Moyamoya disease is a chronic cerebral vasculopathy characterized by slowly progressive carotid artery stenosis and occlusion accompanied by moyamoya vessels. Surgical interventions are often recommended to the patients with progressive cerebral ischemic symptoms, evidence of cerebral blood flow (CBF) insufficiency and/or cerebral vascular reserve impairment while there are no surgical contraindications.1
Surgical interventions for moyamoya disease include direct and indirect revascularizations. There are three types of external carotid artery (ECA) branches that can be used in these procedures: (1) the superficial temporal artery (STA) or the occipital artery supplying the scalp and galea aponeurotica, (2) the deep temporal artery (DTA) supplying the temporal muscle and periosteum, or in some cases, the sphenopalatine artery participating in the supply of the anterior part of temporal muscle, and (3) the middle meningeal artery (MMA) supplying the cerebral dura mater. There are three surgical options in usage of the STA, including its direct use in STA-middle cerebral artery (STA-MCA) anastomosis,2 its indirect use in encephalo-arterio-synangiosis (EAS),3 encephalo-duroarterio-synangiosis (EDAS),4 encephalo-duro-arterio-myosynangiosis (EDAMS),5 and omentum transplantation; which is essentially an approach using the blood flow in the STA without directly using the vessel.6 The DTA can be used alone in encephalo-myo-synangiosis (EMS)7 or as one of the donor vascular grafts in indirect revascularization procedures, such as EDAMS. The MMA has not been reported to be used alone in such procedures. The procedures mentioned above can be used alone or combined, resulting in different therapeutic effects. Generally speaking, STA-MCA bypass combined with EDAS or EDAMS can achieve better results, and the surgical procedures can be carried out differently.1,7–12 According to the current literature, moyamoya disease patients that need surgical treatment were merely classified into ischemic1,8–10,12,13 and hemorrhagic14 cases without more detailed classification in accordance with the preoperative angiographic findings and clinical manifestations. Additionally, the published5,6,9,11–13 clinical and radiological data on the postoperative short-term hemodynamic changes and spontaneous anastomoses of the DTA and MMA with the cortical arteries are insufficient. Without such data, explicit answers cannot be found to such questions as: (1) Should one or both branches of the STA be used for direct anastomosis? (2) Should the temporal muscle be split or preserved intact? (3) How to protect the middle meningeal arteries during the operation and dispose of the dural mater in stage V or VI cases with spontaneous anastomosis between the meningeal arteries and the cortical arteries?
To answer these questions, the effects of different surgical procedures were tested. We then designed a new combined cerebral revascularization procedure in 2005 so as to bring into full play the blood supply of STA, MMA, and DTA.15,16 In this new designed combined revascularization, the direct anastamosis of the STA and the MCA was performed combined with encephalo-duro- myo-synangiosis (EDMS).
From October 2005 to November 2009 a total of 199 patients with moyamoya disease were identified by digital subtraction angiography (DSA) in the Department of Neurosurgery, Huashan Hospital, Fudan University. There were 111 patients who received this combined procedure, including 53 male and 58 female patients with a mean age of 31.4 (range, 11.0–56.0) years. The first author performed all the operations. All the patients had typical moyamoya disease involving both hemispheres.
Abiding by the staging criteria described by Suzuki and Takaku,17 5 patients were in stage II, 38 patients stage III, 40 patients stage IV, 21 patients stage V, and 7 patients stage VI. In accordance with the clinical classification criteria described by Matsushima et al,18 21 patients were type I (transient ischemic attack (TIA) type), 18 patients were type II (recurrent TIA type), 15 patients were type III (TIA-cerebral infarction type), 10 patients were type IV (cerebral infarction-TIA type), and 12 patients were type V (cerebral infarction type). Intracranial hemorrhage as the initial presentation (type VI) was found in 35 patients, including 20 patients with intra-ventricular hemorrhage, 12 patients with paraventricular intracerebral hematoma rupturing into the ventricles, 2 patients with left temporal lobe hematoma, and 1 patient with recurrent cerebral hemorrhage (paraventricular intracerebral hemorrhage two times and intra-ventricle hemorrhage one time). Fourteen patients received external ventricular drainage and/or ventriculoperitoneal shunt. Sixteen patients had concurrent TIA or cerebral infarction, and 2 had a peripheral intracranial aneurysm. Out of these 111 patients, 41 patients had neurological deficits, including 23 cases caused by ischemia and 18 cases caused by hemorrhage. The mean preoperative National Institutes of Health Stroke Scale (NIHSS) score of these 41 patients was 4.7 (range 2.0–14.0). The NIHSS scores could not be evaluated in the other 70 patients who had TIA or reversible ischemic neurological deficits (RIND) without any identifiable symptoms.
Selection of patients for the operation
Prompt surgical intervention is generally thought appropriate upon establishing a diagnosis of moyamoya disease, and the previous literature does not provide explicit specifications of the surgical indications.11,19 In our series, we postulated surgical indications for moyamoya disease based on a comprehensive evaluation of the DSA findings, clinical manifestations, and computed tomographic perfusion (CTP) findings. All the patients with a definite diagnosis of moyamoya disease by angiography of the six cerebral arteries (including the bilateral internal carotid arteries (ICAs), ECAs, and vertebral arteries) underwent CTP examination. The patients in angiographic stage I were recommended to be closely followed up without the operation. Immediate operations were indicated for patients in angiographic stage II with more than two TIAs, or with progressive ischemic symptoms and CBF insufficiency shown by CTP imaging. Preferably, the operation was performed on the hemisphere with relatively more severe ischemia, and in cases with similar ischemic conditions in both hemispheres, the operation was performed on the dominant hemisphere. Operations were also recommended in hemorrhagic cases with a history of intracerebral and/or intraventricular hemorrhage. The operation was performed on the hemorrhagic hemisphere. In cases that the hemorrhagic hemisphere could not be clearly identified due to an intraventricular hemorrhage, CTP imaging was employed so as to decide the more severe ischemic hemisphere. If the ischemic conditions were similar between the hemispheres, the operation would be performed on the dominant hemisphere.
Incision and skin flap
A modified pterional approach was adopted to increase the contact area of the temporal muscle. The skin incision was curved posteriorly as much as possible with the posterior branch of the STA included in the flap. The incision was extended by 1.0–1.5 cm above the superior temporal line (Figure 1A). The superior incision did not need to avoid the posterior branch of the STA, as this branch was used only for direct anastomosis, and its length in the flap was sufficient for anastomosis with any of the cortical arteries in the bone window. The anterior and posterior branches of the STA could be clearly observed on the medial surface of the flap (Figure 1B).
Handling of the temporal muscle, the bone flap, and the dura matter
The temporal muscle was incised along the posterior margin of the flap. The temporal muscle was separated from the temporal bone using a periosteal elevator (Figure 1C) to expose the entire DTA network on the deep surface of the temporal muscle. The bone window was expanded carefully towards the sphenoidal crest until the site where the MMA perforated from the bone (Figure 1D). Holes were then drilled 2 cm along the margin of the bone window for later fixation of the dura mater and temporal muscle.
The dura mater on both sides of the sphenoidal crest residue was suspended and fixed to avoid potential avulsion or rupture of the MMA trunk (Figure 1E). Its main branches were preserved intact and incised on both sides, resulting in dura mater strips with a width of 0.5–1.0 cm (Figure 1F). Radial incisions were made on the remaining part of the dura mater. After the hemostasis, the dura mater was flipped over and spread over the bone window with its cranial surface in close contact with the cortical surface (Figure 1G).
In the early operations, we opened the dura and then checked the cortical arteries. If a receptor artery that has appropriate size for anastomosis existed, we separated one or both branches of the STA. This could accurately determine if the recipient vessels and the STA were a good match, but it would increase the exposure time of the brain tissue. In recent operations, preoperative three-dimensional DSA (3D-DSA) has enabled us to determine the length and the number of STA and cortical branches. Thus, we can separate branch(es) of the STA immediately after opening the scalp. The bone window and the dura matter will then be opened so as to reduce the exposure time of the brain tissue and the volume of bleeding.
Selection of the cortical arteries for anastomosis
The cortical arteries for anastomosis were selected according to the following three criteria: (1) The cortical arteries should be adjacent to the STA for anastomosis. (2) The candidate cortical arteries should have a diameter compatible to that of the STA, preferentially greater than 1 mm. As the patients with moyamoya disease often have thin cortical arteries, we suggest that the cortical arteries with a diameter over 0.6 mm be eligible for STA-MCA bypass. (3) The cortical arteries should have as few perforating branches as possible at the site of anastomosis. When two recipient arteries suitable for anastomosis are located on the cerebral surface, the anterior and posterior branches of the STA were dissociated.
The divided branches of the STA were pulled through the temporal muscle to the vicinity of the target arteries for anastomosis. The STA was then rinsed with pressurized normal saline containing heparin. The STA was anastomosed with the cortical artery in an end-to-side fashion using a single 10–0 nylon atraumatic suture. After the anastomosis, Doppler ultrasound and indocyanine green (ICG) fluorescence angiography were performed to verify the patency of the anastomotic stoma.
Dura mater flip-over and temporal muscle placement
The dura matter should be incised along each main branch of the MMA separately (Figure 2A and 2B). The dura mater strips incised along the MMA trunk and branches were flipped over and sutured at the margin of the bone window. Doppler ultrasound was used to verify the patency of the MMA. The flip-over of the dura mater should be abandoned in the following circumstances: (1) Detection of preoperative formation of distal spontaneous anastomoses of the MMA in stage V or VI patients, with the MMA trunk being maintained in its original state with no flip-over. (2) Identification of diminished blood flow by ultrasound in the MMA after the flip-over. (3) Bleeding of the accompanying veins of the MMA. If the bleeding of the accompanying veins could not be controlled by compression with gelatin sponge, the margin of the dura mater could be sutured into the shape of a barrel (with the cortical surface facing outside) containing the MMA, the accompanying veins, and some gelatin sponge for hemostasis (Figure 2C).
The margins of the temporal muscle was sutured with the dura mater at the fold of the flip-over and fixed to the bone window. The bone flap was trimmed into a compatible shape followed by reduction and fixation without compressing the STA. As the bone flap and the temporal muscle exchanged positions in the operation, the inferior edge of the bone flap needed to be elevated a little to reduce the space-occupying effect of the temporal muscle.
Postoperative management and evaluation
The baseline blood pressure was maintained after the operation. No hemostatic agent, antithrombotic or anticoagulant was administered postoperatively, and the drainage tubes were removed within 48 hours after the operation. DSA or CT angiography (CTA) was performed within 1 week after the operation to further verify the patency of the anastomosed arteries (Figure 3A-3C). CTP was also performed within one week to examine the hemodynamic changes, i.e. the CBF, the cerebral blood volume (CBV), and the time-to-peak (TTP). Neurological examinations were carried out; including intelligence, linguistic ability, motor functions, and sensory functions. Six months after the operation, follow-up DSA was performed to examine the formation of collateral circulation (Figure 3D-3H), as well as CTP and neurological examinations. Then the patients were followed in the outpatient department by CTA, CTP, and neurological examinations. At the last follow-up in December 2011, the mean follow-up time was 72.5 months (range, 26.0–133.0 months).
All statistical analyses were performed using SPSS 18.0 for windows (SPSS Inc., USA). Data are presented as mean value and mean ± standard deviation. The change in relative CBF (rCBF) between preoperation and postoperation was determined using the Student's t test. Differences were considered statistically significant at a P value less than 0.05.
Double bypasses (both branches of the ipsilateral STA to the MCA branches) were performed in 59 patients. Eleven patients received bilateral surgery in 6 months to 1 year after the first operation. One hundred unilateral operations were performed including 55 operations on the left and 45 on the right hemisphere. A total of 198 direct artery anastomoses were completed on 122 hemispheres.
Patency of the anastomoses and changes of the arteries
All the 198 stomas had good patency as verified by intra-operative Doppler ultrasound and ICG fluorescence angiography. DSA or CTA performed within one week after the operation showed all the 198 stomas were patent and the STA had successfully begun to supply the MCA region. In the 88 patients with follow-up DSA at 6 months after the operation, patency was maintained in all the anasotomotic stomas, and in 93.2% (82/88) of cases, the STA was thickened by 74.3% (the mean diameter of the STA: pre-operation 1.13 mm, post-operation 1.97 mm, P <0.01), but there was no obvious change in diameter of the STA in the other 6 cases. The DTA constructed via the indirect procedure formed anastomoses with the cortical arteries, and thickened significantly by 145.3% in all the cases (the mean diameter of the DTA: pre-operation 0.64 mm, post-operation 1.57 mm, P <0.01). In 39.8% (35/88) of the patients, the sphenopalatine artery thickened significantly accompanied with spontaneous anastomoses with the cortical arteries. In 90.9% (80/88) of cases, the MMA formed anastomoses with the cortical arteries and thickened significantly by 115.7% (the mean diameter of the MMA: pre-operation 0.57 mm, post-operation 1.23 mm, P <0.01). There was no obvious change in 15.9% (14/88) cases. The anastomoses formed pre-operation between the distal MMA branches and the cortical arteries were preserved well in all stage V or VI patients (Figure 3D-3H).
The moyamoya vessels diminished significantly in 69 cases (78.4%) or disappeared in 19 cases (21.6%). The operated hemisphere, especially the region supplied by the MCA, was supplied predominantly by the ECA branches.
Changes in TIA and clinical symptoms
Of the 76 ischemic cases, 53 patients had TIA. There were 77.4% of cases (41/53) free of TIA within one week after the operation. In the other 22.6% of cases (12/53), patients still suffered from TIA for the short term after the operation, but the frequency of TIA decreased gradually. TIA was arrested 14 days after the operation in 6 cases, 1 month later in 4 cases, and 4 months later in 2 cases. In 16 patients of the 35 hemorrhagic cases accompanied by TIA, 13 patients were free of TIA within one week after the operation, 3 patients still suffered from TIA but the frequency of TIA decreased gradually, and TIA was not observed 6 months after the operation.
There were 90.2% (37/41) of patients with preoperative neurological deficits that showed significant improvement in one week after the operation, but the symptoms did not change in 3 patients and worsened in one patient. At 6 months post the operation, the NIHSS scores of these 41 patients with neurological deficits were lowered by 2.0–8.0 after the operation to a mean of 2.1. The long-term changes of the NIHSS scores are still under investigation, but seem to be similar as the data obtained 6 months post the operation.
Changes in aneurysm and bleeding
As of the last follow-up in December 2011, there were no cases of rehemorrhage in the 35 hemorrhagic cases. In 2 patients with peripheral aneurysms, no particular intervention was taken but the aneurysms disappeared in the follow-up DSA 6 months after the operation.
Hemodynamic changes in the operative hemisphere
Postoperative CTP showed immediate improvement of the CBF in the operated hemisphere. The CBF of both hemispheres increased significantly after the operation in 18.0% (18/100) of patients who received a unilateral operation.
In the ischemic cases, CTP demonstrated a significant increase of the rCBF by 9.2%, from 1.00±0.25 to 1.09±0.23 within one week after the operation (P <0.05). There was also an increase in the hemorrhagic cases by 9.1% in the rCBF from 0.99±0.25 to 1.08±0.28 (P <0.05). At 6 months follow-up, the rCBF further increased by 15.5% (from 0.98±0.26 to 1.13±0.23, P <0.05) with the gradual establishment of the vascular anastomoses due to the indirect procedure.
Two patients had operation-related complications, and neurological deficits worsened after the operation. One patient received the operation at 15 days after symptom onset, and although postoperative transcranial Doppler sonography (TCD) showed a significant increase of the cortical blood flow near the stoma, a new small infarction was detected in the left frontal lobe 3–4 cm away from the stomas at the third day after the operation. The patient suffered from aphasia, but the symptoms were significantly alleviated 6 months later. Another patient received the operation at 10 days after the onset of symptoms, and their hemiplegia was alleviated shortly after the operation. At the third day after the operation, his symptoms worsened and were accompanied by aphasia due to contralateral hemisphere ischemia. We found that the blood pressure of this patient was below 80% of the base line blood pressure. After appropriate treatment, aphasia was alleviated at the tenth day. Two months later, his aphasia disappeared and hemiplegia was alleviated. Five patients suffered from seizures, including 1 patient who had seizures immediately after surgery and 4 patients who had seizures on 1–3 days after surgery. All seizures were well controlled with routine perioperative antiepileptic medication (2–3 weeks). No space-occupying effect of the temporal muscle or cerebral hyperperfusion was found in this series.
Of the 111 patients who received surgery, a total of 198 direct artery anastomoses had been performed on 122 hemispheres. Neurological outcomes in terms of bypass patency, arterial changes, and changes in TIA frequency, changes in bleeding frequency, changes in NIHSS, hemodynamic changes, and complications are to be reported separately. In brief, our experience using the newly combined approach has been very promising, but a longer follow-up is needed to evaluate the long-term effects of our approach in preventing recurrent TIAs or hemorrhages.
The newly designed indirect revascularization termed EDMS, which is a modification of EDS and EMS but is distinguished from EDAS, EMS, and EDAMS, produced promising therapeutic effects in the treatment of moyamoya disease.20 As mentioned above, the CBF increased immediately after the operation due to direct anastomoses between the STA and the MCA. The rCBF increased significantly by around 9% within one week after the operation, and further increased by 15.5% at 6 months after the operation. This approach also allowed maximal use of the MMA and DTA for indirect revascularization, which promoted spontaneous anastomosis 3 or 4 months after the operation.21 There were some disadvantages in the measurements of CBF by CTP, including the radiation dose and the fact that iodinated contrast material may be unsuitable for some patients.
Patients suffered from hemorrhage that often results from the extensive moyamoya vessels that possess poor resistance to the shear stress of the blood flow, or is caused by the rupture of peripheral aneurysms.22–25 The risk of hemorrhage can be considerably lowered by surgical interventions.14,23 In this series, the moyamoya vessels were diminished significantly in 69 cases (78.4%) or disappeared in 19 cases (21.6%) after the operation, and there was no rehemorrhage in these patients as of the last follow-up. In two cases with peripheral aneurysm, the aneurysm disappeared in the follow-up DSA after the operation, demonstrating a decreased risk of recurrent hemorrhage, which was consistent with the results reported by Kuroda et al.26 However, a long-term follow-up is needed to further evaluate the effect of the cerebral revascularization in preventing rehemorrhage.
Our procedure differs from previously described combined surgical procedures 8,10,12,13 in the manipulation of the STA, DTA and MMA. We used the STA only for direct anastomosis. The results obtained so far showed that the STA as the indirect donor vascular graft produces rather poor effects.27 Generally, we used both branches of the STA to undergo direct bypass. Single bypass was performed under the following conditions: (1) Cases having spontaneous anastomoses from the STA anterior branch to cortical artery, mainly on the site of ventricle drainage in hemorrhagic cases. (2) Cases with an overtly low position of the anterior branch of the STA which was prone to injuries to the frontal branch of the facial nerve during the separation of the vessel. (3) Cases with only one branch of the STA which was fit for anastomosis. (4) Cases with very thin vessels lining the cortical surface with only one eligible candidate recipient vessel in the operative field. The above situations could be weighed from a preoperative 3D-DSA.
We consider that the preservation of the integrity of the DTA network is critical to maximize the effect of the donor vascular graft in indirect revascularization procedures. We therefore avoid incising the temporal muscle as previously described. Some authors described either longitudinal or transverse dissection of the temporal muscle, 8,10,12,13,28 but such attempts, according to our experience, will spoil the integrity of the DTA and the SPA. The approach we adopted maintained the integrity of the supplying arteries of the temporal muscle. Six months after the operation, follow-up of DSA has revealed that anastomoses between the DTA and the cerebral cortical arteries formed in all of the treated cases and that the former arteries thickened in all cases. In 39.8% of cases, the SPA also participated in the formation of the collateral circulation, a finding that has not been documented previously. We therefore recommend that the integrity of the temporal muscle be maintained during cerebral revascularization for moyamoya disease.
The integrity of the DTA network and that of the MMA is equally important. The MMA integrity relies mainly on the bone flap design during craniotomy. We adopted a bone flap design similar to that reported by Houkin et al,8 but with a different approach for manipulation of the MMA. According to the preoperative DSA and the intraoperative findings, we performed individualized manipulation of the MMA. In 90.9% (80/88) cases, the MMA formed anastomoses with the cortical arteries and thickened significantly by 115.7%. For stage V or VI patients, the distal meningeal-cortical artery anastomoses formed preoperatively were all preserved well, whereas this has not been described previously. We recommend that in stage V or VI patients, 3D-DSA of the ECA be performed to determine the anatomic relation between the MMA and the cranium, so as to secure the distal anastomoses (Figure 1B-1D).
Operation-related complications occurred in two cases. Both patients received the operation during the early stage after symptom onset, and the complications might have resulted from prolonged intraoperative exposure of the cerebral cortex, or blood pressure control. In other patients with similar symptoms, but who received the operations at least 4 weeks after the onset of symptoms, operation-related complications were not observed. Thus, we suggest conservative therapy for moyamoya disease in the short-term after onset, and surgical intervention can be performed when the neurological condition is stable, preferentially at least 4 weeks after the symptom onset. But further evaluation is needed to determine the appropriate time for the operation. Postoperative seizure after craniotomy is a well-known clinical presentation. However, seizure after revascularization for treating moyamoya disease has rarely been discussed. Currently, there are just a few articles about seizure after revascularization. And the reported incidence of postoperative seizure is 6.0%-18.9%.29,30 In our series, there were five patients who suffered from seizures, including 1 patient who had seizures immediately after surgery and 4 patients who had seizures 1–3 days after surgery. All seizures were well controlled with routine perioperative antiepileptic medication within 3 weeks. In our study, the overall incidence of postoperative seizure was 4.1% (5/122 hemispheres), which is lower than what other authors have described. According to the literature obtained so far,29,30 selecting frontal branches of the MCA as a recipient for STA-MCA bypass contributes significantly to the development of postoperative seizures. We seldom used the frontal branches of the MCA as the recipient artery, which might be the reason that the overall incidence of postoperative seizure was lower than other authors reported.
In conclusion, this study showed that the superficial temporal artery-middle cerebral artery bypass combined with EDMS can achieve good therapeutic effect in the treatment of moyamoya disease.
1. Sakamoto H, Kitano S, Yasui T, Komiyama M, Nishikawa M, Iwai Y, et al. Direct extracranial-intracranial bypass for children with moyamoya disease. Clin Neurol Neurosurg 1997; 99 Suppl 2: S128-S133.
2. Karasawa J, Kikuchi H, Furuse S, Kawamura J, Sakai T. Treatment of moyamoya disease with STA-MCA anastomosis. J Neurosurg 1978; 49: 679-688.
3. Tu YK, Liu HM, Kuo MF, Wang PJ, Hung CC. Combined encephalo-arterio-synangiosis and encephalo-myosynangiosis in the treatment of Moyamoya disease. Clin Neurol Neurosurg 1997; 99 Suppl 2: S118-S122.
4. Matsushima Y, Fukai N, Tanaka K, Tsuruoka S, Inaba Y, Aoyagi M, et al. A new surgical treatment of moyamoya disease in children: a preliminary report. Surg Neurol 1981; 15: 313-320.
5. Kinugasa K, Mandai S, Kamata I, Sugiu K, Ohmoto T. Surgical treatment of moyamoya disease: operative technique for encephalo-duro-arterio-myo-synangiosis, its follow-up, clinical results, and angiograms. Neurosurgery 1993; 32: 527-531.
6. Karasawa J, Kikuchi H, Kawamura J, Sakai T. Intracranial transplantation of the omentum for cerebrovascular moyamoya disease: a two-year follow-up study. Surg Neurol 1980; 14: 444-449.
7. Karasawa J, Kikuchi H, Furuse S, Sakaki T, Yoshida Y. A surgical treatment of “moyamoya” disease “encephalo-myo synangiosis”. Neurol Med Chir 1977; 17: 29-37.
8. Houkin K, Kamiyama H, Takahashi T, Kuroda S, Abe H. Combined revascularization surgery for childhood moyamoya disease STA-MCA and encephalo-duro-arteriomyo-synangiosis (EDAMS): technical note. Childs Nerv Syst 1997; 13: 24-29.
9. Ishikawa T, Houkin K, Kamiyama H, Abe H. Effects of surgical revascularization on outcome of patients with pediatric Moyamoya disease. Stroke 1997; 28: 1170-1173.
10. Ishikawa T, Kamiyama H, Kuroda S, Yasuda H, Nakayama N, Takizawa K. Simultaneous superficial temporal artery to middle cerebral or anterior cerebral artery bypass with pan-synangiosis for Moyamoya disease covering both anterior and middle cerebral artery territories. Neurol Med Chir (Tokyo) 2006; 46: 462-468.
11. Kuroda S, Houkin K. Moyamoya disease: current concepts and future perspectives. Lancet Neurol 2008; 7: 1056-1066.
12. Kim DS, Kang SG, Yoo DS, Huh PW, Cho KS, Park CK. Surgical results in pediatric moyamoya disease: angiographic revascularization and the clinical results. Clin Neurol Neurosurg 2007; 109: 125-131.
13. Matsushima T, Inoue T, Ikezaki K, Matsukado K, Natori Y, Inamura T, et al. Multiple combined indirect procedure for the surgical treatment of children with moyamoya disease. A comparison with single indirect anastomosis and direct anastomosis. Neurosurg Focus 1998; 5: e4.
14. Houkin K, Kamiyama H, Abe H, Takahashi A, Kuroda S. Surgical therapy for adult moyamoya disease. Can surgical revascularization prevent the recurrence of intracerebral hemorrhage? Stroke 1996; 27: 1342-1346.
15. Xu B, Song DL, Mao Y, Xu H, Gu YX, Chen G, et al. Use superficial temporal artery-middle cerebral artery bypass
combined with encephalo-duro-myo-synangiosis to treat moyamoya disease. Chin J Cerebrovasc Dis (Chin) 2007; 4: 445-448.
16. Xu B, Song DL, Mao Y, Xu H, Liao YJ, Zhou LF. Combine direct and indirect revascularizations to treat moyamoya disease. Chin J Neurosurg (Chin) 2009; 25: 102-105.
17. Suzuki J, Takaku A. Cerebrovascular “moyamoya” disease. Disease showing abnormal net-like vessels in base of brain. Arch Neurol 1969; 20: 288-299.
18. Matsushima Y, Aoyagi M, Niimi Y, Masaoka H, Ohno K. Symptoms and their pattern of progression in childhood moyamoya disease. Brain Dev 1990; 12: 784-789.
19. Roach ES, Golomb MR, Adams R, Biller J, Daniels S, Deveber G, et al. Management of stroke in infants and children: a scientific statement from a Special Writing Group of the American Heart Association Stroke Council and the Council on Cardiovascular Disease in the Young. Stroke 2008; 39: 2644-2691.
20. Dauser RC, Tuite GF, McCluggage CW. Dural inversion procedure for moyamoya disease. Technical note. J Neurosurgery 1997; 86: 719-723.
21. Houkin K, Nakayama N, Kuroda S, Ishikawa T, Nonaka T. How does angiogenesis develop in pediatric moyamoya disease after surgery? A prospective study with MR angiography. Childs Nerv Syst 2004; 20: 734-741.
22. Fukui M, Kono S, Sueishi K, Ikezaki K. Moyamoya disease. Neuropathology 2000; 20 Suppl: S61-S64.
23. Irikura K, Miyasaka Y, Kurata A, Tanaka R, Fujii K, Yada K, et al. A source of haemorrhage in adult patients with moyamoya disease: the significance of tributaries from the choroidal artery. Acta Neurochir 1996; 138: 1282-1286.
24. Iwama T, Morimoto M, Hashimoto N, Goto Y, Todaka T, Sawada M. Mechanism of intracranial rebleeding in moyamoya disease. Clin Neurol Neurosurg 1997; 99 Suppl 2: S187-S190.
25. Kawaguchi S, Okuno S, Sakaki T. Effect of direct arterial bypass on the prevention of future stroke in patients with the hemorrhagic variety of moyamoya disease. J Neurosurg 2000; 93: 397-401.
26. Kuroda S, Houkin K, Kamiyama H, Abe H. Effects of surgical revascularization on peripheral artery aneurysms in moyamoya disease: report of three cases. Neurosurgery 2001; 49: 463-468.
27. Houkin K, Kuroda S, Ishikawa T, Abe H. Neovascularization (angiogenesis) after revascularization in moyamoya disease. Which technique is most useful for moyamoya disease? Acta Neurochir (Wien) 2000; 142: 269-276.
28. Ozgur BM, Aryan HE, Levy ML. Indirect revascularisation for paediatric moyamoya disease: the EDAMS technique. J Clin Neurosci 2006; 13: 105-108.
29. Fujimura M, Kaneta T, Mugikura S, Shimizu H, Tominaga T. Temporary neurologic deterioration due to cerebral hyperperfusion after superficial temporal artery-middle cerebral artery anastomosis in patients with adult-onset moyamoya disease. Surg Neurol 2007; 67: 273-282.
30. Jin SC, Oh CW, Kwon OK, Hwang G, Bang JS, Kang HS, et al. Epilepsy after bypass surgery in adult moyamoya disease. Neurosurgery 2011; 68: 1227-1232.