Course Authors

Nazia Hussain, M.D., and Peter Barland, M.D.

Dr. Hussain is a Fellow in Rheumatology, Department of Medicine, Albert Einstein College of Medicine, New York. Drs. Hussain and Barland report no commercial conflicts of interest.

Estimated course time: 1 hour(s).

Albert Einstein College of Medicine – Montefiore Medical Center designates this enduring material activity for a maximum of 1.0 AMA PRA Category 1 Credit(s)™. Physicians should claim only the credit commensurate with the extent of their participation in the activity.

In support of improving patient care, this activity has been planned and implemented by Albert Einstein College of Medicine-Montefiore Medical Center and InterMDnet. Albert Einstein College of Medicine – Montefiore Medical Center is jointly accredited by the Accreditation Council for Continuing Medical Education (ACCME), the Accreditation Council for Pharmacy Education (ACPE), and the American Nurses Credentialing Center (ANCC), to provide continuing education for the healthcare team.

 

Learning Objectives

Upon completion of this Cyberounds^®, you should be able to:

• Discuss the clinical manifestations of CNS vasculitis

• Differentiate between primary and secondary CNS vasculitis

• Describe the differences in presentation, treatment and prognosis of angiographically versus pathologically defined cases.

 

Case Presentation

A 50-year-old man with no significant pre-morbid illnesses presents to your office complaining of numbness in his right lower extremity. He also reports having frequent biparietal headaches over the last 4-6 months.

On examination, he has mild right-sided hemiparesis and hemi-sensory loss. He has noted some recent memory loss and periods of confusion. His blood count, creatinine and urine analysis are all normal, however his sedimentation rate is 50 mm/hour.

A CT scan of his head shows multiple, small bilateral infarcts and hemorrhages. A lumbar puncture is performed which shows an elevated protein and a lymphocytic pleocytosis. The patient's serum anti-nuclear antibody (ANA), anti-SSA (Ro) and anti-neutrophillic cytoplasmic antibodies (ANCA) tests are all negative. Human immunodeficiency virus (HIV) testing is done and is negative, as are tests for cerebrospinal fluid cytomegalovirus, varicella zoster virus and herpes simplex virus. A cerebral angiogram is reported as 'consistent with vasculitis.' A brain biopsy is performed confirming a diagnosis of central nervous system vasculitis.

CNS Vasculitis

Vasculitis affecting the central nervous system can be viewed as either primary, when the CNS is the sole or dominant target organ, or secondary, when it is the result of some systemic process (e.g., infection, drug toxicity or systemic disease).(1)

Primary Angiitis of the CNS

Vasculitis that predominantly affects the CNS has been referred to by many terms. Granulomatous angiitis of the CNS was formerly used because of the characteristic pathology noted in most of the originally-reported cases. However, patients with primary CNS vasculitis who demonstrate non-granulomatous pathology are not uncommon, so this terminology was inaccurate in some instances. The vasculitis has been called isolated angiitis of the CNS but there have been occasional cases with involvement, albeit limited, outside the CNS. The term Primary Angiitis of the Central Nervous System (PACNS) seems the most appropriate as it reflects the anatomic site of the predominant clinical pathology and is not limited by histologic features.

Clinical Features

CNS vasculitis is an uncommon condition. As a result, the literature on it is rather limited. Analyses of the available literature is further complicated by the fact that not all cases are pathologically documented, so that the case reports fall into two categories: those in whom tissue biopsies are reported and those in whom a presumptive diagnosis of vasculitis was made based on neurodiagnostic imaging [i.e., cerebral angiography, magnetic resonance angiogram (MRA), etc.]. Subgroup analysis of the literature suggests that the pathologically documented cases and the angiographically diagnosed cases are two clinically distinctive subgroups.

Pathologically documented cases of PACNS tend to be predominantly men of a wide range of age (46 + 17 years). Virtually any neurologic deficit can be encountered in this group. Headache is the most common symptom. Eighty-five percent of patients develop a variety of focal neurologic deficits, usually in conjunction with the presence of diffuse neurologic dysfunction, such as decreased mentation or altered level of consciousness. The course of disease is subacute to chronic -- the mean duration of symptoms before diagnosis is 5 months. An acute onset is less common.

Angiographically defined cases, in contrast, tend to be young women, typically with a short duration of symptoms, who present with headache as the most frequent and predominant symptom. These patients characteristically have a relatively normal cerebrospinal fluid(3) and tend to have a more favorable outcome despite less aggressive therapy.

Etiology and Pathogenesis

The etiology and pathogenesis of PACNS are unknown. It has been suggested that two associated co-morbid factors may be involved in the etiopathogenesis, namely, an association with systemic viral illnesses and an association with diseases or conditions that alter host defenses.

Through the end of 1995, over 168 cases of PACNS were documented in the English language medical literature and/or archives of the Cleveland Clinic Foundation. This number represents a mixture of histologically confirmed and unconfirmed cases. Among these 168 cases analyzed by Duna et al.,(3) 29 were associated with a treatment or an illness characterized by an immunosuppressive state, including steroid therapy, lymphoproliferative or myeloproliferative disorders, HIV infection and miscellaneous conditions including post-allograft transplantation. In addition, a variety of pathogens have been documented in association with CNS arteritis, including varicella zoster virus, human immunodeficiency virus and cytomegalovirus.

Pathology

PACNS is primarily a leptomeningeal and cortical vasculitic disease involving small and medium sized leptomeningeal and cortical arteries. Less frequently, the veins and venules are involved. The gross examination of the brain and spinal cord reveals no changes specific to the disorder. The majority of cases have small infarcts or hemorrhages. Although widespread involvement is the rule, selective involvement of the telencephalon, individual lobes or the posterior fossa structures has been documented.(2) The cellular infiltrate is granulomatous in nearly all cases.(2) However, an assortment of different patterns of angiitis may be observed in the same biopsy specimen and even in adjacent areas, given a sample of adequate size.(1) Both foreign body and Langerhans giant cells may be present in granulomatous angiitis or there may be only necrotizing lymphocytic vasculitis.

Laboratory Evaulation

Evaluation of suspected cases of PACNS is hindered by lack of reliable non-invasive studies or lab tests of sufficient predictive value to exclude or confirm such a diagnosis. Hemoglobin levels, white blood cell count and erythrocyte sedimentation rate are normal in 83%, 58% and 35% of cases respectively. In the absence of clinical features suggestive of connective tissue diseases or systemic vasculitides, testing for a variety of autoantibodies is of extremely limited value.

The level of clinical suspicion should guide serologic testing for infectious diseases (e.g., HIV, hepatitis C, syphilis).

CSF analysis, on the other hand, is an essential part or the diagnostic process for PACNS. The CSF is abnormal in 80-90% of pathologically documented cases of PACNS.(1) The fluid results usually reflect an aseptic meningitis with a modest pleocytosis and elevated protein level (mean protein 177 mg%, median 100 mg%; mean number of cells 77/mm,(3) median 55/mm(3)).

CSF abnormalities persist or evolve on repeated examinations as the illness proceeds, taking the form of a rising CSF protein concentration and the emergence, or worsening, of a leucocytic pleocytosis. Increased immunoglobulin G synthesis and the presence of oligoclonal bands are occasionally detected but these findings lack sensitivity and specificity and are generally not useful. The CSF glucose in all but a few reported cases has been more than 54 mg/dl. The major clinical usefulness of glucose determination lies in the exclusion of chronic infection, which would produce very low or progressively low levels in untreated patients.(2)

Other laboratory testing should be directed at ruling out some of the other conditions that can clinically mimic PACNS (see below).

Radiology

CT Scan is less sensitive than MRI (65% vs. 90% respectively); however, both CT and MRI lack specificity.

The most common findings are bilateral supratentorial infarcts distributed in the cortex, deep white matter and/or leptomeninges. With intravenous contrast, enhancement may be seen.

Of note, the combination of normal MRI and CSF results has a strong negative predictive value and will exclude CNS vasculitis in most clinical situations, thus obviating the need for more invasive tests.

In histologically confirmed cases of PACNS, cerebral angiography has a sensitivity of only 60%. Moreover, even when cerebral angiograms are abnormal, 'classic' findings of arteritis (alternating areas of stenosis and ectasia in multiple vascular distributions) occur in <40% of cases. Interpretation of cerebral angiography is further limited by poor specificity (26%), as angiographic findings compatible with vasculitis are commonly encountered in non-vasculitic conditions such as vasospasm, CNS infection and even atherosclerosis. In particular, cerebral vasospasm may be seen in a variety of disorders (e.g., drug exposure, postpartum state, hypertension) and is angiographically indistinguishable from the findings considered consistent with vasculitis. The cerebral angiogram, therefore, like all clinical tests, can only be fully interpreted in the context of the details of the case at hand.

In view of its currently lower spatial resolution, MRA should not be regarded as equivalent to conventional angiography for the detection of CNS vasculitis.

It has been suggested that functional brain imaging studies such as SPECT and PET scans be used for the diagnosis of PACNS. However, results of these tests need to be interpreted with caution because of the following reasons:

1. Defects in blood flow do not always correlate with symptoms.
2. Perfusion defects do not distinguish vasculitis from other forms of vasculopathy.
3. An abnormality may not differentiate primary CNS processes from neurologic manifestations of extra cranial disease.

Biopsy

Histologic confirmation remains the clinical standard for the diagnosis of all forms of vasculitis including PACNS. However, brain biopsy, too, is limited by poor sensitivity, with pre-mortem biopsy yielding false negative results in 25% of autopsy documented cases! Sampling of the leptomeninges, as well as the underlying cortex, will increase diagnostic yield, since vasculitis may be present in only one of the two sites. Biopsy of a radiographically abnormal area similarly improves sensitivity. In the absence of a focal lesion, the temporal tip of the non-dominant hemisphere is the preferred biopsy site. Tissue samples should be stained and cultured for microorganisms.

Although false positive biopsies are rarely reported, areas of vascular inflammation may be encountered in lymphoproliferative diseases and CNS infections. As with all invasive procedures, one must weigh the risks and benefits of this test. The risk of hemorrhage with a stereotactic brain biopsy is variably reported as 0.3 to 4.7% and the mortality ranges from 0 to 2.3%.(5)

Therapy

There are no controlled trials of therapy in PACNS diagnosed by either biopsy or angiography. Regardless of the method of diagnosis, and despite the use of corticosteroids alone in 30.4% of the biopsy proven cases and 49.1% of the angiographically defined cases, mortality has been reported to be 4.4% for biopsy proven and 3.6% for angiographically defined cases. Based on their experience with a small, uncontrolled series of patients, Cupps et al.,(4) in 1983 suggested that aggressive combination therapy with corticosteroids and cyclophosphamide is beneficial and is the preferred therapy for severe, progressive or corticosteroid resistant cases of PACNS.

It seems reasonable, therefore, to reserve aggressive therapy (cytotoxic + high dose steroids) for patients who experience a progressive neurologic illness and when at least an attempt has been made to define the condition by histology. The optimal duration of therapy is unknown, but usually it is continued for 6-12 months after clinical remission is achieved.

For the subgroup of patients with acute focal presentations associated with normal CSF, in whom the diagnosis is inferred on the basis of angiography alone, a diagnosis of benign angiopathy of the CNS (BACNS) is appropriate. In such patients, it is believed that there is rarely an indication for initial therapy with aggressive combination immunosuppressive agents. The use of a relatively brief course (3-6 weeks) of high dose corticosteroids in conjunction with a calcium channel blocker may be appropriate.(1)

Secondary CNS Vasculitis

It is very important, when faced with a clinical situation suspicious for central nervous system vasculitis, to rule out secondary causes of this condition before labeling it as PACNS. Secondary causes of CNS vasculitis include the following:

Infection

Viruses (HIV, CMV, VZV, others), bacteria [Borrelia burgdorferi (the agent that causes Lyme disease)], syphilis, Bartonella, Mycobacterium tuberculosis, fungi (aspergillus, coccidiodes, others) and rickettsiae are associated with CNS vasculitides.

In many cases the organism may be angioinvasive but, in others, the vascular inflammation may result from alterations in host defenses, with secondary damage to host tissues.

Systemic Vasculitides

Vasculitis of the CNS may occur with any of the systemic vasculitides but is more commonly reported in Polyarteritis Nodosa (PAN), Behcet's syndrome, Wegener's granulomatosis, Churg-Strauss syndrome and Giant Cell (Temporal) Arteritis.

The true prevalence of CNS angiitis in these cases is difficult to estimate since the diagnosis is most often presumed on clinical grounds, when focal and/or diffuse neurologic deficits develop in the setting of systemic disease.

PAN affects small and medium sized arteries and less commonly arterioles and venules. Clinical CNS involvement occurs in about a quarter of patients.

Wegener's granulomatosis is a necrotizing granulomatous disease of the upper and lower respiratory tracts associated with systemic necrotizing arteritis of small arteries and veins, and glomerulitis. Clinical and pathologic CNS involvement occurs in 25% of patients and is generally ascribed to contiguous invasion of nasal and paranasal granulomas. Granulomas remote from nasal granulomas and CNS arteritis are uncommon.

Giant cell and Takayasu's arteritis are associated with granulomatous vasculitis of medium and large sized arteries. Neurologic involvement in temporal arteritis generally results from occlusion of extradural cerebral vessels. In Takayasu's arteritis, CNS involvement results from uni- or bilateral occlusion of extra-cranial carotid and vertebral arteries. Intracranial vessel involvement is extremely rare.

Connective Tissue Diseases

CNS involvement is noteworthy in systemic lupus erythematosis (SLE) and Sjogren's syndrome.

In SLE, brain pathology most often reveals a vasculopathy with small vessel thickening, hyalinization, intramural platelet deposition and thrombus formation. Frank vasculitis is uncommon (<7%) but has been documented in patients with acute, dramatic and often fatal CNS presentations. When SLE patients present with neurologic events, it is important to consider causes other than vasculitis such as anti-phospholipid syndrome, cardiac emboli and thrombotic thrombocytopenic purpura.

In Sjogren's syndrome, CNS manifestations may be caused by a mononuclear inflammatory vasculitis involving the small vessels of the cortex and the meninges. However, since angiographic abnormalities consistent with vasculitis are uncommonly seen (20%), and only a few patients have been studied histologically, the prevalence, pathogenesis and treatment of CNS vasculitis in Sjogren's syndrome are largely unresolved.

Sarcoidosis

A pre-mortem diagnosis of CNS granlomatous angiitis was made in two patients with systemic sarcoidosis and in five other patients at autopsy, four of whom also had systemic involvement. In these patients, it is not clear whether the angiitis was related to the sarcoidosis.

Malignancy-related States (Especially Lymphoproliferative Diseases)

CNS vasculitis has been reported in association with Hodgkins lymphoma, non-Hodgkins lymphoma and angioimmunolymphoproliferative disorders. The lymphoproliferative disease itself may be detected outside or within the CNS. The pathology of the CNS lesions may be indistinguishable from that of PACNS with widespread granulomatous angiitis.

Whenever faced with a clinical picture that seems compatible with CNS vasculitis, it is imperative not only to rule out secondary causes but also to evaluate for conditions that can closely mimic vasculitis. The list of 'vasculitis mimics' includes the following conditions:

1. Atherosclerosis
2. Moyamoya disease
3. Emboli (including cardiac myxoma)
4. Hypercoagulable states (including anti-phospholipid syndrome.)
5. Intravascular neoplasm (angioendotheliomatosis)
6. Thrombotic thrombocytopenic purpura
7. Reversible vasoconstriction of any cause, e.g., drugs of abuse (cocaine), sympathomimetic drugs
8. Dissections of cerebral arteries
9. Radiation vasculopathy

Summary

Primary Angiitis of the Central Nervous System (PACNS) is an uncommon entity, and a challenge to diagnose and treat. In order to arrive at this diagnosis, it is most important to rule out secondary causes and the so-called vasculitis mimics.

A normal lumbar puncture, together with a normal MRI, has important negative predictive value and may obviate the need for further invasive testing. Cerebral angiography, while useful in confirming a suspicion of PACNS and helping to rule out other diagnoses such as Moyamoya disease, has only moderate sensitivity and specificity. The gold standard for diagnosis is histological examination of brain and leptomeninges obtained by biopsy. The risk and benefit of pursuing this invasive procedure depend on the clinical situation at hand. Treatment of pathologically documented cases is very different from the 'benign angiopathy of the CNS' variant where the diagnosis is made angiographically.