Ultrasonographic and Magnetic Resonance Imaging Diagnosis of an Oropharyngeal Wood Penetrating Injury in a Dog

Introduction

Penetrating oropharyngeal injuries in dogs are most commonly due to wooden foreign objects in medium to large breed dogs.^1–5 Part of the wooden object may remain in the soft tissues and migrate to various locations including the retroorbital space, retropharyngeal region, and submandibular and cervical soft tissues.^1–3,6–8 The majority of dogs present as chronic cases with common complaints of swelling, draining tracts, recurrent abscessation, or pyogranulomas.^2,9

Diagnostic imaging is commonly used in the diagnosis of wooden foreign bodies. Radiographic examination is often unrewarding for visualization of wood because chronic wooden foreign objects are of similar opacity to soft tissues and fluid due to their ability to absorb and retain water. Ultrasound advocates the diagnosis of wooden foreign bodies but is limited by its inability to image deep to air and bone. The use of CT and MRI has become more common in the diagnosis of oropharyngeal wooden foreign bodies. The imaging modality of choice may change on a case-by-case basis. This report describes the ultrasonographic and MRI appearance of an oropharyngeal wooden foreign body and surrounding affected soft tissues. Complementary multimodal diagnostic imaging allowed for accurate diagnosis, localization, and surgical planning.

Case Report

A 5 yr old, 13.9 kg, intact female English setter was referred to the Kansas State University Veterinary Medical Teaching Hospital for evaluation of recurrent cervical pain and swelling. The dog had a stick removed from the soft palate 17 days prior to presentation by the referring veterinarian. At the time of stick removal, the dog had been prescribed amoxicillin^a (36 mg/kg per os [PO] q 12 hr for 10 days) and carprofen^b (3.6 mg/kg PO q 24 hr for 8 days). One week following stick removal, the dog was painful during extension of the neck. On day 15, an additional course of carprofen^b (3.6 mg/kg PO q 24 hr for 6 days) was prescribed. There was no improvement of the neck pain by day 17 when the dog presented to the authors’ institution. The dog had developed a large swelling in the left cranial cervical region. Physical examination revealed pain upon flexion, extension, and leftward movement of the neck. Neurologic deficits were not noted upon complete neurologic evaluation. The patient's rectal temperature was 39.3°C and the pulse rate was 108 beats per minute. A CBC had been performed by the referring veterinarian on day 15 and was not repeated. The CBC abnormalities included a mild leukocytosis (18,060 μL; reference range, 6,000–17,000 μL) with a granulocytosis (16,690 μL; reference range, 300–12,000 μL) and a lymphopenia (490 μL; reference range, 1,000–4,800 μL). Serum biochemical analysis performed the day of presentation revealed only a mild hypoalbuminemia (2.8 g/dL; reference range, 3.0–4.5 g/dL). Results of a free-catch urinalysis were normal, ruling out a protein-losing nephropathy. The dog was admitted to the hospital and prescribed acetaminophen (APAP) with codeine^c (1 mg/kg PO q 8 hr) for the cervical pain. In light of the history, the primary differential for the cervical swelling was abscess and cellulitis secondary to migration of a wooden foreign body.

A cervical ultrasound was performed and a 0.4 cm × 2.3 cm linear, hyperechoic structure with distal acoustic shadowing was identified in the soft tissues lateral to the first (C1) and second (C2) cervical vertebrae (Figure 1). The caudal third of the structure was divided into two distinct, parallel, linear structures. The adjacent musculature was heterogenous, hyperechoic, and had decreased definition of its linear architecture. There was enlargement of the left medial retropharyngeal lymph node, which measured 1.1 cm in diameter. The ultrasound findings were consistent with a foreign body with surrounding cellulitis, myositis, and reactive lymphadenopathy.

View Figure

• Download as Powerpoint
• Download Figure

Due to the proximity of the foreign body to the C1-C2 intervertebral foramen and the potential for spinal canal and nerve root involvement, MRI was performed. The dog was premedicated subcutaneously with acepromazine^d (0.03 mg/kg) and hydromorphone^e (0.1 mg/kg) and general anesthesia was induced with thiopental^f (8 mg/kg IV to effect). Anesthesia was maintained with 1.5% isoflourane^g in oxygen. A low-field (0.3 Tesla) magnet with a human knee coil^h was used with the following pulse sequences and scan planes: sagittal and transverse plane T1-weighted spin echo (repetition time [TR]=600 msec; echo time [TE]=25 msec; flip angle [FA]=90 degrees), sagittal, transverse, and dorsal plane T2-weighted fast spin echo (TR=3,100 msec; TE=119 msec; FA=90 msec), and transverse postcontrast T1-weighted spin echo. Postcontrast images were obtained after the IV administration of 3 mL (0.1 mmol/kg) gadopentetate dimeglumineⁱ.

T1-weighted precontrast transverse plane images demonstrated a cylindrical structure that was iso- to hyperintense compared to normal muscle adjacent located to the left lateral aspect of C1 and C2. One larger diameter cylindrical structure was present at the caudal aspect of C1 and two smaller cylindrical structures were identified at the midbody of C2. These three structures were surrounded by soft tissue with intensity between that of the cylindrical structures and normal muscle. Additional areas of mixed signal intensity were present throughout the subcutaneous tissue and musculature of the left cervical region. Thickening of the muscle bellies and loss of fascial planes was also present. The left medial retropharyngeal lymph node was enlarged, measuring greater than 1 cm in diameter (Figure 2). On the transverse and dorsal plane T2-weighted images, the cylindrical structure was hypointense to normal muscle. Hyperintense tissues were present surrounding the cylindrical structures and within the left cervical musculature and subcutaneous tissues (Figures 3 and 4). Hyperintense cervical soft tissues extended from the base of the skull to the level of the fifth cervical vertebra. On the transverse plane T1-weighted postcontrast images, the cylindrical structure was isointense compared with normal musculature. There was marked contrast enhancement of surrounding soft tissues as well as inhomogenous contrast enhancement of the subcutaneous tissues and left cervical musculature (Figure 5). These MRI findings were consistent with a foreign body with surrounding cellulitis, myositis, and reactive lymphadenopathy. No abnormalities within the spinal canal or spinal cord were noted.

View Figure

• Download as Powerpoint
• Download Figure

View Figure

• Download as Powerpoint
• Download Figure

View Figure

• Download as Powerpoint
• Download Figure

View Figure

• Download as Powerpoint
• Download Figure

On the transverse and dorsal plane T2-weighted images, the cylindrical structure was hypointense to normal muscle. Hyperintense tissues were present surrounding the cylindrical structures and within the left cervical musculature and subcutaneous tissues. Hyperintense cervical soft tissues extended from the base of the skull to the level of the fifth cervical vertebra. These MRI findings were consistent with a foreign body with surrounding cellulitis, myositis, and reactive lymphadenopathy. No abnormalities within the spinal canal or spinal cord were noted.

The dog was scheduled for surgery for foreign body removal. A single dose of cefazolin^j (21.5 mg/kg IV) was administered preoperatively. An incision was made on the left lateral aspect of the neck, adjacent to C1 and C2. The underlying muscle layers were bluntly dissected and the belly of the brachiocephalicus muscle was partially transected and retracted ventrally. The serratus ventralis muscle was moderately swollen with granulation tissue and purulent material. A foreign body approximately 2 cm in length was found adjacent to the C2 nerve root and removed (Figure 6). A surgical drain consisting of a 19 gauge butterfly catheter and a 7 mL vacutainer tube was placed, exiting the skin through a stab incision made dorsal to the original incision. The dog's anesthetic recovery was prolonged; therefore, a single dose of naloxone^k (0.006 mg/kg IV) was administered.

View Figure

• Download as Powerpoint
• Download Figure

Postoperatively, the dog was prescribed amoxicillin/clavulanic acid^l (13.4 mg/kg PO q 12 hr) and continued on APAP codeine^c (1 mg/kg PO q 8 hr). Pseudomonas aeruginosa was isolated from culture of the surgical site. This isolate was shown to have resistance to amoxicillin/clavulanic acid^l, therefore antibiotics were changed to ciprofloxacin^m (10 mg/kg PO q 12 hr). After 2 days, the drain was removed and the patient was discharged.

Two weeks later, reexamination revealed the dog had gained weight, was eating normally, and the surgery site had healed. Four weeks postoperatively, she had fully recovered and regained normal function in her cervical region.

Discussion

Previous reports have described oropharyngeal penetrating injuries to be classified as either an acute or chronic presentation. In acute injuries, dogs normally present with signs of dysphagia, pyrexia, hypersalivation, dyspnea, and oral pain.^1,2,5,8,10 Chronic injuries, as in this report, have the most common presenting complaints of swelling, intermittent discharging sinuses, and pain.^2,5 Dogs with chronic injuries have loss of soft tissue structures, evidence of lacerations, and cellulitis with gas accumulation between tissue planes on radiography.^1,10 The signalment, history, and presentation of the dog presented herein are similar to these previous reports describing chronic oropharyngeal penetration and foreign bodies in dogs.

Many diagnostic imaging modalities can be used in the evaluation of suspected wooden foreign bodies within the musculoskeletal system. These modalities include radiography, ultrasonography, CT, and MRI. There is a limited amount of literature regarding CT and MRI findings in veterinary patients with wooden foreign bodies.³ Patients presenting with a chronic swelling or a draining tract, radiographs are commonly used as the initial screening modality to identify radiopaque objects and define osseous involvement. The value of radiography is often limited in the detection of wooden foreign material. After penetration into the body, wood generally absorbs fluid resulting in an opacity that is similar to the surrounding soft tissues on survey radiographs.^2,11 Sinography can potentially be useful if there is a draining tract present; however, no draining tract was identified in this case.¹² The history of the dog described in this case indicated a high likelihood of a wooden foreign body. Thus, ultrasound was chosen first. Ultrasound has proven to be very useful for the determination of the size, shape, and location of foreign material in soft tissue.^3,5,6,11,13 Confirmation of a foreign body via ultrasound and knowledge of surrounding anatomic landmarks allows for a more direct surgical approach, which limits tissue damage and decreases surgical time.⁵

A major limitation of ultrasound is its inability to image deep to bone or gas. Because of the proximity of the foreign body to the left of the C1-C2 intervertebral foramen, additional presurgical imaging was elected to assess possible spinal canal involvement. The two main imaging options considered were CT and MRI. In a recent study, CT was shown to be more accurate than ultrasound and MRI in the detection of small wooden foreign bodies placed in cadaver paws.¹⁴ A caveat to that study was the inability to use contrast media to cause enhancement of inflamed tissue, which decreased the sensitivity of the MRI. In addition, placement of the foreign body in the manus pads prevented the use of ultrasound and the cadaver limbs did not allow for imbibition of the wood fragments or surrounding soft tissue changes in the limb.¹⁴ These limitations caused the study to mimic an acute injury situation, which differed from the present case.

In a case series of human patients, CT proved useful for the identification of wooden foreign bodies, but the surrounding inflammatory response was poorly differentiated because of the similarity of the attenuation of the surrounding fat and skeletal muscle. This problematic situation was alleviated by viewing the scanned areas with a wider window, such as a bone setting.¹⁵ The author of case series of human patients stated that as the window width was increased, the foreign bodies were more easily identified and better differentiated from a gas or fluid build up due to the wood's absorptive characteristics and the duration of the injury.¹⁵

In the case described in this report, the primary concern was the extent of the soft tissue component of the disease; therefore, MRI was considered the best cross-sectional imaging option. MRI has been used either alone or concurrently with ultrasound in detecting wooden foreign material in both animals and people.^1,3,7,8,15 MRI has high accuracy in determination of the size, shape, and location of wooden foreign bodies. In general, wooden foreign material is reported to be hypointense relative to muscle on both T1 and T2 weighted images.^3,14,15 The surrounding inflammatory response is usually hypointense on T1-weighted images and hyperintense on T2-weighted images. Gadolinium enhancement of the inflamed tissues is usually present with MRI.¹⁵ The present case varied slightly in that the foreign body was isointense to hyperintense on precontrast T1-weighted images with the surrounding inflammatory reaction also being hyperintense to muscle.

There are limitations of MRI in the diagnosis of wooden foreign bodies. False negatives can occur if the object is small or if a hypointense structure or signal void is mistaken for a tendon, dense collagenous structure, or air.¹⁵ Compared with both ultrasound and CT, MRI is the most limited in spatial resolution.^14,15

In this case, multimodal imaging prior to surgery helped define the relationship of the foreign material to adjacent vital anatomic structures such as vascular and neurologic tissues. An accurate and rapid surgical approach was then performed, minimizing tissue trauma and decreasing surgical time.

Conclusion

When evaluating patients with suspected wooden foreign bodies, ultrasound, CT, MRI or a combination of these modalities should be considered. This case report compares and describes the usefulness of combining the ultrasound and MRI modalities in the detection of wooden foreign material secondary to an oropharyngeal penetration injury. Both ultrasound and MRI proved sufficient for determining the size, shape, and location of the foreign material. MRI was more useful than ultrasound for determining the full extent of the soft tissue inflammation and was able to definitively rule out spinal canal involvement. The use of multiple imaging modalities led to a direct and successful surgery with an uneventful and complete recovery.