Food Sci. Technol. Res.

, 10 (4), 428-431, 2004

A Sports Drink Based on Highly Branched Cyclic Dextrin Generates Few

Gastrointestinal Disorders in Untrained Men during Bicycle Exercise

Hiroshi TAKnl, Takashi KOMETANll, Takahisa NISHIMURA1 Takashi KURIK11 and Tohru FUSHIKI

IBiochemical Research Laboratory, Ezaki Glico Co., Ltd., 4-6-5, Utajima, Nishiyodogawa-ku, Osaka, Japan 555-8502
2Laboratory ofNutrition Chemistry, Division of Food Science and Biotechnology, Graduate School ofAgriculture, Kyoto University,
Kitashirakawa Oiwake-cho, Sakyo-ku, Kyoto, Japan 606-8502

Received April 5, 2004; Accepted August 21, 2004

Gastrointestinal disorders after ingesting a sports drink were investigated during bicycle exercise. The
experiment consisted of a preliminary exercise, a 10 min rest, and 30 min of exercise. Seven healthy untrained
volunteers ingested either water or a sports drink based on highly branched cyclic dextrin (HBCD), com-
mercially available dextrin of DE16 or glucose immediately after the preliminary exercise. The mean gastric
emptying time after ingestion of the HBCD-based sports drink was significantly faster than that of the glucose-
based drink. Gastrointestinal disorders were monitored by a questionnaire. The mean degree of subjective
flatulence and the mean number of belches were few with the HBCD-based drink during exercise. When vol-
unteers drank the sports drink based on 10% HBCD during bicycle exercise, they experienced few gastroin-
testinal disorders and consequently could continue the exercise comfortably with little fatigue.

Keywords: highly branched cyclic dextrin, glucose, sports drink, gastrointestinal disorder, gastric emptying time

Introduction reported that a 10% highly branched cyclic dextrin

Fluid balance plays a key role in maintaining optimal
bodily functions, particularly when an activity that induces
large fluid loss is performed. Fluid loss of as little as 2%
of body weight decreases endurance performance (Craig
& Cummings, 1966; Armstrong et al., 1985). With a water
10ss of 4-5~;~o of body weight, physiologic function is
impaired (Buskirk et al., 1958; Ribish & Herbert, 1970).
The water lost as sweat comes from the blood, and intra-
cellular and interstitial pools (Costill & Sparks, 1973;
Costill et al., 1974), and causes a reduction in circulato-
ry capacity, cooling, and exercise efficiency (Astrand &
Saltin, 1964). Therefore, drinking fluids is advised to
achieve optimal performance and feelings of well-being
during prolonged exercise.
Participants in endurance events frequently suffer from
gastrointestinal (GI) disorders such as belching, flatulence,
heartburn, nausea, abdominal pain, Iateroabdominal
stitches, regurgitation, an urge to defecate, and diarrhea.
These symptoms suggest that exercise may influence GI
function. The type of exercise and the type of food ingested
prior to and during exercise influence GI disorders during
exercise (Brouns et al., 1987; Rehrer et al., 1992). During
strenuous physical exercise, sports drinks are the major
energy source consumed. Liquids are also necessary for
rehydration. Therefore, it is important to investigate the
relation between the type of drink consumed and its gastric
emptying time (GET) during exercise. We previously
Abbreviations: HBCD, highly branched cyclic dextrin; GI, gastroin-
testinal; DE, dextrose equivalent; DE16, commercially available dextrin
of a dextrose equivalent 16; CHO, carbohydrate; GET, gastric emptying
time; RPE, rate of perceived exertion
E-mail: takii-hiroshi@glico.co.jp
(HBCD)-based sports drink was significantly (p < 0.05)
transferred from the stomach to the small intestine faster
than 10% commercially available dextrin of a dextrose
equivalent 1 6 (DE16)-based sports drink, or any other 10%
carbohydrate (CHO) solutions, such as glucose, maltose,
or sucrose at rest (Takii et al., In press).
The purpose of this study was to clarify the GET of
HBCD-based sports drink in comparison with other CHO-
based sports drink during exercise. The effects of ingest-
ing the HBCD-based sports drink on GI disorders and
fatigue during exercise were also investigated.
Materials and Methods
Subjects The subjects were 7 untrained men. Their
mean :!: SD values for age, body weight, and height were
34.3 i 2.8 year, 65.4 d: 9.0 kg, and 171.7 i 5.8 cm, respec-
tively. A11 procedures were performed according to the
Helsinki Declaration of 1964 (revised in 1989).
Sports drinks used as testfluids Water and three pro-
totype sports drinks based on HBCD, DE16, and glucose
were used as test fluids in the experiments. The sports
drinks were composed of 10% CHO, I .4% fructose, 0.01 %
stevia, 0.2% ascorbic acid, O. 1% sodium ascorbic acidate,
0.04% vitamin mixture, 0.02% sodium chloride, and 0.01 %
aroma. The osmolality of the test drinks was determined
by freezing-point depression (Advanced Wide-Range
Osmometer 3W2, Advanced Instrument Co., Ltd., MA,
USA).
Exercise protocol Experiments were started at 14:OO
each day, with several days of rest between trials. All
subjects consumed udon, flour noodles in dried-bonito
soup (250 kcal), 3 h before the onset of the test. A11 subjects
A Sports Drink Based on Highly Branched Cyclic Dextrin
were allowed to consume only a cup of water or green tea
up to 1.5 h before the onset of the test. The exercise room
was kept at 25'C, and conversation was allowed during
the experiment. However, no information on the test drink
was given to the subjects. Each subject completed a stan-
dardized warm-up, consisting of 3 min of cycling at each
of 3 Ioads (56, 71, and 85 W) on a cycling ergometer (Lode,
Groningen, The Netherlands). After the subject consumed
a given fluid as quickly as possible (within I min), he was
allowed to rest for 10 min. Exercise was then continued
for 30 min, consisting of three 10-min blocks at each of 3
10ads (71, 85 and 99 W). The subjects maintained an
approximate pedaling rate of 60 revolutions/min. Every 5
min after drinking the test fluid, the subject was asked to
assess feelings of flatulence; the feeling of remaining fluid
was expressed using a 6-point scale: 5, 100% of test fluid
remaining; 4, 80(~c; 3, 60%; 2, 40%; l, 20%; O, no fluid
present. The rate of perceived exertion (RPE) was expressed
using a 15-point scale: for example, 6, sedentary; 7, very
easy; 11, easy; 13, somewhat severe; 15, severe; 17, very
severe; 20, all out (Noble et al., 1983). The number of
belches after the test fluid was consumed was also recorded.
Measurement of the gastric emptying time of the test
solution The relaxed cross-sectional area of the pylorus
antrum was measured by an extracorporeal ultrasonic echo-
image analyzer (SSA-200A, Toshiba Co., Tokyo, Japan)
in a sitting position every 10 min after the test fluid had
been consumed (Nagao et al., 1998). During exercise, this
area was measured after I min of rest to give the stomach
time to stop moving. The gastric emptying characteristics
of the test solutions are shown as the mean GET derived
from gross half-gastric volumes. The measurement was con-
ducted three times in succession, and the average value
was calculated (Takii et al., In press).
Statistics Results are expressed as mean i SEM. One-
way analysis of variance (ANOVA) with Fisher's PLSD
post hoc test was used to assess the statistical significance
1 oo
75
~
1)
(1:
~ 50
~
~ Drink
25
o
-20 -10 O I O 20 30 40
f Time (min)
Drinking
~ ; Blood sampling
Fig. 1. Protocol of the exercise test.
429
of differences, with a p value of <0.05 used to indicate
significance.
Results
Gastric emptying time and GI disorders during
exercise The osmolality of the HBCD-, DE16-, and
glucose-based sports drink was 150, 269, and 787 mOsm,
respectively. The GET was shortest when subjects ingested
the HBCD-based sports drink (Fig. 2). The HBCD-based
sports drink emptied faster than the glucose-based sports
drinks during exercise (Fig. 2). The HBCD-based sports
drink was also associated with a lower flatulence value
than the glucose- and DE16-based sports drinks from 5
min after the start of exercise (15 min after ingestion) to
the end of exercise (Fig. 3). The total number of belches
for all subjects after drinking water or the HBCD-,
glucose-, or DE16-based sports drinks to the completion
of exercise was 1 1, 9, 23 and 24, respectively. Thus, the
HBCD-based sports drink induced the fewest belches. The
mean number of belches per subject with the HBCD-based
sports drink was significantly lower (p < 0.05) than with
the glucose- or DEI 6-based sports drinks (Fig. 4). Although
no significant difference was observed in the RPE between
water and the sports drinks, the RPE with the HBCD-based
sports drink tended to be lower in the later stage of exercise
(data not shown). Postprandial blood glucose increased with
the sports drinks containing CHO, but decreased with
water (Fig. 5). There was no difference in the increase in
postprandial blood glucose among the three sports drinks
containing CHOs.
40
*
c: 30
E
~o
E
~o)
C 20
'~>
~~
E
o
.~
~*co(Q IC
O
o
Fig. 2. Gastric emptying time of the CHO-based sports drinks. A, Water;
B, HBCD; C, DE16; D, Glucose. Data represent the means :!: SEM for
7 subjects.*: significant difference (p < 0.05).
Discussion
We previously reported that a HBCD-based sports drink
was transferred from the stomach to the small intestine
faster than a 10% DE16-, or glucose-based sports drink
and 10% CHO (glucose, maltose, and sucrose) solutions
430 H.TAKII et al.

5 40

CI Water
l HBCD
4- b
1-
A DE16 T
b
O Glucose ~ 20 -
:5
~o)E
(D
~b l
-O
O
CO 3 l~ b( Tb ~o
co
o
o
T
(D
O acl C ~o)
l
C J-
(D

iT l b 1~
T• - ~
::
t*Ca 2- T o O-
~2 ~
D: ac
:~: ib Tb
oO
<l
__. l i
l ar C bc C:} Water ; T•- T.
i aL
1- L ' HBCD !
i
a
ac -2 O
A DE16 ;
li
a i O Glucose
~
l-
-1 O OI O Time (min) 20 40
Fig. 3. Change in flatulence after drinking CHO-based sports drinks.
30
20 -1 O I O
20 40
30 O
Time (min)

Different superscript letters show significant difference (p < 0.05) among Fig. 5. Change in blood glucose level after drinking CHO-based sports
test drinks at the same period after ingestion. drinks. *: significantly different between water and glucose-based drink
(p < O.05). **: significantly different among test drinks (p < 0.05).

5
that subjects felt little fatigue after ingesting the HBCD-
based sports drink.
As noted above, it is reasonable to investigate the rela-
~o 4(D
tionship between the type of drink and its gastric emptying
~i co rate to control GI disorders during exercise. From previous
~o) ab experimental evidence, it may be expected that the inci-
c3
co a)
dence of GI disorders will increase if liquid remains in the
~~ stomach for a long time.
~o In this study, we were interested in investigating GI dis-
co 2 orders in untrained men who ingested several sports drinks
o
E
~ c:(o
just before low-intensity exercise.
(D GI disorders are induced by altered GI blood flow,
~1 changes in gut permeability, disturbed GI-tract motility,
psychological effects ( 'stress') and pharmacological agents
(Green, 1992). One of the most common theories includes
o an exercise-induced reduction in blood flow. The HBCD-
B c D based sports drink was associated with the shortest GET
A
Drink (Fig. 2) and the least flatulence (Fig. 3) among the test
drinks.
Fig. 4. Number of belches after drinking CHO-based sports drinks. A,
Water; B, HBCD; C, DEI 6; D, Glucose. Different superscript letters mean Postprandial blood glucose increased with the ingestion
significant difference (p < O.05). of the sports drinks containing CHO, but decreased with
the ingestion of water (Fig. 5). There were no obvious dif-
at rest (Takii et al., In press). Therefore, we additionally ferences in the increase in postprandial blood glucose
investigated the GET of a HBCD-based sports drink and levels among the three sports drinks. These results sug-
the effects of ingesting such a drink on GI disorders during gested that HBCD was digested and absorbed easily as well
exercise. The HBCD-based sports drink tended to be trans- as other CHOS (Fig. 5). Therefore, with the HBCD-based
ferred faster than glucose- and dextrin-based sports drink sports drink, the decrease in the volume of blood due to
during exercise (Fig. 2). Moreover, GI disorders, such as perspiration during exercise was thought to be reduced.
flatulence and the mean number of belches, after ingest- This result should lead to a reduction in GI disorders.
ing the drink, were significantly (p < 0.05) attenuated Moreover, if liquid stays in the stomach for a long time,
compared with other drinks (Figs. 3, 4). In addition, RPE the stomach itself could swing with a large amplitude. This
in the later stage of exercise tended to be lower with the leads to stretching of ligaments in the abdominal area. If
HBCD-based sports drink (data not shown), which suggests there is a large amount of residual liquid in the stomach,
A Sports Drink Based on Highly Branched Cychc Dextrin
belching and lateroabdominal stitches will be easily gen-
erated (Peters et al., 1993; Plunkett & Hopkins, 1999). In
this experiment, the fewest number of belches was observed
with the HBCD-based drink (Fig. 4). These results suggest
that subjects who ingested a HBCD-based sports drink
might have experienced few GI disorders because the
HBCD-based sports drink was transported from the
stomach faster than other drinks.
The present results regarding GET during exercise coin-
cided with those at rest, as reported previously (Takii et
al., In press). HBCD is a polymer of 900 glucose residues
with an average molecular mass of 160,000 Da. HBCD
contains few glucose, di-, and oligo-saccharides (DE Iess
than 2) (Takata et al., 1996, 1997), since it has a very narrow
molecular distribution. The osmotic pressure of I O% HBCD
solution is only 9 mOsm. Therefore, it can be used to
produce a sports drink without greatly increasing the
osmotic pressure. Since the GET of an ingested drink is
mainly influenced by its osmotic pressure, we prepared a
sports drink using HBCD. The results of this study suggest
that the minimal GI disorders in subjects who ingest a
HBCD-based sports drink might due to its short GET.
We found that the HBCD-based sports drink had a short
GET and scarcely induced any GI disorder. When consumed
during exercise, the number of belches was significantly
low (Fig. 4). A shorter GET (Fig. 2) and a lower flatulence
score (Fig. 3) were also observed. Furthermore, the subjects
showed a lower RPE during exercise (data not shown). We
previously reported that HBCD could enhance the
endurance capacity of mice (Takii et al., 1999). Our present
results suggest that HBCD might be an excellent ingredi-
ent as a sports ergogenic aid. Further investigations are
needed, since an exercise performance test was not per-
formed in athletes in this study.
In conclusion, the present results suggest that, similar
to the previous results at rest (Takii et al., In press), an
HBCD-based sports drink generated few GI disorders in
untrained men during exercise due to its short GET.
Acknowledgements We thank all of the subjects who participated in
the experiments. We also thank Dr. Shigetaka Okada for his kind
advice and Dr. Takashi Nakae for his technical assistance.
Ref erences
Armstrong, L.E., Costill, D.L, and Fink, W.J. (1985). Influence of
diuretic-induced dehydration on competitive running performance.
43 1
Med. Sci. Sports Exerc. , 17, 456-461.
Astrand, P.O. and Saltin, B. (1964). Plasma and red cell volume after
prolonged severe exercise. J. Appl. Physiol., 19, 829-832.
Brouns, F., Saris, W.H.M. and Rehrer. N.J. (1987). Abdominal com-
plaints and gastrointestinal function during long-lasting exercise.
Int. J. Sports Med., 8, 175-189.
Buskirk, T., Iampietro, P.F. and Bass, D.E. ( 1 958). Work performance
after dehydration: effects of physical conditioning and heat acclima-
tization. J. Appl. Physiol., 12, 189-194.
Costill, D.L. and Sparks, K.E. (1973). Rapid fluid replacement fol-
lowing thermal dehydration. J. Appl. Physiol. , 34, 299-303.
Costill, D.L., Baranam, G., Eddy, D. and Fink, W.J. ( 1974). Alternation
in red cell volume following exercise and dehydration. J. Appl.
Physiol., 37, 912-916.
Craig, F.N. and Cummings, E.G. (1966). Dehydration and muscular
work. J. Appl. Physiol., 34, 299-303.
Green, G. (1992). Gastrointestinal disorders in the athlete. Clin.
Sports Med., 11, 453-470.
Nagao, Y., Kodama, H., Yonezawa, T., Taguchi, A., Fujino, S.,
Nakahara, K., Haruma, K. and Fushiki, T. ( 1998). Correlation
between the drinkability of beer and gastric emptying. Biosci.
Biotechnol. Biochem., 62, 846-851 .
Noble, B.J., Borg, G.A, Jacobs, I., Ceci, R. and Kaiser, P. (1983). A
category-ratio perceived exertion scale: relationship to blood and
muscle lactates and heart rate. Med. Sci. Sports Exerc. , 15, 523-528.
Peters, H.P., van Schelven, F.W., Verstappen, P.A., de Boer, R.W.,
Bol, E., Rrich, W.B., van der Togt, C.R. and de Vries, W.R. (1 993).
Gastrointestinal problems as a function of carbohydrate and mode
of exercise. Med. Sci. Sports Exerc., 25, 121 1-1224.
Plunkett, B.T. and Hopkins, W.G., (1999). Investigation of the side
pam "stitch" induced by running after fluid ingestion. Med. Sci.
Sports Exerc., 31, 1 169-1 175.
Rehrer, N.J., van Kemenade, M., Meester, W., Brouns, F. and Saris,
W.H.M. (1992). Gastrointestinal complaints in relation to dietary
intake in triathletes. Int. J. Sports Nutr., 2, 48-59.
Ribish, P.M. and Herbert, W.G. ( 1 970). Effects of rapid weight reduc-
tion and subsequent rehydration upon the physical working capacity
of wrestlers. Res. Q., 41, 536-541.
Takata, H., Takaha, T., Okada, S., Hizukuri, S., Takagi, M. and
Imanaka, T. (1996). Structure of cyclic glucan produced from
amylopectin by Bacillus stearothemophilus branching enzyme.
Carbohydr Res., 295, 91-101.
Takata, H., Takaha, T., Nakamura, H., Fujii, K., Okada. S., Takagi,
M, and Imanaka, T. (1997). Production and some properties of a
dextrin with a narrow size distribution by cyclization reaction of
branching enzyme. J. Ferment. Bioeng., 84, 1 19-123.
Takii, H., Ishihara, K., Kometani, T., Okada, S, and Fushiki, T.
( 1999). Enhancement of swimming endurance in mice by highly
branched cyclic dextrin. Biosci. Biotechnol. Biochem., 63,
2045-2052.
Takii, H., Nagao-Takii, Y., Kometani, T., Nishimura, T., Nakae, T.,
Kuriki, T, and Fushiki, T. Fluids containing a highly branched cyclic
dextrin influence the gastric emptying rate. Int. J. Sports Med. ,
25, Inpress.