Article Text

Chronic inflammatory demyelinating polyneuropathy as a complication of cat scratch disease
  1. Department of Pediatric Neurology, Neuromuscular Centre Nijmegen, University Hospital Nijmegen, PO Box 9101, 6500 HB Nijmegen, The Netherlands
  2. Department of Neurology
  1. Dr A Verrips A.Verrips{at}
  1. Department of Pediatric Neurology, Neuromuscular Centre Nijmegen, University Hospital Nijmegen, PO Box 9101, 6500 HB Nijmegen, The Netherlands
  2. Department of Neurology
  1. Dr A Verrips A.Verrips{at}

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Cat scratch disease (CSD) was first described in 1950 as a benign regional lymphadenitis. This infection is caused byBartonella henselae. The clinical range of CSD has expanded beyond the classic presentation. In 5%–20% of the infected patients the disease may spread to other organs. However, neurological complications associated with CSD are rare, with encephalopathy being by far the most common form (90%) of nervous system involvement. Encephalopathy occurs in 2%–3% of patients and is more common in adults than in children with the onset varying from a few days to months after diagnosis of CSD.1 Other known neurological manifestations, often in combination with encephalopathy, are neuroretinitis, oculoglandular disease of Parinaud, myelopathy, radiculopathy or abducens nerve, and facial nerve paresis. We report on a 3 year old boy who developed chronic inflammatory demyelinating polyneuropathy (CIDP) 6 weeks after identification of CSD.

(A) Sural nerve. Low power electron micrograph. Demyelinated fibres (arrows) and thinly remyelinated fibres. Bar=2.0 μm. (B) Sural nerve; detail of myelinated fibre. Macrophage has invaded the myelinated fibre; process of macrophage is extending between axon and myelin sheath (arrow head). The damaged inner major dense line is terminating as a small dark swelling (arrow). Bar=0.2 μm.

A previously healthy 3 year old boy presented in a paediatric clinic with regional lymphadenitis of the submandibular and suboccipital glands. He was afebrile and did not have any other symptoms. There was no hepatosplenomegaly. No neurological signs were present on physical examination. As the boy might have been scratched by one of his pet cats, in accordance with a small cut on his scalp, CSD was suspected. It was serologically confirmed (enzyme linked immunosorbent assay (ELISA) B henselae IgG> 850 U/l; IgM> 250U/l) and he was treated with clarithromycin.

Six weeks after the onset of CSD, he showed difficulty in walking, inability to run or climb stairs, and frequent falls. Slowly progressive pain and a gait disorder were noted over a course of a further 8 weeks. On admission to our department there were no general signs and only a small submandibular lymph node remained. Mental state and cranial nerve examination were normal. He showed a symmetric distal muscle weakness in all limbs. A marked sensory ataxia was noted. We found mild root pain on a straight leg raising test. Deep tendon reflexes were very low to absent, especially in the legs. Plantar responses were flexor.

Serum concentrations of liver enzymes and glucose, and a protein spectrum were normal. Laboratory tests for adeno, RS, corona, influenza 1–2–3, parainfluenza 1–2–3, sendai, mumps, measles, herpes simplex and varicella zoster viruses, Mycoplasma pneumoniae, Chlamydia psittaci, Coxiella burnetti, Mycobacterium tuberculosis and atypical mycobacteria were negative. Antinuclear factor and ANCA were negative. Levels of complement factor 3 and 4 were normal. ELISA B henselae IgG in serum, 3.5 months after outbreak of lymphadenitis, was still >850 U/l. IgM specific antibodies now were <200 U/l. Examination of CSF showed 13 mononuclear leucocytes/μl, no polynuclear leucocytes, raised protein concentration (558 mg/l, normally 160–310 mg/l), three oligoclonal bands in CSF on isoelectric focusing, and a slight intrathecal IgG synthesis (51 mg/l, normally 3.2–15 mg/l). IgG and IgM specific antibodies against B henselae were negative in two CSF samples. Polymerase chain reaction (PCR) onB henselae in CSF and in serum was negative. Enolase, myelin basic protein, S100, lactate and glucose concentrations in CSF were normal.

Electroencephalography, cranial CT, and gadolinium enhanced MRI of the thoracolumbar region were normal. Nerve conduction studies showed a marked decrease of motor nerve conduction velocities in all limbs (median nerve 20 m/s). In sensory conduction studies no response could be elicited. The EMG was normal.

The sural nerve biopsy showed many demyelinated axons and signs of early remyelination, but no onion bulbs (figure A). The density of myelinated fibres was slightly below normal, which could be attributed partly to intrafascicular oedema. Myelinated fibre size histogram was bimodal. There were no endoneurial infiltrates, or signs of vasculitis. Electron microscopy showed macrophage induced demyelination (figure B). Polymerase chain reaction (PCR) on B henselae in sural nerve fragments was negative.

The history and clinical presentation, combined with CSF findings and the results of EMG and light and electron microscopy studies, are compatible with the diagnosis of chronic inflammatory demyelinating polyradiculoneuropathy (CIDP).

He was treated with prednisone (15 mg every other day for 1 month) and the dose was reduced slowly in the course of 4 months. His motor and sensory function recovered completely and deep tendon reflexes reappeared. Nerve conduction studies 1 year after the onset of CSD were normal.

Neurological complications of CSD are rare and predominantly of the CNS. Myelitis has been described in combination with encephalopathy2 and radiculopathy was reported only in combination with encephalomyelitis. Ophthalmological problems are oculoglandular disease of Parinaud3 and neuroretinitis.4 In a study by Carithers and Margileth of 76 patients with CSD and neurological complications, 15 patients showed signs of dysfunction of cranial or peripheral nerves. Ten patients had neuroretinitis, two children had paresis of the facial nerve, and three adult women complained of neuralgia.2 One case study presented a peripheral facial nerve paralysis as a complication of CSD.5

Up until now CIDP has never been reported as a neurological complication of CSD.

Given the history and clinical course, the electrophysiological and nerve biopsy findings, coupled with the strongly positive serology toB henselae, we think that the CIDP in this patient is a direct complication of CSD. CIDP is an autoimmune process in which both humoral and cellular factors are thought to participate in the pathogenesis. Wheeler et al also suggested an immune response as a pathophysiological mechanism responsible for CSD encephalopathy.1 In our patient a delayed myelin destruction is induced by sensitised macrophages, originally activated by the Bartonella infection. Therefore, we hypothesise that the pathophysiology of both central and peripheral nervous system complications after a CSD infection shares a similar immunological mechanism.