The Transperiosteal “Inside-Out” Occipital Artery Harvesting Technique

LABORATORY INVESTIGATION J Neurosurg 130:207–212, 2019

The transperiosteal “inside-out” occipital artery harvesting technique

*Arnau Benet, MD,1,2 Halima Tabani, MD,1,2 Xinmin Ding, MD, PhD,1,2 Jan-Karl Burkhardt, MD,1,2 Roberto Rodriguez Rubio, MD,1,2 Ali Tayebi Meybodi, MD,1,2 Peyton Nisson, BS,2 Olivia Kola,2 Sirin Gandhi, MD,1,2 Sonia Yousef, BS,2 and Michael T. Lawton, MD1,2

1Department of Neurological Surgery, and 2Skull Base and Cerebrovascular Laboratory, University of California, San Francisco, California

OBJECTIVE The occipital artery (OA) is a frequently used donor vessel for posterior circulation bypass procedures due to its proximity to the recipient vessels and its optimal caliber, length, and flow rate. However, its tortuous course through multiple layers of suboccipital muscles necessitates layer-by-layer dissection. The authors of this cadaveric study aimed to describe a landmark-based novel anterograde approach to harvest OA in a proximal-to-distal “inside-out” fashion, which avoids multilayer dissection. METHODS Sixteen cadaveric specimens were prepared for surgical simulation, and the OA was harvested using the classic (n = 2) and novel (n = 14) techniques. The specimens were positioned three-quarters prone, with 45° contralater- al head rotation. An inverted hockey-stick incision was made from the spinous process of C-2 to the mastoid tip, and the distal part of the OA was divided to lift up a myocutaneous flap, including the nuchal muscles. The OA was identified using the occipital groove (OG), the digastric muscle (DM) and its groove (DG), and the superior oblique muscle (SOM) as key landmarks. The OA was harvested anterogradely from the OG and within the flap until the skin incision was reached (proximal-to-distal technique). In addition, 35 dry skulls were assessed bilaterally (n = 70) to study additional craniometric landmarks to infer the course of the OA in the OG. RESULTS The OA was consistently found running in the OG, which was found between the posterior belly of the DM and the SOM. The mean total length of the mobilized OA was 12.8 ± 1.2 cm, with a diameter of 1.3 ± 0.1 mm at the suboccipital segment and 1.1 ± 0.1 mm at the skin incision. On dry skulls, the occipitomastoid suture (OMS) was found to be medial to the OG in the majority of the cases (68.6%), making it a useful landmark to locate the OG and thus the proximal OA. CONCLUSIONS The anterograde transperiosteal inside-out approach for harvesting the OA is a fast and easy technique. It requires only superficial dissection because the OA is found directly under the periosteum throughout its course, obviating tedious layer-by-layer muscle dissection. This approach avoids critical neurovascular structures like the vertebral artery. The key landmarks needed to localize the OA using this technique include the OMS, OG, DM and DG, and SOM. https://thejns.org/doi/abs/10.3171/2017.6.JNS17518 KEY WORDS aneurysms; bypass; occipital artery; occipital groove; revascularization; posterior circulation; far-lateral approach; surgical technique

he occipital artery (OA) is a frequently used donor ported early experience with posterior circulation revas- vessel for extracranial-intracranial bypass proce- cularization achieved using the OA,13 the OA has been dures to the posterior circulation, by virtue of its widely used as a donor vessel in treating posterior circula- Tproximity to the recipient vessels and its optimal caliber, tion aneurysms, ischemic neurological disorders, and rees- length, and flow rate.3 After Sundt and Piepgras first re- tablishing blood flow after resection of skull base tumors,

ABBREVIATIONS AICA = anterior inferior cerebellar artery; DG = digastric groove; DM = digastric muscle; OA = occipital artery; OG = occipital groove; OMS = occipitomastoid suture; PICA = posterior inferior cerebellar artery; SOM = superior oblique muscle; VA = vertebral artery. SUBMITTED February 26, 2017. ACCEPTED June 13, 2017. INCLUDE WHEN CITING Published online January 26, 2018; DOI: 10.3171/2017.6.JNS17518. * Drs. Benet and Tabani contributed equally to this work.

©AANS 2019, except where prohibited by US copyright law J Neurosurg Volume 130 • January 2019 207

Unauthenticated | Downloaded 10/05/21 10:15 AM UTC A. Benet et al. particularly those involving the vertebral artery (VA) or distal) approach for harvesting the OA was performed in posterior inferior cerebellar artery (PICA).3,13 the rest of the specimens (n = 14) to test the feasibility Although the OA has sufficient length and caliber to of the technique and to define the surgical landmarks for effectively redirect blood flow to the posterior circula- identification of the OA (Video 1). tion,10 it has a tortuous course in which it passes through VIDEO 1. Clip showing anterograde inside-out OA harvest. The 3D multiple layers of suboccipital muscles, necessitating lay- video depicts stepwise dissection of the OA aided by the antero- er-by-layer dissection for exposure. Therefore, harvesting grade inside-out technique. Copyright Arnau Benet. Published with the OA has been regarded as time consuming and more permission. Click here to view. challenging than harvest of the superficial temporal ar- The specimens were positioned in a three-quarter tery. 2,7 The conventional method used to harvest the OA prone position with the head turned 45° toward the contra- entails removing it from its distal to its proximal segment, lateral side by using the 3-pin head clamp (Mizuho Amer- using layer-by-layer muscle dissection through the nuchal ica), and the neck was slightly flexed, allowing flattening and suboccipital musculature.6,8 The transitional segment of the suboccipital musculature. This position allowed of the OA, which runs between the superior edge of the exposure of the mastoid process as well as easy access splenius capitis muscle and the superior nuchal line, has to the inion and the spinous process of C-2. The classic been deemed the most challenging segment of the OA to inverted hockey-stick incision was performed, connecting harvest because it is known to traverse vertically through the spinous process of C-2 to the tip of the mastoid process several layers, including the tendon of the splenius capitis (Fig. 1A). The incision began along the avascular midline muscle and the galea aponeurotica.7 Recently, Fukuda et through the nuchal ligament and was extended superiorly al. proposed 3 variations of the conventional OA harvest- 3 cm above the superior nuchal line. At this point, it was ing technique, all of which directed the dissection toward turned laterally in a horizontal fashion until reaching the the transitional segment of the OA, using 3 different skin asterion region and then continued inferiorly to end 1 cm incisions.7 The authors concluded that the most critical below the mastoid tip. While performing the incision, care step during OA harvest is the dissection of the transitional was taken to keep the cut shallow near the lateral third of segment, and that layer-by-layer dissection of muscles is the superior nuchal line to prevent inadvertent damage to required to safely and effectively harvest this segment. the distal segment of the OA, which runs subcutaneously. Several studies have described the microanatomy of the The subcutaneous segment of the OA was then iden- OA in detail for its use in posterior circulation bypass.2,3 tified, ligated, and transected. In 2 cases, longer grafts However, an efficient, technically less challenging, and were obtained by dissecting distally (superiorly) into the minimally invasive technique for harvesting the OA is scalp as proof of concept, but measurements were taken still lacking. only to the skin incision for statistical consistency. Blunt In this cadaveric surgical simulation study, we propose dissection was then continued in the avascular plane in a novel technique for harvesting the OA via an antero- the midline until the occipital bone was exposed. The nu- grade (from proximal to distal) approach performed using chal muscles were then elevated from medial to lateral a myocutaneous flap. Our objective was to describe the into the flap, keeping the periosteal sheath intact within approach in a stepwise manner and define key landmarks the flap (Fig. 1B). Dissection then continued through the to safely and efficiently harvest the OA by using the pro- suboccipital space, and the C-1 and C-2 vertebrae were posed technique. exposed between the posterior rectus capitis major and minor muscles. The suboccipital muscles were elevated Methods subperiosteally along the inferior nuchal line into a single Craniometric Study myocutaneous flap. As dissection continued inferiorly, the longissimus capitis was freed from its insertion at the Thirty-five dry skulls were assessed bilaterally (n = 70) posterior margin of the mastoid process and the poste- to study the bony landmarks that can be used to identify rior belly of digastric muscle (DM) was detached from its and locate the proximal segment of the OA. The occipi- attachment at the digastric groove (DG), and both were tal groove (OG) was bilaterally identified in each skull. included in the flap. The shape of the OG was classified into canal, groove, or The OA was then identified in the OG, which was found impression. Next, the asterion was located and the origin medial to the DG (Fig. 1C). It was bound medially by the of the occipitomastoid suture (OMS) was identified. The superior oblique muscle (SOM) and laterally by the poste- OMS was then followed inferiorly to assess its relationship rior belly of DM. Once the OA was identified, it was dis- with the course of the OA: medial to the OA, lateral to the sected proximally by using sharp microdissection under OA, or crossing the OA. the microscope, freeing it from its surrounding connective and periosteal tissue. Dissection was then continued dis- Development and Feasibility Assessment of the Proposed tally with 45° Pott scissors, following the course of the OA Technique under the periosteum. The course of the OA was followed Eight adult cadaveric heads (16 specimens) were pre- distally, freeing it circumferentially from its surrounding pared for surgical simulation following our published connective tissue attachments, ending at the superior edge protocol for embalming.5 The conventional OA harvest- of the hockey-stick incision (Fig. 1D and Video 1). ing technique was performed in 2 specimens to assess The lengths and calibers of the suboccipital and occipi- the shortcomings of this technique and to design the new tal segments of the OA were measured. Also, the spatial approach. The proposed novel anterograde (proximal to relationship of the OA with key defined landmarks (OG,

208 J Neurosurg Volume 130 • January 2019

Unauthenticated | Downloaded 10/05/21 10:15 AM UTC A. Benet et al.

FIG. 1. Stepwise depiction of the anterograde inside-out OA harvesting technique. A: The classic hockey-stick incision (dotted line) was performed, connecting the spinous process of C-2 to the mastoid process (solid curved line). The star depicts the position of the inion. B: The musculocutaneous flap was reflected, detaching the nuchal and occipital muscles from their insertions into a single myocutaneous flap. C: The proximal segment of the OA was identified in the OG, which was found medial to the DG. D: The OA was identified in the OG and was freed from the surrounding occipital musculature and connective tissue in an anterograde (proximal-to-distal) manner. C. = capitis; Ext. = external; LCM = longissimus capitis muscle; M. = muscle; Occ. = occipital; Sup. = superior. Figure is available in color online only.

DM, DG, and SOM) was assessed and recorded. All mea- the OA were found to be arising from the muscular side, surements of length were taken with the vessel fully ex- which was located in a deeper plane of dissection. tended and straight. Any caliber measurements were made In our observations of dry skull specimens, we found using the microscope and a caliper tool. that the OA ran in a groove in 71.5% of the specimens, whereas it carved an impression in the rest of the cases (28.5%). In none of the specimens was the OA found to Results run in a bony canal. The OMS was found to be medial The OA was successfully harvested in all 14 specimens to the OA groove or impression in 68.6% of the cases, by using the proposed technique without any damage to whereas in 31.4% it ran centrally through the OA groove the vessel, as evidenced by preserved adventitia of the or impression (Fig. 2). In none of the specimens was the vessel and absence of latex exposure. The mean total har- OMS found lateral to the OA. vested length of the OA was 12.8 ± 1.2 cm, and the mean length of the suboccipital segment was 8.1 ± 0.6 cm. The Illustrative Case mean diameter of the suboccipital segment was 1.3 ± 0.1 A 59-year-old man presented with complaints of gait mm and that of the occipital segment was found to be 1.1 instability and imbalance. On radiological investigation, ± 0.1 mm. The most consistent landmarks for identifying he was found to have a subacute infarct in the cerebellum and localizing the proximal OA were lateral to the SOM, in the PICA territory. Angiography revealed vertebro- medial to the DG, and within the OG. At the superior nu- basilar insufficiency due to intracranial atherosclerosis; chal line, the OA was located at an average of 3.3 cm from in particular, occluded bilateral VAs beyond the PICAs. the midline, whereas at the incision site it was found at The patient had a history of prior small-vessel middle an average of 5.3 cm from the midline. No muscle perfo- cerebral artery strokes on the left side as well as a right- rator vessels were transected during exposure of the OA sided pontine ischemic stroke, leading to left-sided hemi- from the periosteum because all the muscular branches of paresis. The patient opted for surgical management, and

J Neurosurg Volume 130 • January 2019 209

Unauthenticated | Downloaded 10/05/21 10:15 AM UTC A. Benet et al.

FIG. 2. Dry skull images depicting important bony landmarks for the identification of the OA. A: The relationship between DG, OG, and OMS is depicted. The OG was medial to the DG, and the OMS was medial to the OG in the majority (68.6%) of cases. The foramen for the mastoid emissary vein is also shown. B: In 31.4% of the specimens, the OMS was running through the OG. The asterion is shown; the OMS can be followed from the asterion inferiorly to locate the OG, and thus the OA. Figure is available in color online only. an OA–anterior inferior cerebellar artery (AICA) bypass of the recipient vessels and is commonly encountered en was planned. route to the pathological entity being addressed, sparing The patient was placed in the lateral position with the the patient the need for an additional incision for an in- head fixated in the Mayfield head holder. A hockey-stick terposition graft. Since the first description of its use in skin incision was made, and scalp and muscle flaps were 1967, 13 several variations of the harvesting technique for mobilized inferolaterally, taking care to preserve the peri- the OA have been proposed,6–8 but most of them are te- osteum. The OA was harvested via an inside-out trans- dious and time consuming due to the tortuous course of periosteal approach, pulling the OA through the perios- the OA through the nuchal and suboccipital musculature. teum and harvesting it all the way to the edge of the flap. The complex techniques needed to harvest the OA might A right extended retrosigmoid craniotomy was then per- be an important limiting factor for the widespread use of formed after drilling the sigmoid sinus. After dural open- the OA as a donor vessel for extracranial-intracranial by- ing, the cerebellopontine angle was entered and the AICA pass procedures. was identified. The OA was transected and fish-mouthed, The technique described in this study proposes har- and brought down into the surgical field. An OA-AICA vesting the OA from the myocutaneous flap. In contrast end-to-side bypass was then performed. Intraoperative with the classic technique, our proposed technique in- indocyanine green video angiography and postoperative volves identifying the proximal segment of the OA in the angiography confirmed patency of the bypass (Video 2). OG, and harvesting it circumferentially in a proximal to VIDEO 2. Clip showing a surgical case demonstrating the transperi- distal fashion, from deep to superficial, dissecting away osteal inside-out OA harvesting technique for an OA-AICA bypass. Copyright Arnau Benet. Published with permission. Click here to from critical neurovascular structures. In its proximal view. segment, the OA has been described to course in a single layer of connective tissue termed the styloid diaphragm,7 The patient recovered well from surgery and had no neu- making it easily identifiable. This method requires only rological deficits at discharge. superficial dissection because the OA is found directly under the periosteum throughout its course. This results in Discussion minimal manipulation of the nuchal and occipital muscu- We propose an anterograde transperiosteal technique lature, and does not require layer-by-layer muscle dissec- for harvesting the OA that does not require layer-by-layer tion, decreasing the risk of muscular atrophy. In addition, muscle dissection and is based on reliable, easily identifi- because this technique requires minimal blind dissection, able surgical landmarks. This anterograde, transperiosteal it is relatively safer and has less chance of damaging the dissection is potentially less invasive, more efficient, and graft. Thereby, this technique offers an easier, faster, and technically less challenging than the existing techniques. potentially safer OA harvesting technique than those that The most consistent landmarks for identifying and local- have been described previously (Table 1). On the other izing the proximal OA in this study were the SOM, the hand, inadvertent damage to the proximal portion of the DG, and the OG, all of which are easily identifiable dur- OA can potentially be more hazardous than that to the dis- ing a conventional hockey-stick incision for a far-lateral tal OA. Thus, we propose the use of anatomical landmarks approach. including the OG, DG, and OMS to aid in the localization The OA has been widely used for revascularization of the OA. In addition, adjunct use of modalities like neu- procedures of the posterior circulation because it has been ronavigation and intraoperative Doppler can further aid demonstrated to offer a suitable length, caliber match, and in the identification of the proximal portion of the OA, if flow rate for vessels in the posterior circulation, in particu- required. lar the PICA.1,3,4,9,11 In addition, it is located in the vicinity The OA has traditionally been divided into 3 segments

210 J Neurosurg Volume 130 • January 2019

Unauthenticated | Downloaded 10/05/21 10:15 AM UTC A. Benet et al.

TABLE 1. Comparison of the salient features of the classic versus the anterograde “inside-out” OA harvesting techniques Classic Approach Anterograde “Inside-Out” Approach Distal-to-proximal “retrograde” OA dissection Proximal-to-distal “anterograde” OA dissection Requires layer-by-layer muscle dissection; is tedious & risky Requires transperiosteal dissection of OA from a single myocutaneous flap; is easier & faster Exposure of the OA is difficult, specifically the transitional segment, which Reliable landmark-based approach; it allows easy localization of the OA traverses several muscle layers Requires blind dissection of the OA as it traverses multiple muscle layers Requires minimal blind dissection because the OA is under direct view throughout the dissection Dissection has to be continued until adequate caliber is achieved proximally Dissection starts from the proximal segment, which has larger caliber; the graft length w/ largest caliber can be tailored early Muscular perforators may be transected during layer-by-layer muscle dis- No muscular perforators transected; reduction of blood loss section based on its relationship with key structures throughout its found to be > 1 mm.2 We measured the length of the ar- course, from its origin from the external carotid artery to tery after mobilizing the suboccipital segment and found a its distal bifurcation into terminal segments (i.e., digastric, mean length of 8.1 cm. This is critical because the average 2 suboccipital, and occipital segments). Other authors have length reported for OA bypass to the V3 segment of the proposed a classification based on the course of the OA VA is 4.0 cm, the V4 segment is 5.0 cm, the caudal loop through the musculature that it traverses (i.e., intramus- of PICA is 5.8 cm, and the AICA is approximately 5.9 cular, transitional, and subcutaneous segments).7 Based cm.3 Thus, dissecting the suboccipital segment provides on this classification, the transitional segment of the OA adequate length for bypass, sometimes allowing for omis- has been suggested as the safer segment to identify the sion of the more time-consuming occipital segment of the OA for its harvesting. However, to identify this transitional OA. This reduces the time needed for OA dissection and segment, a clear identification of multiple muscular land- time of the operation, correlating with reduced mortality marks is required after a layer-by-layer dissection, becom- and lower hospital costs.12 ing challenging when classic incisions are used, and still The OMS is a key landmark for localizing the proximal carrying a considerable risk of damaging the OA. To over- OA. The OMS was consistently found within (31.4%) or come the common limitations of all the previous methods, medial (68.6%) to the OG. Because the OMS is exposed we designed a technique that allows shallow, superficial during the approach, it can be used as a landmark to lo- dissection of the OA. We propose the use of a classic hock- cate the proximal OA. The OMS can be followed inferi- ey-stick incision with a myocutaneous flap, and then har- orly from the asterion to the OA groove. Inadvertent dam- vest of the OA from the undersurface of the flap through age to the proximal OA may be avoided by identifying the the periosteum. This brings the OA superficial in the dis- OMS and dissecting medially to it, because the proximal section after the flap is reflected, similar to the depth of the OA will be lateral to it in most cases. None of the 70 OGs superficial temporal artery on an anterolateral approach. studied had a bony canal, although this configuration may This technique preserves the integrity of the nuchal and be found.2 Therefore, the impact of a canalicular configu- occipital muscles. By being relatively atraumatic to this ration of the OG in identifying the OA using our proposed musculature, this technique has a potentially less chance technique remains unknown and of minimal statistical of postoperative pain and muscular atrophy. It also has less relevance. risk of inadvertent transection of the OA and less blood The scope of this study was to design and assess the loss, because no muscular branches of the OA are encoun- feasibility of this new technique by using surgical simula- tered and/or transected during the harvest. tion in cadavers. This study lays the foundation for an ev- In our study, we ended the dissection of the subcuta- idence-based, ethical use of the anterograde OA harvest- neous segment at the incision site. However, in cases in ing technique in patients. The illustrative case included which longer grafts are required, this segment can be validates the feasibility of this technique in clinical set- further dissected distally superiorly into the scalp if the tings. However, further assessment of clinical outcomes, incision is extended superiorly. Another advantage of us- such as bleeding, vasospasm, postoperative pain, muscular ing the proximal-to-distal technique is that because the atrophy, and length of hospital stay, may be studied with diameter of the proximal segment is adequate, the dissec- large-scale clinical application of this technique. More- tion can be terminated once adequate length is achieved, over, evaluation of metrics such as operating room time, which is not the case when harvesting distal to proximal, patency of bypass, intraoperative complications, and tech- where dissection has to be continued until adequate diam- nical ease is required to establish the superiority of this eter is found proximally. technique over the existing ones for harvesting the OA. In our study, the OA diameter was 1.3 mm at the suboccipital and 1.1 mm in the occipital segments. These findings are comparable to previous studies in which the Conclusions average diameter of the suboccipital segment was report- Harvesting the OA for bypass can be a challenging and ed to be 1.4 mm and that of the occipital segment was time-consuming task due to its tortuous course and the

J Neurosurg Volume 130 • January 2019 211

Unauthenticated | Downloaded 10/05/21 10:15 AM UTC A. Benet et al. various layers of muscle it runs within. We propose a novel occipital muscles under a reverse C-shaped skin incision. Br inside-out technique that uses the classic hockey-stick in- J Neurosurg 29:401–405, 2015 cision to harvest the OA in an anterograde manner (from 9. Khodadad G: Occipital artery-posterior inferior cerebellar proximal to distal) from the myocutaneous flap through artery anastomosis. Surg Neurol 5:225–227, 1976 10. Khodadad G: Short- and long-term results of microvascular the periosteum. By obviating the need for layer-by-layer anastomosis in the vertebrobasilar system, a critical analysis. dissection, this technique is potentially relatively safer, Neurol Res 3:33–65, 1981 more efficient, and less technically challenging than the 11. Lewis SB, Chang DJ, Peace DA, Lafrentz PJ, Day AL: Distal conventional techniques. The OMS, OG, DM, DG, and posterior inferior cerebellar artery aneurysms: clinical fea- SOM are key landmarks for identifying and localizing the tures and management. J Neurosurg 97:756–766, 2002 OA by using this technique. 12. Monk TG, Saini V, Weldon BC, Sigl JC: Anesthetic management and one-year mortality after noncardiac surgery. Anesth Analg 100:4–10, 2005 Acknowledgments 13. Sundt TM Jr, Piepgras DG: Occipital to posterior inferior We express our gratitude to the body donors and their families, cerebellar artery bypass surgery. J Neurosurg 48:916–928, who, through their altruism, contributed to making this project 1978 possible.

References Disclosures 1. Abla AA, McDougall CM, Breshears JD, Lawton MT: Intra- The authors report no conflict of interest concerning the materi- cranial-to-intracranial bypass for posterior inferior cerebellar als or methods used in this study or the findings specified in this artery aneurysms: options, technical challenges, and results paper. in 35 patients. J Neurosurg 124:1275–1286, 2016 2. Alvernia JE, Fraser K, Lanzino G: The occipital artery: a mi- Author Contributions croanatomical study. Neurosurgery 58 (1 Suppl):ONS114– ONS122, 2006 Conception and design: Benet, Ding. Acquisition of data: Tabani, 3. Ateş O, Ahmed AS, Niemann D, Başkaya MK: The occipital Ding, Nisson. Analysis and interpretation of data: Tabani, Ding, artery for posterior circulation bypass: microsurgical anato- Nisson, Yousef. Drafting the article: Benet, Tabani, Rodriguez my. Neurosurg Focus 24(2):E9, 2008 Rubio, Nisson. Critically revising the article: Benet, Tabani, Burk 4. Ausman JI, Diaz FG, Vacca DF, Sadasivan B: Superficial hardt, Rodriguez Rubio, Tayebi Meybodi, Gandhi, Yousef, Law- temporal and occipital artery bypass pedicles to superior, ton. Reviewed submitted version of manuscript: Benet, Tabani, anterior inferior, and posterior inferior cerebellar arteries for Ding, Burkhardt, Tayebi Meybodi, Nisson, Kola, Gandhi, Yousef, vertebrobasilar insufficiency. J Neurosurg 72:554–558, 1990 Lawton. Approved the final version of the manuscript on behalf 5. Benet A, Rincon-Torroella J, Lawton MT, González Sánchez of all authors: Benet. Statistical analysis: Tabani. Administrative/ JJ: Novel embalming solution for neurosurgical simulation in technical/material support: Benet, Tabani, Kola. Study supervi- cadavers. J Neurosurg 120:1229–1237, 2014 sion: Benet. 6. Crowley RW, Medel R, Dumont AS: Operative nuances of an occipital artery to posterior inferior cerebellar artery bypass. Supplemental Information Neurosurg Focus 26(5):E19, 2009 Videos 7. Fukuda H, Evins AI, Burrell JC, Stieg PE, Bernardo A: A safe and effective technique for harvesting the occipital Video 1. https://vimeo.com/231692849. artery for posterior fossa bypass surgery: a cadaveric study. Video 2. https://vimeo.com/231692953. World Neurosurg 82:e459–e465, 2014 8. Katsuno M, Tanikawa R, Uemori G, Kawasaki K, Izumi N, Correspondence Hashimoto M: Occipital artery-to-posterior inferior cerebel- Arnau Benet: University of California, San Francisco, CA. lar artery anastomosis with multiple-layer dissection of sub- [email protected].

212 J Neurosurg Volume 130 • January 2019

Unauthenticated | Downloaded 10/05/21 10:15 AM UTC