Keywords

1 Introduction

Unilateral moyamoya disease was formerly considered a distinct clinical entity and has been regarded as “probable” moyamoya disease. The diagnostic criteria of moyamoya disease are gradually changing with growing understanding of the genetics and pathophysiology of the disease, and currently, both bilateral and unilateral lesions are regarded as “definite” moyamoya disease. Unilateral moyamoya disease constitutes 10%–20% of all cases of moyamoya disease. Although the rate of contralateral progression varies among studies, it occurs in both children and adults. Considering that unilateral and bilateral moyamoya disease are classified under a single family and share common genetic factors, i.e., RNF213 mutations [1, 2], they are safely considered to share a common pathological background. However, some reports have shown that the progression pattern or cerebral blood flow (CBF) pattern differs between unilateral and bilateral moyamoya disease, and thus, whether they are the same clinical entity remains to be determined. It is also possible that the contribution of nongenetic factors is different between unilateral and bilateral moyamoya disease.

Regardless of whether unilateral and bilateral moyamoya disease are identical, they share common features that are useful in understanding the pathophysiology of the disease. The distinguishing characteristics of both unilateral and bilateral moyamoya disease are progressive stenosis around the terminal region of the internal carotid artery (ICA) and the development of fine collaterals. Understanding the mechanism of disease progression will improve the management of patients. Although it is possible that the process of progression may differ between unilateral and bilateral moyamoya disease, it is advantageous to be able to observe the progressive changes from a very early stage (Suzuki stage 0–1) in cases of contralateral progression in unilateral moyamoya disease. Therefore, it is necessary to review the characteristics of unilateral moyamoya disease in association with bilateral moyamoya disease and other unilateral arteriopathies such as focal cerebral arteriopathy (FCA) to better understand the pathophysiology of moyamoya disease.

2 Definition of Unilateral Moyamoya Disease

Unilateral moyamoya disease was formerly regarded as a distinct clinical entity and has been regarded as “probable” moyamoya disease. In 2012, childhood-onset unilateral moyamoya disease with stenotic changes on the contralateral side was regarded as “definite” moyamoya disease [3]. According to the current guidelines of the Research Committee on Spontaneous Occlusion of Circle of Willis (moyamoya disease) [4], both bilateral and unilateral lesions are regarded as “definite” moyamoya disease when patients are diagnosed by conventional angiography, although when diagnosis is performed with only magnetic resonance angiography, only bilateral lesions can be diagnosed as moyamoya disease. Patients with moyamoya disease and underlying diseases such as neurofibromatosis type 1, autoimmune diseases, or Down syndrome can also be regarded as having moyamoya disease (correctly known as quasi-moyamoya disease) in the broad classification [5, 6]. However, lesions caused by atherosclerosis or cranial irradiation should not be treated as moyamoya disease. Conventional angiography is necessary for the diagnosis of unilateral moyamoya disease because it is sometimes difficult to distinguish it from non-moyamoya arteriopathy and middle cerebral artery disease.

3 Clinical and Radiological Features

The proportion of unilateral moyamoya disease cases was 19.7% in a Japanese study comprising 180 patients [7]. In other studies, it was reported to be approximately 15% (9.4%–18.0%) [8,9,10,11,12,13,14,15,16]. The age distribution in unilateral moyamoya disease shows two peaks, similar to that in bilateral moyamoya disease, but the proportion of childhood-onset disease is smaller in unilateral moyamoya disease than in bilateral moyamoya disease [7]. A report from China showed only one peak distribution, with a mean age at diagnosis of 30.8 years [2]. Ikezaki et al. reported a female-to-male ratio of 1.65, which is not different from that in moyamoya disease [7]. However, many reports showed a ratio of >2.0. In contrast, Ogata et al. reported that the female-to-male ratio in unilateral moyamoya disease was lower and between those in moyamoya disease and atherosclerotic intracranial arteriopathy [17]. Because the ratio is affected by ethnicity, surgical case-only analysis, year of enrollment, and use of magnetic resonance imaging (MRI) screening for asymptomatic patients, it varies across studies. Furthermore, a nationwide comprehensive study is needed to clarify the difference in the female-to-male ratio between unilateral and bilateral moyamoya disease.

Clinical manifestation also varies across studies. Some reports showed that symptomatic cases and left-side involvement were higher or hemorrhagic onset was common in unilateral moyamoya disease, whereas others showed that asymptomatic cases were more common (55.6%) [10]. This is susceptible to diagnostic and reporting biases. In a series of surgical cases, there were no asymptomatic cases. Screening examination by MRI such as brain checkup increases the incidence rate of asymptomatic unilateral moyamoya disease. Yu et al. compared patterns of hemorrhagic stroke between adult patients with bilateral moyamoya disease and those with unilateral moyamoya disease [18]. Patients with unilateral moyamoya disease and acute intracranial hemorrhage were at the earlier Suzuki stage and had a higher incidence rate of hypertension than patients with bilateral moyamoya disease. Intraventricular hemorrhage was more common in bilateral moyamoya disease, whereas subarachnoid hemorrhage was more common in unilateral moyamoya disease.

Familial cases accounted for 4.1%–17.6% of cases of unilateral moyamoya disease and ranged between 5% and 10% in most studies [2, 7, 10,11,12, 19,20,21]. Zhang et al. reported that familial cases accounted for 12.2% of bilateral moyamoya disease cases and 5.5% of unilateral moyamoya disease cases [2], suggesting that the proportion is higher in bilateral moyamoya disease. In accordance with this finding, the rate of bilateral involvement was higher in mutants than in wild types for the RNF213 p.R4810K mutation [22].

Only a few studies have assessed CBF in unilateral moyamoya disease. Ogata reported the differences in angiographic and CBF characteristics between unilateral and bilateral moyamoya disease [17]. Ethmoidal moyamoya vessels were more prominent in patients with moyamoya disease than in those with unilateral moyamoya disease. CBF at resting state in patients with unilateral moyamoya disease was significantly higher than that in patients with bilateral moyamoya disease or those with atherosclerotic arteriopathy. The vascular reserve capacity was higher in patients with unilateral moyamoya disease, but it was not statistically significant. It was speculated that collateral flow from the unaffected hemisphere prevents the decrease of blood flow. However, caution is advised when interpreting the results of this study because the patients with unilateral moyamoya disease were mostly asymptomatic, whereas most of the patients with bilateral moyamoya disease and those with atherosclerosis had a stroke.

Angiographic characteristics are common in both unilateral and bilateral moyamoya disease. Both show progressive steno-occlusive changes at the terminal region of the ICA, accompanied by fine collaterals at the base of the brain. Nevertheless, some reports suggested that the initiation of stenosis in unilateral moyamoya disease may be different, i.e., stenosis starts at the distal part of the MCA or anterior cerebral artery (ACA) but not at the terminal region of the ICA. For instance, Liu et al. reported a patient with unilateral moyamoya disease associated with the RNF213 mutation who underwent vertebral artery dissection followed by MCA and ACA stenosis before developing the angiographic moyamoya phenotype [23]. However, MCA stenosis in patients with bilateral moyamoya disease is likely to be underestimated. Therefore, the finding does not necessarily indicate that the progression pattern of arterial stenosis differs between unilateral and bilateral moyamoya disease. Additional clinical studies or preclinical models are necessary to examine the difference between unilateral and bilateral moyamoya disease.

4 Biomarkers

Several studies aimed to identify biomarkers for unilateral moyamoya disease. Homocystinuria is a well-known cause of quasi-moyamoya disease. Ge et al. showed that serum homocysteine level was elevated in patients with moyamoya disease and was significantly higher in patients with unilateral moyamoya disease than in those with bilateral moyamoya disease [15]. Jeon et al. reported that CRABP-I expression levels in bilateral moyamoya disease were significantly higher than those in unilateral moyamoya disease (p = 0.044) [24]. Both homocysteine and CRABP-I regulate retinoic acid, which is important for the generation of regulatory T cells, suggesting the role of autoimmunity in the development of unilateral moyamoya disease. In accordance with this finding, Chen et al. showed that autoimmune disease was more common in patients with unilateral moyamoya disease than in those with bilateral moyamoya disease, although no specific disease was identified [25].

The R4810K mutation in RNF213 has been reported to be associated with unilateral and bilateral moyamoya disease. The prevalence of the R4810K mutation was reported to be lower in patients with unilateral moyamoya disease [22]. To date, no specific gene has been identified to cause unilateral moyamoya disease. MRV-1 has been reported to be associated with moyamoya disease in patients with NF1. Such genetic factors may be identified in unilateral moyamoya disease in future studies.

In summary, compared with patients with bilateral moyamoya disease, hypertension and autoimmune disease were more common and homocysteine levels were higher in hemorrhagic stroke cases, whereas RNF213 mutation was less common in patients with unilateral moyamoya disease. These findings suggest that the balance of genetic and nongenetic factors may differ between unilateral and bilateral moyamoya disease.

5 Differentiation of Moyamoya Disease from Non-moyamoya Arteriopathy, Including FCA

Intracranial arteriopathy is a common condition that resembles moyamoya disease but does not fulfill the diagnostic criteria for moyamoya disease. Such an arteriopathy is regarded as non-moyamoya arteriopathy. However, no clear distinction can be made between moyamoya disease and non-moyamoya arteriopathy. In the case of rapid progression of unilateral arteriopathy within a family of familial moyamoya diseases, the development of moyamoya vessels was not evident [26]. Some patients with moyamoya disease exhibit a typical feature of moyamoya arteriopathy on one hemisphere and an atypical feature on the other hemisphere. The frequency of the R4810K mutation was reported to be correlated with a number of diagnostic criteria for moyamoya disease (bilateral and basal moyamoya vessels and involvement of the terminal ICA), suggesting that non-moyamoya arteriopathy may be a spectrum of moyamoya disease or RNF213-related vasculopathies. However, no specific factor was identified other than RNF213, and the rationale for this hypothesis is insufficient. In the same context, quasi-moyamoya disease was also shown to be associated with RNF213 [5], but further evidence is needed if we classify it under the same entity as moyamoya disease.

In cases of unilateral pediatric arteriopathy, differential diagnosis from transient cerebral arteriopathy (FCA) is necessary. FCA is also known as transient cerebral arteriopathy. However, transient indicates that the arterial narrowing is not progressive, and symptoms are not necessarily transient. Moreover, most children are left with permanent arterial abnormalities and residual neurological deficits. Therefore, FCA would be the preferred technical term. Braun et al. reported that among children with FCA in Europe, only 6% had progressive arteriopathy including moyamoya disease [27]. Conversely, Yeon et al. reported that moyamoya disease is more prevalent in East Asian countries and that FCA was not well recognized. They reported that 20% of cases of childhood arteriopathy, whose features are distinct from those of moyamoya disease at diagnosis, were progressive [28]. Stroke was preceded by chickenpox in 44% of patients with FCA and in none of the patients with progressive arteriopathies [27]. Araki et al. reported a case of bilateral FCA with occlusion of the left ICA followed by focal narrowing of the right ACA [29]. The patient did not have the R4810K mutation. In contrast, Echizenya et al. reported a patient with FCA with the R4810K mutation in whom arteriopathy occurred after viral infection and who recovered completely [30]. Thus, a combination of genetic and nongenetic factors does not necessarily induce progression of the arteriopathy or formation of moyamoya collaterals, suggesting that more than two factors may be necessary for disease progression in moyamoya disease. Comparative analysis, especially comparison of genetic architecture, has not been conducted between unilateral moyamoya disease and FCA. Future studies are warranted.

6 Contralateral Progression

Sixteen studies have analyzed contralateral progression of unilateral moyamoya disease (Table 3.1) [2, 7,8,9,10,11,12, 19,20,21, 31,32,33,34,35]. Ten studies included only surgical cases, two included only adult cases, and five included only pediatric cases. The percentage of contralateral progression ranged from 17.8% to 66.6% (median 28.5%) in pediatric cases and from 0 to 50.0% (median 14.6%) in adult cases. Zhang et al. performed one of the largest studies enrolling both pediatric and adult patients with unilateral moyamoya disease (17 children and 84 adults); they reported that the rate of contralateral progression was 22.2% and 14.6%, respectively [2]. Three studies consistently showed that pediatric patients aged <7 years develop contralateral progression faster than older populations [11, 12, 19] as well as develop the contralateral lesion within 3 years. These findings indicate that children have a higher risk of contralateral progression and a shorter progression period.

Table 3.1 Summary of previous studies analyzing contralateral progression of unilateral moyamoya disease
Full size table

Several candidates were determined as predictors of contralateral progression. Age at diagnosis is one of the most reliable factors, as discussed above. Angiographic abnormalities (equivocal findings of stenosis) at the MCA or ACA have been also reported in several studies, although some denied it [12]. Kuroda et al. compared the risk of progression between the untreated hemisphere of moyamoya disease and the unaffected side of unilateral moyamoya disease. They showed that the interval from diagnosis to progression was shorter in patients with moyamoya disease [8], supporting the notion that angiographic narrowing is a predictor of disease progression. Recently, Church et al. analyzed a large cohort of 217 patients with unilateral moyamoya disease including Asian and Caucasian populations and reported that hyperlipidemia is a risk of contralateral progression [21]. Comorbidities related to quasi-moyamoya disease such as NF1 have been proposed as risk factors for contralateral progression, but the results are inconsistent [19, 21]. Further replication studies are necessary to draw conclusions.

Although it has not been reported whether the mutation in RNF213 is associated with the contralateral progression of unilateral moyamoya disease, it has now been examined in the SUPRA Japan study. Considering that Asian ancestry and family history of moyamoya disease are risk factors for contralateral progression, the RNF213 R4810K mutation may also be considered a risk factor.

In terms of patient care, a periodical examination by MRI is recommended to detect progression at an early stage before the onset of symptoms. Pediatric patients, especially those aged less than 9 years, should undergo examinations within a year’s interval at least for 3 years. Because some studies showed that contralateral progression occurs after 5 years, and because long-term follow-up data are insufficient, it is recommended that patients continue to undergo periodical examinations.

7 PCA Involvement

Involvement of the posterior cerebral artery (PCA) is associated with poor prognosis in moyamoya disease. Thus, it is important to understand the risk of PCA involvement in unilateral moyamoya disease as well. It has been reported that PCA involvement tends to occur on the ipsilateral side. Matsushima reported that two of six pediatric patients with unilateral moyamoya disease had PCA involvement on the ipsilateral side [31]. Mugikura showed that PCA involvement occurs predominantly on the ipsilateral hemisphere [36]. Furthermore, the mechanism of disease progression of posterior circulation on the side of the anterior circulation needs to be evaluated in future studies.

8 Perspectives

Unilateral moyamoya disease shares a common genetic background (RNF213 mutation) with bilateral moyamoya disease. However, patients with unilateral moyamoya disease were shown to have a lower frequency of the R4810K mutation, and nongenetic factors such as homocysteine level, hyperlipidemia, hypertension, and autoimmune disease were more predominant. The balance or the number of risk factors may differ between unilateral and bilateral moyamoya disease. Given that even moyamoya disease exhibits laterality, both unilateral and bilateral moyamoya disease should reflect different phases or phenotypes of the same disease; however, current evidence is insufficient to prove this presumption. Further comprehensive analyses comparing bilateral and unilateral moyamoya disease, FCA, quasi-moyamoya disease, and intracranial atherosclerotic disease are warranted. After replicating the effect of hyperlipidemia or homocysteine on unilateral moyamoya disease progression, preclinical models will be strongly required to prove the clinical findings and test the efficacy of therapeutic drugs such as statin or folic acid. Further studies on unilateral moyamoya disease will help better our understanding of the pathophysiology of moyamoya disease and provide a therapeutic strategy.