Eosinophilic vasculitis in an isolated central nervous system distribution

BACKGROUND

Eosinophilic vasculitis has been described as part of the Churg–Strauss syndrome, but affects the central nervous system (CNS) in only 6–7% of cases.1^,2 Presentation in an isolated CNS distribution is rare. We present a case of eosinophilic vasculitis isolated to the CNS.

CASE PRESENTATION

A 39-year-old woman with a history of migraine without aura presented to an institution (located in the borough of Queens, New York, USA; no academic affiliation) in an acute confusional state with concurrent headache and left-sided weakness and numbness. Her symptoms resolved over 2 weeks without residual deficits. After 18 months, the patient presented again with a similar constellation of symptoms, including acute-onset headache, confusion, left visual field flashes, and left-sided numbness and weakness. The patient worked at a perfume counter and identified odours of certain perfumes as a trigger for her migraine attacks. She had no respiratory complaints including asthma, rash, constitutional symptoms or allergies. Other than a multivitamin, she took no drugs regularly, including non-prescription supplements. Further history and physical examination showed no features suggestive of systemic vasculitis, including lymphadenopathy or rash. Neurological examination was notable for mildly impaired attention, a left homonymous lower quadrantanopsia, left hemiparesis and left patchy hemihypesthesia.

INVESTIGATIONS

First presentation

Lumbar puncture showed acellular cerebrospinal fluid (CSF) with a slightly increased protein concentration. She was treated presumptively for herpes encephalitis on the basis of an increased herpes simplex virus (HSV)-2 IgG titre in the serum and CSF (IgM not analysed and CSF polymerase chain reaction negative), as well as bifrontal polar gyral-enhancing brain lesions on magnetic resonance imaging (MRI).

Second presentation

Brain MRI showed new occipital–parietal leptomeningeal-enhancing lesions; lumbar puncture again showed acellular CSF with raised protein concentration (128 mg/dl). Brain MRI showed right occipital, anterior frontal and bilateral parietal cortical lesions with leptomeningeal enhancement on the T1-weighted post-contrast images (fig 1A, black arrowheads). Gradient echo showed patchy low signal in sulci, consistent with persistent intracellular deoxyhaemoglobin or methaemoglobin. FLAIR images showed high signal of the cortical mantle as well as the subarachnoid space in the adjacent gyri (fig 1B, white arrowheads) that showed prolonged diffusion rates on diffusion-weighted images and apparent diffusion coefficient maps, consistent with inflammatory oedema or subacute ischaemia. The bifrontal lesions observed on MRI during the patient’s previous hospitalisation were faintly visible on the FLAIR sequence (fig 1B, white arrow) and showed low signal in the sulci on the gradient echo images, consistent with haemosiderin deposition from prior haemorrhage.

View larger version:

• In a new window

• Download as PowerPoint Slide

Figure 1 (A) Brain magnetic resonance imaging (MRI) post-gadolinium T1 sequence shows right occipital, anterior frontal, and bilateral parietal cortical lesions with leptomeningeal enhancement on the T1-weighted images (black arrowheads).

(B) FLAIR image shows high signal of the cortical mantle as well as the subarachnoid space in the adjacent gyri (white arrowheads). The bifrontal lesions observed on MRI during the patient’s previous hospitalisation show evolution (white arrow). (C, D) Histological examination shows transmural, predominantly eosinophilic, inflammatory infiltrates of medium-sized leptomeningeal arteries (black arrows). Small cortical vessels showed focal fibrinoid necrosis and swollen endothelial cells, without granulomas.

Serum inflammatory markers were increased, with an erythrocyte sedimentation rate of 35 mm/h (normal range 0–22 mm/h) and C-reactive protein level 92.6 mg/l (normal <3 mg/l; table 1). The patient had a normal blood count including only 1% eosinophils. Serum electrolytes, creatinine, liver enzymes and urine analysis were also normal. Antithyroglobulin and antimicrosomal antibodies were not detected. Stool was negative for ova and parasites. CSF examination showed 2 leucocytes/mm³ (100% lymphocytes), 300 erythrocytes/mm³, a protein level of 119 mg/dl and normal glucose. CSF cultures were negative, as were polymerase chain reaction studies for HSV-1 and HSV-2 and varicella zoster virus. Serum rapid plasma reagin was non-reactive and the patient was HIV negative. Rheumatological studies, including antinuclear antibodies, anti-DNA, anti-extractable nuclear antigens, antineutrophil cytoplasmic antibodies (cANCA, pANCA), rheumatoid factor and lupus anticoagulant, were negative. A DNA analysis was carried out to look for mitochondrial myopathy, encephalopathy, lactic acidosis and stroke-like episodes (MELAS), but there was no A3243G mutation. Cerebral angiography showed no evidence of focal arterial narrowing. Computed axial tomography of the chest was normal.

Brain biopsy showed transmural, predominantly eosinophilic, inflammatory infiltrates of medium-sized leptomeningeal arteries (fig 1C, D, black arrows). Small cortical vessels showed focal fibrinoid necrosis and swollen endothelial cells, without granulomas. Immunoperoxidase staining for amyloid-ß peptide in cortical and subpial vessels showed no evidence of amyloid angiopathy. Gram stain, bacterial culture, staining for acid-fast bacilli and mycobacterial culture of the brain biopsy were all negative. No viral inclusions were identified, and immunohistochemical stains for HSV-1, HSV-2 and varicella were negative.

TREATMENT

The patient received high-dose intravenous dexamethasone for one week, followed by a prednisone taper.

OUTCOME AND FOLLOW-UP

The patient’s dysesthesias resolved rapidly on steroid therapy. At one year follow up she had mild residual sensory loss without motor signs and had not suffered a recurrence. Follow up brain MRI showed gyral volume loss and encephalomalacia in previously involved areas.

DISCUSSION

Eosinophilic vasculitis is a feature of certain rheumatological conditions, but is rare in an isolated CNS distribution.3 Stroke caused by vasculitis is well described in the Churg–Strauss syndrome, which is characterised by multisystem granulomatous eosinophilic vasculitis accompanied by peripheral eosinophilia and a history of asthma.4 Granulomatous eosinophilic vasculitis accompanied by mild peripheral eosinophilia has been observed in parenchymal and leptomeningeal vessels affected by cerebral amyloid angiopathy.5 Eosinophilic temporal arteritis and claudicating systemic arteritis, associated with a 24% peripheral eosinophil count, was reported with vasculitic vertebrobasilar stroke, although no aetiology was identified.6 Here we describe a non-granulomatous eosinophilic vasculitis of the CNS with no accompanying eosinophilia of the peripheral blood or CSF, and no involvement of other organ systems. Given their similarities, her two isolated episodes were probably due to the same disorder. Lack of peripheral eosinophilia argues against systemic infection or drug reaction. The absence of granulomas further distinguishes this case histopathologically from nearly all other reported cases of isolated CNS vasculitis. Although her MRI findings could indicate focal lobar oedema with haemosiderin deposition, the condition argues against cerebral amyloid angiopathy or amyloid-ß-related angiitis, which has been described as a reversible cerebral amyloid angiopathy leucoencephalopathy syndrome with associated angiitis.7^,8 No aetiology for this patient’s primary CNS eosinophilic vasculitis has yet been identified. Spontaneous resolution and recurrence of her syndrome is an unusual feature for typical CNS vasculitis and may suggest an environmental epitope with immune reaction as the cause.