Share it:

section content


A 3 year old castrated retired racing greyhound presented with acute paraplegia. The patient had been travelling in a car and developed acute paraplegia on arrival after a 4-5 hour drive. The patient was seemingly clinically normal when the journey began, but did atypically vocalize 1 hour into the journey. There was no known trauma and the journey was otherwise uneventful. The patient did not appear to have a traumatic event in the car.

Clinical and neurological examination findings: Paraplegic, with intact hind limb reflexes. No voluntary tail movement. No spinal pain was identified. The exam localized the lesion to T3-L3. No other clinical or neurologic abnormalities were identified. The patient was imaged approximately 15 hours after the paraplegia was detected by the owner.

T2w sagittal, thoracic spine

HASTE, thoracic spine

T2w sagittal, lumbar spine

STIR sagittal, lumbar spine

HASTE, lumbar spine

T2w transverse, lumbar spine

T1w sagittal pre and post contrast

Most evident in the T2-w transverse sequence, there is a thin T1 and T2 hypointense curvilinear structure within the dorsal aspect of the neural canal centered over the mid-aspect of the L1 vertebral body. The structure drapes over the dorsal half of the spinal cord. A thin margined T2 hyperintensity that nulls in STIR sequence between the aforementioned T2 hypointensity and the spinal cord presumably reflects a small. amount of epidural fat. Dorsal to the aforementioned curvilinear band there is heterogeneous T2 signal with patchy regions of T2 hypointensity interspersed amongst regions that are T2 hyperintense. Ill-defined regions of T1 hyperintensity are present dorsally. Some attenuation of the dorsal aspect of the subarachnoid space is evident in the HASTE sequences and the spinal cord is mildly dorsally and dorsolaterally compressed bilaterally resulting in somewhat of a triangular shape in cross-section. There is a focal region of T2 hyperintensity within the dorsocentral aspect of the spinal cord at the mid aspect of L1. There is mild enhancement within the dorsal aspect of the neural canal dorsal to the aforementioned curvilinear T1 and T2 hypointense structure.

The cranial lumbar discs have normal T2 and STIR signal and there is no evidence of disc bulging or prolapse. The ventral epidural space appears normal.


In light of the rapid onset of clinical signs, the heterogeneous signal within the most dorsal aspect of the neural canal over L1 was thought most likely to reflect an acute hematoma seemingly contained within some sort of thin dorsally located band of tissue. The signal characteristics are consistent with acute hemorrhage (less than 3 days). The curvilinear T1 and T2 hypointense margin was thought likely to reflect a thin band of fibrous tissue.

The most likely differentials in this patient include atypical hematoma formation possibly secondary to a coagulopathy or a regional vascular anomaly. Hemorrhage and inflammation associated with a missile type disc herniation was considered a differential, but possibly less likely given the normal in-situ disc signal and lack of disruption to the ventral epidural space. An inflammatory lesion with focal empyema was thought unlikely based on the overall lack of contrast medium enhancement. The focal intramedullary T2 hyperintensity was thought to be edema (possibly associated with an ischemic vascular event) or gliosis.


A coagulation panel was normal. In the CSF collected via lumbar puncture was moderate to marked neutrophilic pleocytosis. A left T13-L2 hemilaminectomy was performed. A subperiosteal hematoma (confirmed histologically) was identified and removed. No other abnormalities were identified that might explain the presence of a hematoma in this location. At 4 weeks post-Sx the patient is mildly paraparetic and continues to improve.


  1. Theobald A, Dennis R, Beltran E. Imaging Diagnosis – Spontaneous subperiosteal vertebral hemorrhage in a Greyhound. Vet Radiol Ultrasound 2013; 55 (4): 420-423


Surgical photographs