Assessment Anatomic diagnosis

Etiologic diagnosis—Differential diagnoses included conditions resulting in extradural spinal cord compression such as neoplasia, abscess, synovial cyst, hematoma, or granuloma. Other differential diagnoses included EPM, spinal cord trauma with or without vertebral fracture, infarct attributable to fibrocartilaginous embolization, and verminous meningoencephalomyelitis. The diagnostic plan consisted of CSF analysis with cytologic examination of a sample to assess for verminous or protozoal inflammation or neoplasia. Paired serum and CSF samples were assessed for antibodies against Sarcocystis neurona with an ELISA^a to test for EPM. Potential diagnostic imaging included plain cervical radiography to assess for vertebral fractures and cervical contrast myelography to assess for dorsoventral spinal cord compression. Endoscopic examination of the upper respiratory tract could also be used to confirm the negative bilateral thoracolaryngeal reflex.

Diagnostic test findings—None of the aforementioned diagnostic testing was performed, and the horse was euthanized (by means of IV administration of pentobarbital sodium) because of the chronicity and severity of clinical signs coupled with a lack of response to multiple treatments for EPM. Necropsy was performed with the consent of the owner and revealed an extradural hematoma extending along the ventral portion of the spinal cord, corresponding with the second to the eighth thoracic spinal cord segments. There was locally extensive hemorrhage in the epaxial muscles dorsal to the T2 vertebra. Histologic examination of the associated spinal cord segments was not performed.

Comments

To the authors' knowledge, there are limited reports of extradural spinal hematomas in horses, and all have been associated with the mid to caudal cervical segments (C4-C6) and brachial intumescence.^1,2 The pathogenesis of cervical extradural spinal hematomas in horses remains poorly understood. It is hypothesized that hematoma development is induced by stress on vasculature associated with the high range of motion of the vertebral column within this region.^1,2 Three reported equine cases of extradural hematomas were thought to have been associated with damage to the ventral internal venous plexus, spinal branch of the vertebral artery, or intervertebral vein.¹

Extradural spinal hematomas in humans due to trauma (accidental or iatrogenic), abnormal vasculature, epidural venous plexus defects, pachymeningitis, neoplasia (vertebral hemangioma or metastatic carcinoma), coagulopathies (inherited or acquired), and spontaneous (idiopathic) causes have been reported.³ An association of extradural spinal hematomas with prolonged Valsalva maneuvers (ie, sneezing, defecation, or coitus) has been recognized.^3,4 In humans, postulated risk factors include sickle cell anemia, anticoagulant treatments, dual antiplatelet therapy, elevated venous pressure (eg, during pregnancy), and recent trauma.^3,4 These lesions are most often associated with the caudal cervical portion of the spinal cord or brachial intumescence; however, lesions affecting the thoracic spinal cord segments also develop.^4,5 Compared with cervical extradural hematomas, those affecting thoracic segments are more likely to be associated with profound neurologic deficits.^5,6 Both conservative management and surgical decompressive techniques for the treatment of spinal hematomas have been described. In the presence of marked spinal cord compression, prompt surgical evacuation is recommended.^3,4

There is minimal information regarding treatment outcomes of cervical extradural spinal hematoma in horses. In 2 reports,^1,2 4 of 5 affected horses were euthanized within 21 days after presumptive diagnosis because of their guarded prognosis and chronicity of neurologic signs. Reported treatments include administration of NSAIDs (flunixin meglumine or phenylbutazone) or corticosteroids (dexamethasone) as well as mannitol and dimethyl sulfoxide.

In 1 equine case¹ of presumptive cervical extradural spinal hematoma, initial neurologic assessment revealed grade 3/5 ataxia and tetraparesis, and myelography confirmed spinal cord compression (at the level of C5 through T1). Treatments included administration of dimethyl sulfoxide, mannitol, and flunixin meglumine. Reevaluation after 9 months revealed mild-to-moderate atrophy of muscle groups associated with the neck and forelimbs (rhomboid cervicis, splenius, and subclavius muscles) and minimal gait improvement. However, myelography revealed resolution of spinal cord compression. That horse's gait improved over a period of several years, and evaluation 5 years after onset of clinical signs revealed only grade 1 to 2/5 ataxia. The fact that clinical improvement did not correlate with myelographic findings suggests that spinal cord compression may not be the sole cause of neurologic dysfunction in horses with cervical extradural spinal hematoma. The more favorable outcome in that specific case (compared with other published reports) may have been associated with the lesser neurologic grade at time of diagnosis.

To the authors' knowledge, an exclusively thoracic extradural spinal hematoma in horses has not been previously described. In the horse of the present report, the lesion extended along the ventral aspect of the spinal cord from T2 to T8, which correlated well with the neurologic findings and resultant neuroanatomic localization. There was no history or evidence of trauma associated with the onset of neurologic signs and no evidence of trauma to the thoracic vertebrae; thus, the hemorrhage observed grossly in the dorsal epaxial muscles was unlikely to have been related to an inciting cause. Histologic examination of the spinal cord segments or extradural hematoma was not performed. Nonetheless, given the duration, stability, and severity of the horse's neurologic signs, return to normal function seemed unlikely regardless of treatment.^1,2 The case described in the present report has highlighted the use of clinical neurologic assessment to accurately localize a lesion in a situation where antemortem contrast myelography or cross-sectional imaging were unavailable.