Long Wang, MD, PhD, Shuaibin Lu, MD, Li Cai, MD, Hai Qian,MD, PhD, Rokuya Tanikawa, MD, and Xiang’en Shi, MD, PhD

自2009年底至2010年初，我们首次在国内外用桡动脉移植，颌内动脉作为供血动脉与脑血管搭桥，先后治疗了2例大脑中动脉闭塞患者及1例后循环巨大梭形动脉瘤患者。手术后动脉造影见颌内动脉-大脑中动脉或颌内动脉-大脑后动脉搭桥血管血流通畅;脑缺血得以改善，动脉瘤内血栓形成，患者症状明显减轻。此后，我们的研究先后在国内和欧洲神经外科专业杂志《Acta Neurochirurgica》(2011;153(8):1649-1655)发表，被认为是“切实创新的有利方法”，是“巧妙的颅内外血流搭桥，解决了临床的关键问题”。经文献检索国内外尚无临床应用报告。2015年，我们用颌内动脉搭桥治疗复杂性颅内动脉瘤的报告《Management ofcomplex intracranial aneurysms with bypasssurgery: a technique application and experience in93 patients》 (血管搭桥在治疗复杂动脉瘤的技术应用和93例经验)在国际著名专业杂志《NeurosurgicalReview》(8/26/2014)上发表，被美国巴罗神经外科研究所主任M.T. Lawton发表专题评论，“颌内动脉搭桥结果表明，该方法确切实用”。美国教授Linx从67种神经外科专业杂志中选出的最有影响的前10篇文章中，该论文被同行的关注度和检索排名第五名。

近年来的经验，我们的综述文章《The History andEvolution of Internal Maxillary Artery Bypass》(颌内动脉搭桥的历史与发展)于2018年3月在美国世界神经外科杂志《World Neurosurgery》 (2018，113:320-332)发表。美国著名的国际脑血管外科专家，纽约迈蒙尼德医学中心神经外科主任 E. Nossek教授, 扎克医学院神经外科主任 D. J. Langer教授罕见地在美国世界神经外科杂志发表述评(World Neurosurgery (2018)115:44-46.)：“颌内动脉与大脑中动脉搭桥是承担脑血流代替治疗的‘新范式’( new work horse )”，认为颌内动脉至大脑中动脉搭桥与传统的颈内动脉/颈外动脉至大脑中动脉相比有多处优势。首先, 颌内动脉到大脑中动脉搭桥是在显微外科手术视野范围内操作，避免了以往颈部二次手术切口;其次, 较短的血管移植长度完全减少了以往过长移植血管扭曲和皮下隧道穿行压迫性阻塞的并发症。称赞离断颧弓扩大颞下窝开颅，可确实地鉴别颌内动脉，得到较好的手术显露，达到的近心端吻合。认为在颞浅动脉搭桥

专刊封面首次刊登颌内动脉搭桥脑血管重建治疗复杂大脑中动脉瘤照片

2019 年2 月世界著名神经外科聚焦杂志(Nurosurgical Focus)发表关于脑血管重建专刊，发表了我们复杂性大脑中动脉瘤血管重建治疗(NeurosurgFocus2019;46(2):E10)，并在文章封面刊登该我们新颖创新血管重建影像图，这也是该杂志首次向世界专业学者展示中国学者血管重建治疗颅内复杂动脉瘤的新创方法，得到国际同行认可，被赞誉为教课书式经典。

全文如下：

Internal maxillary artery bypass for the treatment of complex middle cerebral artery aneurysms

THE innovation of endovascular tools in clinical prac-tice has put the indications for microsurgical repair in a constant state of ebb and flow. Although the treatment algorithm for intracranial aneurysms has been changed, microsurgery remains superior to endovascu-lar therapy in the territory of the middle cerebral artery (MCA), which is frequently involved with perforating lenticulostriate arteries (LSAs), coursing in a curved mor-phology and situated peripherally. In selected cases, direct clipping across the aneurysm neck without blood flow impairment may pose a technical challenge by virtue of the complex angioarchitecture. In such cases, complex MCA aneurysms (CMCAAs) continue to require bypasses as part of a curative treatment strategy with deliberate oc clusion.

Recently, the internal maxillary artery (IMA) bypass, the only substantive modifcation of cerebral revascular ization techniques,has gained momentum. This technique was heralded as a new “workhorse” in the feld of high-fow bypass to replace conventional cervical artery bypass. However, very limited data exist regarding the experiences with using this novel extracranial artery as a donor, especially in the territory of the MCA. In the present study, we aimed to investigate the surgical outcomes of IMA bypass in the treatment of CMCAAs.

Methods

Data Collection

This study retrospectively evaluates our experience us ing an IMA bypass to treat CMCAAs between January 2010 and July 2018. The protocol was approved by the in stitutional review board and ethics committee of SanBo Brain Hospital, Capital Medical University. An informed consent form was signed by all patients and documented in each case. The patient demographic and clinical data were extracted from our electronic medical record system. The following data were collected: demographic information; clinica parameters (aneurysm characteristics, antiplatelet and anticoagulation regimens, and operative reports); ra diographic studies; intraoperative measurement data and images; and outpatient clinical data. An experienced in vestigator who was blinded to the patients’ clinical charac teristics independently reviewed all the data. The patients’characteristics are given in Table 1.

IMA Bypass Decision-Making

The treatment strategy is tailored on a case-by-case basis by multiple considerations, including patient age, family history of subarachnoid hemorrhage (SAH), medi cal comorbidities, and psychological impact of aneurysm (morphology, location, size, rupture status).The MCA an eurysms that related to at least 2 of the following features were defned as complex in character:1) size>2.5cm;2) aneurysm angioanatomy (involvement of critical perforat ing or branch vessels); 3) previous treatment (endovascular or surgical); 4) complex morphology(fusiform,dissecting,serpentine,and dolichoectatic); 5) intraluminal thrombo sis; or 6) atherosclerotic plaques and calcifcations of the aneurysm wall and/or neck. Both surgical and endovascu lar treatment options are offered.

Therapeutic alternatives were discussed after diagnostic angiography among neurosurgeons,

neurointerven tionalists, and neurovascular neurologists in a multidis ciplinary decision-making process. Patients and families had the freedom to choose a therapy. However, if direct clipping/clip reconstruction or neurointerventional treat ment was deemed futile on

the basis of preoperative im aging fndings, patients were directly recommended for cerebral bypass.

The type of bypass performed in the MCA territory depended on the aneurysm origin, LSA involvement, caliber and location of the occluded artery, viability of the collat eral circulation, and the potential graft fow. The high-fow extracranial-intracranial (EC-IC) bypass was preferred to achieve the deliberate occlusion of large-diameter arteries (prebifurcation or bifurcation lesions), whereas low-fow bypass was frst considered for distal MCA lesions. The cervical high-fow bypass has been replaced by the novel IMA bypass due to the avoidance of a second incision, short graft, and well-matched caliber. Considering many false-negative results, the balloon temporary occlusion test was rarely used in our series and did not affect our strategies.

MCA Aneurysm Location

The aneurysm location was divided into 4 subtypes ac cording to a new classifcation:1) sphenoidal segment of the MCA (M 1 )–LSA; 2) M 1 –early cortical branch (ECB) (i.e., M 1 aneurysms arising on M 1 at the origin of early frontal or temporal ECB); 3) MCA bifurcation or trifurca tion aneurysm; and 4) distal MCA aneurysms that arise on the insular, opercular, or cortical segments of the MCA. However, in our series, a majority of cases manifested with complex characteristics, and specifc location was determined in a setting of multidisciplinary discussion.

Bypass Technique and Perioperative Management

The techniques of performing IMA bypass have been described in our previous articles.Briefy, we approached the aneurysms using a frontotemporal crani otomy with zygomatic osteotomy. The radial artery (RA) was harvested and prepared before being anastomosed in an end-to-end fashion to the pterygoid segment of the IMA (I 2 ) by using 9-0 prolene. The free end of the RA was then brought to the sylvian fssure and anastomosed to the insular segment of the MCA (M 2 ) by using 9-0 sutures in an end-to-side manner. The aneurysm located after the bypass takeoff was then resected or trapped with a perma nent aneurysm clip or ligation.

For patients without a history of SAH, 300 mg of as pirin was given 3 days prior to the procedure. During the procedure, neurophysiological monitoring was used to identify cerebral ischemia during temporary occlusion of the parent artery. Doppler fow

measurements (PRO Focus 2202, BK Medical) and indocyanine green videoangiography were used to analyze the adequacy of treatment and patency of parent vessels. After the procedure, all patients were monitored overnight in a neurointensive care unit under the supervision of neurosurgeons, and blood pressure was kept 10%–20% higher than the baseline. On postoperative day 2, subcutaneous anticoagulation with low-molecular-weight heparin was prescribed for 3 consecutive days, and then patients were kept on 300 mg of aspirin per day as in the preoperative algorithm. Three months after the procedure, each patient was maintained on a 100-mg single daily dose of aspirin. Patients were loaded with 1.0 mg/hr nimodipine for 7–10 days after the intervention.

Imaging Analysis and Follow-Up

Postoperative and 3D CT angiography (CTA), which was evaluated independently by neuroradiologists, was scheduled at 6–8 hours after treatment to confrm the patency of the graft conduit and to look for hemorrhagic complications.If worsening neurological function was ob served,multimodal MRI was performed immediately to identify territorial ischemia. The aneurysm stability was classifed as complete occlusion/obliteration (no residual aneurysm), greatly diminished (residual neck), or incom plete occlusion (aneurysm lumen remaining). The follow up cerebral angiogram was performed after 3 months, and then yearly thereafter.

TABLE 1. Clinical characteristics of 12 patients with CMCAAs treated by IMA bypass

The neurological outcomes were assessed using a modifed Rankin Scale (mRS) by one of the authors (S.L.) in a blinded fashion, and the scores were recorded. Follow up functional outcome was measured based on telephone interviews with the patient or, if the patients were unable to participate, their relatives or primary caregivers were asked to provide information. If complications occurred, mRS scores were determined on an outpatient visit with physical examination.

Results

Patients and Aneurysm Characteristics

During an 8-year period, 12 consecutive patients har boring CMCAAs were treated by IMA bypass; the mgroup included 9 males and 3 females. The average age of pa tients was 31.7 years, with a range of 14–56 years.The most common presentation that led to the diagnosis was headache. Other presenting symptoms varied, and includ ed facial numbness, dizziness, and seizure. SAH was ob served in 3 patients, of which 2 were Hunt and Hess grade II and 1 was Hunt and Hess grade III. One patient had a thrombosed giant aneurysm following a frst attempt at treatment. The pretreatment mRS score (mean 1.6; range 0–4) was 0–1 for 9 patients and 3–4 for 3 patients.

The mean size of the CMCAA was 23.7 mm (range 10–37 mm), and all the aneurysms were large (n = 6) or giant (n = 6). Two aneurysms were located at the M1–LSA segment, 3 at the M1–ECB segment, 2 at the MCA bifur cation, 1 at the distal MCA segment, and 4 extensively in volved the MCA. Treated lesions presented with fusiform morphology in 7 patients, thrombotic in 6 patients, serpen tine in 1 patient, and recurrent after a primary attempt in 1 patient. The baseline patient and aneurysm characteristics are summarized in Table 2. Immediate and Long-Term Outcomes of IMA Bypass Surgery

The proximal occlusion of the parent arteryfollowing IMA bypass was attained in 8 cases (75%) with (n = 2) or without (n=6)aneurysmorrhaphy(AR).The remaining aneurysms were completely trapped with (n =1) or with out (n=2) AR or completely resected(n =1).Intraopera tive complications, such as unexpected rupture, were not observed. Of the 12 patients who were initially included in this study, 10 (83.3%) achieved complete aneurysm elimination on immediate angiographic results, and the aneurysms of the other 2 (16.7%) were greatly diminished. Postoperative angiograms documented graft patency in 10 of 12 patients (83.3%). Surgery-related deaths did not oc cur in the perioperative period. The mRS score (mean 1.7; range 0–4) at discharge was 0–2 in 9 patients and 3–4 in 3 patients, which was equal to the preoperative status. Three patients had unfavorable outcomes, of whom 2 h ad new permanent neurological defcits by virtue ofthegraft occlusion and 1 was slightly disabled as refected in the poor Hunt and Hess status. The angiographic and clinical outcomes after the procedure are summarized in Table 3.

No adverse events (patient death, posttreatment rup ture, or aneurysm recurrence) were observed during hos pitalization and follow-up. In the angiographic follow-up study (mean 28.7 months; range 2–74 months), patent RA was found in 83.3% (n=10) of patients, and the complete occlusion rate of aneurysms was 100%. The median du ration of follow-up focusing on neurological functions was 53.1 months (range 19–82 months), and excellent out comes were achieved in all patients (mRS score: mean 0.3; range 0–1). Regarding the cosmetic issue in the region of the temporalis muscle, no patient complained of fullness and discomfort at a level that was bothersome enough to merit intervention. The outcomes at long-term follow-up are summarized in Table 4.

Illustrative Cases

A 37-year-old woman (case 12) was admitted to our department with a 1-month history of severe headache, and a mass lesion in the sylvian fssure was detected. Her past medical history was unremarkable, and neurological examination on admission revealed no focal defcits. Pre-operative CTA (Fig. 1) was performed and confrmed a giant, fusiform, thrombosed aneurysm that involved the M 1 –M 2 segment of the MCA. Following a multidisci plinary discussion, microsurgical repair or endovascular treatment modalities were deemed likely to fail. There fore, a high-fow IMA bypass was considered, followed by parent artery occlusion to preserve the perforators arising from the aneurysm wall. The RA was harvested and prepared after confrmation of the Allen test, and the aneurysm was approached using a frontotemporal crani otomy with zygomatic osteotomy to expose the I 2 . The ex tracranial IMA-RA bypass was initially performed in an end-to-end manner, and then the free end of the RA was anastomosed to the M 2 in an end-to-side manner. Follow ing the bypass, patency was confrmed by intraoperative Doppler ultrasonography and indocyanine green, and the proximal part of M 1 was occluded with a permanent aneu rysm clip (Aesculap Instruments Corp.) (Fig. 2). Complete elimination of the aneurysm with a patent graft artery was confrmed by postoperative angiography (Fig. 3), and the patient was discharged, having attained independent status (mRS score of 2). She underwent follow-up evalua tion at 19 months after surgery through an outpatient visit without exhibiting any functional neurological limitations (mRS score of 0).

FIG. 1. Preoperative axial (A), coronal (B), and sagittal (C) CTA demonstrated a giant (36 × 25 × 25 mm), thrombosed, fusiform MCA aneurysm. 3D CTA (D–F) images were then reconstructed to identify the specifc location of the aneurysm involving the M1–M2 MCA segment.

Complications and Clinical Impact

Two patients experienced permanent neurological worsening after IMA bypass. The frst patient (case 2) had an MCA bifurcation aneurysm (Fig. 4A and B), andthe atherosclerotic nature of the lesion was noted intraoperatively (Fig. 4C). Following the primary attempt at arterial reconstruction between RA and temporaM₂　, stenosis of theanastomosis site was detected immediately. The bypass was disassembled and the opening of temporal M₂　was sutured. The M2segment of the aneurysm wall was then chosen as a recipient, and microsurgical anastomo sis was performed uneventfully in an end-to-side fashion. Neurophysiological monitoring showed no change dur ing the bypass, and Doppler imaging identifed patency of the graft with abundant distal fow. However, the pa tient’s functions gradually decreased, and neuroimaging demonstrated left frontotemporal ischemia with severe vasospasm of the RA (Fig. 4D – F). Mannitol and nimodi- pine were then prescribed for conservative management. Although the patient had weakness of the right extremity (muscle strength grade 2) at discharge with moderate dis ability (mRS score of 3), her symptoms were completely resolved without neurological defcits (mRS score of 1) at the last follow-up.

In the second patient (case 8), a giant MCA bifurca tion aneurysm was resected following IMA-RA–temporal M₂ 　bypass.Although　the perforators that arose from the aneurysm wall were not observed, the proximal M₁ seg ment that was involved in the lesion was partially sacri fced intraoperatively. Postoperative CTA revealed graft occlusion, and the

patient experienced left hemiparesis on postoperative day 7. The basal ganglia infarction was veri fed by neuroimaging with the suspicion of M₁　perforating artery thrombosis. The patient was discharged with severe disability (mRS score of 4) but recovered uneventfully during the 4-year follow-up.

FIG. 2. IMA-RA-M 2 bypass was performed followed by proximal occlusion of the aneurysm. An intraoperative picture (left) show ing 2 anastomosis sites (white asterisks) and clip ligation of proximal M1 (black asterisk). Artist’s rendering (right) of the patient’s microsurgical treatment with IMA bypass for the management of CMCAA. An = aneurysm; Ant. = anterior; Max. A = maxillary artry; O.N = optic nerve; Pcom = posterior communicating artery; RAG = right arterial graft; Temp. A = temporal artery. Copyright Li Cai (right panel). Published with permission.

These 2 occlusion cases that resulted in temporary neu rological defcits could be explained by the perforator sac rifce and inappropriate recipient site selection.

Discussion

This study of 12 patients with IMA bypasses for the CMCAAs observed on immediate angiographic sequenc es showed that the patency rate was 83.3% and that all the aneurysms were removed from the circulation. In 2 cases (16.7%) our patients experienced ischemic events that re sulted in new permanent neurological defcits. In one case the surgeon had used the aneurysm wall as the recipient, and the other aneurysm was resected intraoperatively, which may compromise the surrounding perforators. How ever, over a long-term follow-up, all patients experienced unremarkable outcomes, and negative events (aneurysm recurrence or late-phase bleeding) were not encountered. We emphasize that radical aneurysm resection or using the aneurysm wall as a recipient should be avoided. To our knowledge, this represents the largest case series reported to date, and no previous study has attempted to describe the failed experience with IMA bypass. Our results predict that, even in the endovascular era, high-fow bypass retains an essential role for complex cerebrovascular disorders.

Endovascular Treatment of CMCAAs

Recently, techniques in which fow diversion devices (FDDs) are used to treat complex cerebrovascular diseases have documented acceptable safety profles, which result in a decrease in the number of indications for a classic surgical approach.11 However, the treatment results remain suboptimal in the territory of the MCA, which should be evaluated as a separate entity for the following reasons. First, the estimated proportion of MCA aneurysms that are amenable to treatment with the FDD was only 5%. More over, this device is not currently indicated for aneurysms on tortuous sidewall or end-wall bifurcations, and those arising from bifurcation represent 85% of all MCA aneu rysms. Even with the latest-generation devices, there is in vitro and in vivo experimental evidence that FDDs are more prone to fail. As for MCA aneurysms, perforator injury is another potential concern. FDDs often cover or “jail” side branches, which can cause unintended occlusions and ischemic complications. The rate of side-branch occlusion was strongly associated with bifurcation aneurysms. In addition, delayed hemorrhage and aneurysm rupture are complications that must be fac tored into the treatment algorithm.The requirement for antiplatelet agents also limits their application in patients with a history of rupture. Another major concern is the high recanalization and retreatment rates, as well as the high risk of incomplete occlusion, which may render the endovascular method ineffective. Therefore, we should also remind ourselves that the clinical course after FDD placement is not inevitably benign or without risk, and the morbidity and mortality rates associated with these endo vascular series consistently exceeded those reported by experienced microneurosurgeons.

FIG. 3. Postoperative axial (A), coronal (B), and sagittal (C) CTA confrmed the total elimination of the aneurysm, and 3D CTA (D–F) illustrated the patent status of the graft vessel, with robust distal fow.

Microsurgical Selection for CMCAAs

The primary treatment strategy for MCA aneurysms re mains microsurgical repair. However, when an aneurysm presents complex characteristics, such as an intraluminal thrombus, mycotic or infectious etiology, atherosclerotic or calcifed neck, nonsaccular morphology, and incorpora tion of perforators, complete surgical clipping is infeasible or at least notoriously diffcult.In those circumstances, cerebral revascularization must be considered as an al- ternative method that depends on the aneurysm location, LSA anatomy, and rupture status.

Recently, intracranial-to-intracranial bypass has been widely used for its simpler and less invasive characteris tic. At the same time, several potential shortcomings including anatomical constraints and technical demands should be considered. Besides, the frequent involvement of perforators, as in most cases in this study, may prohibit this technique from being indicated for proximal MCA aneurysms. However, in situ side-to-side bypass is particularly attractive for arterial reconstruction of distal MCA (M ₂ or M ₃ [opercular segment of the MCA]) arborization because this technique eliminates the need for graft harvesting or the further dissection of extracranial donor vessels. Meanwhile, the low fow volume is adequate for the blood replacement of distal vasculatures.

The EC-IC bypass is a well-established method for treating complex aneurysms and is often performed in combination with a trapping strategy. Proximal occlusion, as a subtype of partial trapping, is preferred for producing a retrograde fow from the distal part of the aneurysm and facilitating intraaneurysmal thrombosis. Complete trapping may be the best alternative for aneurysms free of perforating vessels.

FIG. 4. Preoperative axial (A) and coronal (B) CTA illustrated an MCA bifurcation aneurysm. Intraoperative photograph showing atherosclerotic characteristics of the lesion (C). The patent graft was not observed on postoperative angiography (D and E) stud ies, and left frontotemporal ischemia was detected (F)

In rare cases, efferent outfow occlusion may be preferable when the afferent artery is diffcult to reach and sacrifce, although negative results have been documented in the literature. In addition, AR may be implemented following vascular reconstruction for de bulking purposes.

The Role of IMA Bypass in Cerebral Revascularization

Superfcial temporal artery (STA)–MCA bypass is the standard type of cerebral revascularization. However, insuffcient blood fow may result in negative events following therapeutic occlusion of the main trunk arteries that should be replaced by high-flow revascularization. Recently, the IMA bypass was redefned as a new “workhorse” in the feld of high-fow bypass to replace conventional cervical artery bypass,in which long-term grafts are prone to kink subcutaneously and second-site incision is mandatory.

However, complex and traumatic donor vessel harvesting that required zygomatic osteotomy or middle fossa removal during the IMA bypass discouraged broader acceptance of this technique into neurosurgical practice. Recently, aless invasive method without bone sacrifce to identify I 2 was proposed. Meanwhile, new bone landmarks to locate I 2 in a simpler and more effcient way were also introduced by conducting surgical simulation in cadavers. However, we should keep in mind that the working space on a fxed cadaver head is wider than that of a live patient because cadaver tissue is stiff and atrophied.No tably, the only clinical case presented by the investigators had received 2previous procedures, and the middle fossa had been partially removed before IMA bypass. Therefore this is not the truly minimally invasive IMA bypass operation as defned. Meanwhile, regarding the issues of severe postoperative pain and chewing diffculties, no patient complained of fullness and discomfort of the temporalre gion in the current study. From our clinical experience after performing 130 cases of IMA bypass, exposing I 2 without bone sacrifce is not reasonable and is extremely diffcult due to obstruction by the temporalis muscle. Our position is that a comparative study of the advantages and limitations of each exposure technique should be conducted in the near future.

As for the interposition graft of IMA bypass, the RA and forearm cephalic vein have been used in clinical settings. Recently, preclinical studies focusing on STA trunk grafts have been comprehensively reported in the literature. However, its diameter mismatch withdonor vessels and the short length available for approaching EC-IC vessels make tension-free reconstruction diffcult. Notably, an article with “STA graft” in its title included the only case of IMA bypass with RA graft. We highly recommend that RA be used as the preferred graft in IMA bypass procedures by virtue of suitable arterial diameter and adequate graft length. Meanwhile, the synchronous performance of craniotomy and graft harvesting by 1 primary surgeon and 2 assistants makes the procedure less time-consuming. The forearm cephalic vein is another valid option for those who fail

the Allen test. The STA trunk may only be considered when other reliable grafts are not available. Therefore, IMA bypass with RA following bone drilling remains the benchmark in this novel bypass surgery. Limitations of the Study The main inherent limitation of this study was theretrospective nature of the data analysis, which leads to potential ascertainment bias. Besides, the limited number of patients may have precluded a firm conclusion regard- ing the safety and efficacy of IMA bypass for treating CMCAAs. In addition, this study was not a randomized controlled trial, which is another limitation. Further stud- ies with more cases and a control group are necessary to confirm our findings. Our results also reflected the experi- ence of a single surgeon who used a defined set of proto- cols at a single center, which may not be representative of the neurosurgical cases in other large academic centers and may limit the generalizability of the data.

Conclusions

The satisfactory results in the present study demonstrate the feasibility of treating CMCAAs by using an IMA bypass procedure; the rates of periprocedural com- plications and morbidity were acceptable, and the efficacy of the surgery was high. This novel arterial reconstruction procedure is a promising method for complex cerebrovas- cular cases and should be kept in the armamentarium of neurosurgeons. However, intraoperative radical aneurysm resection or inappropriate recipient site selection should be avoided in this subtype of MCA aneurysm. Despite advances in endovascular and neurointerventional tech- niques, EC-IC bypass remains an essential solution and, indeed, the dominant option when performed by experi- enced hands.

Acknowledgments

We received financial support from the Beijing Municipal Natural Science Foundation (Grant No. 7161005 to X.S.) and the Science and Technology Commission Foundation of Beijing (Grant No. Z161100000516019 to X.S.). Long Wang expresses his sincere gratitude to Xiang’en Shi for supporting him as a clinical research fellow at Barrow Neu- rological Institute and an advanced training fellow in Sapporo Teishinkai Hospital. He also thanks his beloved parents, Yan Wu and Haige Wang, for their gracious consideration and great confi- dence during his entire 29 years of life. Finally, he appreciates the invaluable support and companionship he has received from his wife Lujun Jing over the last 5 years.