Secondary Logo

Journal Logo

Original article

Arterioles in cerebral amyloid angiopathy and vascular dementia

ZHU, Guang-ming; ZHANG, Wei-wei; LIU, Ying; LI, Juan

Author Information
doi: 10.3760/cma.j.issn.0366-6999.2009.24.013
  • Free


Cerebral amyloid angiopathy (CAA) is a disease of the elderly, characterized by the deposition of amyloid in cortical and leptomeningeal vessels, which most commonly presents as lobar hemorrhage or as part of a dementia process.1 It often overlappes with vascular dementia (VaD). Small cerebrovascular lesions are one of the most important factors in CAA and VaD. Recent studies suggest that multiple microvascular lesions with impact on deep cerebral structures, rather than large lobar infarcts, are the major substrate of VaD.2,3 So it is very important for us to understand the pathological changes of small vessels in these diseases.

Although imaging can provide some useful information in some studies about CAA and VaD, pathological studies on the arterioles are scarce. In this study, we analyzed some indicators of arteriolar fibrosis and stenosis in CAA patients (CAAs) and VaD patients without CAA (VaDs), including the external and internal diameter of arterioles, thickness of arteriolar wall and sclerotic index (SI) in gray matter and white matter of cerebral hemisphere, to elucidates the discrepancy between the arterioles of VaDs and CAAs, as well as aging and hypertension.4,5


Patients and controls

We examined more than 50 brains necropsied in Beijing Military General Hospital from 1985 to 2000. Ten deceased CAA patients (CAAs, 6 females and 4 males, mean age (73.1±5.0) years, range from 65 to 80 years) and twelve deceased vascular dementia patients without CAA (VaDs, 5 females and 7 males, mean age (71.6±5.4) years, range from 64 to 82 years) were available for this study. Patients who suffered with both CAA and VaD were included in the CAA group.

All CAAs were confirmed by congo-red and β-amyloid staining after autopsy. All patients had a stroke history, three had hypertension, and two had diabetes. Four of them (3 males and 1 female) were also suffering from dementia. The mini-mental state examination (MMSE) before death were 19, 21, 17 and 13. But mini-mental state examination of the other CAAs were not available.

All VaDs were consistent in clinical and pathologic diagnosis to exclude Alzheimer's disease, CAAs or other non-vascular dementia. All patients had a stroke history. Ten of them had hypertension and six had diabetes. Mini-mental state examination had been done before their mortuus (mean score 13.4±4.0, range from 7 to 20).

Five deceased patients (3 females and 2 males, mean age (71.8±4.0) years, range from 68 to 78 years) without known cerebrovascular diseases served as controls.

Histology and immunohistochemistry

All the brains were fixed in 4% phosphate buffered formaldehyde solution and the selected regions were embedded in paraffin. Representative samples of the frontal lobe were selected for this study. Sections (5-μm thickness) representing both cortical and subcortical areas were cut and stained with hematoxylin and eosin. All CAA and VaD sections were confirmed by Congo red and anti-β-Amyloid monoclonal mouse antibody (Maxim, China) staining. Additional sections were immunostained for detection of α-smooth muscle actin with α-smooth muscle-actin monoclonal mouse antibody (Boster, China). The bound primary antibodies were visualized by using the appropriate peroxidaselabeled secondary antibodies with diaminobenzidine as hematoxylin as counterstain.

Morphometric study

All transversely cut arterioles in gray matter and white matter with an external diameter equal or greater than 30 μm and with a maximum of 300 μm were examined from each patient and control in one representative section of the frontal lobe extending from the surface of the cortex deep into the white matter. Since these sections were cut perpendicular to the brain surface and parallel to the penetrating arteries, virtually all of the measured arterioles seen as cross sections were branches of the primary penetrating arteries.

The internal and external diameters of arterioles were then measured on screen measurements. In arterioles with elliptical profiles, the lesser diameter, i.e., the diameter perpendicular to the maximum axis, was measured. The external diameter was measured at the same orientation and corresponded to the diameter spanning external boundaries of the connective tissue. The internal diameter was measured as the diameter of the arteriolar lumen. The thickness of the arterioles was calculated as the external diameter minus the internal diameter. The sclerotic index (SI) was defined and calculated as: 1-(internal diameter/external diameter).

Statistical analysis

Data are presented as means±standard deviation (SD). The external and internal diameter of the arterioles, thickness of the arteriolar wall and SI in CAAs and VaDs were compared with those of the controls using one-way analysis of variance (ANOVA). P <0.01 was considered significant.


Histology and immunohistochemistry

In the arterioles of the CAAs, congo-red-positive substance could be observed in the wall of CAAs arterioles, although expression was not uniform, even in the same visual field (Figure 1). β-Amyloid accumulated in the tunica media (Figure 2). Clear immunopositivity for α-smooth muscle-actin existed in the tunica media of the control group and the appearance of vessel and α-smooth muscle-actin staining were normal (Figure 3), whereas the staining was irregular or negative in the CAAs and VaDs (Figures 4 and 5). In CAAs the lumen of the vessel was not as narrow as in the VaDs, but α-smooth muscle-actin positive cells were rare (Figure 5). Although immunoreactivity for α-smooth muscle-actin in VaDs was much stronger than in CAAs, the thickness of the vessel was markedly increased and the lumen was nearly obstructed (Figure 5).

Figure 1.
Figure 1.:
CAAs group. Congored-positive substance could be observed in the wall of CAAs arterioles, though the degrees were dissimilar, even in same visual field (Congo-red staining, original magnificantion×100).
Figure 2.
Figure 2.:
CAAs group. β-Amyloid accumulated in the tunica media of CAAs (Immunohistochemistry staining of β-Amyloid, original magnificantion×400).
Figure 3.
Figure 3.:
Control group. In control group, the appearance of vessel and α-SM-Actin staining were normal (Immunohistochemistry staining of α-SM-Actin, original magnificantion×200).
Figure 4.
Figure 4.:
CAAs group. In CAAs, the lumen of vessel was not so narrow, if compare with VaDs. But α-SM-actin positive cells were rare, which demonstrated the loss or degeneration of VSMCs (Immunohistochemistry staining of α-SM-Actin, original magnificantion×400).
Figure 5.
Figure 5.:
VaDs group. More α-SM-actin positive cells existed in the vessel of VaDs, which suggested the over-proliferation of VSMCs. The lumen of the vessel was so narrow that the vessel wall was significantly thickened (Immunohistochemistry staining of α-SM-actin, original magnificantion×200).

Morphometric analysis of the arterioles

The indicators of arteriolar fibrosis and stenosis, external diameter of the arterioles, the internal diameter, thickness of the arteriolar wall and sclerotic index, as well as the numbers of arterioles measured in the cortical gray matter and white matter are given in the Table.

Data of arterioles in gray matter and white matter in CAAs, VaDs and controls (mean±SD)

External diameter of arterioles

The external diameter of the gray matter in the CAAs was significantly greater than the controls (P=0.004), whereas the VaDs showed no significant difference if compared with CAAs (P=0.087) or the controls (P=0.181). Though the external diameter of arterioles in white matter of VaDs was the greatest, there are no marked difference between each group.

Internal diameter of arterioles (diameter of arteriolar lumen)

In gray matter arterioles the diameter of the lumen in the VaDs was markedly smaller than in the CAAs (P=0.006). The lumen of the control did not show a significant difference with CAAs (P=0.962) or VaDs (P=0.029). The differences also did not reach significance in the white matter arteriolar lumen when comparing the three groups (VaDs vs CAAs, P=0.581; CAAs vs Control, P=0.144; VaDs vs Control, P=0.054).

Thickness of arteriolar wall

Analysis of the thickness of the arteriolar wall revealed that both CAAs and VaDs might cause remarkable thickening when compared with controls, not only in white matter (CAAs vs control, P=0.003; VaDs vs control, P <0.001), but also in GRAY MATTER (P <0.001). Although those thickness of VaDs arteriolar wall were larger than CAAs. There was no significant difference in gray matter between VaDs and CAAs (P=0.659). In white matter of the two groups, the result was similar (P=0.235).

Sclerotic index (SI)

The SI of the VaDs in the gray matter was largest followed by the CAAs. Group comparison could reach marked differences with all of the P values less than 0.001. In the white matter we got the same result.

All these differences demonstrated that arteriolar fibrosis and stenosis of the VaDs arterioles were more severe in either gray matter or white matter, when compared with CAAs and controls. The gray matter external and internal diameter of CAAs was much greater than other groups, which may indicate expanding of arterioles, but this tendency was not observed in white matter.


CAA is a disease of the elderly, characterized by the deposition of amyloid in cortical and leptomeningeal vessels.1 Severe CAA, particularly when associated with secondary vascular pathology, may lead to vascular dementia-like ischemic changes.6 Vascular dementia is defined as the loss of cognitive function resulting from ischemic, ischemic-hypoxic, or hemorrhagic brain lesions as a result of cerebrovascular diseases.7 It is difficult to make the distinction between patients with widespread CAA and those suffering from vascular dementia.

CAA-associated vascular changes include smooth muscle cell degeneration, hyalinization, ‘double-barreling’ phenomenon, macrophage infiltration, and aneurysmal dilatation of the vessel wall.6 Accumulation of β-Amyloid in the vascular media and/or adventitia is associated with degeneration of endothelial cells and vascular smooth muscle cells (VSMCs), or even defects in the vessel wall.1,6

VaD is the second most common type of dementia.7 The main clinicopathological subtypes of VaD are large vessel and small vessel disease,7,8 the latter being more prevalent.2,3 Large vessel VaD results from strategic large vessel strokes. Small vessel VaD may result from multiple subcortical lacunar infarcts, bilateral thalamic lesions, or from diffuse white matter lesions. Subcortical ischemic small vessel VaD is currently recognized as the most broad and homogeneous subtype of VaD.9,10

Some well defined vasculopathies known to cause dementia are proven, such as cerebral autosomal-dominant arteriopathy with subcortical infarcts and leukoencephalopathy (CADASIL)4,5 or some types of cerebral amyloid angiopathies. Since, in this study, we describe the microscopic arteriolar findings in CAA and VaD, patients who suffered with both CAA and VaD will be included in the CAA group.

Severe loss or even complete absence of α-SM-Actin was observed in most of the CAA-affected vessels, illustrating the degeneration or loss of VSMCs. But in VaDs, although the sclerosis of arteriolars is more severe than CAAs, the expression of α-SM-Actin was much stronger than CAAs and the number of VSMCs was also greater than CAAs. All of these demonstrated that in CAAs, loss and degeneration of VSMCs were predominant, thus the over-proliferation of VSMCs was more significant in VaDs. CAA is also associated with fibrinoid necrosis and microaneurysm formation. Ultimately, these degenerative changes may lead to impaired vessel wall integrity, cumulating in rupture and cerebral hemorrhages.

In our study, the lumen and external diameter of CAAs arterioles in gray matter is much greater than in other groups, although those in white matter are without this tendency. It demonstrated the gray matter arterioles were dilated and their wall were not markedly thickened. All these points suggested that microaneurysm formation is more severe in gray matter of CAAs, which could explain why lobar hemorrhage occur more often in CAAs.

Arterial sclerosis occurs in both CAAs and VaDs, but the SI of VaDs is the greatest, not only in white matter, but also in gray matter. Hypertension may be the main factor for this. In our CAAs group hypertension history is rare (3/10), but in the VaDs group nearly 90% patients (10/12) had hypertension before mortuus.

In conclusion, fibrotic thickening and stenosis of the walls of arterioles occur both in CAA and VaD, but the tendency is more frequent in VaD. Arterioles of CAA also display the expanding in gray matter, which may be related with lobar hemorrhage.


1. Wong SH, Robbins PD, Knuckey NW, Kermode AG.. Cerebral amyloid angiopathy presenting with vasculitic pathology. J Clin Neurosci 2006; 13: 291-294.
2. Scheltens P, Kittner B. Preliminary results from an MRI/CT-based database for vascular dementia and Alzheimer′s disease. Ann N Y Acad Sci 2000; 903: 542-546.
3. Esiri MM. Which vascular lesions are of importance in vascular dementia? Ann N Y Acad Sci 2000; 903: 239-243.
4. Miao Q, Paloneva T, Tuisku S, Roine S, Poyhonen M, Viitanen M, et al. Arterioles of the Lenticular Nucleus in CADASIL. Stroke 2006; 37: 2242-2247.
5. Miao Q, Kalimo H, Bogdanovic N, Kostulas K, Börjesson-Hanson A, Viitanen M. Cerebral arteriolar pathology in a 32-year-old patient with CADASIL. Neuropathol Appl Neurobiol 2006; 32: 455-458.
6. van Horssen J, de Jong D, de Waal RM, Maass C, Otte-Holler I, Kremer B, et al. Cerebral smyloid angiopathy with severe secondary vascular pathology: A histopathological study. Dement Geriatr Cogn Disord 2005; 20: 321-330.
7. Jellinger KA. Understanding the pathology of vascular cognitive impairment. J Neurol Sci 2005; 229-230: 57-63.
8. Jellinger KA. The pathology of ischemic-vascular dementia: An update. J Neurol Sci 2002; 1203-204: 153-157.
9. Roman GC, Erkinjuntti T, Wallin A, Pantoni L, Chui HC. Subcortical ischaemic vascular dementia. Lancet Neurol 2002; 1: 426-436.
10. Erkinjuntti T, Inzitari D, Pantoni L, Wallin A, Scheltens P, Rockwood K, et al. Research criteria for subcortical vascular dementia in clinical trials. J Neural Transm Suppl 2000; 59: 23-30.

arterioles; cerebral amyloid angiopathy; vascular dementia

© 2009 Chinese Medical Association