Clinicopathological Study of Intracranial Fusiform and Dolichoectatic Aneurysms

Subjects and Methods

From April 1983 to December 1998, we treated 16 patients with chronic fusiform aneurysms. Imaging studies on all cases included cerebral angiography, CT scan, and MRI. All cases had 2 features on cerebral angiography that we chose as criteria for chronic fusiform aneurysms: (1) fusiform arterial ectasia at a nonbranching site of the intracranial cerebral arteries and (2) no obvious atherosclerotic changes found in other intracranial arteries. Dissecting aneurysms that showed typical “pearl–and-string” sign or clear double lumen on cerebral angiography were excluded; we saw 52 such typical dissecting aneurysms during the study period.

Medical records, including all imaging studies, were reviewed. Follow-up MRI was obtained at least every 12 months in all cases except for in 1 patient who died at an acute stage. As previously described, we made a diagnosis of IMH on MRI when abnormal high-signal intensity on T1-weighted images was observed along the tortuous dilated artery.^{10 11}

Aneurysm tissues were obtained during surgery in 2 cases and by autopsy in 6. For histological examination, tissues were fixed in 10% formalin and embedded in paraffin. Tissue sections (4 μm each) were made and stained with hematoxylin and eosin, elastica–van Gieson, and Masson’s trichrome stains. For immunohistochemistry, a polyclonal antibody for factor VIII–related antigen (Dako Corp) was used as a primary antibody at 1:1500 dilution, and the Elite ABC kit (Vecstein Laboratory) with nickel enhancement was used for coloring reaction; manufacturer’s protocols were followed. Degree of IEL fragmentation was classified into 4 groups based on the ratio of the part without IEL relative to the whole circumferential length: none, <15%; slight, 15% to 30%; moderate, 30% to 50%; and severe, >50%. Degree of neoangiogenic vessel formation within the hyperplastic intima or of recanalizing vessel formation in the thrombus was evaluated on axial sections and was rated on the basis of the ratio of the area occupied by these vessel lumen relative to the whole area of the axial section of the aneurysm wall: none, <15%; slight, 15% to 30%; moderate, 30% to 50%; and severe, >50%. Degree of IMH and luminal thrombus formation were also assessed on axial sections and were rated on the basis of the ratio of the area occupied by the hemorrhage or thrombus to the whole area of the aneurysm: none, <15%; slight, 15% to 30%; moderate, 30% to 50%; and severe, >50%. Degree of intimal hyperplasia was rated as the ratio of intimal thickness to the whole wall thickness: none, <15%; slight, 15% to 30%; moderate, 30% to 50%; and severe, >50%.

Results

MRI Findings

IMH was observed at initial examination in 8 of 9 symptomatic cases. CEI was found in all cases (Figures 1B⇓ and 1D⇓). On serial MRIs, these symptomatic aneurysms showed gradual enlargement in all cases with IMH. On the other hand, 6 of 7 asymptomatic aneurysms showed neither IMH nor CEI at initial examination and did not show enlargement on serial examinations. However, in 1 case, CEI already was evident at initial MRI (Figure 1B⇓). Notably, all 8 aneurysms with IMH were >20 mm in diameter, and all 10 aneurysms with CEI were >12 mm in diameter. We did not notice any difference in degree of tortuousness or irregularity of unaffected arteries of studied patients compared with those of the general population in a similar age range. A summary of the MRI findings is presented in the Table 1⇓.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1.

Neuroradiological imagings of representative cases (A through D) with corresponding histology (E through H). Case 3 (A and E), (A) Cerebral angiogram demonstrates relatively small right MCA fusiform aneurysm (8 mm diameter). E, Histology (by van Gieson stain) shows fragmentation of IEL (*) and intimal thickening but no neoangiogenesis. Case 7 (B and F), (B) T1-weighted MRI enhanced on coronal view shows 12-mm fusiform aneurysm of left IC. Marked enhancement of aneurysm wall (arrowhead) is present, but no IMH is seen. F, Histology (van Gieson) reveals IEL fragmentation (*) and intimal thickening accompanied by moderate degree of recanalizing vessel formation (arrowheads). Case 11 (C and G), (C) Nonenhanced T1-weighted MRI shows axial section of 28-mm dolichoectatic aneurysm of right vertebral artery with IMH (arrow). G, Histology (van Gieson) of resected aneurysm reveals marked IEL fragmentation and degeneration (*) and intimal thickening (small arrow) that contains thrombus from old and fresh IMH (large arrow). Case 14 (D and H), (D) T1-weighted MRI without (left) and with (right) gadolinium injection demonstrates advanced-stage dolichoectatic aneurysm of basilar artery, 32-mm diameter. IMH shows slightly high intensity with patchy, laminated enhancement (arrowheads), whereas an area around the recent IMH is markedly enhanced, which indicates neovascularization (arrowheads). H, Histology of aneurysm at autopsy after fatal rupture shows multiple dissection within thickened intima and severely fragmented IEL (*) as well as formation of vasa vasorum within the adventitia (van Gieson). All bars indicate 400 μm.

View this table:

• View inline
• View popup

Table 1.

Clinical Summary of 16 Patients With Fusiform and Dolichoectatic Aneurysms

Histological Examination of Aneurysms

Four characteristic features were noticed: (1) IEL fragmentation and intimal hyperplasia; (2) neoangiogenesis in the thickened intima; (3) IMH and luminal thrombosis; and (4) recanalizing vessel formation in the thrombus. Fragmentation of the IEL and intimal hyperplasia were found in all cases, but extent of IEL fragmentation was more prominent in larger or symptomatic aneurysms (Figures 1E⇑ through –1H). In addition, degeneration of the media was also noticed universally. Neoangiogenic vessel formation within the thickened intima was observed in 5 of 8 cases (Figures 1F⇑ and 2C⇓) and 4 of them accompanied IMH (Figures 1G⇑, 1H⇑, and 2A⇓). Immunohistochemistry for factor VIII–related antigen strongly immunolabeled the endothelial cells of the neoangiogenic vessels, which confirmed the vascular origin of these structures. IMH consisted of fresh IMHs around the neoangiogenic vessels and generally coexisted with old, laminated thrombus. Old thrombus constituted the largest portion of the aneurysms in most cases and occasionally contained recanalizing vessels (Figure 1G⇑, 1H⇑, 2B⇓, and 2D⇓).

• Download figure
• Open in new tab
• Download powerpoint

Figure 2.

Recanalizing vessel formation in case 14. New vessels within thickened intima (A and C) and thrombus (B and D) are demonstrated more clearly by immunostaining for factor VIII–related antigen (C and D) than by van Gieson stain (A and B), which strongly stains endothelial cells lining the vessel lumen (arrowheads). Bar indicates 400 μm. Asterisk indicates severely fragmented IEL.

When the MRI features of the 8 cases were compared with histological findings, CEI was associated with existence of neoangiogenic vessels within the thickened intima: all 5 cases with intimal neoangiogenesis but none of the 3 cases without neoangiogenesis showed CEI on MRI. MRI detected IMH in all 3 cases with histologically confirmed IMH. Similar to the relationship between MRI findings and size of the aneurysm, good correlation existed between the histological features and the size of the aneurysms: IMH was seen exclusively in aneurysms >28 mm in diameter, and neoangiogenesis was seen exclusively in aneurysms >12 mm in diameter. Results of the histological examination are also summarized in the Table⇑.

Discussion

Data given in the present article demonstrate characteristic histological features of chronic fusiform aneurysms, and these histological findings are closely associated with size of the aneurysm. Universal histological features were fragmentation of IEL and intimal hyperplasia. Therefore, these 2 changes are probably among the earliest processes of chronic fusiform aneurysm formation. Neoangiogenesis within hyperplastic intima was the next most common finding, which corresponded well to CEI observed on MRI. Notably, this radiological finding was not usually found in small aneurysms but was observed in aneurysms >10 mm in diameter. The third most common but the most drastic changes was IMH and recanalization of thrombus formed by the IMH, which were observed in aneurysms >20 mm in diameter. Of specific clinical importance were the observations that IMH was detectable on MRI and aneurysms with IMH progressed in a relatively short period without exception, which often resulted in serious outcomes.

Some of the histological features described in our series are similar to atherosclerotic changes, but our observations indicate the causes of fusiform aneurysm may be different from those of atherosclerosis. Recanalizing vessels in a thickened intima, which occasionally cause bleeding, are also found in atherosclerotic arteries at an advanced stage. Disruption or fragmentation of the IEL is found also in atherosclerosis, but not at an early stage of the disease. In atherosclerosis, early changes include intimal cell proliferation around lipid deposits and duplication or thinning of IEL, neither of which was observed in our cases. Therefore, the most prominent difference between atherosclerosis and fusiform aneurysms seems to be at the initial step of the disease: lipid deposition in atherosclerosis and IEL fragmentation in fusiform aneurysms. Therefore, it is likely that the primary abnormality in fusiform aneurysms could lie in the IEL. In support of this notion, 2 of the present cases were similar to those of patients with Ehlers-Danlos syndrome, a systemic disease related to an abnormality in collagen, which is a key component of the IEL.

Data presented in the present study suggest a possible mechanism for progression of chronic fusiform aneurysms, a stepwise progression: (1) The earliest detectable change seems to be IEL fragmentation, almost immediately followed by intimal hyperplasia, possibly as a normal reaction to the damage. (2) When intimal thickening reaches a certain level, neovascularization within the thickened intima occurs. (3) New vessels within the intima then cause bleeding and IMH. (4) Repeated recanalization of thrombus and further bleeding from those vessels lead to rapid growth. Among those steps, IMH seems to be the most critical event, because it apparently forces the aneurysms to progress and, in most cases, become fatal. Although further studies are needed to confirm this hypothesis, knowledge of this possible mechanism of progression and of corresponding MRI characteristics could help to determine the timing of surgical intervention. Specifically, a surgical intervention to decrease the risk of IMH should be considered when CEI is detected on MRI, because this probably indicates neovascularization in thickened intima, which could herald IMH and rapid aneurysm growth. Once IMH occurs, further progression seems to be inevitable. In our series, 6 cases were treated surgically: 3 underwent parent artery occlusion and 3 had clip reconstruction of the parent artery. The aneurysms disappeared in all 6 cases, and no recurrences have been observed. In contrast, 4 patients with IMH who were observed without surgical intervention or with only a distal bypass showed aneurysm growth without exception. Therefore, appropriate surgical procedures should help patients by preventing progression of the aneurysms.

With wider availability of high-resolution MRI, we are now able to diagnose various previously undetectable vascular lesions¹² and are likely to encounter more cases of asymptomatic fusiform aneurysms. The data and hypothesis presented in the present article could be helpful for establishment of management policy for such lesions. These data also indicate the need for further studies to establish the natural course of chronic fusiform aneurysms.