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.
Center for Digestive Diseases, GI Motility Program, 8730 Alden Drive, Thalians Building, Second Floor East, Los Angeles, CA 90048. E-mail: firstname.lastname@example.org
Objective: The objective of this study was to validate the use of impedance for measurement of antral contractions and to determine the relationship between food-induced changes in gastric motility and satiation.
Research Methods and Procedures: In Experiment 1, three dogs were implanted with an antral strain gauge and bipolar electrodes for measurement of local tissue impedance. Impedance and strain gauge recordings were obtained simultaneously during antral contractions to correlate impedance changes with contractile events. In Experiment 2, seven dogs were implanted with two pairs of gastric electrodes for simultaneous recording of slow wave activity and impedance. The changes in the rate of slow waves and of antral contractions assessed by impedance during food intake were characterized.
Results: Variations in strain gauge amplitude were highly correlated with changes in antral impedance (R2: 0.70 to 0.82, p < 0.05). In Experiment 2, slow wave rate was significantly reduced after food intake and reached a nadir at satiation (5.0 ± 0.3 vs. 3.8 ± 0.5 events/min, p < 0.001). Likewise, the amplitude of antral contractions assessed by variations in impedance was significantly increased after food intake, peaking at satiation (5.3 ± 1.4 vs. 12.2 ± 4.3 Ohms, p < 0.01).
Discussion: Measurement of impedance is a reliable tool for assessing gastric contractility. Food ingestion significantly reduces slow wave rate and enhances antral contractions. Peak changes in these two variables occur at the time of satiation. Electrical measurements of both slow waves and impedance may be used to estimate gastric motility and satiation.
The consumption of food normally lasts until people are comfortably full. This perception of fullness is believed to result from activation of satiation signals generated primarily in the gastrointestinal tract (1). These short-term signals are triggered by the chemical properties and mechanical effects of the food (2, 3) and help regulate energy intake by maintaining appropriate meal size. Mechanical factors, such as distension or contractions, activate mechanoreceptors in the stomach that, in turn, activate afferent neurons, primarily vagal, that communicate with the nucleus of the solitary tract (4, 5, 6). Also, the ingestion of food changes the electromechanical activity of the stomach to increase contraction amplitude and to decrease slow wave rate (SWR)1 (7, 8). However, the temporal relationship between satiation and food-induced changes in gastric electromechanical activity is not well defined. Exploring this relationship may provide better insights into the mechanisms that modulate satiation.
The aims of this study were 1) to correlate changes in antral tissue impedance with antral mechanical activity to validate the use of electrical impedance as a tool for assessment of gastric contractile function and 2) to determine the temporal relationship between changes in gastric electromechanical activity and satiation.
Research Methods and Procedures
This experiment was designed to validate the use of impedance to assess gastric mechanical activity, by simultaneous impedance and strain gauge (SG) recording. Three mongrel adult dogs (weight, range: 18.5 to 20 kg) underwent laparotomy after induction with 20 mg/kg thiopental and inhalation of 2% isoflurane. Two pairs of temporary myocardial electrodes (A&E Medical, Farmingdale, NJ) were implanted within the seromuscular layer of the anterior surface of antrum. The distance between the electrodes used to measure impedance was 4 cm. Impedance was measured by delivery of a balanced, non-depolarizing stimulus at a frequency of 2.5 kHz and duration of 60 µs, which delivered a current of a few microamperes. An SG (model F-08ISP; Star Medical Corp., Tokyo, Japan) was placed between this pair of electrodes. This allowed the simultaneous recording of both muscle contractions and impedance at the same location. A muscle contraction lowers the impedance by decreasing the distance between electrodes, increasing tissue cross-section area, and improving electrode–tissue interface, which result in reduced electrical resistance. Thus, changes in impedance can be used to estimate mechanical activity. A second pair of electrodes, spaced 0.5 to 1 cm apart, was placed in parallel and superior to the SG and along the lesser curvature. This pair was used to detect slow wave (SW) activity (Figure 1). The electrodes and SG wires were exteriorized along the flank and connected to a custom made recording device (a prototype of the implantable pulse generator), kept in the pocket of a canine vest. The three signals (SG, SW, and impedance) were recorded simultaneously. Data were band-pass filtered (0.5 to 15 Hz) and sampled at a rate of 200 Hz and transmitted to an outside recorder through a radiofrequency wand.
This experiment attempted to define the temporal relationship between food-induced changes in gastric electromechanical activity and the time of meal termination.
Seven female mongrel dogs (weight, 20.6 ± 2.3 kg) underwent a laparotomy under general anesthesia, with 20 mg/kg thiopental intravenously and 1% to 2% inhaled isoflurane. Two pairs of temporary myocardial electrodes (A&E Medical) were implanted within the seromuscular layer of the stomach as follows. One pair of electrodes was placed on the anterior surface of the antrum at 4 cm from the pylorus. Mechanical activity of the antrum was determined by measuring the impedance between these electrodes. The second pair of electrodes, used to detect antral SWs, was placed superiorly to the first pair. The electrodes for impedance were spaced 4 cm apart, and those used to record SWs were spaced 0.5 to 1 cm apart. Electrodes were exteriorized along the flank and connected to an external data logger/pulse generator (TANTALUS; MetaCure, Orangeburg, NY) capable of sampling impedance and electrical signals.
Recording of Electromechanical Activity
Simultaneous recordings of gastric SWs and antral contractions (as determined by changes in impedance between electrodes) were obtained continuously before and after meals. Data were band-pass filtered (0.5 to 15 Hz), sampled at a rate of 200 Hz, and transmitted to an outside recording device (TANTALUS) using a radiofrequency wand. Data were analyzed using custom-designed software in a portable laptop computer (series tablet PC keyboard; Hewlett-Packard, Seoul, Korea). Impedance was measured as described in Experiment 1.
Experiments started 2 to 3 weeks after the surgery. Dogs were fed once a day. They were started with a preload dose of 100 grams of dry food, followed in 10 minutes by an amount (800 grams) of the same food, which exceeded their normal daily intake. This regimen lengthened the duration of meal consumption and tested the response to both an initial low quantity of food and a full meal. Dogs were fed outside their cages, in a room in the presence of an observer. A habituation period of at least 1 week was applied before the start of experiments. Satiation was defined as time of last food intake, as determined by the observer who watched the dogs during feeding.
The protocol was reviewed and accepted by the Institutional Animal Care and Use Committee at the Cleveland Clinic Foundation (Cleveland, OH).
Contractions by SG and changes in electrical impedance were recorded continuously for 20 minutes before food intake and up to 30 minutes after the meal. The difference between maximum and minimum of contraction-associated deflections in the impedance and SG channels was measured.
Antral impedance and the contractions detected by SG were randomly sampled for 3 to 4 minutes during fasting, just before the meal, during meal, and in the postprandial period for a total of ∼15 minutes in each dog. Recording was performed on 3 separate days, and the correlation between the two variables was assessed by linear regression.
SWR and changes in antral impedance were determined for 5 minutes during each of six periods: 1) baseline, just before eating; 2) at the end of the preload; 3) 10 minutes before satiation; 4) at satiation; 5) 10 minutes after satiation; and 6) 30 minutes after satiation. The interval between SWs was analyzed by proprietary Matlab-based software. A repeated-measures ANOVA was used to assess difference between periods.
Data are presented as mean ± standard deviation and p < 0.05 for significance.
A total of 52.7 ± 2.3 contractions per dog were assessed by SG, both during fasting and in the postprandial period. There was a significant linear correlation (R2: 0.70 to 0.82, p < 0.05) between the amplitude of the contractions detected by SG and the amplitude of the concomitant changes in antral impedance, as detected by the electrodes (Figure 2). As seen in Figure 2, antral contractions determined by SG were associated with negative deflections in the antral impedance channel. Baseline impedance value was 447.5 ± 75 Ohms. The median value of impedance changes below baseline, associated with SG events, was 16 Ohms, with a range of 1 to 75 Ohms, and the value at the fifth percentile level was 3.65 Ohms. Based on these results, we considered a decrease from baseline impedance >3.65 Ohms to represent a contraction.
Total volume consumed was 338 grams (range, 284 to 478 grams). Dogs consumed the preload meal in <1 minute. Initiation of the meal and preload was characterized by an abrupt decrease in SWR and an increase in the amplitude of antral contractions. Changes in SWR and amplitude of contractions increased with consumption of more food until satiation (Figure 3). At satiation, SWR slowed to its nadir, and antral contractions tended to reach peak amplitude. SWR was 5.0 ± 0.3 events/min at baseline, 4.6 ± 0.4 events/min at preload, 4.6 ± 0.4 events/min at 10 minutes before satiation, 3.8 ± 0.6 events/min at satiation, 4.4 ± 0.4 events/min at 10 minutes after satiation, and 4.4 ± 0.4 events/min at 30 minutes after satiation. Amplitude of contractions at the same time-points was 5.3 ± 1.4, 8.5 ± 2.1, 9.2 ± 2.8, 12.2 ± 4.3, 11.7 ± 3.3, and 11 ± 5.6 Ohms. Differences were statistically significant at satiation and during the postprandial period compared with baseline (Figure 4). Dogs usually stopped eating just before or at about the time of these peak changes. If more than one feeding interval occurred, the slowest electrical activity rate often occurred at the end of the last eating episode. Average time from beginning of the meal to satiation was 17.8 ± 5.7 minutes. The mean time interval between satiation point and the slowest rhythm was 80 ± 40 seconds. By this time, contractions either reached maximal amplitude or were >90% of maximal amplitude.
We found that changes in tissue impedance correlated well with antral contractions. Furthermore, food ingestion resulted in progressive reduction of SWR and increased the amplitude of antral contractions that peak at the time of meal termination.
The results of the simultaneous recording from impedance electrodes and SGs in the first part of the study suggest that electrical signals from implanted electrodes provide reliable data on gastric contractile activity. Contractions were always associated with a negative deflection in impedance. The opposite, an increase in impedance, occurs when the distance between electrodes is increased during distension (9). This novel method has a number of advantages over existing technologies. It precludes the need for placement of intraluminal recording devices, which are particularly difficult to maintain in laboratory animals. This method can provide data from ambulatory studies conducted over long periods, and importantly, can provide data from other segments, such as the fundus. Electrodes are also more simple and durable compared with SGs. More importantly, with the use of gastric electrical stimulation in humans, implanted electrodes can provide valuable data on gastric contractile activity.
Gastrointestinal signals are considered to play a major role in inducing satiation (1). These signals modulate eating behavior through the release of various gut peptides and also through activation of mechanoreceptors, primarily in the stomach (10). In turn, activation of such mechanoreceptors generates neural signals in afferent nerves, primarily vagal, that terminate in the nucleus of the solitary tract (5, 10). Food-induced activation of such mechanoreceptors results both from distension of the stomach and from gastric muscle contractions (6, 11). Food ingestion also decreases SW frequency and increases the amplitude of gastric contractions (7, 8). However, the pattern of these changes and their temporal relationship to satiation has not been studied.
This study was designed to examine changes in SWR and the amplitude of gastric muscle contractions produced by food ingestion. It has been shown that SWR is diminished by food ingestion and gastric distension (8, 12, 13). Gastric distension likely plays an important role in the decrease of SWR. In a study by Lin et al. (14) using various volumes of non-nutritive buffered solution on SWR in a canine model while concomitantly measuring gastric emptying, it was found that SWR varied as a function of intragastric volume. These investigators found that SWR decreased linearly with increasing volumes introduced into the stomach or with larger volumes remaining in the stomach. A similar linear decrease in SWR was found in dogs when the stomach was distended with increasing volumes using balloon inflation (15). However, the relationship between these changes and satiation was not assessed.
We found that the SWR gradually decreased, and the amplitude of contractions gradually increased as the volume of food consumed increased. Of particular interest is the observation that the amplitude of antral contractions reached its peak, and SWR reached its nadir, at the time when dogs stopped feeding. These results point toward an association between these events and satiation.
It is well documented that antral distension and antral contractions activate vagal afferents (16). In carefully dissected vagal afferent fibers from the rat stomach, Peles et al. (6) were able to show that both antral distension and induced antral contractions, using non-excitatory electrical stimulation, increase firing of vagal afferent fibers in an intensity-dependent manner. Moreover, the combined stimuli had an additive effect on vagal afferent firing. It is also well known that the activation of gastric vagal afferents is an important factor in the mechanism of satiation, because vagal afferent signals modulate responses to food in centers in the hypothalamus (10). Thus, one can postulate that afferent signaling peaks at the time when contractions reach their peak amplitude and that this time coincides with the onset of satiation and meal interruption. This information may be valuable for the treatment of obesity. Increasing the amplitude of contractions in the early stage of meal intake may reduce food consumption by mechanically induced enhancement of afferent signaling. It was recently shown that non-excitatory stimulation that applies pulse trains synchronized to slow waves after food intake can amplify antral contractions in obese subjects and promote weight loss (17). The findings described in this study suggest that enhanced contractions, timed to the early part of the meal when spontaneous contractions are still not at peak level, may be one of the mechanisms leading to weight loss induced by gastric electrical stimulation (17).
In summary, this study showed that changes in tissue impedance can be used to reliably measure antral contractions. It also showed a clear association between gastric electromechanical events and satiation that may point toward a possible causal relationship between these two events. This association may potentially allow gastric electrical stimulation to modulate satiation signals.
This study was supported, in part, by a grant from the Binational Industrial Research and Development (BIRD) Foundation.