Abstract
Background
The clinical features of aneurysms associated with moyamoya disease (MMD) and risk factors for the formation and rupture of aneurysms are not well defined.
Aims
In this study, we retrospectively analyzed clinical data of MMD patients and examined the potential risk factors for the formation and rupture of aneurysms in these patients.
Methods
The medical records of all MMD patients in our hospital from April 2012 to May 2019 were reviewed. The logistic regression analysis was used to determine the independent association between various potential risk factors and the presence or rupture of intracranial aneurysms in MMD patients.
Results
Of 2230 MMD patients, 182 (8.2%) cases had intracranial aneurysms. The mean age of onset in patients with aneurysms was 47.2 years, which was significantly higher when compared with those without aneurysms (p < 0.001). In logistic regression analysis, age of onset remained significantly associated with the presence of intracranial aneurysms, while female gender, hypertension, diabetes mellitus, and coronary artery disease were not. Besides, intracranial aneurysms were significantly associated with intracranial hemorrhage in MMD patients (odds ratio [OR] = 5.19; 95% confidence interval [CI], 3.80–7.09). About 60% aneurysms >5 mm in size, and 62.1% aneurysms with irregularly shaped morphology were ruptured. Aneurysms located in basilar tip, collateral or moyamoya vessels were more likely to present with rupture.
Conclusions
Age was an important risk factor for intracranial aneurysms formation in MMD patients. Aneurysms increased the risk of intracerebral hemorrhage in MMD patients, and their ruptures were correlated with aneurysms size, location, and morphology.
Introduction
Moyamoya disease (MMD) is a slowly occlusive cerebrovascular disorder characterized by progressive stenosis or occlusion of the terminal part of the internal carotid arteries (ICA) and their main branches such as anterior cerebral arteries (ACA) and middle cerebral arteries (MCA). 1 Some patients may present with cerebral ischemic events such as transient ischemic attack or complete infarction, while other patients may develop intracerebral hemorrhagic events such as subarachnoid hemorrhage (SAH), intraparenchymal hemorrhage (ICH), or intraventricular hemorrhage (IVH).2–4 In a few cases, seizure, headache, or dizziness may be the main clinical manifestations in some MMD patients. 5 Intracerebral hemorrhage is believed to be the most common fatal complication in MMD, while the mechanism of intracranial hemorrhage is thought to be the rupture of intracranial aneurysms or very thin walled collateral vessels of MMD patients.6,7 In daily clinical practice, the presence of intracranial aneurysms is not very uncommon on conventional angiography in patients with MMD. However, the clinical features of those cases, the characteristics of aneurysms, and the risk factors for the formation and rupture of aneurysms in MMD are not well defined.
Aims
In this study, we reviewed the clinical data of patients with MMD in our own institution and described the clinical characteristics, and investigated the potential risk factors for the formation and rupture of aneurysms in MMD.
Methods
Patients data
This study was approved by the medical research ethics board of our hospital, the first affiliated hospital of Zhengzhou University (ss-2019-000092). Because this was a retrospective observational study, and de-identified information was used for the analysis, informed consents were waived as part of the institutional ethics board approval. All methods in the study were performed in accordance with World Medical Association Declaration of Helsinki. A total number of 2612 inpatients diagnosed with MMD in our institution were identified from April 2012 to May 2019. All patients’ medical records including cerebral digital subtraction angiography (DSA), CT angiography (CTA), and MR angiography (MRA) were reviewed. According to the diagnostic criteria for MMD, 8 the cerebral angiographic images must show (1) occlusion or stenosis of the terminal part of ICA or proximal part of ACA or MCA and (2) the formation of abnormal vascular network in the vicinity of occlusive or stenotic lesions. Definite cases were defined as patients with bilateral involvement, while probable cases were defined as unilateral involvement. 9 There were 382 cases excluded in this study due to (1) diagnostic mistake; (2) lack of angiographic imaging done in our hospital; and (3) no visible vascular network and collaterals. The remaining 2230 patients with definite or probable MMD therefore were included in this study. We retrospectively evaluated all patients’ medical records and collected the following information: (1) age of onset, sex, and nationality; (2) previous diseases such as hypertension, diabetes mellitus, coronary artery disease.
Intracranial aneurysms and subclassification
The diagnosis of intracranial aneurysms was made by conventional cerebral DSA, CTA, or MRA. The angiographic images were reviewed and interpreted independently by at least two neurologists, and the results were confirmed by another neurosurgeon in our institution. The results of the angiograms including aneurysm size, morphology (saccular, fusiform, or dissecting), and location were recorded. Patients with MMD were divided into two groups: (1) those with intracranial aneurysms and (2) those without aneurysms. Aneurysm sizes recorded as the maximum 2D angiographic dimension were assigned to four categories: (1) ≤ 5 mm, (2) 5–15 mm, (3) 15–25 mm, or (4) > 25 mm. According to their morphology, the aneurysms were divided into regularly shaped aneurysms (aneurysms with smooth margin) and irregularly shaped ones (aneurysms with irregular margin, aneurysms with daughter sac or multilobulated aneurysms). Besides, the ruptured aneurysms were identified by pattern of hemorrhage visible on CT or MRI. If MMD patients with intracranial aneurysms presented with diffused and symmetric SAH, the ruptured aneurysms were supposed to be the underlying causes of SAH.
Statistical analysis
All data analyses were performed using SPSS 21.0 software. Means and frequencies were compared by the nonparametric test or χ 2 method, respectively. The logistic regression analysis was used to determine the independent effect of the variables under study on the formation or the rupture risk of intracranial aneurysms. Variables associated with formation of aneurysms included age of onset, gender, hypertension, diabetes mellitus, and coronary artery disease, while variables associated with rupture risk of aneurysms included age, gender, aneurysm size, morphology, and location. All values were expressed as means ± standard deviations. A p value of < 0.05 was considered to be significant.
Results
There were 2230 patients diagnosed with MMD included in this study. The female-to-male sex ratio was 1.1:1. Their age of onset ranged from 8 months to 74 years (mean age of onset: 40.8 years). Of the 2230 patients with MMD, 182 cases (8.2%) had intracranial aneurysms, while the remaining 2048 cases did not. Among the 182 cases with intracranial aneurysms, 106 cases (58.2%) were female and 76 (41.8%) were male. Their age of onset ranged from 6 to 71 years (mean age of onset: 47.2 years, Table 1). In the nonparametric test, there was a significant difference in age of onset between MMD patients with aneurysms and those without (p < 0.001, Table 1, Figure 1). MMD patients with elder age of onset were more prone to development of aneurysms than patients with younger age of onset. In univariate analysis, there was no significant difference in the frequency of gender, preexisting hypertension, diabetes mellitus, and coronary artery disease when patients with and without intracranial aneurysms were compared.
Comparison of age of onset distribution between MMD patients with aneurysms and those without. There was much more proportion of MMD patients with aneurysms showing age of onset over 40 years, compared with those without aneurysms. Clinical features of patients with moyamoya disease CI: confidence interval; SD: standard deviation.
In logistic regression analysis, age of onset was associated with the formation of intracranial aneurysms. Other risk factors such as gender, preexisting hypertension, diabetes mellitus, and coronary artery disease were not associated with intracranial aneurysms in either analysis.
In our study, of those 2048 MMD patients without intracranial aneurysms, only 397 cases (19.4%) presented with intracranial hemorrhagic events; while there were 101 cases (55.5%) of 182 MMD patients with aneurysms developed intracerebral hemorrhage: 45 cases (24.7%) with SAH, 40 cases (22.0%) with ICH, and 16 cases (8.8%) with IVH (Figure 2). SAH was the most common hemorrhagic event in MMD patients with aneurysms. Intracranial aneurysms significantly increased the risk of intracerebral hemorrhage in MMD patients (odds ratio [OR] = 5.19; 95% confidence interval [CI], 3.80–7.09).
Clinical manifestations of moyamoya disease patients with aneurysms.
There were a total number of 215 aneurysms found in those 182 MMD patients with anuerysms. The mean number of aneurysms was 1.18 per patient. Among them, 155 cases had one aneurysm, while the other 27 cases had two or more aneurysms. Because 6 of 182 MMD patients with aneurysms underwent surgery for aneurysms in other hospitals before admission, the details about the aneurysm sizes, and morphology of these six patients could not be acquired clearly. Therefore, a total number of 209 aneurysms in 176 MMD patients were included for further analysis, after excluding these six aneurysms without details. The majority of aneurysms associated with MMD were no more than 5 mm, while only three aneurysms were larger than 15 mm (Table 2). These aneurysms were divided into three groups based on the anatomic distribution: 122 located on anterior circulation, 64 on the posterior circulation, and 23 on the collateral or moyamoya vessels (Figure 3 showed some case examples). The ICA were the most common sites on which aneurysms usually arose in MMD. Fifteen aneurysms had nonsaccular morphology, and only accounted for a very small proportion: 11 fusiform and 4 dissection.
Some case examples of MMD patients with aneurysms showed different features of aneurysms. A 43-years-old female MMD patient with acute cerebral infarction showed a regularly shaped aneurysm in left internal carotid artery (a). A 54-years-old male MMD patient with SAH showed an irregularly shaped multilobulated aneurysm in anterior communicating artery (b). A 56-years-old female patient with SAH showed an aneurysm with irregular margin in basilar tip (c). A 52-years-old female patient with IVH showed a multilobar pseudoaneurysm in right anterior choroidal artery (d). A 45-years-old male patient with ICH of right temporal lobe and SAH showed an intracranial aneurysm with oblate shape in a right moyamoya vessel (e). A 6-years-old girl presented with ICH in left basal ganglia showed an aneurysm in left thalamic perforating artery (f). Angiographic characteristics of 209 aneurysms ACoA or ACA: anterior communicating artery or anterior cerebral artery; CI: confidence interval; Collateral or Moyamoya: collateral or moyamoya vessels; ICA: internal carotid artery; MCA: middle cerebral artery; PCoA: posterior communicating cerebral artery; SD: standard deviation; VBA or PCA: vertebrobasilar artery or posterior cerebral artery.
Among 176 MMD patients with intracranial aneurysms who had detailed data of aneurysms, there were 96 cases presented with intracerebral hemorrhage, of which 71 cases (74.0%) were believed to be associated with the rupture of intracranial aneurysms. About 60% intracranial aneurysms > 5 mm in size, and 62.1% aneurysms with irregularly shaped morphology were ruptured. Besides, aneurysms located in basilar tip, collateral or moyamoya vessels were more likely to present with rupture. In logistic analysis, the rupture risk of intracranial aneurysms associated with MMD was associated with aneurysm size, location, and morphology, while age and gender did not significantly correlate with rupture status.
Discussion
In this study, the clinical characteristics of aneurysms associated with MMD from one large single institution were firstly described, as far as we knew. There were 8.2% cases with MMD developing one or more intracranial aneurysms, and more than half of them presented with hemorrhagic events. We also found that the age of onset was associated with the development of intracranial aneurysms in MMD patients, while the gender, preexisting diseases such as hypertension and diabetes mellitus were not. The anterior circulation was the most common site on which aneurysms usually arose in MMD patients. The rupture of intracranial aneurysms associated with MMD was associated with aneurysm size, location, and morphology, but was not associated with age and gender. Intracranial aneurysms > 5 mm in size, with irregularly shaped morphology, or located in basilar tip, collateral or moyamoya vessels were more likely to present with rupture.
Intracranial aneurysms were focal dilatations of cerebral arteries characterized by a pathological wall structure with disruption of internal elastic lamina and media. 10 The underlying pathophysiological mechanisms were believed to be associated with defects in vascular endothelial and smooth muscle cells, hemodynamic forces, vascular remodeling, inflammatory pathway, genetics, and other risk factors.10–12 The prevalence of major artery aneurysms in adult patients with MMD was reported to be 3.4% to 14.8%.6,13,14 In this study, we found that the exact prevalence of all type of aneurysms in general MMD patients was 8.2%, which was much higher than that of aneurysms in a general population.10,15 Previous reports showed that the aneurysms associated with MMD were frequently located in the posterior circulation.6,14,16 However, our data showed that the aneurysms associated with MMD most commonly arose from anterior circulation, with the ICA most frequently being involved. As we all knew, about 90% of saccular intradural aneurysms in non-MMD patients were located on the anterior circulation, with the anterior communicating complex being the most frequent site. 17 The difference in the distribution of aneurysms location between MMD and non-MMD patients might be partly due to severe narrowing or occlusion of ACA and MCA in MMD patients. Even so, the posterior circulation still accounted for a relatively large proportion of aneurysms involvement in MMD, which was very different from aneurysms not associated with MMD. 18 The aneurysms originating from posterior circulation in MMD were believed to be partly associated with hemodynamic stress due to the occlusion or narrowing of anterior circulation arteries. 6 However, there were also other potential reasons that could increase the incidence of aneurysms formation in MMD patients. The most prominent pathological features of MMD were the break of the internal elastic lamina and the destruction and proliferation of smooth muscle cells in the tunica media. 19 In MMD specimens, the common irregularities of internal elastic lamina of MCA were elongated, fragmented and disappearance, and the MCA had a thinner media than control ones.20,21 In addition, moyamoya vessels exhibited various histopathological changes, including fragmented elastic lamina, and weakened media. 1 Therefore, both the hemodynamic stress and the abnormalities of structure of blood vessels might be associated with development of aneurysms in MMD patients.6,19–22 The defects of internal elastic lamina and media of blood vessel walls in MMD patients per se might also be one of main reasons for the relatively higher prevalence of aneurysms in MMD patients versus that in general population.
There had been no previous studies about potential risk factors for the formation of aneurysms associated with MMD at present. In this study, we found that age was an important risk factor, with higher incidence of aneurysms in elder patients. Intracranial aneurysm frequency increased with age and was the highest in the fourth and fifth decades of life. The incidence of aneurysms was relatively lower in children with MMD than in adult patients. Previous researches showed that age was also one of the most important non-modifiable risk factors for the development and rupture of intracranial aneurysms without MMD.23,24 Other important risk factors related to the risk of aneurysms development and rupture included the following: smoking, alcohol abuse, hypertension, presence of familial intracranial aneurysm, and female gender.23–25 However, there was no remarkable association between either female gender or hypertension and intracranial aneurysms formation in MMD in our study. These indicated that the underlying pathophysiological process might be different in these two patients population.
More than half of MMD patients with aneurysms presented with hemorrhagic events. The incidence of hemorrhagic events was much higher in patients with aneurysms than that of without aneurysms. In addition, 74.0% hemorrhagic events in MMD patients with aneurysms who presented with intracerebral hemorrhage were supposed to be attributed to the rupture of intracranial aneurysms in this study. Previous study also showed that aneurysms were more frequently found in the hemorrhagic hemispheres than control or ischemic parts. 7 These results indicated that aneurysms might be important predictors for hemorrhagic stroke, and the rupture of aneurysms was one of the important causes of intracerebral hemorrhage in MMD patients. Many studies claimed that the size, location, and morphology of aneurysms seemed to have much greater significance in predicting rupture risk of aneurysms in non-MMD patients.26–29 Likewise, in the presented study, the aneurysm size, site, and morphology were found to be associated with ruptures of intracranial aneurysms in MMD patients. However, there were some differences between the rupture risk predictors of these two kinds of aneurysms. Some researches revealed that the aneurysms located within the anterior circulation, especially on the anterior communicating artery, had a particularly high rupture risk in non-MMD patients,27–29 while other study showed that the basilar-tip location was predictive of aneurysms rupture in both non-MMD patients with history of SAH and without. 30 By contrast, our study revealed higher rupture risk in aneurysms located in basilar tips, collateral and moyamoya vessels in MMD patients. However, further researches are needed to investigate whether it is rational to treat the aneurysms on basilar tips, collateral or moyamoya vessels in MMD more aggressively. Although hemorrhagic events usually caused more severe morbidity and higher mortality, treatment of aneurysms with MMD was difficult for neurosurgeons, and the optimal strategy options were also unknown. The treatment of some aneurysms remained challenging, especially those aneurysms located in collateral, moyamoya vessels. As some ruptured aneurysms located in collateral arteries in MMD could spontaneously disappear latterly in most of the cases, conservative treatment could be an effective alternative for technically inaccessible aneurysms.6,31 In some reported cases, the aneurysms on the moyamoya vessels disappeared after superficial temporal artery to MCA anastomosis combined with encephaloduro myosynangiosis. 32 Therefore, these studies indicated that the treatment of aneurysms varied according to the location of the aneurysms and their hemodynamic features.6,33
However, there were several limitations in this study. Firstly, selection bias could not be avoided in the study due to our data being based on a single-center institution. Some MMD patients with very severe stroke might not be admitted in our hospital. Secondly, because some risk factors such as the previous history were collected retrospectively, recalling bias was inevitable. Therefore, it was reasonable to perceive our results cautiously.
In conclusion, the prevalence of intracranial aneurysms was 8.2% in general MMD patients in our study. Among several potential risk factors, only age was an important risk factor for intracranial aneurysms formation in patients with MMD. Moreover, intracranial aneurysms increased the risk of intracerebral hemorrhage in MMD, and the rupture of intracranial aneurysms in MMD patients was correlated with aneurysm size, location, and morphology, but not with age and gender.
Footnotes
Acknowledgements
Special thanks to the neurosurgeon Longzhou Zhang in our hospital for collecting data in this study.
Authors’ contributions
Hecheng Yang analyzed the data and drafted the manuscript. The study concept and design were done by Hong Lu; Limin Zhang, Menghan Wang, Jingtao Wang and Lijie Chen collected and analyzed data. All authors read and approved the final manuscript.
Data availability
Deidentified data of this study can be accessed from the corresponding author upon reasonable request.
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
