Abstract
Background
While periventricular anastomosis, a unique abnormal vasculature in moyamoya disease, has been studied in relation to intracranial hemorrhage, no study has addressed its change after bypass surgery. The authors sought to test whether direct bypass surgery could restore normal periventricular vasculature.
Methods
Patients who had undergone direct bypass surgery for moyamoya disease at a single institution were eligible for the study. Baseline, postoperative, and follow-up magnetic resonance angiography (MRA) scans were scheduled before surgery, after the first surgery, and 3 to 6 months after contralateral second surgery, respectively. Sliding-thin-slab maximum-intensity-projection coronal MRA images of periventricular anastomoses were scored according to the three subtypes (lenticulostriate, thalamic, and choroidal anastomosis). Baseline and postoperative MRA images were compared to obtain a matched comparison of score changes in the surgical and nonsurgical hemispheres within individuals (intra-individual comparison).
Results
Of 110 patients, 42 were identified for intra-individual comparisons. The periventricular anastomosis score decreased significantly in the surgical hemispheres (median, 2 versus 1; p < 0.001), whereas the score remained unchanged in the nonsurgical hemispheres (median, 2 versus 2; p = 0.57); the score change varied significantly between the surgical and nonsurgical hemispheres (p < 0.001). Of the 104 periventricular-anastomosis-positive hemispheres undergoing surgery, 47 (45.2%) were assessed as negative in the follow-up MRA. Among the subtypes, choroidal anastomosis was most likely to be assessed as negative (79.7% of positive hemispheres).
Conclusions
Periventricular vasculature can be restored after direct bypass. The likelihood of correction of choroidal anastomosis is a subject requiring further studies.
Similar content being viewed by others
Explore related subjects
Discover the latest articles, news and stories from top researchers in related subjects.Avoid common mistakes on your manuscript.
Introduction
Intracranial hemorrhage is a devastating symptom of moyamoya disease [9], and the effectiveness of bypass surgery in terms of preventing hemorrhage is a subject of special interest [2, 10, 12, 14, 16, 17]. This effect has been explained in relation to the numerous abnormal collateral vessels typical of the disease, which can decrease after bypass surgery. However, the mechanism of such decrease remains unclear because these collateral vessels have rarely been defined from an anatomical perspective. Research conducted with a comparative design is also lacking in this field.
Periventricular anastomosis is a unique phenomenon defined as pathological anastomoses between the perforating or choroidal artery and the medullary artery in the periventricular area that serve as collaterals to the cortex via retrograde flow in the medullary artery (Fig. 1) [4, 8]. Periventricular anastomosis is a novel term representing fragile nature of abnormal collateral vessels in moyamoya disease and is different from a traditional idea of “basal moyamoya vessels.” After a bypass is performed to augment blood flow in the cortex, the resulting flow change in the medullary artery can restore normal periventricular vasculature, although this hypothesis has remained unproven. Testing this hypothesis is clinically relevant because the results might suggest ways of improving the effectiveness of bypass surgery to prevent hemorrhage. The objective of the present study is to test whether periventricular vasculature detected with magnetic resonance angiography (MRA) can be restored after bypass surgery.
Methods and materials
Study design
This was an intra-individual left-right hemisphere comparative study comparing the postoperative change in periventricular anastomosis in both the surgical and nonsurgical hemispheres. The study was approved by the ethics committee of the Kyoto University Graduate School of Medicine (R1600).
Patients and setting
The present study included patients diagnosed with moyamoya disease according to the guideline [18] who had undergone direct bypass surgery of the bilateral hemispheres at Kyoto University Hospital between 2013 and 2017. Patients were excluded from the analysis if they had been diagnosed as quasi-moyamoya disease (secondary moyamoya phenomenon due to underlying disease); had not undergone imaging modalities on schedule; had undergone bypass surgery for vascular territories other than the middle cerebral artery (MCA) territory; or had undergone previous surgical treatment outside our hospital. Patients with unilateral disease involvement (unilateral moyamoya disease) were also excluded from the intra-individual comparison analysis.
Surgical treatment protocol
All patients underwent MRI, single photon emission tomography (SPECT), and angiography to determine whether bypass surgery was indicated. For those manifesting ischemic symptoms, bypass surgery was considered if hemodynamic compromise was apparent in SPECT. For those manifesting intracranial hemorrhage, bypass surgery was considered according to the inclusion criteria of the Japan Adult Moyamoya Trial [17]. Either direct bypass, single superficial temporal artery (STA)–MCA anastomosis, or combined bypass comprising STA-MCA anastomosis and encephalo-myo-synangiosis was adopted as a first-line surgical treatment. The surgical procedure in our institution is described in detail elsewhere [15]. Surgeons used a large horseshoe-shaped scalp incision surrounding the branch of the STA. In children, encephalo-myo-synangiosis was combined with direct bypass. For patients with bilateral involvement of the carotid arteries, a second bypass surgery of the contralateral MCA territory was performed at least 1 month after the first surgery. In all cases, the patency of the bypass was assessed with angiography performed 3 to 6 months after the second surgery.
Imaging schedule
All patients underwent prescheduled 3 MRA scans (Fig. 2): baseline MRA performed a few days before the first bypass surgery, postoperative MRA performed after the first surgery and a few days before the second surgery on the contralateral side, and follow-up MRA performed 3 to 6 months after the second surgery. The time interval between any two consecutive scans was not to exceed 12 months, and postoperative MRA was optionally performed several months before the second surgery in cases with a long time interval between the first and second surgeries. The method used for processing MRA images is described in detail elsewhere [8]. In brief, time-of-flight MRA source images were scanned with a 3-T MR scanner (MAGENETOM Trio, Skyra, Prisma, Siemens Healthnieer AG, Erlangen, Germany) by using 32-channel head coil; these were converted into sliding-thin-slab maximum-intensity-projection (STS-MIP) images with 5- to 15-mm-thick slabs in a coronal plane perpendicular to the lateral ventricle (Fig. 3).
Measurement of periventricular anastomosis
Two raters (A.M. and T.F.), blinded to other clinical information to avoid biases, retrospectively viewed the STS-MIP MRA images to grade the development of periventricular anastomosis by hemisphere. The methods used for classification and grading of the periventricular anastomoses followed those of our previous study [4, 8]. In brief, anastomoses were classified into three subtypes: (1) lenticulostriate, beginning at the lenticulostriate artery and connecting to the medullary artery; (2) thalamic, beginning at the thalamotuberal, thalamoperforating, or thalamogeniculate arteries and connecting to the medullary or insular arteries; (and 3) choroidal, beginning at the plexal segment of the anterior or posterior choroidal arteries and connecting to the medullary artery (Fig. 1). Detection of at least the proximal part of the medullary or insular arteries was considered a positive indicator for the presence of periventricular anastomosis. For the thalamic type, a vessel signal extending outside the thalamus or inside the third ventricle was judged as a positive indicator. Each subtype was scored 1 for positive and 0 for negative, and the sum of these subscale scores for each hemisphere represented the periventricular anastomosis score; the scores thus ranged from 0 to 3.
The presence of anastomosis was determined by a consensus of two raters. Inter-rater agreement was not assessed in the present study because good agreement has already been reported [8].
Suzuki stage [19] and involvement of the posterior cerebral artery (PCA) were recorded with conventional angiography for each hemisphere. PCA involvement was defined as the presence of occlusion or stenosis greater than 50% in the P1 to P3 segments of either PCA [7].
Statistical analysis
The outcome measure was the change in the periventricular anastomosis score after surgery, regarded as an ordinal variable. For intra-individual comparison, the score change was calculated using baseline and postoperative MRA images (Figs. 2 and 3) and a comparison was made between the surgical hemisphere (the hemisphere first receiving surgery) and the nonsurgical hemisphere (the hemisphere receiving surgery thereafter). As this was a matched comparison between the right and left hemispheres within individuals, the Wilcoxon matched pairs signed rank sum test was used for the comparison and no statistical adjustment was required. p values of < 0.05 were considered significant.
We also assessed midterm change in each subtype of periventricular anastomosis in the surgical hemispheres using baseline and follow-up MRA images (Fig. 2). This analysis included cases with unilateral disease involvement. Each subtype of periventricular anastomosis was assessed as a dichotomous variable (positive/negative), and McNemar’s test was used to assess the change. All analyses were performed with JMP software (Version 13, SAS Institute Inc., USA).
Results
A total of 110 patients underwent 170 bypass surgeries at our hospital between 2013 and 2017. Of these patients, 68 were excluded from the intra-individual comparison because they had either undergone only unilateral bypass (45 patients), fulfilled the exclusion criteria (22 patients), or had undergone bilateral bypass surgery at an interval exceeding 12 months (1 patient), as shown in the flow chart (Fig. 4). The remaining 42 patients, accounting for 42 surgical and 42 nonsurgical hemispheres, were analyzed for intra-individual comparison.
Intra-individual comparison
The median age was 10 years (range, 1–65), and the female-to-male ratio was 1.8. At onset, 28 patients (66.7%) presented with ischemic symptoms, 11 (26.2%) with hemorrhagic symptoms, and 3 (7.1%) with other symptoms. The surgical hemispheres—the hemispheres receiving surgery first—comprised 27 (64.3%) left and 15 (35.7%) right hemispheres. The surgical and nonsurgical hemispheres exhibited no significant difference in baseline periventricular anastomosis score (p = 0.12). The median baseline score for the surgical hemispheres was 2, with 9 hemispheres scored as 0, 11 scored as 1, 14 scored as 2, and 8 scored as 3 (Fig. 5). The median baseline score for in the nonsurgical hemispheres was also 2, with 5 scored as 0, 12 scored as 1, 15 scored as 2, and 10 scored as 3 (Fig. 5). The surgical and nonsurgical hemispheres also exhibited no significant difference in baseline Suzuki stage (median, 2 versus 2; p = 0.79). Although the prevalence of PCA involvement was higher in the surgical hemispheres than in the nonsurgical hemispheres, the difference was not significant (21.4% versus 9.5%, p = 0.06 in McNemar’s test). The median time interval between the first surgery and postoperative MRA was 50 days (interquartile range, 40–81).
As shown in Figs. 5 and 6, a significant decrease in the periventricular anastomosis score was observed in the surgical hemisphere after surgery as compared with the baseline (median, 2 versus 1; p < 0.001), whereas no significant difference was observed in the nonsurgical hemisphere (median, 2 versus 2; p = 0.57). A comparison of the surgical and nonsurgical hemispheres revealed a significant difference in score change (p < 0.001).
Midterm assessment
This analysis included 88 patients with 131 hemispheres receiving bypass surgery (Fig. 4) and assessed midterm changes in periventricular anastomosis after surgery. The median age of the 88 patients was 14.5 years (range, 1–67), and the female-to-male ratio was 1.6. At onset, 63 patients (71.6%) presented with ischemic symptoms, 19 (21.6%) with hemorrhagic symptoms, and 6 (6.8%) with other symptoms. Median baseline Suzuki stage in the 131 hemispheres was 3. The median period from surgery to follow-up MRA was 112 days (interquartile range, 95–152).
Of the 104 hemispheres positive for at least one periventricular anastomosis at baseline, 47 (45.2%) were assessed as negative at follow-up. In contrast, of the 27 hemispheres negative for any periventricular anastomosis at baseline, only 2 (7.4%) were assessed as positive at follow-up. The proportion of hemispheres positive for at least one periventricular anastomosis significantly decreased after surgery compared with the baseline (104/131 or 79.4% versus 59/131 or 45.0%, p < 0.001, Fig. 7).
As shown in Fig. 7, the proportion of hemispheres positive for lenticulostriate, thalamic, and choroidal subtypes also significantly decreased (27.5% versus 19.1%, p < 0.01; 56.5% versus 30.5%, p < 0.001; 60.3% versus 12.2%, p < 0.001; respectively). Anastomoses were assessed as negative at follow-up in 11 of 36 (30.5%) lenticulostriate-anastomosis-positive hemispheres, 34 of 74 (45.9%) thalamic-anastomosis-positive hemispheres, and 63 of 79 (79.7%) choroidal-anastomosis-positive hemispheres.
Discussion
The results of the intra-individual comparison suggest that the periventricular anastomosis score decreases after bypass surgery. The results of the midterm assessment suggest that, among the subtypes of periventricular anastomosis, choroidal anastomosis is most likely to decrease.
Our results accord with those in pioneering studies addressing postoperative changes in “basal moyamoya vessels.” Houkin et al. revealed that moyamoya vessels detected with conventional angiography decreased in 25% of adult patients after bypass surgery [10]. A study by Liu X et al. observed a reduction in moyamoya vessels in 13 of 17 hemispheres after bypass surgery for hemorrhagic moyamoya disease [16].
Our results are also consistent with several studies addressing postoperative changes in the choroidal artery. In a study using conventional angiography, Irikura et al. reported that a reduction in abnormal medullary arteries derived from the anterior choroidal artery was observed in 62% of the hemispheres undergoing indirect bypass [11]. Jiang et al. similarly reported that an improvement in the abnormal extension of the anterior choroidal and posterior communicating arteries was observed in 75 of 107 hemispheres receiving surgery (70.1%) [14]. Our results showing that 79.7% of choroidal anastomoses were assessed as negative after surgery are comparable to theirs. Our study, through its unique comparative design using noninvasive MRA, might serve to highlight the importance of the preceding studies.
The reduction in abnormal vessels after surgery might reasonably be explained by the concept of periventricular anastomosis. In normal anatomy, the medullary arteries, emerging from the pial arteries to penetrate the parenchyma and supply the deeper white matter, do not form functional anastomoses until they become capillaries [1]. In moyamoya disease, the medial end of the medullary artery connects to the perforating or choroidal artery and the direction of blood flow in the medullary artery is reversed to supply blood flow to the cortex [4, 8]. After successful bypass to the cortical artery, this retrograde flow in the medullary artery can be restored to normal, resulting in elimination of the pathological anastomosis and normalization of the perforating or choroidal arteries, as schematized in Fig. 3. This change might accordingly be observed as shrinkage of the abnormal vessels because normograde flow in the medullary artery is almost invisible.
Correction of abnormal periventricular vasculature might be considered a key mechanism in the preventive effect of bypass surgery against rebleeding, as suggested in recent studies, including a randomized controlled trial and meta-analysis [2, 12, 13, 16, 17]. Our results on choroidal anastomosis also correspond well with recent sub-analysis results of a randomized controlled trial suggesting that patients with hemorrhage located at the posterior half of the brain accrued greater benefit from bypass surgery [20]. Such a surgical benefit for the posterior hemorrhage group might be explained by the likelihood of correction of choroidal anastomosis as suggested in the present study because choroidal anastomosis is considered a typical bleeding source in posterior hemorrhage [5]. Recent studies have revealed that choroidal anastomosis is a strong predictor of rebleeding in hemorrhagic moyamoya disease [3, 6] and correcting choroidal anastomosis after bypass surgery could become an important subject requiring verification in additional studies. The present results might also promote further studies elucidating why choroidal anastomosis is more likely to be corrected after surgery. Addressing this question might improve the effect of bypass surgery at preventing hemorrhage.
The present study has several limitations. First, the results were derived only from cases treated with a specific bypass procedure at a single center and generalizing from the present results is a questionable assumption. Second, although interrater agreement for diagnosing the periventricular anastomosis score is good [8], the results could have been contaminated by misclassification of periventricular anastomoses. Third, despite the matched intra-individual comparison, the prevalence of PCA involvement was slightly higher in the surgical hemisphere, reflecting the practical trend toward treating the more severely affected hemisphere first. This potential difference, however, could have only a minimal effect on the present results because the periventricular anastomosis scores of the surgical and nonsurgical hemispheres did not differ at baseline; rather, they were slightly higher in the nonsurgical hemisphere, as shown in Fig. 5.
Conclusion
Periventricular anastomosis, a fragile periventricular vasculature present in moyamoya disease, can be corrected with direct bypass surgery. The correction might result from the effectiveness of direct bypass at preventing bleeding. Among the subtypes of anastomosis, correction of choroidal anastomosis appears the most likely to occur and this should be validated in further studies.
Abbreviations
- MCA:
-
Middle cerebral artery
- MRA:
-
Magnetic resonance angiography
- PCA:
-
Posterior cerebral artery
- SPECT:
-
Single photon emission tomography
- STA:
-
Superficial temporal artery
- STS-MIP:
-
Sliding-thin-slab maximum-intensity-projection
References
Akashi T, Takahashi S, Mugikura S, Sato S, Murata T, Umetsu A, Takase K (2017) Ischemic white matter lesions associated with medullary arteries: classification of MRI findings based on the anatomic arterial distributions. AJR Am J Roentgenol 209:W160–W168
Ding J, Zhou D, Paul Cosky EE, Pan L, Ya J, Wang Z, Jin K, Guan J, Ding Y, Ji X, Meng R (2018) Hemorrhagic moyamoya disease treatment: a network meta-analysis. World Neurosurg 117:e557–e562
Fujimura M, Funaki T, Houkin K, Takahashi JC, Kuroda S, Tomata Y, Tominaga T, Miyamoto S (2018) Intrinsic development of choroidal and thalamic collaterals in hemorrhagic-onset moyamoya disease: case control study of the Japan Adult Moyamoya Trial. J Neurosurg 1:1–7. https://doi.org/10.3171/2017.11.jns171990
Funaki T, Fushimi Y, Takahashi JC, Takagi Y, Araki Y, Yoshida K, Kikuchi T, Miyamoto S (2015) Visualization of periventricular collaterals in moyamoya disease with flow-sensitive black-blood magnetic resonance angiography: preliminary experience. Neurol Med Chir (Tokyo) 55:204–209
Funaki T, Takahashi JC, Houkin K, Kuroda S, Takeuchi S, Fujimura M, Tomata Y, Miyamoto S (2018) Angiographic features of hemorrhagic moyamoya disease with high recurrence risk: a supplementary analysis of the Japan Adult Moyamoya Trial. J Neurosurg 128:777–784
Funaki T, Takahashi JC, Houkin K, Kuroda S, Takeuchi S, Fujimura M, Tomata Y, Miyamoto S (2018) High rebleeding risk associated with choroidal collateral vessels in hemorrhagic moyamoya disease: analysis of a nonsurgical cohort in the Japan Adult Moyamoya Trial. J Neurosurg 1:1–8. https://doi.org/10.3171/2017.9.jns17576
Funaki T, Takahashi JC, Takagi Y, Yoshida K, Araki Y, Kikuchi T, Kataoka H, Iihara K, Miyamoto S (2013) Impact of posterior cerebral artery involvement on long-term clinical and social outcome of pediatric moyamoya disease. J Neurosurg Pediatr 12:626–632
Funaki T, Takahashi JC, Yoshida K, Takagi Y, Fushimi Y, Kikuchi T, Mineharu Y, Okada T, Morimoto T, Miyamoto S (2016) Periventricular anastomosis in moyamoya disease: detecting fragile collateral vessels with MR angiography. J Neurosurg 124:1766–1772
Han DH, Kwon OK, Byun BJ, Choi BY, Choi CW, Choi JU, Choi SG, Doh JO, Han JW, Jung S, Kang SD, Kim DJ, Kim HI, Kim HD, Kim MC, Kim SC, Kim SC, Kim Y, Kwun BD, Lee BG, Lim YJ, Moon JG, Park HS, Shin MS, Song JH, Suk JS, Yim MB, Korean Society for Cerebrovascular D (2000) A co-operative study: clinical characteristics of 334 Korean patients with moyamoya disease treated at neurosurgical institutes (1976–1994). The Korean Society for Cerebrovascular Disease. Acta Neurochir 142:1263–1273 discussion 1273–1264
Houkin K, Kamiyama H, Abe H, Takahashi A, Kuroda S (1996) Surgical therapy for adult moyamoya disease. Can surgical revascularization prevent the recurrence of intracerebral hemorrhage. Stroke 27:1342–1346
Irikura K, Miyasaka Y, Kurata A, Tanaka R, Yamada M, Kan S, Fujii K (2000) The effect of encephalo-myo-synangiosis on abnormal collateral vessels in childhood moyamoya disease. Neurol Res 22:341–346
Jang DK, Lee KS, Rha HK, Huh PW, Yang JH, Park IS, Ahn JG, Sung JH, Han YM (2017) Bypass surgery versus medical treatment for symptomatic moyamoya disease in adults. J Neurosurg 127:492–502
Jeon JP, Kim JE, Cho WS, Bang JS, Son YJ, Oh CW (2018) Meta-analysis of the surgical outcomes of symptomatic moyamoya disease in adults. J Neurosurg 128:793–799
Jiang H, Ni W, Xu B, Lei Y, Tian Y, Xu F, Gu Y, Mao Y (2014) Outcome in adult patients with hemorrhagic moyamoya disease after combined extracranial-intracranial bypass. J Neurosurg 121:1048–1055
Karasawa J, Touho H, Ohnishi H, Miyamoto S, Kikuchi H (1992) Long-term follow-up study after extracranial-intracranial bypass surgery for anterior circulation ischemia in childhood moyamoya disease. J Neurosurg 77:84–89
Liu X, Zhang D, Shuo W, Zhao Y, Wang R, Zhao J (2013) Long term outcome after conservative and surgical treatment of haemorrhagic moyamoya disease. J Neurol Neurosurg Psychiatry 84:258–265
Miyamoto S, Yoshimoto T, Hashimoto N, Okada Y, Tsuji I, Tominaga T, Nakagawara J, Takahashi JC (2014) Effects of extracranial-intracranial bypass for patients with hemorrhagic moyamoya disease: results of the Japan Adult Moyamoya Trial. Stroke 45:1415–1421
Research Committee on the Pathology and Treatment of Spontaneous Occlusion of the Circle of Willis; Health Labour Sciences Research Grant for Research on Measures for Intractable Diseases (2012) Guidelines for diagnosis and treatment of moyamoya disease (spontaneous occlusion of the circle of Willis). Neurol Med Chir (Tokyo) 52:245–266
Suzuki J, Kodama N (1983) Moyamoya disease--a review. Stroke 14:104–109
Takahashi JC, Funaki T, Houkin K, Inoue T, Ogasawara K, Nakagawara J, Kuroda S, Yamada K, Miyamoto S (2016) Significance of the hemorrhagic site for recurrent bleeding: prespecified analysis in the Japan Adult Moyamoya Trial. Stroke 47:37–43
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
The authors declare that they have no conflict of interest.
Ethical approval
All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. For this type of study, formal consent is not required.
Informed consent
All subjects gave opt-out consent.
Additional information
Comments
In this study, the authors used MRA to evaluate the periventricular anastomosis in patients with moyamoya disease and score the changes after direct bypass surgery including intra-individual comparison. The study nicely demonstrated the beneficial effect of direct EC-IC bypass on periventricular anastomosis in patients with moyamoya disease. This effect has long been purported to occur and described anecdotally, but this is the first study to demonstrate this effect as related to direct bypass surgery. Further work aiming to elucidate the factors influencing the degree of improvement would be welcome. For example, measuring flow in the bypass during surgery and afterwards and determining whether a robust bypass with greater flow augmentation, might be a factor, would be of great value. Also, longer term correlation of the changes in anastomosis patterns with the risk of bleeding and outcomes would be most welcome. But for the time being, the authors are to be congratulated on this novel work which nicely adds to our understanding of this elusive disease.
Fady Charbel
Illinois, USA
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
This article is part of the Topical Collection on Vascular Neurosurgery - Other
Rights and permissions
About this article
Cite this article
Miyakoshi, A., Funaki, T., Takahashi, J.C. et al. Restoration of periventricular vasculature after direct bypass for moyamoya disease: intra-individual comparison. Acta Neurochir 161, 947–954 (2019). https://doi.org/10.1007/s00701-019-03866-9
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00701-019-03866-9