Cerebral amyloid angiopathy (CAA) refers to the deposition of β-amyloid in the media and adventitia of small and mid-sized arteries (and, less frequently, veins) of the cerebral cortex and the leptomeninges.[1] It is a component of any disorder in which amyloid is deposited in the brain, and it is not associated with systemic amyloidosis.
CAA has been recognized as one of the morphologic hallmarks of Alzheimer disease (AD), but it is also often found in the brains of elderly patients who are neurologically healthy.[2, 3] While often asymptomatic, CAA may lead to dementia, intracranial hemorrhage (ICH), or transient neurologic events. ICH is the most recognized result of CAA.
The Boston Cerebral Amyloid Angiopathy Group has elaborated guidelines for the diagnosis of cerebral amyloid angiopathy (CAA) associated with intracranial hemorrhage (ICH). Four levels of certainty in the diagnosis of CAA are considered: definite, probable with supporting pathologic evidence, probable, and possible. The first 3 require that no other cause of hemorrhage has been identified. The levels are characterized as follows:
Most cases of cerebral amyloid angiopathy (CAA) are sporadic, although genetic predispositions exist; eg, apolipoprotein E (ApoE) subtypes confer different risk profiles.
Most cases of CAA-related intracranial hemorrhage (ICH) are spontaneous. However, some evidence suggests that the amyloid is produced in the smooth muscle cells of the tunica media as a response to damage to the vessel wall (perhaps by arteriosclerosis or hypertension). Hereditary forms of CAA are due to specific gene mutations.
Amyloid deposition is complex and involves the following key processes:
Impaired elimination and accumulation of soluble and insoluble β-amyloid peptide may underlie the pathogenesis of CAA and explain the link between CAA and Alzheimer disease (AD). Amyloid fibrils may deposit in cerebral vessels, as in β-amyloid CAA, or form senile plaques in brain parenchyma.[4]
The dynamic between accumulation and clearance of amyloid may be related to impaired drainage from perivascular basement membranes. Interstitial fluid and solutes drain from brain to cervical lymph nodes along basement membranes of capillaries and arteries powered by the pulsatile flow in these vessels (reverse transport).[5]
Dutch type
Hereditary cerebral hemorrhage with amyloidosis-Dutch type is an autosomal-dominant disorder with complete penetrance. Among affected individuals, 87% have ICH and 13% have infarcts (deep).The age of onset of ICH is in the sixth decade (mean, 55 y). Some patients develop dementia without ICH.
Amyloid deposits are found in cortical and leptomeningeal vessels; parenchymal neurofibrillary tangles are not seen. Deposited amyloid protein in these patients is identical to the amyloid protein seen in sporadic cases, and the likely genetic defect is in the amyloid protein precursor protein (APP) gene on chromosome 21.
Icelandic type
Hereditary cerebral hemorrhage with amyloidosis-Icelandic type is also autosomal dominant. Patients present with their first episode of ICH in the third or fourth decade, with some patients as young as age 15 years dying from ICH. One case report identified a family with late-onset dementia with and without ICH.
The amyloid angiopathy is more widely distributed in this type than in other types, involving arteries in the cerebrum, cerebellum, and brainstem. The amyloid protein is a mutant of the cysteine protease inhibitor cystatin C.
Amyloid damages the media and adventitia of cortical and leptomeningeal vessels, leading to thickening of the basal membrane, stenosis of the vessel lumen, and fragmentation of the internal elastic lamina. These processes result in fibrinoid necrosis and microaneurysm formation, predisposing to hemorrhage.
CAA-related brain changes include lobar cerebral and cerebellar hemorrhage, leukoencephalopathy, small cortical ischemic infarcts, and plaque deposition. Leukoencephalopathy may be related to chronic hypoperfusion of deep WM (meningo-cortical segments of long perforators).
Neuropathologically, mild CAA primarily affects a relatively smaller proportion of the leptomeningeal and superficial cortical vessels, in contrast to the diffuse, significant deposition of amyloid in small arteries and arterioles seen in severe CAA. Medium-sized leptomeningeal arteries are affected with amyloid deposition in the outer portion of tunica media to tunica adventitia.
Frequently, complete erosion occurs, with only endothelium surrounding the deposit, predisposing to hemorrhage. Electron microscopy demonstrates fibrils of amyloid in the outer basement membrane in the initial stage of CAA. As the disease progresses, significant amyloid accumulation leads to tunica media degeneration, capillary and arteriolar infiltration, and formation of dystrophic neuritic plaques.
The risk of intracranial bleeding following head trauma and neurosurgical procedures is increased in patients with CAA. Some evidence suggests that CAA has a role in a substantial proportion of anticoagulant- and thrombolytic-related hemorrhages.
The true incidence and prevalence of cerebral amyloid angiopathy (CAA) are hard to specify, as definite CAA is a pathologic diagnosis typically obtained postmortem. However, estimates can be made based on autopsy series and the incidence of lobar intracranial hemorrhage (ICH). A series of 400 autopsies found evidence of CAA in the brains of 18.3% of men and 28% of women aged 40-90 years.
In a series of 117 brains of patients with confirmed AD, 83% had evidence of CAA.[6] The prevalence of CAA increases with advancing age; some autopsy series have found CAA in 5% of individuals in the seventh decade but in 50% of those older than 90 years. In patients with Alzheimer disease (AD), the incidence in several studies and meta-analyses has ranged from about 80-90%.
CAA is estimated to account for up to 15% of all ICH in patients older than 60 years and up to one half of nontraumatic lobar ICH in patients older than 70 years (approximately 15-20 cases per 100,000 people annually). CAA and CAA-related hemorrhage are particularly common in elderly individuals with AD and Down syndrome.
Upon autopsy, CAA may be found more commonly in women than in men; however, the incidence of ICH is the same in women and men.
The severity of CAA is age related, and more than 50% of patients in the tenth decade of life have evidence of CAA. Increasing age and the presence of AD are the only identified risk factors for CAA. Sporadic CAA-related ICH occurs in patients aged 60 years or older.
Familial forms of CAA are associated with hemorrhage at younger ages, by the third or fourth decade in the Icelandic form and by the sixth decade among the Dutch kindreds. Hemorrhage occurs at the same age in men and women.
The severity of angiopathy and fibrinoid necrosis closely correlate with the occurrence of intracranial hemorrhage (ICH). The most consistent clinical effect of cerebral amyloid angiopathy (CAA) is lobar ICH. Lobar ICH is associated with a lower mortality rate (11-32%) and a better functional outcome than are hypertensive deep ganglionic bleeds.
Of individuals with CAA-related hemorrhage, 25-40% have a recurrence, with the highest risk in the first year. Recurrent hemorrhages can occur simultaneously or several years later. They are associated with a high mortality rate (up to 40%).
In one series investigating lobar hemorrhage, the recurrence rate was reported to be 38% and the mortality rate high at 44%. Of the recurrences, 36% occurred in the same location.
Patients with a previous hemorrhage are at greater risk for subsequent hemorrhages than are those with no history. A history of hemorrhagic stroke before the index lobar hemorrhage can predict early recurrence of ICH.
Hypertension may exacerbate the tendency to suffer CAA-related hemorrhage and vice versa. Cortical petechial hemorrhage can be epileptogenic.
A higher number of ICHs at baseline on gradient-echo (GRE) magnetic resonance imaging (MRI) sequences is associated with a higher risk of future ICH, subsequent cognitive impairment, loss of independence, and death.
Cognitive impairment is a common feature of CAA. CAA is the most significant microscopic abnormality in 10-15% of patients diagnosed with Alzheimer disease (AD) by clinical criteria. More than 40% of patients with ICH have some degree of dementia. In some cases, the cognitive changes precede the ICH.
The relationship between CAA and AD is close. CAA, present in 80-85% of patients with AD, is severe in one third to two thirds of these patients.
Vascular lesions can play a significant pathophysiologic role and can contribute to the development of dementia in AD. The severity of CAA is correlated with the presence of ischemic or hemorrhagic lesions in the brains of patients with AD, and CAA is associated with gross strokes but not with subcortical lacunae.
Although CAA may contribute to the neurodegeneration of AD, a direct causal link between the 2 disorders has not been established. The association could be due to shared risk factors, such as the presence of APOE gene’s e4 allele.
Some patients with CAA present with a progressive dementia, involving rapid cognitive decline over days or weeks. This rapid progression could be due to the additive effects of severe vascular amyloid, cortical hemorrhages and infarctions, white matter destruction, and accumulation of neuritic plaques.
Few cases of vasculitis of various types (giant cell arteritis, rheumatoid vasculitis, primary angiitis of the central nervous system [CNS]) associated with CAA have been reported. No consensus exists as to whether the pathologic abnormalities are related causally or whether the appearance of vasculitis is a reaction to CAA-induced angiopathic changes.
Conditions to consider in the differential diagnosis of cerebral amyloid angiopathy (CAA) include the following:
Cerebral amyloid angiopathy (CAA) is frequently asymptomatic. However, it can manifest as one of several clinicopathologic entities. The most frequent are intracranial hemorrhage (ICH) and dementia.
CAA most often comes to clinical attention because of ICH. Symptoms range from transient weakness to coma, depending on the size and location of the hemorrhage. Patients may have recurrent episodes.
The most common symptom at onset is headache (60-70% of patients), with the location of the pain varying in accordance with the location of the hematoma, as follows:
Vomiting (in 30-40% of patients) tends to occur early. Seizures occur at onset in 16-36% of patients. Seizures are most commonly partial, with symptoms determined by the location of the ICH. As many as half of the patients present in status epilepticus.
Dementia may manifest as several patterns of cognitive dysfunction. Some cognitively normal patients present with rapid progression to profound dementia in a couple of years. Other patients can have a more protracted course, as is commonly seen in AD.
Stereotyped transient neurologic events commonly consist of focal weakness, paresthesias, or numbness. In some cases, these events may be prodromes for larger hemorrhages.
The symptoms spread to contiguous body parts over 2-10 minutes, and they may involve areas in several vascular territories. These events are probably due to small, petechial cortical hemorrhages that lead to focal seizures. The rate of spread is akin to that seen in migraine; some have proposed that these episodes may represent spreading depression of neuronal activity. Some patients present with transient confusion or episodes of visual misperceptions.
Uncommon presentations of CAA include the following:
Physical findings in cerebral amyloid angiopathy (CAA) depend on the disease process associated with CAA in a particular patient.
The features of intracranial hemorrhage (ICH) depend on the location of the bleed. Strict isolation of features from each lobe is frequently not possible because of extension of hematoma to other lobes, mass effect, and increased intracranial pressure. Even so, location-related presentations can be characterized as follows:
Coma at presentation has been reported in a small proportion of patients (0.4-19%) with ICH. A decreased level of consciousness, related to the size and location of the hematoma, results from compression of the contralateral hemisphere or brainstem or from increased intracranial pressure.
No specific laboratory findings are diagnostic of cerebral amyloid angiopathy (CAA). However, some patients may have cerebrospinal fluid (CSF) abnormalities; specifically, increased protein and decreased soluble β-amyloid or ApoE.
Genetic evaluation can be considered, especially in patients with a family history of CAA. In cases of CAA-related ICH, laboratory studies should rule out other possible etiologies.
With regard to electroencephalography, electroencephalograms (EEGs) may be diffusely abnormal, but they usually do not show evidence of seizure focus.
The severity of cerebral amyloid angiopathy (CAA) is associated with ApoE polymorphism. The APOE gene’s e4 and e2 alleles are risk factors for CAA. The e2 allele also confers an increased risk of intracranial hemorrhage (ICH) in patients with CAA. The APOE e4 allele is associated with earlier onset of first hemorrhage and carries a significant risk of concomitant Alzheimer disease (AD).[9] Patients with lobar ICH and the e2 or e4 allele have a greater risk of early recurrence.
These tests lack sensitivity and specificity and are not indicated as screening or diagnostic procedures. However, they may be helpful prognostic tools in identifying patients with a greater risk of early recurrence.
The following consultations are useful:
In computed tomography (CT) scanning, the finding of a single lobar hemorrhage with superficial location and cortical involvement with or without local extension to the subarachnoid and intraventricular spaces is suggestive of hemorrhage related to cerebral amyloid angiography (CAA). Evidence of multiple hemorrhages restricted to lobar regions may be present.
Hemorrhages are more common in the frontal and parietal lobes, involving the cortex and subcortical white matter. Over time, several lobes may be involved. Deep central gray nuclei, the corpus callosum, and the cerebellum are sometimes affected. CAA is rarely the cause of putaminal, thalamic, or brainstem hemorrhage.
Pure subarachnoid, intraventricular, and subdural hemorrhages can be seen but are rare. CAA should never be assumed to be the cause of an isolated subarachnoid hemorrhage unless all other causes, particularly aneurysmal, have been excluded.
Patients with CAA-associated dementia have leukoencephalopathy similar to that seen in Binswanger disease. Atrophy can also be detected, particularly in patients with cognitive impairment and a history of prior hemorrhage.
Florbetapir F18 (AMYViD) was approved by the FDA in April 2012 as a diagnostic imaging agent. It is indicated for PET brain imaging of beta-amyloid neuritic plagues in adults being evaluated for Alzheimer disease or other cognitive decline. A second 18F-labeled Pittsburgh compound B (PIB) derivative, flutemetamol F18 (Vizamyl), was also approved in October 2013.
A third agent, florbetaben F 18 (Neuraceq), was approved by the FDA in March 2014. Images may be obtained between 45-130 minutes following the injected dose. FDA approval was based on safety data from 872 patients who participated in global clinical trials as well as 3 studies that examined images from adults with a range of cognitive function, including 205 end-of-life patients who had agreed to participate in a post-mortem brain donation program. Images were analyzed from 82 subjects with post-mortem confirmation of the presence or absence of beta-amyloid neuritic plaques.[10]
In PET scanning, cortical retention of Pittsburgh Compound B (binds beta-amyloid) may serve as an in vivo marker of CAA in humans.[11]
MRI may show evidence of multiple large and small, petechial cortical and subcortical hemorrhages, even in patients without a history of previous hemorrhage. In asymptomatic patients, clinically silent microhemorrhages may serve as a marker of disease progression.[12, 13]
GRE MRI sequences show evidence of hemosiderin deposition that corresponds to old hemorrhages. In patients who present with lobar hemorrhages, evidence of old petechial bleeds can help in the diagnosis of cerebral amyloid angiopathy (CAA).[14]
On GRE sequences, punctate (usually < 5mm), round hypointensities, termed microbleeds, are frequently identified in white matter. Although these cerebral microhemorrhages are often present in amyloid angiopathy, they are not diagnostic of amyloid pathologically. Any conclusions regarding the significance of cerebral microbleeds must be interpreted given the individual patient or population being evaluated.
Microbleeds may be associated with hemorrhagic transformation of ischemic stroke. Microbleeds may be more common in patients with hypertension, but no characteristic pattern occurs in the distribution of microbleeds. Microbleeds may suggest a hemorrhage-prone angiopathy involving brain parenchyma distant from identified microbleeds.
The presence, or number, of microbleeds may impact decisions to administer thrombolytic, anticoagulant, or antiplatelet therapy.
As previously mentioned, higher number of intracranial hemorrhages (ICHs) at baseline on GRE is associated with a higher risk of future ICH, subsequent cognitive impairment, loss of independence, and death. Leptomeningeal enhancement is seen in patients with associated vasculitis.
Angiographic findings are abnormal only in rare cases of cerebral amyloid angiography (CAA) ̶ related vasculitis. Specificity and positive predictive value in such cases is less than 30%.
Given that some of the features of CAA and vasculitis are similar, a high index of suspicion is required. Angiography should be considered in patients with a history of hemorrhages or ischemic strokes with rapid cognitive decline (over weeks or a few months), prominent headaches, and seizures.
Histologic examination is required for the definitive diagnosis of cerebral amyloid angiopathy (CAA). Pathologic samples are obtained from hematoma evacuation, cortical biopsy, or postmortem specimens. The disease process may be diffuse, however, so pathologic data may be lacking even in biopsy cases. (Brain biopsy has a sensitivity of 53% and a negative predictive value of 70%.)
The presence of vascular amyloid is a sensitive marker for CAA-related hemorrhage. β-amyloid consists of twisted β-sheet fibrils in a vessel wall. It is a homogenous, intensely eosinophilic material that gives a smudged appearance by light microscopy. When stained with Congo red and visualized under polarized light, it gives a characteristic yellow-green (ie, apple green) birefringence. When thioflavin T and S are used and visualized with ultraviolet light, amyloid appears fluorescent.
The presence of fibrinoid necrosis in amyloid-laden vessels is relatively specific for CAA-related intracranial hemorrhage (ICH). CAA, which involves cortical and leptomeningeal vessels, is most common in the parietal and occipital lobes.
Parenchymal features found in the brains of patients with CAA include patchy demyelination and loss of white matter, cortical hemorrhages and infarcts, and neuritic plaques with or without neurofibrillary tangles. Patients with CAA have been found with a progressive increase in white matter lesions; this may suggest a progressive microangiopathy leading to incident lobar hemorrhage.[15] Most patients with CAA-related ICH do not have Alzheimer disease (AD).
Cerebral amyloid angiopathy (CAA) is largely untreatable at this time. The management of CAA-related intracranial hemorrhage (ICH) is identical to the standard management of ICH. Pay special attention to the reversal of anticoagulation, the management of intracranial pressure, and the prevention of complications.
If coexisting vasculitis is found on angiography and brain biopsy, long-term treatment (up to 1y) with steroids and cyclophosphamide is indicated.
A syndrome of subacute cognitive decline, seizures, and white matter changes on MRI, with perivascular inflammatory changes on biopsy, was described, with some patients improving clinically (but not to baseline) when given corticosteroids or cyclophosphamide.
Although early investigations showed Cerebril (Neurochem, Inc), a drug developed to reduce amyloid formation and deposition, to be safe, this drug is currently not being actively studied for CAA.
An analysis of ICH subgroups in the PROGRESS (Perindopril Protection Against Recurrent Stroke Study) trial indicated that, over a follow-up period of 3.9 years, patients treated with the antihypertensive drug perindopril had a reduction in CAA-related ICHs of 77%. However, the total number of CAA-related ICHs was small (16 probable CAA-related ICHs), so definitive conclusions await larger samples.[16]
Hematoma evacuation can be lifesaving when the hematoma causes significant mass effect and predisposes to herniation, particularly when medical management of increased intracranial pressure yields no response. The goal of therapy is to lower intracranial pressure.
No evidence is available from well-designed, randomized clinical trials that can help to determine which patients benefit from evacuation of the hematoma. It is agreed, however, that the intervention should be considered in patients with intermediate-sized hematomas (20-60mL) who have a progressive deterioration in their level of consciousness. Surgery should be performed before coma develops.
Surgery is not beneficial for small or very large hematomas. Patients with small (< 20mL) hematomas and minimally decreased levels of consciousness tend to have good outcomes with conservative treatment. When the hematoma is large (>60 mL) and the patient is lethargic or comatose, the prognosis is poor despite surgical evacuation.
Early concerns about the safety of hematoma evacuation in patients with intracranial hemorrhage (ICH) related to cerebral amyloid angiopathy (CAA) were unfounded. Several series have reported low rates of mortality and postoperative hematoma; surgical evacuation of the hematoma should be performed when clinically indicated.[17] When determining whether evacuation of the hematoma is appropriate, consider the patient's cognitive status.
No evidence supports the belief that evacuation leads to an increased rate of recurrence. A large series that evaluated 50 neurosurgical procedures in 37 patients with CAA-related ICH found an 11% mortality rate and a 5% rate of postoperative hematoma that required intervention.[18] Risk factors associated with an adverse postoperative outcome were age older than 75 years and the presence of a parietal hematoma.
Although transoperative oozing from the walls of the hematoma was a common occurrence in the study, it could be controlled easily with an absorbable hemostat (eg, oxidized cellulose, gelatin sponge) or fibrin glue.
Patients with cerebral amyloid angiopathy (CAA) have an increased risk of bleeding while taking warfarin, even when the level of anticoagulation is in the therapeutic range (ie, international normalized ratio, 2-3). The vasculopathic changes may predispose these patients to small bleeds. The use of anticoagulants may result in the enlargement of small hemorrhages that otherwise would have remained asymptomatic.
Withdrawal of anticoagulant agents is a prudent intervention to prevent recurrences in patients with prior lobar hemorrhages, particularly if GRE MRI suggests earlier petechial hemorrhages. Antiplatelet agents are a safer alternative.
Strong evidence regarding the relationship between CAA and antithrombotic therapy is lacking. Data suggested that GRE microbleeds were more prevalent among a cohort of aspirin users in the Rotterdam Scan Study; however, more research is needed to understand the dynamic between antithrombotic therapy and the risk of intracerebral hemorrhage and microbleed formation.[19]
In the patient with coronary artery disease, cardiac stents, and/or ischemic stroke, the benefit of antithrombotic therapy is clear and withdrawal of antiplatelet therapy requires prudent consideration of multiple factors. Management must be tailored to each individual case, taking into account the risk of hemorrhage, the benefit of stroke prophylaxis, and the preferences of the patient.
Two studies support the treatment of hypertension to prevent hemorrhage recurrence in CAA. The first, an analysis of patients in the PROGRESS trial, showed that those on active therapy with perindopril had fewer recurrences of all types of hemorrhage than those in the control group.[16] A second study, a single center cohort study of patients with lobar and nonlobar hemorrhages, suggested that inadequate blood pressure control during follow-up was associated with higher risk of lobar as well as nonlobar ICH recurrence.[20]
In a small study of liver transplantation in patients with amyloidogenic transthyretin (ATTR) Tyr11, a hereditary cause of CAA, mortality and the occurrence of cerebral hemorrhage and dementia in 3 patients who had transplantation were lower than in 5 patients who did not. However, the small number of patients in the study made it difficult to know how generalized the results were.[21]
Activities in patients with CAA should not be restricted. However, patients should avoid any degree of head trauma.
MRI Brain GRE 3.0 TeslaTop: Multiple scattered microbleeds in multiple vascular territories in patient with hyperhomocysteinemia and methylene-tetrahydrofolate reductase mutation. Middle: Acute parenchymal hematoma with typical appearance of right occipital microbleed and left parietal chronic microhemorrhage.Bottom: Microbleeds involving multiple lobes. Chronic hemorrhage in the left parietal lobe with typical slit-like appearance.
