Application of the lactose 13C-ureide breath test for measurement of equine orocaecal transit time


  • D. G. M. SUTTON,

    Corresponding authorSearch for more papers by this author

    1. School of Veterinary Medicine, College of Medical, Veterinary and Life Sciences, University of Glasgow, and Isotope Biochemistry Laboratory, Scottish Universities Environmental Research Centre, Scottish Enterprise Technology Park, East Kilbride, UK.
    Search for more papers by this author
  • S. LOVE

    1. School of Veterinary Medicine, College of Medical, Veterinary and Life Sciences, University of Glasgow, and Isotope Biochemistry Laboratory, Scottish Universities Environmental Research Centre, Scottish Enterprise Technology Park, East Kilbride, UK.
    Search for more papers by this author



Reasons for performing study: Application of the lactose 13C-ureide breath test (LUBT) for measurement of equine orocaecal transit time (OCTT) has not been reported previously. The ability to assess OCTT noninvasively, and to investigate its relationship to gastric emptying rate and small intestinal transit, would be of both clinical and research value.

Objectives: 1) Assessment of the LUBT in healthy horses, with comparison of induced versus noninduced test protocols. 2) Application of a new dual stable isotope breath test (lactose 13C-ureide and 13C-octanoic acid) for gastrointestinal transit measurement.

Hypothesis: The LUBT will allow quantification of equine OCTT, and test efficacy will be enhanced by prior administration of lactose 12C-ureide as shown in vitro. The dual tracer breath test will permit simultaneous measurement of gastric emptying, OCTT and small bowel transit times.

Methods: Induced and noninduced LUBTs were performed in 3 healthy mature horses in randomised order using a standard test meal and protocol. Combined LUBT and 13C-octanoic acid breath tests (13C-OABT) were performed in 4 individuals on 4 occasions at weekly intervals. Expiratory isotopic recovery was modelled to allow generation of gastric emptying data, small bowel transit times and caecal transit parameters.

Results: The induction protocol for the LUBT increased the rate and magnitude of expiratory 13CO2 significantly. Mean ± s.d. values for OCTT, caecal lag phase (tlag) and caecal t1/2 using the induced LUBT were 3.24 ± 0.65 h, 5.62 ± 1.22 h and 6.31 ± 1.21 h, respectively. Dual stable isotope tests resulted in the production of 2 discrete peaks in expiratory 13CO2 in 15/16 tests from which gastric t1/2, OCTT and small bowel transit (SBT) parameters could be calculated.

Conclusions: The induced LUBT provides a reliable noninvasive measure of equine OCTT and can be paired with the 13C-OABT to provide further information about small intestinal motility.


Disordered small intestinal motility is demonstrable in equine syndromes such as post operative ileus (Gerring and Hunt 1986) and chronic grass sickness (Milne et al. 1996), but also may be involved in conditions such as ileal impaction (Proudman et al. 1998; Little and Blikslager 2002), idiopathic or secondary ileal muscular hypertrophy (Chaffin et al. 1992) and ileo-ileal or ileocaecal intussusception (Ford et al. 1990). Proudman and Holdstock (2000) reported tapeworm infestation as a cause of small intestinal disruption, whilst intraluminal S. vulgaris larval antigen has been demonstrated to reduce ileal myoelectrical activity and cause colic (Berry et al. 1986). Intra-abdominal inflammation, abscessation or neoplasia also may cause colic signs by affecting small intestinal transit of ingesta. A major difficulty in investigating many of these conditions clinically has been the absence of a simple noninvasive test for measurement of equine OCTT. In addition to facilitating the diagnostic investigation of such cases, a noninvasive modality for OCTT measurement would permit determination of the effects of specific drugs, such as potential prokinetic agents and commonly used sedative, analgesic and anaesthetic agents, on small intestinal motility.

The lactose 13C-ureide breath test (LUBT) is a safe, noninvasive, stable isotope test that has been used to evaluate OCTT (Heine et al. 1995) and has been validated against the predicate method of gastroenterocolonic scintigraphy (Camilleri et al. 1998) for measurement of this parameter in Man (Geypens et al. 1999). Since being validated the LUBT has proved useful particularly in the investigation of small intestinal disorders in paediatric gastroenterology (van den Driessche et al. 2000). In vitro intestinal fermentation studies have demonstrated that equine microbial digestion of lactose 13C-ureide is restricted to the large bowel, and this stable isotope has been validated as a marker of equine OCTT (Sutton et al. 2011). A major aim of this study was to apply the LUBT in vivo, and to determine whether the induced test protocol recommended from in vitro studies was optimal for clinical purposes. Prior exposure to lactose 12C-ureide has been shown to increase subsequent bacterial degradation of lactose 13C-ureide in man (Mohr et al. 1999), ruminants (Merry et al. 1982) and horses (Sutton et al. 2011), with a consequent improvement in the accuracy of OCTT detection.

A further aim was to combine the LUBT with the 13C-OABT in a novel dual stable isotope test, to determine whether or not this was of value in investigation of small bowel transit, and its relation to gastric emptying rate and OCTT. The 13C-OABT has been validated against the optimum technique of gastric radioscintigraphy in both Man (Ghoos et al. 1993) and horses (Sutton et al. 2002a, 2003) for the measurement of solid phase gastric emptying rate. The clinical application of the dual stable isotope test has been reported in paediatric medicine (Van den Driessche et al. 1997), where it has proved useful for differentiating disorders of gastric emptying from those of small intestinal motility in a variety of conditions. In human medicine, the dual stable isotope test has provided an alternative to whole gut transit scintigraphic studies (Camilleri et al. 1991; Bonapace et al. 2000) for evaluation of severe gastrointestinal dysfunction. As there is no clinical alternative in equine medicine, the dual stable isotope test if successful could open exciting new avenues of clinical investigation.

Materials and methods


Four healthy mature horses (mean age 14 years, range 10–20 years) with no history or clinicopathological evidence of gastrointestinal disease were used in this study. Each subject was maintained on a constant diet of ryegrass seed hay (30 g/kg bwt daily) throughout the 3 month study period to minimise fluctuations in basal 13CO2 output. A regular daily routine was maintained in order to minimise physiological fluctuations in gastrointestinal motility. The study was approved by the Glasgow University Ethics and Welfare Committee.

Study design

Induced and noninduced 13C-LUBTs were each performed on 2 occasions in 3 healthy mature individuals in randomised order. These tests were performed at a minimum interval of 7 days, and 14 days were allowed for microbial de-adaptation to occur before performing a noninduced test. The study design allowed paired comparisons to be made of the effect of the induction process in vivo in each individual.

Dual stable isotope tests were then performed in 4 subjects, at weekly intervals and repeated on 4 occasions. Once again, fluctuations in diet or management were avoided to minimise changes in gastrointestinal motility.

The lactose 13C-ureide breath test

Fermentation studies in vitro (Sutton et al. 2011) showed that equine lactose 13C-ureide (13C-LU) digestion is optimised by microbial exposure to lactose 12C-ureide (12C-LU) at least 14 h previously in a ratio of approximately 5:1. For the induced LUBT each subject was given 15 mg/kg bwt β-lactosyl 12C-ureide dihydrate1 in a small feed of oats and bran 19 h before administration of 3.0 mg/kg bwt β-lactosyl 13C-ureide dihydrate1 in the standard test meal. The induction time interval was chosen to ensure arrival of 12C-LU in the caecum by 14 h before the start of the test.

Two breath samples were collected for basal 13CO2 measurement before ingestion of the test meal at time 0 h. Thereafter, expiratory breath samples were collected in duplicate at 15 min intervals for 6 h, followed by 30 min intervals for 6 h, then 60 min intervals for a further 2 h. Breath samples were collected using a modified Aeromask2 and QuinTron3 collection bag, and transferred to Exetainers4 prior to isotope ratio mass spectrometric analysis of 13C:12C ratio (ABCA)5. The breath collection and analysis technique has been validated as detailed previously (Sutton et al. 2003).

The noninduced LUBT was performed in the same way but without the administration of the priming dose of 12C-LU. As for every stable isotope test, the subjects were maintained at rest for the duration of each test procedure to minimise fluctuations in VCO2 and basal 13CO2 production, and were fasted for 14 h prior to test meal ingestion so that the stomach was empty.

The dual stable isotope breath test

The protocol for the dual stable isotope test was exactly as for the induced LUBT but with the addition of 1 mg/kg bwt 13C-octanoic acid6 to the test meal. The dose of 12C/13C-LU remained unchanged, as did the schedule for expiratory breath collection.

Test meal composition

A standard meal composition was used for each breath test consisting of 150 g crimped oats, 100 g wheat bran and 200 ml water. This had been validated previously as not affecting baseline 13CO2 production post ingestion (Wyse et al. 2001). In order to keep the stable isotope tracers in the solid phase of gastric emptying, the 13C-LU was homogenised with one egg white, and the 13C-octanoic acid was added to egg yolk (one yolk per 250 mg tracer). The isotope-labelled carriers were then cooked in a microwave until firm, before being finely chopped and mixed thoroughly with the test meal. The average time taken for complete test meal consumption was 5 min.

Calculation of caecal transit parameters

Stable isotope data were expressed and plotted as percentage dose recovery (PDR) of the administered isotope/h, from which transit parameters were calculated. PDR/h is dependent on the enrichment of 13CO2 in the expiratory breath post ingestion, the parts/million (ppm) excess 13C in the administered dose of tracer, and also the total recovery of 13C in labelled CO2, based on estimation of resting VCO2 (Equation 1):

image(Equation 1)

where ppm excess 13Ct is the ppm excess 13C provided by the dose of tracer. The atom% excess is the increase over basal in sample 13C abundance post tracer ingestion, expressed relative to the international Peedee Belemnite standard. For this study, VCO2 was estimated using the formula of Gallivan et al. (1989) where VCO2= 2.84 ml/kg bwt/min.

Orocaecal transit time was defined as that time at which 3% recovery of the final total cumulative dose recovery of the isotope had occurred, and was modelled using a delayed Maes formula (Maes et al. 1994). This power exponential (Equation 2) was fitted by the least squares method using the Excel Solver function7, and the integral of the model used to calculate OCTT (Equation 3):

image(Equation 2)
image(Equation 3)

where y(t) is the % 13CO2 excretion in breath/h; t is time (h); a, b and c are rate constants and d is the delay factor. This estimate of OCTT incorporates a defined point, correlating to the time at which a fixed quantity of the labelled test meal has reached the caecum and been fermented, and is not dependent on fluctuations in baseline 13CO2 output. The time to maximal rate of caecal delivery and fermentation of the 13C-LU, defined as caecal tlag, was calculated using the formula given in Equation 4:

image(Equation 4)

The time after ingestion corresponding to cumulative caecal digestion of 50% of the labelled test meal, caecal t1/2, was calculated by integration, to determine time to 50% recovery of the total recovered 13C dose at infinity (Equation 5):

image(Equation 5)

The time to achieve a cumulative 10% caecal delivery and fermentation of the labelled test meal, caecal t10%, was calculated by the same integrative process (Equation 6):

image(Equation 6)

Calculation of gastric and small intestinal transit parameters

For the dual stable isotope test, a multi-peak Maes formula was used to model the dose recovery data, incorporating multiple ‘d’ or ‘delay’ terms (Equation 7):

image(Equation 7)

where a, b and c are rate constants. Nonlinear regression was used to produce the best modelled fit to the 13C recovery data. The gastric emptying and caecal transit parameters were derived from the first and second modelled peaks, respectively. Gastric tlag and gastric t1/2 were calculated from the first modelled peak in the same manner as for the caecal formulae given in Equations 4 and 5. Gastric tlag represents the time from ingestion prior to maximal rate of delivery of stable isotope to the small intestine, i.e. maximal gastric emptying rate. Gastric t1/2 provides an indirect measure of the time taken for 50% of the labelled test meal to be delivered to, and absorbed from, the duodenum. As post gastric processing of 13C-octanoic acid is constant in rate, gastric emptying rate is the major determinant in generation of gastric t1/2 by the 13C-OABT.

Small bowel transit (SBT) rate has not been reported previously in horses, and the methods of Read et al. (1986) and Bennink et al. (1999), respectively, were used here to calculate SBT in terms of both 50% (SBt1/2) and 10% (SBt10%) transit times:

image(Equation 8)
image(Equation 9)

Small bowel t1/2 has been reported to approximate the scintigraphic deconvolution process (Read et al. 1986), estimating the time taken for half of the test meal to move through the small intestine alone, whilst SBt10% primarily assesses transit of the leading column of chyme from duodenum to caecum (Bennink et al. 1999). The relationship between the different gastric and small intestinal transit parameters was assessed by simple linear regression and statistical significance determined using Pearson's coefficient of correlation.


All individuals produced a peak in expiratory 13CO2 output after ingestion of lactose 13C-ureide allowing estimation of transit parameters. The mean expiratory dose recovery curves for each individual are presented (Fig 1). In each case the induction process resulted in an increased rate of recovery of 13CO2 in the breath. Furthermore, prior exposure to 12C-LU resulted in an increased cumulative dose recovery of 13C-LU. Mean ± s.d. cumulative dose recoveries for the induced and noninduced test protocols were 50.66 ± 13.52 and 34.82 ± 4.74 %, respectively (n = 6, t= 2.708, P<0.05). Initial deviation of expiratory 13CO2 from basal output was more distinct with the induced test protocol, thus increasing accuracy of OCTT estimation.

Figure 1.

Mean expiratory dose recovery of 13C-LU in 3 mature horses, using an induced or noninduced protocol for the LUBT. Each test was performed on 2 occasions at a minimum of 7 days' interval. The mean values and standard deviation are given at each point. Data are modelled using the formula ofMaes et al. (1994).

The OCTT in the 3 subjects was estimated to be significantly less when the induced LUBT protocol was used at 3.24 ± 0.65 h vs. 5.12 ± 1.01 h (n = 6, P<0.05), as was estimation of tlag and t1/2 (Table 1). Mean intraindividual coefficient of variance (CV%) for OCTT using the induced protocol was 14.95%. One of 6 induced LUBTs produced a small early peak in 13CO2 output that returned to baseline prior to the major recovery of isotope, consistent with small intestinal fermentation.

Table 1. Intestinal transit parameters derived from 3 mature horses using the lactose 13C-ureide breath test, giving comparison between mean results gained with the induced (n = 6) and noninduced (n = 6) test protocols. Each protocol was performed twice in each individual in randomised order
Test/parameterOC transit time (h)Caecal tlag (h)Caecal t1/2 (h)
  • *

    denotes significant difference (P<0.05). OC, orocaecal; caecal tlag, time from ingestion to maximal caecal filling rate; caecal t1/2, time until expiratory recovery of 1/2 of total isotopic dose recovered; CV%, mean intra-individual coefficient of variation.

Induced lactose3.24*0.6520.085.62*1.2221.636.31*1.2119.12
13C-ureide BT         
Without induction5.12*1.0119.709.42*1.3714.5810.43*1.5915.28

For the dual stable isotope tests, isotopic recovery was biphasic in 15/16 tests, such that the gastric emptying and caecal filling components of 13C recovery were easily separable, and amenable to fitting with the dual peak model. In the remaining test, a 4 peak model was required to fit the 13C recovery data, reflecting 2 discrete gastric emptying events. On this occasion the subject had paused for 15 min between eating the first and second halves of the test meal. The dual peak model was most accurate to apply when gastric emptying was rapid and complete, due to wider separation of the 2 peaks in 13C output. When there was less separation between 13C peaks due to coincidence of small intestinal 13C-octanoate absorption with arrival of the head of 13C-LU-labelled chyme in the caecum, mathematical curve-fitting was more complex.

The dual isotope breath test results for a 14-year-old Welsh Mountain Pony mare are illustrated (Fig 2) and tabulated (Table 2) for comparison. Gastric t1/2 is seen to be highly variable, ranging from 1.58–3.25 h, while OCTT, SBt10% and SBt1/2 parameters are more conserved. Mean intestinal transit parameters in the 4 individuals are given in Table 3. Using the dual stable isotope test (n = 16), mean ± s.d. OCTT was estimated to be 3.66 ± 0.60 h, similar to the single induced LUBT results, with a CV% of 16.43%. However, intra- and interindividual variability in gastric t1/2 using the dual isotope test (CV% 34.17%) was relatively high and higher than reported previously (Wyse et al. 2001).

Figure 2.

Combined 13C-OABT/LUBT results in a 14-year-old Welsh Mountain Pony mare. Results from weeks 1–4 are shown clockwise from top left. The modelled gastric emptying and caecal transit profiles are shown, plus the combined expiratory recovery of 13CO2 (thick line). Solid diamonds mark the actual isotopic recovery against time.

Table 2. Weekly intestinal transit parameters in a 14-year-old Welsh Mountain Pony mare as measured by the dual stable isotope test on 4 occasions, and corresponding to Figure 2. The mean, s.d. and coefficient of variation (CV%) values are shown for each parameter
Subject 4Gastric parameters (h)Caecal parameters (h)SB transit (h)
  1. The formula used for generation of each intestinal transit parameter is given in the Materials and methods section of the text.

Week 11.101.913.253.964.816.647.413.714.15
Week 20.741.291.582.783.535.126.422.794.84
Week 30.650.752.313.474.235.836.873.584.56
Table 3. Mean, s.d. and coefficient of variation (CV%) intestinal transit parameters in 4 mature horses derived using the combined 13C-OABT/LUBT weekly for 4 weeks
All testGastric parameters (h)Caecal parameters (h)SB transit (h)
(n = 16)t10%tlagt1/2OCTTt10%tlagt1/2t10%t1/2

The correlation between caecal t10% and small bowel t10% was positive and highly significant (n = 16, r2= 0.954, P<0.001) as shown (Fig 3). Significant positive correlations were also present between SBt10% and OCTT (n = 16, r2= 0.937, P<0.001); SBt10% and caecal tlag (n = 16, r2= 0.889, P<0.001); and between SBt10% and caecal t1/2 (n = 16, r2= 0.802, P<0.001). In contrast, a significant negative linear correlation was present between small bowel t1/2 and gastric t1/2 (n = 16, r2= 0.697, P<0.01) as shown (Fig 4). Within each individual there was a tendency for SBt1/2 to be relatively prolonged when gastric emptying was rapid, or shortened if gastric t1/2 was extended. No significant correlations were detected between gastric and caecal transit parameters.

Figure 3.

Linear relationship between caecal t10% (h) and small bowel t10% (h) as determined by dual isotope breath test in 4 healthy mature individuals. The combined test was performed in each subject on 4 occasions. The linear line of ‘best fit’ is shown, with a highly significant positive correlation (n=16, r2=0.954, P<0.001).

Figure 4.

Linear correlation between gastric t1/2 (h) and small bowel t1/2 (h) as measured by 13C-OABT/13C-LUBT in 4 healthy individuals on 4 occasions. A significant negative correlation is present (n=16, r2=0.697, P<0.01).


The results of this study suggest that the LUBT is a reliable indirect test for the measurement of equine OCTT but that adherence to the given induction protocol is essential to its accuracy, as is reported in man (Geypens et al. 1999). The in vivo results presented here are in agreement with in vitro studies in which equine microbial caecal 13C-LU hydrolase activity was restricted to the large bowel and specifically enhanced by prior exposure to 12C-LU (Sutton et al. 2011). The induced LUBT dose recovery curves were modelled accurately by the delayed Maes formula (Maes et al. 1994), allowing calculation of OCTT, caecal tlag and caecal t1/2 in each case. Major advantages of the LUBT for equine use, apart from its noninvasive nature and ease of application, therefore include its repeatability and specificity for OCTT measurement. Although the hydrogen breath test has been reported for the measurement of equine OCTT, it has not been validated for this purpose, and is much less repeatable (Sasaki et al. 1999). Similarly, the sulphasalazine/sulphapyridine test has been used for estimation of OCTT in horses, but has not undergone specific validation (McGreevy et al. 2001), and has greater interindividual variability than reported here for the LUBT. The sulphasalazine/sulphapyridine test also depends on cleavage of the azo-bond by intestinal bacteria, with subsequent detection of plasma sulphapyridine by high performance liquid chromatography.

The dual stable isotope tests produced useful data, which allowed simultaneous measurement of gastric emptying rate, small bowel transit and OCTT to be made. The relationship between these parameters has been difficult to obtain previously in horses due to technical difficulties in assimilating these data. The significant negative correlation found between SBt1/2 and gastric t1/2 suggests that rapid delivery of ingesta into the equine small intestine does not result necessarily in rapid transit of this material to the caecum. In fact, the tendency shown for SBt1/2 to be prolonged when gastric emptying was rapid suggests that, in the presence of nutrient-rich chyme, the equine proximal small intestine may exert control on the delivery of nutrients to the caecum by reducing propulsive contractions. This theory is supported by the significant positive correlations observed between SBt10% and the early caecal transit parameters.

Dyer et al. (2002) showed that the sodium/glucose cotransporter isoform 1 (SGLT I), responsible for the uptake of glucose and galactose, is expressed with highest abundance in the duodenum, and at approximately half that abundance in the ileum. As the SGLT 1 protein has a high affinity and low capacity for sugar substrates, proximal small intestinal regulation of nutrient delivery would have physiological significance, guarding against caecal lactic acidosis.

Excessive ileocaecal transfer of fermentable carbohydrate is of particular consequence in horses as decreased caecal pH, with potential proliferation of lactic acid producing bacteria, may lead to the development of laminitis (Garner et al. 1978). The close correlation observed here between small intestinal and caecal transit rates suggests the presence of an ileal brake mechanism in the horse, as reported in other species (Pappas et al. 1986; Spiller et al. 1988; Read and Houghton 1989; Treacy et al. 1990), in which ileal lipid receptors also regulate small intestinal transit.

The mean ± s.d. values for OCTT derived from the single LUBT (3.24 ± 0.65 h) and dual isotope studies (3.66 ± 0.60 h) were similar, with CV% of 20.08 and 16.43%, respectively. Equine solid phase OCTT is therefore shown to be relatively low, and faster than reported in man (4.86 ± 0.97 h) using the same technique (Geypens et al. 1999). Estimates of equine OCTT derived here were higher than those reported by McGreevy et al. (2001), in which a liquid phase sulphasalazine marker was used. Although small intestinal transit of liquid has been reported to be similar to that of solids (Bennink et al. 1999), such a marker would empty more quickly from the stomach, and solid phase markers are generally recommended for the investigation of clinical disorders of gut motility (Parkman et al. 1995).

Gastric t1/2 mean values (2.87 ± 0.98 h) and interindividual CV% (34.17%) derived in this study using the dual isotope test were comparable to those gained using the 13C-OABT in isolation (Sutton et al. 2002b), and solid phase gastric emptying rate appears rather variable between healthy equine subjects. However, the range in intraindividual CV% (15.84–41.22%) observed in gastric t1/2 using the dual stable isotope test was greater than anticipated. Although intraindividual variation in solid phase gastric emptying of a standard meal has not been well established in horses, the iterative curve-fitting technique used to model the dual curves may have contributed to error. Curve-fitting accuracy, and generation of transit parameters, is reduced when the respective isotopic peaks caused by absorption of 13C-octanoic acid from the small intestine, and fermentation of 13C-LU in the caecum are superimposed.

In order to validate the dual stable isotope test for measurement of equine gastric and caecal transit parameters, and improve data modelling, it would be necessary to compare it to concurrent gastric radioscintigraphy. Alternatively, the combination of a marker of liquid phase gastric emptying, such as sodium 13C-acetate, with the 13C-LU solid phase marker, may allow more clear differentiation of gastric emptying events and orocaecal transit, and improved accuracy of data analysis.

The test results reported in this study have been gained from a small sample of healthy horses. Assessment of test performance in a larger sample of both healthy and unhealthy horses is required before widespread clinical application of the stable isotope tests can be recommended. The presence of significant small intestinal bacterial overgrowth also could be a confounding factor in accurate test interpretation when using the LUBT in horses. However, such peaks in 13CO2 output are likely to be small and more transient than those corresponding to caecal substrate fermentation. The LUBT is a preferred technique for the detection of small intestinal bacterial overgrowth in man, having excellent specificity but suboptimal sensitivity (Berthold et al. 2009).

Although oral trimethoprim did not have a significant effect on equine faecal flora (White and Prior 1982), the precise effect of antibiotic administration on LUBT test efficacy has not been determined in any species. Specific disorders in which intestinal anaerobes are increased in number, such as equine grass sickness, might also prove unsuitable for test application, and have yet to be investigated. Efficacy of the LUBT in equine neonates would depend also on colonisation of the hindgut by the requisite microbes and is not likely to be present for the first few weeks of life.

However, both the LUBT and the dual stable isotope breath test should prove useful tools for investigating the role of small intestinal dysmotility in a variety of equine disorders, in both clinical and research settings. Given the risk factors now identified for the development of colic, further research into the specific effects of these factors on intestinal motility would be beneficial. The noninvasive, quantitative nature of the breath tests also makes them ideal for assessment of the effects of specific drugs, such as anaesthetic and prokinetic agents, on equine small intestinal motility.

Ethical considerations

The study was approved by the Glasgow University Ethics and Welfare Committee.

Authors' declaration of interest

No conflicts of interest have been declared.

Source of funding

Funding was provided by the Horserace Betting Levy Board (D. S.) and the School of Veterinary Medicine, College of Medical, Veterinary and Life Sciences, University of Glasgow.

Manufacturers' addresses

1 Bell College of Technology, Hamilton, South Lanarkshire, UK.

2 Trudell Medical International, London, Ontario, Canada.

3 Quintron Instrument Company, Milwaukee, Wisconsin, USA.

4 Labco Ltd, High Wycombe, Buckinghamshire, UK.

5 PDZ Europa, Northwich, Cheshire, UK.

6 Isotec Inc., Miamisburg, Ohio, USA.

7 Microsoft Corporation, Redmond, Washington, USA.