Neuromodulation: Technology at the Neural Interface Received: September 22, 2013
Revised: February 13, 2014
Accepted: March 24, 2014
(onlinelibrary.wiley.com) DOI: 10.1111/ner.12207
The Effects of Individualized Gastric Electrical Stimulation on Food Craving and Gastrointestinal Peptides in Dogs Xiaojuan Guo, PhD; Yanmei Li, MD; Shukun Yao, MD; Shaoxuan Chen, MD; Yuhui Du, DEng; Zhihua Wang, DEng Background: Using an adjustable stimulator with a wide range of stimulation parameters, the aims of this study were 1) to investigate the effects of long-term gastric electrical stimulation (GES) on appetite and differential food cravings for three different foods and 2) to investigate the effects of GES on plasma gastrointestinal peptide concentrations. Methods: The study was performed in eight Beagle dogs implanted with one pair of serosal electrodes. They were followed during GES and sham GES sessions in a crossover design. GES was conducted using a series of individualized parameters. Food intake and food cravings were observed to evaluate the effects of long-term GES. Enzyme-linked immunosorbent assay was used to measure the plasma concentrations of gastrointestinal peptides. Results: Dogs on GES for three months ate significantly less food than those on sham GES for three months (p < 0.05). A significant change in food cravings was induced by GES. Dogs with GES ate significantly less high-fat food. However, there was no significant difference in consumption of high-carbohydrate food or balanced food between the periods of GES and sham GES. The plasma concentrations of ghrelin, peptide YY3-36, and glucagon-like peptide 1 did not differ significantly between the periods of GES and sham GES. Conclusions: Food intake and food craving were changed significantly by adjustable GES. GES may be used for treating obesity by changing food preferences. Further clinical studies are necessary to highlight the effect of adjustable GES on eating behavior. Keywords: Food craving, gastric electrical stimulation, individual parameters, mechanism, obesity Conflict of Interest: The authors reported no conflict of interest.
INTRODUCTION
www.neuromodulationjournal.com
Address correspondence to: Shukun Yao, MD, Department of Gastroenterology, China–Japan Friendship Hospital, 2 Yinghua East Road, Chaoyang District, Beijing, 100029, China. Email: [email protected] Department of Gastroenterology, China–Japan Friendship Hospital, Beijing, China For more information on author guidelines, an explanation of our peer review process, and conflict of interest informed consent policies, please go to http:// www.wiley.com/bw/submit.asp?ref=1094-7159&site=1 Sources of financial support: The National Natural Science Foundation of China (Grant No. 81070299) and the Capital Medical Development Foundation of China (Grant No. 2009-2018) provided funding for the study.
© 2014 International Neuromodulation Society
Neuromodulation 2014; 17: 483–489
483
Popular attention has been attracted by the overwhelming increase in the prevalence of obesity in recent years. The World Health Organization has estimated that by 2015 about 2.3 billion adults (aged 15+) will be overweight and more than 700 million will be obese. Nowadays at least 2.8 million people worldwide die each year as a result of being overweight or obese (1). Gastric electrical stimulation (GES) has been investigated for the treatment of obesity since 1995. Electrical stimulation is delivered to the stomach via an implanted system that consists of a stimulator and two leads. An external programmer delivers instructions to the stimulator. Possible advantages to GES include less invasiveness than some bariatric surgeries (e.g., gastric bypass) and fewer side effects (e.g., dumping syndrome) (2). Although GES is known to reduce food intake and produce weight loss in animal and human studies, the commercial implantable devices used in clinical studies have mainly been adapted from nerve stimulators, not specifically developed for GES (3,4). These devices have been restricted to the following typical parameters: a pulse width of 0.3 msec, an amplitude of 5–10 mA, a frequency of 40 Hz, and 2 sec on, 3 sec off. Recently, individualized parameters for GES have been proposed for treating obesity. Parameter selection and resistance (i.e., diminution in the response to GES after longterm use) during GES must be considered (5). Fixed parameters are
insufficient, and in this study an adjustable stimulator with a wide range of stimulation parameters was used. This stimulator features a pulse width of 0.3–10 msec, an amplitude of 1–10 mA, a frequency of 40 Hz, and 2 sec on, 3 sec off. Short pulses for GES may not be potent enough, and experiments have proved the importance of pulse width in the stimulation process. The mechanisms by which GES affects food intake and weight are not fully elucidated. The mechanical, neural, and hormonal pathways have close relations with GES in the regulation of energy balance (6–8). A number of studies have shown that GES is effective in reducing appetite and inducing early satiety (8–10). Gastrointestinal peptides such as ghrelin, peptide YY (PYY), and glucagon-like peptide 1 (GLP-1) are important in regulation of food intake (11).
GUO ET AL. Previous studies have shown that GES alters the level of selected satiety-related peptides and the expression of neuropeptides in neurons of the hypothalamus (7). We hypothesized that modulation of the secretion of ghrelin, PYY3-36, and GLP-1 may be one mechanism by which adjustable stimulation reduces appetite and food intake. The aims of this study using adjustable parameters were the following: 1) to investigate the effects of long-term GES on appetite and a possible difference in food craving among three different foods and 2) to investigate the effect of GES on plasma gastrointestinal peptide concentrations.
METHODS AND PROCEDURES Surgical Procedure Eight healthy female Beagle dogs (Academy of Military Medical Sciences, Beijing, China; 8.5–13 kg) were involved in this study and allowed an acclimation period of two weeks. After an overnight fast, the dogs were anesthetized with a combination of fentanyl (2 μg/ kg), propofol (2 mg/kg), and rocuronium (0.6 mg/kg) and maintained on 1–2% sevoflurane in oxygen (1.0 L/min) carrier gases delivered from a ventilator after endotracheal intubation. Vital signs (respiratory rate, pulse, and tongue color) were monitored. By laparoscope, one pair of platinum–iridium electrodes (Noted Technology Development Co. Ltd., Shenzhen, China) was sutured into the seromuscular layer along the greater curvature of the stomach, 3.0 cm above the pylorus. The electrodes were 1.0 cm apart. An adjustable electrical stimulator was embedded under the costal margin of the right upper quadrant subcutaneously and then secured in place by a purse-string suture. After surgery, dogs wore special jackets and were allowed to recover in their individual cages. The dogs received sufficient solid food at a set time (5 PM to 7 PM daily) and had free access to water in their home cages at all times. The postoperative situation, including swelling, secretions, ulcer, and subcutaneous transposition, wound healing, and exposed material, was recorded. All experiments were performed after the dogs were completely recovered, usually two weeks after the surgery. The Animal Care and Use Committee of the China–Japan Friendship Hospital approved the study.
484
Individual Parameters An external programmer delivered periodic rectangular pulses during the experiment. After being completely recovered from the surgery, each dog underwent a pulse width selection period for two weeks, and simultaneously, the other parameters were fixed relative to the pulse width. The pulse widths were increased progressively from 0.3 msec. The maximum pulse width was 10 msec. If the intolerable pulse width was ≤1.0 msec, a lower amplitude was chosen (3 mA, 6 mA, 9 mA). The selection of stimulation parameters was based on the total symptom score. Evaluation of the dog’s symptoms involved observing for licking with the tongue, sialorrhea, vomiting, yawning, barking, groaning, belching, murmuring, and dysphoria. These symptoms were assessed based on their frequency and/or severity (0: never; 1: seldom/mild; 2: often/moderate; 3: continue/ severe). Intolerable parameters were recognized when the total symptom score was ≥3. The maximum pulse widths that induced no or very mild symptoms were selected (i.e., score 500 g balanced food at a set time slot. To determine their food cravings, all dogs were subjected to a food preference test for three days continuously. This test was performed in their individual cages, fitted with three feeders that had the exact same shape. Each day at 5 PM to 7 PM, the three feeders were filled with three different foods (500 g each): BD, CD, and FD. The position of each diet was randomly determined each day. The animals had two hours to eat, after which the remainders in each compartment were weighed to calculate the consumed amount of each food.
Gastrointestinal Peptides Blood was collected from three months after initiation of GES and three months after initiation of sham GES. An indwelling catheter was inserted into a foreleg vein to collect two blood samples (3 mL each) during the three-hour course of the experiment. The first blood sample (preprandial level) was drawn in a fasting state (30 min after GES/sham GES). Then, after GES for one hour, the animals were fed for 30 min. The second blood sample (postprandial level) was drawn one hour after feeding. Blood samples were collected in chilled tubes containing aprotinin (500 IU/mL) and EDTA (1 mg/mL). Plasma ghrelin and PYY3-36 were quantified using enzyme-linked immunosorbent assay (ELISA) kits specifically for canine proteins (Phoenix Pharmaceuticals, Belmont, CA, USA). Total plasma GLP-1 was measured using an ELISA kit (Millipore Research, St. Charles, MO, USA). The assay was carried out according to the instructions provided by the manufacturer.
Experimental Protocol The experiment followed a crossover design. After the periods of recovery from surgery (two weeks) and parameter selection (two weeks), the dogs were randomly studied in two groups. In group A, four dogs underwent a three-month treatment with GES, a twoweek washout period, and a three-month control with sham GES sequentially. During the experiment, GES or sham GES was administered one hour before food intake and then continued for three hours. The other four dogs, in group B, underwent the experiment in the reverse order. At the end of the three months of GES and sham
© 2014 International Neuromodulation Society
Neuromodulation 2014; 17: 483–489
GASTRIC ELECTRICAL STIMULATION IN DOGS
Table 1. Effective Stimulation Energy in Eight Dogs (s·mA2).
Initial effective energy Maximum effective energy
Median
First quartile
Third quartile
90th percentile
95th percentile
15 336
7.3 216
35.3 384
62.4 672
79.2 672
Both the initial effective energy and the maximum effective stimulation energy varied greatly among dogs.
GES, all dogs were subjected to a blood test and a two-hour food cravings test three days in a row. During these tests, GES or sham GES was continued as before. The procedures for sham GES were the same as the GES procedures except that the stimulators were not turned on.
Statistical Analysis Food intake data and gastrointestinal peptide levels are presented as mean ± SD. As the data for stimulation energy were not normally distributed, results are reported as median and first and third quartiles and 90th and 95th percentiles given. Differences in food intake and peptides across GES and sham GES were analyzed by crossover ANOVA. Differences in food intake and peptides within groups were evaluated using paired Student’s t-test or a nonparametric test, as appropriate. The level of statistical significance was set at p < 0.05.
RESULTS All dogs completed the study, and no adverse reactions were observed after surgery. None of the dogs was observed to display severely abnormal behaviors during the application of GES. The impedance was in the range of 500–1000 Ω during the entire study period. These values were within the normal range of electrode impedance.
Stimulation Energy Both the initial effective energy and the maximum effective stimulation energy for the later part of GES showed a wide distribution among dogs (medians 15–336 s·mA2). The initial effective stimulation energy ranged from 5 to 96 s·mA2. The maximum effective stimulation energy ranged from 96 to 672 s·mA2. The maximum effective stimulation energy was more than 10 times the initial effective stimulation energy in six dogs. For the other two dogs, it was, respectively, eight times and seven times as much. The data for effective stimulation energy in all eight dogs are shown in Table 1.
www.neuromodulationjournal.com
months of GES, dogs’ body weight was clearly lower compared with three months of sham GES (p < 0.05). By the end of three months of GES, dogs’ body weight decreased by 9.8 ± 5.6% (group A) and 17.5 ± 6.0% (group B) compared with sham GES.
Food Craving For the three-choice food preference test, a significant change in food cravings was induced by GES. The difference in total consumption of each food choice between the periods of GES and sham GES was significant (p < 0.05). Dogs undergoing GES in group A ate 18.2 ± 5.4% less food overall than with sham GES (272 ± 25.8 vs. 333 ± 31.4 g/day, p < 0.05). Dogs undergoing GES in group B ate 34.5 ± 1.2% less than with sham GES (209 ± 18.0 vs. 319 ± 33.5 g/day, p < 0.05). Dogs undergoing GES ate significantly less high-fat food than with sham GES (p < 0.05). However, there were no differences in consumption of either highcarbohydrate food or balanced food between the periods of GES and sham GES. For high-fat food, during the period of GES, dogs in group A ate 34.6 ± 16.9% less compared with sham GES (151 ± 43.8 vs. 230.3 ± 16.3 g/day, p < 0.05). Dogs in group B ate 47 ± 14.1% less compared with sham GES (138.7 ± 35 vs. 272.3 ± 40.6 g/day, p < 0.05). For high-carbohydrate food, dogs in group A ate 116 ± 43.4 g/day during the period of GES and ate 102 ± 28.7 g/day during the sham GES period. Dogs in group B ate 65 ± 32.3 g/day during the period of GES and ate 43 ± 7.6 g/day during the sham GES period. During the period of GES, a trend of increased consumption of highcarbohydrate food existed in both groups compared with sham GES. However, there was no significant difference between the periods of GES and sham GES. For the balanced food, dogs in group A ate 5 ± 3.8 g/day during the period of GES and 0.25 ± 0.25 g/day during the sham GES period.
© 2014 International Neuromodulation Society
Neuromodulation 2014; 17: 483–489
485
Food Intake There was a statistically significant difference in food consumption between GES and sham GES. Dogs who had undergone GES for three months ate significantly less food than those who had undergone sham GES for three months (p < 0.05). In group A during the period of GES, there was a 24 ± 10% decrease in food consumption compared with sham GES (211 ± 9.0 vs. 279 ± 24.6 g/day, p < 0.05). In group B during the period of GES, there was a 30 ± 4% decrease in the food consumption compared with sham GES (157 ± 12.9 vs. 225 ± 8.8 g/day, p < 0.05). Figure 1 shows the mean daily food consumption with GES and sham GES for each dog. By the end of three
Figure 1. Mean daily food consumption during gastric electrical stimulation (GES) and sham GES for each dog. There was a statistically significant difference in food consumption between GES and sham GES. Dogs on GES for three months ate significantly less food than those on sham GES for three months (p < 0.05).
GUO ET AL. tically significant difference between GES and sham GES. Moreover, during the period of GES, a trend of decrease in total postprandial plasma GLP-1 existed in both groups compared with sham GES, but total GLP-1 levels showed no statistically significant difference between GES and sham GES (23.4 ± 8.1 vs. 32.1 ± 6.5 pM for group A, 22.1 ± 12.7 vs. 23.0 ± 9.3 pM for group B). Neither preprandial nor postprandial plasma levels of PYY3-36 showed a statistically significant difference between GES and sham GES.
DISCUSSION
Figure 2. Mean consumption of three food choices with GES and sham GES (g/day) in group A (a) and group B (b). During the three-choice food preference test, a significant change in food cravings was induced by GES. Dogs on GES ate significantly less high-fat food than those on sham GES. However, there were no differences in consumption of either high-carbohydrate food or balanced food between the periods of GES and sham GES. For high-fat food, during the period of GES, dogs in group A ate 34.6 ± 16.9% less compared with sham GES (151 ± 43.8 vs. 230.3 ± 16.3 g/day, p < 0.05). Dogs in group B ate 47 ± 14.1% less compared with sham GES (138.7 ± 35 vs. 272.3 ± 45.6 g/day, p < 0.05). FD, high-fat diet; CD, high-carbohydrate diet; BD, balanced diet.
Dogs in group B ate 5 ± 1.5 g/day during the period of GES and 3 ± 1.5 g/day during the sham GES period. Although a trend of increased consumption of the balanced food existed in both groups with GES compared with sham GES, there was no significant difference between the periods of GES and sham GES. Figure 2 shows the mean consumption of the three food choices daily during GES and sham GES in group A and group B.
486
Gastrointestinal Peptides As shown in Table 2, the plasma concentrations of ghrelin, PYY, and total GLP-1 did not differ significantly between the periods of GES and sham GES. Although during the period of GES a trend of increase in preprandial plasma ghrelin existed in both groups compared with sham GES (5.7 ± 2.3 vs. 3.9 ± 1.6 ng/mL for group A, 7.3 ± 1.2 vs 6.7 ± 3.1 ng/mL for group B), ghrelin levels showed no statiswww.neuromodulationjournal.com
In this study, we investigated the effects of chronic adjustable GES, including its effects on gastrointestinal peptides. The results showed that food consumption was reduced significantly by adjustable GES. Furthermore, in the three-choice food preference test, a significant change in food cravings was induced by GES. Finally, GES had no significant effects on plasma gastrointestinal peptide levels. All dogs completed the study, and no side effects of GES were observed. GES has been studied as a potential treatment for obesity in many previous animal studies. However, in some clinical studies, no loss of excess body weight was observed (12–14). A randomized clinical trial with 189 obese patients in the USA reported no significant weight loss in those undergoing active stimulation compared with those undergoing sham stimulation over 12 months (12). With more intensive investigation, experiments have documented the importance of individual parameters for the effectiveness of GES (5,15,16). Clinical studies have reported individual variability in visceral sensitivity. There is also a wide diversity of visceral responses to GES among individuals (16). Moreover, resistance (i.e., diminution in the response to GES after long-term use) may occur following chronic stimulation (9). The increase in body weight and food intake that occurred during chronic stimulation indicated a diminution in the response to long-term GES. Based on differences in individual sensitivity, adjustment of the GES parameters was required over a wide range. In our experiment, the reaction of each dog to chronic GES was different. A trend in individual variations was observed with respect to the initial effective stimulation intensity (with energy of 5 to 96 s·mA2) and the maximum effective stimulation intensity (with energy of 96 to 672 s·mA2). In the three-choice food preference test, a significant change in food cravings was induced by GES. Despite free access to three food choices, animals on GES decreased total food intake by 18.2 ± 5.4% in group A and 34.5 ± 1.2% in group B as compared with sham GES. Although no significant differences in consumption of highcarbohydrate or balanced food between GES and sham GES were observed, GES induced a significant decrease in consumption of high-fat food. Animals on GES in groups A and B ate, respectively, 34.6 ± 16.9% and 47 ± 14.1% less high-fat food than with sham GES. Reducing the intake of high-fat food reduced total food consumption. Our results indicate that GES is associated with reduced food intake and food craving. GES not only reduced food consumption but also had an important role in food choices. GES may modulate individual appetite in terms of both quantity and quality. Our study sheds new light on the idea that diet habits may be changed by this treatment. Val-Laillet et al. showed similar effects of vagus nerve stimulation (VNS) on food craving (17). Long-term VNS was performed in eight pigs, and sweet craving was reduced by chronic VNS. Dedeurwaerdere et al. suggested that the olfactory bulbs play a considerable role in food intake during VNS (18). Bodenlos et al.
© 2014 International Neuromodulation Society
Neuromodulation 2014; 17: 483–489
GASTRIC ELECTRICAL STIMULATION IN DOGS
Table 2. Plasma Concentrations of Gastrointestinal Peptides in Different Periods. Peptide Ghrelin (ng/mL) Peptide YY3-36 (ng/mL) Total glucagon-like peptide 1 (pM)
Time Preprandial Postprandial Preprandial Postprandial Preprandial Postprandial
Group A Baseline
GES
Sham GES
Group B Baseline
GES
Sham GES
4.7 ± 1.2 3.1 ± 0.6 5.5 ± 1.9 6.5 ± 1.1 7.2 ± 4.3 29.7 ± 4.2
5.7 ± 2.3 5.3 ± 0.2 8.7 ± 3.6 7.3 ± 3.3 9.7 ± 2.7 23.4 ± 8.1
3.9 ± 1.6 4.3 ± 1.3 5.5 ± 2.7 5.5 ± 0.8 12.8 ± 2.5 32.1 ± 6.5
7.3 ± 3.7 4.7 ± 1.0 4.6 ± 1.1 6.2 ± 2.0 8.6 ± 6.7 29.4 ± 10.7
7.3 ± 1.2 5.0 ± 0.5 3.2 ± 1.3 3.6 ± 0.9 9.5 ± 5.9 22.1 ± 12.7
6.7 ± 3.1 5.9 ± 1.4 3.0 ± 0.9 3.3 ± 1.2 7.3 ± 3.4 23.0 ± 9.3
The plasma concentrations of the three gastrointestinal peptides did not differ significantly between the periods of GES and sham GES. GES, gastric electrical stimulation.
www.neuromodulationjournal.com
As with PYY3-36, GLP-1 is considered a very important anorexigenic hormone. After GLP-1 or GLP-1 receptor agonists are administered centrally or peripherally to animals or peripherally to humans, food intake and body weight are reduced (27,28). Postprandial levels may be elevated as early as two days after gastric bypass (29). On the other hand, fasting levels of GLP-1 are decreased after bariatric surgery (30). Whitson found (31) that GLP-1 levels were significantly altered in morbidly obese nondiabetic patients six months after gastric bypass surgeries, and no significant change was noted in a matched cohort of diabetic patients. These results suggest that changes in GLP-1 levels are not the only mechanism responsible for producing weight loss after gastric bypass. In the present study, although during GES a trend toward a decrease in total postprandial plasma GLP-1 levels was observed in both groups as compared with during sham GES; this difference was not statistically significant. Because the study sample was small and exhibited high interindividual variability, we cannot make a definitive conclusion regarding this outcome. Our results suggest that the peripheral levels of ghrelin, PYY3-36, or total GLP-1 may not influence the effects of GES on food intake. Thus, it is likely that the effects of GES are complex, and levels of other peptide hormones such as cholecystokinin, obestatin, oxyntomodulin, and pancreatic polypeptide could be affected by GES. Both previous studies and the present study indicate that food intake and diet craving can be significantly altered by adjustable GES. Because dietary habits are important to body energy homeostasis, GES may be useful for treating obesity by changing food preferences. Further clinical studies are necessary to highlight the effect of adjustable GES on eating behavior.
Acknowledgements We would like to thank Dr. Kunpeng Liu for his support during the animal anesthesia.
Authorship Statements Drs. Xiaojuan Guo, Yanmei Li, Shukun Yao, Shaoxuan Chen, Yuhui Du, and Zhihua Wang contributed to experiment design, conduct, and analysis. Shukun Yao and Shaoxuan Chen conducted the animal anesthesia and surgery. Xiaojuan Guo and Yanmei Li prepared the manuscript and prepared the figures with important intellectual input from Dr. Shukun Yao. Shukun Yao, Yuhui Du, and Zhihua Wang designed the stimulation and recording system. All authors reviewed the manuscript.
© 2014 International Neuromodulation Society
Neuromodulation 2014; 17: 483–489
487
also proposed, based on data in 33 participants, that interaction between interoceptive and exteroceptive signals may explain altered food craving during VNS (19). However, the mechanisms relating GES and food cravings are still not well elucidated. More studies are necessary to elucidate the interaction between GES and food preferences. Body energy homeostasis is largely regulated by hunger and satiety signals resulting from gut–brain interactions. Satiety signals are regulated by peptides synthesized and released from enteroendocrine cells in the gastrointestinal tract. These cells release various peptide hormones that act as a signal to the central nervous system to regulate appetite. Although there are several studies of the effects of GES on peptide hormones in the central neural system, gastrointestinal tissue, and peripheral plasma, the results remain inconclusive (7,20). Ghrelin is a 28-amino-acid peptide hormone. It is produced predominantly in the stomach and represents the only known orexigenic gut hormone identified to date (21). When administered either peripherally or centrally to rodents, ghrelin rapidly increases food intake and body weight (22). Ghrelin has also been shown to stimulate appetite in humans and has been proposed to function as a meal initiator, in part due to its potent appetite-stimulating effects (23). Recent studies have demonstrated that GES may also regulate the secretion of ghrelin. GES was found to reduce gastric fundus ghrelin in diet-induced obesity rats (20). After six months of GES in obese patients, peripheral plasma ghrelin levels increased, and weight loss was correlated significantly with increased ghrelin levels (24). These data suggest that a compensatory phenomenon may appear after weight loss induced by GES. On the other hand, recent studies showed that GES did not modulate preprandial plasma ghrelin levels (4), and GES also had no effect on postprandial plasma ghrelin levels (25). In the present study, we likewise found that chronic GES did not modulate preprandial or postprandial plasma ghrelin levels. Although during the period of GES a trend towards an increase in preprandial plasma ghrelin levels was clear in both groups as compared with sham GES, the difference between GES and sham GES was not statistically significant because of high interindividual variability. PYY3-36 levels were not affected by chronic adjustable GES in this study. Circulating levels of PYY3-36 are influenced by meal composition and calorie content and elevated within one hour postfeeding (11). PYY is considered a potent anorexigenic hormone. Significant increases in circulating PYY levels have been reported following gastrointestinal surgeries, possibly contributing to the weight loss attributed to these procedures (26). However, in our studies, neither preprandial nor postprandial plasma levels of PYY3-36 were significantly different between GES and sham GES.
GUO ET AL.
How to Cite this Article: Guo X., Li Y., Yao S., Chen S., Du Y., Wang Z. 2014. The Effects of Individual Gastric Electrical Stimulation on Food Craving and Gastrointestinal Peptides in Dogs. Neuromodulation 2014; 17: 483–489
REFERENCES
488
1. World Health Organization. Obesity and overweight. In: Fact sheet no. 311. http:// www.who.int/mediacentre/factsheets/fs311/en/ 2. Zhang J, Maude-Griffin R, Zhu H et al. Gastric electrical stimulation parameter dependently alters ventral medial hypothalamic activity and feeding in obese rats. Am J Physiol Gastrointest Liver Physiol 2011;301:G912–G918. 3. Zhang J, Tang M, Chen JD. Gastric electrical stimulation for obesity: the need for a new device using wider pulses. Obesity (Silver Spring) 2009;17:474–480. 4. Li S, Maude-Griffin R, Sun Y et al. Food intake and body weight responses to intermittent vs. continuous gastric electrical stimulation in diet-induced obese rats. Obes Surg 2013;23:71–79. 5. Guo X, Li Y, Yao S et al. Parameter selection and stimulating effects of an adjustable gastric electrical stimulator in dogs. Obes Surg 2014;24:78–84. 6. Mizrahi M, Ben Ya’acov A, Ilan Y. Gastric stimulation for weight loss. World J Gastroenterol 2012;18:2309–2319. 7. Gallas S, Sinno MH, Boukhettala N et al. Gastric electrical stimulation increases ghrelin production and inhibits catecholaminergic brainstem neurons in rats. Eur J Neurosci 2011;33:276–284. 8. Ouyang H, Xing J, Chen JD. Tachygastria induced by gastric electrical stimulation is mediated via alpha- and beta-adrenergic pathway and inhibits antral motility in dogs. Neurogastroentero Motil 2005;17:846–853. 9. Shikora SA. Implantable gastric stimulation for the treatment of severe obesity. Obes Surg 2004;14:545–548. 10. Yin J, Chen JD. Retrograde gastric electrical stimulation reduces food intake and weight in obese rats. Obes Res 2005;13:1580–1587. 11. Perry B, Wang Y. Appetite regulation and weight control: the role of gut hormones. Nutr Diabetes 2012;16:e26. 12. Shikora SA, Bergenstal R, Bessler M et al. Implantable gastric stimulation for the treatment of clinically severe obesity: results of the SHAPE trial. Surg Obes Relat Dis 2009;5:31–37. 13. Hoeller E, Aigner F, Margreiter R et al. Intragastric stimulation is ineffective after failed adjustable gastric banding. Obes Surg 2006;16:1160–1165. 14. Shikora SA. “What are the Yanks doing?” The U.S. experience with implantable gastric stimulation (IGS) for the treatment of obesity—update on the ongoing clinical trials. Obes Surg 2004;14:S40–S48. 15. Yao SK, Ke MY, Wang ZF et al. Visceral sensitivity to gastric stimulation and its correlation with alterations in gastric emptying and accommodation in humans. Obes Surg 2005;15:247–253. 16. Yao SK, Ke MY, Wang ZF et al. Visceral response to acute retrograde gastric electrical stimulation in healthy human. World J Gastroenterol 2005;11:4541– 4546. 17. Val-Laillet D, Biraben A, Randuineau G et al. Chronic vagus nerve stimulation decreased weight gain, food consumption and sweet craving in adult obese minipigs. Appetite 2010;55:245–252. 18. Dedeurwaerdere S, Cornelissen B, VanLaere K et al. Small animal positron emission tomography during vagus nerve stimulation in rats. A pilot study. Epilepsy Res 2005;67:133–141. 19. Bodenlos JS, Kose S, Borckardt JJ et al. Vagus nerve stimulation acutely alters food craving in adults with depression. Appetite 2007;48:145–153. 20. Xu J, McNearney TA, Chen JD. Gastric/intestinal electrical stimulation modulates appetite regulatory peptide hormones in the stomach and duodenum in rats. Obes Surg 2007;17:406–413. 21. Hameed S, Dhillo WS, Bloom SR. Gut hormones and appetite control. Oral Dis 2009;15:18–26. 22. Castañeda TR, Tong J, Datta R et al. Ghrelin in the regulation of body weight and metabolism. Front Neuroendocrinol 2010;31:44–60. 23. Druce MR, Wren AM, Park AJ et al. Ghrelin increases food intake in obese as well as lean subjects. Int J Obes (Lond) 2005;29:1130–1136. 24. Cigaina V, Hirschberg AL. Plasma ghrelin and gastric pacing in morbidly obese patients. Metabolism 2007;56:1017–1021. 25. Sanmiguel CP, Haddad W, Aviv R et al. The TANTALUS system for obesity: effect on gastric emptying of solids and ghrelin plasma levels. Obes Surg 2007;17:1503– 1509. 26. Korner J, Bessler M, Cirilo LJ et al. Effects of Roux-en-Y gastric bypass surgery on fasting and postprandial concentrations of plasma ghrelin, peptide YY, and insulin. J Clin Endocrinol Metab 2005;90:359–365. 27. Tourrel C, Bailbe D, Meile MJ et al. Glucagon-like peptide-1 and exendin-4 stimulate beta-cell neogenesis in streptozotocin-treated newborn rats resulting in persistently improved glucose homeostasis at adult age. Diabetes 2001;50:1562– 1570. 28. Vilsboll T, Zdravkovic M, Le-Thi T et al. Liraglutide, a long-acting human glucagonlike peptide-1 analog, given as monotherapy significantly improves glycemic
www.neuromodulationjournal.com
control and lowers body weight without risk of hypoglycemia in patients with type 2 diabetes. Diabetes Care 2007;30:1608–1610. 29. le Roux CW, Welbourn R, Werling M et al. Gut hormones as mediators of appetite and weight loss after Roux-en-Y gastric bypass. Ann Surg 2007;246:780– 785. 30. Reinehr T, Roth CL, Schemthaner GH et al. Peptide YY and glucagon-like peptide-1 in morbidly obese patients before and after surgically induced weight loss. Obes Surg 2007;17:1571–1577. 31. Whitson BA, Leslie DB, Kellogg TA et al. Entero-endocrine changes after gastric bypass in diabetic and nondiabetic patients: a preliminary study. J Surg Res 2007;141:31–39.
COMMENTS Appropriate and adjustable parameters in neuromodulation continue to be an intense focus of clinical and basic science research. For many years empirical parameters were utilized for various forms of pain and other treatments. Stimulus frequencies were generally kept between 40 and 125 Hz but the frequency generally centered around 50 Hz and pulse widths often varied between 100–200 μs. Our personal experience that encouraged us to expand various parameters resulted from using an animal model to test higher frequencies of spinal cord stimulation on changes in hind limb peripheral blood flow (1). For several years we used 50 Hz as the “gold standard” for studying peripheral blood flow. However, in the study by Gao et al. (2010), spinal cord stimulation at 500 Hz significantly increased cutaneous blood flow and decreased vascular resistance compared to changes induced by frequencies of 50 and 200 Hz. Recently clinical and basic science studies have expanded to study effective frequencies extending from 4 Hz to 10 K Hz and higher. In addition to variation in frequencies, studies have shown that selection of appropriate pulse widths is also a key parameter that can improve treatment with neuromodulation. In the present study Gastric Electrical Stimulation (GES) was used to treat obesity by affecting food craving in a canine model. The authors pointed out that commercial implanted devices mainly adapted from a nerve stimulator were not appropriate for producing effective responses for GES. In fact, Dr. Jiande Chen and his colleagues (2) published an article stating that a new device with wider pulses was essential for treating obesity with GES. The present study amplifies the importance using an adjustable stimulator with a wide range of stimulation parameters, particularly pulse widths. This study highlighted the importance using adjustable parameters. After the animal completely recovered from the surgery to implant the electrodes on the greater curvature of the stomach, each animal was tested with a selection of pulse widths to define optimal settings; the other parameters were relatively fixed. The selection of the appropriate pulse width was based on the total symptom score. Their results suggested that GES was associated with reduced food intake. An intriguing result was that GES with different parameters not only reduced food consumption but also had an impact on the choice of food. One surprising result was that the parameters selected for reducing food consumption and choice did not require significant changes in gastrointestinal peptides including ghrelin, PYY3-36 and GLP-1. Would these peptides levels change if the electrodes using adjustable parameters were placed on another region of the stomach? The authors suggested that this study may provide new insights about using their approach to change diet habits.
© 2014 International Neuromodulation Society
Robert Foreman, PhD Oklahoma City, OK, USA
Neuromodulation 2014; 17: 483–489
GASTRIC ELECTRICAL STIMULATION IN DOGS REFERENCES (1) Gao J, Wu M, Li L, Qin C, Farber JP, Linderoth B, Foreman RD. Effects of spinal cord stimulation with “standard clinical” and higher frequencies on peripheral blood flow in rats. Brain Res 2010 Feb 8;1313:53–61. (2) Zhang J, Tang M, Chen JD. Gastric electrical stimulation for obesity: the need for a new device using wider pulses. Obesity (Silver Spring) 2009;17:474–480.
*** This work showed that gastric electrical stimulation (GES) in dogs decreased their food intake and body weight after three months of GES. Although such effects of GES are well-known phenomena (reviewed in [1]), the authors applied a novel approach consisting in adjusting parameters of GES over time to achieve the maximal tolerated energy delivered by GES. As long as there is no efficient pharmacological anti-obesity treatment, GES can be considered as a therapeutic option being less invasive than bariatric surgery. Several questions remain however unanswered such as the molecular mechanism underlying food intake reducing effects of GES and if these effects of GES in normal weight dogs can be directly applicable to obese humans. In fact, the obese animals often do not respond to the same treatment that reduce food intake and body weight in the non- obese [2]. Furthermore, the authors did not find any significant effect of GES on plasma gastrointestinal hormone concentration including ghrelin, GLP1 and PYY that would explain observed changes in food intake. However, a recent study showed that in obese subjects increased affinity of plasma immunoglobulins for ghrelin enhances ghrelin’s
orexigenic effect with otherwise normal or reduced plasma ghrelin levels [3]. It would be, hence, of interest to see if long-term effects of GES on appetite may involve modulation of production and properties of immunoglobulins reactive with hunger and satiety hormones after either high or low energy GES. In fact, acute GES with Enterra parameters (low energy) in rats was shown to stimulate ghrelin secretion [4] but also increased plasma levels of ghrelin-reactive IgG [1]. Sergueï O. Fetissov, MD, PhD Rouen University, France
REFERENCES 1. Gallas S, Fetissov SO. Ghrelin, appetite and gastric electrical stimulation. Peptides 2011;32:2283–2289. 2. Fetissov SO, Meguid MM. Serotonin delivery into the ventromedial nucleus of the hypothalamus affects differently feeding pattern and body weight in obese and lean Zucker rats. Appetite 2010;54:346–353. 3. Takagi K, Legrand R, Asakawa A et al. Anti-ghrelin immunoglobulins modulate ghrelin stability and its orexigenic effect in obese mice and humans. Nat Commun 4:2685, doi: 10.1038/ncomms3685 (2013). 4. Gallas S, Sinno MH, Boukhettala N et al. Gastric electrical stimulation increases ghrelin production and inhibits catecholaminergic brainstem neurons in rats. European Journal of Neuroscience 2011;33:276–284.
Comments not included in the Early View version of this paper.
489
www.neuromodulationjournal.com
© 2014 International Neuromodulation Society
Neuromodulation 2014; 17: 483–489
Revised: February 13, 2014
Accepted: March 24, 2014
(onlinelibrary.wiley.com) DOI: 10.1111/ner.12207
The Effects of Individualized Gastric Electrical Stimulation on Food Craving and Gastrointestinal Peptides in Dogs Xiaojuan Guo, PhD; Yanmei Li, MD; Shukun Yao, MD; Shaoxuan Chen, MD; Yuhui Du, DEng; Zhihua Wang, DEng Background: Using an adjustable stimulator with a wide range of stimulation parameters, the aims of this study were 1) to investigate the effects of long-term gastric electrical stimulation (GES) on appetite and differential food cravings for three different foods and 2) to investigate the effects of GES on plasma gastrointestinal peptide concentrations. Methods: The study was performed in eight Beagle dogs implanted with one pair of serosal electrodes. They were followed during GES and sham GES sessions in a crossover design. GES was conducted using a series of individualized parameters. Food intake and food cravings were observed to evaluate the effects of long-term GES. Enzyme-linked immunosorbent assay was used to measure the plasma concentrations of gastrointestinal peptides. Results: Dogs on GES for three months ate significantly less food than those on sham GES for three months (p < 0.05). A significant change in food cravings was induced by GES. Dogs with GES ate significantly less high-fat food. However, there was no significant difference in consumption of high-carbohydrate food or balanced food between the periods of GES and sham GES. The plasma concentrations of ghrelin, peptide YY3-36, and glucagon-like peptide 1 did not differ significantly between the periods of GES and sham GES. Conclusions: Food intake and food craving were changed significantly by adjustable GES. GES may be used for treating obesity by changing food preferences. Further clinical studies are necessary to highlight the effect of adjustable GES on eating behavior. Keywords: Food craving, gastric electrical stimulation, individual parameters, mechanism, obesity Conflict of Interest: The authors reported no conflict of interest.
INTRODUCTION
www.neuromodulationjournal.com
Address correspondence to: Shukun Yao, MD, Department of Gastroenterology, China–Japan Friendship Hospital, 2 Yinghua East Road, Chaoyang District, Beijing, 100029, China. Email: [email protected] Department of Gastroenterology, China–Japan Friendship Hospital, Beijing, China For more information on author guidelines, an explanation of our peer review process, and conflict of interest informed consent policies, please go to http:// www.wiley.com/bw/submit.asp?ref=1094-7159&site=1 Sources of financial support: The National Natural Science Foundation of China (Grant No. 81070299) and the Capital Medical Development Foundation of China (Grant No. 2009-2018) provided funding for the study.
© 2014 International Neuromodulation Society
Neuromodulation 2014; 17: 483–489
483
Popular attention has been attracted by the overwhelming increase in the prevalence of obesity in recent years. The World Health Organization has estimated that by 2015 about 2.3 billion adults (aged 15+) will be overweight and more than 700 million will be obese. Nowadays at least 2.8 million people worldwide die each year as a result of being overweight or obese (1). Gastric electrical stimulation (GES) has been investigated for the treatment of obesity since 1995. Electrical stimulation is delivered to the stomach via an implanted system that consists of a stimulator and two leads. An external programmer delivers instructions to the stimulator. Possible advantages to GES include less invasiveness than some bariatric surgeries (e.g., gastric bypass) and fewer side effects (e.g., dumping syndrome) (2). Although GES is known to reduce food intake and produce weight loss in animal and human studies, the commercial implantable devices used in clinical studies have mainly been adapted from nerve stimulators, not specifically developed for GES (3,4). These devices have been restricted to the following typical parameters: a pulse width of 0.3 msec, an amplitude of 5–10 mA, a frequency of 40 Hz, and 2 sec on, 3 sec off. Recently, individualized parameters for GES have been proposed for treating obesity. Parameter selection and resistance (i.e., diminution in the response to GES after longterm use) during GES must be considered (5). Fixed parameters are
insufficient, and in this study an adjustable stimulator with a wide range of stimulation parameters was used. This stimulator features a pulse width of 0.3–10 msec, an amplitude of 1–10 mA, a frequency of 40 Hz, and 2 sec on, 3 sec off. Short pulses for GES may not be potent enough, and experiments have proved the importance of pulse width in the stimulation process. The mechanisms by which GES affects food intake and weight are not fully elucidated. The mechanical, neural, and hormonal pathways have close relations with GES in the regulation of energy balance (6–8). A number of studies have shown that GES is effective in reducing appetite and inducing early satiety (8–10). Gastrointestinal peptides such as ghrelin, peptide YY (PYY), and glucagon-like peptide 1 (GLP-1) are important in regulation of food intake (11).
GUO ET AL. Previous studies have shown that GES alters the level of selected satiety-related peptides and the expression of neuropeptides in neurons of the hypothalamus (7). We hypothesized that modulation of the secretion of ghrelin, PYY3-36, and GLP-1 may be one mechanism by which adjustable stimulation reduces appetite and food intake. The aims of this study using adjustable parameters were the following: 1) to investigate the effects of long-term GES on appetite and a possible difference in food craving among three different foods and 2) to investigate the effect of GES on plasma gastrointestinal peptide concentrations.
METHODS AND PROCEDURES Surgical Procedure Eight healthy female Beagle dogs (Academy of Military Medical Sciences, Beijing, China; 8.5–13 kg) were involved in this study and allowed an acclimation period of two weeks. After an overnight fast, the dogs were anesthetized with a combination of fentanyl (2 μg/ kg), propofol (2 mg/kg), and rocuronium (0.6 mg/kg) and maintained on 1–2% sevoflurane in oxygen (1.0 L/min) carrier gases delivered from a ventilator after endotracheal intubation. Vital signs (respiratory rate, pulse, and tongue color) were monitored. By laparoscope, one pair of platinum–iridium electrodes (Noted Technology Development Co. Ltd., Shenzhen, China) was sutured into the seromuscular layer along the greater curvature of the stomach, 3.0 cm above the pylorus. The electrodes were 1.0 cm apart. An adjustable electrical stimulator was embedded under the costal margin of the right upper quadrant subcutaneously and then secured in place by a purse-string suture. After surgery, dogs wore special jackets and were allowed to recover in their individual cages. The dogs received sufficient solid food at a set time (5 PM to 7 PM daily) and had free access to water in their home cages at all times. The postoperative situation, including swelling, secretions, ulcer, and subcutaneous transposition, wound healing, and exposed material, was recorded. All experiments were performed after the dogs were completely recovered, usually two weeks after the surgery. The Animal Care and Use Committee of the China–Japan Friendship Hospital approved the study.
484
Individual Parameters An external programmer delivered periodic rectangular pulses during the experiment. After being completely recovered from the surgery, each dog underwent a pulse width selection period for two weeks, and simultaneously, the other parameters were fixed relative to the pulse width. The pulse widths were increased progressively from 0.3 msec. The maximum pulse width was 10 msec. If the intolerable pulse width was ≤1.0 msec, a lower amplitude was chosen (3 mA, 6 mA, 9 mA). The selection of stimulation parameters was based on the total symptom score. Evaluation of the dog’s symptoms involved observing for licking with the tongue, sialorrhea, vomiting, yawning, barking, groaning, belching, murmuring, and dysphoria. These symptoms were assessed based on their frequency and/or severity (0: never; 1: seldom/mild; 2: often/moderate; 3: continue/ severe). Intolerable parameters were recognized when the total symptom score was ≥3. The maximum pulse widths that induced no or very mild symptoms were selected (i.e., score 500 g balanced food at a set time slot. To determine their food cravings, all dogs were subjected to a food preference test for three days continuously. This test was performed in their individual cages, fitted with three feeders that had the exact same shape. Each day at 5 PM to 7 PM, the three feeders were filled with three different foods (500 g each): BD, CD, and FD. The position of each diet was randomly determined each day. The animals had two hours to eat, after which the remainders in each compartment were weighed to calculate the consumed amount of each food.
Gastrointestinal Peptides Blood was collected from three months after initiation of GES and three months after initiation of sham GES. An indwelling catheter was inserted into a foreleg vein to collect two blood samples (3 mL each) during the three-hour course of the experiment. The first blood sample (preprandial level) was drawn in a fasting state (30 min after GES/sham GES). Then, after GES for one hour, the animals were fed for 30 min. The second blood sample (postprandial level) was drawn one hour after feeding. Blood samples were collected in chilled tubes containing aprotinin (500 IU/mL) and EDTA (1 mg/mL). Plasma ghrelin and PYY3-36 were quantified using enzyme-linked immunosorbent assay (ELISA) kits specifically for canine proteins (Phoenix Pharmaceuticals, Belmont, CA, USA). Total plasma GLP-1 was measured using an ELISA kit (Millipore Research, St. Charles, MO, USA). The assay was carried out according to the instructions provided by the manufacturer.
Experimental Protocol The experiment followed a crossover design. After the periods of recovery from surgery (two weeks) and parameter selection (two weeks), the dogs were randomly studied in two groups. In group A, four dogs underwent a three-month treatment with GES, a twoweek washout period, and a three-month control with sham GES sequentially. During the experiment, GES or sham GES was administered one hour before food intake and then continued for three hours. The other four dogs, in group B, underwent the experiment in the reverse order. At the end of the three months of GES and sham
© 2014 International Neuromodulation Society
Neuromodulation 2014; 17: 483–489
GASTRIC ELECTRICAL STIMULATION IN DOGS
Table 1. Effective Stimulation Energy in Eight Dogs (s·mA2).
Initial effective energy Maximum effective energy
Median
First quartile
Third quartile
90th percentile
95th percentile
15 336
7.3 216
35.3 384
62.4 672
79.2 672
Both the initial effective energy and the maximum effective stimulation energy varied greatly among dogs.
GES, all dogs were subjected to a blood test and a two-hour food cravings test three days in a row. During these tests, GES or sham GES was continued as before. The procedures for sham GES were the same as the GES procedures except that the stimulators were not turned on.
Statistical Analysis Food intake data and gastrointestinal peptide levels are presented as mean ± SD. As the data for stimulation energy were not normally distributed, results are reported as median and first and third quartiles and 90th and 95th percentiles given. Differences in food intake and peptides across GES and sham GES were analyzed by crossover ANOVA. Differences in food intake and peptides within groups were evaluated using paired Student’s t-test or a nonparametric test, as appropriate. The level of statistical significance was set at p < 0.05.
RESULTS All dogs completed the study, and no adverse reactions were observed after surgery. None of the dogs was observed to display severely abnormal behaviors during the application of GES. The impedance was in the range of 500–1000 Ω during the entire study period. These values were within the normal range of electrode impedance.
Stimulation Energy Both the initial effective energy and the maximum effective stimulation energy for the later part of GES showed a wide distribution among dogs (medians 15–336 s·mA2). The initial effective stimulation energy ranged from 5 to 96 s·mA2. The maximum effective stimulation energy ranged from 96 to 672 s·mA2. The maximum effective stimulation energy was more than 10 times the initial effective stimulation energy in six dogs. For the other two dogs, it was, respectively, eight times and seven times as much. The data for effective stimulation energy in all eight dogs are shown in Table 1.
www.neuromodulationjournal.com
months of GES, dogs’ body weight was clearly lower compared with three months of sham GES (p < 0.05). By the end of three months of GES, dogs’ body weight decreased by 9.8 ± 5.6% (group A) and 17.5 ± 6.0% (group B) compared with sham GES.
Food Craving For the three-choice food preference test, a significant change in food cravings was induced by GES. The difference in total consumption of each food choice between the periods of GES and sham GES was significant (p < 0.05). Dogs undergoing GES in group A ate 18.2 ± 5.4% less food overall than with sham GES (272 ± 25.8 vs. 333 ± 31.4 g/day, p < 0.05). Dogs undergoing GES in group B ate 34.5 ± 1.2% less than with sham GES (209 ± 18.0 vs. 319 ± 33.5 g/day, p < 0.05). Dogs undergoing GES ate significantly less high-fat food than with sham GES (p < 0.05). However, there were no differences in consumption of either highcarbohydrate food or balanced food between the periods of GES and sham GES. For high-fat food, during the period of GES, dogs in group A ate 34.6 ± 16.9% less compared with sham GES (151 ± 43.8 vs. 230.3 ± 16.3 g/day, p < 0.05). Dogs in group B ate 47 ± 14.1% less compared with sham GES (138.7 ± 35 vs. 272.3 ± 40.6 g/day, p < 0.05). For high-carbohydrate food, dogs in group A ate 116 ± 43.4 g/day during the period of GES and ate 102 ± 28.7 g/day during the sham GES period. Dogs in group B ate 65 ± 32.3 g/day during the period of GES and ate 43 ± 7.6 g/day during the sham GES period. During the period of GES, a trend of increased consumption of highcarbohydrate food existed in both groups compared with sham GES. However, there was no significant difference between the periods of GES and sham GES. For the balanced food, dogs in group A ate 5 ± 3.8 g/day during the period of GES and 0.25 ± 0.25 g/day during the sham GES period.
© 2014 International Neuromodulation Society
Neuromodulation 2014; 17: 483–489
485
Food Intake There was a statistically significant difference in food consumption between GES and sham GES. Dogs who had undergone GES for three months ate significantly less food than those who had undergone sham GES for three months (p < 0.05). In group A during the period of GES, there was a 24 ± 10% decrease in food consumption compared with sham GES (211 ± 9.0 vs. 279 ± 24.6 g/day, p < 0.05). In group B during the period of GES, there was a 30 ± 4% decrease in the food consumption compared with sham GES (157 ± 12.9 vs. 225 ± 8.8 g/day, p < 0.05). Figure 1 shows the mean daily food consumption with GES and sham GES for each dog. By the end of three
Figure 1. Mean daily food consumption during gastric electrical stimulation (GES) and sham GES for each dog. There was a statistically significant difference in food consumption between GES and sham GES. Dogs on GES for three months ate significantly less food than those on sham GES for three months (p < 0.05).
GUO ET AL. tically significant difference between GES and sham GES. Moreover, during the period of GES, a trend of decrease in total postprandial plasma GLP-1 existed in both groups compared with sham GES, but total GLP-1 levels showed no statistically significant difference between GES and sham GES (23.4 ± 8.1 vs. 32.1 ± 6.5 pM for group A, 22.1 ± 12.7 vs. 23.0 ± 9.3 pM for group B). Neither preprandial nor postprandial plasma levels of PYY3-36 showed a statistically significant difference between GES and sham GES.
DISCUSSION
Figure 2. Mean consumption of three food choices with GES and sham GES (g/day) in group A (a) and group B (b). During the three-choice food preference test, a significant change in food cravings was induced by GES. Dogs on GES ate significantly less high-fat food than those on sham GES. However, there were no differences in consumption of either high-carbohydrate food or balanced food between the periods of GES and sham GES. For high-fat food, during the period of GES, dogs in group A ate 34.6 ± 16.9% less compared with sham GES (151 ± 43.8 vs. 230.3 ± 16.3 g/day, p < 0.05). Dogs in group B ate 47 ± 14.1% less compared with sham GES (138.7 ± 35 vs. 272.3 ± 45.6 g/day, p < 0.05). FD, high-fat diet; CD, high-carbohydrate diet; BD, balanced diet.
Dogs in group B ate 5 ± 1.5 g/day during the period of GES and 3 ± 1.5 g/day during the sham GES period. Although a trend of increased consumption of the balanced food existed in both groups with GES compared with sham GES, there was no significant difference between the periods of GES and sham GES. Figure 2 shows the mean consumption of the three food choices daily during GES and sham GES in group A and group B.
486
Gastrointestinal Peptides As shown in Table 2, the plasma concentrations of ghrelin, PYY, and total GLP-1 did not differ significantly between the periods of GES and sham GES. Although during the period of GES a trend of increase in preprandial plasma ghrelin existed in both groups compared with sham GES (5.7 ± 2.3 vs. 3.9 ± 1.6 ng/mL for group A, 7.3 ± 1.2 vs 6.7 ± 3.1 ng/mL for group B), ghrelin levels showed no statiswww.neuromodulationjournal.com
In this study, we investigated the effects of chronic adjustable GES, including its effects on gastrointestinal peptides. The results showed that food consumption was reduced significantly by adjustable GES. Furthermore, in the three-choice food preference test, a significant change in food cravings was induced by GES. Finally, GES had no significant effects on plasma gastrointestinal peptide levels. All dogs completed the study, and no side effects of GES were observed. GES has been studied as a potential treatment for obesity in many previous animal studies. However, in some clinical studies, no loss of excess body weight was observed (12–14). A randomized clinical trial with 189 obese patients in the USA reported no significant weight loss in those undergoing active stimulation compared with those undergoing sham stimulation over 12 months (12). With more intensive investigation, experiments have documented the importance of individual parameters for the effectiveness of GES (5,15,16). Clinical studies have reported individual variability in visceral sensitivity. There is also a wide diversity of visceral responses to GES among individuals (16). Moreover, resistance (i.e., diminution in the response to GES after long-term use) may occur following chronic stimulation (9). The increase in body weight and food intake that occurred during chronic stimulation indicated a diminution in the response to long-term GES. Based on differences in individual sensitivity, adjustment of the GES parameters was required over a wide range. In our experiment, the reaction of each dog to chronic GES was different. A trend in individual variations was observed with respect to the initial effective stimulation intensity (with energy of 5 to 96 s·mA2) and the maximum effective stimulation intensity (with energy of 96 to 672 s·mA2). In the three-choice food preference test, a significant change in food cravings was induced by GES. Despite free access to three food choices, animals on GES decreased total food intake by 18.2 ± 5.4% in group A and 34.5 ± 1.2% in group B as compared with sham GES. Although no significant differences in consumption of highcarbohydrate or balanced food between GES and sham GES were observed, GES induced a significant decrease in consumption of high-fat food. Animals on GES in groups A and B ate, respectively, 34.6 ± 16.9% and 47 ± 14.1% less high-fat food than with sham GES. Reducing the intake of high-fat food reduced total food consumption. Our results indicate that GES is associated with reduced food intake and food craving. GES not only reduced food consumption but also had an important role in food choices. GES may modulate individual appetite in terms of both quantity and quality. Our study sheds new light on the idea that diet habits may be changed by this treatment. Val-Laillet et al. showed similar effects of vagus nerve stimulation (VNS) on food craving (17). Long-term VNS was performed in eight pigs, and sweet craving was reduced by chronic VNS. Dedeurwaerdere et al. suggested that the olfactory bulbs play a considerable role in food intake during VNS (18). Bodenlos et al.
© 2014 International Neuromodulation Society
Neuromodulation 2014; 17: 483–489
GASTRIC ELECTRICAL STIMULATION IN DOGS
Table 2. Plasma Concentrations of Gastrointestinal Peptides in Different Periods. Peptide Ghrelin (ng/mL) Peptide YY3-36 (ng/mL) Total glucagon-like peptide 1 (pM)
Time Preprandial Postprandial Preprandial Postprandial Preprandial Postprandial
Group A Baseline
GES
Sham GES
Group B Baseline
GES
Sham GES
4.7 ± 1.2 3.1 ± 0.6 5.5 ± 1.9 6.5 ± 1.1 7.2 ± 4.3 29.7 ± 4.2
5.7 ± 2.3 5.3 ± 0.2 8.7 ± 3.6 7.3 ± 3.3 9.7 ± 2.7 23.4 ± 8.1
3.9 ± 1.6 4.3 ± 1.3 5.5 ± 2.7 5.5 ± 0.8 12.8 ± 2.5 32.1 ± 6.5
7.3 ± 3.7 4.7 ± 1.0 4.6 ± 1.1 6.2 ± 2.0 8.6 ± 6.7 29.4 ± 10.7
7.3 ± 1.2 5.0 ± 0.5 3.2 ± 1.3 3.6 ± 0.9 9.5 ± 5.9 22.1 ± 12.7
6.7 ± 3.1 5.9 ± 1.4 3.0 ± 0.9 3.3 ± 1.2 7.3 ± 3.4 23.0 ± 9.3
The plasma concentrations of the three gastrointestinal peptides did not differ significantly between the periods of GES and sham GES. GES, gastric electrical stimulation.
www.neuromodulationjournal.com
As with PYY3-36, GLP-1 is considered a very important anorexigenic hormone. After GLP-1 or GLP-1 receptor agonists are administered centrally or peripherally to animals or peripherally to humans, food intake and body weight are reduced (27,28). Postprandial levels may be elevated as early as two days after gastric bypass (29). On the other hand, fasting levels of GLP-1 are decreased after bariatric surgery (30). Whitson found (31) that GLP-1 levels were significantly altered in morbidly obese nondiabetic patients six months after gastric bypass surgeries, and no significant change was noted in a matched cohort of diabetic patients. These results suggest that changes in GLP-1 levels are not the only mechanism responsible for producing weight loss after gastric bypass. In the present study, although during GES a trend toward a decrease in total postprandial plasma GLP-1 levels was observed in both groups as compared with during sham GES; this difference was not statistically significant. Because the study sample was small and exhibited high interindividual variability, we cannot make a definitive conclusion regarding this outcome. Our results suggest that the peripheral levels of ghrelin, PYY3-36, or total GLP-1 may not influence the effects of GES on food intake. Thus, it is likely that the effects of GES are complex, and levels of other peptide hormones such as cholecystokinin, obestatin, oxyntomodulin, and pancreatic polypeptide could be affected by GES. Both previous studies and the present study indicate that food intake and diet craving can be significantly altered by adjustable GES. Because dietary habits are important to body energy homeostasis, GES may be useful for treating obesity by changing food preferences. Further clinical studies are necessary to highlight the effect of adjustable GES on eating behavior.
Acknowledgements We would like to thank Dr. Kunpeng Liu for his support during the animal anesthesia.
Authorship Statements Drs. Xiaojuan Guo, Yanmei Li, Shukun Yao, Shaoxuan Chen, Yuhui Du, and Zhihua Wang contributed to experiment design, conduct, and analysis. Shukun Yao and Shaoxuan Chen conducted the animal anesthesia and surgery. Xiaojuan Guo and Yanmei Li prepared the manuscript and prepared the figures with important intellectual input from Dr. Shukun Yao. Shukun Yao, Yuhui Du, and Zhihua Wang designed the stimulation and recording system. All authors reviewed the manuscript.
© 2014 International Neuromodulation Society
Neuromodulation 2014; 17: 483–489
487
also proposed, based on data in 33 participants, that interaction between interoceptive and exteroceptive signals may explain altered food craving during VNS (19). However, the mechanisms relating GES and food cravings are still not well elucidated. More studies are necessary to elucidate the interaction between GES and food preferences. Body energy homeostasis is largely regulated by hunger and satiety signals resulting from gut–brain interactions. Satiety signals are regulated by peptides synthesized and released from enteroendocrine cells in the gastrointestinal tract. These cells release various peptide hormones that act as a signal to the central nervous system to regulate appetite. Although there are several studies of the effects of GES on peptide hormones in the central neural system, gastrointestinal tissue, and peripheral plasma, the results remain inconclusive (7,20). Ghrelin is a 28-amino-acid peptide hormone. It is produced predominantly in the stomach and represents the only known orexigenic gut hormone identified to date (21). When administered either peripherally or centrally to rodents, ghrelin rapidly increases food intake and body weight (22). Ghrelin has also been shown to stimulate appetite in humans and has been proposed to function as a meal initiator, in part due to its potent appetite-stimulating effects (23). Recent studies have demonstrated that GES may also regulate the secretion of ghrelin. GES was found to reduce gastric fundus ghrelin in diet-induced obesity rats (20). After six months of GES in obese patients, peripheral plasma ghrelin levels increased, and weight loss was correlated significantly with increased ghrelin levels (24). These data suggest that a compensatory phenomenon may appear after weight loss induced by GES. On the other hand, recent studies showed that GES did not modulate preprandial plasma ghrelin levels (4), and GES also had no effect on postprandial plasma ghrelin levels (25). In the present study, we likewise found that chronic GES did not modulate preprandial or postprandial plasma ghrelin levels. Although during the period of GES a trend towards an increase in preprandial plasma ghrelin levels was clear in both groups as compared with sham GES, the difference between GES and sham GES was not statistically significant because of high interindividual variability. PYY3-36 levels were not affected by chronic adjustable GES in this study. Circulating levels of PYY3-36 are influenced by meal composition and calorie content and elevated within one hour postfeeding (11). PYY is considered a potent anorexigenic hormone. Significant increases in circulating PYY levels have been reported following gastrointestinal surgeries, possibly contributing to the weight loss attributed to these procedures (26). However, in our studies, neither preprandial nor postprandial plasma levels of PYY3-36 were significantly different between GES and sham GES.
GUO ET AL.
How to Cite this Article: Guo X., Li Y., Yao S., Chen S., Du Y., Wang Z. 2014. The Effects of Individual Gastric Electrical Stimulation on Food Craving and Gastrointestinal Peptides in Dogs. Neuromodulation 2014; 17: 483–489
REFERENCES
488
1. World Health Organization. Obesity and overweight. In: Fact sheet no. 311. http:// www.who.int/mediacentre/factsheets/fs311/en/ 2. Zhang J, Maude-Griffin R, Zhu H et al. Gastric electrical stimulation parameter dependently alters ventral medial hypothalamic activity and feeding in obese rats. Am J Physiol Gastrointest Liver Physiol 2011;301:G912–G918. 3. Zhang J, Tang M, Chen JD. Gastric electrical stimulation for obesity: the need for a new device using wider pulses. Obesity (Silver Spring) 2009;17:474–480. 4. Li S, Maude-Griffin R, Sun Y et al. Food intake and body weight responses to intermittent vs. continuous gastric electrical stimulation in diet-induced obese rats. Obes Surg 2013;23:71–79. 5. Guo X, Li Y, Yao S et al. Parameter selection and stimulating effects of an adjustable gastric electrical stimulator in dogs. Obes Surg 2014;24:78–84. 6. Mizrahi M, Ben Ya’acov A, Ilan Y. Gastric stimulation for weight loss. World J Gastroenterol 2012;18:2309–2319. 7. Gallas S, Sinno MH, Boukhettala N et al. Gastric electrical stimulation increases ghrelin production and inhibits catecholaminergic brainstem neurons in rats. Eur J Neurosci 2011;33:276–284. 8. Ouyang H, Xing J, Chen JD. Tachygastria induced by gastric electrical stimulation is mediated via alpha- and beta-adrenergic pathway and inhibits antral motility in dogs. Neurogastroentero Motil 2005;17:846–853. 9. Shikora SA. Implantable gastric stimulation for the treatment of severe obesity. Obes Surg 2004;14:545–548. 10. Yin J, Chen JD. Retrograde gastric electrical stimulation reduces food intake and weight in obese rats. Obes Res 2005;13:1580–1587. 11. Perry B, Wang Y. Appetite regulation and weight control: the role of gut hormones. Nutr Diabetes 2012;16:e26. 12. Shikora SA, Bergenstal R, Bessler M et al. Implantable gastric stimulation for the treatment of clinically severe obesity: results of the SHAPE trial. Surg Obes Relat Dis 2009;5:31–37. 13. Hoeller E, Aigner F, Margreiter R et al. Intragastric stimulation is ineffective after failed adjustable gastric banding. Obes Surg 2006;16:1160–1165. 14. Shikora SA. “What are the Yanks doing?” The U.S. experience with implantable gastric stimulation (IGS) for the treatment of obesity—update on the ongoing clinical trials. Obes Surg 2004;14:S40–S48. 15. Yao SK, Ke MY, Wang ZF et al. Visceral sensitivity to gastric stimulation and its correlation with alterations in gastric emptying and accommodation in humans. Obes Surg 2005;15:247–253. 16. Yao SK, Ke MY, Wang ZF et al. Visceral response to acute retrograde gastric electrical stimulation in healthy human. World J Gastroenterol 2005;11:4541– 4546. 17. Val-Laillet D, Biraben A, Randuineau G et al. Chronic vagus nerve stimulation decreased weight gain, food consumption and sweet craving in adult obese minipigs. Appetite 2010;55:245–252. 18. Dedeurwaerdere S, Cornelissen B, VanLaere K et al. Small animal positron emission tomography during vagus nerve stimulation in rats. A pilot study. Epilepsy Res 2005;67:133–141. 19. Bodenlos JS, Kose S, Borckardt JJ et al. Vagus nerve stimulation acutely alters food craving in adults with depression. Appetite 2007;48:145–153. 20. Xu J, McNearney TA, Chen JD. Gastric/intestinal electrical stimulation modulates appetite regulatory peptide hormones in the stomach and duodenum in rats. Obes Surg 2007;17:406–413. 21. Hameed S, Dhillo WS, Bloom SR. Gut hormones and appetite control. Oral Dis 2009;15:18–26. 22. Castañeda TR, Tong J, Datta R et al. Ghrelin in the regulation of body weight and metabolism. Front Neuroendocrinol 2010;31:44–60. 23. Druce MR, Wren AM, Park AJ et al. Ghrelin increases food intake in obese as well as lean subjects. Int J Obes (Lond) 2005;29:1130–1136. 24. Cigaina V, Hirschberg AL. Plasma ghrelin and gastric pacing in morbidly obese patients. Metabolism 2007;56:1017–1021. 25. Sanmiguel CP, Haddad W, Aviv R et al. The TANTALUS system for obesity: effect on gastric emptying of solids and ghrelin plasma levels. Obes Surg 2007;17:1503– 1509. 26. Korner J, Bessler M, Cirilo LJ et al. Effects of Roux-en-Y gastric bypass surgery on fasting and postprandial concentrations of plasma ghrelin, peptide YY, and insulin. J Clin Endocrinol Metab 2005;90:359–365. 27. Tourrel C, Bailbe D, Meile MJ et al. Glucagon-like peptide-1 and exendin-4 stimulate beta-cell neogenesis in streptozotocin-treated newborn rats resulting in persistently improved glucose homeostasis at adult age. Diabetes 2001;50:1562– 1570. 28. Vilsboll T, Zdravkovic M, Le-Thi T et al. Liraglutide, a long-acting human glucagonlike peptide-1 analog, given as monotherapy significantly improves glycemic
www.neuromodulationjournal.com
control and lowers body weight without risk of hypoglycemia in patients with type 2 diabetes. Diabetes Care 2007;30:1608–1610. 29. le Roux CW, Welbourn R, Werling M et al. Gut hormones as mediators of appetite and weight loss after Roux-en-Y gastric bypass. Ann Surg 2007;246:780– 785. 30. Reinehr T, Roth CL, Schemthaner GH et al. Peptide YY and glucagon-like peptide-1 in morbidly obese patients before and after surgically induced weight loss. Obes Surg 2007;17:1571–1577. 31. Whitson BA, Leslie DB, Kellogg TA et al. Entero-endocrine changes after gastric bypass in diabetic and nondiabetic patients: a preliminary study. J Surg Res 2007;141:31–39.
COMMENTS Appropriate and adjustable parameters in neuromodulation continue to be an intense focus of clinical and basic science research. For many years empirical parameters were utilized for various forms of pain and other treatments. Stimulus frequencies were generally kept between 40 and 125 Hz but the frequency generally centered around 50 Hz and pulse widths often varied between 100–200 μs. Our personal experience that encouraged us to expand various parameters resulted from using an animal model to test higher frequencies of spinal cord stimulation on changes in hind limb peripheral blood flow (1). For several years we used 50 Hz as the “gold standard” for studying peripheral blood flow. However, in the study by Gao et al. (2010), spinal cord stimulation at 500 Hz significantly increased cutaneous blood flow and decreased vascular resistance compared to changes induced by frequencies of 50 and 200 Hz. Recently clinical and basic science studies have expanded to study effective frequencies extending from 4 Hz to 10 K Hz and higher. In addition to variation in frequencies, studies have shown that selection of appropriate pulse widths is also a key parameter that can improve treatment with neuromodulation. In the present study Gastric Electrical Stimulation (GES) was used to treat obesity by affecting food craving in a canine model. The authors pointed out that commercial implanted devices mainly adapted from a nerve stimulator were not appropriate for producing effective responses for GES. In fact, Dr. Jiande Chen and his colleagues (2) published an article stating that a new device with wider pulses was essential for treating obesity with GES. The present study amplifies the importance using an adjustable stimulator with a wide range of stimulation parameters, particularly pulse widths. This study highlighted the importance using adjustable parameters. After the animal completely recovered from the surgery to implant the electrodes on the greater curvature of the stomach, each animal was tested with a selection of pulse widths to define optimal settings; the other parameters were relatively fixed. The selection of the appropriate pulse width was based on the total symptom score. Their results suggested that GES was associated with reduced food intake. An intriguing result was that GES with different parameters not only reduced food consumption but also had an impact on the choice of food. One surprising result was that the parameters selected for reducing food consumption and choice did not require significant changes in gastrointestinal peptides including ghrelin, PYY3-36 and GLP-1. Would these peptides levels change if the electrodes using adjustable parameters were placed on another region of the stomach? The authors suggested that this study may provide new insights about using their approach to change diet habits.
© 2014 International Neuromodulation Society
Robert Foreman, PhD Oklahoma City, OK, USA
Neuromodulation 2014; 17: 483–489
GASTRIC ELECTRICAL STIMULATION IN DOGS REFERENCES (1) Gao J, Wu M, Li L, Qin C, Farber JP, Linderoth B, Foreman RD. Effects of spinal cord stimulation with “standard clinical” and higher frequencies on peripheral blood flow in rats. Brain Res 2010 Feb 8;1313:53–61. (2) Zhang J, Tang M, Chen JD. Gastric electrical stimulation for obesity: the need for a new device using wider pulses. Obesity (Silver Spring) 2009;17:474–480.
*** This work showed that gastric electrical stimulation (GES) in dogs decreased their food intake and body weight after three months of GES. Although such effects of GES are well-known phenomena (reviewed in [1]), the authors applied a novel approach consisting in adjusting parameters of GES over time to achieve the maximal tolerated energy delivered by GES. As long as there is no efficient pharmacological anti-obesity treatment, GES can be considered as a therapeutic option being less invasive than bariatric surgery. Several questions remain however unanswered such as the molecular mechanism underlying food intake reducing effects of GES and if these effects of GES in normal weight dogs can be directly applicable to obese humans. In fact, the obese animals often do not respond to the same treatment that reduce food intake and body weight in the non- obese [2]. Furthermore, the authors did not find any significant effect of GES on plasma gastrointestinal hormone concentration including ghrelin, GLP1 and PYY that would explain observed changes in food intake. However, a recent study showed that in obese subjects increased affinity of plasma immunoglobulins for ghrelin enhances ghrelin’s
orexigenic effect with otherwise normal or reduced plasma ghrelin levels [3]. It would be, hence, of interest to see if long-term effects of GES on appetite may involve modulation of production and properties of immunoglobulins reactive with hunger and satiety hormones after either high or low energy GES. In fact, acute GES with Enterra parameters (low energy) in rats was shown to stimulate ghrelin secretion [4] but also increased plasma levels of ghrelin-reactive IgG [1]. Sergueï O. Fetissov, MD, PhD Rouen University, France
REFERENCES 1. Gallas S, Fetissov SO. Ghrelin, appetite and gastric electrical stimulation. Peptides 2011;32:2283–2289. 2. Fetissov SO, Meguid MM. Serotonin delivery into the ventromedial nucleus of the hypothalamus affects differently feeding pattern and body weight in obese and lean Zucker rats. Appetite 2010;54:346–353. 3. Takagi K, Legrand R, Asakawa A et al. Anti-ghrelin immunoglobulins modulate ghrelin stability and its orexigenic effect in obese mice and humans. Nat Commun 4:2685, doi: 10.1038/ncomms3685 (2013). 4. Gallas S, Sinno MH, Boukhettala N et al. Gastric electrical stimulation increases ghrelin production and inhibits catecholaminergic brainstem neurons in rats. European Journal of Neuroscience 2011;33:276–284.
Comments not included in the Early View version of this paper.
489
www.neuromodulationjournal.com
© 2014 International Neuromodulation Society
Neuromodulation 2014; 17: 483–489
