Probable lumbar acute non-compressive nucleus pulposus extrusion in a cat with acute onset paraparesis

Abstract

A spinal cord lesion localised caudal to the L6 spinal segment was diagnosed in a 2-year-old female spayed domestic longhair cat with acute onset paraparesis. Magnetic resonance imaging findings were consistent with an acute, non-compressive nucleus pulposus extrusion of the L5–L6 intervertebral disc. The cat was successfully managed with supportive care, including cage confinement.

Case Report

A 2-year-old, female, spayed domestic longhair cat was referred to the Valentine Charlton Cat Centre for acute onset pelvic limb paraparesis. The cat had no outdoor access and no history of trauma. Earlier that morning it had been reluctant to jump onto a feeding platform and was inappetent. The cat was in good general health with a normal gait before this episode. A littermate had been diagnosed with an aortic thromboembolism approximately 1 year earlier and had been euthanased.

On presentation, aortic thromboembolism was quickly ruled out by palpation of femoral pulses bilaterally and audible detection of distal pelvic limb pulses using a Doppler probe. The cat was obese (body weight 7.3 kg, condition score 5/5) and had an elevated rectal temperature (39.7°C). No abnormalities were detected on palpation of soft tissue, bony or joint structures of the pelvic limbs. The cat consistently became tense and vocalised when the lumbosacral spinal region was palpated. Neurological examination revealed mentation and cranial nerve function to be normal. A crouched pelvic limb gait and flaccid tail were present. Gait abnormalities were restricted to the pelvic limbs, including a plantigrade stance and mild ataxia. Proprioception paw positioning and hopping was normal in the thoracic limbs, but delayed in both pelvic limbs, as evidenced by knuckling, and was slightly worse on the right side (1+). Spinal reflexes were normal in the thoracic limbs. Flexor withdrawal reflexes were reduced (1+), while the patella reflexes were exaggerated in both pelvic limbs (3+). Anal tone and perineal sensation were reduced. Cutaneous nociception of the tail was present, as evidenced by vocalisation when the skin was pinched, but the tail lacked voluntary motor function. The neuroanatomical location of the lesion was assessed as caudal to the L6 spinal cord segment, with sciatic nerve involvement but intact femoral nerve function resulting in pseudohyperreflexia of the patellar reflex.

The presence of pyrexia, although potentially unrelated, raised suspicion for the presence of inflammatory central nervous system (CNS) disease (including, but not limited to, bacterial meningomyelitis, protozoal myelitis, fungal myelitis, discospondylitis, spinal empyema or feline infectious peritonitis) or lymphoma, given the signalment and unknown feline leukaemia virus (FeLV) status. Traumatic intervertebral disc herniation and fibrocartilagenous embolism (FCE) were also considered. Other differential diagnoses, including congenital lesions and other neoplastic aetiologies (primary CNS or metastatic neoplasia), could not be excluded. Haematology was unremarkable and a serum biochemistry profile revealed elevated creatine kinase [568, reference interval (RI) <200 U/l], hyperproteinaemia (81.2, RI 54–73 g/l), hypercholesterolaemia (4.32, RI 1.90–3.90 mmol/l), hyperbilirubinaemia (5.3, RI 2.5–3.5 umol/l), elevated creatinine (195, RI 90–180 umol/l) and low urea (6.11, RI 7.2–10.7 mmol/l). Serological tests for FeLV antigen, feline immunodeficiency virus antibody and Dirofilaria immitis antigen (SNAP Feline Triple Test; Idexx Laboratories) were negative. No abnormalities were identified on thoracic radiography or abdominal ultrasonography. On plain spinal radiographs, a narrowing of the intervertebral disc space between the L5 and L6 vertebral bodies was present on orthogonal views (Figure 1).

Figure 1

Lateral (a) and ventrodorsal (b) spinal radiographs. There is narrowing of the L5–L6 intervertebral disc space

Magnetic resonance imaging (MRI) of the lumbosacral spine was performed on day 2 of hospitalisation (Vet-MR Grande 0.4T low-field MRI). On T2-weighted fast spin echo (FSE) and short TI inversion recovery (STIR) images there was loss of signal intensity of the L5–L6 intervertebral disc and narrowing of the corresponding disc space (Figure 2). The remaining discs appeared normal. A focal hyperintensity was present in the spinal cord directly above the L5–L6 disc space on T2-weighted FSE and STIR images. The same region was isointense to the remainder of the spinal cord on T1-weighted images. There was no evidence of spinal cord compression. A presumptive diagnosis of non-compressive, intramedullary, focal myelopathy, consistent with an acute, non-compressive nucleus pulposus extrusion (ANNPE) was made. Cerebrospinal fluid analysis, which may have helped to rule out inflammatory disease, was declined by the owner.

Figure 2

Sagittal T2-weighted STIR (a) There is loss of signal intensity of the L5–L6 intervertebral disc, narrowing of L5–L6 disc space and adjacent discs appear normal. Transverse T2-weighted FSE (b). A focal transverse hyperintensity is present in the spinal cord directly above the L5–L6 disc space

The cat was managed with strict cage rest to allow adequate time for nucleus pulposus material to dissipate and for cord oedema to resolve. After 12 h of hospitalisation the urinary bladder became distended and flaccid, consistent with a ‘lower motor neuron’ bladder. The bladder was decompressed by manual expression twice daily. Intravenous ticarcillin clavulanate (40 mg/kg q8h) was commenced on the day of presentation because of the pyrexia and concern about potential bacterial infection. This was discontinued on day 3, as the pyrexia had resolved. After 5 days, the cat was urinating voluntarily and the pelvic limb gait was improved. The cat was discharged from hospital with guidelines for restriction of activity and caloric intake to manage obesity. At recheck 2 weeks after presentation only mild ataxia remained. Pelvic limb flexor withdrawal reflexes were improved, but mild, pelvic limb proprioceptive deficits and tail flaccidity remained. Four weeks after presentation, voluntary tail movement had returned, pelvic limb proprioception was normal and there was only mild residual ataxia. Ten months after presentation the cat’s body weight had decreased to 5.7 kg after caloric restriction and the only neurological deficits noted were a slightly plantigrade right pelvic limb stance and low tail carriage.

ANNPE is characterised by nucleus pulposus extrusion of a non-degenerate intervertebral disc. ¹ This is in contrast to Hansen type I disc extrusions which generally undergo chondroid degeneration.^2,3 Synonyms for ANNPE include: high-velocity, low-volume disc extrusions; traumatic discs; disc explosions; type III discs; missile discs; and hydraulic discs. ⁴ Little is known about ANNPE in cats. To our knowledge, there are only two cases of presumptive ANNPE described.^5,6 In contrast, presumptive ANNPE in dogs has been described in a number of case series from which information regarding aetiopathogenesis, clinical features and characteristic MRI findings may be extrapolated.^1,4,7,8 In ANNPE, the nucleus pulposus is forcibly extruded through a tear in the dorsal annulus fibrosus, presumably because of a traumatic aetiology. The nucleus pulposus causes spinal contusion and, although there can be very mild transient compression due to focal oedema and/or haemorrhage, there is minimal-to-absent ongoing compression.^1,4,7,8 The non-degenerate nucleus pulposus rapidly dissipates within the epidural space. However, cases have been described in dogs in which the nucleus pulposus lacerates the dura mater^9–11 or lodges within the intramedullary parenchyma.^12–14

Canine cases are generally associated with strenuous exercise or a traumatic injury.^7,8 In the case described here, no traumatic event or strenuous activity had been witnessed, but these possibilities cannot be completely excluded. Alternatively, obesity may have predisposed the cat to this type of injury. ¹⁹ In humans, non-exertional activities (coughing, sneezing, straining to defecate) can be associated with spontaneous intervertebral disc herniation. ²⁴ As the cat had not been closely observed prior to the onset of paraparesis, such an event may have gone unnoticed and precipitated the intervertebral disc extrusion.

ANNPE is clinically similar to ischaemic myelopathy due to FCE, ¹⁵ as both are characterised by peracute-to-acute onset, non-progressive myelopathy. While the exact aetiopathogenesis of FCE-associated ischaemic myelopathy remains unknown, FCE has been demonstrated to be histologically identical to nucleus pulposus material. Proposed theories for FCE as a cause for spinal ischaemia include vascular entry of intervertebral disc material via veins, arteries or arteriovenous anastomoses, vascular entry of intervertebral disc material via anomalous intra-disc vasculature (either because of remnant embryonic vessels or chronic inflammation) or entry of intervertebral disc material into vertebral venous sinusoidal channels. ¹⁵ If FCEs are, indeed, fragments of non-degenerate nucleus pulposus and trauma is a common aetiology, then FCE, intramedullary nucleus pulposus lesions and ANNPE may be manifestations of the same disease process.

Without histopathology, disc degeneration (suggestive of Hansen type I disc extrusion) could not be completely excluded. However, most clinically significant cases of IVDD in cats involving Hansen type I disc extrusions are secondary to chondroid degeneration, which is often evident radiographically. ²⁴ Additionally, the case presented fulfilled MRI criteria for diagnosis of ANNPE which include a narrowed disc space with a loss of volume and signal intensity of the affected disc on T2-weighted images, a focal hyperintense lesion dorsal to the affected disc space on T2-weighted images with the corresponding region, being iso- to hypo-intense on T1-weighted images and little-to-no spinal cord compression.^5–7 Intervertebral disc space narrowing is not a feature generally associated with FCE.^8,16 In addition, MRI images of ANNPE lesions show very focal, hyperintense intramedullary lesions involving both grey and white matter regions, whereas MRI findings in FCE suggest ischaemic myelopathy predominantly affects grey matter, has greater lesion length-to-vertebral length ratio, and is often lateralised to correspond with the vascular supply to the spinal cord and occlusion of a single blood vessel.^16,17

Our case differed in clinical presentation from the two previous reports of presumptive ANNPE in cats, which both had lesions in the cervical region (C1–C6 spinal segment) and markedly lateralised neurological deficits.^5,6 The most common site for ANNPE lesions in cats is unknown. Hansen type I disc extrusions are reported to occur mainly in the thoracolumbar or mid-to-caudal lumbar regions (eg L4–L5).^18–20 A report of six cats with degenerative IVDD showed 66% of lesions were in the thoracolumbar area, with 2/6 cats requiring a L4–L5 hemilaminectomy. ¹⁹ A series of 10 cats with degenerative IVDD had a predominance of lesions occurring at L4–L5. ¹⁸ Lumbosacral degenerative IVDD has been described in two case series.^21,22 The latter case series described an incidence of 0.05% of the total feline hospital population during the study period. ²² It is not known whether ANNPE follows similar patterns. Owing to the paucity of feline ANNPE cases described thus far, the clinical presentation remains to be established in this species.

Definitive diagnosis of ANNPE requires histopathological identification of nucleus pulposus material within the spinal canal either post mortem or during surgery. Other diagnostic criteria include loss of the nucleus pulposus in a corresponding intervertebral disc and visualisation of spinal contusions. ⁷ As the non-degenerate nucleus pulposus can dissipate rapidly, histopathological diagnosis is not always achievable, even with surgery. ⁶ Additionally, owing to lack of ongoing compression, decompressive surgery is not indicated, meaning that in the majority of cases histological specimens are unattainable except at necropsy.^4,5,8 Therefore, it is important to establish characteristic MRI changes, to allow a presumptive diagnosis of ANNPE, as well as to differentiate it from FCE. ⁴

The outcome and prognostic factors for ANNPE in cats is unknown. One previously documented case that received conservative management showed improvement, although residual ataxia remained at 6 months post-presentation. ⁵ Another describes improvement with residual ataxia 2 months following surgery. ⁶ In contrast, the case described here demonstrated full return to function within 8 weeks. This may have been a result of less severe initial spinal cord trauma.

A study of 42 cases of suspected ANNPE in dogs demonstrated a good outcome in 28 dogs and a poor outcome in 14 dogs. ⁷ Factors that were negatively linked to outcome included increased severity of neurological signs at presentation, increased cross-sectional area and length of the intramedullary lesion on T2 weighted MRI images, and the presence of hypointensity on T2-weighted gradient echo MRI images.

There are no incidence data for ANNPE in small animal populations, but it seems to be uncommon. It is likely that ANNPE is currently under-diagnosed because of limited access to advanced imaging modalities. ANNPE is probably less common in cats than dogs owing to the lower incidence of IVDD in cats compared to dogs.^3,18,23 Increased awareness of ANNPE as a differential for acute or peracute, non-progressive myelopathy in cats may encourage referral for MRI, where this is available, particularly as these cases can have a good outcome.