watermelon + 75 g peanuts + 67 g raw egg white +
113 g spinach (raw) + 300 ml soy milk + 113 g raw spinach +
500 ml gatorade + 25 g sugar + 510 g white rice (cooked) + 100 g yam +
113 g choy sum + 171 g pork loin chop with lean and fat + 260 g rice noodle +
10 g strawberry jam + 142 g choy sum + 100 g low-fat fruit yoghurt +
374 g canned peach + 250 ml lemon tea with 3 tsp sugar (15 g) 60 g chicken breast without skin +
150 g banana + 200 g grapes +
250 ml lemon tea + 250 ml lemon water with 4 tsp sugar (20 g)
3 tsp sugar (15 g) +
100 g chicken breast without skin +
550 ml Coca-Cola classic

Results at the same hydration status and under similar environmental

! conditions.
In the three trials, the body mass in Day 1 (pre-depletion) (HH The resting body mass did not change after the 3-day dietary
vs. HL vs. LL: 64.8 ± 5.4 vs. 64.8 ± 4.7 vs. 64.9 ± 5.1 kg), the labora- treatments with the high-CHO meal (HH: 64.8 ± 5.4 kg vs. 64.9 ±
tory temperature (HH vs. LL vs. HL: 22.0 ± 1.5 vs. 21.2 ± 1.1 vs. 5.6 kg for Day 1 and pre-Ex; LL: 64.8 ± 4.7 kg vs. 64.8 ± 5.2 kg for
21.7 ± 2.2), and relative humidity (HH vs. LL vs. HL: 64.2 ± 2.4% Day 1 and pre-Ex) and light training. However, body mass de-
vs. 62.9 ± 3.5% vs. 66.1 ± 2.7%) in Day 4 (pre-exercise) were not creased (p < 0.05) when the subjects ate the low-CHO diet (HL:
different, indicating that the subjects started all the experiments 64.9 ± 5.1 kg vs. 64.6 ± 5.6 kg for Day 1 and pre-Ex) with light
training.

Chen Y et al. Effect of CHO … Int J Sports Med 2008; 29: 598 – 606
602 Nutrition

Table 3 Total CHO oxidation and fat oxidation before, during, and after exercise in the H-H, L-L, and H-L trials (n = 9, mean ± SEM)

Trial Total CHO oxidized (g) Total fat oxidized (g)

Postprandial Exercise Recovery Postprandial Exercise Recovery
HH 34.12 ± 6.17 251.40 ± 16.59 36.15 ± 4.09 12.57 ± 1.55 27.44 ± 5.19 15.62 ± 3.02
LL 32.59 ± 7.15 245.25 ± 11.36 35.29 ± 7.71 13.05 ± 1.83 29.64 ± 6.26 16.57 ± 4.18
HL 26.74 ± 8.11a 218.89 ± 10.00a 30.00 ± 5.64 14.81 ± 3.56 37.22 ± 3.15a 14.31 ± 3.75
a
p < 0.01 from H-H and L-L

Table 4 Heart rate, rate of perceived exertion (RPE), abdominal discomfort (AD), and perceived thirst (PT) before, during, and after exercise during H-H, L-L, and
H-L trials (n = 9, mean ± SEM)

Pre- 20 min 40 min 60 min 2.5-km 5-km 7.5-km 10-km Post- Post-
meal 60 min 120 min
Heart rate (beats • min–1)
" HH 61 ± 3 169 ± 3a 169 ± 3a 170 ± 3a 172 ± 2a 180 ± 3a 182 ± 3a 187 ± 2a 75 ± 5 68 ± 2
" LL 59 ± 3 164 ± 2a 164 ± 2a 166 ± 2 a 168 ± 3a 180 ± 2a 181 ± 2a 191 ± 2a 78 ± 2 73 ± 4
" HL 63 ± 5 163 ± 3a 169 ± 2a 171 ± 2 a 171 ± 4a 175 ± 3a 178 ± 2a 180 ± 2a 84 ± 5 66 ± 2
RPE
" HH 6±0 11 ± 1a 12 ± 1a 13 ± 1 a 15 ± 0a 15 ± 1a 16 ± 0a 17 ± 1a 7±0 6±0
" LL 6±0 10 ± 1a 11 ± 1a 13 ± 1 a 15 ± 1a 15 ± 1a 16 ± 1a 17 ± 1a 8±0 6±0
" HL 6±0 10 ± 1a 12 ± 1a 13 ± 1 a 15 ± 0a 15 ± 1a 17 ± 1a 17 ± 1a 8±0 7±0
AD
" HH 0.8 ± 0.2 2.3 ± 0.4 2.2 ± 0.4a 2.6 ± 0.4 a 2.9 ± 0.4a 2.6 ± 0.4a 3.8 ± 0.5a 3.6 ± 0.3a 0.6 ± 0.6 0.5 ± 0.2
" LL 0.6 ± 0.5 2.6 ± 0.2a 3.1 ± 0.5a,b 3.8 ± 0.4a,b 3.8 ± 0.6a,b 4.4 ± 0.5a,b 4.4 ± 0.9a 4.8 ± 0.5a,b 0.6 ± 0.7 0.6 ± 0.3
" HL 0.3 ± 0.4 2.1 ± 0.9a 2.9 ± 0.3a 3.6 ± 0.5a,b 3.4 ± 0.5a 5.0 ± 0.3a,b 5.6 ± 0.8a,b 5.8 ± 0.6a,b 0.9 ± 0.4 0.5 ± 0.4
PT
" HH 1 ± 0.4 3 ± 0.3a 4 ± 0.2a 4 ± 0.3 a 4 ± 0.5a 4 ± 0.5a 4 ± 0.3a 4 ± 0.6a 2 ± 0.4 1 ± 0.1
" LL 1 ± 0.2 4 ± 0.3a 4 ± 0.3a 5 ± 0.4 a 5 ± 0.3a 5 ± 0.5a 5 ± 0.5a 5 ± 0.5a 2 ± 0.4 1 ± 0.2
" HL 1 ± 0.1 4 ± 0.4a 5 ± 0.5a 5 ± 0.3 a 5 ± 0.4a 5 ± 0.4a 5 ± 0.4a 5 ± 0.5a 2 ± 0.5 1 ± 0.4
a
p < 0.01 from pre-meal; b p < 0.05 from H-H

There were no differences in the time to complete the 10-km run
between the two high-CHO trials (HH vs. LL: 51.3 ± 5.3 vs.
48.6 ± 1.3 min, p = 0.127). However, the performance in the LL tri-
al was improved as compared to that in the HL trial (LL vs. HL:
48.6 ± 1.3 vs. 55.3 ± 6.9 min, p < 0.05).
Average RER values were higher in the two high-CHO trials (HH
and LL) compared with the low-CHO trial (HL) (p < 0.05)
(l
" Fig. 4) before exercise. The average RER values were mostly
Blood glucose
The 2-h incremental area under the blood glucose curve after the
last breakfast was about 1.5 times larger in the HH trial than in
the LL and HL trials (HH vs. LL vs. HL: 164.3 ± 6.2 vs. 110.4 ± 6.0 vs.
101.4 ± 6.4 mmol • L–1 • min–1, p < 0.01). Higher blood glucose con-
centrations were observed in the two high-CHO trials (HH and
LL) during the exercise and the recovery period (p < 0.05) com-
pared with that in the low-CHO trial (HL) (l " Fig. 2).

higher in the HH trial compared with those in the HL trial

throughout the experimental period, except for the initial period
of T1. However, there were no differences between the two high-
CHO trials.
During the 2-h postprandial period, the total CHO oxidation was
lower (p < 0.05) in the HL trial than in the two high-CHO trials.
During exercise, the total CHO oxidation remained lower
(p < 0.05) in the HL trial, with a compensatory increase in fat ox-
idation (p < 0.05) as compared to that in the HH and LL trials
" Table 3). There were no differences in CHO and fat oxidation
(l
Serum insulin
Serum insulin concentrations peaked at 60 min after consump-
tion of the test meals in all trials and then decreased gradually
towards the end of the exercise period. Just prior to exercise,
the plasma insulin concentrations in the HH trial were still high-
er (p < 0.05) than the premeal value and that of the other two
low-GL trials (LL and HL). No differences in the plasma insulin
values were observed throughout the exercise and recovery pe-
riod among the three trials (l" Fig. 3).

between the two high-CHO trials (HH and LL). The overall energy
expenditure in the three trials was similar irrespective of their Serum NEFA and glycerol
difference in substrate oxidation (data not shown). Serum non-esterified fatty acids (NEFA) (l " Fig. 5) and serum

There were no differences in HR, RPE, and PT at any time points glycerol concentrations (l Fig. 6) were greater before exercise
"

among the three trials (l " Table 4). The AD ratings were higher
(p < 0.05) during exercise in the three trials compared to rest,
while they were lower (p < 0.05) in the HH trial at the later stage
of T1 and during the 10-km TT than that in the HL and LL trials.
after the 3-day dietary treatments and light training in the low-
CHO trial (HL) than in the two high-CHO trials (HH and LL)
(p < 0.05). The NEFA concentrations increased at the onset of
the exercise and were greater starting from the 60-min point of
T1 to the end of the 2-h recovery in the HL trial compared with
those in the HH and LL trials (p < 0.05). Similar to the NEFA, the

Chen Y et al. Effect of CHO … Int J Sports Med 2008; 29: 598 – 606
 Nutrition 603

Fig. 2 Blood glucose concentrations (mmol • L–1)

before, during, and after exercise in the HH, LL and
HL trials (n = 9; mean ± SEM). a: p < 0.01 from Pre-
meal; b: p < 0.01 from HL; c: p < 0.01 from HH.

Fig. 3 Serum insulin concentrations (mU • L–1) be-

fore, during, and after exercise in the HH, LL, and
HL trials (n = 9; mean ± SEM). a: p < 0.01 from Pre-
meal; b: p < 0.01 from HL; c: p < 0.01 from HH.

Fig. 4 Respiratory exchange rate (RER) before,

during, and after exercise during the HH,LL, and HL
trials (n = 9; mean ± SEM). a: p < 0.05 from HH;
b: p < 0.05 from HL.

Chen Y et al. Effect of CHO … Int J Sports Med 2008; 29: 598 – 606
604 Nutrition
serum glycerol concentrations were higher in the HL trial than
those in the HH and LL trials throughout the exercise (p < 0.05).
However, there were no differences for both serum NEFA and se-
rum glycerol concentrations between the HH and LL trials
throughout the experiments.
Blood lactate
The blood lactate concentrations during exercise were higher
than the pre-exercise value in all trials (p < 0.01). Higher blood
lactate concentrations were observed in the two high-CHO trials
(HH and LL) at the end of the 10-km TT compared with the low-
CHO trial (HL) (p < 0.05) (l
" Fig. 7).
Discussion
! higher NEFA and glycerol concentrations during the 2-h post-
The major finding of this study was that there were no differ-
ences in the time to complete a preloaded 10-km TT between
the two high–CHO trials (HH and LL). However, the performance
in the LL trial was improved as compared to that in the HL trial.
The results also indicated that a 3-day CHO loading with differ-
ent GI and GL meals and daily light training altered the glycemic
Chen Y et al. Effect of CHO … Int J Sports Med 2008; 29: 598 – 606
Fig. 5 Serum NFFA concentrations (mmol • L–1)
before, during, and after exercise in the HH, LL, and
HL trials (n = 9; mean ± SEM). a: p < 0.01 from Pre-
meal; b: p < 0.01 from HL; c: p < 0.01 from HH.
Fig. 6 Serum glycerol concentrations (mmol • L–1)
before, during, and after exercise in the HH, LL and
HL trials (n = 9; mean ± SEM). a: p < 0.01 from Pre-
meal; b: p < 0.01 from HL; c: p < 0.01 from HH.
and insulinemic responses, pattern of substrate availability in
plasma, substrate utilization, and Pre-Ex CHO status, only be-
tween the high CHO (HH and LL) and low-CHO trials (HL).
Despite the lack of a measurement of muscle glycogen by biopsy
or MRS, the pre-exercise CHO status could also be assessed indi-
rectly via such indications as changes in body mass and the sub-
strate availability in plasma. The loss of body mass during the 3-
day CHO loading in the low-CHO trial (HL) probably represented
the loss of water that had been stored by the liver, and muscle
glycogen. Although the Pre-Ex blood glucose concentrations
were similar among the three trials, lower mean glucose values,
i.e., 3.32 mmol • L–1 at the end of 10-km TT and 3.26 mmol • L–1
during the 1-h recovery, were observed in the HL trial, which
was comparable to that typically associated with fatigue en-
countered at the end of an endurance exercise [11]. In addition,
prandial period and the whole experiment period appeared to
be a reduced CHO availability in the HL trial.
The relative contribution of CHO and fat as fuel for working
muscles is largely dependent on exercise intensity and duration
[29]. In the present study, there were no differences in the RPE,
HR, and V̇O2 responses throughout the exercise period among

the three trials, which indicates that the subjects were running
at a similar physical exertion level for each trial. However, higher
NEFA availability, higher fat oxidation, and lower blood glucose
values were still observed in the low-CHO trial compared with
those in the two high-CHO trials (HH and LL). Without consider-
ing the GI or GL value of the meals, these results were similar to
previous studies in which increased fat oxidation during exercise
in response to short-term (3 days) or long-term (> 7 days) con-
sumption of a low-CHO diet was observed [22]. It seems that in-
creased plasma fatty acid availability is responsible for the in-
crease in fat oxidation in response to low-CHO diets both at rest
and during exercise [20]. However, Schrauwen et al. [22] and
Bosher et al. [4] reported that the fat and CHO oxidation were
not affected despite the difference in the concentrations of glu-
cose, insulin, and NEFA after endurance or resistance exercise
following ingestion of high-CHO or high-fat diets. Therefore, the
increased fat oxidation after the low-CHO loading in the HL trial
may be due to the contribution of plasma TG or intramuscular
triglyceride (IMTG). This notion may be supported to some ex-
tent by findings from studies that long-term adaptation to
high-fat diets increase the capacity of muscle for plasma triglyc-
eride (TG) uptake by increasing lipoprotein lipase activity and
IMTG storages [14], whereas high-CHO diets increase stored
muscle glycogen [14]. However, it could not be determined if a
similar “adaptation” to high-fat diets in the “long-term adapta-
tion” to a high-fat diet had occurred in the HL trial in our present
study after the 3-day low-CHO loading.
A higher blood lactate concentration immediately after exercise
in the two high-CHO trials (HH and LL) might suggest that CHO
loading in the high-CHO trials had increased glycogenolysis, per-
haps by the greater utilization of muscle glycogen. Previously,
studies have reported that higher lactate levels were observed
after CHO loading compared with normal or depressed glycogen
levels [1], indicating that glycogen levels influence its utilization
during exercise. It has also been shown that when lactate and
glucose are infused simultaneously during exercise, lactate turn-
over exceeds that of glucose, indicating that muscle glycogen
provides most of the CHO oxidized during exercise [5].
Nutrition 605
Fig. 7 Blood lactate concentrations (mmol • L–1)
before, during, and after exercise in the HH, LL and
HL trials (n = 9; mean ± SEM). a: p < 0.01 from Pre-
meal; b: p < 0.01 from HL; c: p < 0.01 from HH.
It has been suggested that muscle glycogenolysis is not affected
by pre-exercise CHO ingestion if the reduction in fat oxidation
due to hyperinsulinemia is offset by the increase in muscle glu-
cose uptake and subsequent oxidation [3]. However, if the in-
crease in muscle glucose uptake is not as large as the insulin-
mediated decline in fat oxidation, muscle glycogen utilization
may be augmented [3]. A recent study suggested that the inges-
tion of an HGI meal increased muscle glycogen storage at rest
but augmented its utilization during subsequent exercise as
compared to an isoenergetic LGI meal [31]. In relation to this,
Burke et al. [6] observed that consuming an HGI meal during a
24-h recovery period from prolonged heavy exercise resulted in
higher muscle glycogen resynthesis as compared to an isoener-
getic LGI diet. However, improved endurance performance and
improved recovery from prolonged exercise were observed fol-
lowing the consumption of LGI meals compared with the results
following the isoenergetic HGI meals in the 24-h recovery period
[27]. It seems that the effect of pre-exercise CHO loading with
different GI and GL is a mixed phenomenon in the present study.
Although the 2-h incremental area under the blood glucose
curve after the last breakfast was larger in the HH trial than
those in the LL and HL trials as predicted [12], higher blood glu-
cose concentrations were similarly observed in the two high-
CHO trials (HH and LL) during the exercise and recovery period.
The HH meal had a GI of 80 and a GL of 553 per day, compared
with the LL meals with a GI of 36 and a GL of 249 per day in the
present study; however, there was no difference in the glycemic,
insulinemic, and lipolytic responses between the two high-CHO
trials throughout the exercise and recovery period. Similar re-
sults have been reported from other studies which suggested an
almost similar muscle glycogen synthesis rate after the con-
sumption of an HGI or LGI diet with large amounts of CHO pro-
vided [17, 28].
After the last HH breakfast in the present study, the plasma glu-
cose concentrations at 20 min during exercise were lower than
those in the LL trial, which was probably the result of higher in-
sulin concentrations at Pre-Ex. It has been well established that
the pre-exercise ingestion of an HGI CHO produces immediate
Chen Y et al. Effect of CHO … Int J Sports Med 2008; 29: 598 – 606
606 Nutrition
hyperglycemic responses, followed by a rapid decline in plasma
glucose concentrations [30, 31].
In conclusion, 3-day isoenergetic CHO loading with low GI and
low GL is more effective in improving endurance performance
as compared to the low-CHO diets with high GI but low GL. Re-
gardless of their difference in GI and GL, the CHO loading with an
equally high-CHO amount (HH and LL) may produce similar
muscle glycogen supercompensation status and exercise meta-
bolic responses during subsequent endurance running. It ap-
pears that the amount, rather than the nature, of the CHO con-
sumed during the 3-day regimen may be the most overriding
factor on metabolism and endurance run performance.
References
1 Andrews JL, Sedlock DA, Flynn MG, Navalta JW, Ji H. Carbohydrate load-
ing and supplementation in endurance-trained women runners. J
Appl Physiol 2003; 95: 584 – 590
2 Bishop NC, Walsh NP, Haines DL, Richards EE, Gleeson M. Pre-exercise
carbohydrate status and immune responses to prolonged cycling: I.
Effect on neutrophil degranulation. Int J Sport Nutr Exerc Metab
2001; 11: 490 – 502
3 Bogardus C, Thuillez P, Ravussin E, Vasquez B, Narimiga M, Azhar S. Ef-
fect of muscle glycogen depletion on in vivo insulin action in man. J
Clin Invest 1983; 72: 1605 – 1610
4 Bosher KJ, Potteiger JA, Gennings C, Luebbers PE, Shannon KA, Shannon
RM. Effects of different macronutrient consumption following a resis-
tance-training session on fat and carbohydrate metabolism. J Strength
Cond Res 2004; 18: 212 – 219
5 Brooks GA, Butterfield GE, Wolfe RR, Groves BM, Mazzeo RS, Sutton JR,
Wolfel EE, Reeves JT. Decreased reliance on lactate during exercise after
acclimatization to 4300 m. J Appl Physiol 1991; 71: 333 – 341
6 Burke LM, Gregory CR, Hargreaves M. Muscle glycogen storage after
prolonged exercise: effect of the glycemic index of carbohydrate feed-
ings. J Appl Physiol 1993; 75: 1019 – 1023
7 Burke LM, Hawley JA, Schabort EJ, St Clair Gibson A, Mujika I, Noakes TD.
Carbohydrate loading failed to improve 100-km cycling performance
in a placebo-controlled trial. J Appl Physiol 2000; 88: 1284 – 1290
8 Chen YJ, Wong SH, Wong CK, Lam CW, Huang YJ, Siu PM. Effect of pre-
exercise meals with different glycemic indices and glycemic loads on
metabolic responses and endurance running performance. Med Sci
Sports Exerc 2006; 38: S 37
9 Doyle JA, Martinez AL. Reliability of a protocol for testing endurance
performance in runners and cyclists. Res Q Exerc Sport 1998; 69:
304 – 307
10 Chryssanthopoulos C, Williams C, Wilson W, Asher L, Hearne L. Compar-
ison between carbohydrate feedings before and during exercise on
running performance during a 30-km treadmill time trial. Int J Sport
Nutr 1994; 4: 374 – 386
11 Coggan AR, Coyle EF. Reversal of fatigue during prolonged exercise by
carbohydrate infusion or ingestion. J Appl Physiol 1987; 63: 2388 –
2395
12 Foster-Powell K, Holt SHA, Brand-Miller JC. International table of glyce-
mic index and glycemic load values. Am J Clin Nutr 2002; 76: 5 – 56
13 Goforth HW Jr, Laurent D, Prusaczyk WK, Schneider KE, Petersen KF,
Shulman GI. Effects of depletion exercise and light training on muscle
glycogen supercompensation in men. Am J Physiol 2003; 285: E1304 –
E1311
14 Helge JW, Watt PW, Richter EA, Rennie MJ, Kiens B. Fat utilization dur-
ing exercise: adaptation to a fat-rich diet increases utilization of plas-
Chen Y et al. Effect of CHO … Int J Sports Med 2008; 29: 598 – 606
ma fatty acids and very low density lipoprotein-triacylglycerol in hu-
mans. J Appl Physiol 2001; 15: 1009 – 1020
15 Indar-Brown K, Noreberg C, Madar Z. Glycemic and insulinemic re-
sponses after ingestion of ethnic foods by NIDDM and healthy sub-
jects. Am J Clin Nutr 1992; 55: 89 – 95
16 Jenkins DJA, Wolever TMS, Taylor RH, Barker H, Fielden H, Baldwin JM,
Bowling AC, Newman HC, Jenkins AL, Goff DV. Glycaemic index of foods:
a physiological basis for carbohydrate exchange. Am J Clin Nutr 1981;
34: 362 – 366
17 Jentjens R, Cale C, Gutch C, Jeukendrup A. Effects of pre-exercise inges-
tion of differing amounts of carbohydrate on subsequent metabolism
and cycling performance. Eur J Appl Physiol 2003; 88: 444 – 452
18 McArdle WD, Katch FI, Katch VL. Exercise Physiology: Energy, Nutrition,
and Human Performance. 4th edn. Baltimore: Williams & Wilkins,
1986
19 Mikines KJ, Sonne B, Tronier B, Galbo H. Effects of acute exercise and de-
training on insulin action in trained men. J Appl Physiol 1989; 66:
704 – 711
20 Mittendorfer B, Sidossis LS. Mechanism for the increase in plasma tri-
acylglycerol concentrations after consumption of short-term, high-
carbohydrate diets. Am J Clin Nutr 2001; 73: 892 – 899
21 Salmeron J, Manson JE, Stampfer MJ, Colditz GA, Wing AL, Willett WC.
Dietary fiber, glycemic load, and risk of non-insulin-dependent diabe-
tes mellitus in women. J Am Med Assoc 1997; 277: 472 – 477
22 Schrauwen P, Wagenmakers AJ, van Marken Lichtenbelt WD, Saris WH,
Westerterp KR. Increase in fat oxidation on a high-fat diet is accompa-
nied by an increase in triglyceride-derived fatty acid oxidation. Diabe-
tes 2000; 49: 640 – 646
23 Sherman WM, Costill DL, Fink WJ, Miller JM. Effect of exercise-diet ma-
nipulation on muscle glycogen and its subsequent utilization during
performance. Int J Sports Med 1981; 2: 114 – 118
24 Sherman WM, Costill DL, Fink WJ, Hagerman FC, Armstrong LE, Murray
TF. Effect of a 42.2-km footrace and subsequent rest or exercise on
muscle glycogen and enzymes. J Appl Physiol 1983; 55: 1219 – 1224
25 Siu PM, Wong SH, Morris JG, Lam CW, Chung PK, Chung S. Effect of fre-
quency of carbohydrate feedings on recovery and subsequent endur-
ance run. Med Sci Sports Exerc 2004; 36: 315 – 323
26 Slyper A, Jurva J, Pleuss J, Hoffmann R, Gutterman D. Influence of glyce-
mic load on HDL cholesterol in youth. Am J Clin Nutr 2005; 81: 376 –
379
27 Stevenson E, Williams C, McComb G, Oram C. Improved recovery from
prolonged exercise following the consumption of low glycemic index
carbohydrate meals. Int J Sport Nutr Exerc Metab 2005; 15: 333 – 349
28 van Loon LJ, Saris WH, Kruijshoop M, Wagenmakers AJ. Maximizing
postexercise muscle glycogen synthesis: carbohydrate supplementa-
tion and the application of amino acid or protein hydrolysate mix-
tures. Am J Clin Nutr 2000; 72: 106 – 111
29 van Loon LJ, Greenhaff PL, Constantin-Teodosiu D, Saris WH, Wagen-
makers AJ. The effects of increasing exercise intensity on muscle fuel
utilisation in humans. J Physiol 2001; 536 (Pt 1): 295 – 304
30 Wee SL, Williams C, Gray S, Horabin J. Influence of high and low glyce-
mic index meals on endurance running capacity. Med Sci Sports Exerc
1999; 31: 393 – 399
31 Wee SL, Williams C, Tsintzas K, Boobis L. Ingestion of a high-glycemic
index meal increases muscle glycogen storage at rest but augments
its utilization during subsequent exercise. J Appl Physiol 2005; 99:
707 – 714
32 Williams C, Brewer J, Walker M. The effect of a high carbohydrate diet
on running performance during a 30-km treadmil time trial. Eur J Appl
Physiol 1992; 65: 18 – 24