Introduction

The North American Porcupine (Erethizon dorsatum) occupies much of the United States and Canada. A retrospective study of porcupine injury in dogs found that 81.2% had quills in the head and neck region.¹ Identification of quills via imaging can be a challenge. One case report found that a quill within the vertebral canal could be seen retrospectively on MRI after it was surgically removed.² Computed tomography (CT) has also been successfully used to diagnose an intracranial intra-axial porcupine quill in a dog.³ Ultrasound has been used to identify ocular porcupine quills.⁴ To the authors’ knowledge, methods for diagnosing intranasal porcupine quills in a dog, and the related limitations of CT, have not been previously reported.

Case Report

A 6 yr old spayed female Rhodesian ridgeback was presented for a 6–8 wk history of nasal sensitivity and bilateral serous nasal discharge after being quilled by a porcupine (Figure 1). The patient had many quills removed just after the incident. A few weeks later, she was seen by her primary care veterinarian who lanced a swelling that was present on the mid dorsal maxilla from which clear fluid was extruded. Antibiotic therapy was initiated, and the swelling significantly reduced in size. Despite treatment, the patient continued to sneeze, paw at her nose, and have less interest in playing since the quilling. The owner also perceived discomfort associated with her left ear.

View Figure

• Download as Powerpoint
• Download Figure

On physical exam the patient had a sensitive, 2 × 1 × 0.3 cm swelling on midline dorsal to her nasal bones and a 2 mm diameter red area on her central philtrum where a tiny piece of firm white foreign material was removed during the exam. It is unclear whether this removed material was quill-related. A scar was appreciated over the dorsal head, rostral to the left ear. There were also other smaller dermal scars. The rest of her physical exam was unremarkable. Preanesthetic blood work, including venous blood gas, packed cell volume, and total solids, was also unremarkable.

A Toshiba Aquillion 64-detector CT scanner was used to acquire 3 mm thick slices from the rostral nares to C2–C3. A soft-tissue algorithm (W: 400, L: 40) and high-definition bone algorithm (W: 4500, L: 450) were used to acquire initial precontrast images. Postcontrast images were collected following intravenous administration of Iohexol. All images were reviewed by both a board-certified small animal internist and radiologist. Evaluation revealed fluid within the left tympanic bulla containing a central hyperattenuating thin linear structure not recognizable as part of the normal anatomy, bilaterally mineralized auricular cartilages, and faint, nonenhancing soft-tissue attenuating wisps observed within the subcutaneous tissue superficial to the left maxillary salivary gland. Otitis media was suspected; a quill could not be definitively identified in association with this finding. Edema or cellulitis was suspected within the soft tissues. The nasal cavity was interpreted as normal without evidence of soft-tissue densities or alterations in turbinate densities.

An Olympus 2.8 mm flexible video bronchoscope (BF-1TQ180) was used for direct rhinoscopic examination and to evaluate the nasopharynx (retroflex rhinoscopy). Both the left and right nasal cavities had normal turbinate density consistent with the CT. Nasal fluid collections were not appreciated. The mucosal appearance was overall normal without overt hyperemia or irregularity. However, within the left nasal cavity, there was a 1 cm long porcupine quill in the dorsal meatus extending dorsally into the soft tissue (Figure 2). Rigid cup biopsy forceps were used for extraction under endoscopic visualization. Within the right nasal cavity, one 1.5 cm long quill was removed from the dorsal lateral meatus ∼5 cm into the nasal passage, and an additional quill of equal length was removed from the right middle nasal meatus. Oral and nasopharyngeal exams with the scope were unremarkable. Otoscopic evaluation under anesthesia, performed as a result of the history of left ear sensitivity and suspicion on CT for otitis media, did not reveal evidence of tympanic membrane disruption or of quills within the ear canal.

View Figure

• Download as Powerpoint
• Download Figure

The patient recovered from anesthesia without complications. She was prescribed a 3 wk course of cefpodoxime for otitis media and cellulitis. One week after the porcupine quill removal, sneezing had resolved, and she was clinically doing well. At 6 mo follow-up, the patient was doing well, and no further intervention had been required.

Discussion

CT, MRI, and ultrasound have been used to identify foreign bodies, but they have limitations. Ultrasound is not useful for nasal foreign bodies because of the bone and air in the area.⁵ In one human study looking for foreign bodies in the foot, overall sensitivity and specificity for foreign body detection was 63 and 98% for CT and 58 and 100% for MRI.⁶ CT detection rates depended on the attenuation values of the foreign bodies, whereas MRI hinged on susceptibility to artifacts.⁶ A study comparing the sensitivity of CT with MRI for diagnosing infraorbital foreign bodies in pig eyes also showed that CT was more reliable as a result of artifacts seen on MRI that prevented foreign body identification.⁷ In a study comparing plain radiography, CT, MRI, and ultrasonography identification of metal, tooth, wood, plastic, stone, glass, and graphite in sizes 0.5, 1, 2, and 3 mm inserted into the maxillofacial region of sheep heads, MRI was least useful in foreign body identification.⁸ CT, on the other hand, was shown to be the most effective tool for identifying small foreign bodies in a variety of densities.⁸ Another benefit to CT is that hounsfield units (HU) can help predict the type of foreign material present.⁷ CT has been used successfully to identify foreign material in the nasal cavity of a dog and to identify porcupine quills in other areas of the canine body.^9,3 For these reasons, CT was chosen to investigate the dog’s nasal cavity in this case.

In this case, the nasal CT was unremarkable and failed to explain the patient’s clinical signs, but rhinoscopy revealed porcupine quills in both nasal passages, thereby demonstrating the limitation of CT in identifying intranasal porcupine quills. Although advanced imaging, such as CT, and rhinoscopy are often synergistic, this was not true in this case. The limitations of CT for identifying nasal foreign bodies have been previously documented. A study on the use of CT for diagnosing the cause of chronic nasal disease in dogs found that the positive predictive value was significantly less for foreign body rhinitis than for neoplasia, aspergillosis, or nonspecific rhinitis when the foreign body could not clearly be detected.⁹ Another study of CT identification of nongastrointestinal foreign bodies in 13 dogs, 7 of which were nasal foreign bodies, found that the foreign body was not visible in 5 cases.¹⁰ To the authors’ knowledge, the limitations of CT for identifying intranasal porcupine quills have not been previously reported.

Possible explanations for why the quills were missed include size and density. The porcupine quills were ∼1–2 mm thick. As the CT slices were 3 mm thick, the quills may have been missed or the cross-sectional appearance of the quills not recognized. However, given the length of the quills (1–1.5 cm) and the angulation of the quills on rhinoscopy, this was considered unlikely.

HU are measurements of density. A linear scale was created by defining the radiodensity of air as –1000 HU and the radiodensity of distilled water at standard temperature and pressure to 0 HU. Therefore, substances more dense than distilled water result in a positive HU. The density of different sites on a CT image can sometimes help determine whether they are bone, soft tissue, or foreign material. The density of porcupine quills has been reported previously as 52–68 HU.³ The density of the nasal turbinates in this patient ranged from 110 to 700 HU. However, given the thin nature of the turbinates and the bone to air interface, it is easy to obtain lower readings if not exactly centered on bone. Therefore, distinction between the nasal turbinates and quills could be challenging, as the quills would not necessarily stand out. The small size and similar density could lead the quills to easily be mistaken as artifact or part of the normal anatomy.

Changing the CT settings to 1 mm slices may be beneficial for future cases of suspected intranasal porcupine quills since this may help address their thin width; however, the density of the quills will remain an obstacle for CT diagnosis.

Conclusion

In conclusion, we described a case of three intranasal porcupine quills in a dog that were not visible on CT. This case demonstrates the importance of recognizing the limitations of diagnostic imaging. Decreasing CT slice thickness may be beneficial; however, the density of porcupine quills will continue to make them challenging to identify within the nasal cavity. These findings suggest that rhinoscopy should be performed in all cases suspicious for foreign body rhinitis in dogs regardless of the imaging findings.