Ventral occipito-atlanto-axial fluid-filled lesion causing dynamic spinal cord compression in a cat

Abstract

Cystic lesions affecting the vertebral canal or spinal cord have rarely been reported in cats. A 3-year-old female neutered domestic longhair cat presented for evaluation of a 2-year-history of episodes of ataxia and paresis affecting all limbs. Neurological examination was consistent with a lesion in the C1–C5 spinal cord segments. Magnetic resonance imaging (MRI) showed a fluid-filled lesion at the occipito-atlanto-axial region causing dynamic spinal cord compression on flexion of the neck. The imaging characteristics were compatible with a juxta-articular cyst. To our knowledge, this is the first report of a fluid-filled lesion causing dynamic cervical spinal cord compression in a cat and highlights the importance of performing flexion–extension MRI views in diagnosing cases with dynamic spinal cord compression.

Case Report

A 3-year-old female neutered domestic longhair cat was referred to the Neurology and Neurosurgery Service of the University of Glasgow Small Animal Hospital for evaluation of a 2-year-history of episodic gait abnormalities of variable frequency, severity and duration. The owner reported that the cat had no contact with other cats and was kept strictly indoors. The clinical signs were waxing and waning during the last 2 years ranging from being completely normal to having moderate ataxia in all limbs. Intermittent treatment with meloxicam (0.1 mg/kg once daily) (Metacam; Boehringer-Ingelheim) was given when the clinical signs were considered worse. The cat showed an acute deterioration of the clinical signs and became non-ambulatory tetraparetic 2 weeks prior to presentation followed by progressive improvement regaining ambulation over a few days.

On presentation, physical examination was normal. Neurological examination showed normal mentation and cranial nerve function. The cat was ambulatory tetraparetic with moderate spinal ataxia in all limbs. Proprioceptive deficits were observed in all limbs. Segmental spinal reflexes were normal. There was no evidence of pain on palpation of the vertebral column or on passive flexion of the head and neck. Based on these findings, the neuro-anatomical localisation was the C1–C5 spinal cord segments. Differential diagnoses included atlanto-axial subluxation, other congenital vertebral malformation causing spinal cord compression, arachnoid diverticulum and, less likely, intervertebral disc herniation and an inflammatory, infectious or neoplastic process.

A complete blood count and serum biochemistry profile were unremarkable. Previous radiographs of the limbs and spine did not show any abnormality. Magnetic resonance imaging (MRI) of the cervical spine was performed using a 1.5-Tesla MRI unit (Siemens Magnetom Essenza; Siemens). A large (1 cm × 0.8 cm), well circumscribed lesion was seen between the occipital condyles with some wrapping around the lateral aspects of the condyles and C1 vertebral body extending ventrally at this level (Figure 1a) and caudally to the C1–C2 articulation (Figure 1b). The signal characteristics corresponded to fluid in all sequences performed: hyperintense on T2-weighted images (WI) (Figure 1a), hypointense on T1WI (Figure 1c) and suppressed in fluid attenuation inversion recovery. After intravenous administration of gadopentetate dimeglumine (94 mg/kg, Magnevist; Bayer HealthCare Pharmaceuticals) there was a discrete peripheral contrast enhancement (Figure 1d). These imaging characteristics were compatible with a fluid-filled lesion and a juxta-articular cyst was considered most likely. No obvious spinal cord compression was identified on neutral views. However, when the neck was flexed, the fluid-filled lesion was compressed and pushed dorsally between the occipital condyles and C1 vertebral body resulting in dorsal ‘extrusion’ of part of the lesion into the ventral aspect of the cervical vertebral canal causing moderate-to-marked compression of the spinal cord (Figure 2). The flexed T2WI sagittal views showed a connection between the more cranial part of the fluid-filled lesion and the C1–C2 articulation. The distance between the dorsal lamina of C1 and the dorsal spinous process of C2 was considered normal, suggesting no C1–C2 instability, and the occipital condyles appeared normal.

Figure 1

Occipito-atlanto-axial fluid-filled lesion. Sagittal (a) and parasagittal (b) T2-weighted images (WI), and T1WI pre- (c) and post-contrast (d) sagittal views of the cervical spine of a 3-year-old domestic longhair cat. The lesion (white arrow) is hyperintense on T2WI (a), hypointense in T1WI (c) and shows mild peripheral enhancement after contrast administration (d). The parasagittal view (b) shows the extension of the lesion around the occipital condyles cranially and the C1–C2 articulation caudally (red arrows)

Figure 2

Neutral (a, b) and flexed (c,d) sagittal (a,c) and transverse (b,d) T2-weighted images of the cranial cervical spine of a 3-year-old domestic longhair cat. The fluid-filled lesion is located ventrally to the spinal cord (white arrows) in the neutral views and not causing spinal cord compression, but is displaced dorsally in the flexed views causing marked spinal cord compression (d, red arrow)

The cat deteriorated after the investigations and showed more severe spinal ataxia affecting all limbs likely caused by the temporary spinal cord compression during flexion views. The owner declined surgical treatment and a tapering anti-inflammatory dose of prednisolone (1 mg/kg a day for 1 week and then 0.5 mg/kg for 1 week; Prednidale 5, Dechra) was prescribed for 2 weeks. One month later the cat was back to normal. On telephonic update 7 months later, it was reported that the cat continued doing well on restricted exercise, with just a couple of short episodes of mild ataxia affecting all limbs observed.

Cystic lesions affecting the spinal cord or the vertebral canal have rarely been reported in cats. Previous reports include subarachnoid diverticulum^1–5 and an intradural epithelial cyst ⁶ affecting the cervical and thoracic spinal cord. To our knowledge this is the first report of a fluid-filled lesion causing dynamic cervical spinal cord compression in a cat. The owner declined surgical treatment and thus it was not possible to characterise the histological appearance of this fluid-filled lesion and make a final diagnosis. The MRI characteristics and its possible connection with the joint make a juxta-articular cyst (synovial or ganglion cyst) most likely. The MRI appearance of the lesion reported here resembles a juxta-articular cyst, as reported in dogs^7–11 and humans.^12–14 These cysts usually have signal intensity equal to or slightly greater than cerebrospinal fluid in T1WI and T2WI, and can show peripheral contrast enhancement, as observed in this case.

Juxta-articular cysts at the C1–C2 junction are extremely rare in humans and have been reported in just one dog. ¹¹ The pathogenesis is unclear. The main theory is that they occur secondary to degenerative changes of the joint or to micro-trauma.^12,14 In humans and the previously reported dog, synovial cysts at the C1–C2 junction have been associated with atlanto-axial subluxation or instability.^11,14 However, the cat presented herein had no evidence of atlanto-axial subluxation or instability that could have caused the formation of the cyst. Although a history of trauma was not reported, we cannot exclude microtrauma as the cause of cyst formation in this cat.

Previous studies of the anatomy of the occipito-atlanto-axial joint cavity in cats ¹⁵ and dogs ¹⁶ found that, unlike humans, a common joint cavity exists composed by five synovial communicating joints. This is consistent with the MRI findings in the presented case and explains the extension of the cystic structure between the occipital condyles to the cranial aspect of C2 vertebral body (Figure 1b), with the cyst originating from this composite joint.

Flexion, extension or traction views of the cervical spine are commonly performed in dogs ¹⁷ and humans ¹⁸ with cervical spondylomyelopathy and are useful in assessing dynamic lesions. The presented case shows the importance of performing flexion–extension MRI views in obtaining a diagnosis when a dynamic spinal cord compression is suspected. The neutral views in this case showed no evidence of spinal cord compression, but there was marked spinal cord compression in the flexion views. Unfortunately, there is a risk of deterioration owing to spinal cord compression when performing flexed views; temporary clinical deterioration occurred in this cat after the MRI, supporting a dynamic disease process as the cause of the presenting complaint. This could also explain the episodic nature of the clinical signs shown by this cat owing to the episodic spinal cord compression being present only in specific cervical positions.

In humans, different surgical and non-surgical treatments have been reported for spinal synovial cysts, but spontaneous resolution is also described.^19,20 Reported surgical treatments include excision and decompression of the cyst and, in some cases, vertebral fusion. In the previously reported dog ¹¹ the cyst was fenestrated and ventral fixation of the atlanto-axial joint was achieved using a locking plate. A similar surgery could have been an option in the present case. No recurrence after surgery has been reported in humans or in the previously reported dog.^11–14 Medical treatments are based on rest, physiotherapy, anti-inflammatory medications and, in some cases, corticosteroid injections into the cysts. ²⁰ The owner of the presented cat elected conservative management owing to the episodic nature of the ataxia and financial constraints. A repeat MRI scan was offered to monitor disease progression, but was also declined by the owner.

The cause of the improvement in the presented cat remains obscure. It is possible that the implemented exercise restriction resulted in a reduced occurrence of dynamic spinal cord compression without change in the size of the cyst. Although the corticosteroid therapy was only of a 2-week duration, this may have reduced spinal cord oedema secondary to trauma leading to improvement. Lastly, spontaneous resolution, cyst rupture or decrease in cyst size cannot be totally excluded based on reports in human literature. ¹⁹

Conclusions

This case report describes the clinical and imaging finding in a cat with a fluid-filled lesion affecting the occipito-atlanto-axial joint and causing dynamic spinal cord compression. It shows the importance of performing flexion–extension MRI views when a dynamic lesion is suspected.