Download Anatomical study of endoscope assisted far lateral keyhole

Anatomical study of endoscope-assisted far lateral keyhole approach to the ventral
craniocervical region with neuronavigational guidance
GUAN Min-wu, CHEN Ling, FENG Dong-xia, LI Ming-chu, FU Paul, CHEN Li-hua, ZHANG Qiu-hang, SAMII
Amir4, SAMII Madjid, KONG Feng and ZHANG Zhi-ping
Department of Neurosurgery, Xuanwu Hospital, Capital Medical University; China International Neuroscience
Institute, Beijing 100053, China (Guan MW, Chen L, Chen LH, Zhang QH, Kong F and Zhang ZP)
Department of Neurosurgery, University of Arkansas for Medical Science, little rock, Arkansas, 72205, USA (Feng
DX)
Department of Neurosurgery, Wayne State University School of Medicine, Detroit, Michigan, 48236, USA (Paul
FU)
Department of Neurosurgery, International Neuroscience Institute, Hannover, 30625, Germany (Samii A and Samii
M)
Correspondence to: Dr. CHEN Ling, Department of Neurosurgery, Xuanwu Hospital, Capital Medical University,
Beijing 100053, China (Tel: 86-10-83198852. Fax: 86-10-83154745. E-mail: [email protected])
Conflict of interest: none.
Abstract
Background: Image-guided neurosurgery, endoscopic-assisted neurosurgery and the keyhole approach are three
important parts of minimally invasive neurosurgery and have played a significant role in treating skull base lesions.
We investigated the potential usefulness of coupling of the endoscope with the far lateral keyhole approach and
image guidance at the ventral craniocervical junction in a cadaver model.
Methods: We simulated far lateral keyhole approach bilaterally in five cadaveric head specimens (10 cranial
hemispheres). Computed tomography-based image guidance was used for intraoperative navigation and for
quantitative measurements. Skull base structures were observed using both an operating microscope and a rigid
endoscope. The jugular tubercle and one-third of the occipital condyle were then drilled, and all specimens were
observed under the microscope again. We measured and compared the exposure of the petroclivus area provided by
the endoscope and by the operating microscope. Statistical analysis was performed by analysis of variance followed
by the Student-Newman-Keuls test.
Results: With endoscope assistance and image guidance, it was possible to observe the deep ventral craniocervical
junction structures through three nerve gaps (among facial-acoustical nerves and the lower cranial nerves) and
structures normally obstructed by the jugular tubercle and occipital condyle in the far lateral keyhole approach. The
surgical area exposed in the petroclival region was significantly improved using the 0° endoscope (1147.80 mm2)
compared with the operating microscope (756.28 ± 50.73 mm2). The far lateral retrocondylar keyhole approach,
using both 0° and 30° endoscopes, provided an exposure area (1147.80±159.57 mm2 and 1409.94±155.18 mm2,
respectively) greater than that of the far lateral transcondylar transtubercular keyhole approach (1066.26±165.06
mm2) (p < 0.05).
Conclusion: With the aid of the endoscope and image guidance, it is possible to approach the ventral
craniocervical junction with the far lateral keyhole approach. The use of an angled-lens endoscope can significantly
improve the exposure of the petroclival region without drilling the jugular tubercle and occipital condyle.
Key words: far lateral approach; neuroendoscope; neuronavigation; keyhole approach; ventral craniocervical
junction
【摘要】
背景：影像引导神经外科、内镜辅助神经外科和锁孔入路是微侵袭神经外科三个重要组成部分，在治疗颅
底疾病中发挥了重要作用。我们通过尸头解剖研究发现导航引导下内镜辅助远外侧锁孔入路能够良好的显
露腹侧颅颈交界区的结构。
方法：对 5 例（10 侧）尸头标本模拟远外侧锁孔入路，术中用神经导航实时定位，并做定量研究，分别用
显微镜和内镜观察颅底结构。随后磨除后内侧 1/3 枕髁和颈静脉结节，再次用显微镜观察，最后测量和比较
内镜和显微镜下各标本岩斜区的显露面积。实验数据采用 Student-Newman-Keuls 检验和方差分析进行统计
学研究。
结果：借助神经导航和角度内镜，通过面听神经、后组颅神经间的三个间隙能够近距离观察颅底结构，还
能观察被颈静脉结节和枕髁遮挡的结构。0 度内镜辅助远外侧髁后锁孔入路时岩斜区的显露面积为 756.28 ±
50.73 mm2，明显大于单纯手术显微镜下的显露面积 756.28 mm2，0 度和 30 度内镜辅助下的显露面积分别为
1147.80±159.57 mm2 and 1409.94±155.18 mm2，优于远外侧经髁经结节锁孔入路(1066.26±165.06 mm2) (p <
0.05)。
结论：借助内镜和神经导航，远外侧髁后锁孔入路能够良好的显露腹侧颅颈交界区，角度内镜能够明显扩
大岩斜区的显露范围，避免磨除颈静脉结节和部分枕髁。
关键词：神经导航，神经内镜，远外侧入路，腹侧颅颈交界区
Introduction
The concept of minimally invasive neurosurgery was first proposed in 1992 by Bauer and Helliwing, and
promoted reduced brain tissue injury, increased operational accuracy and functional recovery. It focused the
sophisticated technologies available in neurosurgery and would become the direction of future development in the
field
[1,2]
. Image-guided neurosurgery and endoscopic-assisted neurosurgery are two important parts of minimally
invasive neurosurgery and play a significant role in treating skull base lesions [3]. Neuronavigation can aid in the
reconstruction of intracranial lesions, blood vessels and nerve fiber tracts using image fusion techniques. Real-time
positional information allows for precise localization and resection of lesions and dramatically improves the
reliability of these operations[4-8]. The use of neuroendoscopes combined with a keyhole approach can avoid the
drawbacks of conventional craniotomy and does not increase corresponding surgical complications.
Neuroendoscopy combined with neuronavigation has the advantages of both systems and allows for
anatomical study of the sellar region and prepontine cistern, treatment of intraventricular lesions and tumor biopsies
in the deep brain parenchyma[4, 8-11]. However, until now there have been few anatomical studies of neuroendoscopy
combined with neuronavigation in the posterior fossa, especially in the ventral craniocervical junction[12,13]. This
region is characterized by a dense distribution of nerves and blood vessels, a rugged skull base and a narrow
operating space, therefore requiring particularly accurate exposure and surgical technique. Because the far lateral
approach remains the most common surgical route to the ventral craniocervical junction and because of the
established advantages of the endoscope (multi-angle panoramic view through narrow surgical corridors), we
investigated the feasibility of a minimally invasive endoscopic-guided far lateral keyhole approach to the ventral
craniocervical junction in a cadaver model. The goal of this research was to help clarify the benefits of
neuronavigation and endoscopy and provide anatomical information for future clinical application.
METHODS
Material: Five formalin-fixed adult cadaveric heads (10 hemispheres) were used. Internal carotid and vertebral
arteries were injected with red latex. The internal venous system was injected with blue latex. The following
instruments were used: computed tomography (GE, Waukesha, WI, USA); operating microscope (Zeiss,
Oberkochen, Germany); endoscopic system (Storz, Tuttlingen, Germany); neuronavigation (BrainLAB, Feldkirchen,
Germany); power drill (Xishan, Chongqing, China); digital camera (Canon, Beijing, China); neurosurgical and
microsurgical instruments and a stainless steel head holder. All procedures were performed at the Skull Base
Training Center of Capital Medical University in Beijing, China.
Neuronavigation data collection: All cadaveric head specimens underwent continuous computed
tomography scans. The imaging data (Dicom pattern) were stored in CD format and imported into the BrainLAB
neuronavigation workstation to perform three-dimensional reconstructions.
Far lateral retrocondylar keyhole approach simulation: Each cadaveric head specimen was fixed
in a head holder, and the mastoid process was positioned and maintained at the highest point. All specimens were
matched and registered with the neuronavigation system through infrared facial recognition technology. A 8-cm
longitudinal S-shaped incision was made approximately 3 cm dorsal to the mastoid process (incision was 3 cm
above the mastoid process plane and 5 cm below). The sternocleidomastoid, splenius capitis and semispinalis
muscles were divided. Muscles were cut at their upper attachment and reflected down to expose the suboccipital
triangle. The superior oblique and rectus capitis posterior major muscles were cut at their attachments to expose the
vertebral artery and venous plexus. A 3 × 3 cm small bone window was created behind the occipital condyle, with
the lateral boundary at the sigmoid sinus, the lower boundary at the occipital condyle and the internal-lower
boundary up to the foramen magnum (Figure 1). The dura was then cut and skull base structures were observed
with the microscope. Next, the microscope was replaced with the registered 0° and 30° endoscopes to inspect the
skull base structures at short distance with neuronavigational guidance. The jugular tubercle and one-third of the
posteromedial occipital condyle were then drilled with the navigational instructions. The anatomical structures of
this region were observed using the microscope for future anatomical reference.
Measuring the exposure area: The exposure area of the petroclival region was calculated by defining
eight points to create a polygon. The trigeminal notch, the internal auditory canal, the jugular foramen and the
hypoglossal canal (if two hypoglossal canals were present, the inferior one was chosen) were selected as the
reference points of the extreme periphery of the exposure area and labeled A, B, C and D respectively. The
reference points for the medial border of the exposure area were defined as the most medial points of the clivus
observable with the microscope/endoscope through the upper gap between the facial-acoustic nerve and the
glossopharyngeal nerve, the middle gap between the vagus nerve and the cranial root of the accessory nerve, and
the inferior gap between the caudal hypoglossal nerves. These points were labeled with 0.05 ml of various coloring
agents. Therefore, the areas of polygons ABCDEFGH, ABCDIJK, ABCDLMN and ABCDEFOP represented the
exposure range of the microscope, 0° endoscope, 30° endoscope and microscope after the occipital condyle and
jugular tubercle were drilled, respectively (Figure 2A-B). The distance between adjacent reference points was
measured using the neuronavigation system. With endoscopic guidance, a probe was placed on a reference point,
which was defined as the zero point. When the probe touched a second point, the machine calculated the distance
between adjacent reference points automatically (Figure 2C). The lengths of the imaginary triangle were recorded,
and the area of each triangle/polygon was calculated.
Statistical Analysis: Data are reported as means ± SEM. Analysis of variance (ANOVA) was used, followed
by the Student-Newman-Keuls test. For all tests, P < 0.05 was considered statistically significant. For statistical
analysis we used the commercially available SPSS software (Version 16.0; SPSS Inc., Chicago, IL, USA).
RESULTS
Exposure with single far lateral retrocondylar keyhole approach
All superficial structures were observable, including the ipsilateral lower cranial nerves, facial acoustic nerve,
posterior inferior cerebellar artery (PICA), anterior inferior cerebellar artery (AICA), jugular foramen, inner
auditory meatus, flocculus cerebelli and choroid plexus of the fourth ventricle. This exposure range was the same
as that obtained with the routine far lateral retrocondylar approach.
Exposure with 0° endoscope-assisted far lateral retrocondylar keyhole approach with
neuronavigational guidance
The superficial structures observable with the endoscope were the same as with the microscope, but the 0°
endoscope had more advantages in exploring deep structures through three nerve gaps. The superior gap was the
space between the facial nerve and the glossopharyngeal nerve. Through this gap, we were able to observe the
origin of the trigeminal nerve, the medial wall of the trigeminal notch, the dorsum sellae, the terminal portion of the
basilar artery, the superior cerebellar artery, the oculomotor nerve, the caudal portion of the third ventricle (from an
outside-inferior to inside-superior direction), the start site of the abducent nerve, Dorello’s canal, the entire length
of the AICA, the onset of the basilar artery and the lateral wall of the confluence of vertebral and basilar arteries
(from an outside to inside direction horizontally) (Figure 3A–C). The medial gap was the space between the vagus
nerve and the cranial root of the accessory nerve. Through this gap we were able to observe the terminal portion of
the ipsilateral vertebral artery, the confluence of the vertebral artery and basilar artery with small branching vessels
and the anterior spinal artery (Figure 3D–F). The inferior gap was the space below the hypoglossal nerve through
which we could observe the hypoglossal foramen, the origin of the hypoglossal nerve and the midline of the clivus.
It was also possible to observe the contralateral vertebral artery and hypoglossal canal in four of the cadaver cranial
hemisphere specimens (Figure 3G and H).
Exposure with 30° endoscope-assisted far lateral retrocondylar keyhole approach
with neuronavigational guidance
The 30° endoscope had the same viewing path as the 0° endoscope, but was able to obtain a larger exposure
range. The 30° endoscope allowed for better observation of the ventrolateral side of the brainstem, especially the
angles between the brain stem and four cranial nerves (the facial-acoustic nerve, the trigeminal nerves and the
vagus nerve), the hypoglossal foramen, the inferior edge of the vertebrobasilar confluence and the anterior spinal
artery with some of its branches. We were also able to observe the contralateral partial clivus, the contralateral
entrance of Dorello's canal with the abducent nerve, the contralateral hypoglossal foramen and vertebral artery and
the anterior margin of the foramen magnum (Figure 3H).
Exposure with far lateral transcondylar transtubercular keyhole approach with
microscope
After drilling one-third of the occipital condyle, we could observe the ipsilateral hypoglossal canal and the
midline clivus. In six of the hemisphere specimens, the contralateral vertebral artery was viewable. Drilling of the
jugular tubercle allowed observation of the vertebrobasilar confluence, a portion of the basilar artery, the origin of
the AICA and the origin of the abducent nerve.
Exposure of the petroclival region using different methods
With the aid of the neuronavigation system, the exposed area of the petroclival region could be measured and
calculated. The mean exposure range of the far lateral retrocondylar keyhole approach was 756.28 ± 50.73 mm2.
The mean exposure range was 1147.80±159.57 mm2 using the 0° endoscope-assisted technique and 1409.94±155.18
mm2 using the 30° endoscope-assisted technique. The far lateral transcondylar transtubercular approach had a mean
exposure range of 1066.26±165.06 mm2. Using ANOVA and the Student-Newman-Keuls test, the petroclival
exposure range was clearly different with the various viewing methods (F = 298.370, P < 0.001), with the 30°
endoscope-assisted far lateral retrocondylar approach giving the largest exposure range (P < 0.05).
DISCUSSION
The far lateral approach is the most favored approach for the surgical treatment of intradural pathologies at the
ventral craniocervical junction. Drilling of the occipital condyle and jugular tubercle is useful in exposing midline
lesions in the inferior clivus region[14-16], but there are disadvantages to extensive bone drilling. Vishteh reported
that drilling the occipital condyle can cause atlanto-occipital joint instability and significantly increased in range of
motion sometimes requiring atlanto-occipital joint arthrodesis[17]. Spektor et al reported that drilling of the jugular
tubercle and occipital condyle increased the risk of injuring the internal jugular vein, jugular bulb, vertebral artery
and cranial nerves, and may lead to cerebrospinal fluid leakage from tearing of the dura[16]. These processes
increase operating time and may result in significant postoperative complications like hemorrhage and infection.
Recently, endoscopes have gained popularity and acceptance as part of the neurosurgical armamentarium.
Endoscopes also have been used to access the cranial base. Hayashi first reported an endoscope-assisted far lateral
transcondylar transtubercular approach to observe the ventral craniocervical junction of cadaver head specimens in
2002[18]. This was the first time that endoscopy and the far lateral approach were united, and thus began the era of
inspecting the ventral craniocervical junction with the endoscope. Still, the endoscope use only occurred after
drilling the occipital condyle and the jugular tubercle. Research had not demonstrated whether the endoscope could
observe areas blocked by bony barriers. In this study, we investigated the usefulness of the endoscopic technique in
the far lateral keyhole approach. We also quantified and compared the surgical exposure obtained with the
endoscope and the microscope in a cadaver model. Our data suggest that using the endoscope in the far lateral
keyhole approach significantly improves surgical exposure without compromising surgical freedom. Our data also
indicate that angled-lens endoscopes significantly improve the exposure of the petroclival region without drilling
the jugular tubercle or one-third of the occipital condyle.
Using the endoscope, we inspected the deep structures through three nerve gaps. Because of the wide upper
gap, the large ventral space of the pons and a clear path (except for the AICA), the endoscope could easily pass
through the nerve gap and observe deep structures in two directions. Although the bony eminence of the posterior
border of the internal acoustic meatus was large in some specimens, it did not hamper the observations or the
endoscope’s operation. When the endoscope passed through the narrower middle gap in an outside inferior to
inside upper direction, we had to gently move the vagus nerve and the cranial roots of the accessory nerve. It was
important to maintain the stability of the endoscope and instruments to avoid damage to the nerves. If the loop of
the PICA obstructed the corridor, we gently pulled it outwards using the probe. It was important to choose an
avascular and nerve-sparse region caudal to the hypoglossal nerve as an entrance through the inferior gap to
prevent damage to the cranial roots of the accessory nerve and the hypoglossal nerve, and to avoid injuring the
perforating branches of the vertebral artery and PICA supporting the medulla oblongata. The denticulate ligament
could be transected to view the anterior margin of the foramen magnum. Appropriate movement of the vertebral
artery can increase the gap and was beneficial for operation.
Endoscopic-assisted far lateral keyhole approach with navigation guidance combined the advantages of
neuronavigation, neuroendoscopy and the keyhole approach. With this approach, a practitioner is able to make full
use of the location information received from the neuronavigation and the image data obtained by the endoscope.
Neuronavigation can match the patient’s preoperative or intraoperative imaging data with practical surgical
anatomy to locate lesions
[19]
. This location information was especially helpful in real-time positioning of the
endoscope and in choosing the best operating route. Neuronavigation combined with neuroendoscopy overcomes
the problem of depth deficiency from an endoscopic image. It also overcomes some of the disadvantages of the
angled endoscope like the inability to see straight ahead, resulting in possible injury to adjacent structures[20,21]. In
addition, neuronavigation can be used before surgery to determine the direction and extent of jugular tubercle and
occipital condyle drilling needed for far lateral approach. Real-time monitoring of the drilling size allows the
surgeon to avoid injuring the hypoglossal canal. In this study, neuronavigation was also used to measure the
distance between adjacent skull base structures, which saved time and made results more accurate.
The endoscopic-assisted far lateral keyhole approach exposed the deep operative field well with an only 3 × 3
cm bone window. This diminished exposure of the surrounding brain tissue, shortened the neck incision and
avoided excessive injury to the muscles and ligaments. The endoscope that we used was only 4 mm wide and could
pass through neurovascular gaps ventrolateral to the brain stem to observe anatomical structures. Depending on the
magnification of the endoscope, we were able to observe even the small vessels and nerves. The “fish eye” effect of
the endoscope and a special light source can provide a full viewing range, thereby avoiding the difficulty of
identifying anatomical structures and resolving the difficulty of poor lighting in the deep operative field [22,23]. Using
the endoscope allowed for observation of the entire clivus with no blind areas, especially of the structures blocked
by the jugular tubercle in the keyhole approach.
Until now, there have been no reports of endoscopic-assisted far lateral keyhole approach with
neuronavigation guidance. This technique has the potential for clinical application, especially in dealing with
lesions extending to the contralateral clivus. Before surgery, the surgeon can use neuronavigation to reconstruct
intracranial structures and create a precise three-dimensional anatomical concept of the surgical target area and the
trajectory leading to it. After craniotomy, the surgeon conducts a preliminary survey of the surrounding structures
using the endoscope, especially the tiny vessels and compressed or displaced nerves on the surface of the tumor.
Once tumor resection is completed under the microscope, the endoscope and neuronavigation can again be used to
search for and dissect tumor remnants in the microscope’s blind spots, thereby avoiding drilling of other bony
structures or requiring multiple surgical approaches. This protocol reduces the chance of residual tumor tissue left
behind and therefore decreases the probability of tumor recurrence. For smaller hidden or deep lesions, using
neuronavigation can aid in difficult intraoperative orientation and identify the positional relationship between the
endoscope and the tumor, which is convenient for designing a manipulation trajectory.
Although anatomical research has demonstrated that endoscopic-assisted far lateral keyhole approach with
neuronavigation guidance dramatically increases the safety and accuracy of surgery, this technique has some
shortcomings in clinical application[21,24,25]. First, the endoscope provides vision only at its tip. The inability of the
endoscope to look sideways or backwards when positioned in the operative field allows for the possibility of injury
to adjacent structures by the shaft of the endoscope. Further complicating the issue is that the nerve gaps between
the vagus nerve, the cranial roots of the accessory nerve and the hypoglossal nerve are narrow. Incorrect
manipulation of the endoscope increases the risks of injuring these nerves and perforating blood vessels supplying
the medulla oblongata. Third, all structures in the endoscopic image are slightly deformed. Most neurosurgeons are
not familiar with the endoscopic image and operative technique. A steep learning curve is still required to master
manipulation and treatment using the endoscope. Fourth, there is still a lack of microinstruments specifically
designed for use with the 30° endoscope in the subarachnoid space. In sort, the improved ability in visualization is
not yet matched by an improved ability to perform the needed procedures.
CONCLUSION
With the aid of the endoscope and image guidance, it is possible to approach the ventral craniocervical
junction with the far lateral keyhole approach. In addition, an angled-lens endoscope can significantly improve the
exposure of the petroclival region without drilling the jugular tubercle or one-third of the occipital condyle.
Furthermore, the use of image guidance systems during endoscopic procedures gives the surgeon a constant
orientation in the surgical field, thus increasing the accuracy and the safety of the approach.
REFERENCES
[1] Hellwig D, Bauer BL. Minimally invasive neurosurgery by means of ultrathin endoscopes. Acta Neurochir
Suppl (Wien). 1992; 54: 63-68. PMID: 1595411
[2] Ishii M, Gallia GL. Application of technology for minimally invasive neurosurgery. Neurosurg Clin N Am.
2010; 21(4): 585-594. PMID: 20947028
[3] Mehta RP, Cueva RA, Brown JD, Fliss DM, Gil Z, Kassam AB et al. What's new in skull base medicine and
surgery? Skull Base Committee Report. Otolaryngol Head Neck Surg. 2006; 135(4): 620-630. PMID: 17011428
[4] Alberti O, Riegel T, Hellwig D, Bertalanffy H. Frameless navigation and endoscopy. J Neurosurg, 2001; 95(3):
541-543. PMID: 11565885
[5] Kurtsoy A, Menku A, Tucer B, Oktem IS, Akdemir H. Neuronavigation in Skull Base Tumors. Minim Invasive
Neurosurg. 2005; 48: 7-12. PMID: 15747210
[6] Rohde V, Spangenberg P, Mayfrank L, Reinges M, Gilsbach JM, Coenen VA. Advanced Neuronavigation in
Skull Base Tumors and Vascular Lesions. Minim Invasive Neurosurg. 2005; 48: 13-18. PMID: 15747211
[7] Wu JS, Lu JF, Gong X, Mao Y, Zhou LF. Neuronavigation surgery in China: reality and prospects. Chin Med J
(Engl). 2012; 125(24): 4497-503. PMID: 23253726
[8] Nonaka Y, Oi S, Samii A, Paterno V, Feigl GC, Lüdemann W et al. Neuronavigational neuroendoscopic surgery.
Frameless free-hand maneuvering of a handy rigid-rod neuroendoscope on visualized three-dimensional
computerized image guidance: trajectory to the prepontine cistern in cadaver study. Childs Nerv Syst. 2006; 22(1):
18-27. PMID: 16091964
[9] Di Rocco F, Oi S, Samii A, Paternó V, Feigl GC, Lüdemann W et al. Neuronavigational endoscopic endonasal
sellar and parasellar surgery using a 2-mm-diameter lens rigid-rod endoscope: a cadaver study. Neurosurg. 2007;
60(4 Suppl 2): 394-400. PMID: 17415180
[10] Lu XJ, Chen KL, Wang Q, Ji WY, Li B, Sun JY et al. Endoscopic expand transnasal approach to the
suprasellar region: anatomical study and clinical considerations. Chin Med J (Engl). 2009; 122(20): 2444-8. PMID:
20079157
[11] Gumprecht H, Trost HA, Lumenta CB. Neuroendoscopy combined with frameless neuronavigation. Br J
Neurosurg. 2000; 14(2): 129-131. PMID: 10889885
[12] Wu CY, Lan Q. Quantification of the presigmoid transpetrosal keyhole approach to petroclival region. Chin
Med J (Engl). 2008;121(8):740-4. PMID: 18701030.
[13] Xia Y, Li XP, Han DM, Zheng J, Long HS, Shi JF. Anatomic structural study of cerebellopontine angle via
endoscope. Chin Med J (Engl). 2007; 120(20):1836-9. PMID: 18028782
[14] Karam YR, Menezes AH, Traynelis VC. Posterolateral Approaches to the Craniovertebral Junction. Neurosurg.
2010; 66: A135-140. PMID: 20173516
[15] Wanebo JE, Chicoine MR. Quantitative Analysis of the Transcondylar Approach to the Foramen Magnum.
Neurosurg, 2001; 49(4): 934-945. PMID: 11564256
[16] Spektor S, Anderson GJ, McMenomey SO, Horgan MA, Kellogg JX, Delashaw JB Jr. Quantitative description
of the far-lateral transcondylar transtubercular approach to the foramen magnum and clivus. J Neurosurg. 2000; 92:
824–831. PMID: 10794297
[17] Vishteh AG, Crawford NR, Melton MS, Spetzler RF, Sonntag VK, Dickman CA. Stability of the
craniovertibral junction after unilateral occipitial condyle resection: a biomechanical study. J Neurosurg. 1999; 90(1
Suppl): 91-98. PMID: 10413132
[18] Hayashi N, Cohen AR. Endoscope-assisted far-lateral transcondylar approach to the skull base, Minim Invas
Neurosurg, 2002; 45(3): 132-135. PMID: 12353157
[19] Willems PW, van der Sprenkel JW, Tulleken CA, Viergever MA, Taphoorn MJ. Neuronavigation and surgery
of intracerebral tumors. J Neurol. 2006; 253: 1123-1136. PMID: 16988793
[20] Di X. Multiple Brain Tumor Nodule Resections under Direct Visualization of a Neuronavigated Endoscope,
Minim Invasive Neurosurg, 2007; 50(4): 227-232. PMID: 17948182
[21] Mayberg MR, LaPresto E, Cunningham EJ. Image-guided endoscopy: description of technique and potential
applications. Neurosurg Focus. 2005; 19(1): E10. PMID: 16078813
[22] Garrett M, Consiglieri G, Nakaji P. Transcranial minimally invasive neurosurgery for tumors. Neurosurg Clin
N Am. 2010; 21(4): 595-605. PMID: 20947029
[23] Perneczky A, Fries G. Endoscope assisted brain surgery: part-1 evolution, basic concept, and current technique.
Neurosurgery, 1998; 42(2): 219 -224. PMID: 9482171
[24] Cappabianca P, Cavallo LM, Esposito F, de Divitiis E, Tschabitscher M. Endoscopic examination of the
cerebellar pontine angle. Clinical Neurology and Neurosurgery. 2002; 104(4): 387-391. PMID: 12140112
[25] Liu YG, Wu CY, Liu M. Neuroendoscopic anatomy and surgery of the cerebellopontine angle. Journal of
Clinical Neuroscience. 2005; 12(3): 256–260. PMID: 15851077
Figure 1. Simulated far lateral retro-condylar small bone window craniotomy. A: An 8-cm long longitudinal
S-shaped incision was made 3 cm behind the mastoid process. B: The skin and subcutaneous tissues were incised to
expose the superficial muscles. C. The attachments of the sternocleidomastoideus muscle and splenius capitis
muscle were cut and the muscles reflected to expose the deep muscles and occipital artery. D: The semispinalis
capitis muscle was manipulated to expose the suboccipital triangle. E: The venous plexus and surrounding fascia
were dissected to expose the vertebral artery. F: 3 × 3 cm far lateral small bone window craniotomy. 1, mastoid
process; 2, sternocleidomastoideus muscle; 3, splenius capitis muscle; 4, longissimus capitis muscle; 5,
semispinalis capitis muscle; 6, occipital artery; 7, superior oblique muscle; 8, inferior oblique muscle; 9, rectus
capitis posterior major muscle; 10, vertebral artery; 11, posterior arch of atlas; 12, C1 nerve root; 13, occipital bone;
14, occipital condyle; 15, edge of foramen magnum; 16, sigmoid sinus.
Figure 2. A-B, Exposure of the petroclival region with different methods. The areas of polygons ABCDEFGH
(blue), ABCDIJK (orange), ABCDLMN (yellow) and ABCDEFOP (black) represent the exposure ranges of the
microscope, 0° endoscope, 30° endoscope and microscope after occipital condyle and jugular tubercle drilling,
respectively.C. Neuronavigation and endoscopy can provide real-time positional information during the operation.
Figure 3
A. Inspect the structures through the gap between the facial-acoustic nerve complex and the glossopharyngeal
nerve under the endoscope.
B. Inspect the structures in the superior clivus region.
C. Elevate the ipsilateral vertebral artery and observe the ventral surface of the vertebrobasilar confluence.
D. Observe the structures through the gap between the vagus nerve and cranial roots of the accessory nerve.
E. Observe the region in the ventral side of glossopharyngeal and vagus nerves and on the medial side of jugular
tubercle.
F. Observe the pons and pontomedullary sulcus from the dorsal aspect of the vertebral artery and basilar artery.
G. Observe the origin of the hypoglossal nerve from the ventrolateral sulcus of the medulla oblongata.
H. Observe the triangle clivus region composed of the bilateral vertebral artery and vertebrobasilar confluence.
AICA, anterior inferior cerebellar artery; AspA, anterior spinal artery; BA, basilar artery; Bri. V, bridging vein; CN
III, oculomotor nerve; CN V, trigeminal nerve; CN VI, abducent nerve; CN XII, hypoglossal nerve; L-VA, left
vertebral artery; R-VA, right vertebral artery; SCA, superior cerebellar artery.