Fatal Intracranial Aneurysms and Dissections Causing Subarachnoid Hemorrhage: An Epidemiological and Pathological Analysis of 607 Legal Autopsy Cases

ARTICLE IN PRESS

Fatal Intracranial Aneurysms and Dissections Causing Subarachnoid Hemorrhage: An Epidemiological and Pathological Analysis of 607 Legal Autopsy Cases Shinjiro Mori, MD, PhD,* Shirushi Takahashi, MD, PhD,*,† Akira Hayakawa, MD, PhD,* Kazuyuki Saito, MD, PhD,*,‡ Aya Takada, and Tatsushige Fukunaga, MD, PhD*

,

MD, PhD,* §

Background: There are no detailed reports, in terms of epidemiology and pathology, on intracranial aneurysms and on dissections that were found in unexpected fatal subarachnoid hemorrhage (SAH) cases. In this report we analyzed, based on large-sized medicolegal autopsy cases, the detailed epidemiology and pathological aspects of both lesions. Methods: We analyzed 607 autopsy cases of unexpected fatal SAHs including 496 aneurysms and 111 dissections. Results: The following results were obtained: (1) Patients who died of dissections were younger than those who died of aneurysms; (2) symptom prevalence rates of aneurysms were 31.9%, appearing to be lower than those in previous studies; (3) a significantly higher prevalence of clinical symptoms was found in patients with dissections (60.5%) than patients with aneurysms; (4) hypertensive cardiomegaly was indicated in more than 80%, while no obvious difference in incidence in hypertensive cardiomegaly was noted between aneurysms and dissections; (5) aneurysms were found to occur much more frequently in the anterior communicating artery (31.9%) and vertebral arteries (7.5%), while dissections were found much more commonly in vertebral arteries (93.7%); and (6) the size of aneurysms was much smaller in general than that previously regarded as a risk factor of rupturing. Conclusions: These data might help in the prompt intervention in SAH and also in the prevention of lethal SAH in clinical settings. Key Words: Subarachnoid hemorrhage—intracranial aneurysm—intracranial arterial dissection—unexpected sudden deaths—medicolegal autopsy. © 2017 National Stroke Association. Published by Elsevier Inc. All rights reserved.

Introduction From the *Tokyo Medical Examiner’s Office, Tokyo, Japan; †Department of Forensic Medicine, Hirosaki University Graduate School of Medicine, Hirosaki, Japan; ‡Department of Forensic Medicine, Juntendo University Graduate School of Medicine, Tokyo, Japan; and §Department of Forensic Medicine, Saitama Medical University, Saitama, Japan. Received July 8, 2017; revision received August 25, 2017; accepted September 20, 2017. Address correspondence to Shinjiro Mori, MD, PhD, Otsuka, Bunkyo-ku, 112-0012, Japan. E-mail: Shinjirou_Mori@member .metro.tokyo.jp. 1052-3057/$ - see front matter © 2017 National Stroke Association. Published by Elsevier Inc. All rights reserved. https://doi.org/10.1016/j.jstrokecerebrovasdis.2017.09.031

Unexpected sudden deaths caused by subarachnoid hemorrhage (SAH) are commonly experienced among forensic pathologists and medical examiners. Indeed, in 2314 administrative autopsies conducted in 2015 at the Tokyo Medical Examiner’s Office, 171 cases (7.4%) were diagnosed as SAH. The risk of rupture of intracranial aneurysms is said to be dependent on the patient’s race, age, sex, the position in cranial arteries, the size, and the presence of antecedent clinical symptoms.1-8 These previous studies, however, were based on patients with obvious symptoms and thus were submitted medical care. Hence, we

Journal of Stroke and Cerebrovascular Diseases, Vol. ■■, No. ■■ (■■), 2017: pp ■■–■■

1

ARTICLE IN PRESS S. MORI ET AL.

2

thought it might be interesting to analyze cases of unexpected sudden death because those data may reflect the real aspect in the epidemiology and pathology of those diseases, eliminating medically biased features of these diseases. There are a few previous reports dealing with this line of study based on autopsy cases.9 Sugai and Kono, and Akimoto performed pathological studies of intracranial aneurysms, 10,11 and Murai et al performed pathological studies of vertebral arterial dissections.12 However, they are again based on a small number of cases, and thus, statistical analyses were imperfect. Therefore, we performed a statistical analysis of largesized medicolegal autopsy cases involving ruptured intracranial aneurysms and dissections in terms of their pathological and epidemiological features.

Materials and Methods Selection of Cases There were 1078 cases of SAH that were confirmed by medicolegal autopsies performed at the Tokyo Medical Examiner’s Office during April 1999 and March 2015. Among those, we excluded cases caused by trauma and also those who died of causes other than intracranial aneurysm or dissection, such as arteriovenous malformation and/or Moyamoya disease. Thus, 607 cases composed of 496 aneurysms and 111 dissections were finally submitted for this study. This study was approved by our institutional ethics committee.

Epidemiological Analysis The patients’ age, sex, premonitory symptoms, and past histories were retrospectively reviewed as epidemiological characteristics. The presence of premonitory symptoms was submitted only when complaints or signs were witnessed within 24 hours before the death. The past histories were based on the clinical records and submitted to this study only when they were available. Antecedent symptoms or previous histories, when adequate information was not described in medical records, were excluded from the data.

Autopsy Findings The size and location of aneurysms and/or dissections as well as the patients’ body length and cardiac weight were cited from the autopsy records. The ratio of heart weight (gram) to body length (centimeter) (HW/ BL ratio) greater than 2.00 was regarded as cardiomegaly following Sato’s criteria.13 Lesions found in the anterior communicating artery bifurcations of anterior cerebral arteries were classified in this report as “lesions of the anterior communicating

artery,” and those in posterior cerebral arteries–basilar artery were classified as “lesions of the basilar artery.” The sizes of the lesions were divided into 5 categories according to previous studies2,3: 0.4 cm or smaller, 0.5-0.6 cm, 0.7-0.9 cm, 1.0-2.4 cm, and 2.5 cm or bigger. Such cases where no aneurysms or dissections were detected and/or the size of the lesions was uncertain were excluded from this study.

Statistical Analysis All statistical analyses were performed with Microsoft Excel 2016 (Microsoft Corporation, Redmond, WA) and the statistics add-in software BellCurve Ekuseru-Toukei (Social Survey Research Information Co., Ltd., Tokyo, Japan). The chi-squared test was used to evaluate the independence of qualitative variables. With regard to sex differences, for the correction of the ratio, expected values were calculated from data from all autopsies performed at the Tokyo Medical Examiner’s Office from April 1999 to March 2015. Differences in quantitative variables between the 2 groups were evaluated using Mann– Whitney’s U-test, differences among multiple groups were determined using the Kruskal–Wallis test, and multiple comparisons were performed using the Steel–Dwass test. A P value of less than 0.05 was considered statistically significant. Correlations were evaluated using Spearman’s rankcorrelation coefficient.

Results Overview of Cases We included 607 cases (496 cases of intracranial aneurysm and 111 cases of intracranial arterial dissection) in this study (Table 1). They constituted of 371 males and 236 females. Male predominance was statistically confirmed both in the aneurysm and dissection groups: the male-to-female ratios were 1.3 in the aneurysm and 5.5 in the dissection groups, respectively. The median age of overall cases was 54.0 (ranging from 20 to 93) and was significantly higher in females (60.0) than in males (50.0) (P < 0.05). Meanwhile, the median age of the aneurysm group (56.0) was significantly higher than that of the dissection group (46.0). The antecedent clinical symptoms were reviewed in 401 patients. Among those, 151 (37.7%) were found to have suffered from clinical symptoms (Table 1). They included headache (n = 90; 22.4%), vomiting (n = 18; 4.5%), dizziness (n = 12; 3.0%), cold-like symptoms (n = 10; 2.5%), and neck pain (n = 8; 2.0%). Symptom prevalence was significantly higher in the dissection group (60.5%) than in the aneurysm group (31.9%). The incidence of headache was especially high (n = 34; 42.0%) in the dissection group.

ARTICLE IN PRESS INTRACRANIAL ANEURYSMS AND DISSECTIONS CAUSING SUBARACHNOID HEMORRHAGE

3

Table 1. Sex, age, and premonitory symptoms of patients Total

Aneurysm

Dissection

n

Total Male Female

607 371* 236*

496 277* 219*

111 94* 17*

Age

Total Male Female

54.0 (45.0,63.0) 50.0 (44.0,58.0)† 60.0 (51.8,70.3)†

56.0 (47.0,65.0)‡ 52.0 (45.0,59.0)† 60.0 (52.5,71.0)†

46.0 (41.0,52.0)‡ 45.5 (41.0,51.0)† 51.0 (47.0,59.0)†

Symptom

Total None Any Headache Vomiting Dizziness Cold-like Neck pain

401 250 (62.3%) 151 (37.7%) 90 (22.4%) 18 (4.5%) 12 (3.0%) 10 (2.5%) 8 (2.0%)

320 218 (68.1%)§ 102 (31.9%) 56 (17.5%) 14 (4.4%) 11 (3.4%) 7 (2.2%) 4 (1.3%)

81 32 (39.5%)§ 49 (60.5%) 34 (42.0%) 4 (4.9%) 1 (1.2%) 3 (3.7%) 4 (4.9%)

As for age, the median, first quartile, and third quartile values are noted (quartiles are in brackets). *Female < male (P < 0.05). †Male < female (P < 0.05). ‡Dissection < aneurysm (P < 0.05). §Aneurysm < dissection (P < 0.05).

Among 532 patients (including 8 with unruptured intracranial aneurysm) whose clinical records were available, 152 (28.6%) were found to have had hypertension. There was no significant difference in those hypertensives between the aneurysms and dissections (28.0% versus 30.9%). The HW/BL ratio greater than 2.00 was found in 502 of 604 cases (82.7%; aneurysm in 81.5%, and dissection in 88.3%).

Anatomical Location and Size The locations of the lesions are listed in Tables 2 and 3 (aneurysm) and Table 4 (arterial dissection). Much more cases were found in the anterior circulation than in the posterior circulation (overall lesions, 68.8% versus 32.2%; aneurysm, 83.7% versus 16.3%, P < 0.05). The most common site of aneurysms was the anterior communicating artery (31.9%), and the next most common site was the middle cerebral artery (25.6%). On the other hand, the majority of dissections were located in the vertebral arteries, while in 6.3%, they were found in other locations, such as the internal cervical artery. No significant laterality was indicated in any of the lesions. The sizes of the lesions are described in Tables 5 and 6 and in Figure 1. The median size of all of the lesions was 0.5 cm (0.4 cm-1.0 cm), and 289 cases (47.6%) had lesions of 0.7 cm or bigger. Aneurysms were significantly smaller than dissections (median size: 0.5 cm versus 0.8 cm; prevalence of lesion larger than 0.7 cm: 41.7% versus 73.9%, respectively, P < 0.05). Aneurysms of 0.6 cm or

smaller were significantly more frequent than those of 0.7 cm or bigger (P < 0.05). Aneurysms at the anterior communicating artery were significantly smaller than those at the vertebral artery and middle cerebral artery (P < 0.05).

Pathological Findings and Their Backgrounds The median age of patients with lesions in the posterior circulation (49.0) was significantly younger than those in the anterior circulation (55.5) (P < 0.05), while no significant difference was found in median age between those with aneurysm in posterior and anterior circulations (56.0 versus 56.0). In the dissection group, a statistical comparison could not be performed because the number of dissections in the anterior circulation was too small (n = 3). A much higher symptom prevalence was found on those with lesions in the posterior circulation than on those with lesions in the anterior circulation: 55.5% versus 28.4%. The same was noted on aneurysms (28.5% versus 47.4%; P < 0.05). In the dissection cases, symptoms had occurred in all patients with dissections in the posterior circulation, while in 1 patient with a dissection in the anterior circulation (n = 1), no premonitory symptoms were found. Concerning the size of the lesions and the patient’s age, a weak correlation was noted (r = 0.42) only in the females in the dissection group (n = 17), while no correlations were noted in any other groups (r = 0.04-0.19). No significant differences were found between antecedent symptoms and the sizes of the lesions in aneurysms as well as dissections.

ARTICLE IN PRESS S. MORI ET AL.

4

Table 2. Position and number of intracranial aneurysms (anterior circulation) ACA

37(7.9%) Left Left A1 Left A2 Left unclassifiable

12 4 1 7

Right Right A1 Right A2 Right unclassifiable

24 8 6 10

Unclassifiable Acom

ICA

IC-PC

Left ACA-Acom Acom Right ACA-Acom Unclassifiable

6 10 74(14.9%)

Left Right Pcom

1 158(31.9%)

16(3.2%) Left Right

32 42 3(0.6%)

Left Right MCA

23 72 23 40

1 2 127(25.6%)

Left Left M1 Left M1-M2 Left M2 Left unclassifiable Right Right M1 Right M1-M2 Right M2 Right unclassifiable

49 27 0 11 11 78 38 3 11 26

Abbreviations: A1, horizontal portion of ACA; A2, vertical portion of ACA; ACA, anterior cerebral artery; Acom, anterior communicating artery; ICA, internal carotid artery; IC-PC, internal carotid artery–posterior communicating artery; M1, horizontal portion of MCA; M2, insular portion of MCA; MCA, middle cerebral artery; Pcom, posterior communicating artery. Percentages are noted in brackets.

Discussion Sex and Age In contrast to previous studies, our results demonstrated male predominance in both aneurysm and dissection groups. Previous studies dealing with clinical cases7 and animal experiments14 have suggested that females are at higher risk in the formation, growth, and rupture of cerebral aneurysms. This female predominance has largely been explained by hormonal factors (i.e., the decrease in estrogen secretion after menopause). Mean-

while, as an exception, cerebral arterial dissection is reported to affect males predominantly.12,15 Those previous data differ from our present study, which apparently shows male predominance. A few speculations might be possible on this difference. One is that our data reflect reality and thus should be emphasized because they are based on hitherto unreported large-scale autopsy cases. On the other hand, another speculation might be possible that our data might have been biased because postmenopausal women are much less exposed to administrative autopsies. In Japan, administrative autopsies

Table 3. Position and number of intracranial aneurysms (posterior circulation and others) PCA

2 (0.4%) Left Right

BA Tip Others Unclassifiable

VA

0 2

37 (7.5%)

38 (7.7%)

VA-BA Left Left PICA Left unclassifiable

1 19 6 13

33 3 2

Right Right PICA Right unclassifiable

17 5 12

Others

4 (0.8%)

Abbreviations: BA, basilar artery; PCA, posterior cerebral artery; PICA, posterior inferior cerebellar artery; VA, vertebral artery. Percentages are noted in brackets.

ARTICLE IN PRESS INTRACRANIAL ANEURYSMS AND DISSECTIONS CAUSING SUBARACHNOID HEMORRHAGE

5

Table 4. Position and number of intracranial dissections ACA

1 (0.9%) Right Right A1

BA

2 (1.8%)

1 1

Tip

2

VA Acom Unclassifiable IC-PC

104 (93.7%)

1 (0.9%)

VA-BA

1

Left Left unclassifiable Right

1 (0.9%) Left

1

1

Right PICA Right unclassifiable Others

46 46 57 3 54 2 (1.8%)

Abbreviations: A1, horizontal portion of ACA; A2, vertical portion of ACA; ACA, anterior cerebral artery; Acom, anterior communicating artery; BA, basilar artery; IC-PC, internal carotid artery–posterior communicating artery; PICA, posterior inferior cerebellar artery; VA, vertebral artery. Percentages are noted in brackets.

Table 5. Size of intracranial aneurysms ≦0.4cm ACA Acom ICA IC-PC Pcom MCA PCA BA VA Others Total

11 70 7 25 2 32 1 6 14 0 168 (33.9%)

0.5-0.6cm

0.7-0.9cm

1.0-2.4cm

2.5cm≦

Total

13 37 4 14 0 35 0 10 6 2

6 32 2 11 0 19 0 5 7 1

5 17 3 21 1 36 1 15 9 1

2 2 0 3 0 5 0 2 1 0

37 158 16 74 3 127 2 38 37 4

121 (24.4%)

83 (16.7%)

109 (22.0%)

15 (3.0%)

496

Abbreviations: ACA, anterior cerebral artery; Acom, anterior communicating artery; BA, basilar artery; ICA, internal carotid artery; ICPC, internal carotid artery–posterior communicating artery; MCA, middle cerebral artery; PCA, posterior cerebral artery; Pcom, posterior communicating artery; VA, vertebral artery. Percentages are noted in brackets.

Table 6. Size of intracranial dissections ≦0.4cm

0.5-0.6cm

0.7-0.9cm

1.0-2.4cm

2.5cm≦

Total

ACA Acom IC-PC BA VA Others

0 1 0 0 15 0

0 0 0 1 10 2

0 0 1 1 28 0

1 0 0 0 49 0

0 0 0 2 2 0

1 1 1 2 104 2

Total

16 (14.4%)

13 (11.7%)

30 (27.0%)

50 (45.0%)

2 (1.8%)

111

Abbreviations: ACA, anterior cerebral artery; Acom, anterior communicating artery; BA, basilar artery; IC-PC, internal carotid artery– posterior communicating artery; VA, vertebral artery. Percentages are noted in brackets.

ARTICLE IN PRESS S. MORI ET AL.

6

Figure 1. Size of intracranial aneurysms and dissections. Extracted data are shown graphically. Percentages and numbers (in brackets) are noted. Abbreviations: ACA, anterior cerebral artery; BA, basilar artery; IC-PC, internal carotid artery–posterior communicating artery; MCA, middle cerebral artery; and VA, vertebral artery.

of elderly people who died from SAH tend not to be performed if the symptoms are typical or if ancillary tests, including postmortem computed tomographic imaging, are positive. Thus, the male preponderance on those intracranial vascular events should be further studied epidemiologically hereafter. Cerebral arterial dissections were shown to occur in a younger age group than that in cerebral aneurysms. This result supports previous reports.8,12,16 However, because these previous reports were based on smaller size of cases and thus were not submitted to statistical analysis, our results might be evaluated thus in that it substantiated those previous postulations statistically.

Antecedent Clinical Symptoms Previous studies have suggested that cerebral aneurysms with clinical symptoms have a higher risk of rupture than those without symptoms.1,5,12 Morita et al reported that as much as 47.4% of patients with unruptured cerebral aneurysms had symptoms (headache or dizziness) that urged them to undergo imaging.5 The rate of symptom prevalence in our cases was 31.9% in cerebral aneurysms. This rate is apparently lower than those of the previous studies. Indeed, our results suggest that one third of the patients who died of cerebral aneurysm rupture do not present antecedent symptoms. These data may reflect the real aspect of symptom expression on cerebral aneurysms. Meanwhile, our study indicated a significantly higher prevalence of symptoms among patients with dissections (60.5%). We also indicated that symptom expression did not correlate with the size of the dissection. Murai et al12 reported that cerebral artery dissection histopatho-

logically shows various degrees of inflammatory cell reaction in adventitia, and this may explain the occurrence of different intervals between the onset and aggravation of the dissection. Thus, an explanation for symptom prevalence in dissection may be that patients who died from dissections might have had enough time to complain about a symptom.

Hypertension and Associated Cardiomegaly It remains controversial whether hypertension is an independent risk factor in the rupture of cerebral aneurysms.2,5,6 Also, there is no detailed statistical analysis on the association of cerebral dissection and hypertension. In our study, the prevalence of hypertension was 28.6%, which is much lower than the result of Sonobe et al’s study (85.7%).6 However, as much as 81.5% of the cerebral aneurysms in the present study were shown to have an HW/BL ratio greater than 2.00, suggesting the presence of cardiomegaly.13 As most of the patients with cardiomegaly are hypertensive, our results may suggest that a considerable number of undiagnosed or nonmedicated patients with hypertension were included in the nonhypertensive group.

Location of Lesions Our results show a higher incidence of both lesions in vertebral arteries (37 cases [7.5%] among aneurysms and 104 cases [93.7%] among dissections, for an overall incidence of 23.2%). Previously, Sugai et al reported in 1955 that the locations of ruptured cerebral artery aneurysms were as follows: 30.9% in the anterior communicating artery, 3.6% in the vertebral arteries, and no detailed sta-

ARTICLE IN PRESS INTRACRANIAL ANEURYSMS AND DISSECTIONS CAUSING SUBARACHNOID HEMORRHAGE 10

tistical analyses for cerebral dissection. These previous data should be reevaluated because the importance of cerebral artery dissection was recently indicated,12 and the diagnostic yield of the lesions has been increasing recently in forensic practices. Our cases with intracranial aneurysms showed higher incidences in the anterior communicating artery (31.9%) and vertebral arteries (7.5%) than the previous work by Morita et al, which reported incidences of nonruptured intracranial aneurysms as follows: 15.5% in the anterior communicating artery and 1.8% in the vertebral arteries.5 Schievink et al assumed that the real frequency of posterior circulation aneurysms must be significantly higher than the incidence of those who received medical attention, because of their lower rate of receiving treatment.9 The incidence of ruptured aneurysms in the anterior communicating artery was guessed to be higher than that of unruptured aneurysms in those (approximately 30% versus 10%-15%).6 However, these previous reports were based on a much lower number of cases for demonstrating statistical analysis. Our results may support this hypothesis. We demonstrated in this study that the patients who died from cerebral aneurysm rupture were largely with no antecedent symptom. By a difference in the configuration of the location, cases with cerebral aneurysms in carotid– posterior communicating arteries must have complained of much more symptoms than those in the anterior communicating artery or vertebral arteries (e.g., optic nerve). Therefore, the reason for the difference in incidences between our results and those in the previous clinical study was that in cases of sudden and unexpected deaths the patients who died from cerebral aneurysms in carotid– posterior communicating arteries were decreased. Our data showed no significant lateralities concerning the incidence of cerebral artery dissection. This result contravenes the results of previous studies indicating a laterality to the right side of intracranial arteries.11,16 However, the numbers of cases in these studies are too small (6 cases in Akimoto; 27 cases in Manz and Luessenhop). Therefore, we consider our results to have a higher statistical validity.

Size of Lesions In this study, intracranial aneurysms of 0.6 cm or smaller were shown to be much more frequent than those of 0.7 cm or bigger. Previous clinical studies have suggested that an increased size of aneurysm bigger than 0.7 cm is a risk of rupture.1,2,5 We believe this is the real aspect of ruptured aneurysm in SAH cadaver. However, a few discussions will be needed for its evaluation. Wiebers et al suggested that certain aneurysms may decrease in size after the rupture.17 Such possible shrinkage of aneurysms after rupture might explain the smaller size of aneurysms in the present autopsy-based study. Also, such shrinkage might be the reflection of the difference between

7

the antemortem state, as measured by in vivo images, in which intravascular pressure increases the size, and the postmortem state of aneurysm, in which intravascular pressure is absent. On the other hand, there are reports saying that a certain group of small aneurysms is prone to rupturing within a short time after its development, as short as a few days to a few weeks,6,17 and that this type of small aneurysm is difficult to recognize clinically and radiologically.6 Hence, another explanation might be possible that the higher incidence of small aneurysms in the present SAH-based study can be the reflection of the high incidence of ruptures of small aneurysms in SAH that are not clinically recognized before rupture and thus not counted in previous studies. Thus, a further detailed clinical and postmortem collaborative large-scale study will be needed on the size of aneurysms and their incidence of rupture. Certainly, such work will add significant new information concerning the risk of rupture of intracranial aneurysms.

Appendix Finally, comments on a few possible limitations of this study might be needed. Actually, the following are the inevitable limitations on studies based on medicolegal autopsy cases: The backgrounds of the patients studied here might be somehow different from previous clinical studies. Medical histories and antemortem clinical symptoms might be somehow ambiguous. Clinical information regarding intracranial aneurysms or dissections before rupture could not be evaluated in detail.

Conclusion We studied, based on our large-sized medicolegal autopsy data sets, the epidemiological and pathological aspects of untreated intracranial aneurysms and dissections resulting in sudden and unexpected death from SAH. Significant new findings are found and described in terms of the age of onset, the location in intracranial arteries, the size of intracranial vascular lesions, and the clinical symptom-prevalence rates. The data described here can be applied for the prompt intervention in such lesions and the prevention of lethal SAH in clinical settings.

References 1. Wiebers DO, Whisnant JP, Huston J III, et al. Unruptured intracranial aneurysms: natural history, clinical outcome, and risks of surgical and endovascular treatment. Lancet 2003;362:103-110. 2. Wermer MJ, van der Schaaf IC, Algra A, et al. Risk of rupture of unruptured intracranial aneurysms in relation to patient and aneurysm characteristics: an updated meta-analysis. Stroke 2007;38:1404-1410. 3. Morita A, Fujiwara S, Hashi K, et al. Risk of rupture associated with intact cerebral aneurysms in the Japanese population: a systematic review of the literature from Japan. J Neurosurg 2005;102:601-606.

ARTICLE IN PRESS 8 4. Raghavan ML, Ma B, Harbaugh RE. Quantified aneurysm shape and rupture risk. J Neurosurg 2005;102:355362. 5. UCSA Japan Investigators, Morita A, Kirino T, et al. The natural course of unruptured cerebral aneurysms in a Japanese cohort. N Engl J Med 2012;366:2474-2482. 6. Sonobe M, Yamazaki T, Yonekura M, et al. Small unruptured intracranial aneurysm verification study: SUAVe study, Japan. Stroke 2010;41:1969-1977. 7. Juvela S, Poussa K, Porras M. Factors affecting formation and growth of intracranial aneurysms: a long-term follow-up study. Stroke 2001;32:485-491. 8. Kubo Y, Koji K, Kashimura H, et al. Female sex as a risk factor for the growth of asymptomatic unruptured cerebral saccular aneurysms in elderly patients. J Neurosurg 2014;121:599-604. 9. Schievink WI, Wijdicks EFW, Parisi JE, et al. Sudden death from aneurysmal subarachnoid hemorrhage. Neurology 1995;45:871-874. 10. Sugai M, Kono R. Pathological studies on subarachnoid hemorrhage, especially on the origin and localization for aneurysms of the basal arteries of brain. Proc Jpn Soc Pathol 1955;44:330-345. (article in Japanese with English abstract).

S. MORI ET AL. 11. Akimoto Y. A pathological study of intracranial aneurysms, particularly of aneurysms other than saccular ones. Acta Pathol Jpn 1980;30:229-239. 12. Murai M, Saito K, Takada A, et al. Subarachnoid hemorrhage from ruptured dissecting aneurysm of the vertebral artery: a clinic-pathological study. Res Pract Forens Med 1993;36:175-183. (article in Japanese with English abstract). 13. Sato K. Studies on heart weight and the degree of cardiomegaly. Hirosaki Medical Journal 1955;6:1-6. (article in Japanese). 14. Jamous MA, Nagahiro S, Kitazato KT, et al. Role of estrogen deficiency in the formation and progression of cerebral aneurysms. Part II: experimental study of the effects of hormone replacement therapy in rat. J Neurosurg 2005;103:1052-1057. 15. Yamaura A. Diagnosis and treatment of vertebral aneurysms. J Neurosurg 1988;69:345-349. 16. Manz HJ, Luessenhop AJ. Dissecting aneurysm of intracranial vertebral artery: case report and review of literature. J Neurol 1983;230:25-35. 17. Wiebers DO, Whisnant JP, Sundt TM Jr, et al. The significance of unruptured intracranial saccular aneurysms. J Neurosurg 1987;66:23-29.