- Does this patient have CNS vasculitis?
- What tests to perform?
How should patients with CNS vasculitis be managed?
What happens to patients with CNS vasculitis?
How to utilize team care?
Are there clinical practice guidelines to inform decision making?
- Other considerations
Does this patient have CNS vasculitis?
Vasculitides are a heterogeneous group of conditions that have in common blood vessel wall inflammation. Vasculitis can involve both the central and the peripheral parts of the nervous system. Central nervous system (CNS) vasculitis is a rare condition, but it remains one of the most challenging disorders to diagnose and treat.
Broadly, CNS vasculitis can be divided into primary angiitis of the CNS (PACNS), where the inflammation is restricted to the vessels of the brain and/or spinal cord, and secondary CNS vasculitis, when it is associated with other systemic disorders.
The latter group includes CNS vasculitis secondary to:
Idiopathic systemic vasculitides (e.g., polyarteritis nodosa (PAN)).
Systemic vasculitides from autoimmune diseases (e.g., systemic lupus erythematosus (SLE).
Systemic vasculitis from non-autoimmune disorders including infections (e.g., Human immunodeficiency virus (HIV), hepatitis C virus–associated cryoglobulinemic vasculitis), herpes zoster, and use of certain drugs (e.g., hydralazine-associated microscopic polyangiitis), cancer etc.
Diagnosis of PACNS
PACNS causes diffuse inflammation of mostly small- to medium-size blood vessels in the CNS. PACNS can present with any neurologic complaint, thus the differential diagnosis is broad. Neurological symptoms and examinations findings depend on the area of the brain or spinal cord involved.
Onset of symptoms can be acute, subacute or chronic with a relapsing-remitting course. Most of the symptoms of vasculitis are due to ischemia or infarction. These present as acute or subacute development of focal neurological deficits including weakness, sensory deficits, dysarthria, aphasia or visual deficits depending on which area of the CNS is affected. There are usually no associated systemic features in PACNS. Headache is the most common symptom in PACNS, followed by altered cognition and persistent neurological deficits.
Stroke and transient ischemic attacks occur in 30–50% of patients with PACNS involving multiple vascular territories.
A new neurologic deficit in patient with systemic vasculitis warrants an evaluation for possible secondary central nervous system vasculitis. Extensive investigations are often needed to make the diagnosis. CNS infections, tumors, demyelinating disorders and paraneoplastic syndromes are a few of the neurological syndromes that can have presentations similar to CNS vasculitis irrespective of etiology.
The diagnosis of CNS vasculitis is considered when patients present with:
Recurrent episodes of cerebral ischemia or infarction, in multiple vascular territories with an inflammatory cerebrospinal fluid (CSF) profile.
Subacute or chronic headache with cognitive decline.
Aseptic meningitis (usually with mild abnormalities in the CSF), or chronic meningitis when alternate etiologies including infections and malignancy have been excluded.
Reversible cerebral vasoconstriction syndrome
An important mimic of CNS vasculitis is reversible cerebral vasoconstriction syndrome (RCVS). RCVS is a syndrome characterized by acute onset of thunderclap headache (severe intensity headaches that reach maximum intensity usually in less than 1 minute), with or without neurological deficits, with evidence of reversible cerebral vasoconstriction on cerebrovascular imaging.
PACNS is almost never associated with thunderclap headache and it usually has a subacute onset headache with a chronic course.
Other distinctive features between RCVS and PACNS include normal CSF findings in RCVS as compared to PACNS, and the predominance of diffuse cerebral vessels involved in RCVS compared to limited vessel involvement in PACNS. A variety of factors that precipitate RCVS onset have been reported. These include drugs such as cannabis and cocaine; medications like selective serotonin reuptake inhibitors, ergotamines, nasal decongestants and triptans; postpartum period; or head trauma and post neurosurgical procedures.
Other mimics of PACNS
Other less common angiographic mimics of PACNS include intracranial atherosclerosis, fibromuscular dysplasia, moyamoya disease, radiation vasculopathy and angiotropic or intravascular lymphoproliferative disorders. Any of the systemic vasculitides or systemic autoimmune disease discussed below that cause CNS vasculitis can mimic PACNS.
Secondary CNS vasculitis
Occurs in the setting of systemic vasculitides such as PAN, microscopic polyangiitis (MPA), granulomatosis with polyangiitis (GPA), eosinophilic granulomatosis with polyangiitis (EGPA) and Behçet’s syndrome.
Infectious causes of CNS vasculitis
Many infectious pathogens including bacteria, viruses, parasites, and fungi can cause cerebrovascular complications and vasculitis is a commonly implicated mechanism. Infection-related vasculitis can closely mimic CNS vasculitis; thus, a high degree of suspicion and thorough evaluation is required to make the diagnosis because treatment options are drastically different.
The human immunodeficiency virus (HIV) has been associated with a systemic and CNS vasculitis and the latter accounts for 13 to 28% of strokes in HIV. However, these are more commonly seen in patients with secondary infections such hepatitis B virus, hepatitis C virus, neurosyphilis, tuberculous meningitis, varicella zoster virus (VZV), cytomegalovirus, and cryptococcal meningitis, which are themselves associated with vasculitis. Other causes of strokes in HIV are related to hemorrhagic stroke due to immune-related thrombocytopenia or coagulopathy, cardioembolic stroke due to AIDS-related cardiomyopathy or endocarditis and accelerated atherosclerosis. Whether HIV-related vasculitis exists independent of these secondary infections is unclear. However, a vasculopathy of large and small vessels have been described in patients with advanced disease, some who have pathological confirmation of vasculitis in the absence of alternate etiologies such as concurrent opportunistic infections. Direct viral invasion of the endothelium or immune complex deposition are implicated in the pathogenesis. In children with advanced disease, a large-vessel vasculopathy of the arteries near the circle of Willis consisting of fusiform aneurysmal dilatations and intracranial vessel stenosis has been described. This is characterized histopathologically with medial fibrosis, marked intimal hyperplasia, and relative lack of inflammation.
Varicella zoster virus vasculopathy is an important cause of stroke. In children, primary infection accounts for nearly one-third of ischemic stroke. In adults, ischemic and less commonly hemorrhagic stroke as a result of VZV vasculitis is usually from reactivation of the virus. Traditionally, it was described in elderly immunocompetent people who developed unifocal ipsilateral large artery stenosis (anterior cerebral artery, MCA, or ICA) after zoster ophthalmicus (usual latent period 3–4 days), causing an acute monophasic hemiparesis contralateral to the ocular disease. However, parts of this classic clinical picture may be missing including absence of a rash and development of neurologic signs and symptoms occurring months after the rash. A multifocal vasculopathy involving both small and large vessels develops in immunocompromised patients.
Spirochetes like treponema pallidum are a well-known cause of stroke by causing an inflammatory arteritis, mostly in the middle cerebral artery. Viruses like Hepatitis B and C, Cytomegalovirus can cause vasculitis. Fungal infections with Cryptococcus and Coccidioides immitis can cause stroke through multiple mechanisms including cerebral vasculitis. These are usually opportunistic infections in severely immunocompromised hosts.
Systemic autoimmune diseases
Certain systemic autoimmune diseases that are associated with vasculitis may have prominent neurological features. These diseases include SLE, antiphospholipid syndrome, rheumatoid arthritis (RA), Sjögren’s Syndrome and sarcoidosis.
In SLE, neurological damage can be a result of direct antibody-mediated tissue damage or damage to the vessels supplying the tissues (cerebral vasculitis). However, there is little histological support for lupus cerebritis, and cerebral vasculitis in SLE is uncommon. Vasculopathy on the other hand is seen often in SLE.
Sarcoidosis affects the nervous system in 5-15% of cases, commonly affecting the brain parenchyma, leptomeninges, hypothalamus, pituitary, cranial nerves, and dura mater. Intracerebral hemorrhage occurs in 0.6% of unselected sarcoidosis and neurosarcoidosis series. The pathogenesis of intracerebral hemorrhage in neurosarcoidosis is not understood but may be related to arterial and venous bleeding from an aggressive vasculopathy due to granulomatous vessel invasion.
CNS vasculitis has been reported in association with Hodgkin’s and non-Hodgkin’s lymphoma and angioimmunolymphoproliferative lesions. These can be indistinguishable from PACNS, but appropriate immunohistochemistry staining as well as B-cells and T-cells markers should be performed because the presence of vasculitic changes does not exclude an underlying lymphoproliferative condition.
What tests to perform?
Routine evaluation of ischemic strokes should be performed when suspecting CNS vasculitis to first evaluate for the more common causes of strokes; this should include cardiac rhythm monitoring, transthoracic echocardiography, and transesophageal echocardiography when suspecting a source of cardioembolism.
Unless contraindicated, every patient being evaluated for CNS vasculitis should get a lumbar puncture for CSF analysis. This is most important for diagnosis of infections as detailed above under infectious causes of CNS vasculitis and a thorough evaluation with appropriate stains and cultures should be conducted. Patients with RCVS have a normal CSF profile. In contrast, patients with PACNS have abnormal CSF results in 80–90% of cases, although findings are non-specific. These include elevated protein levels, increased IgG synthesis, mild pleocytosis and normal glucose. A thorough evaluation of the CSF for an infectious process is extremely important. Unbiased metagenomic next-generation sequencing (mNGS) is an emerging technique that can offer a different approach to detect CNS infections. mNGS have enabled unbiased sequencing of biological samples to determine the source of all the non-host (e.g., pathogen-derived) nucleic acid in a sample. mNGS can be useful to detect an infectious pathogen that has a low yield with a conventional detection technique.
MRI and CT
Magnetic Resonance Imaging (MRI) and computed tomography (CT) can be used for imaging the brain when suspecting CNS vasculitis. Contrast enhanced images provide information about blood brain barrier breakdown.
MRI of the brain is more sensitive than CT and provides superior contrast resolution and anatomic detail. However, neither MRI or CT is specific for the diagnosis of CNS vasculitis. MRI findings in CNS vasculitis may include both cortical and subcortical areas showing ischemic and hemorrhagic lesions, periventricular white matter lesions, and parenchymal or meningeal contrast enhancement.
In RCVS, which is frequently misdiagnosed as PACNS, brain MRI can often be normal despite an abnormal cerebrovascular imaging. The combination of normal MRI and abnormal cerebrovascular study along with normal CSF findings in the patient with the appropriate clinical presentation should favor the diagnosis of RCVS.
Vessel imaging can be obtained using MR or CT angiography, but cerebral angiography provides imaging of smaller vessels up to 500 µm in diameter. The resolution of vessels smaller than that is low using cerebral angiography thus decreasing sensitivity in smaller vessels, and this limitation must be considered when interpreting results. In pathologically proven disorders such as granulomatous angiitis of the CNS, the sensitivity of cerebral angiography findings is as low as 10–20%.
Cerebral angiography is often considered the gold standard for the diagnosis of CNS vasculitis, and this is a fallacy since it does not provide a pathological diagnosis, which is the only gold standard for diagnosis.
Cerebral angiographic findings in PACNS include vessel beading which refers to alternating areas of stenosis and dilatation that can be smooth or irregular, and typically occur bilaterally but can also include single vessels; these findings are far from specific and is also seen in atherosclerosis, radiation vasculopathy, infection, and vasospasm. The specificity of cerebral angiography for diagnosis of PACNS can be as low as 30%.
RCVS highly mimics angiographic findings in CNS vasculitis, but there is reversibility of these vessel abnormalities on repeat imaging within 3 months.
There is ongoing research using high resolution contrast enhanced MR imaging of vessels to look at vessel wall characteristics of wall thickening and vessel wall enhancement to characterize and differentiate between CNS vasculitis, RVCS and intracranial atherosclerosis.
Brain biopsy remains the gold standard for the diagnosis for CNS vasculitis and should be entertained in patients with a chronic meningitis picture, or when there is a suspicion for infection or malignancy.
The sensitivity of brain biopsy still remains low with a false negative rate of up to 25% in autopsy-detected cases; the utility of brain biopsy lies in the definitive identification of cases where laboratory testing and neuroimaging were inconclusive, and again to rule out alternate etiologies including infections and malignancies.
Open wedge biopsy of a radiologically identified lesion enhances the yield of the biopsy. If this is not possible, biopsy of the temporal lobe tip on the non–dominant side with a sampling of the overlying leptomeninges and underlying cortex is an alternative. Many clinicians fear the risks of brain biopsy; however, the risks are generally low in experienced centers. Small hemorrhage at the biopsy site without permanent neurological deficit can occur. In a large series examining the safety of stereotactic brain biopsy in over 7000 patients, the mortality rate was less than 1%.
Since the sensitivity and specificity of all tests described above are suboptimal, making a diagnosis should include a comprehensive review of the history, physical examination, laboratory, neuroimaging and pathological data. Even when a pathological diagnosis of vasculitis is made, there should be a thorough search for infections and malignancy. The importance of this cannot be overemphasized.
How should patients with CNS vasculitis be managed?
Randomized controlled trials to direct treatment in CNS vasculitis are non-existent given the rarity of the disease. Most of the evidence is based on retrospective observational data and expert opinion.
PACNS is initially treated with glucocorticoids alone or in combination with cyclophosphamide, depending on the severity and the involvement of the disease; cyclophosphamide is continued for 3-6 months till remission is achieved and then is switched to maintenance therapy such as azathioprine or mycophenolate mofetil. Limited data are available on the long term outcome of patients with PACNS; thus the duration of maintained therapy is unknown. In our experience, patients with pathologic findings of granulomatous angiitis should be treated more aggressively than others since granulomatous angiitis is historically associated with a worse outcome.
Assessing disease activity with serial MRI scans, looking for subclinical strokes or new lesions on the MRI and clinical symptoms is essential. Serial MRI scans are especially important during periods of tapering and switching medications to monitor for disease activity. CSF studies can also be done to document improvement in the inflammatory response.
Along with the immunosuppressants, prophylaxis for pneumocystis carinii infection and adequate prophylaxis for osteoporosis should be started to prevent associated side effects.
What happens to patients with CNS vasculitis?
Earlier studies suggested that biopsy-proven PACNS was a highly fatal disease; however, the use of glucocorticoids and cyclophosphamide has resulted in a higher degree of good outcomes in patients with PACNS.
The estimated mortality rate of PACNS varies between 10% and 17%. It has been more difficult to estimate the disability rate as different studies have used different outcome measures. 20%-22% of PACNS patients tend to have severe disability from PACNS or from complications of the disease. Disability scores tended to improve over time.
For secondary CNS vasculitis, the prognosis for the most part depends on the detection and management of the underlying systemic condition.
How to utilize team care?
The diagnosis and management of CNS vasculitis requires a concerted team effort.
Initial consultation or admission is usually to a neurological service. Evaluation involves the participation of vascular neurologists, neuroradiologists, neurosurgeons, neuropathologists and rheumatologists. Infectious disease specialists should be included when there is suspicion for an infectious etiology.
Once the diagnosis is established, long term follow-up is set up with rheumatology and vascular neurology. Depending on symptoms which may arise during the course of the illness, subspecialty consultation with epilepsy or headache may be considered.
Physical therapy, occupational therapy, speech therapy and cognitive rehabilitation are extremely important, especially after a stroke, depending on the deficits.
Are there clinical practice guidelines to inform decision making?
There are no official guidelines for the management of PACNS.
Typical lengths of stay
Length of hospital stay depends on the reason for admission. Large ischemic strokes, intracranial hemorrhage or seizures may warrant intensive care unit level management prior to transferring the patient to a regular nursing floor.
For undiagnosed patients who are admitted with less acute presentations, usual hospital length of stay is less than a week. For clinically stable patients, diagnostic evaluation can be pursued as an outpatient.
Copyright © 2017, 2014 Decision Support in Medicine, LLC. All rights reserved.
No sponsor or advertiser has participated in, approved or paid for the content provided by Decision Support in Medicine LLC. The Licensed Content is the property of and copyrighted by DSM.