Effects of Cisapride, Buprenorphine, and Their Combination on Gastrointestinal Transit in New Zealand White Rabbits

Abstract

Due to their effective analgesic properties, opioids are worthy of consideration for pain management in rabbits. However, this class of drugs causes undesirable effects including reduced gastrointestinal (GI) motility, reduced fecal output, and delays GI transit times and thus increases the risk of GI stasis. The risk of stasis discourages the use of opioids in rabbits, which could affect animal welfare. Gastroprokinetic agents such as cisapride are effective in promoting gastric emptying in many species, but whether this effect occurs in rabbits is unknown. This study assessed the efficacy of cisapride when administered as a single agent and in combination with buprenorphine in rabbits; efficacy was assessed by measuring GI transit times, fecal output, body weight, and food and water intake. Female New Zealand White rabbits (n = 10) were studied in a crossover, randomized design and received either vehicle and buprenorphine, cisapride and saline, cisapride and buprenorphine, or vehicle and saline (control) every 8 h for 2 d. Rabbits were anesthetized and administered radio-opaque, barium-filled spheres via orogastric tube. Feces was assessed via radiography for detection of the barium-spheres to determine GI transit time. GI transit time was significantly longer in buprenorphine groups than in control groups, regardless of the use of cisapride. Fecal output and food and water intake were lower for buprenorphine groups than control groups. Cisapride did not significantly alter GI transit, fecal output, or food and water intake. In addition, treatment group did not significantly affect body weight. In conclusion, buprenorphine treatment (0.03 mg/kg TID) prolonged GI transit time and reduced fecal output and food and water consumption in rabbits. Coadministration of buprenorphine and cisapride (0.5 mg/kg) did not ameliorate these effects, and the administration of cisapride at this dose did not appear to affect GI motility in female rabbits.

In the United States, more than 2 million rabbits are client-owned, many of which may undergo moderate to markedly painful procedures such as surgery that require alleviation of pain and distress.¹ Furthermore, according to the USDA's 2018 annual animal usage report, approximately 133,000 rabbits were used for research, many of which were used in potentially painful procedures requiring analgesia.³ The opioid analgesic most commonly used in rabbits is a partial μ-opioid receptor agonist, buprenorphine.⁸ Buprenorphine prolonged gastrointestinal (GI) transit times and reduced fecal output and food and water intake in male rabbits.^7,18 These established side effects of opioids leave rabbits at risk of developing a potentially life-threatening condition known as GI stasis. GI stasis is characterized by decreased muscular contractions of the stomach and intestines, leading to blockage and the proliferation of harmful bacteria in the rabbit's intestinal tract. Initial clinical signs include reduced appetite, small or no fecal pellets, and abdominal discomfort.⁴

Although opioid-induced GI stasis is reported in other species, rabbits are more prone to GI stasis due to their unique digestive anatomy, and they can have severe clinical signs.^13,24 The rabbit GI system combines a simple stomach with hindgut cecal fermentation. The cecal digestive process of rabbits differs from that of other hindgut-fermenting animals in that, in rabbits, indigestible fiber is eliminated as rapidly as possible through constant digestion and peristalsis. Other fine particles and solutes are retained in the cecum as substrates for microbial fermentation. This rapid separation of fiber and energy concentrates is useful in view of rabbits’ small size and high metabolic rate.²⁴

A rabbit can become prone to developing GI stasis when GI motility, appetite, and thirst are decreased due to stress, inappropriate diet, GI blockage, or other causes of hypomotility, such as the administration of opioids. Stasis can lead to proliferation of harmful GI bacteria and the development of potentially fatal intestinal blockages.^24,30 Current standard-of-care treatment for GI stasis in rabbits includes restoring appetite through forced feedings, correcting dehydration, and administering agents that promote GI motility.^16,24,30 However, the recommended dosages of gastroprokinetic agents such as cisapride in rabbits are extrapolated from other species. Little scientific evidence exists regarding whether these prokinetic drugs are effective in rabbits at published doses.

Cisapride is a serotonin 5-hydroxytryptamine 4 receptor agonist with some 5-hydroxytryptamine 3 antagonist activity, which results in increased GI motility. Cisapride has broad prokinetic activity, increasing the motility of the colon, esophagus, stomach, and small intestine in humans, dogs, cats, and mice.^15,23,27,29 Cisapride and the antacid ranitidine have synergistic effects on increasing GI contractility in postmortem isolated rabbit intestine.^13,23 In addition, cisapride improved contractility in ex-vivo newborn and adult rabbit ileums.¹⁴ In mice, cisapride reversed morphine-induced delayed GI transit time.²⁷ Another option for treating GI opioid-induced side effects is a peripheral opioid receptor antagonist, such as methylnaltrexone. However, methylnaltrexone was found to be ineffective at reducing negative GI side effects associated with buprenorphine in rabbits.¹⁸ Given the broad prokinetic effects of cisapride and the different mechanism of action of cisapride as compared with peripheral opioid antagonists, veterinary clinicians may reasonably administer cisapride to ameliorate GI stasis in rabbits. Multiple veterinary formularies include the use of cisapride as a prokinetic agent in rabbits.^6,10,23

The effects of buprenorphine on GI motility were assessed previously in male rabbits.¹⁸ One aim of the current study was to characterize the GI motility effects of buprenorphine in healthy, female New Zealand White (NZW) rabbits. Furthermore, given the lack of information regarding gastroprokinetic agents in rabbits and the potential positive analgesic use of opioids during invasive procedures, the second aim of our study was to evaluate the efficacy of cisapride in healthy, female NZW rabbits and its potential to ameliorate the negative GI side effects of buprenorphine. Demonstrating cisapride's effectiveness in ameliorating buprenorphine-induced GI side effects could encourage veterinarians and researchers to use this drug to treat opioid-induced stasis in rabbits, thus directly benefitting animal welfare. To our knowledge, this study is the first to evaluate cisapride's potential to treat opioid-induced GI stasis in rabbits.

Previous methods used to measure GI transit times in rabbits require training to interpret data and the use of expensive techniques such as Doppler ultrasonography, computed tomography, MRI, or wireless motility capsules.^5,9,22 In our current study, GI transit times were measured by using an established method involving biologically inert, barium-filled, radiopaque spheres administered via orogastric tube and quantified through serial radiography of the feces of each animal. Little specialized training is required to administer the spheres and to interpret results. The efficacy of this technique for measuring GI transit time was previously validated in horses.¹⁷ This method was used previously to document the effects of buprenorphine and methylnaltrexone on GI transit time in rabbits.^17,18,26

Based on published data, we hypothesized that buprenorphine at 0.03 mg/kg SC every 8 h would reduce GI motility in female rabbits as evidenced by increased GI transit time and decreased fecal output when compared with a saline control group.¹⁸ Furthermore, we hypothesized that buprenorphine would decrease food and water intake when compared with a saline control group.¹⁸ For our second aim of the study, we hypothesized that cisapride alone at 0.5 mg/kg every 8 h would increase GI motility as evidenced by more rapid GI transit time when compared with a saline control group. The buprenorphine and cisapride dosing regimens selected for this study are commonly cited in many veterinary drug formularies.^6,10,23,24 Finally, we hypothesized that coadministration of cisapride and buprenorphine would alleviate the opioid-associated side effects of GI stasis by increasing GI motility, such that we would find no significant differences in GI transit time, fecal output, and food and water intake when comparing treated and control rabbits.

Materials and Methods

Animals, husbandry, and welfare.

Healthy, intact female NZW rabbits (Oryctolagus cuniculus; n = 10; age, 8 mo; weight, 3.5 to 4.6 kg) were received from Charles River Laboratories (Oxford, MI). All rabbits were acclimated for 3 weeks prior to study initiation. The rabbits were experimentally naive prior to this study. Rabbits were negative for Pasteurella spp., Helicobacter spp., Bordetella bronchiseptica, Salmonella spp., Clostridium piliforme, Lawsonia spp., Treponema spp., cilia-associated respiratory bacillus, Eimeria spp., Passalurus ambiguus, Cheyletiella parasitovorax, Psoroptes cuniculi, Leporacarus gibbus, reovirus, rabbit hemorrhagic disease virus, lymphocytic choriomeningitis virus, and rotavirus. Rabbits were housed individually in stainless-steel cages (Allentown, NJ) with slotted floors to allow collection of feces from a tray. Stainless-steel mailboxes (Wheaton Fabrication, Ithaca, NY) were provided as shelters, and stainless-steel or plastic balls were provided as enrichment manipulanda. When not on study and during washout periods, rabbits were rotated weekly into one of 2 floor pens that included extra enrichment, such as a wooden shelter box and plastic toys. In addition, rabbits were provided with timothy hay (Western timothy hay; crude fiber maximum, 32%; Oxbow, Murdock, NE) during the acclimation and washout periods. The hay was removed from all rabbit cages 2 d prior to each treatment week. Rabbits were fed a pelleted diet (Country Feeds 16% Rabbit Feed, Nutrena, Minneapolis, MN), and municipal tap water was provided without restriction in glass water bottles with sipper tubes. The room was maintained on a 12:12-h light:dark cycle, with a temperature of 66 to 70 °F (18.9 to 21.1 °C) and relative humidity of 30% to 70%. All work was conducted in an AAALAC-accredited facility in compliance with the Animal Welfare Act and Guide for the Care and Use of Laboratory Animals and was approved by the IACUC of Cornell University (Ithaca, NY).^2,11

Experimental design.

Prior to study initiation, rabbits were assigned to 4 treatments in a crossover, randomized design, with at least a 1-wk washout period between treatment periods, with the washout period beginning after day 5 of the study. According to a previous study, the half-life of cisapride in rabbits was 4 to 10 h, so we allowed at least 1 wk between treatment periods to allow for adequate drug excretion.²⁰ A 1-wk washout period was used previously in a similar model when evaluating buprenorphine and methylnaltrexone in male rabbits.¹⁸ Our current study occurred over 12 wks. Each rabbit was randomly assigned a treatment prior to each study period, and no rabbit received the same treatment more than once. The cross-over design promoted consideration of the ‘reduction’ aspect of the 3 Rs, allowing appropriate animal numbers for each treatment group while reducing the number of rabbits needed for the study. Personnel collecting the data were blind to the treatment groups. Data were collected for 5 d during each treatment period. The animals were divided into the following 4 treatment groups: an oral, fruit-flavored, oil-suspension vehicle (Wedgewood Pharmacy, Swedesboro, NJ) and buprenorphine (Veh+Bup; 0.03 mg/kg SC; buprenorphine hydrochloride injection, MWI Animal Health, Boise, ID); a compounded, fruit-flavored preparation of cisapride (0.5 mg/kg PO; Wedgewood Pharmacy, Swedesboro, NJ) and subcutaneous saline (Cis+Sal); cisapride and buprenorphine combined at the same doses as in previous groups (Cis+Bup); and oral vehicle and saline (control, Veh+Sal). All treatments were administered every 8 h for 48 h. The regimen of administering buprenorphine at this dose every 8 h is common in rabbits after noxious procedures.^6,7,10,23 The cisapride dose evaluated in the current study was derived from formularies and clinical textbook recommendations.^6,10,16,23 During the dark phase, a red light was used to facilitate administration of treatments and collection of fecal pellets.

On day 1 of each study period, rabbits were anesthetized with isoflurane (3% to 4%; Patterson Vet Supply, Greeley, CO) in an induction chamber until loss of the righting reflex was observed. Rabbits were then removed from the chamber and maintained on isoflurane (2% to 3%) via face mask until an appropriate plane of anesthesia was reached for orogastric tube placement. All animals were under isoflurane anesthesia for a total of 15-20 minutes. Ophthalmic lubricant was administered in both eyes to prevent corneal irritation or injury. The orogastric tube (diameter, 6.6 mm; Stallion Catheter, EquiVet, Lebanon, OH) was measured from the tip of the nose to the last rib to ensure that the tube could reach the stomach. Sterile lubricant (Patterson Veterinary) was placed on the tip of the tube. The orogastric tube was then placed and confirmed to be in the correct position through palpation of the tube at the side of the neck, confirming esophageal placement. Next, 20 barium-filled spheres (diameter, 1/8 in.; low-density polyurethane spheres, Precision Plastic Ball Company, Franklin Park, IL) and either cisapride or the vehicle control (equal volume to cisapride dose) were administered through the orogastric tube, which was then flushed with 25 mL of municipal tap water. Because spheres can become lodged in the orogastric tube during administration, successful placement of 20 spheres was not always possible, and a minimum of 15 spheres was administered. The remaining spheres in the orogastric tube were counted and recorded to document the total number of spheres each rabbit received. Buprenorphine or saline (equal volume to buprenorphine dose) was then administered subcutaneously between the shoulder blades (Figure 1 A). Administration of the treatment was hour 0. Before the rabbit recovered from anesthesia, day 1 weights were obtained. To facilitate accurate fecal and sphere collection, an Elizabethan collar (E-collar) was placed to prevent ingestion of excreted spheres and feces. The rabbits were monitored in their cages until fully recovered from anesthesia. A total of 6 treatments were administered every 8 h for 2 d. On days 1 and 2, a complete physical exam of all body systems was performed on each rabbit every 8 h. Signs of abdominal distension, dehydration, or abdominal discomfort were evaluated via abdominal palpation and auscultation by 2 veterinarians who were blind to treatment, including an ACLAM board-certified veterinarian. Abdominal discomfort was assessed by these same veterinarians by observing orbital tightening, ear positioning, and reaction to abdominal palpation. On days 3 through 5, a complete physical exam was performed on the rabbits daily by the same veterinarians. E-collars were removed on day 5. Noon on day 5 was considered to be hour 100.

Figure 1.

(A) Rabbits were anesthetized with isoflurane (3% to 4% in induction chamber, maintained via face mask at 2% to 3%). An orogastric tube was placed, and rabbits were administered radiopaque barium spheres in tap water containing either cisapride (blue) at 0.5 mg/kg or an oil-suspension vehicle control of equal volume (yellow). Rabbits were allowed to recover from anesthesia and then were returned to home cage. (B) Radiograph of rabbit feces with barium spheres (red arrow) used to determine gastrointestinal transit times and percentage of spheres recovered. The radiograph features fecal collections from 3 different rabbits at the same time point.

Fecal collection and sphere detection

Rabbit feces were collected and weighed every 4 to 6 h, for a total of 100 h. At every collection time point, feces were radiographed for the detection of radio-opaque spheres (Figure 1 B). The GI transit time was defined as time to the appearance of the first sphere. Feces found outside of the cage were collected and radiographed for lost spheres, but these spheres were not included in any data set because the rabbit from which the sphere came could not be identified. By 100 h, if the recovery of beads was less than 75% of those administered on day 1, abdominal radiographs were taken for the detection of spheres remaining in the GI tract. The number of spheres recovered at each time point was divided by total number of spheres administered to obtain “percent spheres recovered”.

Measurement of body weight, food, and water.

Baseline body weight was obtained on day 1, and rabbits were weighed daily at noon for 4 more days. Food and water consumed from the previous day was measured daily at noon throughout the study period. The last collection was performed at noon on day 5, so total measurements of food and water for day 5 were not obtained. Fresh food and water were provided daily. If rabbits did not drink water from a bottle or did not produce feces within 24 h, a bowl of water was provided to improve access to water. Water consumed from bottles or bowls was measured by weighing the container and water together and then subtracting the weight of the container (in grams). The weight was then converted to milliliters.

Statistical analysis.

All statistical analyses were performed by using R (R Core Team, 2020).²⁵ A sample size of 10 rabbits per group was calculated to detect a delay in transit of 6.4 h with 80% power and a significance level of 0.05.⁵ GI transit time was analyzed by using a linear mixed-effects model. The model contained fixed effects of the oral treatment, the injected treatment, and their interaction and a random effect of rabbit identification to control for the repeated measurements across the 4 treatment combinations. The significance of the fixed effects was tested by using F tests with Satterthwaite approximation. Model assumptions were assessed through visual assessment of the residuals. Fecal weight, sphere recovery, food intake, water intake, and body weight also were analyzed by using linear mixed-effects models. The models contained fixed effects of the oral treatment, injected treatment, day, all 2-way interactions, and 3-way interaction. The models included random effects of rabbit ID and rabbit ID with treatment combinations to control for the repeated measurements taken on rabbits across the 4 treatment combinations and across days within each treatment combination. The significance of the fixed effects was tested by using F tests with Satterthwaite approximation, and pairwise comparisons between treatment combinations within each day were made by using the Tukey HSD method to control the type I error rate. Summary data are expressed as mean ± SD.

Results

Clinical observations and body weight changes.

No signs of abdominal pain or distension were observed during treatment. Four rabbits did not produce feces within 24 h of administration of spheres in the buprenorphine treatment groups, regardless of which oral treatment was administered. In these situations, the introduction of a water bowl encouraged drinking, likely due to ease of access to water, and the animal defecated within the next 24 h. No other clinical interventions were necessary. For all treatments, average body weight decreased from day 1 through 5, but body weight did not differ between any of the 4 groups at any time point (data not shown).

Effects of oral cisapride and subcutaneous buprenorphine on gastrointestinal motility.

Barium spheres were recovered from the feces of all 10 rabbits in the Veh+Sal, Veh+Bup, and Cis+Sal groups and from 8 of the 10 rabbits in the Cis+Bup treatment group. The two rabbits were excluded from the Cis+Bup treatment group due to incorrect bead placement in one rabbit and due to no spheres detected within 100 h in the other rabbit; this latter rabbit produced approximately 75% of the introduced spheres over the next 3 d after 100 h. The time to first appearance of a sphere (mean ± 1 SD) was 12.6 ± 7.2 h in the Veh+Sal (control) group, 40.8 ± 19.9 h in the Veh+Bup group, 10.4 ± 4.3 h in the Cis+Sal group, and 37.8 ± 22.3 h in the Cis+Bup group. GI transit time for the buprenorphine groups (Veh+Bup and Cis+Bup) was significantly (P < 0.001) longer than for the Veh+Sal and Cis+Sal groups. Transit time was not significantly different between Cis+Bup and Veh+Bup rabbits (P = 0.991) or between the Cis+Sal group and the control group (P = 0.265), indicating that cisapride had no significant effect on GI transit time (Figure 2 A). Pairwise comparisons between the treatment groups on days 1 through 3 showed a statistically significant decrease in the percentage of spheres recovered in the buprenorphine groups as compared with the injected saline groups (P < 0.001), regardless of oral treatment group. By day 4, a statistically significant difference between the Cis+Bup and Veh+Sal groups was no longer present (Figure 2 B). To better characterize the variability in total sphere recovery, feces were collected daily for an additional 3 d from 5 rabbits that each had less than 75% sphere recovery by hour 100. Four of the 5 rabbits from either the Veh+Bup or Cis+Bup treatment groups excreted spheres in feces for as long as 8 d after bead administration. Abdominal radiographs of these rabbits showed no retention of beads in the GI tract.

Figure 2.

(A) Transit time in hours (mean ± SE), as defined by time to observation of first barium sphere in feces, in 10 female, 8-mo old NZW rabbits that received an oral vehicle control and saline (Veh+Sal), vehicle and buprenorphine (Veh+Bup), cisapride and saline (Cis+Sal) or cisapride and buprenorphine combined (Cis+Bup). Cisapride did not appear to effect gastrointestinal transit time, as the interaction between the Cis+Bup and Veh+Bup was not statistically significant (P = 0.991) nor was the main effect of Cis+Sal (P = 0.265). (B) Cumulative spheres recovered (%) in the feces of 10 NZW rabbits receiving the treatments described for panel A. On days 1 through 3, there was a statistically significant (P < 0.001) decrease in the percentage of spheres recovered between the buprenorphine treatment groups and saline groups, regardless of oral treatment. By day 4, there was no longer a statistically significant difference between the Cis+Bup and Veh+Sal groups.

Effects of oral cisapride and subcutaneous buprenorphine on fecal output

Fecal output on days 1 through 3 was significantly (P < 0.001) lower in the buprenorphine groups when compared with injected saline groups, regardless of which oral treatment was administered (Figure 3). By day 4, no significant difference in fecal output was detected between the 4 groups. Cisapride treatment had no significant effect on fecal output (P = 0.3). However, on days 1 and 2, feces from the buprenorphine groups were smaller, harder, and irregularly shaped as compared with the feces from the saline control group (Figure 4).

Figure 3.

Fecal weights (mean ± SE) in 10 NZW rabbits receiving an oral vehicle control and saline (Veh+Sal), oral vehicle and buprenorphine (Veh+Bup), oral cisapride and saline (Cis+Sal), or oral cisapride and buprenorphine combined (Cis+Bup). Pairwise comparisons between the treatment groups each day show that on days 1 through 3, regardless of oral treatment, the 2 treatment groups that received buprenorphine produced significantly (P < 0.01) less feces than the 2 treatment groups that received saline. There was no statistically significant difference between the 4 treatment groups at day 4.

Figure 4.

Feces collected from a Bup+Cis treatment group (left) and a control group (right) on day 2 of the second treatment week. Feces from the buprenorphine group tended to be smaller, harder, and more irregularly shaped when compared with a control treatment group.

Effects of oral cisapride and subcutaneous buprenorphine on food and water intake.

Average food and water intake increased from day 1 through 4, regardless of treatment group (Figure 5 A and B). Daily pairwise comparisons between the treatment groups showed that on days 1 through 3, the buprenorphine groups consumed significantly less food than did the saline-injected groups (P < 0.001). On all 4 d, water consumption was lower in the buprenorphine groups than in the saline groups. On day 2, a statistically significant decrease in water consumption was detected between all buprenorphine groups and saline groups, regardless of which oral treatment was administered. Administration of cisapride had no significant effect on food and water intake whether given with saline or buprenorphine (P = 0.8).

Figure 5.

(A) Food (g) and (B) water consumed (mL, mean ± SE) in 10 NZW rabbits receiving an oral vehicle control and saline (Veh+Sal), oral vehicle and buprenorphine (Veh+Bup), oral cisapride and saline (Cis+Sal), or oral cisapride and buprenorphine combined (Cis+Bup). Average food and water intake increased from day 1 through day 4, regardless of the treatment group. Pairwise comparisons between the treatment groups each day show that on days 1 through 3, the buprenorphine groups consumed significantly (P < 0.001) less food than the saline groups. Overall, buprenorphine treatment groups consumed less water than saline groups. There was a statistically significant (P < 0.05) decrease in water consumption on day 2 in all buprenorphine groups, regardless of oral treatment, compared with saline groups.

Discussion

This study examined the effects of 4 treatments—1) an oral vehicle control and SC saline control, 2) an oral vehicle control and SC buprenorphine, 3) oral cisapride and SC saline, and 4) oral cisapride and SC buprenorphine—on GI transit time, fecal output, food and water consumption, and body weight in 10 healthy, female, NZW rabbits. Buprenorphine is one of the most common analgesic drugs administered after potentially painful procedures in rabbits. Our results showed that buprenorphine administered 3 times daily at 0.03 mg/kg for 2 d prolonged GI transit time and reduced fecal output and food and water consumption in female rabbits. These side effects are risk factors for the development of GI stasis. To combat these negative effects of buprenorphine administration, we evaluated cisapride as a potential treatment for opioid-induced GI stasis. When administered with a SC saline injection, oral cisapride at 0.5 mg/kg 3 times daily for 2 d had no significant effect on GI transit time, fecal output, or food and water consumption. Oral cisapride also had no significant effect on these parameters when administered with buprenorphine. None of the rabbits in the 4 treatment groups showed significant changes in body weight.

The GI transit times were lowest in the control group and significantly higher after buprenorphine administration, regardless of which oral treatment was administered (Figure 2 A). On days 1 through 3, the percentage of sphere recovery in feces was lowest in the buprenorphine groups. These 2 results suggest that we observed buprenorphine-induced GI hypomotility in buprenorphine treated rabbits. The administration of cisapride had no significant effect on GI transit time or sphere recovery, suggesting that cisapride at 0.5 mg/kg PO 3 times daily would not ameliorate buprenorphine-induced hypomotility. Because cisapride is a prokinetic agent used to treat opioid-induced stasis in humans and is effective in dogs, cats, and mice, we anticipated a similar effect in rabbits.^15,27,29 Serotonin (5- hydroxytryptamine) is present in rabbit gut mucosa, suggesting cisapride should be effective in rabbits.¹⁹ Due to the lack of peer-reviewed information on the use of cisapride in rabbits, higher doses could be efficacious and the subject of future studies.

Cisapride was administered to healthy, adult animals receiving opioids. Therefore, effects of cisapride may be more apparent in rabbits experiencing ileus related to causes other than opioid administration, such as stress-induced ileus. Furthermore, buprenorphine is typically given to animals after potentially noxious procedures and anesthetic events, both of which potentially reduce GI motility.²⁴ To better emulate postsurgical analgesia conditions, we anesthetized our rabbits with isoflurane to administer the barium spheres via orogastric tube. No published studies show that isoflurane itself influences GI transit time in rabbits. In rats, brief exposure to isoflurane increased GI transit times and repeated exposure to isoflurane increased aversion behaviors.^21,28 In horses, isoflurane increased GI transit time, but return to a normal GI transit time was faster with isoflurane than halothane.¹⁴ Therefore, repeated exposures to isoflurane could have prolonged GI transit time or reduced appetite and thirst, thereby affecting our model. However, given that all control and treatment rabbits underwent isoflurane anesthesia for similar time intervals, it likely did not notably affect GI transit times in the 4 treatment groups.

When compared with saline-injected groups, buprenorphine groups excreted significantly fewer spheres in the feces on days 1 through 3, consistent with the time course of reduced GI motility. The total percentage of sphere recovery by 100 h ranged from 60% (Veh+Bup) to 98.5% (Cis+Sal) among the 4 treatment groups (Figure 2 B). Each rabbit from which less than 75% of the administered spheres were recovered by the end of the 100 h was radiographed the following week to look for spheres retained in the GI tract. All rabbits with less than 75% of spheres recovered were from a buprenorphine treatment group (Veh+Bup or Cis+Bup). No spheres were detected in the GI tract via abdominal radiographs, indicating the loss of spheres elsewhere. To better understand sphere loss, feces were collected daily for an additional 3 d from 5 rabbits with less than 75% sphere recovery. Four of the 5 rabbits from either the Veh+Bup or Cis+Bup treatment groups excreted spheres in feces for as long as 8 d after bead administration. This finding suggests buprenorphine's effect on GI motility may persist for days after termination of buprenorphine administration, and careful consideration of longer washout periods may be necessary in future studies. During sphere administration, spheres occasionally became lodged in the orogastric tube, leading to variability in the total number of spheres administered (15 to 20 spheres administered to each rabbit). In addition, some feces containing intact spheres were found outside the cages, leading to the variability of total spheres recovered because the rabbit that produced these feces could not be determined. These feces were radiographed for detection of spheres, and a sphere was noted in 2 of the collected fecal pellets, accounting for some loss in the total spheres recovered. In a previous study measuring GI transit time in rabbits, approximately 25% to 60% of spheres were recovered from each rabbit, with the fewest spheres recovered in a buprenorphine-treated group.¹⁸ The authors of that study hypothesized that cecotroph ingestion allowed rabbits to ingest and masticate spheres, accounting for the absence of intact spheres.¹⁸ To prevent sphere ingestion, we placed E-collars on all rabbits for the experimental period. E-collars may have influenced digestion by preventing ingestion of cecotrophs or may alone have been a source of stress for the rabbits. Stress has been implicated as a risk factor for GI stasis.²⁴ Another source of stress in our study could have been individual housing. However, given that all animals wore collars and were singly housed, we presume the lack of cecotrophy and stress induced by the E-collars or single-housing effected all groups similarly and therefore did not contribute to the differences we observed. After removal of E-collars on day 5, rabbits may have ingested spheres via cecotroph ingestion, or spheres could have been lost in feces found outside of the cage pan. Using E-collars resulted in a better sphere recovery rate than in the previous study,¹⁸ ranging in our study from 60% (Veh+Bup) to 98.5% (Cis+Sal) recovery (Figure 2 B). Measuring GI transit time after bead administration may be subject to some variability in sphere recovery but overall proved to be a successful way to measure GI transit time in NZW rabbits. Variability could be reduced further by keeping E-collars on and extending fecal collections to at least 8 d after buprenorphine administration.

At the dose used in our study, buprenorphine reduced fecal excretion and food and water intake as compared with an injected saline control (Figure 5 A and B). This finding shows that not only did buprenorphine prolong GI transit time, but it also reduced the amount of feces produced, likely due to the reduction in food intake. Lack of hay supplementation may have affected GI motility, appetite, or thirst. Fiber is an important component for supporting GI motility in a rabbit diet and is typically provided by supplementation with hay.^1,7 Although rabbits did not receive hay during the study periods, the pelleted diet provided adequate fiber, and all control and treated rabbits received the same diet. Therefore, lack of hay during study periods likely did not contribute significantly to differences in fecal excretion or food and water intake among the 4 treatment groups. Decreased water consumption has been linked to decreased food intake in rabbits and likely contributed to our findings.¹ One potential variable in our water consumption findings was water spillage from water bottles or bowls. Spillage could have affected the results between the treatment groups. However, minimal water loss from the water bottles was noted. The water bowls were checked at every fecal collection, and no spilled water was noted in the collection pan at any time point. Because of published studies and the expected difficulty of accurately measuring hay consumption, hay was not provided during study periods.¹⁸

A previously published study using a similar research model found that buprenorphine administered to male rabbits twice daily at 0.05 mg/kg SC resulted in decreases in fecal excretion and food and water consumption, consistent with our findings.¹⁸ We also found that feces excreted by rabbits in the buprenorphine group were smaller, drier, and more irregularly shaped when compared with feces from a control animal (Figure 4). The abnormal feces were likely due to a combination of reduced water consumption and GI hypomotility in the buprenorphine-treated rabbits. Typically, rabbit day feces are round, firm, and friable due to excretion of fiber. Episodes of GI hypomotility or stasis are associated with significant changes in GI function. Fecal pellets are significantly reduced or absent, and those that pass are much smaller and irregularly shaped due to loss of liquid from the gut contents in the stomach.²⁴ We saw similar changes in fecal characteristics in the buprenorphine-treated rabbits. By day 4, fecal excretion and food and water consumption in the buprenorphine groups had normalized in comparison to the saline control groups, even though administration of buprenorphine was discontinued on day 2. This observation further suggests that the side effects of buprenorphine may outlast perceived analgesic effects.^6,23

We anticipated that cisapride administration would alleviate buprenorphine-induced GI side effects and regulate GI motility, as evidenced by normalized GI transit time, fecal output, and food and water intake when compared with a control group. However, cisapride did not decrease GI transit time, increase fecal output, or increase food and water consumption, suggesting that cisapride at the dose used in this study was not suitable for ameliorating the negative GI side effects of buprenorphine administration in female NZW rabbits.

Although no statistically significant differences were detected in body weight between the treatment groups, average body weight decreased from days 1 through 5 for all treatments. This decrease may have resulted from increased stress due to daily handling of the rabbits. However, this decrease was unanticipated, given that the total handling decreased throughout the week and that animals appeared to acclimate to handling as the study progressed.

In a review of publications from 2005 through 2007, approximately half of rabbits undergoing experimental surgical procedures did not receive systemic analgesic therapy.¹⁷ An explanation for the lack of analgesia is that authors may have excluded the information from their reports either by mistake or lack of attention to such details in manuscript creation and review. Alternatively, the lack of systemic analgesic therapy was possibly due to the perceived risk of opioid-associated GI stasis in rabbits and therefore was not administered postoperatively. This perceived risk discourages researchers and clinicians from using opioids, leading to concerns regarding animal welfare and the possibility of stress-induced biased results. In our study, GI side effects were statistically significant between treatment groups, but medical intervention was not necessary at this dose and frequency of buprenorphine administration. Some rabbits in the buprenorphine treatment groups did not produce feces within 24 h after sphere administration. The addition of a water bowl to the home cage resulted in defecation within 24 h after placement of the bowl. Dehydration contributes to ileus in rabbits, and by increasing ease of access to water through supplying a water bowl, hydration improved, eliminating the need for clinical intervention.¹ In contrast to our study, another group found that twice-daily (0.03 mg/kg) administration of buprenorphine to Dutch belted rabbits who underwent ovariohysterectomy did not cause any significant differences in feed consumption or fecal output, when compared with using a NSAID. However, additional studies such as the one presented here are necessary for evaluating buprenorphine administration at the recommended 8- to 10-h frequency.^6,7,23 In addition, a single high dose of buprenorphine did not appear to have adverse effects on GI motility in healthy rabbits.⁹ Our findings suggest that although buprenorphine delays GI transit and decreases fecal output and food and water consumption, it is possible to use buprenorphine safely as an analgesic, given that no clinical intervention was necessary for the rabbits in this study. Nevertheless, fecal output and food and water intake should be monitored carefully after buprenorphine administration, because intervention may be necessary in some cases, such as in other breeds of rabbits or in postsurgical patients.

Although cisapride (0.5 mg/kg PO) was not effective at ameliorating the GI side effects of buprenorphine, validating this experimental method allows for the evaluation of other common prokinetic drugs and opioids used in rabbits. In addition, the model used in this study could be used to explore different doses of cisapride in rabbits or consider evaluating cisapride prior to opioid-administration. Because only 8-mo old female rabbits were evaluated in this model, future studies could expand on the use of prokinetics and opioids in young or aged female rabbits, or in male rabbits. Other applicable studies include assessing the efficacy of other prokinetic agents such as metoclopramide on rabbit GI transit times. Future experiments can evaluate full-μ opioid receptor agonists, such as hydromorphone or methadone, to characterize their effects on rabbit GI motility. Full μ-opioid receptor agonists can induce more significant GI side effects than a partial μ-opioid receptor agonist such as buprenorphine.¹² We hope to characterize the effects of a full μ-opioid receptor agonist on GI transit times and establish whether negative effects can be ameliorated by another prokinetic agent, appetite stimulant, or peripheral opioid-receptor antagonist in rabbits. Investigating full μ-opioid receptor agonists may encourage the clinical consideration of these drugs instead of a partial μ-opioid receptor agonist such as buprenorphine in rabbits. A full μ-opioid receptor agonist may provide better analgesia than a partial μ-opioid receptor agonist during invasive procedures.

In conclusion, our results suggest that thrice-daily treatment of buprenorphine (0.03 mg/kg SC) for 2 d slowed GI transit time, decreased fecal output, and reduced food and water consumption in healthy, female NZW rabbits. In addition, the administration of cisapride (0.5 mg/kg PO) did not significantly affect fecal output, food and water consumption, or GI transit times when compared with a control, nor did cisapride ameliorate the prolonged GI transit time when coadministered with buprenorphine. However, with diligent monitoring of fecal output and of food and water intake, buprenorphine may be used safely as an analgesic in rabbits.