Equine piroplasmosis treatment protocols: Specific effect on orocaecal transit time as measured by the lactose 13C-ureide breath test



Reasons for performing study

Imidocarb dipropionate is the drug of choice for equine piroplasmosis but its administration causes severe colic and diarrhoea. An imidocarb protocol that reduces these effects is needed.


1) Quantification of the effects of imidocarb dipropionate on equine orocaecal transit time (OCTT), with and without atropine or glycopyrrolate premedication and 2) investigation of an improved pretreatment regimen for imidocarb administration.


Treatment with imidocarb dipropionate will result in colic and reduced OCTT as demonstrated by the lactose 13C-ureide breath test which will be ameliorated by premedication with either atropine or glycopyrrolate.


The effects of 3 drug therapies on OCTT were compared in 6 healthy horses in a randomised double-blind study vs. a saline control: 1) imidocarb dipropionate 2.4 mg/kg bwt administered intramuscularly (i.m.) with saline administered intravenously (i.v.; imidocarb/saline); 2) imidocarb dipropionate 2.4 mg/kg bwt administered i.m. with atropine 0.035 mg/kg bwt administered i.v. (imidocarb/atropine) and 3) imidocarb dipropionate 2.4 mg/kg bwt administered i.m. with glycopyrrolate 0.0025 mg/kg bwt administered i.v. (imidocarb/glycopyrrolate). The lactose 13C-ureide breath test was used to measure OCTT in each case and significance of treatment effect determined by a linear model analysis of variance.


Imidocarb/atropine treatment caused an increase in OCTT (P<0.05) whereas imidocarb/saline produced a nonsignificant decrease in OCTT. Imidocarb/saline caused colic and diarrhoea in 4 of 6 horses, which were not seen in any of the horses treated with imidocarb/atropine or imidocarb/glycopyrrolate or administered the saline control. Intestinal borborygmi were increased in imidocarb/saline and decreased in imidocarb/atropine treated horses, respectively.


Imidocarb/saline treatment induced colic signs and a potential reduction in OCTT while imidocarb/atropine treatment increased OCTT significantly when compared with imidocarb/saline. Both atropine and glycopyrrolate premedication ameliorated the clinical gastrointestinal effects of imidocarb but atropine produced significant inhibition of gastric and/or small intestinal motility not detected with glycopyrrolate. Premedication with glycopyrrolate is recommended when using imidocarb for treatment of equine piroplasmosis.


Theileria equi and Babesia caballi are the causative agents of equine piroplasmosis. This is the most common tick-borne disease of equidae in southern Africa [1] and piroplasmosis has been detected worldwide, including in recent studies in Europe [2], the United States [3], South America [4], Jordan [5] and Asia [6]. Transmission of the parasites via the infected tick may lead to haemolytic anaemia, anorexia, weight loss, transplacental infection and even death in naive individuals. Imidocarb dipropionate, a carbamate, is the drug of choice for the treatment of equine piroplasmosis [7, 8] but severe colic and diarrhoea may occur due to cholinesterase inhibition [9, 10]. Concurrent administration of atropine [1] or glycopyrrolate [7] has been advocated to minimise or prevent the side effects observed with imidocarb treatment.

Imidocarb administration causes a significant increase in frequency of defaecation, total faecal output and faecal water content in horses [11], effects ameliorated by premedication with glycopyrrolate or atropine. However, semi-quantitative assessment of gastrointestinal motility using transabdominal ultrasonography did not reveal significant differences in either small or large intestinal motility in individuals administered imidocarb [11]. Thus, neither the specific impact of imidocarb on equine orocaecal transit time (OCTT) nor its interaction with glycopyrrolate and atropine have been quantified.

The induced lactose 13C-ureide breath test has been validated against gastroenterocolonic scintigraphy for the measurement of OCTT in people [12] and, more recently, has been validated in vitro for OCTT measurement in horses [13]. The underlying principle of the lactose 13C-ureide breath test is that enzymatic splitting of the glycosylureide moiety of the stable isotope is performed only by intestinal microbes in the large bowel. Subsequent hydrolysis of the released 13C-urea component results in rapid liberation of stable 13CO2 and its appearance in the bicarbonate pool. As glycosylureide cleavage is the rate-limiting step in this process, mass spectrometric analysis of the 13C:12C ratio in expiratory breath after ingestion of the labelled test meal provides an indirect measurement of OCTT [14].

Using the lactose 13C-ureide breath test to measure OCTT, one aim of this study was to improve our knowledge of the specific effects of imidocarb treatment on equine small intestinal transit and its interaction with the anti-cholinergic compounds atropine and glycopyrrolate. The goal of quantifying the effects of these drugs was to validate an imidocarb protocol for treatment of equine piroplasmosis that resulted in minimal short-term disruption to gastrointestinal motility and function.

Materials and methods


Six healthy adult equidae (4 Thoroughbreds, 2 Basuto ponies) from the Equine Research Centre herd at The Faculty of Veterinary Science, University of Pretoria were used in this blinded, randomised prospective study. The animals' age range was 4–13 years (mean 8.2 years), with bodyweight (bwt) ranging from 339 to 545 kg (mean 460 kg).

All animals were in good body condition with no history or physical evidence of gastrointestinal disease. Biochemical and haematological parameters and faecal egg counts were within reference limits prior to inclusion in the study. During the study period, subjects were fed a constant diet of alfalfa and teff hay only (ratio 1:3, total 2% bwt/day) and allowed access to grazing. Exercise was restricted for the duration of the trial to minimise external influences on gastrointestinal motility. All subjects were stabled and fasted for 14 h prior to each test to reduce baseline shifts in 13C output [15] and ensure each had an empty stomach. The animals were primed to maximise bacterial lactose ureide hydrolase enzymatic activity by feeding unlabelled lactose 12C-ureide (15 mg/kg bwt) 16 h prior to each test as reported previously [14].

Study design

The relative effects of 3 drug therapies (imidocarb dipropionate 2.4 mg/kg bwt administered i.m. with saline administered i.v. [imidocarb/saline group]; imidocarb dipropionate 2.4 mg/kg bwt administered i.m. with atropine 0.035 mg/kg bwt administered i.v. [imidocarb/atropine group]; imidocarb dipropionate 2.4 mg/kg bwt administered i.m. with glycopyrrolate 0.0025 mg/kg bwt administered i.v. [imidocarb/glycopyrrolate group]) on OCTT were compared. The lactose 13C-ureide breath test was used in 6 equine subjects and the results compared with a saline control (saline i.m. with saline i.v. [saline control group]). In the crossover design, each animal received one blinded treatment per week in randomised order with a 7 day washout period between treatments.

Expiratory breath samples were collected in duplicate 60 and 0 min before ingestion of the labelled test meal to estimate baseline 13C output. Immediately after the test meal had been eaten, each animal was administered the designated drug combination. Breath sample collection was then continued at 30–60 min intervals for 12 h. The time at which the test meal was given was designated as zero time (0 min). In a preliminary pilot study, basal 13C:12C expiratory ratio was measured in the subjects under experimental conditions after ingestion of an unlabelled test meal to ensure that this parameter remained constant and that the test meal was palatable to all subjects.

The study was approved in advance by the Faculty's Animal Use and Care Committee and the Research Committee.

The induced lactose 13C-ureide breath test

After ingestion of a primer dose of lactose 12C-ureide (15 mg/kg bwt) 16 h previously, each animal voluntarily ingested a test meal containing lactose 13C-ureide1 (2.7 mg/kg bwt β-lactosyl 13C-ureide dihydrate), prepared in cooked egg white as described previously [14]. The test meal consisted of 150 g oats, 100 g wheat bran and 200 ml water for horses and exactly half of each component for ponies.

After test meal ingestion, expiratory samples were collected in duplicate using a previously validated and described technique [16] and stored at room temperature prior to analysis.

Measurement of 13C:12C ratio

The 13C:12C content of each sample was measured relative to the international Pee Dee Belemnite limestone fossil standard (δ13CPDB) by continuous flow isotope ratio mass spectrometry. Analytical accuracy was ensured by calibration to reference gases at the beginning of each batch run and every 5 to 10 samples thereafter. The δ13C ratio was converted to absolute units (parts per million excess 13C), and 13C enrichment determined by subtraction of average baseline breath samples.

Data were expressed subsequently as percentage dose recovery (PDR) of the administered isotope/h. VCO2 estimation needed for generation of PDR/h data used the formulae of Gallivan et al. [17] and Orr et al. [18] for horses and ponies, respectively.

Modelling and data analysis

Expiratory breath isotopic PDR/h data were plotted against time and modelled using the power exponential formula of Maes et al. [19], as given in (1). Fit of the modelled 13CO2 recovery curve was optimised using nonlinear least squares regression analysis using the iterative Solver function of Microsoft Excel2.

display math(1)

where y(t) is the % 13CO2 excretion in breath per h; t is time (h); a, b and c are rate constants and d is a delay factor.

Cumulative PDR (PDRcum) for each breath test was calculated by integration of PDR values over the period of the test [20]. The OCTT was estimated to be that point at which PDRcum approximated 3% ((2)), i.e. time after ingestion at which 3% of cumulative isotope recovery at time infinity had been achieved [14].

display math(2)

The parameters caecal t1/2 (duration from ingestion to recovery of 50% of total cumulative isotope recovery at time infinity) and caecal tlag (duration from ingestion to time of maximal dose recovery rate of the expiratory 13C label) were calculated as given in (3), (4):

display math(3)
display math(4)

where b and c are rate constants.

A general linear model ANOVA and Tukey pairwise 95% simultaneous confidence interval (CI) 2-tailed tests were used to assess whether each different compound caused a significant (P<0.05) difference to OCTT relative to the saline control. Paired t tests (CI 95%) were used to determine intra-individual significance of the treatment groups or the Mann–Whitney rank sum test where results were not normally distributed.

Clinical monitoring

Each subject was physically examined 1 h before drug administration and then at 1, 2, 3, 5, 7, 9, 11 and 12 h after treatment by the same observer blinded to treatment protocol (JK). Evaluated parameters included mucous membrane colour, capillary refill time, heart and respiratory rates, temperature, intestinal borborygmi, faecal output and consistency and behavioural signs of abdominal pain. At each time point, borborygmi were evaluated in right and left dorsal and ventral abdominal quadrants over a 2 min period and rated on a scale from 0–4 [21], with 0 assigned to absent gut sounds and 4 to marked increase in borborygmi (more than 4 loud sounds). A mean borborygmi score was calculated for each individual for separate time points and for the cumulative 12 h post treatment period. Mean borborygmi scores in the 12 h after treatment were compared between treatment and control groups by one-way ANOVA on ranks for repeated measures followed by Dunnett's test, as were mean heart and respiratory rates (significance at P<0.05). Faecal output was assessed as the number of piles of faeces present in the stall at each time and the total cumulative number produced in the 12 h after treatment. Output was assessed in this manner per individual and per treatment group. Colic signs were recorded descriptively as absent, mild (pawing), moderate (flank watching, lying down) or severe (rolling). Diarrhoea was categorised as absent (formed faeces), mild (soft faeces), moderate (completely unformed) or severe (profuse, projectile).


Clinical changes following specific piroplasmosis treatment protocols

Clinical datasets were combined over the 12 h post treatment period for every individual in each treatment group to increase sample size and improve reliability of intergroup comparison. The imidocarb/saline group had a significant (P<0.001) increase in cumulative intestinal borborygmi scores after treatment (mean ± s.d. = 2.64 ± 0.09), when compared with the saline control group (mean ± s.d. = 2.44 ± 0.09), with borborygmi score peaking at 1–3 h after treatment. In contrast, cumulative intestinal borborygmi scores were significantly decreased in both the imidocarb/atropine (1.86 ± 0.06) and imidocarb/glycopyrrolate (2.22 ± 0.10) groups when compared with the saline control group (P<0.001). These data are summarised in Table 1.

Table 1. Effect of different piroplasmosis treatment protocols vs. a saline control on specific clinical parameters in the study population (n = 6) in the 12 h post treatment period
Treatment (mg/kg bwt)Borborygmi mean (± s.d.)Heart rate (beats/min) mean (± s.d.)Respiratory rate (breaths/min) mean (± s.d.)
  1. Matching superscript letters denote a significant difference between parameter. (P>0.001). Parameters recorded by the same observer (J.K.), blinded to treatment protocol. In each case the given mean represents the mean group value (n = 6) in the 12 h period after treatment administration.

Saline i.m.

+ saline i.v. (control)

a,b2.44 ± 0.09c,d33.78 ± 1.32e12.80 ± 1.48

Imidocarb 2.4 i.m.

+ saline i.v.

a2.64 ± 0.0936.24 ± 5.08e16.37 ± 2.76

Imidocarb 2.4 i.m.

+ glycopyrrolate 0.0025 i.v.

2.22 ± 0.10c38.42 ± 5.3712.38 ± 1.81

Imidocarb 2.4 i.m.

+ atropine 0.035 i.v.

b1.86 ± 0.06d47.62 ± 6.0913.16 ± 2.65

Four of 6 horses in the imidocarb/saline group also displayed moderate to severe colic signs (lying down, flank watching and rolling) from 30 min to 5 h after treatment with severe diarrhoea occurring from 1 to 3 h after treatment in these individuals. Cumulative defecation frequency was also increased in the imidocarb/saline group, with a group total of 23 piles of faeces passed in the 12 h post treatment period, compared with a range of 12–15 in the other groups. No diarrhoea or colic signs occurred in the imidocarb/atropine, imidocarb/glycopyrrolate or saline control groups after treatment.

The effect of treatment on mean heart and respiratory rates is shown in Figure 1. Cumulative mean (± s.d.) heart rates in the imidocarb/atropine, imidocarb/glycopyrrolate, imidocarb/saline and saline control groups in the 12 h period after treatment were 47.62 ± 6.09 beats/min; 38.42 ± 5.37 beats/min; 36.02 ± 5.08 beats/min and 33.75 ± 1.32 beats/min, respectively, such that all groups maintained a significantly higher heart rate than the saline control group (P≤0.006). Maximal increase in heart rate (range 44–96 beats/min) in the imidocarb/atropine group was present in every individual 30 min after treatment, returning to pretreatment values after a further 30 min in 4 of 6 horses and only after 3 h in the remaining 2 horses. Five of 6 horses in the imidocarb/glycopyrrolate group had a significantly increased heart rate after treatment, but this was maintained for a comparably shorter duration (30–90 min) than in the imidocarb/atropine group, with a maximum rate recorded of 58 beats/min. The respiratory rate in the imidocarb/saline group was significantly higher (mean ± s.d. = 16.37/min ± 2.76, P<0.001) than in all the other groups, in which mean respiratory rate after treatment ranged from 12.80 (±1.48) to 13.16 (±2.65) breaths/min. Mean respiratory rate in the imidocarb/saline group peaked at 2 h, coincident with the development of colic signs in this group (Fig 1).

Figure 1.

Graph showing the mean (± s.d.) heart and respiratory rates in 6 healthy adult horses at specific times after administration of different imidocarb treatment protocols for equine piroplasmosis. HR = heart rate; RR = respiratory rate; SC = saline control; I/A = imidocarb/atropine combination; I/G = imidocarb/glycopyrrolate combination; I/S = imidocarb/saline.

Effect of specific piroplasmosis treatment protocols on OCTT

The effect of the different treatment protocols on mean values for OCTT, caecal tlag and caecal t1/2, as measured by the lactose 13C-ureide breath test, is summarised in Table 2.

Table 2. Effect of different piroplasmosis treatment protocols vs. a saline control on specific parameters of equine intestinal transit as measured using the lactose 13C-ureide breath test (n = 6)
Treatment (mg/kg bwt)Mean (± s.d.) OCTT (h)Mean (± s.d.) caecal tlag (h)Mean (± s.d.) caecal t1/2 (h)
  1. OCTT = orocaecal transit time; caecal tlag = caecal lag phase; caecal t1/2 = caecal half-emptying time. Matching superscript letters denote a significant difference between parameters (P<0.05).

Saline i.m.

+ saline i.v. (control)

a5.77 ± 1.67d7.70 ± 1.6514.19 ± 3.55

Imidocarb 2.4 i.m.

+ saline i.v.

b5.85 ± 1.72e7.29 ± 2.0311.36 ± 3.25

Imidocarb 2.4 i.m.

+ glycopyrrolate 0.0025 i.v.

c6.10 ± 1.01f7.66 ± 1.2112.19 ± 2.88

Imidocarb 2.4 i.m.

+ atropine 0.035 i.v.

a,b,c12.98 ± 8.22d,e,f16.75 ± 11.5827.91 ± 22.16

Analysis of variance revealed that the effect of drug treatment on OCTT, caecal tlag and caecal t1/2 was significant in each case (P<0.05), while the effect of horse or week was not significant, suggesting that these were not confounding factors. Mean values for OCTT and caecal tlag in the saline control group were estimated by the lactose 13C-ureide breath test as 5.77 (± 1.67) h and 7.70 (± 1.65) h, respectively; both of these values were greater than that reported previously using the same test meal [14]. These values were not significantly different from those in either the imidocarb/saline or imidocarb/glycopyrrolate groups. In contrast, OCTT and caecal tlag were prolonged in the imidocarb/atropine group at 12.98 (± 8.22) h and 16.75 (± 11.58) h, respectively and were significantly higher in each case than in all other treatment groups. Similarly, mean caecal t1/2 was markedly prolonged in the imidocarb/atropine group, although the small sample size and large variance meant that this difference was not significant. In the imidocarb/saline group there was a tendency toward shortening of both mean caecal tlag and caecal t1/2 when compared with the saline control group but this was not significant. In contrast with the imidocarb/atropine group, none of the transit parameters in the imidocarb/glycopyrrolate group were significantly different from those of the saline control group. The effect of treatment on the distribution of the parameter OCTT is presented as boxplots in Figure 2.

Figure 2.

Boxplot illustrating the effect of different piroplasmosis treatment protocols on parameters of intestinal transit in 6 healthy horses in a blinded crossover study. The boxplot illustrates the range, interquartile range and median values for orocaecal transit time for each treatment group.

The dose of atropine used in this study resulted in intra-individual increases in OCTT, caecal tlag and caecal t1/2 in all cases, but this was extreme in 3/6 horses with an increase in OCTT to over 10 h in these individuals (range 10.46–27.47 h). In these same 3 individuals, the imidocarb/glycopyrrolate regimen (OCTT range 5.51–7.51 h) did not result in a significant difference in transit parameters when compared with the saline control.


In this prospective randomised study, 4 of 6 healthy horses developed signs of colic together with tachypnoea, tachycardia, increased intestinal borborygmi and an increase in total faecal output when administered a standard i.m. dose of imidocarb accompanied by saline. The symptoms started approximately 30 min after drug administration and lasted up to 5 h, also coinciding with the presence of projectile diarrhoea. The occurrence of both colic [22] and diarrhoea [23] after imidocarb administration has been observed previously and was reported to be of similar duration [9]. This duration of systemic side effects observed following i.m. administration of imidocarb correlates to its peak plasma concentration, with a decline measured after 2–4 h [24]. The significant increase in borborygmi noted in the imidocarb/saline group also was in agreement with the ‘violent peristalsis’ [9] or hypermotility of the gastrointestinal tract [22] previously associated with the cholinergic properties of imidocarb.

Given the clinical evidence that imidocarb caused increased large intestinal motility, it was anticipated that its administration also would cause a significant reduction in OCTT. Cholinergic agents increase jejunal contractility in vitro by increasing both circular and longitudinal smooth muscle activity [25]. However, there was no significant difference in OCTT between the imidocarb/saline or saline control groups in this study, suggesting that there was minimal effect of imidocarb (2.4 mg/kg bwt) on gastric emptying and subsequent small intestinal transit of the labelled test meal. This selective difference in the action of imidocarb on the equine small and large intestinal tract has not been reported previously.

In contrast to combining glycopyrrolate with imidocarb, the combination of atropine at 0.035 mg/kg bwt and imidocarb resulted in variable and often extreme prolongation of OCTT and caecal t1/2, with significant reduction in orocaecal movement of the test bolus when compared with all other test groups. Intestinal borborygmi were reduced significantly in both the imidocarb/glycopyrrolate and imidocarb/atropine groups when compared with the saline control, but this effect was most marked after atropine administration, both in terms of borborygmi score and duration of effect. Individual animal response to atropine also was more variable than to glycopyrrolate. Atropine at the dose used in this study has been shown previously to cause a profound reduction in the rate of solid phase gastric emptying [26], which also is likely to result in prolongation of OCTT.

Both glycopyrrolate (0.0025 mg/kg bwt) and atropine (0.035 mg/kg bwt) when given i.v. prevented the immediate signs of colic associated with concurrent imidocarb administration in this study. This was in contrast to previously reported findings [11] that atropine at 0.02 mg/kg bwt did not prevent imidocarb-related colic. The difference in the results of the 2 studies is likely to be due to dose-dependent effects of atropine on equine large intestinal motility. Atropine given at 0.044 mg/kg bwt has been shown to cause a significant reduction in large intestinal motility and development of impactions [27]. Thus, the 0.035 mg/kg bwt dose of atropine used in this study may be the optimum dose to inhibit the action of imidocarb on large intestinal motility while minimising potential side effects. However, as found in this study and previously reported [26], the 0.035 mg/kg bwt atropine dose will have a deleterious effect on gastric emptying rate and potentially small intestinal motility and may be linked to subsequent colonic impaction [27].

In the present study, glycopyrrolate (0.0025 mg/kg bwt) successfully prevented the development of diarrhoea and colic associated with imidocarb administration, without causing significant prolongation of OCTT. A small, but significant, reduction in large intestinal borborygmi scores was detected in the imidocarb/glycopyrrolate group compared with the imidocarb/saline group but this was less marked than recorded in the imidocarb/atropine group. This same dose of glycopyrrolate has been previously reported to be sufficient to reduce the cardiovascular effects of xylazine without causing clinical detriment to large intestinal motility [21, 28]. The results of the present study suggest that glycopyrrolate blocks the inhibition of cholinesterase induced by imidocarb [29] without affecting gastric emptying rate or small intestinal motility. This is in contrast to atropine, which counters the clinical effects of imidocarb administration but at the cost of significant detriment to both small and large intestinal transit. The potential difference found in this study between the effects of the 2 parasympatholytic compounds on equine intestinal motility has not been previously reported and is worthy of further investigation.

The administration of imidocarb/atropine led to a tachycardia of up to 3 h duration, in agreement with effects of atropine described elsewhere [30], whereas subjects were tachycardic for a maximum of 1 h following imidocarb/glycopyrrolate treatment. This may indicate that atropine 0.035 mg/kg bwt is of greater systemic duration than glycopyrrolate 0.0025 mg/kg bwt in horses.

The lactose 13C-ureide breath test proved sufficiently sensitive in this study to detect significant differences in OCTT caused by drug administration and was a useful diagnostic tool that was simple to perform. Although several digestive processes might contribute to the OCTT, the test results were well described by a one-curve model and a two-curve model of fit was not required [31].

The lactose 13C-ureide breath test has been validated using in vitro studies for the measurement of OCTT in horses [13] but has not been reported previously in applied pharmacological studies. Limitations of the stable isotope breath tests in horses might include natural fluctuations in basal 13CO2 production during the test period [32]. This was minimised by removing dietary variations during the study period and by avoidance of changes in exercise level [33]. However, the habitual consumption of C4-rich kikuyu grass by subjects between test days meant that the δ 13C value of the test meal was lower than that of the standard diet. A preliminary study in the test subjects revealed a tendency for the expiratory 13C:12C ratio to fall with time under test conditions after consumption of the unlabelled test meal. This occurred in all subjects with minimal variation and may have been caused by metabolism of the test meal. A progressive decline in basal 13CO2 production could have resulted in uniform overestimation of transit times by the lactose 13C-ureide breath test and values for OCTT in this study were higher than those found in a previous study [14]. Increased enrichment of the test meal with the 13C-isotope would have further reduced any inaccuracies in parameter measurement but was prohibitively expensive for this study. Although direct comparison of drug effect on intestinal transit parameters was not considered to be affected by this technical issue, use of a larger number of subjects would have enhanced the statistical power of the study.

The study was performed during the drought season in the north of South Africa, during which a peak in impaction colics is common in equine practice. However, the randomisation of drug therapies with time was sufficient to remove this potential confounding factor from the interpretation of test results.

It is concluded from this study that imidocarb has deleterious effects on large bowel motility with production of diarrhoea and colic in a majority of animals. As imidocarb/saline did not cause a significant reduction in OCTT, these side effects of imidocarb likely result principally from induction of large intestinal hypermotility rather than from modulation of small intestinal motility. Atropine and glycopyrrolate each prevented these side effects of imidocarb in this study but atropine was associated both with prolongation of OCTT and a significant reduction in large intestinal motility that was profound in certain subjects. Glycopyrrolate in contrast prevented the extreme side effects of imidocarb at therapeutic doses without causing significant changes to OCTT as compared with control individuals. Therefore, based on the drugs evaluated in this study, an effective protocol to ameliorate the gastrointestinal side effects observed with imidocarb is to administer a concurrent dose of glycopyrrolate at 0.0025 mg/kg bwt.

Authors' declaration of interests

No competing interests have been declared.

Sources of funding

Financial support was gratefully received from the University of Pretoria, specifically from the Department of Companion Animal Clinical Studies, the Faculty Research Fund and the Equine Research Centre.


We would like to thank Thireshni Chetty, Stellest de Villiers, Claire Malik, Chris Joone and Chris Matjiane for their technical assistance as well as Kenneth Joubert for his statistical advice.


J.K., A.G. and D.S.: study design (20%, 40%, 40%); J.K.: data collection (100%); T.P.: stable isotope analysis (100%); J.K. and D.S.: data analysis and interpretation (50%, 50%); J.K. and D.S.: preparation of the manuscript (60%, 40%).

Manufacturers' addresses

  1. 1

    Bell College of Technology, Hamilton, Scotland.

  2. 2

    Microsoft Corporation, Redmond, USA.