Site-Specific Relationship Between Intracranial Aneurysm and Aortic Aneurysm
Background and Purpose—The high prevalence of intracranial aneurysms (IAs) in patients with a bicuspid aortic valve or coarctation of the aorta suggests a link between IA and aortic pathology. However, studies reporting this link do not sufficiently address the heterogeneity of IAs arising from different anatomic locations. This study aimed to explore whether a location-specific relationship exists between the 2 kinds of aneurysms.
Methods—Retrospective institutional analysis of patients aged ≥18 years with both IA and an aortic aneurysm (AA) was performed from 2005 to 2014. IAs were categorized based on their locations: internal carotid artery, other anterior circulation, and posterior arteries. AAs were classified as ascending, descending, infrarenal, or multiple. We analyzed the clinical characteristics and the distribution of IA in each AA group.
Results—Of 2375 patients, 660 with available intracranial angiography were screened for IA. We identified 71 patients with 97 IAs. The frequency of both anterior circulation-IAs and internal carotid artery-IAs differed significantly among the AA groups (P=0.001 and P=0.01, respectively). Anterior circulation-IAs were most frequently observed in ascending AA group and least frequently in infrarenal AA group. In contrast, internal carotid artery-IAs were found mostly in infrarenal AA group, least in ascending AA group. Proportions of patients having anterior circulation-IA and internal carotid artery-IA were also highest in ascending AA group and infrarenal AA group, respectively. The number of posterior arteries-IAs was too small to characterize.
Conclusions—The differing distribution patterns of IA among AA groups suggest a site-specific sharing of pathomechanism between the 2 types of aneurysms.
Despite the high prevalence (3%) of intracranial aneurysms (IAs) in the general population1 and the devastating consequences of their rupture,2 the pathogenic mechanisms underlying IA are still poorly understood. A significant proportion of patients with IA are young, without risk factors, and few clues have been found to predict who will develop an IA and in which artery. There have been anecdotal reports concerning the relationship between IAs and other vasculopathies, such as cervicocephalic arteriopathies, bicuspid aortic valve, coarctation of the aorta, aortic aneurysm (AA), and dissection,3–5 suggesting a common pathophysiology.
Thoracic and abdominal aneurysms have different developmental origins, clinical features, and pathomechanisms.6 IAs are also known to have different aneurysmal profiles, such as size, number, and rupture risk, according to their location.7 Previous studies have proposed a potential link between intracranial and ascending AAs.4,5,8 Based on these findings, we hypothesized that IA and AA may share site-specific pathomechanisms and that this would be evident in the anatomic distribution of coexisting IAs and AAs.
We reviewed records of 2375 patients presenting with AA who were aged ≥18 years and evaluated at the Seoul National University Hospital, Seoul, South Korea, between January 2005 and December 2014. Coexistence of IA was screened using all available magnetic resonance angiography or computed tomography angiography (CTA). Fusiform dilatation, mostly from arterial dissection, was excluded from IA screening. Magnetic resonance angiography or CTA was performed for a variety of reasons, including, to screen for intracranial disease in patients with vascular risk factors, to preoperatively evaluate patients planning an aortic surgery or other surgical procedures, to evaluate patients with stroke or dementia, and to evaluate patients with head trauma and headache. All patients with IA identified by CTA also had magnetic resonance angiography. Of the 660 individuals with IA screening preformed, 25 (3.8%) were identified as having a genetic arteriopathy and excluded (Figure 1). Four (16%) of this group had aneurysms in both aorta and intracranial arteries. An additional 24 (3.6%) individuals were excluded for inflammatory or acquired arteriopathies. Two (8.3%) of this group had aneurysms in both vascular beds. All patients presenting with characteristic clinical, imaging, and laboratory features of an underlying genetic or inflammatory condition underwent further appropriate work-up. Each diagnosis was made by specialists in the field of the disease and genetic testing was done for 3 of 660 patients (1 with autosomal dominant polycystic kidney disease and 2 with Loeys–Dietz syndrome). Two Marfan syndrome, 3 autosomal dominant polycystic kidney disease, and 2 Loeys–Dietz syndrome patients were diagnosed after their presentation with AA in our hospital. All except 1 patient with systemic lupus erythematosus and 5 patients with Takayasu’s arteritis were also diagnosed after CTA of the aorta. Additional 33 patients were further evaluated in suspicion of an underlying inflammatory condition that was found to be unrelated to aortopathy. From the remaining 611 patients, we identified 71 patients (11.6%) with 97 IAs. Three patients with a history of subarachnoid hemorrhage because of the rupture of IA, 2 of whom had coexisting ascending thoracic aneurysm (asc-AA) and 1 had multi-AA, were also included in the analysis.
Demographic data and medical history were collected from each patient. Hypertension, diabetes mellitus, and dyslipidemia were defined as previously described.8 The study was approved by the institutional review board of Seoul National University Hospital with waiver of informed consent.
The AA locations were divided into 3 segments demarcated by the left subclavian and renal arteries: asc-AA, descending thoracic-suprarenal aneurysm, and infrarenal aneurysm (IR-AA). Suprarenal aneurysms usually coexist with or extend from a thoracic aneurysm (thoracoabdominal aneurysm) and were therefore categorized as descending thoracic-suprarenal aneurysm. AAs existing in >1 segment were considered as multiple aneurysms (multi-AA). IAs were categorized into 3 groups according to their locations: intracranial internal carotid artery, also including anterior choroidal, superior hypophyseal, and ophthalmic arteries (ICA-IA); anterior circulation excluding the ICA, comprising anterior cerebral, anterior communicating, and middle cerebral arteries (ant-IA); and posterior circulation, consisting of arteries not included in the ICA-IA or ant-IA group. The size of the IA was determined as maximal diameter measured from the time of flight sequence. AAs were evaluated based on CTA measurements. Ascending aortic, descending thoracic, and abdominal aneurysms were each defined by aortic diameters of ≥4.5, ≥3.5, and ≥3 cm or containing an endovascular stent graft. All imaging studies were reviewed by experienced neurologists and radiologists.
We compared the distribution of IA locations, using both the number of IAs and the number of patients in each of the 3 IA sites, among the 4 AA groups. Statistical analysis was conducted using IBM SPSS 21.0 (IBM Co, Armonk, NY) and R software (http://www.r-project.org/). Fisher exact test (with Freeman–Halton extension) was used for categorical variables, and Kruskal–Wallis test was used for continuous variables. Two-tailed P values <0.05 were considered significant.
Details of the inclusion/exclusion and characteristics of study population are presented in the Table and Figure 1. Patients with AA had an 11.6% prevalence of IA and the coprevalences in each AA group ranged 9.0% to 13.6%. The prevalence of IA in patients with AA is higher than that in the general population, and the prevalence in our sample is comparable with a prior estimate of 9.0% in patients with thoracic AA.9 Male-to-female ratio showed significant difference among the 4 AA groups, with the highest ratio in IR-AA. Other demographic and clinical characteristics were not significantly different except for the location of IAs. Both ant-IAs and ICA-IAs were unevenly distributed among the 4 AA groups. Ant-IAs were observed most frequently in the asc-AA group (1.1 per person) and least frequently in IR-AA group (0.3 per person). Conversely, ICA-IAs were found most frequently in the IR-AA group (0.9 per person) and least frequently in the asc-AA group (0.2 per person). The number of posterior circulation-IAs was too small to reveal any distinct pattern.
Next, we analyzed the number of patients in each IA locations. The proportion of patients who had concurrent ant-IAs was highest in the asc-AA group and least in the IR-AA group (Table I in the online-only Data Supplement). Likewise, the proportion of patients having ≥1 ICA-IAs was highest in IR-AA group and lowest in asc-AA group. To evaluate the overall distribution of patients with both IA and AA, we separated patients with multiple IAs from patients with a single IA (Table II in the online-only Data Supplement). Of 17 patients with multiple IAs, 10 patients had IAs unambiguously categorizable to 1 of the 3 IA locations, whereas 7 had multiple IAs in >2 of 3 IA locations (Table III in the online-only Data Supplement). The latter were classified separately and the former were combined with the patients having only 1 concurrent IA. When analyzed thus, there was a significant difference in the distribution of patients with IA according to the AA locations (Figure 2; Table III in the online-only Data Supplement).
AAs arising from multiple sites may imply a distinct pathophysiology not related to a specific aortic location, and the concern that inclusion of the multi-AA group could confound the overall result may arise. However, comparison of the 3 groups excluding the multi-AA group resulted in consistently significant differences in the frequencies of ant-IA and ICA-IA (P=0.001 and P=0.007, respectively) and the distribution of patients (Table I in the online-only Data Supplement).
Given the discrepancy in male-to-female ratios among AA locations (Table), we also tested the possibility that the sex difference might affect the IA distribution. However, the number of IAs and proportions of patients in each IA locations showed similar patterns even when male and female patients were considered separately (Table IV in the online-only Data Supplement).
The distribution of IAs differed significantly according to the AA locations. More ant-IA coexisted with asc-AA and more ICA-IA with IR-AA. These results suggest a shared pathomechanism between IA and AA that contains a distinct site-specific component. Genetic factors may play a more significant role for ant-IAs and asc-AAs, as several congenital defects or syndromes involve the ascending aorta6 and as genetic risk load of IA was shown to be higher in the middle cerebral artery than in other sites.10 However, the abdominal aorta is more atherosclerosis-prone than thoracic arota,6 whereas intracranially, stenosis is more frequent in ICA than in middle or anterior cerebral arteries,11,12 suggesting that acquired mechanisms may contribute more significantly to aneurysms at these sites. Although the exact mechanism is still unclear, further investigation into the site-specific nature of aneurysm formation may reveal new insights on how aneurysms occur. Emerging data supporting shared genetic risk for IA and AA,13,14 and differing clinical characteristics of IA by coexisting aortic pathology8 underscore the importance of genetic and clinical investigation for their overlapping pathophysiology.
There are several limitations to this study. Our small sample size reduces statistical power, and multiple comparisons from the small number of subjects may raise the false-positivity issue. Retrospective collection of patients who underwent intracranial imaging study may bias the subject pool toward an older population with a larger number of cardiovascular risk factors. Generalizability to other non-Korean populations specifically related to variances in aneurysm pathophysiology also needs consideration. A population-based study is necessary to lead to more concrete and revealing findings.
Sources of Funding
This study was supported by a grant from the Korean Health Technology R&D Project, Ministry of Health and Welfare, Republic of Korea (HI12C0421). Dr Jung was supported by the Seoul National University Hospital Research Fund (0420120990).
The online-only Data Supplement is available with this article at http://stroke.ahajournals.org/lookup/suppl/doi:10.1161/STROKEAHA.115.009254/-/DC1.
- Received February 25, 2015.
- Revision received April 24, 2015.
- Accepted April 24, 2015.
- © 2015 American Heart Association, Inc.
- Southerland AM,
- Meschia JF,
- Worrall BB.
- Norman PE,
- Powell JT.
- Kappelle LJ,
- Eliasziw M,
- Fox AJ,
- Sharpe BL,
- Barnett HJ.