LEADING ARTICLE Sports Med.

1997 Aug; 24 (2): 73-81

01 12-1642/97/00J8.0073/$04.50/0

© Adis International Umited. All rights reserved.

Carbohydrate-Loading and
Exercise Performance
An Update
John A. Hawley, Elske]. Schabort, Timothy D. Noakes and Steven C. Dennis
Sports Science Institute of South Africa, incorporating the Liberty Life Chair of Exercise and
Sports Science, the Medical Research Council and the University of Cape Town Bioenergetics
of Exercise Research Unit, Department of Physiology, University of Cape Town Medical School,
Cape Town, South Africa

Summary This review suggests that there is little or no effect of elevating pre-exercise
muscle glycogen contents above normal resting values on a single exhaustive
bout of high-intensity exercise lasting less than 5 minutes. Nor is there any benefit
of increasing starting muscle glycogen content on moderate-intensity running or
cycling lasting 60 to 90 minutes. In such exercise substantial quantities of glyco-
gen remain in the working muscles at the end of exercise. However, elevated
starting muscle glycogen content will postpone fatigue by ",,20% in endurance
events lasting more than 90 minutes. During this type of exercise, exhaustion
usually coincides with critically low (25 mmollkg wet weight) muscle glycogen
contents, suggesting the supply of energy from glycogen utilisation cannot be
replaced by an increased oxidation of blood glucose. Glycogen supercompensa-
tion may also improve endurance performance in which a set distance is covered
as quickly as possible. In such exercise, high carbohydrate diets have been re-
ported to improve performance by 2 to 3%.

The concept that the performance of strenuous
[>70% of maximal oxygen uptake CV0 2max )] ex-
ercise could be influenced by the preceding diet
is not new. Some 50 years ago Christensen and
Hansen[l] reported that endurance capacity was im-
paired when individuals consumed a carbo-
hydrate- (CHO) restricted diet prior to exercise, but
was improved after the consumption of a diet high
in CHO. It was in the late 1960s that Swedish in-
vestigators first described an extreme regimen in-
volving exercise to exhaustion, a period of re-
stricted CHO intake and subsequent consumption
of large quantities of dietary CHO, to increase or
'supercompensate' the stores of muscle and liver
glycogen prior to exercise.[2,3] Such a regimen was
shown to significantly elevate muscle glycogen
content and prolong submaximal (75% -V0 2max )
exercise time to exhaustion compared with individ-
uals who ingested their habitual or a low CHO
diet. [3] Since this investigation and other early stud-
ies,[4,5] Sherman et aI.l6] have proposed a so-called
'modified regimen' for glycogen supercompensa-
tion which does not require the athlete to exercise
to exhaustion or restrict CHO intake in the week
immediately prior to a competition. Now it is
widely accepted that several days of an increased
CHO intake in association with a reduction in ex-
ercise will supercompensate glycogen content in
those muscles frequently active in training.
Although CHO-Ioading was thought to benefit
74
only the performances of those athletes who par-
ticipate in prolonged, continuous activities, several
recent studies have shown decreased times to ex-
haustion during short duration «15 min), intense
(>95% \T02max) exercise tests after a low CHO diet.
Accordingly, the aim of this article is to evaluate
the role of diet and starting muscle glycogen con-
tent on performances over a range of exercise in-
tensities and duration.
1. Muscle Glycogen Stores in Humans
The glycogen content of skeletal muscle of un-
trained individuals consuming a mixed diet is typ-
ically 80 mmollkg of muscle wet weight (ww).[7]
For individuals involved in regular endurance
training and consuming a mixed diet, resting mus-
cle glycogen content is somewhat higher at ",,125
mmollkg ww muscle.[6.8.9] After several days of a
high [8 glkg body mass (BM)J CHO diet and a de-
crease in training (the so-called 'modified' regimen),
muscle glycogen content can be elevated to values
in excess of 200 mmol/kg ww. [6] As 1g of glycogen
is stored in tissue with 3 to 5g of water, [10, II] an
athlete's BM may increase by 2 to 3% after several
days of CHO-loading.l6 ]
There is some evidence that trained athletes
who habitually consume a moderate-to-high
CHO diet (26g CHO/kg BM/day) do not have the
same capacity to increase their muscle glycogen
contents as untrained individuals.[l2- 161 Indeed, if
a moderate-to-high CHO diet is consumed, muscle
glycogen 'supercompensation' qm occur on a day-
to-day basis in well trained athletes undertaking
prolonged, strenuous daily workouts. In this re-
spect, Costill et aJ.f8] have previously reported that
muscle glycogen content was not significantly dif-
ferent when trained runners consumed either 525
or 650g CHO/day, suggesting that the extent of
muscle glycogen supercompensation is not further
increased by the ingestion of very large (>600
g/day) quantities of dietary CHO.
A recent study has suggested that women do not
increase their muscle glycogen content to the same
extent as men in response to the modified regi-
men.[17] However, this is the only report of a failure
of increased dietary CHO to elevate muscle glyco-
© Adis International Limited, All rights reserved.
Hawley et ai,
gen stores. Most studies show that an additional
intake of dietary CHO results in muscle glycogen
supercompensation.
Finally, almost all of the laboratory investiga-
tions of the effects of different diets on exercise
capacity have used the time to exhaustion at a fixed
work rate as a measure of an individual's athletic
performance. It should be noted that the variability
of this measurement can often be greater than the
expected treatment effect. For example, McLellan
et aJ.f 181 have reported a high (17%) coefficient of
variation (CV) in cycling times to exhaustion at
80% \T02max in untrained individuals, while
Jeukendrup et aU19] found an even higher figure
(27%) for a similar task in well trained male
triathletes. On the other hand, when individuals are
asked to perform a time-trial where they cover a
predetermined distance as fast as possible, the CV
is reduced to "" 1%. [20] In future investigations of
the ergogenic effects of nutritional substances, it
should therefore be ensured that the variability of
the dependent variable is small enough to detect
physiologically significant differences between
treatments.
2. Effects of Exercise Intensity on
Muscle Glycogen Utilisation
The contribution of glycogen to energy metabo-
lism during exercise depends on its availability and
on the relative intensity of exercise. Fig. I shows
that as exercise intensity increases there is a more
rapid exponential rate of muscle glycogen utilisa-
tion.[21,22] During short duration (<15 min), high-
intensity (>95% \T02max) exercise, activation of
oxygen-independent glycogenolysis produces lac-
tate and decreases muscle pHJ23,24] The importance
of muscle glycogen as the main fuel for this type
of work is demonstrated by reductions in \T02max
after glycogen depletionPS,26]
During exercise performed at 80 to 85% \T02max
and lasting approximately 1 hour, muscle glycogen
remains the dominant source of energy,[9] but gly-
cogen depletion per se is not the cause of fatigue
in exercise of this intensity and duration.l 9,12] It is
during prolonged (>90 min) moderate-intensity ex-
ercise at a fixed work rate of70 to 75% \T02max that
Sports Med, 1997 Aug; 24 (2)
Muscle Glycogen and Exercise Performance
f 80
~ diet for 3 to 4 days prior to exercise impaired cy-
~ 60
E cling time to exhaustion at 100% V0 2max compared
.s
lii 40
8'
u
.a; 20
~ 83% 64%-
~ 0~~V~O~2rm~~-.~V_·O~2~ma~x~r-r-,V_OT2m=~.--
30 60 90 120
Exercise duration (min)
Fig. 1. The effect of exercise intensity on the rate of muscle
glycogen utilisation.[21] Abbreviation and symbol: ww = wet
weight; V02max= maximal oxygen uptake.
the onset of fatigue is associated with critically low
(,.,25 mmol/kg ww) levels of muscle glycogenP-5,27]
Such fatigue can occur despite the existence of eu-
glycaemia (5 mmoIlL), suggesting that there is an
absolute requirement for muscle glycogen within
the contracting muscle fibres which cannot be re-
placed by an increased uptake and oxidation of
blood glucoseJ28]
In ultra-endurance events lasting 4 to 5 hours
and typically conducted at less than 65% \,02max,
it has been estimated that CHO and fat contribute
equally to power productionJ29] However, after 6
hours of submaximal exercise the contribution to
energy production from CHO oxidation falls to be-
tween 10 and 20%J30]
3. Effect of Carbohydrate-Loading on
Exercise Capacity
3.1 High-Intensity Exercise Lasting Less than
20 Minutes
Table I summarises the reported effects of
CHO-Ioading and CHO restriction on the perfor-
mance of high intensity exercise lasting less than
20 minutes. Unfortunately, muscle glycogen con-
tents before and after exercise were determined in .
only 2 of the 7 investigations.[17,31] In the other
studies, increases or decreases in starting muscle
glycogen contents after high or low CHO diets
have to be assumed.
© Adis International limited. All rights reserved.
75
Four of the studies in table I showed that the
consumption of a Iow (<15% of total energy) CHO
with when a high (>65%) CHO diet was consumed
for the same period.[32-35] However, these diets are
extreme; athletes never consume such low
amounts of CHO either during training or prior to
competitions.£36] When the effects of a high CHO
diet on subsequent exercise capacity were com-
pared with the individual's habitual diet, there
were generally only very small differences in per-
formance.
Vandenberghe et aU31] examined the effects of
3 days' high CHO diet on muscle glycogen content
and its subsequent utilisation during 105 seconds
of cycling at 125% V0 2max in 18 moderately
trained individuals. Although starting muscle gly-
cogen content was 56% higher after CHO-Ioading,
net glycogen breakdown during supra-maximal
exercise was similar (table I). Subsequently, the
same investigators repeated the studies on 14 indi-
viduals who cycled to exhaustion at the same work
rate. Performance times were identical after nor-
mal and high CHO diets. This study confirmed the
findings of earlier workers who demonstrated that
there is little[32] or no[33-35] effect of elevating pre-
exercise muscle glycogen contents above normal
resting values on a single exhaustive bout of high
intensity exercise lasting less than 5 minutes.
Only when high intensity exercise to exhaustion
is preceded by a bout of submaximal exercise does
a CHO-Ioading regimen have some effect on sub-
sequent exercise capacity. Pizza et al.[37] examined
the effects of a CHO-Ioading regimen on running
time to exhaustion at a speed and gradient eliciting
\,02max following a l5-minute submaximal run at
75% V0 2max . In this study, 6 of the 8 participants
increased their times to exhaustion after the high
CHO diet (table I).
Tarnopolsky et aU17] compared the effects of a
CHO-Ioading regimen on exercise time to exhaus-
tion at 85% V0 2max following 1 hour of submaxi-
mal (75% V0 2max ) cycling in 7 male and 8 female
athletes (table I). Exercise time to exhaustion after
the high CHO diet was increased in 6 of the 7 men
Sports Med. 1997 Aug; 24 (2)
@
» Table I. Effect of carbohydrate-loading on high intensity exercise lasting less than 20 minutes (all values are mean ± SD) O:'!
0.
<n'
Dietary treatment Muscle glycogen Performance measure Results Reference
~ (mmol/kg wet weight)a
3
o pre-exercise post-exercise
Cf
:OJ
Q. Ma/e(M)
c
3 Group A: 4 days' HCHO (8.0 ± 0.02 g/kg BM CHO/day) 131 ± 20 31 ± 10 Cycle for 60min at 75% 18.9 ± 8.1 min (n = 7) 17
Group B: 4 days' NORM (6.4 ± 0.02) 93± 18 25 ± 6 V02m." then to exhaustion 13.0 ± 7.6 min*
a at 85% V02max
~
Fema/e(F)
cO'
~ Group A: 4 days' HCHO (6.2 ± 0.02) 95 ± 13 15 ±9 14.7 ± 14.5 min (n = 8)
Group B: 4 days' NORM (4.6 ± 0.02) 95 ± 18 18 ± 10 14.0 ± 11.3 min
~<: *B < A (p < 0.05, for M)
<D
0.
Group A: 5 days' NORM (4.5 g/kg BM CHO/day) 1A: 185 ± 17 1A: 173 ± 14 1. Cycle for 100s at 2A: 2.9 ± 1.3 min (n = 15 M, 17 F) 13
Group B: 2 days' 4.5 g/kg BM CHO/day and depletion exercise, 1B: 132 ±26 1B:121±26 125% V02max 2B: 2.9 ± 1.0 min
then 3 days' HCHO (7.5 g/kg BM CHO/day) 2. Cycle to exhaustion at
125% V02m.,

Group A: NORM (4.6 ± 0.1 g/kg BM CHO/day) Cycle to exhaustion at A: 5.1 ± 2 min* (n = 11 M) 33
Group B: 3 days' LCHO (0.8 ± 0.1) 100% V02max B: 3.7 ± 1 min
Group C: 3 days' HCHO (7.1 ± 0.3) C: 5.8 ± 3 min*
*A, C> B (p < 0.05)

Group A: NORM (5.5 ± 2 g/kg BM CHO/day) Cycle to exhaustion at A: 4.0 ± 0.5 min (n = 7 M) 34
Group B: 3 days' LCHO (0.6 ± 0.3) 100% V02max B: 3.3 ± 0.4 min
Group C: 3 days' HCHO (8.9 ± 3.0) C: 3.8 ± 0.3 min*
'B < A (p < 0.05),
*C > B (p < 0.01)

Group A: NORM (3.7 ± 0.3 g/kg BM CHO/day) Cycle to exhaustion at A: 3.5 ± 1.1 min (n = 6 M) 5
Group B: 4 days' LCHO (0.3 ± 0.1) 100% V02max B: 3.0 ± 1.1 min*'
Group C: 4 days' HCHO (6.6 ± 0.2) C: 3.7 ± 1.2 min
*B < A (p < 0.01),
'B < C (p < 0.05)

Group A: NORM (3.8 ± 0.3 g/kg BM CHO/day) Cycle to exhaustion at A: 4.9 ± 1.1 min (n = 6 M) 32
Group B: 3 days' LCHO (0.3 ± 0.04) 104 ± 5% V02max B: 3.3 ± 0.9 min'
en C: 6.6 ± 1 .4 min*
Group C: 3 days' HCHO (5.9 ± 0.1)
'8 'B < A (p < 0.005),
~
;;: *C > A (p < 0.05)
<D
0.

Group A: 6 days' NORM (4.0 ± 0.5 g/kg BM CHO/day) Run to exhaustion at A:4.7±1.1 min(n=8M) 37
~
..... Group B: 3 days' HCHO (4.5 ± 0.5), 15.5 km/h, 3.3% gradient B: 5.1 ± 1.0 min* If
»
c then 3 days' 8.2 ± 0.4 g/kg BM CHO/day *B > A (p < 0.05) 8
'fl
~ a 1 mmol/kg wet weight = 4.3 mmollkg dry weight.
§ Abbreviations: BM = body mass; CHO/day = carbohydrate per day; HCHO = high carbohydrate intake; LCHO = low carbohydrate intake; NORM = normal diet; V02max= maximal oxygen uptake. r!
Muscle Glycogen and Exercise Performance 77

., <Ii

.,a;" ""c.'"
e
by an average of 45%, whereas time to fatigue was ~
::>
only prolonged in 4 of the 8 women and this (5%) a: co 0> ;:! .,e
difference was not statistically significant. Indeed, '"x>-
0
iL "iii
when the sexes were pooled and the number of E
observations increased to 15, there was no signifi- '" ~ 'x
~
co
~ co 'E"
cant increase in the exercise time to exhaustion '"II .s"
.s"
after the high versus normal CHO diets.
In agreement with Saltin and Karlsson,[221 these
'E
e
e
'E
e
'E
.s
e
EE
e
E E
"'"d:" ""~ J
'>
~ ~ 00
"5
data suggest that the availability of muscle glyco- w +1
0>
+1 +1 '"0+1 '"....+1 +1 +1
...ci '"a 'C

gen does not limit exercise capacity at work rates

'§
.,w '"
co '"
co '"co .... .... ...... "iii
E
a: « Iii (j «Iii « Iii 0e
eliciting ~100% "V0 2max when that exercise is com- ~
II
::> 15 :2;
menced with a normal (i.e. 85 to 100 mmollkg ww) .,'"
w a:
e
0 ., 0
muscle glycogen content. Exercise times to ex- .,E ~
::>
~
E "e
l!l~
~
Z
iii
haustion at these work rates are usually between 3 "
e e
2
'"
~O N
x ,> .~ c
-0 ,- ""l!l
to 5 minutes and, with glycogen depletion rates in '"
E E 6# ",-0
e ~
.!:

the order of 10 mmol glycosyl units/kg ww/min,[ 381 .,

0
"t: ""c:i0> -0
e co :=
">- 0>
Q)
~
~~ -0
>-
there should be adequate glycogen present through-
0-
(f)
Cl.
'" o " ~
0
.c
.,w
+1
OJ
out the exercise bout. e
.,'" "
;:
.,
m
'" ...
.
co
'" .Q
E .~ co '" 0
~ '"+1
~

~ x +1
+1 '"co+1 +1 +1
m
~
+1 "
.,'w::>" 1 '"~ 0
~
co 0'" ~
... co
co I
3.2 Moderate Intensity Exercise Lasting 60 to E '"0a.
0> m
0
« Iii (j «Iii « Iii
.,;:'"
..J
"iii
90 Minutes >
c'Qj
iii
~ 8'1D ., "'l!l"
.,'" ~;:
"e.,'"
0
'"co+1
co
'"+1 ~ co 0>
'"+1~
0> ... ,",

There also appears to be no benefit from in- "S "5~

.,"" .,x
+1
m
+1
It) co
+1 +1
.... ~
0'"
e ~
-0
E
(30 0 It)
'" co 0'"
creasing starting muscle glycogen content on ei- '"
>-
'" E
::>E ~
« Iii'" «Iii «Iii
~
0 :2;_ (j 0
ther moderate intensity running or cycling lasting m a. -e
"'"
.9
60 to 90 minutes (table 11). Sherman et aI.l 61exam- 0
co ~

ined the effects oflow, moderate or high CHO diets '"

e
E'"
.~
"
on muscle glycogen utilisation during a 20.9km .,
.!!! 0
I
time-trial performance in 6 well trained runners. .,'"x
.~ °
I
;;:
They found that marked differences in pre-exercise ., '"
"0
-'=' 0;
muscle glycogen contents had no influence on per- 'iii
.,
e
a.

formances. If anything, running times were actu- £., 0 0 ~

"0
I I >-
0
ally slower when athletes started the trials with .,
"§ :2; °
:2;
0
>:
'"
.E
'"
'iii
~

-e0
higher levels of muscle glycogen. Interestingly, the "0 ID ID
I ;: '"
""C,'" '"" 0 e:- "
0
a
absolute amount of muscle glycogen left at the end E
e
"'C, I _I "0 " >-
.... .... iU'0
of the 3 runs was similar regardless of the initial
0
,...: ,...:
"#- '" '"
'" .... ~:2; "" "0
0
e
'C
'", '{/J
'w
OlD '6 0
E I

..'" °'"
>- >-
muscle glycogen content (see table II). '0" '" '" >- Io~
'"
~
"0 "0
'"
"0 E oj
The results of Sherman et aI.l61 were sub- '".,e '".,e :2; '"

6'"
ID ~
{/J

"§
sequently confirmed by Madsen et aI.l91 They re- "0
>- -= -= I
'" +1
~~ E >-
E
"
°
"0
'"
~
"'0
~~~~~%
0 0
ported that 25% higher starting muscle glycogen -e 'a;
"#- ~G ;: .c
contents did not improve treadmill run time to ex- "'" C.,
"6
~aIaZa
I I -en I
-(I) -(I) :2;
0
It)
'{/J
!£.I Q) :2;
:2; '{/J ;: ID
"
haustion at 75 to 80% "V02max , Again, the total .,
t3 E iU'0 iU'o iU'o a: iU' a: iU'
S'"
Oi

amount of glycogen utilised during the 2 runs was =

w
15
~
"0:2;"0:2;"0:2;
M[J)M[J)<O[J)
0"0
Z'"
«Iii
0-0
Z '"
« Iii
0
t::
c
'"
-..
E .~
«~Cri~o~
e:- §,eng-o,g-en a. a. E
similar, but perhaps the most important finding '"
0.0.
::> ::> ::> ::> ~
eE E e
~

:is Q) 011)0000 .0
C5::'C5eC5e .0
was that, at 'exhaustion', muscle glycogen content
was still quite high (table II).
~ is
"" "" '" «:

© Adis International Urnited. All rights reserved. Sports Med. 1997 Aug; 24 (2)
@
Table III. Effect of carbohydrate loading on prolonged exercise lasting longer than 90 minutes (all values are mean ± SO) I OJ
»
0.
"''5. Dietary treatment Muscle glycogen Performance measure Results Reference
(mmollkg wet weigh!")
cD
3 pre-exercise post-exercise
~
0
:J
Q.
Group A: NORM A:85±12 A:20±12 Cycle to exhaustion at A: 67 ± 25 min (n =22 M) 4
c 85% of a workload that
3 Group B: 1-3 days' < 0.1 glkg BM CHO/day, B: 35 ±9 B: 34 ± 25 min' (n =15 M)
~ 1.4-1.9 glkg BM fallday, 6.4-7.4 glkg BM protein/day elicits a heart rate of
p. Group C: 3 days' HCHO > 9.5 glkg BM CHO/day, C: 152±28 170 beats/min C: 102 ± 56 min' (n =19 M)
~ <0.2 glkg BM fallday, < 0.5 glkg BM protein/day ·C > A (p < 0.001),
cIT ' B < A (p < 0.01 )
::r
;;t
Group A: NORM A: 100±27 A: 11 ±5.5 Cycle to exhaustion at A: 114 ± 16 min' (n =9 M) 3
Group B: 3 days' 2.0 g/kg BM fallday, B: 33± 16 B: 5.5 ±5.5 75% V02m,x B: 57±5 min
<
<D 5.2 glkg BM protein/day
p. Group C: 3 days' HCHO (8.0 glkg BM CHO/day,
* C: 183±50 C: 22±11 C: 167 ± 53 min'
1.7 g/kg BM protein/day) ·C>A(p<O.OI), C>B(p<O.OOI)
'A> B (p < 0.001)

Group A: HCHO (± 8.1 glkg BM CHO/day) A: 194±11 60±26 Cycle for 180 min at A: 180 min (7 of 7) (n =15 M) 27
Group B: NORM B: 124±20 B: non-finishers 70% V02max B: 180 min (4 of 8); 150 min (4 of 8)"
22±8 '4 of 8 vsO of 7 (p < 0.05)
B: finishers 34 ± 30

Group A: 4 days' LCHO (10.5% CHO, 76% fat, 13.5% protein) A: 45 ± 19 A: 35±21 Run to exhaustion at A: 64 ± 16 min (n =7 M) 39
Group B: 4 days' HCHO (77% CHO, 13.5% protein, 9.5% fat) B: 112±61 B:75±19 70% V02max B: 106 ± 13 min

Group A: NORM trial 1: 4.6 ± 1.3 glkg BM CHO/day; Run to exhaustion at Atl: 119 ± 19 min (n =15 M, 15 F); 40
trial 2: 5.1 ± 1.4 70% V02max t2: 122 ± 22 min
Group B: Complex CHO trial 1: 4.6 ± 1.3 glkg BM CHO/day; Btl: 106 ±24 min;
trial 2: 7.7 ± 1.8 t2: 133 ± 46 min'
Group C: Simple CHO trial 1: 4.0 ± 0.7 !]Ikg BM CHO/day; C tl: 114 ± 16 min;
trial 2: 7.0 ± 1.2 t2: 141 ± 27 min'
'B2>B" C2>C, (p<O.OI)

Group A: 3.5 days' 6.1 glkg BM CHO/day (pasta) A: 103±49 Run to exhaustion at A: 153 ± 49 min (n =14 M) 13
Group B: 3 days' 2.4 g/kg BM CHO/day and depletion exercise, B: 130±47 75% V02max B: 169 ± 30 min
then 3.5 days' 11.2 g/kg BM CHO/day (pasta)
Group C: 3.5 days' 6.3 glkg BM CHO/day (beverage) C: 107±32 C: 139 ± 26 min
Group 0: 3 days' 2.5 glkg BM CHO/day and depletion exercise, 0: 150±44 0: 168 ± 27 min'
then 3.5 days' 11.6 g/kg BM CHO/day (beverage) '0 > C (p < 0.05)
"0
'"0
a-
~ Group A: NORM A: 100±39 A: 29±33 30km running race A: 143 ± 20 min (n =10M) 41
<D
P. Group B: HCHO - 3 days no CHO, B: 194±66 B: 105 ±72 B: 135.3 ± 18 min
then 3 days 9 glkg BM CHO/day
~ iF
"»
c Group A: NORM 30km treadmill run 42
<R 1-
trial 1: 5 ± 1 g/kg BM CHO/day A tl: 135.3 ± 14.1 min (n =12 M, 6 F); ~
'"
.. trial 2: 7d 5.4 ± 0.8 12: 135.3 ± 14.1 min
3 :-
'"
Muscle Glycogen and Exercise Performance 79

II Recently, we investigated the effect of CHO-

~
o
->
loading on muscle glycogen content and utilisation
during a I-hour cycle time-trial in highly trained
individuals. [12] Although muscle glycogen content
was significantly elevated above normal values af-
ter 3 days of a high CHO diet, CHO-Ioading had
no effect on subsequent time-trial performance (ta-
~
ble II). Nor did the ",,25% elevation of pre-exercise
<Xl
II
muscle glycogen content affect the rates of muscle
C
glycogen utilisation or total CHO and fat oxidation
E'E ii)
-><-><0 during the performance rides.[12] At the end of the
co ...... a
M (1) v
time-trials substantial quantities of glycogen were
+1 +1 C.
I':~< still present in the working muscles. Together,
~ ffl 1\
«co~ these studies[5,6,9,12] suggest that CHO-Ioading has
no benefit to performance for athletes who partic-
ipate in moderate intensity events lasting 60 to 90
minutes.

3,3 Prolonged Exercise Lasting Longer than

90 Minutes

In contrast, there is evidence to indicate that el-

evated starting muscle glycogen contents extend
COO!)O o
N ............ O)
+1 +1 +1 +1
co'"
+1 +1 endurance in events lasting ~90 min (table III). The
<Xl co
LO,....r-1.()
.q(,)C\lC\I
~'" first study to show an influence of pre-exercise diet
«coob «co on time to fatigue in prolonged exercise was con-
ducted by Christensen and Hansen.[I] They ob-
",co
"""0)"""\.0
C\lC\lT"""T"'"
+1 +1 +1 +1 "''''
+1 +1
served that a very low CHO diet impaired the en-
0000
~~~;:::
... '"
00!) durance of 3 men who exercised to exhaustion at
«coob 60 to 65% \!02max. In this study, fatigue was asso-
ciated with ketosis, a low rate of CHO oxidation
and symptoms of hypoglycaemia.[l]
The findings of Christensen and Hansen [I] were
subsequently confirmed by other groups of Scan-
dinavian physiologists in the late 1960s. Ahlborg
et aJ.l4] and Bergstrom et alP] showed that low
CHO diets reduced starting muscle glycogen con-
tents by 60 to 80% and decreased cycling times to
exhaustion by ",,50%. Conversely, high CHO diets
elevated starting muscle glycogen contents by ",,80%
and increased cycling times to exhaustion by ",,50%
(see table III). The data of Bergstrom et al.l 3] indi-
cated a particularly close relationship between ini-
tial muscle glycogen content and cycling endur-
ance at 75% \!02max (fig. 2). Above a muscle glycogen
content of ",,25 mmol/kg ww, each increase in start-
ing muscle glycogen content of ",,70 mmol/kg ww

© Adis International Limited. All rights reserved. Sports Med. 1997 Aug: 24 (2)
80 Hawley et al.

extended endurance by ===Ih. The requirement of • LowCHO

~ 300
;: o Normal
===70 mmol/kg ww/h is in accord with the findings • HighCHO
of Bosch et alP7] who found that the initial rates ~
0
E • •
of muscle glycogen utilisation in individuals cy- .sc 200
cling at 70% "VOZmax were ===50 mmol/kg ww/h and
that the muscle glycogen content at exhaustion was
CD
C)

===22 mmol/kg ww (table III).
Evidence for an important role of muscle glyco-
gen in endurance exercise also comes from studies
of the effects of high CHO diets on running times
to fatigue at 70 to 75% "VOZmax . The largest in-
creases in running endurance were found in a study
by Galbo et alp9] In that study, the individuals in-
gested either a low (10%) or a high (77%) CHO
diet. Compared with the low CHO diet, the high
CHO diet increased muscle glycogen contents by
===150% and extended running times to exhaustion
by ===66% (table III). Diets high in 'complex' or
'simple' CHO have also been shown to improve
running endurance by ===23 and 26%[40] and by ===10
and 20%[13] when compared with normal diets. In
the latter studies, the 10 and 20% longer running
times to fatigue were associated with 26 and 40%
increases in starting muscle glycogen contents re-
spectively (table III).
In addition to increasing cycling and running
times to fatigue, glycogen supercompensation may
also improve athletic performances during pro-
longed exercise in which a set distance must be
covered as quickly as possible. Karlsson and
Saltin[41] found that a high CHO diet before exer-
cise decreased the times taken to run a 30km road
race by ===6% (table III) . Interestingly, the ===94%
higher starting muscle glycogen contents did not
increase the initial running speed but allowed the
athletes to maintain their race pace for longer.
Williams et al.l4Z] reported a similar finding. They
observed that a high CHO diet before exercise in-
creased the speed over the last 5km of a 30km tread-
mill running time-trial and improved perform-
ance by ===2%. Widrick et aU43 ] also showed that a
===70% higher initial muscle glycogen content in-
creased the speed over the last 14% of a simulated
70km time-trial on an isokinetic cycle ergometer.
A maintenance of higher work rates towards the
end of the 70km time-trial improved performances
,.,u0
• ••
0.
~ 100 0
0
'0
f/)
•
.
:J 0
E
'iii
E 0
• •••
2 3 4
Endurance at 75% V02max
Fig. 2. The effects of starting muscle glycogen content on sub-
sequent exercise time to exhaustion at 75% of maximal oxygen
uptake (V02max),13] Abbreviations: ww =wet weight; CHO =carbo-
hydrate.
by 3.2 to 3.4%, irrespective of whether or not CHO
was ingested during the trial. Rauch et al.l 14] also
found that ===50% increases in starting muscle gly-
cogen contents improved prolonged cycling perform-
ances by ===3.5%, despite the ingestion of CHO dur-
ing exercise. In those studies, a high CHO diet
before 2 hours of cycling at ===75% "V02max in-
creased the distances covered in a subsequent 1-
hour time-trial from 36.7 to 38.0km (table III).
Thus the only benefit of increasing pre-exercise
muscle contents above normal values comes in ex-
ercise lasting >90 minutes. In such exercise, mus-
cle glycogen 'supercompensation' increases endur-
ance by ===20% and performance by ===2 to 3%.
Acknowledgements
The studies of carbohydrate metabolism reported in this
review and undertaken in the authors' laboratory were sup-
ported by grants from the Medical Research Council of South
Africa, the Nellie Atkinson and Harry Crossley Research
Funds of the University of Cape Town, Bromor Foods (Pty)
Limited, the South African Potato Producers Association,
and the South African Sugar Association.
References
I. Christensen EH, Hansen O. Arbeitsfahigkeit und Errichtung.
Skandinavische Archlv fUr Physiologie 1939; 8: 160-71
2. Bergstrom J, Hultman E. Muscle glycogen synthesis after exer-
cise: An enhancing factor localised to the muscle cells in man.
Nature 1966; 210: 309-10
3. Bergstrom J, Hermansen L, Hultman E, et al. Diet, muscle gly-
cogen and physical performance. Acta Physiol Scand 1967;
71: 140-50

© Adis Internafional Umlted. All rights reserved. Sports Med. 1997 Aug; 24 (2)
Muscle Glycogen and Exercise Performance
4. Ahlborg B, Bergstrom J, Brohult J, et al. Human muscle glyco-
gen content and capacity for prolonged exercise after differ-
ent diets [in Swedish]. Foersvarsmedicin 1967; 85-99
5. Hermanssen L, Hultman E, Saltin B. Muscle glycogen during
prolonged severe exercise. Acta Physiol Scand 1967; 129-39
6. Sherman WM, Costill DL, Fink WJ et al. The effect of exercise
and diet manipulation on muscle glycogen and its subsequent
use during performance. Int J Sports Med 1981; 2: 114-8
7. Bergstrom J, Hultman E, Roch-Norlund AE. Muscle glycogen
synthase in normal subjects: basal values, effect of glycogen
depletion by exercise and of a carbohydrate-rich diet follow-
ing exercise. Scand J Clin Lab Invest 1972; 29: 231-6
8. Costill DL, Sherman WM, Fink WJ et al. The role of dietary
carbohydrates in muscle glycogen resynthesis after strenuous
running. Am J Clin Nutr 1981; 34: 1831-6
9. Madsen K, Pedersen PK, Rose P, et al. Carbohydrate super-
compensation and muscle glycogen utilization during ex-
haustive running in highly trained athletes. Eur J Appl Physiol
1990; 61: 467-72
10. Olsson KE, Saltin B. Variations in total body water with muscle
glycogen changes in man following exercise. Acta Physiol
Scand 1970; 80: 11-8
II. Sherman WM, Plyley MJ, Sharp RL, et al. Muscle glycogen
storage and its relationship with water. Int J Sports Med 1982;
3: 22-4
12. Hawley JA, Palmer G, Noakes TD, et al. Effect of carbohydrate
supplementation on muscle glycogen content and subsequent
utilisation during I hr cycle time trial. Eur J Appl Physiol
1997; 75 (5): 407-12
13. Lamb DR, Snyder AC, Baur TS. Muscle glycogen loading with
a liquid carbohydrate supplement. Int J Sport Nutr 1991; 1:
52-60
14. Rauch LH, Rodger I, Wilson GR, et al. The effects of carbohy-
drate loading on muscle glycogen content and cycling per-
formance. Int J Sport Nutr 1995; 5: 25-36
15. Roedde S, MacDougall JD, Sutton JR, et al. Supercompensation
of muscle glycogen in trained and untrained subjects. Can J
Appl Sports Sci 1986; 11: 42-6
16. Tremblay A, Sevigny J, Jobin M, et al. Diet and muscle glyco-
gen in vastus lateralis of runners for the marathon. J Can
Dietetic Assoc 1980; 41: 128-35
17. Tarnopolsky MA, Atkinson SA, Phillips SM, et al. Carbohy-
drate loading and metabolism during exercise in men and
women. J Appl Physiol1995; 78: 1360-8
18. McLellan TM, Cheung SS, Jacobs I. Variability of time to ex-
haustion during submaximal exercise. Can J Appl Physiol
1995; 20: 39-51
19. Jeukendrup A, Saris WHM, Brouns F, et al. A new validated
endurance performance test. Med Sci Sports Exerc 1996; 28:
266-70
20. Palmer GS , Dennis SC, Noakes TD, et al. Assessment of the
reproducibility of performance testing on an air-braked cycle
ergometer. Int J Sports Med 1996; 4: 293-8
21. Gollnick PD. Selective glycogen depletion pattern in human
muscle fibers after exercise of varying intensity and at vary-
ing pedalling rates. J Physiol 1974; 241: 45-57
22. Saltin B, Karlsson J. Muscle glycogen utilization after work of
different intensities. In Pernow B, Saltin B, editors. Muscle
metabolism during exercise. New York: Plenum Press, 1971:
289-300
23. Cheetham ME, Boobis LH, Brooks S, et al. Human muscle meta-
bolism during sprint running. J Appl Physiol 1986; 61: 54-60
© Adls Interna~onal limited. All rights reserved.
81
24. Sahlin K. Intracellular pH and energy metabolism in skeletal
muscle. Acta Physiol Scand 1978; Suppl. 455
25. Astrand PO, Hallback I, Hedman R, et al. Blood lactates after
prolonged severe exercise. J Appl Physiol1963; 18: 619-22
26. Heigenhauser GJF, Sutton JR, Jones NL. Effect of glycogen
depletion on the ventilatory response to exercise. J Appl Phys-
io11983; 54: 470-4
27. Bosch AN, Dennis SC, Noakes TD. Influence of carbohydrate
loading on fuel substrate turnover and oxidation during pro-
longed exercise. J Appl Physiol1993; 74: 1921-7
28. Gollnick PD. Energy metabolism and prolonged exercise. In:
Lamb DR, Murray R, editors. Perspectives in exercise science
and sports medicine. Volume 1: prolonged exercise. Indianapolis:
Benchmark Press Inc., 1988: 1-42
29. Hawley JA, Hopkins WG. Aerobic glycolytic and aerobic lipo-
lytic power systems. A new paradigm with implications for
endurance and ultraendurance events. Sports Med 1995; 19
(4): 240-50
30. Edwards HT, Margaria R, Dill DB. Metabolic rate, blood sugar
and the utilization of carbohydrate. Am J PhysioI1934; 108: 203-9
31. Vandenberghe K, Hespel P, Vanden Eynde B, et al. No effect of
glycogen level on glycogen metabolism during high intensity
exercise. Med Sci Sports Exerc 1995; 27: 1278-83
32. Maughan RJ, Poole DC. The effects of a glycogen-loading reg-
imen on the capacity to perform anaerobic exercise. Eur J
Appl Physiol1981; 46: 211-9
33. GreenhaffPL, Gleeson M, Maughan RJ. The effects of dietary
manipulation on blood acid-base status and performance of
high intensity exercise. Eur J Appl Physiol 1987; 56: 331-7
34. Greenhaff PL, Gleeson M, Whiting PH, et al. Dietary compo-
sition and acid-base status: limiting factors in the perfor-
mance of maximal exercise in man? Eur J Appl Physiol1987;
56: 444-50
35. GreenhaffPL, Gleeson M, Maughan RJ. Diet-induced acidosis
and the performance of high intensity exercise in man. Eur J
Appl Physiol 1988; 57: 583-90
36. Hawley JA, Dennis SC, Lindsay FH, et al. Nutritional practices
of athletes: are they sub-optimal? J Sports Sci 1995; 13: S63-74
37. Pizza FX, Flynn MG, Duscha BD, et al. A carbohydrate loading
regimen improves high intensity, short duration exercise per-
formance. Int J Sport Nutr 1995; 5: 110-6
38. Sutton JR, Jones NL, Toews CJ. Effect of pH on muscle glycol-
ysis during exercise. Clin Sci 1981; 61: 331-8
39. Galbo H, Holst J, Christensen NJ. The effect of different diets
and of insulin on the hormonal response to prolonged exer-
cise. Acta Physiol Scand 1967; 107: 19-32
40. Brewer J, Williams C, Patton A. The influence of high carbo-
hydrate diets on endurance running performance. Eur J Appl
Physiol1988; 57: 698-706
41. Karlsson J, Saltin B. Diet, muscle glycogen, and endurance
performance. J Appl Physiol 1971; 31: 203-6
42. Williams C, Brewer J, Walker M. The effect of a high carbohy-
drate diet on running performance during a 30-km treadmill
time trial. Eur J Appl Physiol1992; 65: 18-24
43. Widrick JJ, Costill DL, Fink WJ, et al. Carbohydrate feedings
and exercise performance: effect of initial muscle glycogen
concentration. J Appl Physiol 1993; 74: 2998-3005
Correspondence and reprints: John A. Hawley, PhD, Sports
Science Institute of South Africa, P.O. Box 115, Newlands
7725, South Africa.
Email: jhawley@sports.uct.ac.za
Sports Med. 1997 Aug; 24 (2)