History
A 10-month-old 26.8-kg castrated male Golden Retriever was referred because of a 5-month history of progressive pelvic limb ataxia and paresis. Prior to referral, the dog was evaluated by the primary veterinarian and empirically treated with prednisone (20 mg, PO, q 12 h, on a tapering schedule), doxycycline (unknown dosage), and physical therapy. The dog’s pelvic limb ataxia improved with medical treatment but worsened as the prednisone dosage was tapered.
On referral examination, the dog was bright, alert, and responsive and had abrasions on the dorsal aspect of digits 2 and 3 of the pelvic limbs bilaterally. The nails of these digits were also very short. The dog was ambulatory but had moderate pelvic limb ataxia and paresis. No abnormalities were detected for the dog’s cranial nerves or spinal reflexes; however, the dog had delayed postural reactions in its pelvic limbs. The dog had no evidence of hyperesthesia, and results of a CBC and serum biochemical analyses were within reference limits. The referring veterinarian’s radiographic images of the dog’s thoracic region of the vertebral column (Figure 1) were reviewed.
Diagnostic Imaging Findings and Interpretation
Within the spinous processes of T5, T6, and T8, there were 3 smoothly marginated, ovoid, mineral opaque nodules with a maximum dimension approximately the size of the T6 vertebral body (Figure 2). These nodules had slightly increased opacity, compared with the surrounding spinous processes. The T7 and T5 spinous processes were concave along their caudal borders, conforming to the nodules along the cranial aspects of T6 and T8. These findings were not evident on the ventrodorsal projection. A nodule with a similar appearance and size was identified in the midbody of the left 12th rib. There was semiformed, heterogenous, soft tissue opaque material in the stomach, consistent with normal ingesta. Additionally, there was an increased soft tissue opacity superimposed over the right cranial, right middle, and right caudal lung lobes attributed to the effects of patient obliquity.
Consideration for the nodules within the spinous processes and the left 12th rib was given to multiple cartilaginous exostoses (MCE). Lesser consideration was given to fungal granuloma or aggressive primary or metastatic bone neoplasia due to the lack of periosteal proliferation, lack of cortical destruction, presence of smooth margins, and lack of osteolysis other than the pressure remodeling of the adjacent spinous processes.
The dog underwent MRI, which revealed that the nodules identified on radiography expanded the spinous process and, compared with the adjacent spinous processes, were isointense on T1-weighted, T2-weighted, and T1-weighted FLAIR sequences and peripherally hyperintense on T2-weighted STIR sequences (Figure 3). This hyperintensity along the outer margins of the nodules evident on T2-weighted STIR sequences was consistent with active remodeling. In addition, we identified a nodule that arose from the ventral aspect of the dorsal lamina of T8, extended into the vertebral canal, and caused ventral deviation of the subarachnoid and epidural spaces and compression of the spinal cord. This compressive nodule was not identified on radiography due to summation and superimposition of the nodule with the vertebral body and pedicles. The use of MRI allowed visualization of this nodule in multiple planes, evaluation of the soft tissue structures of the spinal cord, and assessment of the amount of compression by the nodule. This extradural spinal cord compression was presumed to have caused the abnormal clinical signs in this dog. Adjacent to the compressive lesion of T8, there was mild hyperintensity of the spinal cord evident on T2-weighted images. There was mild dilation of the central canal of the spinal cord cranial to T8. A smaller nodule with the same signal intensity characteristics was present along the right dorsal aspect of the vertebral canal at L7 that did not cause displacement of the cauda equina or attenuation of the lumbar cistern.
Treatment and Outcome
The dog underwent dorsal laminectomy of T7 through T9 to remove the abnormal spinous processes and decompress the spinal cord. Histologic examination of the nodules revealed benign proliferative nodules capped by a thick zone of hyaline cartilage that exhibited layers of endochondral ossification toward the center and that was outlined by a layer of thin fibrous stroma (Figure 4), consistent with osteochondromas or MCE. No malignant features were noticed throughout any of the sections examined.
The dog remained hospitalized for 11 days for supportive care and physical therapy. At recheck examination 5 weeks after surgery, the dog was strongly ambulatory with no obvious ataxia or paresis.
Comments
Multiple cartilaginous exostoses are characterized by osteochondromatous outgrowth in the metaphyseal region of bones that develop by osteochondral ossification, most commonly seen in the vertebrae, ribs, and long bones in dogs.1,2 Two main theories as to the origin of MCE pose that the lesions arise from the growth plate cartilage due to a defect in the perichondral ring or that the lesion arises from the osteogenic layer of the periosteum that fails to differentiate into osteoblasts.2,3 In cats, a viral cause has also been proposed.4
Considered to be a disease of immature dogs, MCE lesions arise during active bone growth2 and are thought to cease when the dog reaches skeletal maturity and the adjacent epiphyseal plates close. It is generally considered to be a benign proliferative disease of cartilage and bone that is often discovered incidentally and is not clinically significant. However, MCE can be the cause of clinical problems (eg, neurologic deficits, pain, and decreased function) when the proliferations compress the spinal cord or other nerves, blood vessels, tendons, or ligaments.2 In the dog of the present report, an MCE lesion on T8 caused severe dorsal compression of the spinal cord that led to ambulatory paraparesis and moderate proprioceptive ataxia of the pelvic limbs. With MRI, we were able to detect the compressive lesion, which was not evident on radiography.
Malignant transformation of MCE is possible in dogs.3 Although humans tend to have sarcomatous change most commonly in the ilium and proximal aspect of the femur, dogs appear to have chondrosarcomas arise more commonly in the lumbar vertebrae.3 Pulmonary metastasis may also occur in dogs. Malignant transformation gives MCE a bimodal age of clinical onset, with < 6 months of age typical for MCE and adulthood typical for malignant transformation.3
A presumptive diagnosis of MCE can be made on the basis of radiographic findings; however, histologic examination is required for definitive diagnosis.5 As evident in the dog of the present report, the bone proliferation seen with MCE does not exhibit aggressive bone disease characteristics, such as medullary lysis or irregular margins due to periosteal reaction, that are usually seen with malignant neoplasms or fungal infections of bone.6 Fracture callus or benign bone tumors, however, are other possible differential diagnoses for the radiographically benign lesions.5
It is important to note that the most radiographically apparent lesions may not be the cause of a patient’s clinical signs. In the dog of the present report, radiography readily revealed the benign proliferations of the spinous processes of T5 and T6 and the dorsal aspect of T8 but did not reveal the proliferative lesion of T8 that caused severe compression of the spinal cord. Thus, although radiography provided evidence for a presumptive diagnosis of MCE in the dog of the present report, MRI revealed the cause for the dog’s neurologic signs and was used to evaluate the extent of the lesions and to plan for surgical correction. Of note on the MRI was the peripheral hyperintense signal around each lesion evident of T2-weighted STIR sequences. Given that tissues rich in hyaline cartilage have hyperintense signals on T2-weighted MRI images, the areas of hyperintense signals around each of the nodules in the dog of the present report represented the cartilage cap of an active MCE lesion. For selection of advanced imaging, it is important to note that CT may not provide reliable detection of cartilage caps < 2.5 cm in thickness and that MRI may be better in detecting active MCE lesions.7
Treatment for MCE is generally only required if the lesions cause clinical signs. Surgical removal of the exostosis can be undertaken, and if removal is complete and the animal has reached skeletal maturity, regrowth is uncommon. However, the prognosis is considered guarded because there is a risk of regrowth and malignant transformation.5
References
- 1. ↑
Gee BR, Doige CE. Multiple cartilaginous exostoses in a litter of dogs. J Am Vet Med Assoc. 1970;156(1):53–59.
- 2. ↑
Gambardella PC, Osborne CA, Stevens JB. Multiple cartilaginous exostoses in the dog. J Am Vet Med Assoc. 1975;166(8):761–768.
- 5. ↑
Bailey CS, Morgan JP. Congenital spinal malformations. Vet Clin North Am Small Anim Pract. 1992;22(4):985–1015.
- 7. ↑
Silver GM, Bagley RS, Gavin PR, Kippenes H. Radiographic diagnosis: cartilaginous exostoses in a dog. Vet Radiol Ultrasound. 2001;42(3):231–234.