SEARCH

SEARCH BY CITATION

Keywords:

  • atropine;
  • cisapride;
  • gastric emptying;
  • rats;
  • stable isotopes

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgment
  9. References

Abstract Currently available rat models for measuring gastric emptying are hampered by the necessity to kill the animals at the end of each experiment, which makes repetitive testing impossible. We have developed and validated a noninvasive test model, adapted from the13C-octanoic breath test in humans, for repetitive measurements of gastric emptying in rats. Male Wistar rats were trained on a fixed protocol to eat a piece of pancake doped with 1 μg13C-octanoic acid after 12 h fasting, and to stay thereafter in cylindrical glass cages. Breath tests were performed by a fully automated system of computer-guided switching valves, which collected consecutive breath samples. All breath samples were analysed by gas chromatography and isotope mass spectrometry. The area under the curve (AUC) from the cumulative13CO2excretion from 0 to 6 h was determined by the trapezium method to calculate the gastric half-emptying times (t½). Inter-day variability was determined. The effect of subcutaneous or intraperitoneal injection of saline was studied. The test was further validated for pharmacological interventions by oral administration of cisapride and parenteral administration of atropine, to induce, respectively. acceleration and delay of gastric emptying. Mean gastric emptying times ± SD of 24 rats were 119.3 ± 28.2 min, 138.7 ± 26.0 min, and 124.5 ± 30.9 min on three different test days. The mean coefficient of variation of three repeated measurements in the same 24 rats was 17.5%. No significant differences were observed after subcutaneous or intraperitoneal injection of saline. In a second test series of eight rats, cisapride significantly accelerated gastric emptying (mean t½ 112.7 ± 33.1 min, P < 0.05), while atropine caused a significant delay (mean t½ 205.9 ± 24.9 min, P < 0.05) when compared to control test results (mean t½ 140.7 ± 16.7 min) in the same rats. We validated the13C-octanoic breath test to study gastric emptying in rats. This test method obviates the necessity to kill laboratory animals and allows repetitive measurements of gastric emptying to study its physiology or pathophysiology as well as the effect of pharmacological agents.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgment
  9. References

Radioscintigraphy is generally accepted as the ‘gold standard’ for measuring gastric emptying in humans.1 The 13C-octanoic breath test has been widely applied in clinical diagnosis of gastric emptying disorders since the first report of this test by Ghoos et al.2 It is more suitable for repetitive measurements and applications in children because no radioactive substance is used. However, neither of these two test methods has been validated for laboratory animals. Most rat models currently used for measuring gastric emptying are hampered by the necessity to kill the animals at the end of the test, which makes repetitive testing impossible, therefore, these gastric-emptying studies are limited to transversal comparisons of single measurements. For solids, most investigators use Amberlite pellets or glass beads, administered by gavage, to measure proportional retention in the stomach after fixed time intervals.3 For liquid gastric emptying, phenol red is dissolved in a fairly viscous solution and administered by gavage.4 Concentration and volume measurements of intragastric contents are used to calculate the percentage of dye retention after killing the animal. Both methods limit scientific yield to one figure for one animal and do not allow intraindividual comparisons. This last would be of utmost importance for interventional pharmaceutical trials. In addition, anaesthesia is often necessary to permit a quick and reliable administration of dye or pellets intragastrically by gavage. These manipulations cause an nonphysiological stress to the laboratory animal, which can significantly influence motility patterns. Trials with imaging techniques, mostly based on radioscintigraphy, are flawed by important movement artefacts. These can be overcome by movement-limiting chambers or general anaesthesia, however, these also impose an important stress on the laboratory animal.5–8

Breath tests in small laboratory animals have recently been reported as a noninvasive method to measure liver function9 or gastric emptying.10 However, laboratory experiments of CO2 sampling from living or anaesthesized rats were performed previously, and were designed to study the in vivo metabolism of sodium acetate11 or aminopyrine.12,13 H2 sampling from expired breath was reported to measure orocaecal transit time in rats after oral ingestion of lactose.14 Symonds et al.10 reported a method to measure gastric emptying repetitively in mice, by adapting the13C-octanoic breath test. They measured intra-animal reproducibility by reporting coefficients of variation for three parameters: gastric half-emptying time, gastric emptying coefficient and lag time.

We have developed a fully automated method to measure gastric emptying repetitively in rats, based on the 13C-octanoic breath test for humans. We compared gastric half-emptying times intraindividually in basal conditions and after administration of cisapride and atropine.

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgment
  9. References

Test animals and training

Five series of eight healthy male Wistar rats were trained twice weekly, for 3 weeks, on a fixed time schedule, to eat a piece of pancake after 12-h overnight fasting. After finishing this meal, the rats were placed in 3 L glass cages, allowing free movement, for 6 h. In between the training periods, they were given continuous access to water ad libitum and standard rat feed (Pavan Service, Turnhowt Belgium). They were kept in a room with constant temperature at 22 °C with free exposure to daylight. Their mean weight at initiation of the training was 248 g (range 221–270 g), and this had increased to 310 g (range 248–360 g) the morning before starting the first breath test.

Test meal

One egg yolk, doped with 400 µg 13C-octanoic acid, was mixed with 450 mL skimmed milk and 200 g of commercial pancake powder (Oetker, Belgium). This resulted in a dough with a caloric content of 580 kJ 100 g−1 (caloric contents 72% carbohydrates, 10% lipids, 18% proteins). Following this, 50 mL of the mixture was baked for 4 min and then divided into pieces of 1 g, each containing 1.0 µg of 13C-octanoic acid.

Breath test set-up

Eight glass containers were adapted with inlet and outlet valves, connected to 2-mm rubber tubes Fig. 1. Through the inlet valve a continuous flow of humidified CO2-free synthetic air (Alphagaz, Luik, Belgium) passed for 6.5 min, after which the inlet and outlet valves were closed to allow accumulation of expired CO2 from the rat. After exactly 3.5 min, the outlet valve was automatically opened and a breath sample was aspired trough the outlet tubing via a multiswitch valve. After closure of the outlet valve, the inlet valve opened and renewal of the air was started. Aspirated breath samples passed through a multichannel computer-guided switching valve, and were thereafter injected automatically by an autosampler (Gilson, Analis, Belgium) in silicone-free glass tubes (BD Vacutainer Systems, Plymouth, UK). Each of the eight cages was consecutively opened at 1-min intervals, resulting in a complete breath sampling of eight rats within a period of 8 min. This automated system was repeated every 10 min for 6 h. Before the meal, two baseline breath samples per rat were collected. After finishing their meal, rats were placed immediately in their cage to start the breath test.

image

Figure 1. Breath test set-up.

Download figure to PowerPoint

Test protocols

We set the time interval between two breath tests at 3 or 4 days, to eliminate possible interference by accumulation or retention of 13CO2. This interval corresponded to the interval of 3–4 days between each training session during the previous 3 weeks. All tests were started at 09.00 hours to preclude any possible influence from circadian rhythm. To exclude possible interference from other 13C sources in the test meal, breath tests were performed in eight rats who had eaten a pancake after the addition of 12C-octanoic acid. In a first experimental protocol, day-to-day variability was measured in 24 rats, based on three consecutive tests. In a similar protocol, one series of eight rats received an intraperitoneal (i.p.) injection of 1 mL saline at the second test day, 20 min before the start of the breath test. Another series of eight rats was given 1 mL saline subcutaneously (s.c.), 20 min before the second breath test.

A second experiment was designed to evaluate pharmacological interventions with cisapride and atropine. Eight rats underwent four consecutive breath tests on day 1, 4, 8 and 11. On day 1 and 11, control breath tests measured gastric emptying in basal conditions to calculate long-term reproducibility. On day 4, cisapride was administered intragastrically by gavage at a dose of 3 mg kg−1 of a 1-mg mL−1 oral solution (Prepulsid syrup; Janssen-Cilag, Beerse, Belgium), 20 min before starting the test meal. On day 8, atropine was given simultaneously s.c. and i.p. at equivalent doses (2.5 mg kg−1), 12 min before the test meal.

Data analysis

Breath samples were analysed by gas chromatography followed by isotope ratio mass spectrometry to determine the enrichment delta of the ratio of exhaled 13CO2 vs. 12CO2. From these measurements, it is possible to calculate the cumulative fractions of the administered dose excreted per unit of time (C%dose/h, Fig. 2). The area under the curve (AUC) over 6 h was calculated by the trapezium method. We defined the gastric half-emptying time (t½) as the time interval after which half of the total AUC had been obtained.

image

Figure 2. Example of an excretion curve. x, percentage dose points; o, cumulative percentage dose points. The area under the curve (AUC) of the cumulative fractions over 6 h was calculated by the trapezium method. The gastric half-emptying time was defined as the time after which half of the total AUC had been obtained.

Download figure to PowerPoint

Statistical analysis

Based on the Kolmogorov–Smirnov test, a normal distribution was accepted and data were presented as mean ± standard deviation (SD). To evaluate intraindividual reproducibility, the coefficient of variation was calculated for the gastric half-emptying time. To compare gastric emptying data between different test conditions, one-way analysis of variance for repeated measures (anova) and paired Student's t-test were used. P < 0.05 was considered significant.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgment
  9. References

Excretion curves showed a steep increase of 13CO2 during the first hour, with a gradual decrease thereafter for several hours. Excretion values remained significantly above baseline, however, until 6 h after the start time (Fig. 2). After ingestion of pancake with the addition of 12C-octanoic acid, there was no detectable 13CO2 excretion above the baseline values.

In the first experiment, the mean t½ of 24 rats on three different test days was 119.3 ± 28.2 min, 138.7 ± 26.0 min, and 124.5 ± 30.9 min, respectively (Fig. 3, anovaP > 0.05). The mean coefficient of variation calculated on three repeated measurements of 24 rats (intraindividual comparisons) was 17.5%. On pairwise comparisons, we observed however, a significant slowing down of gastric emptying between test day 1 and test day 2 (paired t-test P < 0.05), and a nonsignificant acceleration on test day 3 compared with test day 2. Test day 1 and 3 showed comparable results. No significant differences could be observed after i.p. or s.c. injection of saline (Fig. 4ab). The second test series (Fig. 5) showed a significant acceleration of gastric emptying after administration of cisapride (mean t½ 112.7 ± 33.1 min), compared with the status on day 1 (mean t½ 140.7 ± 16.7 min) and on day 11 (mean t½ 142.5 ± 33.7 min; t-test P < 0.05). After administration of atropine, a significant delay of gastric emptying (mean t½ 205.9 ± 24.9 min) was observed compared with both control series (t-test P < 0.05). The results from control series on day 11 were not different from those of the first control series on day 1.

image

Figure 3. Gastric half-emptying times (mean ± SD) of three consecutive breath tests in 24 rats, with a time interval of 3 days. No significant differences between groups were observed (one-way anova for repeated measures, P < 0.05). On pairwise comparison, test day 2 was significantly slower than test day 1 (Student's t-test, P < 0.05).

Download figure to PowerPoint

image

Figure 4. (a) Gastric half-emptying times (mean ± SD) of three consecutive breath tests in eight rats, with intraperitoneal injection of saline 20 min before the second test. No significant differences were found. (b) Gastric half-emptying times (mean ± SD) of three consecutive breath tests in eight rats, with subcutaneous injection of saline 20 min before the second test. No significant differences were found.

Download figure to PowerPoint

image

Figure 5. Evaluation of pharmacological interventions with cisapride and atropine in eight rats. Gastric half-emptying times (mean ± SD) of four consecutive tests with intragastric administration of cisapride 20 min before the second test, and intraperitoneal + subcutaneous administration of atropine 12 min before the third test. *Cisapride caused a significant acceleration of the gastric emptying when compared to both control series (Student's t-test, P < 0.05). Atropine caused a significant delay when compared to both control series (Student's t-test, P < 0.05). There was no significant difference between the two control series.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgment
  9. References

We have developed a fully automated system to measure gastric emptying simultaneously in eight rats by a 13C-octanoic breath test. We validated this test for research applications. Day-to-day variability for the gastric half-emptying time on three repeated measurements was 17.5%. Injection of saline (s.c. or i.p.) did not influence gastric emptying rates. As no other validated technique exists to compare intraindividual gastric emptying in rats, we performed an intrinsic validation. After oral administration of cisapride we demonstrated a significant acceleration, and after parenteral administration of atropine we demonstrated a significant delay of gastric emptying.

The 13C-octanoic breath test has been widely applied in clinical diagnosis of gastric emptying disorders in humans since the first report of this test by Ghoos et al.2 The test is based on the assumption that postgastric processing of octanoic acid, in normal conditions, is a quick and uniform process. Although criticism of this test methodology has risen recently, it remains a valuable technique in pharmacodynamic or intraindividual studies in humans.15–17 The initially reported coefficient of variation for gastric half-emptying time of 26.7%2 was comparable with earlier reports of reproducibility of scintigraphic data.1,18,19 As a major advantage, the breath test allowed repetitive measurements without radiation burden, so that it could be applied in children and for pharmacological studies.

The majority of gastric-emptying studies in laboratory animals, almost invariably rats, has until now been performed in a setting that necessitated killing of the animals after each single experiment. The cost in time and more importantly, in animal life, to obtain nonreproducible test results is high. The first report of noninvasive repeated measurements of gastrointestinal transit in rats dates from 1989, when Enck14 and colleagues described a hydrogen breath test to determine orocaecal transit time. They used test boxes of 2.5 L volume, from which cumulative exhaled air was sampled at fixed time intervals in a syringe inserted through a rubber valve. No anaesthesia or gavage was applied, in order to avoid stress reactions with increase of plasma cortisol or catecholamines. The first experiments of 13CO2 or 14CO2 measurements in respired breath of rats studied metabolism of sodium acetate after oral administration11 and of aminopyrine after intravenous (i.v.) administration,12,13 necessitating an implanted jugular cannula or i.v. injection during anaesthesia.

The only study in which the 13C-octanoic acid breath test in animals was applied was published very recently.10 The mouse was used as a test animal, and different test meals were compared. The authors used curve fitting by least-squares analysis and nonlinear regression analysis as previously described by Ghoos et al.,2 to calculate three parameters: lag time, half-emptying time and gastric emptying coefficient. However, neither the mathematical model nor the test method as a whole was validated by an intrinsic or extrinsic validation procedure, and duration of the test was changed to 180 min. A problem encountered by these authors was difficult curve fitting due to a biphasic gastric emptying curve in one rat (of a total of seven tested animals), resulting in rejection of these data for analysis. We avoided this shortcoming by using the trapezium method to calculate an AUC based on the raw data. Half-emptying time was derived directly from the time point at which half of the AUC had been obtained. Background 13CO2 excretion was ruled out as a confounding variable by performing breath tests on rats after they had ingested pancakes doped with 12C-octanoic acid. The mean coefficient of variation of gastric half-emptying time within the same rat in this study was comparable to figures from breath test studies with humans (12–27%)2,15,16 and mice (24.3%).10 The significant difference in the control series (Fig. 3) between the first and second test days could theoretically indicate a spontaneous slowing down of gastric emptying by repetitive testing, but this is refuted by the acceleration on the third test day. The test series with injection of saline, also performed in triplicate, did not show a similar trend, nor did the control series at day 11 of the second experiment, compared with day 1. In preparation for pharmacological intervention studies, we examined the possible confounding influence of s.c. and i.p. saline injections as a stressful event; there was no detectable effect of these injections on gastric emptying.

Scintigraphic methods for measuring gastric emptying in rats have been reported.6–8 This technique allows for repeated measurements in vivo. However, no standardized method has been described. Diverse techniques of data acquisition and calculation of gastric emptying have been reported, and two papers.6,7 refer to liquid gastric emptying. In addition, some important differences from our test setting exist, such as the use of restraining cages,6 a restraint period by holding the rat in front of a camera,7 or general anaesthesia.8 These differences would hamper a comparison of one of these scintigraphic methods with our technique. The only way of validating this breath test would be to administer simultaneously Amberlite pellets in a solid phase to the rats before the breath test, but then killing of different rats would be necessary at fixed time intervals. This would interrupt the breath test prematurely, so that half-emptying times could not be measured any longer. Therefore, we decided to validate our test pharmacologically by demonstrating an acceleration as well as a slowing down of gastric emptying. Cisapride is a well-known prokinetic agent with a dose-dependent effect on gastric emptying in rats.20–23 We preferred gavage without anaesthesia as an oral method of administration to ascertain an exact ingested dose. The observed acceleration was significant, but more specific studies of dose-dependency and preferred route of administration are needed. Atropine is also well validated as a gastroparetic agent after intramuscular, subcutaneous or intravenous injection in rats.3,4,20,24 Differences from control series were highly significant in our series.

The advantages of our technique in comparison with scintigraphy are the minimal stress for the test animal (no restraining cage, no anaesthesia, spontaneous eating of the test meal), the use of a solid test meal as a more physiological way of examining gastric emptying, and the possibility of examining eight rats at the same time, under identical conditions. Moreover, no gamma irradiation or gamma camera equipment is necessary.

Conclusion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgment
  9. References

We have developed a 13C-octanoic acid breath test model for Wistar rats, and validated this test method by studying reproducibility, and the effect of cisapride and atropine on gastric emptying rates. The major advantage of this noninvasive method is the possibility of intraindividual comparisons at baseline conditions. It allows to study physiology and pathophysiology of gastric emptying in rats, and to perform pharmacological intervention studies by repeating the test with varying doses or formulations of a molecule presumed to influence gastric emptying. The present study gives strong evidence for applicability of the 13C-octanoic acid breath test in gastric emptying studies in rats. In addition, this test should give an alternative to euthanasia of laboratory animals for gastric-emptying studies in the future.

Acknowledgment

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgment
  9. References

This work was supported by a grant from NFWO Belgium, n°31521401 and from Glaxo–Wellcome Belgium.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgment
  9. References
  • 1
    Heading R, Tothill P, McLoughlin G, Shearman D. Gastric emptying rate measurement in man. A double isotope scanning technique for simultaneous study of liquid and solid components of a meal. Gastroenterology 1976; 71: 4550.
  • 2
    Ghoos Y, Maes B, Geypens B, Mys G, Hiele M, Rutgeerts P, Vantrappen G. Measurement of gastric emptying rate of solids by means of a carbon-labeled octanoic acid breath test. Gastroenterology 1993; 104: 16407.
  • 3
    Matsuda M, Aono M, Moriga M, Okuma M. Centrally administered neuropeptide Y (NPY) inhibits gastric emptying and intestinal transit in the rat. Dig Dis Sci 1993; 38: 84550.
  • 4
    Tamhane M, Thorat S, Rege N, Dahanukar SJ. Effect of oral administration of Terminalia chebula on gastric emptying: an experimental study. Postgrad Med 1997; 43: 123.
  • 5
    Griffith G, Owen G, Kirkman S, Skields R. Measurement of rate of gastric emptying using chromium-51. Lancet 1966; i: 12445.
  • 6
    Haruta S, Kawai K, Jinnouchi S, Ogawara K, Higaki K, Tamura S et al. Evaluation of absorption kinetics of orally administered theophylline in rats based on gastrointestinal transit monitoring by gamma scintigraphy. J Pharm Sci 2001; 90: 46473.
  • 7
    Green G, Guan D, Schwartz J, Phillips W. Accelerated gastric emptying of glucose in Zucker type 2 diabetic rats. role in postprandial hyperglycemia. Diabetologia 1997; 40: 13642.
  • 8
    Duval-Araujo I, Queiroz D, Magnago A, Simal C, Marino V, Carvalho S et al. Increased gastric emptying induced by Helicobacter heilmannii type 1 infection in rats. J Med Microbiol 2000; 49: 62734.
  • 9
    Ishii Y, Asai S, Kohno T, Suzuki S, Ishii M, Hosoi I et al. 13CO2 peak value of L-(1–13C) phenylalanine breath test reflects hepatopathy. J Surg Res 1999; 86: 1305.DOI: 10.1006/jsre.1999.5705
  • 10
    Symonds E, Butler R, Omari T. Assessment of gastric emptying in the mouse using the (13C)-octanoic acid breath test. Clin Exp Pharmacol Physiol 2000; 27: 6715.DOI: 10.1046/j.1440-1681.2000.03318.x
  • 11
    Ampulski R, Boggs R. A comparison between 14CO2 and 13CO2 measurements in the respired breath of rats fed sodium acetate. Biomed Mass Spectrom 1977; 4: 98102.
  • 12
    Willson R, Hart F, Hew J. Breath analysis of 14CO2 production from aminopyrine in the normal rat. Res Comm Chem Pathol Pharmacol 1979; 23: 50521.
  • 13
    Kotake A, Schreider B, Latts J. The in vivo measurement of expired 14CO2 derived from the N-methylation of aminopyrine as a reflection of the in vitro hepatic cytochromeP-450 drug-metabolism activity in rats. Drug Metabol Disposition 1982; 10: 2518.
  • 14
    Enck P, Wienbeck M. Repeated noninvasive measurement of gastrointestinal transit in rats. Physiol Behav 1989; 46: 6337.
  • 15
    Choi M-G, Camilleri M, Burton D, Zinsmeister A, Forstrom L, Nair K. 13C-octanoic acid breath test for gastric emptying of solids. accuracy, reproducibility, and comparison with scintigraphy. Gastroenterology 1997; 112: 115562.
  • 16
    Choi M-G, Camilleri M, Burton D, Zinsmeister A, Forstrom L, Nair K. Reproducibility and simplification of13C-octanoic acid breath test for gastric emptying of solids. Am J Gastroenterol 1998; 93: 928.
    Direct Link:
  • 17
    Delbende B, Perri F, Couturier O, Leodolter A, Mauger P, Bridgi B et al. 13C-octanoic acid breath test for gastric emptying measurement. Eur J Gastroenterol Hepatol 2000; 12: 8591.
  • 18
    Collins P, Horowitz M, Cook D, Harding P, Shearman D. Gastric emptying in normal subjects-a reproducible technique using a single scintillation camera and computer system. Gut 1983; 24: 111725.
  • 19
    Jonderko K. Short- and long-term reproducibility of radioisotopic examination of gastric emptying. Nucl Med Biol 1990; 17: 297301.
  • 20
    Yamano M, Kamato T, Miyata K. Participation of a cholinergic mechanism in 5-hydroxytryptamine (5-HT)3,5 HT4 receptor-mediated stimulation of gastric emptying in rats. Arzneim-Forsch Drug Res 1997; 47 (II): 12426.
  • 21
    Yamano M, Kamato T, Nagakura Y, Mijata K. Effects of gastroprokinetic agents on gastroparesis in streptozotocin-induced diabetic rats. Arch Pharmacol 1997; 356: 14550.
  • 22
    Ueki S, Matsunaga Y, Yoneta T, Tamaki H, Itoh Z. Gastroprokinetic activity of Nizatidine during digestive state in the dog and in the rat. Arzneim-Forsch Drug Res 1999; 49 (II): 61825.
  • 23
    Hatanaka S, Kawarabayashi K, Iseri M, Tsubokura K, Furuhama K. Enhancing effect of DQ-2511 on gastric emptying of spontaneously hypertensive rats. Life Sci 1995; 56: 37782.
  • 24
    Borella LE, Lippmann W. A simple non-radioactive method for simultaneous quantitative determination of stomach emptying and intestinal propulsion in the intact conscious rat. Digestion 1980; 20: 3649.