Central Nervous System Vasculitis

Systemic vasculitis

Henoch-Schönlein purpura, Kawasaki disease, granulomatosis with polyangiitis, microscopic polyangiitis, eosinophilic granulomatosis with polyangiitis, polyarteritis nodosa, Behçet’s disease

Systemic rheumatologic disease

Systemic lupus erythematosus (usually a vasculopathy), dermatomyositis, scleroderma, rheumatoid arthritis, Cogan’s, inflammatory bowel disease, sarcoidosis

21.4 Evaluation of Patients Suspected with CNS Vasculitis

The clinical presentation of CNS vasculitis is protean which often causes a delay in diagnosis. In our practice CNS vasculitis is primarily suspected after an unexplained neurologic deficit, stroke, or multi-strokes in younger-aged patients with no cardiovascular risk factors and when an abnormal cerebrovascular study emerges while working up a patient with neurologic abnormalities. In most instances, an alternative diagnosis is made giving the rarity of the CNS vasculitis.

Diagnostic criteria for PACNS were proposed in 1988 by Calabrese and Mallek (Table 21.2) [9]. Although not validated prospectively, these criteria remain the mainstay of the workup of patients with PACNS. These criteria emphasize the importance or ruling many mimics of the disease before making the diagnosis of PACNS.

Table 21.2

Diagnostic criteria for primary angiitis of the central nervous system

The presence of an acquired and otherwise unexplained neurologic deficit

• With presence of either classic angiographic or histopathologic features of angiitis within the CNS

• And no evidence of systemic vasculitis or any condition that could elicit the angiographic or pathologic features

21.4.1 Laboratory Testing

Laboratory testing in PACNS are normal giving the isolated nature of the disease. Inflammatory markers such as sedimentation rate and C-reactive protein are normal in PACNS. Abnormalities in inflammatory markers should prompt an evaluation for a systemic involvement whether inflammatory or infectious. Autoimmune serologies, such as antinuclear antibodies (ANA), rheumatoid factor (RF), and antineutrophilic cytoplasmic antibodies (ANCA), are also negative in PACNS. These serologic tests are often requested during the evaluation of CNS vasculitis, but should be indicated in the right clinical setting to assist in ruling in or refuting a systemic inflammatory disease diagnosis. Evaluation for an infectious process by blood cultures and/or serology and PCR, for suspected bacterial endocarditis or other infectious diseases. In patients who have unexplained multiple strokes, coagulation studies and antiphospholipid testing are essential for evaluation. Other laboratory testing should be tailored according to the individual exposures and risk factors of the patient.

21.4.2 Cerebrospinal Fluid Studies

Cerebrospinal fluid findings are abnormal in 81–90% of cases of CNS vasculitis; in a cohort of 101 patients, 88% of the CSF tests showed at least one abnormality, with the majority of cases being a mildly increased protein, white blood cells (WBC), or both [2, 3, 10]. In the combined series published by Calabrese et al., the average CSF white blood cell count was 60 cells/high-power field (hpf), with the highest reported value of 330 cells/hpf. The average protein level was 118 mg/dL, with a maximum value seen of 825 mg/dL [9]. Other series have reported lower values with medium CSF white blood cell count of 5 cells/μL (range of 0–535 cells/μL) and a median total protein of 0.7 g/dL (range of 0.15–1.03 g/dL) [2]. This discrepancy of CSF findings among the cohorts could be related to the contamination of some cohorts with cases of RCVS, thus making the amount of pleocytosis seen in CNS vasculitis CSF samples artificially lower. CSF is a vital component of the workup for CNS vasculitis, but it is important to note that abnormalities seen in the CSF are not specific for CNS vasculitis, and it’s critical that other causes of CSF pleocytosis such as infection be ruled out.

21.4.3 Magnetic Resonance Imaging

Magnetic resonance imaging (MRI) is a very sensitive modality in the diagnosis of CNS vasculitis. In the Mayo cohort of patients, 97% of patients with CNS vasculitis had MRI abnormalities with ischemic infarcts being the most common (53%). The majority of MRI abnormalities follow a supratentorial distribution. Infratentorial lesions in the absence of supratentorial lesions are extremely rare. Radiographic lesions are distributed evenly in subcortical white matter, cortical gray matter, deep white matter, and cerebellum (Fig. 21.1a). One third of the cases can have gadolinium enhancement (Fig. 21.1b). CNS vasculitis can also present with a mass lesions found on MRI mimicking a neoplasm in upwards of 5–15% [2, 3]. MRI has the highest sensitivity compared to any other imaging modality and should be the neuroimaging of choice in the first step in evaluating a patient with CNS vasculitis. Importantly these lesions are not specific for the diagnosis of CNS vasculitis and can be seen in other neurologic disease such as demyelinating or ischemic disorders.

Fig. 21.1

Imaging of patients with primary CNS vasculitis. (a) Cerebral angiogram shows alternating stenosis and dilatation of the distal middle cerebral artery (arrows) and the anterior cerebral artery (arrowheads). (b) Magnetic resonance angiography of the brain shows a short segment stenosis of the anterior cerebral artery (green arrow) and stenosis of the distal middle cerebral artery (white arrow). (c) Fluid-attenuated inversion recovery (FLAIR)-weighted MRI shows a large abnormality within the right cerebral hemisphere consistent with ischemia (arrowheads). (d) MRI shows diff use, asymmetric, nodular, and linear leptomeningeal enhancement, with dura only slightly affected. Reproduced by permission of Lancet (Salvarani et al. Lancet 2012;380:767–777)

21.4.4 Cerebral Vascular Studies

Cerebral angiography is part of the diagnostic criteria proposed by Calabrese and Mallek (Table 21.2). The classic finding of CNS vasculitis is either vascular narrowing or focal dilations, which can alternate causing the appearance of a “string of beads” that are typically bilateral. The middle cerebral artery is affected the most, with an average of 2.3 lesions per patient compared to the anterior cerebral artery of 0.2 lesions per patient and the posterior cerebral artery of 0.7 lesions per patient [10] (Fig. 21.1c). It is important to understand that angiography provides information on the contour and the lumen of the vessels without providing the underlying cause of the abnormality thus limiting its specificity to as low as 30% for the diagnosis of CNS vasculitis [11]. Multiple noninflammatory etiologies can cause the classic findings of CNS vasculitis such as infection, vasospasm, atherosclerosis, fibro muscular dysplasia, malignancies, or neurofibromatosis.

Cerebral angiography is the most invasive of imaging studies, but up to this point provides the most sensitive spatial and temporal resolution of the vessel anatomy. Angiography can be falsely negative, as the imaging modality resolution is limited to vessels larger than 0.2 mm; thus small-vessel vasculitis can be missed. In various cohorts of CNS vasculitis, the sensitivity of angiography ranges between 50 and 90% [2, 3]. Other vascular modalities such as magnetic resonance angiography (MRA) and computed angiography (CTA) have much lower spatial resolution compared to cerebral angiography and are more useful in large- and medium-sized vessels (Fig. 21.1d). Other vascular modalities such as high-resolution MRI (HR-MRI) have emerged in the field of CNS vasculitis. HR-MRI focuses on the vessel wall and provides more information than lumenography and can be helpful in differentiating patients with CNS vasculitis and RCVS.

21.4.5 Brain Biopsy

Although biopsy is not mandated for the diagnosis of CNS vasculitis, it is extremely important to consider this diagnostic modality in each case as it is the only definitive way of making the diagnosis. Brain biopsy not only ensures the diagnosis of CNS vasculitis but it can reveal an alternative diagnosis. A meta-analysis reported in 2015 revealed a diagnostic yield of brain biopsy of 74.7% for suspected PACNS [12]. In this report a brain biopsy for suspected PACNS had the highest yield compared to other indications. Furthermore, alternative diagnosis can be found in 30% to 39% of the biopsies [13]; these include cerebral amyloid angiopathy, viral meningoencephalitis, CNS lymphoma, demyelination, and other miscellaneous diagnoses such as posterior reversible encephalopathy syndrome, progressive multifocal leukoencephalopathy, and Alzheimer’s disease. These studies highlight the importance of histologic confirmation of the diagnosis which can reveal an alternative diagnosis in many cases. False negatives are possible and have been reported as high as 47% likely secondary to the skipped lesion nature of the vasculitic process and inaccessible locations for biopsy [11]. It is not clear if there is a difference in diagnostic yield between the different surgical techniques obtaining the biopsy. In the retrospective study of 29 biopsy samples, there was no difference in biopsy yield between open and closed technique [13].

Three histologic patterns are generally recognized in CNS vasculitis: granulomatous, lymphocytic, and necrotizing. Granulomatous angiitis is characterized by destructive mononuclear inflammation with well-formed granulomas and multinucleated giant cells. Granulomatous angiitis usually affects the small and medium leptomeningeal and cortical arteries [14]. The inflamed vessels cause ischemia and stroke once they become narrowed, occluded, and thrombosed preventing flow to their vascular territory [14] (Fig. 21.2a). Lymphocytic inflammation has similar findings to granulomatous, except a lymphocytic predominant infiltrate instead of a mononuclear infiltrate, and without associated granulomas (Fig. 21.2b). Lymphocytic infiltrate can also be seen in malignancy and infections. In these circumstances a more diligent workup should be performed on the pathologic tissue to rule out an infectious or malignant process. Necrotizing vasculitis is characterized by necrotizing vasculitis with transmural fibrinoid necrosis. Necrotizing vasculitis usually involves small muscular arteries with disruption of the internal elastic lamina (Fig. 21.2c).

Fig. 21.2

Histopathological features of primary CNS vasculitis. (a) Granulomatous pattern of primary CNS vasculitis. Left-hand image shows transmural inflammation of a leptomeningeal artery with prominent mononuclear (bracket) and granulomatous (arrow) adventitial inflammation and focal fibrin thrombus formation (asterisk; hematoxylin and eosin [H&E] stain). Inset picture on right shows noticeable thickening and luminal obliteration of several leptomeningeal vessels (H&E stain). The right-hand image shows focal collections of epithelioid histiocytes arranged in granuloma-like aggregates. Where the lumen is preserved, the vessel wall is thickened by an amorphous eosinophilic material (amyloid). Ischemic neurons can be seen in the adjacent parenchyma (H&E stain). (b) Granulomatous pattern with amyloid angiopathy in primary CNS vasculitis. Left-hand image shows destructive vasculitis with well-formed granulomas in leptomeningeal vessels (arrows) and wall thickening with eosinophilic material (asterisk; H&E stain). The right-hand image shows amyloid-β deposits in all vessels (immunoperoxidase stain for βA4 amyloid). (C) Lymphocytic pattern of primary CNS vasculitis. Left-hand and right-hand images show substantial thickening and luminal obliteration (asterisks mark lumen remnant) of several leptomeningeal vessels. The infiltrate is predominated by lymphocytes, but has few histiocytes and granulocytes. Granuloma-like features are not seen (H&E stain). (d) Necrotizing pattern of primary CNS vasculitis. Left-hand image shows a small leptomeningeal artery with transmural acute inflammation (H&E stain). Right-hand image shows segmental transmural fibrinoid necrosis (asterisk), which displays as red-staining material in the vessel wall (Masson’s trichrome). Hemorrhage and acute infarction are evident in the underlying cortical parenchyma (right-hand image, bottom). Reproduced by permission of Lancet (Salvarani et al. Lancet 2012;380:767–777)

Only gold members can continue reading. Log In or Register to continue