General Neurology
Brain death/death by neurologic criteria
Nov. 09, 2024
MedLink®, LLC
3525 Del Mar Heights Rd, Ste 304
San Diego, CA 92130-2122
Toll Free (U.S. + Canada): 800-452-2400
US Number: +1-619-640-4660
Support: service@medlink.com
Editor: editor@medlink.com
ISSN: 2831-9125
Toll Free (U.S. + Canada): 800-452-2400
US Number: +1-619-640-4660
Support: service@medlink.com
Editor: editor@medlink.com
ISSN: 2831-9125
Worddefinition
At vero eos et accusamus et iusto odio dignissimos ducimus qui blanditiis praesentium voluptatum deleniti atque corrupti quos dolores et quas.
Diffuse inflammation of small- and medium-sized blood vessels confined exclusively to the brain, meninges, or spinal cord is called primary angiitis of the central nervous system (PACNS). In the last few years, there have been significant advances in the understanding of this disorder due to the identification of pathological and clinical subsets, along with advances in the imaging modality of vessel wall abnormalities. A considerable gap still exists, however, particularly with respect to its pathophysiological mechanisms. Although various laboratory and neuroimaging findings may support the diagnosis of primary CNS angiitis, they are not highly specific, and neurologists often face a substantial challenge when diagnosing this disorder because they need to rule out a vast array of nonvasculitic conditions and secondary causes of central nervous system vasculitis. Therefore, a correct and timely diagnosis of primary CNS angiitis requires a high degree of suspicion coupled with knowledge of other diseases that can masquerade as primary CNS vasculitis. In this article, the authors focus on the clinical and pathological findings, diagnostic work-up, differential diagnosis, and current therapeutic options of this still enigmatic and complex entity.
• Primary angiitis of the CNS is an uncommon disease with protean clinical manifestations that often make diagnosis and treatment very challenging. | |
• The current diagnostic approach is characterized by a low sensibility and specificity; hence, diagnosis relies on a high degree of clinical awareness and thorough investigation aimed at ruling out a broad list of alternative diagnoses and mimickers. | |
• Brain biopsy is currently the gold standard for definite diagnosis of primary angiitis of the CNS (PACNS). | |
• Reversible cerebral vasoconstriction syndrome is the most common mimic of primary CNS angiitis. | |
• Immunosuppressive therapy with glucocorticoids alone or in combination with cyclophosphamide is the first line of treatment, despite lack of evidence based on controlled clinical trials. |
Vasculitis or angiitis is defined as an inflammatory disease of arteries, veins, or both that results in histologically demonstrable structural injuries to the vessel wall, often accompanied by thrombosis and evidence of ischemic damage to the tissues served by the affected blood vessels. A vasculitis is considered primary when it occurs without an identifiable cause or when it is unassociated with an underlying disease and secondary when it occurs as a manifestation of a diverse group of underlying diseases.
The central nervous system vasculature can be affected secondarily by numerous forms of vasculitis such as generalized autoimmune diseases (systemic lupus erythematosus, Sjögren syndrome) or systemic vasculitis (Wegener granulomatosis, polyarteritis nodosa) as well as by drugs (amphetamines, cocaine), infections (bacterial, fungal, protozoal, mycoplasmal, rickettsial, viral), malignancy (Hodgkin lymphoma, non-Hodgkin lymphoma, leukemia, lung cancer), Behçet disease, Cogan syndrome, and sarcoidosis (96).
Primary CNS vasculitis was first described by Harbitz in 1922, but it was not until 1959 that Cravioto and Feigin introduced the concept of vasculitis with a unique predilection for the CNS and coined the term "noninfectious granulomatous angiitis with a predilection for the nervous system” (64; 31). In their classical description, they explained two cases from their own series and also gave a literature review of suspected cases including one case from Harbitz (64), two cases from Newman and Wolf (101), and two from McCormick and Neubuerger (89). They questioned the diagnosis of Churg-Strauss syndrome in Newman and Wolf's cases and also temporal arteritis in McCormick and Neuberger's cases. They suggested that along with their own two cases, these six cases should be grouped together as delineating a distinct clinical and pathological entity. They also suggested the possible utility of biopsy in establishing the diagnosis and cortisone/ACTH in therapy. Another case was added in the literature in 1966 by Hughes and Brownell (69).
By the 1970s, this disease was considered a severe, uniformly fatal disorder and unresponsive to therapy with few notable exceptions like the case reported by Snyder and McClelland in 1978 (138). In all published case reports until that time, the entity had been recognized at postmortem examinations. Cupps and colleagues described the efficacy of cyclophosphamide and high-dose glucocorticoids, with the idea driven by the successful use of these agents in systemic angiitic syndromes (33). Sigal in 1987 wrote an excellent review paper shedding light on the neurologic manifestations of vasculitic and rheumatological syndromes (136). Fauci and colleagues in 1979 and then Calabrese and Mallek in 1988 proposed the diagnostic criteria that relied equally on histopathology or, alternatively, cerebral angiography in the appropriate clinical setting (46; 24). These criteria include:
(1) An unexplained neurologic deficit despite aggressive diagnostic evaluation. | |
(2) A high probability angiogram for arteritis or histopathological evidence of arteritis confined to the CNS. | |
(3) Exclusion of all those disorders capable of mimicking the angiographic findings or associated with vascular inflammation of the CNS. |
Over time it became increasingly common for patients to be diagnosed with primary angiitis of the CNS solely on angiographic grounds without supporting histopathology. As a result, patients, diagnosed solely on the basis of abnormal angiogram, were treated with prolonged and intensive immunosuppressive regimens based on the supposition that cases diagnosed by angiographic findings were clinically equivalent to biopsy-proven cases (150).
The achievements of 1980s proved to be lessons of the 1990s. First, enthusiasm for the empiric treatment of cerebral vasculitis waned because of recognition of the unreliability of cerebral angiography in its diagnosis. A second factor that lessened interest in empiric therapy with cyclophosphamide for CNS vasculitis was the recognition of permanent side effects in up to 40% of patients treated with oral cyclophosphamide for Wegener granulomatosis. Calabrese and colleagues described so-called benign angiopathy of the CNS among young women with the prior diagnosis of primary angiitis of the CNS (23). They differed in the onset with a focal cerebral deficit, normal CSF, lack of progression, and spontaneous resolution. They proposed primary angiitis of the CNS as a heterogeneous disease with granulomatous angiitis of the CNS and benign angiopathy of the CNS, and with atypical forms as clinical subsets. Hajj-Ali and coworkers provided evidence that the pathologic process in benign angiopathy of the CNS is vasoconstriction rather than vasculitis, thus, indicating that benign angiopathy of the CNS should not be included in the spectrum of primary angiitis of the CNS (59). Reversible cerebral vasoconstriction is the term now used to describe entities such as postpartum angiopathy, Call-Fleming syndrome (25), benign angiopathy of the CNS, and others. In 2005, MacLaren and colleagues explained that the clinical course of patients with primary angiitis of the CNS differed markedly depending on the size of the vessel involved (87). They proposed two subsets of primary angiitis of the CNS: small-vessel disease and medium-vessel disease.
But the debate still continues as these classifications continue to evolve. Calabrese rightly noted: "A penumbra is a space of partial illumination, such as in eclipse, between perfect shadow and full light. Despite significant progress, primary angiitis of the CNS remains foursquare in the penumbra awaiting full illumination" (19). The knowledge about this intriguing disease continues to expand, and excellent reviews were done in the last decades (154; 152; 151; 61).
Vasculitis predominantly affecting the CNS has been given several diagnostic labels in the past 90 years. Harbitz first described it as "unknown forms of arteritis" (64). The same condition was later described as, "non infectious granulomatous angiitis involving the central nervous system" (101), as "giant cell arteritis involving small meningeal and intracerebral vessels" (89), as " noninfectious granulomatous angiitis with a predilection for the nervous system" (31), as" granulomatous angiitis of the nervous system" (16) or simply "granulomatous angiitis" (17), and as "isolated angiitis of CNS" (33; 95). The nomenclature of this unique disease entity continues to evolve, and terms such as granulomatous angiitis of the CNS and isolated angiitis of the CNS fall out of use as neither is appropriate because the histology can be granulomatous or lymphocytic and subclinical involvement of extracranial arteries (eg, pulmonary and abdominal visceral angiitis) have been reported. Granulomatous pathology in cerebral vessels is highly nonspecific as it had been shown to be associated with various disease entities like temporal giant-cell arteritis (72), herpes zoster (113; 67), lymphoproliferative tumors (109; 73), sarcoidosis (26), amyloidal angiopathy (50), and systemic lupus erythematosus (129). The final diagnostic label of "primary angiitis of the central nervous system" proposed by Calabrese and Mallek in 1988 and endorsed by Lie in 1991 and 1992 is the most appropriate and correct one according to our current understanding of this unique disease (24; 83; 84).
Inflammatory processes may affect any area of the CNS. Thus, virtually every neurologic sign or symptom has been reported at least once in relation to primary CNS angiitis (21).
Nonfatal symptoms such as headaches and confusion are the most common presentation, with hemiparesis being the most common sign. Primary CNS angiitis is characterized by subacute or chronic insidious-onset headache, whereas single or recurrent thunderclap headache usually reveals reversible cerebral vasoconstriction syndrome (137; 62). Multiple and recurrent strokes or transient ischemic attacks, insidious encephalopathy, and cognitive impairment are also common hints of primary CNS angiitis. Movement disorders like chorea and myoclonus, radiculopathy, optic and cranial neuropathies, seizures, and ataxia have been reported, though infrequently (53). Nonspecific visual complaints occur in approximately 15% of patients. Signs and symptoms of systemic vasculitis, such as fever, weight loss, peripheral neuropathy, or rash are absent (62).
Scolding and colleagues defined three broad categories of clinical presentation (133):
(1) Acute or subacute encephalopathy, commonly presenting as confusional state, progressing to drowsiness and coma if left untreated. | |
(2) Atypical multiple sclerosis in phenotype, with a relapsing-remitting course and features such as optic neuropathy and brainstem episodes but also accompanied by other features less common in multiple sclerosis such as seizures, severe and persistent headaches, encephalopathic episodes, or hemispheric stroke-like episodes. | |
(3) Intracranial mass lesions with headaches, drowsiness, focal signs, and, often, raised intracranial pressure. |
A meta-analysis of 911 patients diagnosed with primary angiitis of the CNS (43% confirmed by angiography, 41% confirmed by biopsy, and the diagnostic procedure was unclear in the remainder) enrolled in 46 cohort studies between 1988 and 2020 has been reported (09). Focal neurologic deficit (63%), headache (51%), and cognitive impairment (41%) were the most common clinical manifestations at onset. Compared to patients with angiographic diagnoses, those with a biopsy diagnosis were more likely to show cognitive impairment and seizures and less likely to manifest focal neurologic signs (09).
Along with classic granulomatous angiitis of the CNS (which accounts for 20% of all patients with primary CNS angiitis), in the last few years lymphocytic vasculitis, mass-lesion presentation, amyloid-beta-related cerebral angiitis, and angiographically defined primary angiitis of the CNS have also been considered subtypes of primary angiitis of the CNS (61; 62).
Moreover, based on clinical findings from a retrospective cohort study initially involving 101 patients and later updated to include 131 patients (all treated at a single center over a 25-year period), Salvarani and colleagues have suggested that primary angiitis of the CNS is a highly variable syndrome that may consist of several subsets of heterogeneous diseases, and it involves a catastrophic rapidly progressive clinical course, cerebral amyloid angiitis, intracranial hemorrhage, solitary tumor-like mass lesion, spinal cord abnormalities, prominent leptomeningeal enhancement on brain MRI imaging, angiographically-negative/biopsy-positive primary CNS vasculitis, and a benign outcome (114; 120; 123; 124; 54).
Thus, the presumably different subtypes of primary angiitis of the CNS so far proposed need to be definitely delineated based on epidemiological, clinical, histological, and neuroimaging findings in order to outline optimal treatment and establish outcomes.
Twenty percent of patients will have signs or symptoms that fluctuate early in the course of disease. This can be explained by the size of the vessels involved and by the fact that the fundamental pathophysiological mechanism in all angiitides is tissue ischemia; it is not surprising that early manifestations of the disorder might fluctuate in intensity or even spontaneously improve. Although the classic picture is one of progressive, cumulative, and multifocal neurologic dysfunction, there are abundant exceptions, including patients whose presentations suggest any of the following:
• cerebral tumor (144) |
Stroke occurs frequently in the setting of primary CNS angiitis, often manifesting as recurrent cerebral ischemia or transient ischemic attack involving several vascular territories, with associated inflammatory changes in the cerebrospinal fluid; however, stroke is rarely the initial manifestation of primary CNS angiitis (61). In around 12% to 15% of patients, parenchymal hemorrhage (most often in the frontal lobe) and, less commonly, subarachnoid hemorrhage (restricted to the brain convexity) may occur at or around the time of primary angiitis of the CNS diagnosis (10; 119). These patients usually have less-impaired cognition, a persistent neurologic deficit, and fewer brain infarctions on MRI; they need therapy at follow-up less frequently and are more prone to presenting with a necrotizing histopathological pattern than primary angiitis of the CNS patients without intracranial hemorrhage (119). On the other hand, primary angiitis of the CNS patients with intracranial hemorrhage may present with cerebral amyloid angiopathy, although vasculitis (granulomatous and necrotizing histopathological pattern) was also found in these patients, precluding reliable determination of the degree of contribution of beta-amyloid peptide vascular deposition in the development of intracranial bleeding, particularly in cases of necrotizing primary angiitis of the CNS (119). It rarely causes cerebral infarct (145) or transient ischemic attacks in the absence of clinical or laboratory evidence of a widespread CNS inflammatory disorder, such as CSF pleocytosis (24).
In a cohort of 131 patients with primary angiitis of the CNS, a subset of 11 (8%) patients with an average age of 58 was found to have a form of the disease that rapidly progressed after onset, with a median of 24 days between symptom onset and diagnosis (120). Cognitive dysfunction and persistent neurologic deficits occurred in over 50% of these 11 patients, 45% had headaches, and 36% had acute quadriparesis or paraparesis; one patient had simultaneous spinal involvement. Six patients had a median protein level above 105 mg/dl, and most showed multiple and bilateral infarctions involving both the cortex and subcortical regions on brain MRI. Cerebral angiography revealed multiple vessel abnormalities with involvement of both large- and small-vessel arteries. Pathological results (open brain biopsy or postmortem analysis) revealed granulomatous, necrotizing, or combined granulomatous and necrotizing pattern. A rapid cognitive decline whose main clinical manifestation was a frontal lobe dysexecutive syndrome mimicking a behavioral variant of frontotemporal dementia was reported in a 47-year-old man with lymphocytic pleocytosis on CSF, no vessel wall abnormalities on contrast-enhanced MRA, and perivascular infiltrates of inflammatory T-cells along with deposition of hemosiderin around the small vessels – both histopathological features suggestive of isolated small-vessel vasculitis (13).
De Boysson and colleagues compared the clinical findings and outcome of 26 patients with isolated small-vessel vasculitis of the central nervous system with those of 76 patients with large- and medium-vessel involvement (37). Patients with isolated small-vessel vasculitis were younger (median age: 41.5 years; range: 18 to 61) than patients with large- and medium-vessel vasculitis (median age: 48.5 years; range: 19 to 80). Seizures, dyskinesias, abnormal CSF, and relapse were significantly more frequent in patients with small-vessel vasculitis than in the other group. Cognitive deficits and impaired consciousness were also more frequent in patients with small-vessel involvement, though not significantly. Conversely, focal deficits were significantly less observed in patients with small-vessel vasculitis. MRI findings between the two groups also showed differences. On brain MRI, leptomeningeal or parenchymal gadolinium enhancement and tumor-like lesions were significantly more common in patients with small-vessel vasculitis, whereas fewer acute ischemic lesions were observed in this group. All patients with small-vessel vasculitis had an abnormal biopsy--a lymphocytic pattern being most frequent, followed by a granulomatous inflammatory pattern. The mortality rate was similar in both groups, and there were no significant differences in good functional outcome (37).
Isolated spinal cord vasculitis was initially reported by Feasby and colleagues in 1975 when they described a case presenting with subacute progressive cervical myelopathy with negative workup for secondary pathologies (47). Granulomatous angiitis involving the spinal cord is seen in about 14% of cases of primary angiitis of the CNS (20). A lower frequency (5%) of spinal cord involvement was found in 100 primary angiitis of the CNS patients (115). In this series, all patients had thoracic spinal cord abnormalities. Myelopathy was the initial manifestation of primary angiitis of the CNS in three patients; however, all these patients had concurrent or subsequent cerebral involvement. In rare cases, spinal cord involvement can be the only manifestation of primary angiitis of the CNS (56). Patients usually present with back pain, progressive myelopathy, and an elevated CSF protein level with pleocytosis and normal erythrocyte sedimentation rate. MRI is sensitive enough to reveal diffuse enlargement of the cord, but this finding is highly nonspecific as such enlargement may be associated with diffusely infiltrating astrocytoma, lymphoma, leukemia, metastasis, sarcoidosis (neurovascular form) (153), an infectious process, and cord ischemia (55). Therefore, the diagnosis should be confirmed by angiography or biopsy before starting immunosuppressive regimens.
As previously mentioned, primary angiitis of the CNS can clinically mimic an intracranial mass lesion (tumor-like presentation), with a frequency between 4% and 12% among biopsy-proven primary angiitis of the CNS. The median age at onset ranges from 30 to 66 years, with males predominating (94; 36; 126; 141; 146). The median time from symptom onset to PACNS diagnosis is around 1 month. The most common clinical manifestations are headache, seizures, cognitive dysfunction, and focal neurologic deficits (94; 36; 126; 141; 146). Confusion, aphasia, ataxia, and visual and neuropsychiatric symptoms are less common (75; 68; 126; 141; 146). Single and, less often, multiple asymmetric tumor-like lesions involve the cortical and subcortical areas or the deep white matter and, less commonly, the thalamus, basal ganglia, or the infratentorial region (126; 68; 141; 146). Vasogenic edema, mass effect, and a variable pattern of parenchymal and meningeal gadolinium enhancement are observed on neuroimaging. Macro- and microhemorrhages are common (94; 36; 126; 141; 146). In addition, cerebral infarcts may be found in about 30% of patients (36). Acute phase reactants (erythrocyte sedimentation rate and C reactive protein) are usually normal. CSF usually showed mild to moderate pleocytosis with elevated protein level (94; 126; 141). This subset of patients had fewer or no abnormal findings on both brain MR angiography and digital subtraction angiography, suggesting mainly small-vessel involvement (94; 36; 126; 141; 146). The most common histopathological patterns are granulomatous vasculitis along with beta-amyloid peptide vascular deposition and lymphocytic vasculitis (126; 141; 146). A favorable response to glucocorticoids and immunosuppressants is usual in all patients (126; 141). The relapse rate in this subset of patients ranges between 10% and 50%, with good functional outcome after treatment intensification (36; 126; 141).
The clinical findings, response to therapy, and outcome have been reported in a cohort of 78 patients with biopsy-proven inflammatory or noninflammatory cerebral vascular amyloid-beta angiopathy who were seen retrospectively at a tertiary center over a period of 25 years (127). In this cohort, nearly half of the patients showed evidence of inflammatory infiltrate; 36% (28) patients had cerebral amyloid angiopathy with transmural inflammation (most often granulomatous pattern); and 13% (10 patients) had cerebral amyloid angiopathy with mild nondestructive perivascular inflammation. The rest of the patients exhibited extensive cerebral amyloid-beta angiopathy and no inflammatory infiltrate (127).
These results were further compared with a retrospective cohort of 118 patients seen at the same tertiary center with a diagnosis of primary CNS angiitis without amyloid-beta deposition (127). Patients with a pattern of amyloid-beta-related angiitis were older, more likely to have altered cognition and seizures, and less likely to have hemiparesis and visual symptoms at presentation than patients with primary CNS angiitis. Radiological differences were also found between both groups. Intracerebral hemorrhage and leptomeningeal enhancement after administration of gadolinium were significantly higher in patients with amyloid-beta-related angiitis, whereas cerebral infarction was significantly less common in this group. In the CSF, only the protein levels were higher in patients with amyloid-beta-related angiitis. Finally, the response to treatment, duration of treatment, and outcomes were quite similar between both groups, although the relapse rate was nonsignificantly higher in patients with amyloid-beta-related angiitis compared to patients with primary CNS angiitis (127).
The results after comparison of these two groups suggest that cerebral amyloid angiopathy with transmural inflammation (true angiitis) is not a different nosological entity from primary CNS angiitis, but rather, conversely, another subset of primary CNS angiitis. In the 78-patient cohort, the number of patients with cerebral amyloid-related angiopathy was low, precluding any conclusion as to whether it constitutes a definable subset of primary CNS angiitis (127).
In a study of 168 patients with primary angiitis of CNS (PACNS), 10 cases (5.9%) also had a diagnosis of lymphoma, six had Hodgkin, and four had non-Hodgkin lymphomas (121). The PACNS diagnosis was established by biopsy or angiography in eight and two patients, respectively. The lymphoma diagnosis was established in seven patients during the investigation of PACNS and in two after this diagnosis. A granulomatous angiitis, found in seven patients, was the most common histopathological pattern; in two cases, it was associated with amyloid beta deposition in the vessel wall. Eight patients were male, with a mean age of 50.5 years at diagnosis; progressive cognitive decline and headache were the most common first symptoms at disease onset, and two patients had spinal cord involvement. The time between the first symptoms and the PACNS diagnosis was 4 months. Onconeural antibodies were negative in four patients; these antibodies were not investigated in the remaining patients. In addition, investigation of infectious vasculitis was negative in all patients. Patients with lymphoma and PACNS had more leptomeningeal contrast enhancement on initial brain MRI and a worse outcome at last follow-up (two patients had an aggressive and fatal course, and four patients had a severe disability according to modified Rankin score) compared with those PACNS patients with no lymphoma. Treatment response was similar between both groups. A paraneoplastic mechanism may explain the simultaneous diagnosis of both pathologies (121).
Primary angiitis of the CNS may be a devastating disease. The consequence of missing the diagnosis is the death of the patient; the consequence of delay in diagnosis is likely to be severe disability. The prognosis is more favorable with early recognition and prompt treatment with an immunosuppressive regimen. But these therapies are not benign, with an estimated risk of 40% for severe, permanent side effects.
Twenty-five percent of primary angiitis of the CNS patients had at least one relapse. Relapse was less frequent in patients with small-vessel primary angiitis of the CNS. Relapsing primary angiitis of the CNS patients required a longer immunosuppressive treatment than those without relapse (114).
The mortality rate in a large cohort of primary angiitis of the CNS patients was 17% during a mean follow-up period of 13 months (114). Cerebral infarction was the most common cause of death in these patients. Large-vessel involvement, cerebral infarction, cognitive impairment, and focal neurologic deficit at presentation were associated with high mortality, probably reflecting widespread and severe neurologic lesions. Moreover, the survival rate or functional outcome was not influenced by treatment (glucocorticoids alone or in combination with cyclophosphamide) or diagnostic method (biopsy or angiography). Functional outcome in patients with moderate disability at diagnosis improved over the course of treatment, but this trend was not found in patients with low disability at diagnosis, in whom functional status remain unchanged.
Better outcomes have been seen in patients with prominent leptomeningeal enhancement on brain MRI and small-vessel primary angiitis of the CNS (angiogram negative) than in other primary angiitis of the CNS patients (116; 117; 54). Outcomes in 53 patients with positive biopsy for primary angiitis of the CNS were relatively favorable. The rate of mortality and severe disability (modified Rankin Scale > 5) was 14% after a mean follow-up of 1.14 years following brain or spinal biopsy (92). On the other hand, a fatal outcome occurred in 10 out of 11 patients with a rapidly progressive subset of primary angiitis of the CNS, despite combined immunosuppressive treatment (120). Six patients died after a mean of 16 days from the start of treatment, and four died several months later. After 1 year, only one patient was still alive, though severely disabled.
A report describes the relapse and survival rate and the functional outcomes in 97 patients with primary CNS angiitis enrolled in a European multicenter cohort study (38). The median patient age at diagnosis was 46, with males making up 52% of the cohort. Patients were included if remission was achieved after an induction treatment--mostly glucocorticoids, usually combined with cyclophosphamide--and were followed for a median of 55 months. A mainly azathioprine or methotrexate maintenance treatment, started at a median of 4 months after induction therapy, was administered to nearly half of patients. Approximately 40% of patients had a good outcome (no relapse and modified Rankin score of 2 or lower at last follow-up), and patients who received maintenance therapy had a significantly better outcome (67%) than those who did not (20%). One third of patients underwent a relapse; relapses were twice as common in those who had not received maintenance therapy as in those who had (45% vs. 21%, respectively) (38).
An observational retrospective study of 44 patients with a diagnosis of PACNS by either biopsy (57%) or neuroimaging (43%) established that male patients had a higher rate of relapse after a median follow-up of 5 years (132). The mean age at diagnosis was 43 years, 26 patients were male, headache was the most common neurologic manifestation, and most cases involved the lymphocytic histopathological pattern. Twelve different immunotherapy regimens were used. Cyclophosphamide, the most common first-line therapy agent, was administered to 33 patients; glucocorticoids were administered to 30 patients, most often as comedication with other immunosuppressants; 86% of patients had more than 24 months of maintenance therapy. Relapse occurred in 26 (59%) of the 44 patients at a median of 1.5 years after PACNS diagnosis; 12 patients had more than one relapse and relapses usually occurred while patients were on immunosuppressant therapy. The annual relapse rate was 1.4 and male patients had a 3-fold higher risk of relapse than the female patients did. After 6 years of follow-up, two thirds of patients achieved remission and a comparable number had a good functional outcome (modified Rankin scale ≤ 2) (132).
Prolonged remission rate and functional long-term outcome were determined according to three main therapeutic regimens in a cohort of 112 patients with a PACNS diagnosis based on angiographic (71%) or histopathological (29%) findings (35). Remission was achieved with the initial induction treatment in 106 patients (95%), 14 of whom received only corticosteroids (group 1) whereas 40 were treated with corticosteroids and an immunosuppressant in the induction phase but received no maintenance therapy (group 2) and 45 received a combined treatment with corticosteroids and an immunosuppressant for induction followed by maintenance therapy (group 3). The therapeutic schedules for the remaining seven patients could not be classified in any group. Overall, 52 patients in remission received an immunosuppressant as maintenance therapy. Prolonged remission without relapse occurred in two thirds (70) of the patients after a mean follow up of 57 months. The overall mortality rate was 8%. A good functional outcome (modified Rankin scale ≤ 2) at last follow-up occurred in 56% (63) of the patients. More long-lasting remissions and a better functional outcome at the last follow-up occurred in patients who received the combined treatment with corticosteroids and an immunosuppressant in the induction phase followed by maintenance therapy (group 3). Conversely, patients with gadolinium-enhanced lesions (usually associated with small-vessel involvement) on MR imaging had a higher risk of relapse (35). The use of maintenance therapy was, thus, significantly associated with prolonged remission and a better functional outcome. These findings suggest that maintenance therapy with an immunosuppressant combined with corticosteroids may lead to the best long-term clinical and functional outcomes in patients with PACNS who have already achieved remission with either corticosteroid alone or in combination with another immunosuppressant.
Depression rate, quality of life, and functional independence were evaluated in 27 (34.6%) out of 78 PACNS patients who answered a questionnaire (60). Patients were mean aged 50.8 years, 52% were female, and their mean follow-up was 60 months after diagnosis established by biopsy (20 patients) or angiography (seven patients). The depression rate as evaluated by Patient Health Questionnaire (PHQ-9) was 30%; functional independence, assessed by the Barthel Index for Activities of Daily Living, showed mild disability in 70% and severe disability in 18.5% of patients. Quality of life was evaluated by the European Quality of Life–5 Dimensions (EQ-5D-5L) questionnaire, with 48% of patients showing mobility difficulties, 33% having self-care difficulties, 44% having problems with usual activities, 48% with pain, and 33% with moderate-severe anxiety or depression. Treatment history was known in 25 patients, all of whom received glucocorticoids: 13 (52%) patients as a single pharmacological agent, 11 (44%) along with cyclophosphamide, and one patient along with mycophenolate mofetil. Longer duration of immunosuppressants showed better Barthel Index and PHQ-9 scores and less disability in self-care and usual activities, though these results were not statistically significant.
Neuropsychological outcome in patients with primary angiitis of the central nervous system has not received much attention. The cognitive profile assessed by the Montreal Cognitive Assessment (MoCA) tool at baseline and after 3 and 10 months of follow-up has been reported in a young adult man with primary angiitis of the central nervous system, which was confirmed by brain biopsy and multiple cortical and subcortical microbleeding on brain MRI and was treated with immunosuppressants. A score of 15 out of 30 on MoCA was observed at baseline; attention, executive functioning, speed of processing, and auditory-verbal memory were severely affected. Reassessment at 3- and 10-month follow-ups showed an improvement in cognitive performance with scores of 20 out of 30 and 26 out of 30, respectively. Language, constructional praxis, visuospatial skills, and visuoperceptual organization showed greater improvement, whereas processing speed, attention, executive functioning and memory showed less improvement (90).
A case report from the landmark paper of Cravioto and Feigin that formed the basis of this intriguing disease is summarized below (31):
A 56-year-old female with no significant past medical history was admitted for an episode of aphasia followed by convulsive movements in the right side of the face and the right arm. She reported persistent vomiting with frontal headaches for the past 2 years. She signed off against medical advice.
A month later she had another episode of aphasia and a sudden episode of loss of consciousness that lasted for 30 minutes. She continued to have intermittent episodes of headache and vomiting. Later, she developed progressive weakness of both legs along with gradual mental deterioration and suffered from right-sided convulsive seizures. Physical examination revealed slight weakness and increased deep-tendon reflexes of the right extremities, right Babinski sign, and bilateral Hoffmann reflexes. All modalities of sensation appeared normal. Laboratory workup was largely unrevealing, with erythrocyte sedimentation rate noted to be slightly elevated at 35. Cerebrospinal fluid showed 62 RBC, three lymphocytes, 88 mg protein, negative Wassermann reaction, and a pressure of 220 mm of H2O.
The patient’s mental status deteriorated, and she developed right hemiparesis. She had severe convulsions involving the face and left leg and smacking movements of the lips. These were treated with intravenous amobarbital sodium. Repeat spinal tap showed slight xanthochromia with 363 RBC, 57 mg protein, and a pressure of 180 mm of H2O. Later she again developed convulsive movements that were partially responsive to intravenous and intramuscular phenobarbital. She suddenly became apneic and expired.
At autopsy, the brain appeared externally normal apart from the thickened leptomeninges at the base. Microscopically, the severe granulomatous process diffusely involved arteries and veins. The process involved the entire thickness of the vessel walls, with arteries being more affected than veins. Leptomeninges and subependymal white matter were particularly affected. The vessel lumens were narrowed or completely occluded. No fungi, tubercle bacilli, or other bacteria were found. Study of other organs revealed no lesions resembling those in the brain.
Cause and pathogenesis are unknown (54). Younger and colleagues suggested that primary angiitis of the CNS is a nonspecific inflammatory reaction of diverse etiology (153). Many systemic vasculitis syndromes are a result of deposition of immune complexes, but there is little support for this mechanism in primary CNS vasculitis. The infiltrating cells were CD4+ T lymphocytes in the few cases in which leukocyte typing was done. Reyes and colleagues reported a patient in whom they found intranuclear virus-like particles resembling herpes virus on electron microscopy of formalin-fixed brain samples (111), but to date there is no convincing evidence of infection with any microorganism in isolated CNS vasculitis, although Mycoplasma does cause CNS vasculitis in animals. There is no animal model for pure primary angiitis of the CNS. The one most often cited is cerebral arteritis induced by intravenous injection of Mycoplasma gallisepticum (143), but the model has its own shortcomings.
The credit for the first pathological description of primary CNS angiitis goes to Harbitz (64), but it was not until 1959 that Cravioto and Feigin elaborately discussed this entity (31). Further case reports were published by Budzilovich and colleagues between 1959 and the end of the following decade (16). By 1968, Kolodny and colleagues had accurately characterized the essential pathological features of the disorder (76).
Gross examination of brain and spinal cord showed no changes specific to the disorder. The majority of cases have had infarcts or hemorrhages, but they are usually small. Light microscopic examination confirms the presence of infarcts, hemorrhages, and areas of ischemic cell changes. The vascular inflammation is usually of a chronic granulomatous nature, with monocytes and histocytes, lymphocytes, and plasma cells infiltrating the walls of small (200 µm) arteries and veins, particularly in the leptomeninges and the branches that arise from them to penetrate the cortex. Larger vessels may be involved but never to an extent exceeding that of smaller ones. Angiitic vessels up to 1 mm in diameter have been observed (76; 77). Several case reports showed the involvement of the vessels of the circle of Willis or their larger branches but never without severe concomitant involvement of smaller vessels. Involvement of veins up to 2.5 mm has been seen, but that of larger veins or dural sinuses has not. Although present in most of the autopsy cases, giant cells of either Langerhans- or foreign body-type are not required to make the diagnosis. Nevertheless, in autopsied cases of primary CNS angiitis, multinucleated giant cells and macrophages may occasionally predominate over T-lymphocytes in transmural infiltrates (91). Skipped lesions are common as the disease is notoriously segmental in nature, resulting in only 50% of biopsy cases showing classic histopathology of primary angiitis of the CNS. Granulomas that seem to separate from vessels are occasionally encountered. Serial sectioning reveals an involved vessel adjacent to most of these. Reticulin stains often will allow identification of a small residual lumen within a granulomatous mass (17). Intimal fibrosis usually signifies healing or healed lesions. In most positive biopsies, acute healing and healed vasculitic lesions frequently coexist in different segments of the same artery or in adjacent arteries of the same biopsy. Kolodny and colleagues proposed the following sequence for the development of angiitic lesions (76):
(1) Intimal swelling and hyperplasia in arteries and adventitial lymphocyte infiltration in veins. | |
(2) Subintimal fibrinoid change and adventitial accumulation of histocytes. | |
(3) Necrosis and fibrinoid changes in the media, fragmentation of internal elastic lamina, and panmural infiltration of lymphocytes and histocytes. | |
(4) In typical lesions, granulomatous mass replaces all or part of vessel wall-lymphocytes, large mononuclear cells, fibroblast, multinucleated giant cells, and invariable number of plasma cells. | |
(5) Possibly a later stage, abundant fibrosis and relatively sparse lymphocytic infiltration, accompanied by large mononuclear and giant cells. |
Small, clinically silent foci of vasculitis are occasionally present in one or more viscera; lungs (76), heart (103), kidney (76), prostate (04), lymph nodes (17), and aorta (103) but by definition produce neither laboratory nor symptomatic evidence of organ dysfunction.
Among the other syndromes that can cause CNS angiitis, polyarteritis nodosa only rarely displays a granulomatous inflammation and does not affect the veins. Fibrinoid necrosis and transmutable inflammation both are more prominent in polyarteritis nodosa than in isolated CNS angiitis, with a special predilection for bifurcation. Polyarteritis-type necrotizing vasculitis may be seen in up to 25% of positive biopsy results. Churg-Strauss syndromes (30) and Wegener granulomatosis have extravascular granuloma, and eosinophils are prominent in the former. The angiitis form of sarcoidosis is characterized by extravascular noncaseating granulomas and the absence of fibrinoid necrosis.
The etiopathogenesis of primary CNS angiitis is very limited due to the scant availability of necessary pathological tissues. The potentially invading organisms need to have culture, immunohistochemistry, and in situ identification studies done so the mechanisms can be understood. The pathogenesis can be reviewed in the following sequence:
(1) Initiating factors (infection or immunocompromised state). |
The basic pathogenesis of all forms of angiitis almost certainly is immunologic; the inciting event may be nonspecific vascular damage, allowing deposition of immune complexes or exposing new antigens to the bloodstream. Crowe discusses four different mechanisms of immunologic tissue injury that might apply to the pathogenesis of angiitis (32):
(1) Immune complexes |
The granulomatous nature of the inflammation suggests a primary role of cell-mediated immunity with persistent antigenic stimulus that attracts T lymphocytes specific to that antigen as opposed to an antibody-mediated process. Immune complexes do induce granulomatous formation in certain circumstances. Although circulating immune complexes, hypocomplementemia and vascular deposition of immunoglobulin or complement have not been found in the cases of isolated CNS angiitis studied to date. The cellular infiltrate in primary angiitis of the CNS has not been immunophenotyped in large series of patients in order to make any concrete conclusion.
Once the immunological mechanisms have initiated an angiitis, proximal mechanisms of disease will lead to tissue damage, mainly by ischemia. This is simply the result of stenosis of vessels by the masses of inflammatory cells in their walls. The endothelial surface normally is the site of a delicate balance between the prothrombotic and antithrombotic tendencies, with the later obviously dominant in the normal circulation. Angiitis can alter this balance to favor thrombosis. The cytokines TNF and IL-1 are products of stimulated macrophages; in cultured endothelial cells they cause a number of changes that would encourage thrombosis. There is convincing in vitro and emerging in vivo evidence of endothelial activation by TNF and IL-1 that increase the permeability of vessel walls and expose thrombogenic basal lamina to the bloodstream.
In the case of a tumor-like presentation, breakdown of the blood-brain barrier of the small vessels due to infiltration of inflammatory cells in the perivascular and parenchymal regions has been proposed as the mechanism of mass lesion formation (81).
Cerebral vasculitis is a rare but potentially life-threatening condition with an annual incidence of 1 to 2 per million compared to 39 per million for systemic vasculitis (148). In a 2007 report, the annual incidence rate of primary angiitis of the CNS was estimated at 2.4 cases/million persons per year in the retrospective Salvarani cohort of 101 patients, over a 21-year period (114). To date, there has been no primary angiitis of the CNS-specific epidemiological study, despite there being an estimated 700 published cases of primary angiitis of the CNS worldwide (93). Of note, the disorder has been reported from many centers in Europe and North America, as well as from Australia and New Zealand. Whether the paucity of reports from Asia, Africa, and South America reflects a truly lower incidence in these regions is unknown. Most cases of primary CNS angiitis occur from the fourth to the sixth decade of life; the mean age ranges from 43 to 50 years of age, although patients anywhere from 3 to 71 years of age have been described (53). In earlier postmortem studies, males predominated 4:3 (84), but studies show an equal sex ratio.
The differential diagnosis of this unique entity must be examined in various ways, such as clinically as well as with neuroimaging modalities (MRI) and angiography (28; 79).
As mentioned above, primary angiitis of the CNS is a syndrome characterized by a range of pathology, at one extreme predominantly involving large intracranial vessels and at the other mainly involving cerebral microvasculature. It is useful to keep the pathological spectrum in mind when approaching the diagnosis. Thus, predominantly large-vessel disease is more likely to manifest at least in part with stroke-like features, whereas predominantly microvascular disease is more likely to manifest as a parenchymal mass, encephalopathy, or myelopathy. The differential diagnosis for these two extremes differ considerably (99).
Predominantly microvascular disease | ||
• Glioma or glioblastoma multiforme | ||
Predominantly macrovascular disease | ||
• Thromboembolic stroke | ||
- Idiopathic (eg, thunderclap headache) | ||
• Post-trigeminal zoster vasculopathy | ||
Mixed | ||
• Tuberculosis |
• Multiple infarcts (atherosclerotic vs. embolic) |
Numerous entities, both inflammatory and noninflammatory, induce changes (sometimes transitory) resembling primary CNS vasculitis on cerebral angiography (130). | ||
• Neoplastic angioendotheliosis | ||
- With pheochromocytoma |
PACNS diagnosis remains challenging because of its diverse and nonspecific clinical and neuroimaging features. Rice and Scolding proposed binary PACNS diagnostic criteria, “definite” or “possible,” which may be a useful tool in the therapeutic research field (112). Both categories include appropriate clinical findings for CNS vasculitis as well as reliable exclusion of alternative diagnoses (PACNS mimics and secondary CNS vasculitis). Along with clinical findings suggesting PACNS, the “definite” category requires histopathological proof of a granulomatous, lymphocytic, or necrotizing pattern of CNS vasculitis, either by biopsy or by autopsy, as well as evidence of vessel wall involvement. The “possible” PACNS diagnosis lacks these histopathological findings.
Sarti and colleagues have proposed a first step algorithm in the PACNS diagnostic process that should be properly validated (128). The purpose of this algorithm is to provide guidance in the clinical setting when PACNS could be suspected in a patient, before performing extensive and sometimes invasive investigations. Clinical and neuroradiological features were categorized as major and minor according to the median of frequencies reported in 24 case series involving 585 patients with a PACNS diagnosis (41% biopsy proven). Major clinical features included new-onset or modified headache, cryptogenic ischemic stroke, subacute cognitive impairment, and neurologic focal deficit, whereas minor clinical features included seizures, alteration of consciousness, and psychiatric disorders. Major neuroradiological features included multiple parenchymal lesions, parenchymal or meningeal contrast enhancement, vessel abnormalities (single or multiple stenoses/occlusion), and vessel wall contrast enhancement, whereas minor neuroradiological features included parenchymal or subarachnoid hemorrhages and single parenchymal lesions. According to this algorithm, PACNS should be initially suspected if patients have one clinical feature (major or minor) together with one major neuroradiological feature, or two clinical features (at least one major) associated with one minor neuroradiological feature. There was no other explanation for the presenting clinical and neuroradiological features in either scenario (128).
The diagnosis of granulomatous angiitis begins with a high index of suspicion based on the clinical picture and the need to develop an organized approach. The clinical manifestations are diverse, with the typical patient presenting with headache of gradual onset, often accompanied by signs and symptoms of encephalopathy, and later developing focal symptomatology. Of particular note is the lack of systemic and constitutional symptoms. Absence of headache and mental status changes has good negative predictive values. A careful history and physical examination with attention to skin, eyes, testicles, paranasal sinuses, and lungs are likely to exclude systemic vasculitis. General medical laboratory investigations are frequently normal and unhelpful in confirming the diagnosis but extremely valuable in excluding other primary angiitis of the CNS mimickers. Acute phase reactants (erythrocyte sedimentation rate, C-reactive protein) are usually normal (08). Screening blood studies, including white blood cell counts, hematocrit, rheumatoid factor, antinuclear antibody, cryoglobulin, total hemolytic complement, C1q binding for circulating immune complexes, SSA/SSB, p-ANCA, c-ANCA, syphilis serology, and hypercoagulable profile (anticardiolipin and antiphospholipid antibodies) are all normal but clinically helpful to exclude other conditions (24; 08).
Cerebral spinal fluid examination should be performed in all patients with suspected primary angiitis of the CNS unless a contraindication exists; the abnormalities of CSF observed in primary angiitis of the CNS generally reflect an inflammatory process within the CNS. Mildly elevated protein was observed in 52% to 72% of patients and mildly lymphocytic pleocytosis in 50% to 68% (115; 79). In the Salvarani cohort of 101 primary angiitis of the CNS patients, a higher median leukocyte count and a higher median total protein concentration were observed in patients with primary angiitis of the CNS confirmed by biopsy versus primary angiitis of the CNS patients diagnosed by angiography: 17 versus 4 cells/ml and 98 versus 54 mg/dl, respectively (114). Nevertheless, in the same cohort, 8% of patients had a normal cerebral angiogram and a positive biopsy for small-vessel vasculitis with a predominating granulomatous histologic pattern. These patients had a CSF profile characterized by a higher protein level (median 1180 mg/L), perhaps reflecting a diffuse spread of the intracranial inflammatory process, and a mildly lymphocytic pleocytosis (115c). The CSF analysis has a high negative predictive value, although a completely normal CSF cannot rule out primary angiitis of the CNS. Thus, the primary value of CSF examination in investigating suspected CNS vasculitis is to rule out infections including syphilis or neoplastic infiltration of meninges (58). Moreover, the CSF profile can help to distinguish between primary angiitis of the CNS and reversible cerebral vasoconstriction syndrome (the most frequent mimic of primary central nervous system vasculitis) when faced with angiographic abnormalities. A normal CSF profile is usually found in the latter case, unless there is a coexisting cortical hemorrhage (137).
Abnormal electroencephalographic findings have been reported in 27% to 74% of primary angiitis of the CNS patients (114; 78). Unfortunately, the lack of specificity of results greatly compromises its usefulness in confirming the diagnosis. The most common abnormalities identified are the diffuse slow-wave pattern, but at times the pattern may be focal (114; 78). Nevertheless, this pattern can also be observed in a variety of infectious and metabolic disorders affecting the CNS. As with other diagnostic tests, the EEG may be useful in identifying other causes of unexplained neurologic symptoms, particularly when the clinical presentation is one of unexplained dementia (24). When this situation is secondary to prion disease, the EEG may reveal diagnostic changes and, thus, would eliminate primary angiitis of the CNS as a further consideration.
Brain magnetic resonance rather than cranial CT is the first-line imaging modality for detecting brain parenchyma lesions. Magnetic resonance imaging with special sequences, including T2-weighted and FLAIR images, are found to have high sensitivity but lack specificity. Earlier case reports indicate low sensitivity for MRI, but unfortunately, they did not include the FLAIR sequence (71). Later, Wasserman and colleagues provided evidence that advanced neuroimaging sequences tailored to detect infarction (ie, diffusion-weighted, perfusion imaging and fluid-attenuated inversion recovery sequences) are needed to enhance the MRI sensitivity (147). Alhalabi and Moore found abnormalities in 13 of 17 patients. The other four patients had a normal MRI, thus, yielding a sensitivity of 74% (02). MRI sensitivity approaching 100% has been reported. A normal brain MRI makes the diagnosis of primary CNS angiitis unlikely (78; 62; 53; 79). A broad spectrum of abnormalities not specific for primary CNS angiitis may be found on brain MRI, including multiple infarcts, confluent white matter lesions, large intraparenchymal hematomas, multiple microhemorrhages, leptomeningeal and parenchymal enhancement, and small and large, single and multiple tumor-like mass lesions (108). In very rare cases, brain MRI may prove to be normal in patients with histologically confirmed primary central nervous system angiitis (108). In the last few months, T2 hyperintense lesions with ring enhancement and associated vasogenic perilesional edema found at baseline showed a fluctuating course with disappearance of previous lesions and appearance of new ring enhancement lesions at 3 and 6 years on brain MRI follow-up in a 20-year-old patient. Remarkably, all the brain MRI lesions at baseline and during 6 years of follow-up were observed exclusively in the right hemisphere (125). This radiological pattern has never been reported before. No spinal cord involvement occurred either at baseline or during follow-up on spinal MRI. The patient had biopsy-proven PACNS (lymphocytic histologic pattern), as well as multifocal chronic relapsing course characterized by clinical exacerbations of hemiparesis restricted to the left side. One extremely unusual pattern of PACNS may be multiple hyperintense lesions with contrast enhancement distributed in the white and grey matter on brain MRI and exclusively confined to one cerebral hemisphere. A targeted brain biopsy is essential to discriminate between PACNS and a broad spectrum of pathologies (142).
The most common lesions found by brain MRI are cerebral infarctions with diffusion disturbances, which occur in up to 75% of patients. They are often multiple and bilateral, involving both grey and white matter (02; 114; 78).
Less commonly reported are hemorrhages, both parenchymal and subarachnoid, and mass lesions are rare occurrences (144; 58). MRI may show ischemic and hemorrhagic lesions of different ages. Gadolinium-enhanced intracranial lesions can be observed in around one third of cases (114). Variable patterns of contrast enhancement have been observed, including irregular subcortical streaks, leptomeningeal enhancement with comparatively little parenchymal involvement, focal cortical rib boning, and diffuse parenchymal vessel enhancement (135).
Prominent gadolinium leptomeningeal enhancement on MRI, most often diffuse rather than localized, was found in 8% of patients with proven primary small-vessel angiitis of the CNS. MRI imaging showed bilateral symmetrical linear gadolinium-enhancement in a radial pattern through the cortical white matter and brainstem of a 47-year-old Caucasian man with biopsy-proven (granulomatous pattern) primary CNS angiitis. This peculiar contrasted-enhancement pattern is thought to follow the pathway of the white matter vessels involved (52). Fifty percent of patients had both leptomeningeal enhancement as the only abnormality observed on brain MRI and vascular deposit of beta-amyloid peptide on brain biopsy (115b).
Brain MRI and MR angiography findings have been reported for 60 patients (mean age: 45 years; 57% male) with primary angiitis of the CNS proven either angiographically (63%) or by biopsy (37%) (14). In this multicenter cohort study, acute infarcts were observed in 75% of patients at the time of diagnosis. The most common distribution (42%) was multiple bilateral distal ischemic lesions, mainly in the carotid circulation, followed by large subcortical and large cortical infarcts (24% and 19% of patients, respectively). The majority of patients had multiple infarcts mostly in the supratentorial region; 25% had infratentorial infarcts whereas 27% had a single ischemic lesion. Half of the patients had acute ischemic lesions coexisting with subacute and chronic infarcts. The vasculitis diagnosis was arrived at either by angiography or by biopsy (with no vascular involvement on neurovascular imaging) in 77% and 23%, respectively, of patients with disseminated ischemic lesions (14).
Hemorrhagic lesions were also frequent in this cohort. Acute or old parenchymal hemorrhages were found in 55% of patients, with a broad spectrum of manifestations including microbleeds and lobar, deep, or multifocal hemorrhages. Interestingly, over 50% of hemorrhages were found within or in the proximity of acute infarcts, suggesting a hemorrhagic transformation of ischemia. An acute convexity subarachnoid hemorrhage was observed in 14 patients (26%), seven patients showed diffuse bleeding, and the remainder had focal bleeding (14).
Gadolinium-enhanced parenchymal and leptomeningeal lesions occurred in 75% and 42% of patients, respectively. Both patterns of enhancement were more common in patients with biopsy-proven vasculitis diagnosis. Twelve percent of patients presented with vasculitis resembling a brain tumor (14).
Three-dimensional time-of-flight MR angiography was abnormal in 77% of 56 patients. Multifocal narrowing and segmental stenosis occurred in at least two cerebral arteries in 56% of patients; 45% of patients had proximal and distal occlusion or arterial stenosis; and 18% of patients had affected arteries in the distal segment. One fifth of patients had a typical beading pattern on MRA. Four patients with normal MRA showed a distal stenotic lesion on digital subtraction angiography (14).
The previously mentioned meta-analysis of 911 PACNS patients found cerebral infarcts in 68% and 34% of patients with angiographic and biopsy diagnoses, respectively, whereas intracranial hemorrhages were found in 25% of cases confirmed by biopsy and in 18% of those confirmed by angiogram (09). Parenchymal or leptomeningeal enhancement on neuroimaging were more common in biopsy-proven cases (69%) than in angiogram-confirmed cases (22%). Parenchymal enhancement was more than twice as frequent (37%) as leptomeningeal enhancement (16%). A third of patients with PACNS biopsy diagnosis who had digital subtraction angiography showed any of the following vascular abnormalities: beading, vessel occlusion, collateral circulation, or microaneurysms. Vasculitis findings were observed on brain MRA in 29% of patients with PACNS biopsy diagnosis (09).
High-resolution anatomical and blood-sensitive MRI sequences aimed at providing information about the vessel wall structure of intracranial blood vessels may be a useful technique in the diagnosis of both inflammatory and noninflammatory intracranial arteriopathies (61; 88; 43). In a small sample of seven patients with multifocal segmental narrowing of large intracranial arteries of unclear etiology, this high-resolution contrast-enhanced technique showed persistent arterial narrowing, circumferential arterial wall thickening, and arterial wall enhancement in patients with primary angiitis of the CNS, whereas in those with reversible cerebral vasoconstriction syndrome, it showed transient arterial narrowing, arterial wall thickening, and a lack of arterial wall enhancement (88). According to the European Stroke Organization (ESO) guidelines on PACNS, high-resolution vessel wall imaging MRI is a promising, though not yet validated, technique for PACNS diagnosis. It remains to be seen, however, whether this technique improves diagnostic accuracy compared to digital subtraction angiography (105).
Advanced MR imaging techniques such as 1.5T MRI, 3D-enhanced susceptibility-weighted angiography (SWAN), a new method for susceptibility-weighted imaging with a 3-dimensional multi-echo gradient-echo MRI may provide a higher sensitivity and better spatial resolution than conventional T2-gradient-echo MRI. Thus, in a female aged 37 with transient hemifacial paresthesias, multiple hyperintense supra- and infratentorial white matter lesions were observed on fluid-attenuated inversion recovery and diffusion-weighted imaging sequences; most of these lesions showed contrast enhancement on T1-weighted images, whereas T2 sequences were unremarkable. The 3D-enhanced susceptibility-weighted angiography (SWAN) revealed multiple supratentorial hypointense lesions within frontal and temporal convexity (51). These lesions found on SWAN sequences were initially thought to be either cortical hemosiderin deposits or small artery or venous dilations. Later, a digital subtraction angiography confirmed that they corresponded to fusiform microaneurysms and segmental narrowing within distal branches of anterior, middle, and vertebral arteries compatible with vasculitis (51). Interestingly, microaneurysms were rarely seen and, moreover, this patient never developed parenchymal or subarachnoid hemorrhage. At 24 months of follow-up after immunosuppressive therapy, most lesions had disappeared in FLAIR sequences, and a decrease in size of supratentorial hypointense lesions was found in SWAN sequences. Based on these findings, the authors suggest that this new MR technique may become a useful noninvasive tool in the setting of a multimodal MRI approach when vasculitis is suspected.
The potential diagnostic value of special imaging techniques, such as black-blood 3.0-Tesla brain MRI (T1-weighted, fat-and-blood-suppressed sequences before and after contrast injection), was evaluated in 12 patients with suspected large primary angiitis of the central nervous system (107). Moderate or strong vessel wall contrast enhancement was observed in four patients who had a final diagnosis of primary CNS angiitis. Contrast enhancement on the vessel wall correlated with stenotic areas. After immunosuppressive therapy, contrast enhancement diminished in two patients but remained unchanged in the other two patients. Nevertheless, contrast enhancement was not specific because it was also observed in six patients with intracranial artery disease (atherosclerosis, moyamoya disease, and dissection) (107).
Moreover, the potential diagnostic usefulness of contrast-enhanced high-resolution MRI at 3-Tesla in visualizing vessel-wall inflammation was corroborated in a patient with recurrent ischemic strokes restricted to the vertebrobasilar circulation due to primary angiitis of the central nervous system established by angiography after appropriate exclusion of mimics and secondary vasculitis. In this patient, high-resolution MRI (axial proton density and post-contrast T1-weighted images) was able to show eccentric vessel-wall thickness with enhancement at the proximal basilar artery on day 10 of evolution, whereas no luminal changes in the vessel wall were observed at onset, at day 3, and at day 7 of evolution by standard brain MRA or digital subtraction angiography (102). A new contrast-enhanced high-resolution MRI at 3-Tesla performed at 1 month of immunosuppressant therapy showed a nearly complete resolution of the vessel-wall inflammatory abnormalities. Consequently, this noninvasive special imaging modality may be useful in the diagnostic work-up, particularly when standard MRA and conventional cerebral angiography are not diagnostic, as well as in monitoring therapy response.
A unilateral form of PACNS has been reported. In this rare variant, clinical and radiological features are strictly limited to one hemisphere (01). AbdelRazek and colleagues report seven such cases, termed “unilateral relapsing primary angiitis of the central nervous system” (UR-PACNS) because of its potential for relapse. Males accounted for five of these cases; the median age at clinical onset was 26 years (ranging from 10 to 49 years); all patients presented with seizures; three patients had associated aphasia, and four had hemiparesis unrelated to Todd paralysis at presentation. CSF examination was unremarkable, and no features of vasculitis occurred in five patients with conventional cerebral angiography. All patients exhibited lymphocytic vasculitis of small- and medium-sized vessels (first brain biopsy was not diagnostic in two patients). Mean follow-up was 7.5 years, and the mean annual relapse rate was 0.62. Five patients had multiple relapses in the left hemisphere by clinical and radiological criteria, whereas four patients exhibited a progressive unilateral volume loss on neuroimaging. Steroids alone were unsuccessful in preventing disease progression. Cyclophosphamide and rituximab as induction therapy followed by a steroid-sparing agent appeared to be associated with disease remission (01). Due to its main form of clinical presentation, this uncommon unilateral variant of PACNS expands the differential diagnosis, which should include Rasmussen encephalitis and unilateral cortical FLAIR-hyperintense lesions in anti-MOG-associated encephalitis with seizures (FLAMES). To explain the unilateral hemispheric predominance of clinical, pathological, and radiological features at onset and relapses, AbdelRazek and colleagues speculate the existence of a functional difference in immune response rather than a structural antigenic difference between both cerebral hemispheres.
Chances of positive biopsy are increased if the tissue is from the contrast-enhancing areas. Ehsan and colleagues showed that serial imaging is useful in monitoring the clinical course of the disease (44). An MRI scan should be performed to exclude other diagnoses, such as multiple cerebral metastasis, multicentric primary CNS tumors, and hydrocephalus or demyelinating diseases. The general consensus among clinicians is that MR angiography is not very useful due to its low spatial resolution. In two series comprising 21 and 101 primary angiitis of the CNS patients, MR angiography was suggestive of vasculitis in 47% and 59% of patients, respectively (114; 78). Nineteen patients in the 101-patient cohort had findings suggestive of vasculitis on MRA; 14 of them showed involvement of both large- and small-vessel arteries, whereas the remaining five had exclusive involvement of either large- or small-vessel arteries (114). Yuh and colleagues pointed out the importance of perfusion imaging in assessing microcirculatory vasculopathy and diffusion imaging in detecting early changes of microcirculatory ischemia or infarctions that are characteristic findings of cerebral vasculitis and may not be appreciated by conventional MRI sequences (T2 or FLAIR) (155).
The enthusiasm for angiographic diagnosis of primary angiitis of the CNS has markedly fallen in the last decade since the realization of the fact that the findings are highly nonspecific and can be produced by a large number of inflammatory and noninflammatory conditions (61). The vascular changes revealed by conventional or noninvasive (magnetic resonance or computed tomography) angiography include beading, aneurysms, circumferential or eccentric vessel irregularities, multiple occlusions with sharp cutoffs, and an avascular mass effect. Among the pathologically documented cases, cerebral angiography may be normal in up to 40% of patients (sensitivity of 60%) (24; 63; 61). A subset of primary angiitis of the CNS involving small vessels beyond the resolution of cerebral angiography has been described in adults (115c; 54). Additionally, Hellmann and colleagues demonstrated that biopsy of angiographically abnormal vessels is safe and may help the surgeon to obtain tissue that will be most likely to show histopathological changes (66). Alhalabi and Moore pointed out the fact that serial angiography can be used for monitoring the clinical course, but the procedure does carry the risk of transient (10%) or permanent (1%) neurologic deficits (02).
There are very few reports of MR spectroscopy in patients with primary angiitis of the central nervous system. MR spectroscopy in patients with tumor-like presentation has revealed nonspecific findings with varying degrees of elevation of choline/N-acetyl aspartate ratio and lipid-lactate peaks (81). In another patient with primary CNS angiitis who exhibited an unusual pattern on MRI (linear gadolinium enhancement in a radial pattern), MR spectroscopy demonstrated elevated choline, decreased N-acetyl aspartate, and no evidence of lactate (52).
Brain and meningeal biopsy is the gold standard for the diagnosis of primary angiitis of the CNS because it is the only way to identify the characteristic pathology and to exclude other disorders like lymphoproliferative diseases, infections, sarcoidosis, and others (96). Brain biopsy is limited by poor sensitivity (20). Premortem biopsies yield false-negative results in about 25% of autopsy proven cases (22); sampling of the leptomeninges as well as the underlying cortex will likely increase the diagnostic yield because the vasculitis may be present in only one of the two sites. (22; 96; 29; 70; 54). Biopsy of radiographically abnormal areas, particularly in the presence of abnormal enhancement, improves the sensitivity of the procedure (92; 54). In the absence of focal lesions, the temporal tip of the dominant hemisphere in an area with longitudinally oriented surface vessels is the preferred site (20). An infratentorial approach should be considered in patients when there is suspicion of sarcoidosis or tuberculosis because the basilar meninges are preferentially involved (45). Deep or stereotactic biopsy samples are not required unless approaching a mass lesion (96; 86). 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 disease and CNS infections (20). Because the specificity of biopsy is not 100%, even a positive biopsy result should be interpreted in light of the entire clinical picture. Alrawi and colleagues proposed biopsy criteria for definite and probable primary angiitis of the CNS (03):
(1) Minimum of two layers of lymphocytes within or around the walls of parenchymal or leptomeningeal and dural vessels (“lymphocytic inflammation”). | |
(2) Structural alterations of the vessel wall, such as prominence of the endothelial cells or indistinct appearance with or without necrosis. | |
(3) Pink neuronal cytoplasm and pyknotic neuronal nuclei with or without pyknotic glial nuclei and astrocytic gliosis ("ischemic changes"). | |
(4) Neuronophagia. | |
(5) Parenchymal (including perivascular) edema. | |
(6) Exclusion of alternative diagnoses. |
Probable primary angiitis of the CNS requires fulfillment of criteria 2 through 6.
In a series of 51 brain biopsies and two spinal nerve root biopsies from primary angiitis of the CNS patients (of which nine had stereotactic biopsies and the rest had open biopsies), 63% were diagnostic of vasculitis (92). Eighty-nine percent of the stereotactic biopsies were diagnostic of vasculitis, and 57% of the open biopsies provided a vasculitis diagnosis; approximately 60% of the diagnostic biopsies included leptomeningeal samples. Targeted biopsies (which included a radiologically identified lesion) were diagnostic in 79% of cases, whereas blind biopsies were negative for primary angiitis of the CNS in all cases. A granulomatous histologic pattern was most common (58% of cases), whereas lymphocytic and necrotizing vasculitis were found in 33% and in 15% of patients, respectively. A deposit of β/A4 amyloid was found exclusively in the granulomatous pattern (nearly half of those cases), whereas intracranial hemorrhage occurred exclusively in patients with necrotizing primary angiitis of the CNS (92). The explanations advanced for the high proportion of nondiagnostic biopsies in this cohort were that the affected vessels may have been larger and may not have extended to the parenchymal surface and leptomeninges, and that a highly focal vasculitis might be missed by biopsy. Nonspecific gliosis and mild perivascular mononuclear inflammation were the most common findings in nondiagnostic biopsies (92).
An extended, multimodal-imaging-guided, stereotactic brain biopsy performed in 23 patients with suspected PACNS based on clinical manifestations established a definitive diagnosis of pathological vasculitis in seven (30%) of the patients, whereas a diagnosis other than PACNS was also established in seven patients, and the biopsies of the remaining nine patients yielded unspecific gliosis results (140). Biopsies included tissue samples from three different compartments (meninges, cortex, and white matter) in angiography-negative patients with MRI lesions and clinical manifestations suggestive of PACNS. Minor and transient complications occurred in 13% of the patients (140).
A multicenter retrospective cohort study in which half of the 200 patients diagnosed with PACNS underwent brain biopsy compared baseline findings between 61 biopsy-positive and 39 biopsy-negative patients (100). Biopsies were obtained more often when patients had seizures, cognitive impairment, non-ischemic parenchymal or leptomeningeal enhancement on MRI, or one or more microbleeds on neuroimaging. Vasculitis findings were significantly twice as frequent from brain parenchyma samples (57%) as from leptomeninges samples (26%). Biopsy-positive PACNS was more likely to occur in patients with seizures (OR 8.31), cognitive impairment (OR 2.58), and in female patients (OR 2.9). Patients with non-ischemic parenchymal or leptomeningeal enhancement or one or more microbleeds on neuroimaging had a higher likelihood of presenting vasculitis on brain biopsy (100). Pure lymphocytic, granulomatous, and necrotizing vasculitides were found in 57%, 41%, and 2% of patients, respectively. Amyloid-beta–related angiitis was found in nine (15%) of 57 patients, whereas eight patients showed the granulomatous subtype, which was more commonly associated with headache, inflammatory CSF, acute subarachnoid hemorrhage, one or more microbleeds, and one or more intracranial stenoses on digital subtraction angiography.
To date, there is insufficient experience with SPECT and PET scanning to know how, if at all, such modalities may be used in the evaluation of these patients. Noninvasive imaging of activated microglia/macrophages by PET/MRI targeting the translocator protein (TSPO) using [18F]DPA-714 (a second-generation tracer) has been evaluated in four patients with a PACNS diagnosis (06). Patients were mean aged 51.7 years and half were male. Two patients showed an elevated tracer uptake extending beyond the ischemic area demonstrated by MRI; one, diagnosed by biopsy, received no treatment before PET scan, and the other, diagnosed by angiography, received glucocorticoids for 4 days before PET scan. The other two patients had PACNS diagnoses based on clinical findings and showed no pathologic tracer uptake on PET; their interval between glucocorticoid therapy and PET scan ranged from 9 to 13 days (06).
In a retrospective analysis of 31 patients, the patterns of clinical manifestations, laboratory investigations, and neuroimaging findings were compared according to whether the PACNS diagnosis was established by biopsy (17 of 26 biopsies were positive for vasculitis) or (as in 14 patients) by neuroimaging criteria (MRA or digital subtraction angiography) (131). The most common pattern of PACNS, identified in six patients, was lymphocytic; four patients had a granulomatous pattern and two had a necrotizing type. Patients with a biopsy-proven diagnosis (small-vessel vasculitis) had a more progressive chronic course of clinical manifestations and more headache, cognitive deficits, and seizures than those with neuroimaging-based diagnoses (medium-sized vessel vasculitis), although the differences were not statistically significant. In addition, patients with a biopsy-proven diagnosis did not have significantly higher leukocyte count, higher protein levels, or more oligoclonal bands in CSF than patients with a neuroimaging-based diagnosis. Concerning radiological findings, patients with biopsy-confirmed PACNS were found to have significantly fewer brain infarcts and vessel wall abnormalities, initially and during the course of the disease, when compared with the group of patients with an imaging-based diagnosis. Contrast enhancement (parenchymal and leptomeningeal) was more frequently observed in patients with biopsy-proven diagnoses. In the first study, high-resolution MR showed contrast enhancement in the vessel wall in 10 (76%) of 14 patients with imaging-based diagnoses, whereas in the biopsy-proven group, no patient had this finding initially, though it appeared later in the course of the disease in three patients (21%). The clinical outcome (modified Rankin score) did not show any difference between the groups after a median follow-up of 46 months (131).
Emerging biomarkers in peripheral venous blood are promising tools in PACNS diagnosis. Endothelial damage markers such as circulating endothelial cells and von Willebrand factor antigen or markers for endothelial and vascular repair, such as endothelial progenitor cells, may show higher levels in patients with PACNS, systemic vasculitis, and noninflammatory cerebral vasculopathies (reversible vasoconstriction syndrome and moyamoya disease – PACNS mimics) (39; 40). In addition, levels of these biomarkers may vary according to disease activity, thereby making them useful tools for monitoring the response to treatment. Further studies are needed to determine whether these biomarkers or others can enhance diagnostic certainty.
CSF biomarkers may also be helpful in a PACNS diagnostic work-up, though there is scant information to date. Patients with PACNS may exhibit elevated CSF interleukin-17 (IL-17) and neurofilament light chain levels (39; 106). CSF IL-17 levels may remain elevated regardless of disease activity, whereas elevation of CSF neurofilament light chain levels preceded the appearance of vasculitis changes on MRA and digital subtraction angiography by 6 months (39; 106). Elevated soluble TREM2 levels on serum and CSF have been observed in a small cohort of PACNS patients compared to control patients. These elevated levels were positively correlated with proinflammatory factors (tumor necrosis factor-alpha), PACNS severity (NIHSS score and modified Rankin score), neuroimaging findings (lesion volume), and poor outcome (57).
Ophthalmoscopic examination. Dynamic recording of erythrocyte flow using video slit lamp microscopic recording and low dose fluorescein angiography to examine the vasculature of the anterior ocular chamber can be a useful additional investigation. Scolding and colleagues conducted a study in which four of five patients had abnormal findings (133). Typical abnormalities are marked slowing of flow, multifocal attenuation of arterioles and erythrocyte aggregation. Fluorescein studies may confirm these changes and demonstrate areas of small vessel infarction along with post capillary leakage.
The current management scheme for primary angiitis of the CNS was first proposed by Cupps and colleagues and was based on reports of successful therapy in systemic vasculitides such as polyarteritis nodosa and Wegener granulomatosis with a combination immunosuppressive regimen (46; 98; 33; 07).
To date, no standardized treatment protocols are available, and no controlled clinical trials have been performed. Prospective randomized controlled trials are difficult because of the rarity of the condition and the lack of unifying diagnostic criteria. Retrospective analysis has revealed significant support for the use of steroids with cyclophosphamide therapy in confirmed cases (74). High-dose prednisone plus cyclophosphamide is the treatment of choice for most of the subtypes of primary CNS angiitis, particularly for those patients with rapidly progressive course or more severe forms (21; 114; 54; 123). Eighty percent of patients showed a favorable response to this therapeutic approach (corticosteroid either combined or not with cyclophosphamide) (114).
The management has been broadly divided into induction and maintenance phases (74). In the induction phase, the consensus opinion is to start with a daily dose of 1 mg/kg of prednisone or its equivalent along with 1 to 2 mg/kg of cyclophosphamide; however, the course and burden of the disease should be considered when introducing cytotoxic agents. Intravenous methylprednisolone (1 g/day for 3 to 5 days) may be a valid alternative, particularly in patients with a rapidly progressive form of primary angiitis of the central nervous system or acute alteration of mental status (11; 123). Side effects of high-dose glucocorticoids should be anticipated, such as bone mineral loss, opportunistic infections, and neurologic side effects like delirium, mania, and schizophrenia-like states. All patients under cyclophosphamide and glucocorticoid treatment should receive prophylactic treatment against Pneumocystis jirovecii, a common pathogen in immunosuppressed hosts. A combination of trimethoprim and sulfamethoxazole (80 mg/400 mg daily) is recommended (11). The major toxicity of cyclophosphamide in the acute phase is myelosuppression, so its dose should be adjusted to keep the total WBC count above 4000 and the absolute neutrophil count above 1000 to 1500. A complete blood cell count is recommended at least every 2 weeks. Other side effects include hemorrhagic cystitis, which can be minimized with hydration, frequent voids, and administration of mesna. The most serious long-term toxicity is an increased risk of malignancy, especially leukemias, lymphomas, and bladder cancer. Patients should be informed about the risks and benefits before starting this cytotoxic agent. Based on the efficacy observed in systemic vasculitis treatment and the relatively fewer adverse effects compared with cyclophosphamide, mycophenolate mofetil combined with corticosteroids may be an alternative immunosuppressant agent, particularly in young patients with primary angiitis of the CNS likely requiring long-term immunosuppressive therapy (27). The ESO guidelines on PACNS suggest that mycophenolate mofetil should be considered for long-term maintenance therapy to reduce the toxicity of cyclophosphamide (105). Recommendations on when and how rapidly to taper off prednisone treatment vary widely, but, in any case, should be tailored to the severity, extent, and course of the disease; adverse effects of glucocorticoids; and patient and physician preference.
Once clinical remission has been achieved, treatment with cyclophosphamide should be continued for 3 to 6 months and then switched to a lower-risk immunosuppressant agent like azathioprine (2 mg/kg/day), mycophenolate mofetil, or methotrexate (11; 123). This maintenance therapy should be continued for at least 12 to 18 months (123) or even for as long as 2 to 3 years after induction of remission (11). The ESO guidelines on PACNS suggest continuing maintenance therapy for at least 2 years in PACNS patients with no recurrences (105).
Monitoring of induced remission had been a matter of intense debate. There is no substitute for careful serial clinical evaluation. Although some symptoms may improve with treatment, neurologic deficits due to cerebral infarction or hemorrhage may be irreversible. Use of serial angiography was proposed initially (139), but it exposes the patient to unnecessary risks of angiography. Moreover, the broad spectrum of angiographic abnormalities found on direct or indirect cerebral vascular studies may not undergo expeditious modifications; therefore, such modifications may not be reliable parameters for checking disease activity (58). Subsequently, serial MRI scans were also suggested to monitor remission; thus, a brain MRI and MRA should be obtained between 4 to 6 weeks of the start of treatment and at least quarterly thereafter, particularly during the first year of treatment or on a new onset of neurologic deficit (123). Calabrese and colleagues proposed that if CSF abnormalities are present before therapy begins, they are more likely to be sensitive indicators of disease activity (22). Transcranial Doppler monitoring of blood flow has been applied to follow postpartum angiopathy, and this type of noninvasive monitoring may have some role but has not been studied (12).
Deterioration, failure to respond initially, or intolerance of the above regimen may require the use of alternative agents like methotrexate. Intravenous IgG has been administered with success a few times, but its use has been poorly documented. Tumor necrosis factor blockers have been used with acceptable results in two patients who were refractory to immunosuppressive therapy (118). Successful clinical and radiological responses with rituximab were observed in two patients with a diagnosis of primary CNS angiitis, one of whom was initially treated with rituximab (375 mg/m2/week for 4 weeks) and prednisone, and the other with rituximab after cyclophosphamide failure (34). Rituximab proved safe and effective in six selected patients with symptomatic PACNS who had failed to respond to corticosteroids and immunosuppressants (126). Four of these patients were diagnosed with PACNS based on histological findings and the remaining two were diagnosed based on imaging results. Five patients had a refractory disease whereas the sixth patient’s initial treatment-induced remission was due to rituximab along with corticosteroids. The mean age at diagnosis was 50 years and mean disease duration at the start of rituximab therapy was 15 months. Rituximab substantially reduced the number of flares in four patients; the fifth patient showed an initial improvement but then failed to respond to rituximab and had two subsequent relapses when B cells were depleted. The remaining patient had a relapse after which a second course of rituximab was associated with remission. Before treatment with rituximab was begun the six patients had a median mRS score of 3. At their last visit they showed an improved functional outcome with a median mRS score of 2 (126).
Salvarani and colleagues reported the long-term follow-up of treatment response in 191 PACNS patients from a single center over a period of 35 years (122). The median follow-up was 19 months, though nearly 25% of patients had a longer than 8-year follow-up. All except five patients received an induction-phase treatment. A glucocorticoid (oral or intravenous) was administered to 184 patients: to 72 as a single agent and to 112 in combination with another immunosuppressant, mostly cyclophosphamide and, less often, azathioprine or mycophenolate mofetil. Nineteen patients treated with prednisone and mycophenolate mofetil or azathioprine showed a higher favorable response rate (95%) compared to those patients treated with prednisone and cyclophosphamide (85%) or prednisone alone (83%). In all therapeutic schemes, improvement usually occurred within the first 2 months. Patients with histopathological evidence of small vessel involvement had a better treatment response compared to angiography diagnosed PACNS patients.
Thirty-five patients (19%) underwent maintenance therapy (17-month median duration) after induced remission; azathioprine, mofetil mycophenolate, and methotrexate were the immunosuppressants prescribed. A third of patients experienced at least one relapse; relapses were not significantly associated with the initial therapeutic scheme. Patients with angiography-based large-vessel involvement or both large- and small-vessel involvement had more relapses than those with small-vessel involvement alone. About a fifth of the patients were considered to be in long-term remission (longer than 12 months) after discontinuing maintenance therapy; this condition was not associated with the initial therapeutic regimen. Nearly a third of the 191 patients died during the follow-up; mortality was associated with increasing age, cognitive dysfunction, and cerebral infarction on diagnostic neuroimaging. Patients with lymphocytic PACNS showed a better functional outcome and lower mortality rates than patients with granulomatous or necrotizing PACNS (122).
Beneficial effects of other biological agents such as infliximab, etanercept, and tocilizumab in PACNS treatment have been scantily reported (48).
Peripartum cerebral angiopathy is a clinical-angiographic syndrome that initially was confused with primary CNS angiitis because of the similar angiographic appearances, but now there is convincing evidence that it behaves differently with normal CSF analysis and recovers without the use of immunosuppressive agents. It is a clinical syndrome characterized by reversible multifocal brain ischemia due to multilocular segmental narrowing of large- and medium-sized cerebral arteries (15). Patients usually present with severe generalized or occipital headaches with abrupt onset, similar to those seen in subarachnoid hemorrhage. Brain ischemia resulting from severe vasoconstriction typically leads to generalized motor seizure. Neurologic deficits such as cortical blindness, visual field deficits, confusion, aphasia, hemiparesis, and ataxia have been described. MRI usually shows edema evident as T2 hyperintensities that is mostly reversible. Calabrese and colleagues, acknowledged that two of the patients whom they reported as examples of “primary CNS angiitis" should now be considered cases of postpartum angiopathy (22).
All contributors' financial relationships have been reviewed and mitigated to ensure that this and every other article is free from commercial bias.
Julien Bogousslavsky MD
Dr. Bogousslavsky of the Swiss Medical Network has no relevant financial relationships to disclose.
See ProfileJorge Moncayo-Gaete MD
Dr. Moncayo-Gaete of Universidad Internacional del Ecuador has no relevant financial relationships to disclose.
See ProfileSteven R Levine MD
Dr. Levine of the SUNY Health Science Center at Brooklyn has no relevant financial relationships to disclose.
See ProfileNearly 3,000 illustrations, including video clips of neurologic disorders.
Every article is reviewed by our esteemed Editorial Board for accuracy and currency.
Full spectrum of neurology in 1,200 comprehensive articles.
Listen to MedLink on the go with Audio versions of each article.
MedLink®, LLC
3525 Del Mar Heights Rd, Ste 304
San Diego, CA 92130-2122
Toll Free (U.S. + Canada): 800-452-2400
US Number: +1-619-640-4660
Support: service@medlink.com
Editor: editor@medlink.com
ISSN: 2831-9125
General Neurology
Nov. 09, 2024
Neurogenetic Disorders
Oct. 23, 2024
General Neurology
Oct. 10, 2024
Neuroimmunology
Oct. 10, 2024
General Neurology
Oct. 05, 2024
Neuro-Oncology
Oct. 03, 2024
Neuro-Ophthalmology & Neuro-Otology
Sep. 25, 2024
Neuro-Ophthalmology & Neuro-Otology
Sep. 25, 2024