Ruptured Peripheral Cerebral Aneurysms Associated With Moyamoya Disease: A Systematic Review
Article information
Abstract
Background and Purpose
A ruptured peripheral cerebral aneurysm (PPCA) associated with moyamoya disease (MMD) is a rare but potentially life-threatening condition with controversial management strategies. We aim to summarize the clinical characteristics, treatment strategies, and prognostic factors of PPCAs in MMD.
Methods
We reviewed studies published in PubMed between 1980 and 2023 and used logistic regression analysis to identify the risk factors for adverse outcomes.
Results
Of 425 identified studies, 48 eligible studies involving 121 participants were included in the current study. The mean age at diagnosis was 40.8±15.1 years, with a peak age of onset between 41 and 50 years. Among the identified participants, 59.6% were female, and 55.9% presented with impaired consciousness. Aneurysms were present in the posterior (35.5%) or anterior (30.6%) choroidal arteries in 66.1% of the cases, and 71.1% of the patients presented with intraventricular hemorrhage (IVH) with or without intracerebral hematoma (ICH). The treatment strategies were embolization (28.9%), direct surgery (21.5%), revascularization (22.3%), and conservation (27.3%). Favorable outcomes were achieved in 86.8% of all cases, with 97.1% for embolization, 65.4% for direct surgery, 96.3% for revascularization, and 84.8% for conservative treatment. Aneurysm rebleeding occurred in 11 (26.8%) of 41 patients managed conservatively, leading to worse outcomes in 7 patients (63.6%). Impaired consciousness (odds ratio [OR], 8.61; 95% confidence interval [CI], 2.06–36.00) and aneurysm rebleeding (OR, 16.54; 95% CI, 3.08–88.90) independently predicted poor outcomes.
Conclusion
PPCA should be considered in patients with hemorrhagic MMD, particularly those with IVH with or without ICH. Endovascular and bypass treatments are recommended as first-line options, with direct open surgery as an alternative in urgent hematoma evacuation cases. Detailed preoperative planning and intraoperative technical assistance are necessary to reduce procedure-related complications. Conservative management should be selected with caution because of the high risk of rebleeding and poor outcomes. Impaired consciousness and aneurysm rebleeding appeared to be independent risk factors for adverse prognoses. We emphasize that treatment selection should be personalized, and the potential benefits should be weighed against the associated risks.
Introduction
Cerebral aneurysms associated with moyamoya disease (MMD) remain a potential hemorrhagic factor, with a reported incidence of 3%–18% [1-3]. Based on the anatomic location, these aneurysms can be broadly classified as major trunk aneurysms located on the circle of Willis and peripheral cerebral aneurysms (PPCAs) originating from the choroidal, lenticulostriate, meningeal, or moyamoya arteries. MMD-associated aneurysms are more prone to bleeding because of their fragile aneurysmal wall and increased hemodynamic stress, which predisposes them to a poor prognosis [4-8].
Considering these risks, prompt and appropriate interventions are required to treat such aneurysms. Direct surgical or endovascular treatment is widely accepted for hemorrhagic aneurysms of the major trunk of the intracranial arteries. However, optimal treatment for ruptured PPCAs remains controversial. Some authors recommend direct treatment, including endovascular embolization or open surgery, whereas others advocate conservative observation or cerebral revascularization because of the tendency for spontaneous regression [2,5-51].
Notably, studies on ruptured PPCA associated with MMD are relatively few, and all of them are small, single-center, retrospective studies and case reports. To the best of our knowledge, there has been no comprehensive literature review on these studies. Consequently, we conducted a systematic review of MMD-related ruptured PPCAs published over the past 40 years to synthesize evidence on the clinical characteristics, treatment strategies, and prognostic factors, with the aim of improving the current understanding of this rare but important clinical entity.
Methods
Search strategies and study selection
A comprehensive literature search of the National Center for Biotechnology Information’s PubMed database was performed according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. The search term “(moyamoya [title/abstract]) AND (aneurysm [title/abstract])” was used to identify articles published between January 1980 and December 2023. The cited references were also searched to identify additional relevant studies.
Documents reporting intracranial aneurysms associated with MMD, regardless of the design, were considered eligible for inclusion. All the retrieved articles were screened based on their titles and abstracts. Studies were excluded based on the following criteria: irrelevant or non-English articles; reviews, letters to the editor, commentaries, abstracts, or book chapters; nonhuman studies; unruptured intracranial aneurysms; intracranial major artery aneurysms; and insufficient individual patient data.
Data extraction
We extracted the following information from each study: patient demographics; clinical presentation; hemorrhagic type; number of patients with acute hydrocephalus; aneurysm location; treatment modality; number of patients with aneurysm rebleeding; time of aneurysm rebleeding after the first ictus; time between treatment and aneurysm disappearance; and follow-up outcome.
Clinical outcomes, assessed using the Glasgow Outcome Scale (GOS) or the modified Rankin Scale (mRS), were dichotomized as good (GOS 4–5 or mRS 0–3) or poor (GOS 1–3 or mRS 4–6).
Two reviewers (ZF and YC) independently screened the full texts to identify potentially relevant studies and extracted the data. Discrepancies were resolved by consensus with a third reviewer (CF).
Statistical analyses
Continuous data are expressed as mean±standard deviation. Categorical variables were analyzed using Pearson’s chi-square test or Fisher’s exact test. Univariate and multivariate logistic regression analyses were used to identify the risk factors for poor outcomes. In the multivariate analysis, we performed pairwise comparisons between treatment modalities.
All statistical analyses were performed using SPSS version 19.0 (IBM Corp., Armonk, NY, USA). A two-sided P-value of less than 0.05 was considered statistically significant.
Results
A total of 48 studies with 121 eligible patients were included in this systematic review (Figure 1). The characteristics of these publications are described in Supplementary Table 1. Regarding the bias concern in this systematic review of case reports, a JBI critical appraisal checklist for case reports (Joanna Briggs Institute, Adelaide, Australia; https://synthesismanual.jbi.global) is provided for the included articles (Supplementary Table 2).
Demographic and aneurysm characteristics
The general patient characteristics are summarized in Table 1. Regarding ethnicity, 41 (33.9%) were Chinese, 39 (32.2%) were Korean, 29 (24.0%) were Japanese, and 12 (9.9%) were Caucasian. Age data were available for 85 patients, and the mean age was 40.8±15.1 years (range, 7–74 years). Of these patients, eight were younger than 18 years, and the remaining 77 (90.6%) were 18 years or older. The peak age of onset was between 41 and 50 years in both male and female participants (Figure 2). Of the 104 patients with available sex data, 62 (59.6%) were female, resulting in a female-to-male ratio of 1.48:1. Admission presentation data were available for 68 patients, of whom 38 (55.9%) exhibited impaired consciousness, and 26 (38.2%) experienced headaches (with or without nausea/vomiting). The hemorrhagic types included subarachnoid hemorrhage (SAH), intraventricular hemorrhage (IVH), intracerebral hemorrhage (ICH), and their combinations. The most common type was IVH (61/121, 50.4%), followed by IVH+ICH (25/121, 20.7%) and ICH (18/121, 14.9%). Acute hydrocephalus was observed in 21 patients (17.4%). Aneurysms were frequently found in the posterior choroidal artery (PChA; 43/121, 35.5%) and anterior choroidal artery (AChA; 37/121, 30.6%).
Treatment modalities and clinical outcomes
The relationships among the number of cases, year of report, treatment selection, and follow-up outcomes are presented in a Sankey diagram (Figure 3). Of the 121 patients with PPCA, 35 (28.9%) underwent embolization, 26 (21.5%) underwent direct surgery, 27 (22.3%) underwent revascularization, and 33 (27.3%) were managed conservatively (Figure 4). The technical success rates for targeted embolization, direct surgery, and cerebral revascularization were 83.3% (35/42), 96.3% (26/27), and 100% (27/27), respectively. Two patients underwent bypass surgery upon diagnosis of MMD before PPCA formation [26,40].
In the embolization group, aneurysms were embolized with or without sacrificing the parent arteries. Liquid adhesives were used for embolization in 25 patients (71.4%), and detachable coils were used in 5 patients (14.3%).
In the direct surgery group, pathological histology was examined in 10 (38.5%) of 26 patients; 6 patients (60%) had saccular aneurysms and 4 (40%) had pseudoaneurysms.
In the revascularization group, 22 (81.5%) of 27 PPCAs disappeared and 5 (18.5%) either decreased in size or remained stable during follow-up. The time from aneurysm diagnosis to disappearance was available for 7 patients, ranging from 1 to 24 months, with an average of 7 months.
In the conservative therapy group, follow-up angiographic imaging was available for 24 (72.7%) of 33 patients. Among these, 22 (91.7%) aneurysms disappeared, with time data available for 9 cases, ranging from 3 weeks to 12 months (mean, 6.1 months), and 2 aneurysms (8.3%) remained stable in size.
Overall, 105 (86.8%) of 121 patients achieved a favorable outcome, whereas the remaining 16 patients had unfavorable outcomes. The rates of good outcomes were 97.1% (34/35) in the embolization group, 65.4% (17/26) in the direct surgery group, 96.3% (26/27) in the revascularization group, and 84.8% (28/33) in the conservative therapy group.
Aneurysm rebleeding
Aneurysm re-rupture occurred in 11 (9.1%) of the 121 participants, all of whom underwent conservative treatment [10,12,14,16,26,32,35,40,47,50]. The rate of rebleeding in cases that received conservative treatment was 26.8% (11/41). The timing of rebleeding was available for 10 patients, which varied from 1 day to 7 years (mean, 12 months). Most rebleeding episodes (7/10, 70%) were observed within 35 days of the initial bleeding. Of the 11 patients, only four achieved a good outcome, and the remaining 7 (63.6%) had a poor outcome.
Prognostic factors
In the univariate logistic regression analysis, there were significant differences in the outcomes of impaired consciousness (P=0.042), treatment modality (P=0.001), and aneurysm rebleeding (P<0.001) between the two groups (Table 2). No significant correlation was found between the outcome and ethnicity, age, sex, hemorrhagic type, acute hydrocephalus, or aneurysm location.
Multivariate logistic regression analysis revealed that impaired consciousness (odds ratio [OR], 8.61; 95% confidence interval [CI], 2.06–36.00; P=0.003) and aneurysm rebleeding (OR, 16.54; 95% CI, 3.08–88.90; P=0.001) were independent risk factors for poor outcome, whereas treatment modality was not (Table 3).
Discussion
MMD-associated PPCA has been rarely reported. Only 121 ruptured PPCAs have been described in patients with MMD in the past 40 years. We hypothesized that these aneurysms may be underestimated in patients with MMD based on the following findings: this lesion often remains undiagnosed unless it ruptures; it tends to be missed by routine brain imaging, especially when the size is small. Therefore, catheter-based cerebral angiography is recommended for the detection of aneurysms, when possible.
The exact mechanisms contributing to the formation, growth, and rupture of such aneurysms have not been clearly elucidated. Several factors may be involved, including pathological vascular architecture, long-term hemodynamic stress, and aneurysm location [5-7,41]. Unilateral MMD is susceptible to the formation of anterior circulation aneurysms on the unaffected side, whereas bilateral MMD is prone to the formation of posterior circulation aneurysms [1]. Under normal circumstances, high Suzuki stages are often associated with uncoordinated distribution of the cerebral circulation and development of collateral compensation in the posterior circulation, thereby predisposing to the PPCA formation owing to the long-term hemodynamic burden on the fragile collateral vessels [1,52,53]. Recent clinical studies suggested the impact of choroidal anastomosis (so-called choroidal channel) as dangerous collaterals with a high bleeding risk among the variety of “periventricular anastomosis” in adult patients with hemorrhagic MMD [52-54]. Based on this observation, PPCAs at choroidal channel are considered to be at a high risk of rupture, which requires prompt management.
In terms of ethnicity, the largest number of participants were from Asia, particularly China, followed by Korea and Japan. Our study revealed that the mean age at diagnosis was 40.8±15.1 years, and there was a higher prevalence in females than in males, which is consistent with the findings of previous studies [55,56]. Compared with SAH in main trunk aneurysms, the most common hemorrhagic type in PPCAs was IVH with or without ICH. Such aneurysms occurred more frequently in the PChA and AChA. Therefore, PPCA should be considered as a potential cause of hemorrhagic MMD, particularly when manifested as IVH with or without ICH, and catheter-based angiography is essential to identify this underestimated but important lesion.
Theoretically, the presence of MMD-associated PPCA represents a potential hemorrhagic risk if not obliterated. Some studies show that hemorrhagic MMD has higher rates of rebleeding (28%–35%) and mortality (7%–10%) [4,5,41,57,58]. The rate of good outcome after the first rupture was 60%, whereas it decreased to 40% after further bleeding [4,57]. Rhim et al. [41] reported that the rate of rebleeding in aneurysm-related hemorrhagic MMD (8/19, 42.1%) was significantly higher than that of aneurysm-unrelated hemorrhagic MMD (13/58, 22.4%); all aneurysmal rebleeding events occurred at least 6 months after the initial hemorrhage. Iwama et al. [59] reported 7 participants with rebleeding at the same site, 4 with MMD-associated aneurysms, and most rebleeding episodes (6/7, 85.7%) were observed within 2 months after the first onset. In the present study, aneurysmal rebleeding occurred in 9.1% of the cases, and the rate of poor outcome was 63.6% among them; in terms of timing, 70% of aneurysmal rebleeding occurred within the first 35 days. Therefore, elimination of such aneurysms to prevent recurrent rupture is plausible but remains controversial. Several authors have preferred to treat this lesion with direct endovascular or surgical intervention, whereas others have opted for revascularization surgery or conservative management, because of its tendency to regress spontaneously [2,5-51,60].
Endovascular embolization is recommended as a safe, feasible, and effective treatment option for most PPCAs in patients with MMD [2,5,7,21,22,24,25,27,28,30,33,37-39,41,43,45,46,51]. Kim et al. [5] reported on a cohort of 8 ruptured PPCAs in MMD cases treated with endovascular therapy, 7 of which were uneventful. Endovascular treatment was attempted in 42 patients and was successful in 35. In contrast to the treatment of main trunk aneurysms, liquid adhesive embolization, rather than coiling with or without a stent-assisted technique, is the preferred option for aneurysms of the collateral vessels. In the cases we identified, 25 (71.4%) of the 35 patients who underwent successful endovascular treatment were treated with n-butyl cyanoacrylate, ethylene-vinyl alcohol copolymer (Onyx), or cyanoacrylate (Glubran). Typically, aneurysms and their adjacent parent arteries are occluded together. Some authors have explained the use of liquid embolic agents for aneurysm obliteration because the lesion and the collateral vessel tend to be small and fragile, and the delivery of the detachable coil or stent might predispose them to rupture or dangerous tearing [5,15]. In addition, accessible selective catheterization is critical to the technical success of embolization because the parent artery is usually thin and tortuous. Our current study showed that 7 (16.7%) of 42 PPCAs could not be embolized because of approach inaccessibility [5,9,26,28,36,39,47].
Direct surgery, such as neck clipping or aneurysm resection, has been considered too risky for the treatment of MMD-associated ruptured PPCAs for the following reasons: these aneurysms are generally small and deep-seated, making them difficult to localize and identify; surgical clipping is difficult to perform in some cases due to the lack of a defined aneurysm neck; these aneurysms are prone to re-rupture with significant morbidity and mortality during intraoperative dissection and retraction, given their fragile characteristics; there is a potential risk of injury to the abnormal, fragile collateral vessels during surgical exposure; patients with MMD might be more susceptible to cerebral ischemia because of the poor reserve capacity of hemodynamics [2,39,40,42]. Under such circumstances, direct surgical repair might be a significant challenge and is not recommended for these aneurysms. Nevertheless, this therapeutic strategy has unique advantages over the other three, including surgical hematoma evacuation and aneurysm acquisition for further histopathological examination. The pathological features of MMD-associated PPCAs remain unclear owing to limited histopathological data and because they are typically considered as pseudoaneurysms based on their anatomic location and tendency for early morphologic change [2,25,48,49,51]. However, in this review, of the 10 PPCAs with available histopathological diagnosis, 6 (60%) were saccular aneurysms and the other 4 (40%) were pseudoaneurysms [1,6,10-12,14,17,18,29].
Cerebral revascularization has been reported as an alternative treatment to eliminate MMD-associated PPCAs [2,15,16,19,31,39,48,49]. Ni et al. [2] treated 13 MMD-related PPCAs with pure revascularization (superficial temporal artery–middle cerebral artery bypass combined with encephalo-duro-myo-synangiosis) because of microcatheter inaccessibility, of which 12 were obliterated and 1 remained stable during the follow-up period. Liu et al. [39] performed bilateral encephalo-duro-arterio-synangiosis to treat 5 pediatric PPCAs, 3 of which regressed spontaneously and the other 2 were stable at follow-up. Here, we observed 27 ruptured PPCAs treated with revascularization, of which 22 disappeared, 4 were stable, and 1 decreased in size. The exact mechanism underlying the spontaneous resolution of this lesion after bypass surgery is currently unknown. However, the reduction of the hemodynamic burden on the aneurysm and its parent artery by bypass surgery explains this phenomenon to some extent [48,49]. In particular, Lee et al. [49] found dynamic changes in diameter associated with altered hemodynamics in the parent artery (from 2.42 mm initially to 5.33 mm and finally to 1.51 mm), which poses a risk of aneurysm rupture. Moreover, there is a potential risk of PPCA formation and rupture if hemodynamic stress is not effectively reduced, even after surgical revascularization, as shown in two reported cases [26,40]. The authors suggested that impaired collateral circulation after bypass could result in aneurysm formation, and diminished collateral circulation during the follow-up period might be a sign of aggravated hemodynamic stress leading to aneurysm formation. Notably, de novo aneurysm formation around the anastomosis site or within the collateral vascular network after revascularization surgery cannot be ignored [61-67]. This postoperative complication has been rarely reported but may be clinically underestimated because it is often asymptomatic and undiagnosed [62]. Regarding the mechanism of aneurysm formation after intracranial anastomotic surgery, the following hypotheses have been proposed: disruption of the internal elastic layer and tunica media of the arterial anastomosis during the revascularization procedure might lead to traumatic aneurysm formation at the suture line; the jet blood flow from the donor arteries could directly impact the artificial T-bifurcation, causing wall stress shear and leading to spindle-like dilatation around the site of anastomosis, predisposing to the aneurysm formation under the influence of increased hemodynamic stress; a patient with MMD with the genetic variant RNF213 c.14576G>A might carry a higher risk of repeated de novo aneurysm formation. The time from bypass surgery to aneurysm diagnosis in the reported cases ranged from 6 months to 20 years, suggesting that aneurysm formation is probably time-dependent [61-67]. Therefore, long-term surveillance after cerebral revascularization seems to be essential.
Conservative therapy was selected as a viable alternative for MMD-related hemorrhagic PPCA in the following situations: high-risk surgical procedures; inaccessible endovascular approach; and the potential tendency of the aneurysm to resolve spontaneously [5,7-9,11-13,17,28,34-36,41,43]. Our review of the literature showed that 33 ruptured PPCAs were treated with conservative observation, of which 22 (66.7%) spontaneously regressed and 2 (6.1%) remained stable. The mean time to aneurysm disappearance was 6.1 months, ranging from 3 weeks to 1 year. Some authors have speculated that cerebral vasospasm and secondary thrombosis might explain the phenomenon of spontaneous aneurysm regression in the setting of conservative treatment [8,36]. Considering that such aneurysms usually originate from small-caliber collateral branches, vasospasm after aneurysmal hemorrhage might lead to the retention of blood flowing into the aneurysm sac, thereby resulting in thrombosis obliteration. In addition, some authors have suggested spontaneous regression as the natural history of pseudoaneurysms [68,69]. Notably, conservatively managed MMD-associated ruptured PPCAs are at a high risk of recurrent hemorrhage, which correlates with high morbidity and mortality. In the present review, we identified 11 cases of aneurysm re-rupture; all rebleeding occurred during the observation period [10,12,14,16,26,32,35,40,47,50]. Most rebleeding events (70%) occurred within the first 35 days after ictus. The rates of rebleeding and the associated poor prognosis in these patients were as high as 26.8% and 63.6%, respectively, suggesting that aggressive intervention should be undertaken to prevent future rupture.
The overall outcomes of these patients were favorable (86.8%). Good outcomes were achieved in 97.1%, 65.4%, 96.3%, and 84.8% of the cases treated with endovascular embolization, direct surgery, cerebral revascularization, and conservative observation, respectively. These results indicate that the therapeutic effect of embolization may be comparable to that of revascularization, all of which are superior to conservative observations. The treatment outcomes of direct open surgery were the worst. However, it should be emphasized that most participants in the open surgery group were treated 15 years ago, when knowledge about this lesion was relatively inadequate and the surgical procedures were potentially more invasive. Studies have reported that the advancement and application of stereotaxy, neuronavigation, and neuroendoscopic techniques have made open surgery safer, more effective, and less invasive [10,23,29,42]. However, the prognostic factors for patients with MMD-associated PPCAs remain unclear. Our review identified the independent risk factors for poor outcomes in these patients, including impaired consciousness and aneurysm rebleeding. Patients with impaired consciousness or aneurysm rebleeding showed an 8.6- and 16.5-fold increased risk of adverse prognosis, respectively, compared with those without symptoms.
This systematic review has several limitations. First, some eligible records may have been excluded because only PubMed was searched. Second, we only included articles published in English; thus, articles from high-incidence regions in other languages, such as Chinese, Japanese, and Korean, were excluded. Third, there is the potential for study heterogeneity, and a statistical bias exists. Fourth, our results were based on retrospective case series and single-case reports, which are considered lower-quality evidence. Fifth, the reliability of our study is weakened by the small sample size. Nevertheless, this review provides preliminary evidence for the selection of appropriate treatment options for such aneurysms.
Conclusions
MMD-associated ruptured PPCAs are more prevalent in females, with a peak age of onset at 41–50 years. These aneurysms most commonly arise from the choroidal arteries and frequently present with IVH with or without ICH. When hemorrhagic MMD is encountered, PPCA should be considered, and catheter-based angiography is essential to identify such lesions. Endovascular intervention or bypass surgery may be a first-line therapeutic option. In certain cases, particularly when urgent hematoma evacuation is required, direct open surgery may be an alternative. However, detailed preoperative planning and intraoperative technical assistance are required to reduce procedure-related complications. Conservative observation should be chosen with caution because of the high risk of aneurysm rebleeding and the associated worse outcomes, unless individuals are deemed unsuitable for surgical or endovascular treatment. The overall outcome seemed to be favorable, and impaired consciousness and aneurysm rebleeding may be independent risk factors for an unfavorable prognosis. We emphasize the importance of personalized treatment plans to balance the benefits and risks. Further studies with larger sample sizes or randomized controlled trials are warranted to facilitate a better understanding of this rare but clinically important condition.
Supplementary materials
Supplementary materials related to this article can be found online at https://doi.org/10.5853/jos.2024.02061.
Notes
Funding statement
None
Conflicts of interest
The authors have no financial conflicts of interest.
Author contribution
Conceptualization: ZF, CF. Study design: ZF, CF. Methodology: all authors. Data collection: all authors. Investigation: all authors. Statistical analysis: CF. Writing—original draft: ZF, CF. Writing—review & editing: CF. Approval of final manuscript: all authors.