Gastrointestinal stasis is one of the most common clinical disorders of pet rabbits and can be primary or secondary in nature.1–4 Underlying or contributing pathological conditions can include gastric impaction, gastric gas accumulation, intestinal impaction, intestinal gas accumulation, intestinal obstruction, primary gastroenteritis, intestinal adhesions, neoplasia such as uterine adenocarcinoma, pancreatitis, and liver abnormalities such as hepatic lipidosis, torsion, and cholangiohepatitis.1,2,4 Associated clinical signs are generally attributable to secondary alterations in fluid balance, gastrointestinal motility, or both1–4 and are often nonspecific.
Rabbits with gastrointestinal stasis typically need supportive care and a diagnostic workup to determine underlying causes. Gastrointestinal obstruction needs to be ruled out whenever possible because of the need for surgical management in most cases and because the prognosis for rabbits undergoing gastrointestinal surgery is guarded owing to the high risk of peritonitis, postoperative gastrointestinal stasis, endotoxemia, or acute renal failure, even with prompt foreign body removal.4,5
Abdominal radiography is useful in the evaluation of rabbits with gastrointestinal stasis, and serial radiography can be used to monitor response to treatment.4–6 Drawbacks of radiography include distress to the rabbit, risk of injury or radiation exposure to the rabbit or handler, possible need for sedation, lack of quantitative results, and inability to distinguish between mechanical and paralytic ileus.5 A more sensitive and specific imaging modality is needed to better guide treatment recommendations for affected rabbits and avoid unnecessary surgery whenever possible.
Ultrasonography, which lacks ionizing radiation, has been diagnostically useful for human and veterinary patients with gastrointestinal disorders. For example, in a prospective study7 of people with ileus and gastrointestinal obstruction, abdominal ultrasonography revealed abnormalities in 25% of people with normal radiographic findings, and the cause of the obstruction was identified in 13% of those patients. Furthermore, in a study8 of dogs and cats with confirmed gastrointestinal foreign bodies, only 9 of 16 foreign bodies were identified via radiography, whereas all 16 were identified via ultrasonography.
Although ultrasonography can be helpful to identify gastrointestinal obstructions and foreign bodies, its use for rabbits has previously been limited because of perceived difficulty obtaining images owing to gastrointestinal gas accumulation.4 To the authors’ knowledge, the sole report9 of ultrasonography of the gastrointestinal tract in rabbits consists of descriptions of normal ultrasonographic features in 21 healthy rabbits, including those of the ventral aspect of the stomach wall, pylorus, descending duodenum, jejunum, sacculus rotundus, appendix, cecum, spiral loop of the ascending colon, and descending colon.
Traditional B-mode ultrasonography is commonly used in small animal practice and generates a grayscale image based on the amplitude and depth of returning echoes.10 Doppler ultrasonography identifies an apparent shift in the sound frequency of a moving target within an imaged area and is quantitative, providing velocity of returning echoes. Pulsed-wave Doppler ultrasonography allows identification of that Doppler shift from a specific depth or location, with the Doppler shift displayed as a graph mapping velocity over time.
Duplex Doppler ultrasonography, which consists of simultaneous display of real-time B-mode and PWD images, has been used to provide subjective (B-mode) and more quantitative (PWD) analysis of gastrointestinal peristalsis in humans, horses, and dogs and can be used to differentiate between mechanical and paralytic ileus.11–14 This noninvasive technique analyzes Doppler signals produced by placing the PWD ultra-sonography sample volume on the wall or within the lumen of the gastrointestinal tract. Doppler signals are displayed on a graph, with the x-axis representing time and the y-axis representing velocity. Distinction of peristaltic contractions from nonprogressive movements is based on degree of amplitude and duration of the PWD signal. More specifically, when crescendo-decrescendo waveforms or organized large excursions from baseline (the emitting frequency of the ultrasound transducer) are observed (indicating large shifts of the contents within the sample volume), activity is described as probably peristaltic. Alternatively, when no or minimal deviation from baseline is detected, the gastrointestinal tract is classified as having no peristaltic activity.13 In rabbits, gastrointestinal activity is reported to infrequently involve true aborad peristaltic motion and more frequently segmentation motion throughout the intestinal tract,15 although such observations have not been confirmed via ultrasonography.
Given findings in humans, horses, and dogs, duplex Doppler ultrasonography could provide a minimally stressful and objective technique for assessing gastrointestinal activity in rabbits and has potential as an additional diagnostic technique for differentiating between obstructive and nonobstructive gastrointestinal disease. At the authors’ institution, abdominal ultrasonography has typically been performed on nonsedated rabbits, which generally appear to tolerate the procedure well. However, distressed rabbits brought for evaluation have typically been sedated prior to any extensive handling or diagnostic testing, which may affect gastrointestinal motility. For example, xylazine, ketamine, and acepromazine have different effects on small intestinal motility in horses and dogs.11,14
The purpose of the study reported here was to explore the usefulness of duplex Doppler ultrasonography for evaluation of gastrointestinal activity in healthy nonsedated and sedated rabbits and to evaluate agreement between B-mode and PWD ultrasonographic measurements. We believed that focused ultrasonography of the gastrointestinal tract would be well tolerated by nonsedated and sedated rabbits. We hypothesized that the number of peristaltic contractions as measured via B-mode and PWD ultrasonography would not differ significantly, regardless of sedation status, and that there would be no difference in the number of peristaltic contractions between non-sedated and sedated rabbits.
Materials and Methods
Ethics statement
The protocol for the study was approved by the Tufts University Institutional Animal Care and Use Committee. Owner consent was obtained for all rabbits prior to ultrasonographic evaluation.
Animals
When recruitment began, the first 10 client-owned rabbits brought to the Cummings Veterinary Medical Center for routine wellness examination and for which owner consent was obtained were used to establish B-mode and duplex Doppler ultrasonographic measurements in nonsedated rabbits. An additional 11 client-owned rabbits brought for routine ovariohysterectomy or castration were used to evaluate the effects of sedation on gastrointestinal motility. Rabbits of any age, sex, breed, or body weight were eligible, provided they were deemed healthy on physical examination by an exotic animal specialty veterinarian (JEG and SEK). No food-withholding period was observed, and owners were instructed to allow their rabbit typical access to food prior to evaluation.
Ultrasonography of rabbits undergoing physical examination only
All ultrasonographic evaluations were performed by a board-certified veterinary radiologist (TJO) or senior radiology resident (SA). Each rabbit was first positioned in dorsal recumbency by means of gentle manual restraint. The ventral abdominal region was clipped of hair, and acoustic coupling gel or alcohol was applied. The linear transducer (frequency range, 5 to 18 MHz) of an ultrasound unita was used to identify the stomach, duodenum, a representative segment of small intestine (jejunum), cecum, and colon on the basis of anatomic location and appearance as described elsewhere.9 B-mode still images and 5- to 10-second video recordings of each segment of the gastrointestinal tract were obtained and stored in a local picture archiving and communication systemb for evaluation of luminal contents. The target of the PWD sample gate was placed on the near-field gastrointestinal wall whenever possible as previously described.13 Owing to the small size of the rabbits, placement of the PWD sample gate within the gastrointestinal lumen, as described for horses,14 or at the mucosal-luminal interface could have occurred when the rabbit, gastrointestinal segment, or operator moved during the sample acquisition period, although attempts were made to keep the sample gate as close to the wall as possible. Sample volume was set at 1.5 mm for all rabbits.
Gastrointestinal segments could often be imaged well in only a single plane because of the small size of the rabbits and the large size of the ultrasound transducer, resulting in a near perpendicular angle of the sample volume with the gastrointestinal segment. Given that a previous study13 showed that the Doppler signal from gastrointestinal motility is not influenced by angle of incidence between the Doppler signal emission and the orientation of the gastrointestinal segment, no attempt was made at angle correction. A minimum of three 30-second video clips of the duplex Doppler ultrasonographic evaluation of each gastrointestinal segment were obtained by use of digital frame–capture softwarec and video production software.d Videos were digitally compressed by means of H.264 or MPEG-4 advanced video coding and imported to a video-editing software programe to generate a single compilation video of three 30-second duplex Doppler ultrasonography video clips of each gastrointestinal segment for each rabbit. The same compilations were then edited to crop out the PWD ultrasonography information, generating a B-mode–only video for each rabbit. Videos were saved in MPEG-4 format and reviewed by 1 investigator (TJO) using commercially available software.f
The number of peristaltic contractions in each gastrointestinal segment was determined as reported for horses14 by use of a subjectively identified, high-amplitude, organized, crescendo-decrescendo pattern of Doppler excursion lasting 1 to 2 seconds (Figure 1). Motion due to subject respiration, subject movement, and nonperistaltic waves that did not result in an organized excursion were excluded (Figure 2). Measurements were made for each of the 30-second video clips for each gastrointestinal segment. Similarly, B-mode videos were evaluated for apparent evidence of peristalsis, which was defined as a unidirectional motion of luminal contents (in the sagittal plane) or a single rhythmic dilation and contraction of the lumen (in the transverse plane) lasting 1 to 2 seconds. To-and-fro motion of luminal contents (in the sagittal plane) or lumen dilation without subsequent contraction or a churning appearance (in the transverse plane) was excluded.
Ultrasonography of rabbits undergoing routine ovariohysterectomy or castration
Ultrasonography was performed prior to sedation for routine ovariohysterectomy or castration as described for the other rabbits. Rabbits were then sedated with butorphanol tartrateg (0.5 mg/kg, IM), midazolam hydrochlorideh (0.5 mg/kg, IM), and ketamine hydrochloridei (15 mg/kg, IM). Ultrasonography was repeated as previously described 15 minutes after sedative administration.
Statistical analysis
All statistical analysis was performed with the aid of statistical software.j For each gastrointestinal segment, in both nonsedated and sedated rabbits, the mean ± SD number of contractions during a 30-second period was calculated. The paired t test (for normally distributed data) and nonparametric Wilcoxon signed rank test (for nonnormally distributed data) were used to compare the number of peristaltic contractions in the duodenum and jejunum between nonsedated and sedated rabbits. Values of P < 0.05 were considered significant. Bland-Altman plots of agreement were generated to compare B-mode and PWD ultrasonographic contraction counts of the duodenum and jejunum in all rabbits included in the study (n = 21). The rabbits undergoing gonadectomy (n = 11) were included twice, with nonsedated and sedated contraction counts recorded for a total of 32 measurements for each ultrasound mode (21 nonsedated and 11 sedated measurements). Bland-Altman plots depict instrument bias (both fixed and proportional) on the y-axis (as the difference between the 2 instruments [here, B-mode vs PWD ultrasonography]) as a function of the mean measurement (best estimate of the real value), which is displayed on the x-axis. Thus, the generated Bland-Altman plots showed the agreement between 2 ultrasound modes and any bias in the measurements.
Results
Animals
Of the 21 rabbits included in the study, 7 were castrated males, 7 were sexually intact females, 4 were sexually intact males, and 3 were spayed females. This group included 5 mixed-breed rabbits, 3 English Spot crosses, 3 Netherland Dwarfs, 2 French lops, 2 mini lops, 2 New Zealand rabbits, 1 American Chinchilla, 1 Dutch, 1 Lion-head, and 1 Rex. Mean ± SD age and body weight were 2.14 ± 1.93 years (range, 0.47 to 8.21 years) and 2.05 ± 0.96 kg (range, 0.97 to 4.0 kg), respectively. The sedation protocol appeared to be well tolerated by the 11 rabbits sedated for routine ovariohysterectomy or castration, allowing for minimal manual restraint during image acquisition. All 21 rabbits completed the study.
Ultrasonographic findings
For nonsedated rabbits, only the duodenum and jejunum (and not the stomach, cecum, and colon) had evidence of reliable propulsive contractions. Respective mean ± SD numbers of peristaltic contractions observed during a 30-second period via B-mode and PWD ultrasonography were 1.10 ± 0.55 (range, 0 to 3) and 0.70 ± 0.64 (range, 0 to 2) for the duodenum and 0.67 ± 0.70 (range, 0 to 4) and 0.23 ± 0.42 (range, 0 to 3) for the jejunum. For the cecum, the mean number of contractions observed via B-mode ultrasonography was 0.37 ± 0.46 (range, 0 to 2), but no contractions were observed via PWD ultrasonography. Data for the stomach, cecum, and colon were consequently excluded from further analysis.
Overall, excellent agreement was achieved between B-mode and PWD ultrasonography in the number of peristaltic contractions observed in the duodenum and jejunum of both nonsedated (n = 21 measurements) and sedated (11 measurements) rabbits (Figures 3 and 4), with 2 of 32 measurements outside of the range of agreement (94% agreement). Little difference was observed between nonsedated and sedated rabbits in the number of peristaltic contractions in the duodenum (mean difference, −0.24 and 0.12 for B-mode and PWD ultrasonography, respectively; P ≥ 0.44) and jejunum (mean difference, −0.3 and −0.45 for B-mode and PWD ultrasonography, respectively; P ≥ 0.81).
Discussion
Duplex Doppler ultrasonography, a technique that has been previously used to evaluate in real time gastrointestinal motility in humans, dogs, and horses, was successfully used in the present study to observe and measure peristaltic contractions in healthy rabbits. Although very few to no propulsive movements were observed in the stomach, cecum, and colon of any rabbit, peristaltic waves in the duodenum and jejunum were readily observed with this imaging modality. In addition, no significant difference was observed in the number of peristaltic contractions in the duodenum and jejunum as identified via B-mode versus PWD ultrasonography or in sedated versus nonsedated rabbits.
The lack of consistently visible contractions in the stomach of healthy rabbits was not unexpected given what is known about gastrointestinal physiology in rabbits. The stomach has few inherent contractions, relying on adjacent bowel motility and animal locomotion to churn contents.15 Nevertheless, the lack of reliably identified motility in the cecum and colon was unexpected because multiple types of colonic contractions have been described, including progressive segmental aborad contractions and high-frequency churning motions that vary in timing depending on colonic contents.15 In a previous study14 of healthy fed horses, no contractions were noted in the stomach; however, the cecum and large intestine were always visible with active propulsive motion in the fed horses.
The lack of a difference in duodenal and jejunal contractility between nonsedated and sedated rabbits in the present study was an expected finding. Although many of the routine abdominal ultrasonographic evaluations of rabbits performed at our institution have been performed without the use of sedation, we are aware that rabbits that are distressed or perceived to be easily distressed may receive sedation prior to handling or full diagnostic workup, particularly in an emergency setting. Therefore, it was reassuring to find that sedation had no identifiable adverse impact on ultrasonographic identification of peristalsis.
High agreement between peristalsis measurements obtained via B-mode and PWD ultrasonography was expected in the present study. Although, to the authors’ knowledge, agreement between B-mode and PWD ultrasonographic measurements has not been quantified in any previously reported study involving duplex Doppler ultrasonographic evaluation of the gastrointestinal tract in rabbits, B-mode and PWD ultrasonography had moderate agreement (κ coefficient = 0.51) for assessments of motility in the aforementioned study14 in horses. Subjectively, the investigators in the present study who observed the real-time ultrasonographic assessments (TJO, JEG, and SA) found that the crescendo-decrescendo pattern was easier to identify on the PWD than the B-mode tracings, given that churning, segmentation motion and patient motion could complicate identification of peristalsis during real-time B-mode image acquisition. In the human literature,13 peristaltic waves have been defined as gastrointestinal wall Doppler shifts > 1 kHz lasting 2 seconds; however, this specific criterion did not appear to apply to the rabbits in our study because PWD crescendo-decrescendo patterns were brief. Therefore, we recommend the subjective, more practical use of a visual crescendo-decrescendo pattern if PWD ultrasonography is used.
Duplex Doppler ultrasonography, like other ultrasonographic techniques, is subject to some limitations and artifacts. Respiratory motion, patient movement, and inadvertent excessive manual pressure from the operator may create artifactual excursions from baseline on the PWD graphs that must be distinguished from true intestinal contractions. These artifactual motions would be most likely to cause anomalous high-velocity spikes in the PWD tracing that lack an organized crescendo-decrescendo pattern. In addition to motion artifacts, gas within the rabbit gastrointestinal tract could be considered a potential limitation; indeed, as reported for horses,14 gastrointestinal luminal contents resulted in various artifacts in the rabbits of the present study, including reverberation due to gas and acoustic shadowing due to dense ingesta. In horses, reliable Doppler patterns could not be obtained for the stomach, cecum, or large intestine because of the presence of intraluminal gas.14 Conversely, the presence of intraluminal gas has no impact on Doppler patterns in humans,13 and we were able to identify the stomach, proximal aspect of the duodenum, jejunum, cecum, and colon in all rabbits as previously described.9
Gas and shadowing luminal contents as well as rabbit motion in the present study resulted in variability in the orientation of the transducer and, hence, the PWD sampling gate. Although this may have resulted in data acquisition for both the gastrointestinal wall and lumen in some rabbits, we were still able to obtain typical peristaltic waveforms with no artifact from the Doppler signal of mural vessels, an artifact that was similarly avoided in humans.13 In addition, Doppler ultrasonographic measurements are most accurate when the Doppler beam is parallel to the structure of interest,10 but the size of the rabbits and the size of the ultrasound transducer resulted in many measurements being acquired with larger angles of incidence to the structure of interest. For true quantitative Doppler ultrasonographic measurements, this large angle of incidence can result in considerable changes in calculated Doppler shift velocities.10
Another potential source of error in the study reported here was the brief data acquisition period. We chose a 30-second period (rather than a typical 1-minute period) given our experience with rabbit motion, which made maintaining the location of the PWD sample gate difficult. However, use of this briefer period may have resulted in underestimation of the amount of peristaltic activity. In addition, no attempt was made to gauge the amount of distress the rabbits may have had (eg, by measuring heart rate) before and after sedation, and the lack of an observable difference in measurements between these 2 states may have been due to minimal effects of sedation.
No well-established criteria are available for the use of widely used imaging modalities such as radiography and ultrasonography to differentiate between nonobstructive and obstructive gastrointestinal disease in rabbits. Because B-mode and PWD ultrasonographic measurements of peristalsis had good agreement in the present study, and B-mode ultrasonography is more readily available to clinicians, we propose the use of B-mode ultrasonography to visualize rabbit gastrointestinal motility. However, given that some healthy rabbits in the present study had no duodenal or jejunal peristaltic contractions during this brief period, additional research with a larger number of rabbits and longer observational periods would be important to better define the number of contractions that might be expected in healthy rabbits, before comparisons can be made with rabbits with gastrointestinal disorders. In addition, because previous data indicate that true peristaltic waves are infrequent in rabbits,15 future studies should include evaluation of segmentation motion of the duodenum and jejunum as well.
Acknowledgments
Funded by the Association of Exotic Mammal Veterinarians 2016 Benjamin and Bella Rabbit Research Grant.
Presented in abstract form at the 15th Annual Conference of the Association of Exotic Mammal Veterinarians, Dallas, September 2017.
The authors thank Dr. Joanna Webb for data collection and patient recruitment.
ABBREVIATIONS
PWD | Pulsed-wave Doppler |
Footnotes
Philips EPIQ 7, Philips, Bothell, Wash.
Carestream Health Inc, Rochester, N Y.
Epiphan DVI2USB, Epiphan Systems Inc, Palo Alto, Calif.
Wirecast, Telestream LLC, Sterling, Va.
Camtastia, TechSmith Corp, Okemos, Mich.
Quicktime, Apple, Cupertino, Calif.
Torbugesic, Fort Dodge Animal Health, Overland Park, Kan.
Midazolam, West-Ward Pharmaceutical Corp, Eatontown, NJ.
Putney Inc, Portland, Me.
SAS, version 9.4, SAS Institute Inc, Cary, NC.
References
1. Graham JE. Lagomorpha (pikas, rabbits, and hares). In: Miller E, ed. Fowler's zoo and wild animal medicine current therapy. 8th ed. St Louis: WB Saunders, 2014;375–384.
2. DeCubellis J, Graham J. Gastrointestinal disease in guinea pigs and rabbits. Vet Clin North Am Exot Anim Pract 2013;16:421–435.
3. Huynh M, Vilmouth S, Gonzalez MS, et al. Retrospective cohort study of gastrointestinal stasis in pet rabbits. Vet Rec 2014;175:225.
4. Lichtenberger M, Lennox A. Updates and advanced therapies for gastrointestinal stasis in rabbits. Vet Clin North Am Exot Anim Pract 2010;13:525–541.
5. Oglesbee BL, Jenkins JR. Gastrointestinal diseases. In: Quesenberry KE, Carpenter JW, eds. Ferrets, rabbits, and rodents: clinical medicine and surgery. 3rd ed. St Louis: Elsevier, 2012;193–204.
6. Harcourt-Brown FM. Gastric dilation and intestinal obstruction in 76 rabbits. Vet Rec 2007;161:409–414.
7. Meiser G, Meissner K. Ileus and intestinal obstruction—ultrasonographic findings as a guideline to therapy. Hepatogastroenterology 1987;34:194–199.
8. Tyrrell D, Beck C. Survey of the use of radiography vs ultrasonography in the investigation of gastrointestinal foreign bodies in small animals. Vet Radiol Ultrasound 2006;47:404–408.
9. Banzato T, Bellini L, Contiero B, et al. Abdominal ultrasound features and reference values in 21 healthy rabbits. Vet Rec 2015;176:101.
10. Mattoon JS, Nyland TG. Fundamentals of diagnostic ultrasound In: Mattoon JS, Nyland TG, eds. Small animal diagnostic ultrasound. 3rd ed. St Louis: Elsevier, 2015;32–36.
11. An YJ, Lee H, Chang D, et al. Application of pulsed Doppler ultrasound for the evaluation of small intestinal motility in dogs. J Vet Sci 2001;2:71–74.
12. Gimondo P, La Bella A, Mirk P, et al. Duplex-Doppler evaluation of intestinal peristalsis in patients with bowel obstruction. Abdom Imaging 1995;20:33–36.
13. Gimondo P, Mirk P. A new method for evaluating small intestinal motility using duplex Doppler sonography. AJR Am J Roentgenol 1997;168:187–192.
14. Mitchell CF, Malone ED, Sage AM, et al. Evaluation of gastrointestinal activity patterns in healthy horses using B mode and Doppler ultrasonography. Can Vet J 2005;46:134–140.
15. Davies RR, Davies JA. Rabbit gastrointestinal physiology. Vet Clin North Am Exot Anim Pract 2003;6:139–153.