Potential of Intestinal Electrical Stimulation for
Obesity: A Preliminary Canine Study
Jieyun Yin,* Hui Ouyang,* and Jiande D.Z. Chen*†
YIN, JIEYUN, HUI OUYANG, AND JIANDE D.Z.
CHEN. Potential of intestinal electrical stimulation for
obesity: a preliminary canine study. Obesity. 2007;15:
Objective: The aims of this study were to investigate the
therapeutic potential of intestinal electrical stimulation
(IES) for obesity. Experiments were performed to investi-
gate the effects of IES on food intake, gastric tone, gastric
accommodation, and its possible pathway.
Research Methods and Procedures: Ten normal dogs and
six dogs with truncal vagotomy were used in this study.
Each dog was equipped with a gastric cannula for the
measurement of gastric tone and accommodation by
barostat and one pair of duodenal serosal electrodes for IES.
The experiment on food intake was composed of both
control session without IES and IES session after a 28-hour
fast. The experiment on gastric tone and accommodation
was performed in the fasting and fed states and composed of
three sessions: control, IES, and IES with NG-nitro-L-argi-
Results: IES significantly reduced food intake in the normal
dogs (459.0 vs. 312.6 grams, p ? 0.001). The food intake
was negatively correlated with the fasting gastric volume
during IES. IES significantly decreased fasting gastric tone
in the normal dogs reflected as a decrease in gastric volume
(89.1 vs. 261.3 mL, p ? 0.01), which was abolished by
vagotomy and NG-nitro-L-arginine.
Discussion: IES reduces food intake and inhibits gastric
tone in the fasting state. The inhibitory effect of IES on
gastric tone is mediated by both vagal and nitrergic path-
Key words: intestinal electrical stimulation, gastric mo-
tility, gastric accommodation, satiety, food intake
Obesity is a complex, multifactorial, and chronic condi-
tion characterized by excess body fat. Obesity results from
an imbalance between energy expenditure and caloric in-
take. It is one of the most prevalent public problems in the
United States. Severe obesity or clinically severe obesity
(BMI ? 40 or ?100 pounds over normal weight) affects
?15 million Americans (1,2). In addition, obesity is asso-
ciated with an increased prevalence of socioeconomic hard-
ship due to a higher rate of disability, early retirement, and
widespread discrimination (3,4).
Delayed satiety is one of the major symptoms in obese
patients, it increases food intake, and furthermore, it in-
creases body weight. Several studies have shown that gas-
tric distension acts as a satiety signal to inhibit food intake
(5,6) and that impaired gastric accommodation is closely
related to the symptoms of early satiety (7). Although an
earlier study reported that obese patients may have a larger
gastric capacity (8,9), not too much attention was paid to the
correlation between food intake and gastric volume or dis-
Gastric electrical stimulation (GES)1has been shown to
alter gastric motility in both humans and animals (10–13).
Recently, there is a growing interest in the therapeutic role
of GES for obesity. In animal models, GES was reported to
reduce food intake and result in significant weight loss
(14,15). Weight loss in severely obese patients was reported
in clinical studies using a pacemaker-like device: implant-
able gastric stimulator (IGS) (16). Similar to GES, intestinal
electrical stimulation (IES) was reported to alter gastroin-
testinal motility (17). However, little is known about the
Received for review June 2, 2005.
Accepted in final form November 17, 2006.
The costs of publication of this article were defrayed, in part, by the payment of page
charges. This article must, therefore, be hereby marked “advertisement” in accordance with
18 U.S.C. Section 1734 solely to indicate this fact.
*Veterans Research and Education Foundation, Veterans Administration Medical Center
and Transneuronix Inc., Oklahoma City, Oklahoma; and †Division of Gastroenterology,
University of Texas Medical Branch, Galveston, Texas.
Address correspondence to Jiande Chen, Division of Gastroenterology, Route 0632, 1108
The Strand, Room 221, Galveston, TX 77555-0632.
Copyright © 2007 NAASO
1Nonstandard abbreviations: GES, gastric electrical stimulation; IGS, implantable gastric
stimulator; IES, intestinal electrical stimulation; MDP, minimal distending pressure; L-NNA,
NG-nitro-L-arginine; NO, nitric oxide; NOS, nitric oxide synthase.
OBESITY Vol. 15 No. 5 May 20071133
therapeutic potential of IES for the treatment of obesity. In
our laboratory, IES was found to reduce fat absorption in a
rat model (18); however, it has never been reported whether
IES affects food intake. It is known that there are vagal
afferent and efferent pathways (vagovagal reflex) along the
gut. It is unknown whether this pathway is involved in the
effect of IES on gastric motility and food intake.
The aim of this study was, therefore, to investigate the
therapeutic potential of IES for obesity in dogs. Experi-
ments were designed to study the effects of IES on food
intake, gastric volume, and its mechanisms.
Research Methods and Procedures
Sixteen healthy female hound-mix dogs (18 to 24 kg)
were involved in this study. After an overnight fast, the dogs
were anesthetized with initial intravenous infusion of so-
dium thiopental (5 mg/kg; Abbott Laboratories, North Chi-
cago, IL), and maintained on IsoFlo (isoflurane 1.5%, in-
halation anesthesia; Abbott) in oxygen-nitrous oxide (1:1)
carrier gases delivered from a ventilator after endotracheal
intubation. Laparotomy was performed, and a gastric can-
nula was implanted on the anterior wall of the stomach, 10
cm proximal to the pylorus. One pair of bipolar electrodes
(A&E Medical, Farmingdale, NJ) was implanted on the
duodenal serosa 10 cm distal to the pylorus. In six of the 16
dogs, bilateral truncal vagotomy was performed at the level
of diaphragmatic hiatus using a previously established
method (19). A segment of ventral and dorsal trunks of the
vagus innervating the stomach (1 cm) was excised to pre-
vent the regeneration of these nerves. The ramifications to
the stomach were also severed to ensure that all these vagal
nerves innervating the stomach were denervated. The study
was approved by the Animal Care and Use Committee of
the Veterans Affairs Medical Center (Oklahoma City, OK).
All experiments were performed after the dogs were
completely recovered from the surgery, ?14 days after the
surgery. The vagotomized animals were studied within 30
days after the surgery.
Experiment 1: Effects of IES on Food Intake. This study
was performed in 10 dogs in the regular canine cage to
avoid possible environmental disturbances on food intake.
Before the study, each dog was given enough days to
acclimate to eat in the dog cage with a portable stimulation
device (Transneruonix, Inc., Mt. Arlington, NJ) attached to
the back. Nine of the dogs took 3 to 5 days for the accli-
mation; only one dog took 7 days. During this acclimation
period, food was given and taken away at certain times
according to the actual study protocol. After the acclimation
period, each dog was studied in two randomized sessions
(sham stimulation and stimulation) at an interval of 3 to 5
days. Twenty-nine hours before the test, the dog was given
the regular amount of food for 1 hour and then fasted until
the test. During the test, an unlimited amount of solid food
(Labdiet; PMI Nutrition International, Inc., Brentwood,
MO) and water were given to the dog for 1 hour with or
without IES. IES was initiated at the beginning of the
experiment and performed by the implanted serosal elec-
trodes connected to the portable stimulation device with a
stimulation frequency of 20 cycles/min, pulse width of 300
ms, and amplitude of 6 mA. This set of parameters was
previously known to delay gastric emptying (20). In addi-
tion, preliminary testing was performed in a few dogs before
the formal experiment to ensure that the animals could
tolerate the stimulation with the above parameters. No be-
havioral changes were noted during the testing. As a matter
of fact, a recent study showed that even with higher output
(10 mA), IES inhibited small intestinal contractions without
inducing any symptoms, and the inhibitory effect disap-
peared when the pulse amplitude was reduced to below 5
mA (21). Sham IES was the same as IES except that the
stimulator was not turned on. The amount of food consumed
by the dog during the experimental period was calculated.
Experiment 2: Effects of IES on Gastric Tone/Distension.
This experiment was performed in the same 10 dogs used in
Experiment 1 and was initiated 3 to 5 days after the first
experiment. After an overnight fast, the dog was brought to
the laboratory and maintained in a standing position with
slight restriction of movement. A polyethylene bag (CT-
BP800; H & A Mui Enterprise Inc., Mississauga, Ontario,
Canada) was positioned in the proximal stomach by the
surgically implanted gastric cannula. The bag was non-
compliant, 700 mL in maximal capacity, and attached to the
distal end of a double-lumen catheter. The procedure of
balloon insertion was tested during the surgery when the
abdomen was open to ensure accurate placement. The di-
rection of the catheter heading, the depth of the catheter
inserted, and the placement of the bag had been carefully
verified before the initiation of the study. The catheter was
connected to a computer-controlled electric gastric barostat
device (Distender Series IIR; G & J Electronics Inc., Wil-
lowdale, Ontario, Canada). The bag was briefly inflated
with 300 mL of air for unfolding and deflated completely
thereafter. After a brief rest period (5 minutes), the minimal
distending pressure (MDP) was determined by inflating the
bag in 1-mm Hg increment until a pressure at which evident
respiratory excursions were recorded and the bag volume
was equal to or larger than 30 mL. Fasting gastric volume
was recorded at an operating pressure of 2 mm Hg higher
than the MDP for 20 minutes at baseline and 20 minutes
with IES. The parameters of stimulation were the same as
Experiment 3: Mechanisms of IES. Eight normal and six
vagotomized dogs were used in this study. Each of the eight
normal dogs was studied in three randomized sessions:
IES for Obesity, Yin, Ouyang, and Chen
1134OBESITY Vol. 15 No. 5 May 2007
control, IES, and IES plus infusion of NG-nitro-L-arginine
(L-NNA; Sigma-Aldrich, St. Louis, MO), a nitric oxide
(NO) synthase (NOS) inhibitor. Gastric tone was recorded
with barostat at an operating pressure of 2 mm Hg higher
than the MDP for 20 minutes at baseline, 20 minutes with or
without IES, and 60 minutes with or without IES after a
liquid test meal (237 mL of Boost; Mead Johnson Nutri-
tionals, Evansville, IN). The L-NNA infusion session was
the same except that L-NNA was infused for 40 minutes in
the fasting state after the 20-minute baseline, and IES was
performed from the second 20-minute infusion period until
60 minutes after the meal. L-NNA (5 mg/kg) was infused at
a speed of 1.0 mL/min with an infusion pump (KD, Scien-
tific, Inc., Boston, MA) by a 20-gauge intravenous catheter
placed in the antecubital vein of the dog. In the six vagot-
omized dogs, only two sessions (control and IES) were
performed to verify the involvement of vagal pathway.
Gastric tone was assessed from the measured gastric
volume. At the constant operating pressure, an increase in
gastric volume reflects a decrease in gastric tone and vice
versa. Gastric accommodation was calculated by the differ-
ence in gastric volume between the fed and fasting states.
All data are presented as means ? standard error. Paired
Student’s t test was applied to investigate the effect of IES
on acute food intake in normal dogs. Paired Student’s t test
was also used to investigate the effects of IES on gastric
tone and gastric accommodation. Unpaired Student’s t test
was applied to investigate the difference in gastric volume
between the normal and vagotomized dogs. Spearman rank
correlation was used to investigate the correlation between
food intake and gastric volume in the normal dogs. A p
value ? 0.05 was considered statistically significant.
Effects of IES on Food Intake and Gastric Tone
IES significantly reduced food intake in the normal dogs.
As shown in Figure 1, the mean amount of food intake was
459.0 ? 42.8 g in the control session and significantly
reduced by 31.9% to 312.6 ? 40.5 g (p ? 0.001) in the IES
session. IES reduced gastric tone in the fasting state. The
gastric volume in the control session was 77.7 ? 5.4 mL
and significantly increased to 281.6 ? 30.7 mL with IES, an
increase of ?3-fold.
Food intake was found to be correlated with reduced
gastric tone or increased gastric volume. In the control
session without IES, there was a positive correlation (r ?
0.73, p ? 0.013) between food intake and fasting gastric
volume (Figure 2A). However, in the stimulation session, a
negative correlation (r ? ?0.69, p ? 0.025) was noted
between the amount of food intake and gastric volume
during stimulation (Figure 2B), suggesting that IES-induced
gastric relaxation or distension was related to the reduced
food intake with IES.
Involvement of Vagal Pathway in IES-Induced Gastric
In Experiment 3, the mean value of the MDP in the
control sessions was 3.0 ? 0.6 mm Hg in the normal dogs.
IES significantly increased gastric volume in the fasting
state. The fasting gastric volume, reflecting basal gastric
tone, was 89.3 ? 8.3 mL at baseline and 261.3 ? 37.0 mL
during IES (p ? 0.002, Figure 3). There was no difference
in the postprandial gastric volume between the control and
IES sessions (427.1 ? 25.4 vs. 431.6 ? 23.9 mL, p ? 0.7).
The mean value of MDP in the vagotomized dogs was 2.4 ?
0.5 mm Hg, which was similar to that in the normal dogs
(p ? 0.3); similarly, there was no difference in the fasting
gastric volume between the normal and vagotomized dogs
(89.3 ? 8.3 vs. 119.4 ? 17.7 mL, p ? 0.2, Figure 3).
However, the postprandial gastric volume in vagotomized
dogs was significantly lower than that in the control dogs
Figure 1: Effect of IES on food intake in normal dogs. Individual
values are shown as dots; mean and standard error of each group
are shown as bars (n ? 10). IES significantly reduced food intake
compared with the control (p ? 0.001).
IES for Obesity, Yin, Ouyang, and Chen
OBESITY Vol. 15 No. 5 May 20071135
(300.3 ? 23.6 vs. 427.1 ? 25.4 mL, p ? 0.03), indicating
a reduced gastric accommodation after vagotomy.
Vagotomy abolished the IES-induced gastric relaxation.
In the vagotomized dogs, IES failed to increase gastric
volume from its baseline level (p ? 0.2, Figure 3).
Involvement of NO in IES-Induced Gastric Relaxation
In the normal dogs, intravenous infusion of L-NNA pre-
vented the IES-induced gastric relaxation. As shown in
Figure 4, in the fasting state, the infusion of L-NNA resulted
in a slight but significant reduction in gastric volume from
115.9 ? 17.3 mL at baseline to 81.0 ? 14.7 mL after the
infusion (minutes ?40 to ?20) (p ? 0.04), suggesting an
increase in gastric tone. At the presence of L-NNA, IES
failed to increase the gastric volume (115.1 ? 18.7, p ? 1.0
vs. baseline), suggesting the involvement of the nitrergic
pathway in the IES-induced gastric relaxation.
In this present study, we have found that IES reduced
food intake and reduced gastric tone or induced gastric
distension, the reduced food intake was correlated with the
Figure 2: Correlation of food intake and gastric distention. (A)
There was a positive correlation between food intake and prepran-
dial gastric volume in the control session (r ? 0.73, p ? 0.013,
n ? 10). (B) There was a negative correlation between food intake
and preprandial gastric volume in the stimulation session (r ?
?0.69, p ? ?0.025, n ? 10).
Figure 3: IES significantly increased preprandial gastric volume in
the normal dogs (p ? 0.002, n ? 8). However, vagotomy abol-
ished the IES-induced gastric relaxation (p ? 0.05, n ? 6).
Vagotomy resulted in a significantly lower postprandial gastric
volume compared with the normal dogs (p ? 0.03).
Figure 4: Line graph of gastric volume averaged every 5 minutes
in the normal dogs. Infusion of L-NNA (total infusion length, time
?40 to time 0) resulted in a slight but significant reduction of
gastric volume in the fasting state compared with the baseline
(time, ?40 to ?20 minutes) (p ? 0.04). With the presence of
L-NNA, IES failed to increase the gastric volume during the time
of ?20 to 0 minutes, suggesting that L-NNA reversed IES-induced
gastric relaxation (p ? 0.005, n ? 8).
IES for Obesity, Yin, Ouyang, and Chen
1136OBESITY Vol. 15 No. 5 May 2007
IES-induced gastric relaxation, and the IES-induced gastric
relaxation was abolished by both vagotomy and L-NNA.
GES as a potential therapy for obesity has recently been
explored in both animals and humans (14–16,22). The first
human study using an IGS for the treatment of severe
obesity performed by Cigaina et al. (22) showed a substan-
tial and sustained weight loss. Since then, a number of
follow-up studies have been reported in the literature to
further confirm the initial results and explore possible
mechanisms of IGS (14,16,23–26). Although the exact
mechanisms of IGS in reducing weight remain largely un-
known, it has been reported that IGS induces gastric dis-
tension or reduces gastric tone, inhibits antral contractions
(25), and impairs intrinsic gastric myoelectrical activity
possibly by inhibitory sympathetic and nitrergic pathways
IES, similar to GES, has been reported for the treatment
of gastrointestinal disorders (17,28). To the best of our
knowledge, this is the first study to show that IES is capable
of reducing food intake and, thus, may have a therapeutic
potential for obesity. Our study showed a significant reduc-
tion (?31.9%) in food intake with IES and a significant
correlation between the amount of food intake and the
gastric tone. The positive correlation between the basal
gastric volume and the amount of food intake observed in
the dogs without stimulation in this study was in agreement
with a recent study in humans (29). Interestingly, however,
when IES was applied, the gastric tone was decreased, and
this IES-induced relaxation in gastric tone or increase in
gastric volume was correlated with the amount of food
intake. That is, a decrease in gastric tone with IES leads to
a decrease in food intake. It may be theorized that IES
relaxes the stomach, and this relaxation leads to gastric
distention that activates stretch receptors, which in turn
triggers a satiation signal in the central nervous system.
Apparently, future studies are needed to prove this hypoth-
Gastric accommodation, defined as the reflex-mediated
postprandial augmentation of the gastric volume, has been
associated with food intake (7,30,31). The amount of the
reflex-mediated postprandial augmentation determines the
volume of food, which can be ingested without inducing
bloating, whereas the ingestion of food exceeding this re-
flex-mediated gastric volume is believed to induce a feeling
of bloating. Impaired gastric accommodation has been as-
sociated with upper gastrointestinal symptoms including
early satiety, bloating, epigastric pain, weight loss, etc.
(7,32,33). Tack et al. have reported that 40% of dyspeptic
patients have impaired accommodation, and this is associ-
ated with early satiety and weight loss. The presence of
early satiety as a relevant or severe symptom is a good
predictor of impaired accommodation (7). In our study, we
have noticed that the gastric tone was decreased with IES in
the fasting state but not in the postprandial state. That is, the
postprandial gastric volume remained unchanged with or
without IES. Accordingly, an increase in gastric volume in
the fasting state with IES was equivalent to a decrease in
To understand the mechanism underlying the IES-in-
duced gastric relaxation or impaired gastric accommoda-
tion, an NOS inhibitor, L-NNA, was used in this study to
investigate the possible involvement of the nitrergic path-
way. NO has been postulated as a mediator of the non-
adrenergic non-cholinergic neural inhibition of the gastro-
intestinal smooth muscle (34). In humans, the inhibition of
NOS was found to dose dependently inhibit gastric accom-
modation to a meal and enhance meal-induced satiety (35).
In this study, the IES-induced gastric relaxation was com-
pletely blocked by L-NNA, suggesting the involvement of
the nitrergic pathway. Moreover, this nitrergic-mediated
relaxation of the stomach with IES was also associated with
the vagal pathway because the effect was also blocked by
vagotomy. This suggests a vagovagal reflex with IES, sim-
ilar to other types of stimulation, such as meal ingestion or
nutrient stimulation (36,37). The vagovagal reflex pathway
requires activation of intrinsic nitrergic neurons in the stom-
ach (38); NO is released from the non-adrenergic non-
cholinergic nerve, resulting in the relaxation of the stomach.
A number of technical issues should be considered before
this method is applied in clinical patients. Firstly, the place-
ment of stimulation electrodes could be more challenge for
IES than for GES because the intestinal wall is thinner. A
small lead is necessary for IES. Secondly, there is a need for
a future study to determine the best location for implanting
the stimulation lead both functionally and surgically.
Thirdly, caution has to be made with long pulse stimulation.
Deterioration of electrodes may occur if stimuli are not
charge-balanced. Fourthly, a special algorithm or circuitry
may be needed to automatically detect food intake and to
turn on stimulation only during a certain postprandial pe-
riod. Apparently, more studies are needed before the
method of IES could be applied in clinical practice.
In conclusion, IES results in a significant reduction in
food intake and a significant increase in gastric volume in
the fasting but not fed state in dogs. The reduction in food
intake is correlated with the IES-induced gastric relaxation,
have a therapeutic potential for the treatment of obesity.
This work was partially supported by NIH Grant 1 R43-
1. Kuczmarski RJ, Flegal KM, Campbell SM, et al. Increasing
prevalence of overweight among US adults: the National
Health and Nutrition Examination Surveys, 1960 to 1991.
IES for Obesity, Yin, Ouyang, and Chen
OBESITY Vol. 15 No. 5 May 20071137