While primary angiitis of the central system (PACNS) remains a rare entity, the poor specificity of the available diagnostic tests and its multiple mimics create a major diagnostic challenge. Recently, there have been advances in understanding PACNS and differentiating it from its mimics. A recent breakthrough is the proposal of reversible cerebral vasoconstriction syndromes (RCVS) as a unifying concept for a group of disorders that highly mimics PACNS. RCVS are characterised by acute-onset, recurrent headaches, with or without additional neurologic events, with reversible vasoconstriction of the central nervous system (CNS) vasculatures, mimicking CNS vasculitis. RCVS are considered the most common mimics of PACNS. Advances in our understanding of RCVS have allowed for identification of patients previously confused with PACNS. The scope of this article focusses on the work-up, differential diagnosis and evaluation of PACNS, as well as a discussion of the secondary CNS vasculitides with emphasis on their clinical findings, diagnoses and treatment.
What is primary angiitis of the central nervous system?
Central nervous system vasculitis is divided into two nosologic categories based on whether it is primary or secondary. PACNS is a vasculitic process confined to the central nervous system, with no identifiable underlying systemic diseases or aetiologic attribute. By contrast, CNS vasculitis is considered secondary when an identified systemic disease or infectious process is being implicated as the origin of the CNS vasculitis.
PACNS was initially described in the mid-1950s as a fatal disease and was considered a rare disorder with only 46 cases reported until 1986 . PACNS carried a dreadful prognosis until 1983, when Cupps et al. reported successful treatment with a combination of cyclophosphamide and glucocorticoids. This resulted in enthusiasm for early diagnosis and treatment. Moreover, advances in neuroradiologic diagnostic modalities led to better recognition of PACNS but created an admixture of cases diagnosed by different modalities. The temporal trends in diagnosis, treatment and outcomes created a need for a re-appraisal of PACNS as a single nosologic entity. In 1993, Calabrese et al. analysed 108 patients with CNS vasculitis drawn from the English literature in addition to their own patient experiences. They recognised that cases referred to as PACNS compromised a heterogeneous disorder with clinical subsets that significantly differed in terms of prognosis and therapy. It is currently well accepted that PACNS comprises different subsets, including granulomatous angiitis of the CNS (GACNS), and atypical cases. RCVS are not subset of PACNS, but rather a group of disorders characterised by reversible vasoconstrictive processes confined to the CNS, and not true vasculitis. RCVS have marked clinical and pathophysiological heterogeneity compared with PACNS and are considered the most common mimics of PACNS.
Calabrese et al. proposed a set of criteria for the diagnosis of PACNS ( Table 1 ) . Despite the expansion of different diagnostic modalities and our knowledge and ability to better identify certain disease entities, these criteria remain the essential guidelines that should be followed in the work-up of PACNS.
|
Granulomatous angiitis of the CNS
The diagnosis of GACNS is derived from the pathologic appearance of PACNS; thus this subset of PACNS is diagnosed by pathologic findings of granulomatous vasculitis on brain tissue. Clinically the disease appears to be male predominant and occurs at any age. It is characterised by a long prodromal period, with few patients presenting acutely. Signs and symptoms of systemic vasculitis such as peripheral neuropathy, fever, weight loss or rash are usually lacking. The disorder is usually suspected when stroke, more often recurrent, occurs in young patients with no identifiable cardiovascular or hypercoagulable risk factors, or in the setting of chronic meningitis, recurrent focal neurologic symptoms or unexplained diffuse neurologic dysfunction not associated with systemic disease or any other process. Table 2 summarise the signs and symptoms of GACNS.
|
Pathologic findings of GACNS
GACNS is diagnosed by the pathologic finding of granulomatous angiitis affecting the small and medium leptomeningeal and cortical arteries . Findings of Langhans or foreign body giant cells, necrotizing vasculitis or a lymphocytic vasculitis are usually present. The classic finding of granulomatous segmental vasculitis with Langhans or foreign body giant cells is present in less than 50% of the biopsies . The inflammation causes the vessels to become narrowed, occluded and thrombosed, causing tissue ischaemia and necrosis of the territories of the involved vessels. Skipped lesions are common. Because of the focal and segmental distribution of CNS vasculitis, a positive biopsy is diagnostic for the disease demonstrated, but a single isolated negative biopsy does not necessarily exclude primary or secondary CNS vasculitis .
Atypical CNS vasculitis
Atypical cases of PACNS are the most common among PACNS patients. This subset does not fit the diagnostic criteria for either GACNS or its major mimics including RCVS, yet the diagnosis of PACNS is suspected by angiographic or histopathologic evidence. It cannot be overemphasised that these cases should only be diagnosed as CNS vasculitis after a thorough evaluation that excludes any of the imitating entities discussed later in this article. Included in the atypical PACNS group are patients with angiographic findings suggestive of CNS vasculitis but with abnormal cerebrospinal fluid (CSF) findings that preclude a diagnosis of RCVS, or patients with GACNS-like presentation but without granulomatous features on CNS biopsies. It should be noted that CNS vasculitis should not be diagnosed only on the basis of an abnormal angiogram. CSF analysis is often abnormal in this category.
In addition, patients presenting with PACNS at unusual anatomic sites such as the spinal cord or those presenting with mass lesions are included in this category. Five percent of cases of PACNS may present as a solitary mass lesion . These lesions are frequently misdiagnosed as a brain tumor or abscess. The diagnosis of a PACNS mass lesion is performed by histologic evidence of vasculitis, with careful exclusion of infection and malignancy; granulomatous or lymphocytic vasculitis may be seen. While excision of the lesion may be curative, aggressive immunosuppressive therapy with corticosteroids and cyclophosphamide is associated with better outcomes .
- •
PACNS is a vasculitis confined to the CNS, which affects small- and medium-sized cerebral blood vessels.
- •
PACNS is referred to as granulomatous angiitis of the CNS (GACNS) when granulomatous vasculitis is confirmed on brain biopsy.
- •
Atypical PACNS cases are diagnosed based on angiographic or non- granulomatous histopathologic evidence of CNS vasculitis, in addition to abnormalities on CSF and exclusion of the imitating entities.
How to make the diagnosis of PACNS
Correct diagnosis is the key to achieving treatment success in any disease state. In PACNS there is no precise test or marker that is specific to the disease. Thus establishing the diagnosis depends on high clinical suspicion coupled with thorough investigations to exclude the mimics in patients with qualified pathologic or angiographic findings.
Laboratory studies
Acute-phase reactants, such as sedimentation rate and C-reactive protein, are usually normal in patients with PACNS. If these serum markers of inflammation are elevated, other forms of CNS vasculitis should be considered. These include CNS vasculitis secondary to connective tissue diseases, systemic vasculitides or infections. In these cases, testing should proceed accordingly and clues from the history, physical exam and the epidemiologic risk factors should be incorporated in directing the diagnostic work-up. Different serologic tests and biomarkers should be performed to exclude other disease entities. The selection of these serologic tests should be performed in the right clinical setting for accurate interpretation of the laboratory tests. These include antinuclear antibodies (ANAs), rheumatoid factor, antibodies to the Ro/SSA, La/SSB, Sm and RNP antigens, antibodies to double-stranded DNA, antineutrophil cytoplasmic antibodies (ANCAs), serum C3 and C4, serum cryoglobulins, serum and urine protein electrophoresis with immune electrophoresis and hypercoagulable profile to include anticardiolipin and antiphospholipid antibodies. Infectious work-up may include testing for human immunodeficiency virus (HIV), syphilis, mycobacterium, fungi and Lyme disease.
- •
Laboratory testing for PACNS is non-specific.
- •
Elevation of acute-phase reactants indicates a systemic process rather than PACNS.
- •
Laboratory testing and serologies are used primarily to exclude other aetiologies of CNS dysfunction.
Cerebrospinal fluid analysis
Cerebrospinal fluid analysis is a crucial part of the work-up of patients with suspected PACNS. Although the findings are non-specific in PACNS, CSF analysis should be the first step performed to exclude infectious or malignant diseases. Abnormal CSF findings occur in 80–90% of pathologically documented cases of PACNS. Findings usually reflect aseptic meningitis, with modest pleocytosis, normal glucose, elevated protein levels and occasionally the presence of oligoclonal bands and elevated IgG synthesis. The importance of the CSF analysis lies in the exclusion of any infectious aetiologies, especially in patients presenting with chronic meningitis-like pictures. The CSF analysis becomes extremely significant, especially in the setting of highly abnormal cerebral angiogram. Angiographic studies cannot distinguish PACNS from RCVS; however, in contrast to the abnormal CSF in patients with PACNS, the CSF profiles in patients with RCVS are typically normal unless there is an associated cortical haemorrhage. The combination of normal CSF analysis and highly abnormal cerebral angiogram with involvement of multiple vascular beds should dramatically increase the suspicion for RCVS.
- •
CSF analysis is a crucial part of the diagnostic work-up of PACNS.
- •
Findings are non-specific and include pleocytosis and elevated protein.
- •
Adequate stains and cultures should be performed to rule out infectious or malignant processes.
Neuroimaging modalities
The neuroradiologic findings of CNS vasculitis should be interpreted in recognition of the capabilities and limitations of specific imaging methods, and the assimilation of the imaging features with clinical and laboratory findings to exploit the correct diagnosis. The findings on neuroimaging studies such as computed tomography (CT) and magnetic resonance imaging (MRI) are not specific for CNS vasculitis regardless of whether it is primary or secondary. MRI is a more sensitive diagnostic imaging technique than CT, except when cerebral haemorrhage is suspected. The sensitivities of CT and MRI in CNS vasculitis diagnosed by positive angiogram are estimated 65% and 75%, respectively .
Magnetic resonance imaging
MRI findings include multiple and often bilateral infarcts in the cortex, deep white matter or leptomeninges, with or without contrast enhancement . The MRI findings of vascular involvement of the CNS associated with systemic diseases were recently reviewed in 24 patients with a diagnosis of secondary CNS vasculitis. Underlying causes for CNS vasculitis included systemic lupus erythematosis, tuberculosis, bacterial meningitis, Takayasu’s arteritis, polyarteritis nodosa, syphilis, drug abuse, yellow fever and Varicella. Non-specific high-intensity T2WI/FLAIR lesions in white matter were the most common finding, present in 42% of patients. Thirty-three percent had infarctions in the territory of large cerebral arteries, sometimes, but not always, associated with high-intensity T2WI/FLAIR small foci .
A finding of T2-hyperintense foci in the MRI can pose serious diagnostic challenges, giving the low specificity of such a finding, and should be interpreted carefully. These lesions should not simply be attributed to CNS vasculitis; they must be analysed in consideration with other clinical data and other diagnostic test results. MRI findings of T2-hyperintense foci can be associated with conditions or disease states other than CNS vasculitis: widened perivascular spaces, brain ageing, migraine, ischaemic changes, hypoxic–ischaemic changes, haemorrhagic foci, inflammatory changes, multiple sclerosis, central pontine and extrapontine myelinolysis, metastases, changes caused by radio-and chemotherapy, lesions associated with neurometabolic diseases, CNS degenerative diseases and eclampsia . The interpretation of the MRI findings should be performed by an expert neuroradiologist who is familiar with the findings of CNS vasculitis and its radiologic mimics.
The sensitivity of the MRI varies depending on the gold standard applied for the diagnosis of PACNS. Thus, the available data from the literature should be interpreted in the context of the poor specificity of the angiographic findings, especially when this is used as the modality for the diagnosis of PACNS. Several series of PACNS cases included patients who qualify for the diagnosis of RCVS, creating a challenge in estimating the true sensitivities and specificities of neuroimaging.
The sensitivity of MRI in GACNS approaches 100% . By contrast, normal brain MRI findings are not infrequent in RCVS, the most common mimics of PACNS. However, parenchymal haemorrhages and small non-aneurysmal subarachnoid haemorrhages overlying the cortical surface are very common in RCVS and should raise the possibility of RCVS in the right clinical settings.
- •
Non-specific high-intensity T2WI/FLAIR lesions in white matter are the most common finding in PACNS; however, these are not specific for the diagnosis.
- •
In the absence of compatible CSF and MRI findings, the diagnosis of PACNS will be less likely.
Cerebral vascular studies
Magnetic resonance angiography
Although MR angiography (MRA) is a less invasive study than cerebral angiogram, its resolution remains inadequate for the demonstration of vasculitic changes in the small vessels usually affected in PACNS. MRA studies are generally of little use in the evaluation of PACNS.
Cerebral angiography
The low rate of complications in cerebral angiography and the advances in this technique have led to its ready use in the diagnosis of cerebral vasculitis . The typical findings of vasculitis in cerebral angiography include alternating areas of stenosis and dilatation. Other findings include irregularities and stenosis. In contrast to systemic vasculitides, cerebral micro-aneurysms are rare and their presence in the cerebral circulation should raise the suspicion of septic aetiology. While the findings of ectasia and dilatation are characteristic of cerebral arteritis, they are not specific, and are commonly encountered in non-vasculitic disorders such as atherosclerosis, infection and vasospasm . The specificity of cerebral angiogram in the diagnosis of PACNS has been found to be as low as 30% . More recently the sensitivity and specificity of cerebral angiography for the diagnosis of PACNS were re-appraised. Thirty-eight patients with possible PACNS who had undergone cerebral angiography followed by cortical and leptomeningeal biopsies were analysed by Kadkhodayan et al. . Angiography reports were reviewed by investigators blinded to the biopsy results. Fourteen patients had typical angiographic findings of vasculitis. None had primary angiitis of the CNS at brain biopsy (60% specificity). Primary angiitis of the CNS was found at biopsy in only two of the remaining 24 patients. In this series, patients suspected of having primary angiitis of the CNS on the basis of clinical and angiographic findings did not have primary angiitis of the CNS at biopsy. This series highlighted the importance of reliance not only on the angiographic findings in the diagnosis of PACNS but also the need of incorporating the clinical picture, CSF and pathologic findings in controversial cases .
Angiography is insensitive as a screening test for patients with GACNS. Vollmer et al. reviewed the published literature of all GACNS cases not associated with Hodgkin’s disease, herpes zoster, sarcoidosis or other proximate cause and included only those cases that had been evaluated with angiography or other neuroimaging techniques before death or biopsy, and included their own experience. They concluded that 56% of angiograms performed in patients with histologically proven GACNS were abnormal, but only 27% were diagnostic for vasculitis . The low sensitivity of the cerebral angiogram is derived from the poor spatial resolution of the digital subtraction for defining small-vessel diseases, which are the primary vascular beds affected in GACNS.
Interpreting the angiographic results should be carried out with caution. The aetiological value of radiological examinations is poor, and in most cases the morphology and distribution of the lesions do not point to any specific origin except in rare incidence such as herpes zoster angiitis . In patients with RCVS, the most common mimic of PACNS, the typical characteristic findings on angiogram are those of stenosis and ectasia, which are referred to as ‘beading’ in multiple vascular beds ( Fig. 1 ). This can erroneously lead to misdiagnosis of CNS vasculitis when considering the angiographic findings alone. It cannot be overemphasised that the diagnosis of PACNS should not be based on the findings of high-probability angiogram alone. The angiographic findings should be associated with an appropriate and high-probability clinical setting along with inflammatory changes in the CSF with all appropriate exclusions for the consideration of a PACNS diagnosis.
- •
Abnormalities of cerebral angiography in PACNS include beading, circumferential or eccentric luminal narrowing, occlusions of one or more arteries and/or a vascular mass effect.
- •
The sensitivity and the specificity of cerebral angiography in the diagnosis of PACNS are poor.
- •
A negative angiogram cannot be used to exclude the diagnosis of PACNS.
- •
A positive angiogram is not diagnostic of PACNS.
Emerging neuroimaging modalities
The recent advances in imaging such as functional diffusion-weighted imaging (DWI) may play an important role in the evaluation of CNS vasculitis by demonstrating pathology on the microcirculatory level. A positive finding of microvascular ischaemia may assist in the diagnosis of CNS vasculitis. More importantly, the demonstration of normal microcirculation may reliably exclude CNS vasculitis . DWI detects small and active ischaemic changes not visible on conventional MRI and may be useful in assessing treatments and patient outcomes . Further studies on a larger scale are needed to better evaluate the operating characteristics of this technique in the diagnosis of CNS vasculopathies. The role of this technique in CNS vasculitis is still in its embryonic state.
Brain biopsy
Pathologic evaluation of the CNS often provides additional information in the diagnosis of possible PACNS. Nonetheless, brain biopsies are an invasive procedure, and most physicians and patients are hesitant to consider it despite its low complication rate. The possibility of biopsy is usually entertained when other non-invasive modalities have failed to identify the diagnosis. It should be noted that brain biopsy is limited by its low sensitivity; the presence of a negative biopsy does not rule out the diagnosis of PACNS. False-negative biopsies can be as high as 25% of autopsy-documented cases . Despite the relatively low sensitivity of the brain biopsy, the indication for this procedure is highly appreciated to identify mimicking lesions, in particular, infection or malignancy . These may be identified in up to 30–40% of selected patients. Brain biopsies should also be performed when there are MRI findings that suggest an inflammatory or infectious process, but non-invasive approaches to diagnosis have failed.
Open-wedge biopsy of a radiographically involved area, rather than random sampling of the non-dominant hemisphere , yields better sensitivity. Alternatively, if this approach is not amenable, directing the biopsy to the tip of the non-dominant temporal lobe with sampling of the overlying leptomeninges and underlying cortex, especially areas of leptomeningeal enhancement, may increase the sensitivity .
Finally, the presence of vasculitis in the biopsy specimen should not preclude performing special stains and cultures for occult infections that may produce secondary vasculitis. It cannot be emphasised enough that the role of CNS biopsies in most cases is to rule out an alternative diagnosis. Adequate stains, cultures and viral testing as well as histological studies and immunophenotyping are invaluable in the diagnosis of suspected PACNS.
- •
Brain biopsy is limited by its lack of sensitivity.
- •
The preferred choice is open-wedge biopsy with sampling of the overlying leptomeninges and underlying cortex.
- •
The preferred site is the radiographic involvement or, if non- accessible, the tip of the non-dominant temporal lobe.
- •
Biopsies are often performed to rule out other entities.
- •
Adequate stains and immunohistochemistry should be performed on the biopsies even with the findings of vasculitis.
How to make the diagnosis of PACNS
Correct diagnosis is the key to achieving treatment success in any disease state. In PACNS there is no precise test or marker that is specific to the disease. Thus establishing the diagnosis depends on high clinical suspicion coupled with thorough investigations to exclude the mimics in patients with qualified pathologic or angiographic findings.
Laboratory studies
Acute-phase reactants, such as sedimentation rate and C-reactive protein, are usually normal in patients with PACNS. If these serum markers of inflammation are elevated, other forms of CNS vasculitis should be considered. These include CNS vasculitis secondary to connective tissue diseases, systemic vasculitides or infections. In these cases, testing should proceed accordingly and clues from the history, physical exam and the epidemiologic risk factors should be incorporated in directing the diagnostic work-up. Different serologic tests and biomarkers should be performed to exclude other disease entities. The selection of these serologic tests should be performed in the right clinical setting for accurate interpretation of the laboratory tests. These include antinuclear antibodies (ANAs), rheumatoid factor, antibodies to the Ro/SSA, La/SSB, Sm and RNP antigens, antibodies to double-stranded DNA, antineutrophil cytoplasmic antibodies (ANCAs), serum C3 and C4, serum cryoglobulins, serum and urine protein electrophoresis with immune electrophoresis and hypercoagulable profile to include anticardiolipin and antiphospholipid antibodies. Infectious work-up may include testing for human immunodeficiency virus (HIV), syphilis, mycobacterium, fungi and Lyme disease.
Related posts:
Inflammatory osteoarthritis of the hand
Management of osteoporosis in a pre-menopausal woman
Prevention and cure of rheumatoid arthritis: Is it possible?
Autoinflammatory conditions: when to suspect? How to treat?
Evidence-based management of rapidly progressing systemic sclerosis
Treatment-resistant inflammatory myopathy
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
