Introduction

In dogs, colorectal masses are uncommon. Rectal leiomyoma,¹ leiomyosarcoma, plasmacytoma,² adenomatous polyps, carcinoma in situ,³ and adenocarcinoma⁴ are the most common colorectal neoplasms in dogs. Biopsy techniques for colorectal tumors include cup biopsy via colonoscopy or proctoscopy,⁵ endoscopic polypectomy,⁶ and incisional biopsy performed via direct visualization.⁷

If a mass is located orad to the terminal rectum, then it must be accessed via flexible or rigid endoscopy^8,9 or mucosal eversion¹⁰ in order to maintain direct visualization. Flexible colonoscopy necessitates retroflexion to view the rectum, which may not be possible with a space-occupying mass; however, biopsies may still be acquired.¹¹ Snare polypectomy with electrocautery for excisional biopsy of colorectal polyps with or without saline dissection is another minimally invasive technique.¹² Suture-assisted prolapse or “pull out” is limited by how much of the colorectum may be tractioned to achieve visualization and by the extent and distance of the mass from the rectal orifice.^9,10,13 Allis tissue forceps have also been suggested as a means of achieving exposure without suture.¹⁴ A technique using a rigid resectoscope for examination and treatment of colorectal masses has also been described.⁹ This alternative to the “pull-through” technique is used for debulking of nonresectable masses as a salvage technique and is therefore not appropriate for discrete small masses, which would benefit from a pull-out technique using mucosal eversion.⁹ In the setting of extensive lesions, it is associated with fewer complications but requires equipment availability and surgical expertise. It is important to note that this method was produced as an alternative to the pull-through technique, which is associated with such a poor postoperative prognosis that some authors do not consider it humane.⁹

Many of these techniques require special equipment and/or extensive training. A more important consideration, however, is the diagnostic quality of the acquired sample. Although the pull-through technique provides the clinician with full-thickness samples of the colorectal wall, the complications of this procedure may be too significant. Although it has more complication potential than minimally invasive options, mucosal eversion when possible can provide full-thickness diagnostic samples. At the other extreme, colonoscopy provides only a small, superficial sample. Extent and depth of sample are both expected to influence pathologist diagnosis.¹⁵

The authors propose that with correct case selection of discrete noncircumferential masses, obturator-assisted prolapse allows for atraumatic direct observation, stabilization, and isolation of mass tissue for diagnostic purposes in a manner that is simple and economical. Although full-thickness excisional biopsies are not attainable via this technique, the size of the sample collected via core needle biopsy is likely to be comparable to cup biopsies but with the potential of deeper margins. This may prove to be a useful, inexpensive alternative for clinicians to achieve diagnostic sampling if standard measures such as endoscopic biopsy or surgical biopsy are not available or financially feasible.

Case Report

There were seven patients with colorectal masses in this case series (Table 1). The median patient age was 9 yr, with a range of 6.5–11 yr of age. The patient weights ranged from 5.9 to 46.8 kg with a median weight of 18.4 kg. There were five female spayed dogs and two male neutered dogs. The breeds consisted of three mixed-breed dogs, one Labrador retriever, one cocker spaniel, one standard schnauzer, and one shih tzu.

TABLE 1 Results from the Rectal Mass Access Procedure

View Fullscreen

Computed tomography (CT) of the abdomen, both prior to and following administration of IV radiographic contrast agent, was performed in all patients. The longest axis of colorectal masses identified via CT was a median of 5 cm (range of 2.2–8.9 cm). The distribution of masses is described in Table 1. Three patients had colorectal masses and four patients had solely rectal wall involvement on CT scan. Of note, three of the patients with rectal masses had focal masses (2.2–3.3 cm) present on the right lateral aspect of the rectal wall located proximal to the terminal rectum at the level of the second to fifth coccygeal vertebrae. The other patient with a rectal mass had a rectal wall mass originating at the level of the sacral body and extending to the anus, occluding 100% of the lumen of the terminal rectum.

Staging was performed with thoracic radiographs in addition to the abdominal CT scan in all patients prior to sampling. Metastasis was suspected in three patients based on contrast enhancement with regional lymph nodes enhancement and/or liver nodules. There were two remaining patients that had less certain metastatic disease noted (one regional at the internal iliac lymph node, one distant pulmonary nodule).

Mass location in all seven patients was confirmed by digital rectal examination prior to procedure. In the patients with carcinomas that appeared ill-defined on CT scan, rectal examination demonstrated a clear mass with a roughened margin. Despite wall thickening being demonstrated as a primary feature on CT imaging, conformation of a mass on rectal examination allowed for decision making. Cases were selected based upon several factors including owner financial limitations, minimally invasive sampling requested because of suspect metastasis, difficulty of eversion of the tissues, and aboral location of the mass limiting effective eversion. Case recruitment was limited because cases were selected with these specifications. In many of the cases, the exposure was such that a needle core could be used to sample the mass but an incisional biopsy would have been challenging. A cup biopsy would have been possible, but the length of throw of the core needle biopsy was considered to have an advantage for extent of the sample.

To facilitate safe acquisition of a diagnostic biopsy (Figure 1), the plastic rigid obturator from a commercially available disposable-type proctoscope in either a 19 mm × 20 cm or 20 mm × 25 cm size^a,b was first maneuvered past a mass in the colorectum or rectum. This was performed with the tip oriented facing away from a mass (Figure 1A) and deflecting the colorectal wall. Once the tip of the obturator was past a mass, the handle was rotated 180 degrees (Figure 1B) to engage the rim of the tip on the proximal side of a mass. This mass was then deflected and moved caudally by several centimeters (Figure 1C), exposing it. This allowed for gross examination, ease of mass palpation, and improvement of exposure for sampling. Several 14 gauge core needle^c biopsies were then collected (Figure 1D), with the core needle angled toward the shaft of the plastic proctoscope obturator that was used to shield the opposite colorectal wall.

View Figure

• Download as Powerpoint
• Download Figure

Obturator-assisted prolapse was performed in all patients following manual evacuation of feces and an enema. All of the masses in this report were effectively visualized and their morphology and association with the mural surface noted (Table 1). After the biopsies were performed, minor hemorrhage was noted in five patients, and one patient had difficulty defecating for 1 wk after the procedure. An additional patient had perhaps the most significant complication noted. This patient had moderate self-limiting hemorrhage with brief bradycardia and hypotension. The bradycardia (heart rate = 70) and hypotension (systolic blood pressure 70 mm Hg) occurred at the time of the procedure when the obturator was initially inserted and responded to a hetastarch bolus within 5–10 min; thus, these complications were unlikely related to the biopsy itself. There was another brief episode of hypotension 25 min later at the end of anesthesia when the biopsies were acquired that resolved spontaneously.

Histopathology revealed three rectal carcinomas (one papillary, two in situ), two rectal polyps, one leiomyoma (suspect), and one pleocellular inflammation (differential diagnosis including soft tissue sarcoma and histiocytic sarcoma). One colorectal mass was described on histopathological diagnosis as likely an adenomatous polyp, although the pathologist stated that carcinoma could not be ruled out by histopathology. No cases included the serosal wall of the viscus sampled. In six out of seven cases, there was tissue present below the level of the basement membrane. Three patients received tumor-directed care following the biopsy: one patient underwent definitive surgical excision of the mass, one underwent palliative radiation therapy, and one received medical treatment with piroxicam. One patient had subsequent surgical excision of the mass via a ventral midline approach with a pubic osteotomy to remove a mass that proved to be a colorectal leiomyoma. The needle core biopsy of this patient was consistent with the surgical pathology. Of the remaining four patients, one received palliative care (analgesics) only and two were euthanized for reasons unrelated to biopsy complications. There were six patients with follow-up information recorded ranging from 33 to 475 days (median 51 days) after the procedure, and four of these patients were later lost to follow-up. Of the patients for which follow-up was available, there were no major complications at 65–475 days after the procedure. When the patient that lived 65 days after the procedure was euthanized because of anorexia, it was interpreted as resulting from progression of cancer rather than biopsy complications.

Discussion

A diagnostic quality sample for subsequent treatment planning was achieved in all cases biopsied using obturator-assisted prolapse. No major postprocedural complications suggestive of colonic wall perforation occurred. This technique was of significant benefit in cases with discrete, narrow- to medium-base, relatively firm mural masses that allow for ease of engagement of the obturator and partial prolapse. Also, when consideration of the pull-out technique is complicated by difficulty of mucosal eversion, this may be a preferable solution for biopsy. This is not a technique that the authors would recommend for flat diffuse thickening that cannot be engaged with the obturator, or for highly friable or necrosed masses. Additionally, the clinician should consider the diameter of the obturator, mass, and rectum prior to selecting which obturator to use. If too large an obturator is selected, then there is the risk of lodging the device cranial to the mass and creating trauma and a foreign body. Another consideration is that tru-cut biopsies must seat the core needle deep enough in the tissue to acquire a sample. The authors did not select cases in which there was a significant concern that the colorectum could not be adequately shielded by the obturator.

Five patients had possible metastasis noted on CT scan. Another consideration for the clinician when performing biopsies is the invasiveness of the procedure. The significant complications of the pull-through technique have already been mentioned as a reason to consider less invasive methodology. Far less traumatic is the mucosal eversion technique that may provide margins and full-thickness samples. The authors suggest that especially in cases where metastasis is already suspected and access is challenging, finding the least invasive method of sampling is desirable. In addition to lymph node aspirates, this is a setting in which obturator-assisted prolapse is perhaps of maximal benefit.

The technique described in this paper has some limitations. First, depending upon the amount of debris present and the degree of occlusion of the lumen, flexible colonoscopy may be able to show a magnified full field of view and allow for better assessment of postbiopsy hemorrhage and wall integrity. Additionally, the obturator sizes were 19 mm × 20 cm and 20 mm × 25 cm, respectively, and therefore, there could be access limitations imposed by the diameter of the obturator if it cannot be maneuvered beside a mass. Second, the size of sample provided via core needle biopsy is smaller than that provided by excisional biopsy or open surgical approaches and thus provides less tissue for evaluation. In the second, fourth, and fifth cases (Table 1), there may have been minor diagnostic discordance between the biopsy results and actual rectal tumors present because the biopsy did not include the entire rectal wall. The clinical findings in all three dogs were suggestive of invasive and/or recurrent colorectal adenocarcinoma, but none of the biopsies were consistent with this degree of malignancy because they were partial-thickness samples. However, the needle core biopsies obtained in this study provided useful information as to tissue of origin (i.e., epithelium) for each of these three tumors and would have been sufficient to guide definitive therapy had the dogs’ owners elected surgical intervention or radiation oncology consult. Therefore, the diagnosis of carcinoma in situ may not have been accurate.

Clinicians should rely on comparison of biopsy results to clinical signs, physical exam findings of tumor extent, and imaging studies when interpreting results of colorectal biopsies. If a biopsy suggests a benign mass with a highly invasive tumor, then clinical judgement must be used. Third, if a concurrent colonoscopy is not performed, and advanced imaging is not performed, then additional lesions within the colon or even colorectum may not be evaluated. Such extensive evaluation is not, however, within the financial means of many clients, and obturator-assisted prolapse may also provide a salvage sampling method when other methods are unavailable. Fourth, this technique is mainly possible with polypoid and distinct wall masses. The technique does not work well with thickening or completely circumferential masses that would prevent passage of the obturator or risk damage when withdrawing the obturator. If poor integrity of the colonic wall is suspected, then this technique should be avoided. Lastly, the clinician should be aware of potential complications of the procedure. One patient in this series experienced hypotension, bradycardia, and moderate hemorrhage that were likely procedure related. We suspect that this patient experienced a brief vagal event secondary to stretch within the colorectum. This is a complication that should be anticipated. Additionally, like with any other biopsy acquisition, the possibility of hemorrhage should be considered. In this case, the hemorrhage was controlled with local tamponade.

This new technique improves access to colorectal masses in dogs with no major complications seen up to 475 days of follow-up. A recent paper involving 82 dogs undergoing colonoscopy showed that rectal masses are typically not accompanied by additional masses in the proximal colon.¹⁶ In light of these findings, an efficient method of biopsy that does not require extensive preparation and can be performed under brief anesthesia, such as the method described here, is potentially useful.

Conclusion

Obturator-assisted prolapse has several advantages over routine proctoscopy, including technical simplicity, minimal expense, excellent direct visualization, and reduced risk of viscus organ perforation by use of the proctoscope obturator to shield the rectal wall during biopsy. Further study with a larger cohort and long-term follow-up is recommended to document more completely the safety and efficacy of this method.