Acute motor and sensory polyganglioradiculoneuritis in a cat: clinical and histopathological findings

Abstract

Polyneuropathies can have a variety of clinical presentations and tend to be rare in cats. In this report we describe a 6-year-old domestic shorthair cat with an acute and rapidly progressive onset of lower motor neuron and sensory signs affecting the spinal and cranial nerves. Histopathological examination revealed moderate-to-severe multifocal inflammatory infiltrates at the ventral and dorsal nerve roots, and dorsal spinal ganglia at the level of the L4 and cauda equina. The type and severity of inflammation varied between nerve roots, being composed of mainly neutrophils in some and mainly lymphocytes and macrophages in others. Immunohistochemistry showed a combination of neutrophils, macrophages and lymphocytes infiltrating the nerve roots and ganglia. The majority of the lymphocytes were T lymphocytes; only a few B lymphocytes were seen. Neurons within the affected ganglia showed central chromatolysis and necrosis. Wallerian-like degeneration and demyelination were observed in the nerve roots. A sensory and motor polyganglioradiculoneuritis was diagnosed. An autoimmune process similar to the acute motor and sensory neuropathy subtype of Guillain–Barré syndrome in humans or an infection by an unidentified agent were considered most likely.

Case Report

A 6-year-old, 4 kg, male neutered domestic shorthair cat was referred to the Neurology and Neurosurgery Service of the University of Glasgow Small Animal Hospital for evaluation of an acute onset of pelvic limb weakness that progressed to non-ambulatory tetraparesis over 3 days. The cat suddenly developed ambulatory paraparesis and ataxia progressing to non-ambulatory paraparesis over 24 h and over the next 2 days to non-ambulatory tetraparesis and a change in facial expression. He was an active outdoor cat without a history of previous medical problems or recent vaccination, yet intoxication could not be excluded.

On presentation, physical examination was normal apart from tachypnoea and an abdominal breathing pattern. Neurological examination showed normal mentation. On examination of the cranial nerves the palpebral and corneal reflexes were bilaterally absent. Pupils were dilated with reduced pupillary light reflexes; fundic examination and vision were normal. Facial and nasal sensation was decreased, and the jaw was dropped with reduced muscle tone. The cat was non-ambulatory tetraparetic. Proprioceptive deficits were observed in all limbs and segmental spinal reflexes were absent in all limbs. The muscle tone was decreased in all limbs, and superficial pain perception was absent, but deep pain perception was present. Anal tone, defaecation and urination were normal. There was no evidence of pain on palpation of the vertebral column or on passive flexion of the head and neck. The neuro-anatomical localisation was considered to be the peripheral nervous system with motor and sensory involvement, but a multifocal localisation affecting the central nervous system could not be excluded. Differential diagnoses included an infectious process (feline infectious peritonitis, toxoplasmosis, fungal), an immune-mediated process (polyradiculoneuritis and fulminant myasthenia gravis for the motor involvement), a neoplastic process (lymphoma), metabolic (hypokalemia for the motor involvement) or toxic (tick paralysis, ionophores intoxication and botulism toxin all for the motor involvement) disorders.

A complete blood count showed mild lymphopaenia (1.302 × 10⁹/l, reference interval [RI] 1.5–7.0 × 10⁹/l). Serum biochemistry profile showed moderate hyperglobulinaemia (50 g/l, RI 27–45 g/l). Serological tests for feline leukaemia virus, feline immunodeficiency virus, feline coronavirus and Toxoplasma gondii were negative. Thoracic radiographs and abdominal ultrasound were normal.

The cat was initially treated with intravenous fluid therapy (lactated Ringer’s solution, at 4 ml/kg/h) and supplemented with thiamine (a single injection of Vitamin B1; Bimeda), as the cat was anorexic. However, the disease continued to progress quickly and it developed ventilatory problems within 24 h; the owner elected euthanasia.

Necropsy examination was performed the same day. Gross examination was unremarkable. Representative samples of body tissues (kidneys, liver, spleen, lungs and heart), the central nervous system (brain, cerebellum and spinal cord) and peripheral nerves (sciatic nerve) were taken and fixed in 10% buffered formalin. The tissue samples were embedded in paraffin, sectioned at 5 µm and stained with haematoxylin and eosin (H&E). Sections of the spinal cord at the level of L4 vertebra and cauda equina were stained with Luxol fast blue (LFB) and immunohistochemical investigation was performed on the formalin-fixed tissue sections using primary antibodies specific for CD3 (Dako), PAX-5 (Dako), feline CD18 (UC Davis) and MAC387 (Dako). Heat-induced epitope retrieval was used for CD3 and PAX5 and enzymatic antigen retrieval using Proteinase K for CD18 (10 mins) and MAC387 (5 mins). Primary mouse antibody for CD18 (1/20), PAX5 (1/100), MAC387 (1/500) and rabbit antibody for CD3 (1/100) were visualised using the Dako EnVision system.

Histopathological examination of the brain, cerebellum, brainstem, intracranial part of multiple cranial nerves (oculomotor, trigeminal, abducens and facial), one trigeminal ganglion, both sciatic nerves and systemic organs was unremarkable. The spinal cord was examined at three representative levels (at the level of C4, T11 and L4 vertebrae), as well as the cauda equina. Cervical and thoracic sections (at the level of C4 and T11 vertebrae) had no changes within the ganglia or nerve roots. The grey matter was mildly hypercellular with occasional neuronal body satellitosis by mononuclear cells. The changes within the white matter were multifocal with no clear ventral or dorsal distribution and characterised by occasional, multifocal dilation of the myelin sheaths with the occasional Gitter cell. H&E examination of the spinal cord at the level of L4 vertebra and cauda equina revealed moderate-to-marked inflammatory cell infiltrates bilaterally within the dorsal and ventral nerve roots, ganglia, and also sparsely within the adjacent pia mater space and subarachnoid space. The inflammatory cell infiltrates were composed of high numbers of neutrophils admixed with lymphocytes, plasma cells and occasional Gitter cells. Nerve roots and ganglia in the cauda equina were more severely affected than those at the level of L4 vertebra. Ganglion neuronal cells were often shrunken (necrosis) and surrounded by numerous macrophages and lymphocytes (satellitosis), chromatolytic neurons were also seen (Figure 1). There was multifocal evidence of neuronophagia with loss of the neuron and replacement by an aggregate of inflammatory cells. Satellitosis was observed only occasionally in the grey matter. Scattered axons within peripheral nerves and occasionally within the white matter showed Wallerian-like degeneration with frequent infiltration of Gitter cells. LFB stain showed multifocal variation in the degree of staining pattern among axons within the nerve roots. There was variation in staining of the myelin sheaths suggestive of a degenerative process, and digestion chambers frequently contained accumulations of positively stained granular material (myelin fragments). Intact axons were occasionally identified within affected areas.

Figure 1

Histology of a sensory ganglion at the level of the cauda equina. Ganglion neural cells showed chromatolysis are multifocally swollen, necrotic (arrows) and with increased perineuronal cellularity or satellitosis (arrowhead). H&E, x 200

Immunohistochemical characterisation of the inflammatory cell infiltrate, both intraneural and perineural, revealed a predominance of CD18-positive cells within ganglia and nerve roots. The predominant CD18-positive cell type were neutrophils. Numerous sparse MAC387-positive cells were seen predominantly in nerve roots and within myelin sheaths. CD3+ lymphocytes (T cells) and few PAX-5 positive (B cells) were sparse throughout the ganglia, ventral and dorsal nerve roots and meninges (Figure 2). A sensory and motor polyganglioradiculoneuritis of L4 and cauda equina was diagnosed.

Figure 2

Histology and immunohistochemistry at the level of the L4 nerve root. (a) Histology shows moderate-to-marked inflammatory infiltrate multifocally distributed between axons (arrowheads). Multiple myelin sheaths are dilated and filled with shrunken, irregular eosinophilic material (myelin fragments). Haematoxylin and eosin (H&E), × 400. (b) CD18 expression in ventral nerve root is positive for >70% of the inflammatory infiltrate, most of them multilobulated cells (neutrophils). Feline CD18, × 400. (c) MAC387-positive cells infiltrating the nerve roots and within surrounding tissues. Feline MAC387, × 400

Polyneuropathies are particularly rare in cats and they can affect the motor, sensory or autonomic peripheral nerves. ¹ They can be inflammatory, as seen in infectious or autoimmune conditions, or non-inflammatory, as seen in most metabolic, degenerative, toxic and nutritional disorders. Inflammatory polyneuropathies have been reported in a small number of cats with acute or chronic presentations, and report involvement of different regions of the peripheral nervous system. To our knowledge this is the first report of an acute sensory and motor polyganglioradiculoneurits in a cat.

Acute idiopathic polyneuropathy has been previously reported in cats, causing acute onset of tetraparesis and loss of spinal reflexes due to mononuclear inflammation of the ventral nerve roots, but no sensory signs or significant histological changes at the level of the dorsal nerve root or ganglion was reported. ² Acute polyneuritis affecting the peripheral nerves has been reported in a couple of cats, but none had sensory involvement and only the peripheral nerves were examined.^3,4 An acute predominantly sensory polyganglioradiculoneuritis was diagnosed in a cat with normal segmental spinal reflexes, unlike the present case, and mononuclear inflammation was observed mainly at the level of the dorsal nerve root ganglia with mild changes in the ventral nerve roots without axonal or neuronal damage. ⁵ Chronic relapsing polyradiculoneuritis with motor and sensory involvement has been reported in a cat with mononuclear infiltration of the cranial nerves, ganglia, peripheral nerves, ventral and dorsal nerve roots, but again no axonal or neuronal damage was described as seen in this case. ⁶ A chronic recurrent demyelination and remyelination polyneuropathy with mononuclear infiltration of the intramuscular nerve branches has been reported in Bengal cats causing tetraparesis, weakness and in a few cases reduced superficial sensation, but no histopathological examination of the nerve roots or central nervous system has been reported. ⁷

In humans, Guillain–Barré syndrome (GBS) is an acute inflammatory polyneuropathy that has different subtypes. The most common subtype is the acute inflammatory demyelinating polyradiculoneuropathy, but other axonal subtypes with motor (acute motor axonal neuropathy) and motor and sensory (acute motor and sensory axonal neuropathy) involvement have been reported.^8,9 The acute motor and sensory axonal neuropathy subtype share some similarities with the present case. The axonal forms of GBS usually exhibit a more rapid and severe course, with frequent respiratory and cranial nerve involvement, as seen in the present case. ⁸ Electrophysiological studies are frequently used to diagnose and differentiate the subtypes of GBS.^8,9 No electrophysiological tests were performed as the cat deteriorated quickly and euthanasia was elected instead. Another important finding in humans with GBS is the presence of specific antigangliosides antibodies in the different subtypes.^8,9 A recent study in dogs demonstrated the presence of anti-GM2 ganglioside antibodies in acute canine polyradiculoneuritis. ¹⁰ If this is similar in cats is unknown and anti-GM2 ganglioside antibodies were not assessed in this case.

The predominance of T lymphocytes and macrophages seen in our case correspond to the pathological findings in humans with GBS. It is postulated that activated macrophages are targeted to antigens on the surface of the Schwann cell or axon, depending on the subtype, by activated T lymphocytes.^8,9 Despite the fact that the serology results were negative and no organisms were observed in histopathological examination, the increased number of neutrophils seen in the present case makes an unidentified infectious process a possibility.

The lack of correlation between the neurological signs and the distribution of the inflammatory infiltrates found in this case is unusual. The inflammatory infiltrates were seen mainly in the lumbar nerve roots with no significant changes in the cervical and thoracic nerve roots or the cranial nerve roots. This corresponds to the progression of the clinical signs that started in the pelvic limbs and then progressed to the thoracic limbs and cranial nerves. Using necropsy material obtained early in the course of the disease in humans with GBS, it was found that the initial damage is caused by the binding of the antibodies and complement activation preceding inflammatory cell invasion. ¹¹ Moreover, a similar distribution to our cat was reported in two dogs diagnosed with polyradiculoneuritis with more severe inflammatory changes affecting the cauda equina. ¹² One of the limitations of the present case report is that no histological examination of the nerve roots in the cervico-thoracic intumescence (C6–T2) or the extracranial portion of the cranial nerves was performed, so we can not rule out that inflammatory infiltrates could have been present at those levels.

Conclusions

To our knowledge this is the first report of an acute motor and sensory polyganglioradiculoneuritis in a cat. An autoimmune or an unidentified infectious process was considered most likely as no cause was identified.