- Email: [email protected]
Anatomy of the petrous apex as related to the endoscopic endonasal approach
Journal of Clinical Neuroscience 19 (2012) 1695–1698
Contents lists available at SciVerse ScienceDirect
Journal of Clinical Neuroscience journal homepage: www.elsevier.com/locate/jocn
Neuroanatomical study
Anatomy of the petrous apex as related to the endoscopic endonasal approach Kong Feng a, Zhang Qiuhang a,⇑, Zhang Wei b, Liu Jiabin d, Wei Yukui a, Chen Ge a, Zhang Zhiping c a
Skull Base Surgery Center, Capital Medical University Xuanwu Hospital, 45 Changchunjie Street, Xicheng District, Beijing 100053, China Department of Otorhinolaryngology Head and Neck Surgery, Capital Medical University Xuanwu Hospital, Beijing, China c Department of Neurosurgery, Capital Medical University Xuanwu Hospital, Beijing, China d Department of Radiology, Capital Medical University Xuanwu Hospital, Beijing, China b
a r t i c l e
i n f o
Article history: Received 14 June 2011 Accepted 10 September 2011
Keywords: Anatomy Endoscope Internal carotid artery Meckel’s cave Skull base
a b s t r a c t The endoscopic endonasal approach (EEA) has been reported to be an efficient approach for treating lesions of the petrous apex. However, there have been only limited anatomic studies for the EEA. Furthermore, most of the relevant distances for EEA cannot be measured easily on a cadaveric skull. Two fresh adult cadaver heads and five formalin-fixed adult cadaver heads were dissected using the EEA to identify groups of landmarks for safe guidance during this approach. The distances between these landmarks were then measured by CT angiography by using three-dimensional software. The EEA to the petrous apex can be divided into five phases. In each phase, a group of landmarks, rather than a single landmark, can be identified easily for guiding the next phase of the approach. There was no significant difference between males and females in any of the distances reported in the present study. The EEA can be performed to manage a petrous apex lesion more safely by referring to multiple landmarks and the distances between them. Ó 2012 Published by Elsevier Ltd.
1. Introduction The petrous apex is a challenge for the surgeon due to its deep location and the proximity of critical structures, such as the internal carotid artery (ICA), cavernous sinus and Meckel’s cave. It is considered one of the most difficult areas to approach surgically. In recent years, the endoscopic endonasal approach (EEA) has been reported to be an efficient approach for treating petrous apex lesions, such as cholesterol granuloma, cholesteatoma and schwannoma.1–4 As opposed to the three-dimensional (3D) images provided by a microscope, the image provided by an endoscope is two dimensional (2D). Thus, during the endoscopic procedure, surgeons must refer to multiple sequential landmarks by moving the endoscope to gain 3D information. Some cadaveric studies have described the view of the petrous apex region under the endoscope,5,6 and a few landmarks, such as the vidian canal, have been selected as guides for the EEA. However, the previously described landmarks are so limited that if some are destroyed by the lesion before surgery, or are removed for the purpose of extending the surgical corridor, there will not be enough to guide the procedure. In addition, the relevant distances for the EEA should be measured from the ventral side of the skull base. However, it is usually difficult to measure these distances on a dry skull or cadaveric head. Although Chatrath et al. have attempted to ⇑ Corresponding author: Tel.: +86 13701267977; fax: +86 010 83198836. E-mail address: [email protected] (Z. Qiuhang). 0967-5868/$ - see front matter Ó 2012 Published by Elsevier Ltd. http://dx.doi.org/10.1016/j.jocn.2011.09.042
identify some surgical landmarks, and have measured some distances between the landmarks using a navigation system to gain access to the petrous apex,7 detailed anatomic information related to the petrous apex for the EEA requires further study. To our knowledge, the use of CT angiography (CTA) in conjunction with the EEA has not been described previously. The goals of the present study were to identify additional landmarks (and define groups of consistent anatomical landmarks rather than a single landmark) in each phase of the EEA to the petrous apex, and to quantify the distances between the landmarks by CTA using 3D software. These data would provide the surgeon with practical information so that the petrous apex can be approached safely and the vital structures around it can be protected during endoscopic endonasal surgery. 2. Materials and methods 2.1. Cadaveric anatomic study Four male and three three female adult cadaver heads were dissected. All seven specimens (two fresh and five formalin-fixed) had well-pneumatised sphenoid sinuses (SS). The arterial and venous systems were injected with red and blue latex, respectively. Rigid endoscopes (Karl Storz and Co., Tuttlingen, Germany), 4 mm in diameter, 18 cm in length, with 0° and 30° lenses, were used as appropriate for different steps of the dissection. The endoscope was connected to a xenon light source (Cold Light Fountain
1696
K. Feng et al. / Journal of Clinical Neuroscience 19 (2012) 1695–1698
Xenon 300; Karl Storz) through a fiber-optic cable and to a digital camera fitted with three charge-couple-device sensors (Image 1; Karl Storze). A digital videorecorder system was used. The endoscope was advanced along one side of the nasal cavity. The unilateral inferior turbinate and middle turbinate were removed to enlarge the surgical corridor. Anatomic landmarks within the nasal cavity and the nasopharyngeal cavity were identified and captured by the video camera. After the uncinate process was removed, the ethmoid sinus was opened until the lamina papyracea was exposed. A posterior nasal septectomy and bilateral sphenoidotomies were performed. Anatomic landmarks within the SS were identified and recorded. Then, the bone of the lateral posterior wall of the SS was drilled away. The ostium of the maxillary sinus was extended posterior to the posterior wall of the sinus. The sphenopalatine foramen and the sphenopalatine artery were identified and transected. The contents of the pterygopalatine fossa were then displaced laterally, and the anterior opening of the vidian canal was exposed. A high-speed drill was then used to reduce the root of the pterygoid process inferior, medial to the vidian artery and superior to the Eustachian tube. The dura mater of the middle cranial fossa was exposed to the level of the foramen ovale and foramen rotundum. 2.2. Measurement using CTA From February 2007 to September 2008, CTA images of 100 adult patients (69 male, 31 female; 20–79 years of age [mean age, 55.29 years]) (200 sides) with suspected cardiac vascular disease were selected and reconstructed into 3D images of the head, using reconstruction software. Patients with head or cranial disorders were excluded. CTA was performed using a GE Light speed V CT-XT scanner (GE Healthcare, Chalfont St. Giles, UK). Contrast (60 mL UltravistÒ [iopromide], Bayer HealthCare Pharmaceuticals, Wayne, NJ, USA) and saline (40 mL) were injected intravenously (i.v.) at 4 mL/s. The images were acquired at 120 kV and 500 mA with a slice thickness of 0.625 mm and a spacing of 0.969 mm. Image data were transferred to the workstation with 3D visualisation software (GE Advantage Workstation 4.3) to be reconstructed by the volumerendering method. The distances between the anatomic landmarks were measured on both sides. In an effort to eliminate observer error and bias, all measurements were taken twice. The mean of the two values was used to calculate the final results. 2.3. Statistical analysis The mean distances and standard deviation (SD) were calculated for each measurement by the Statistical Package for the Social Sciences for Windows (version 17.0) (SPSS, Chicago, IL, USA) and are presented as the mean ± SD.
3. Results In the EEA to the petrous apex, the access stage could be divided into five phases. Each phase was defined by the cavities or spaces along the surgical corridor: (i) nasal, (ii) nasopharyngeal, (iii) sinus, (iv) pterygopalatine fossa, and (v) petrous apex (Fig. 1). In each phase, there was a group of landmarks, rather than a single landmark, that could be identified easily; these groups were then used for guiding the next phase of the approach. The landmarks in each group are listed in Table 1. In the nasopharyngeal phase the torus tubalis (TT) was a consistent landmark that provided the orientation required to define the positions of many surgically important structures in the subse-
Fig. 1. The access stage of the endoscopic endonasal approach to the petrous apex could be divided into five phases: (i) nasal, (ii) nasopharyngeal, (iii) sinus, (iv) pterygopalatine fossa, and (v) petrous apex phase. (1 = nasal cavity, 2 = nasopharyngeal cavity, 3 = sphenoidal sinus, 4 = maxillary sinus, 5 = pterygopalatine fossa, 6 = petrous apex) (Copyright: Capital Medical University, Xuan Wu Hospital, Beijing, China).
quent phases of the EEA to the petrous apex (Supplementary Fig. 1). The distances from the TT to the landmarks of the sinus phase, the pterygopalatine fossa phase, and the petrous apex phase, as measured by CTA (Supplementary Fig. 1) are shown in Table 2. The distance from the TT to the optic nerve protuberance in the SS, the medial border of the petroclival ICA, and to the lateral border of the petroclival ICA are shown in Table 2. In the sinus phase, the SS was the central cavity of the EEA. This group of landmarks was composed of bony depressions in the posterior wall and the lateral wall of the SS, which included the bulge of the sellar turcica in the SS, called the sellar floor; the bony depression at the upper clivus region, called the clival indentation; the carotid artery protuberance, the optic nerve protuberance and the carotico–optic recess (Supplementary Fig. 2). The carotico–optic recess is a depressed area at the base of the optic strut where it joins the sphenoid body. The distances from the carotico–optic recess to other landmarks (Supplementary Fig. 2) are listed in Table 2. In the maxillary sinus, the infraorbital nerve, the posterior wall of the maxillary sinus and the sphenopalatine foramen comprised the group of landmarks with which to reach the pterygopalatine fossa. In the pterygopalatine fossa phase, the landmarks were the vidian canal, the root of the pterygoid process, the foramen rotundum and the maxillary division of the trigeminal nerve. In the petrous apex phase, the course of the ICA was an important landmark. The other landmarks around the ICA were the foramen ovale, the mandibular division of the trigeminal nerve and the fibrocartilaginous tissue of the foramen lacerum (Supplementary Fig. 3). The average distance from the foramen rotundum to other landmarks is listed in Table 2. There was no significant difference between males and females in any distance reported.
1697
K. Feng et al. / Journal of Clinical Neuroscience 19 (2012) 1695–1698 Table 1 Groups of landmarks used for guidance in the endoscopic endonasal approach (EEA) to the petrous apex
Table 2 Distance between the landmarks related to the endoscopic endonasal approach to the petrous apex Distance (mm) Left
TT–OP TT–FR TT–FO COR–FR COR–FO TT–mICA TT–lICA COR–mICA COR–lICA FO–aICA FO–lICA FR–lICA
Right
Total
Minimum
Maximum
Mean ± SD
Minimum
Maximum
Mean ± SD
Mean ± SD
25.70 16.70 13.10 12.00 24.30 4.90 6.30 19.90 19.10 2.90 4.40 17.50
41.9 32.40 27.10 26.30 32.90 19.90 25.40 32.90 37.00 9.00 15.20 26.50
34.76 ± 3.14 24.44 ± 3.26 18.73 ± 3.06 20.80 ± 2.75 28.14 ± 1.76 9.68 ± 2.66 11.96 ± 2.79 25.74 ± 1.92 29.64 ± 2.69 4.72 ± 1.13 8.98 ± 2.21 21.50 ± 1.94
26.20 17.80 13.00 14.80 24.50 4.6 6.90 19.40 17.3 2.40 3.90 15.60
41.70 31.90 26.60 28.30 31.90 18.60 23.40 31.70 37.40 9.60 13.00 26.20
34.81 ± 3.12 24.54 ± 3.18 18.83 ± 3.12 21.07 ± 2.74 28.05 ± 1.60 9.97 ± 2.68 12.68 ± 2.91 25.60 ± 2.02 30.18 ± 2.63 4.55 ± 1.30 8.56 ± 2.12 21.58 ± 2.00
34.79 ± 3.12 24.49 ± 3.21 18.78 ± 3.08 20.94 ± 2.74 28.09 ± 1.68 9.82 ± 2.67 12.32 ± 2.86 25.67 ± 1.97 29.91 ± 2.67 4.63 ± 1.22 8.77 ± 2.17 21.54 ± 1.97
aICA = anterior border of horizontal segment of petrous ICA, COR = carotico–opic recess, FO = foramen ovale, FR = foramen rotundum, ICA = internal carotid artery, lICA = lateral border of the lacerum segment of ICA, mICA = medial border of the lacerum segment of the ICA, OP = intracranial opening of the optic canal, SD = standard deviation, TT = torus tubalis.
4. Discussion Since the early 2000s, EEA has gained popularity because of advances in illumination techniques and optics. Based on the successful management of sellar region lesions, endoscopic endonasal surgery has been performed for lesions of the petrous apex.1–4 Understanding the relevant anatomy is a prerequisite for the EEA to the petrous apex. The key to the EEA to the petrous apex is to locate the ICA and the second and the third divisions of the trigeminal nerve. Other approaches to the petrous apex during endoscopic endonasal surgery have been described. Zanation et al. divided it into three approaches: a medial approach, a medial approach with ICA lateralisation, and a transpterygoid infrapetrous approach.3 In Theodore’s categorisation scheme, the transmaxillary transpterygoidal approach with opening of the ethmoid and SS can be used to reach the petrous apex.8 According to the results of the present study, the EEA to the petrous apex can be understood as an approach from a certain cavity to another cavity or space (that is, from the nasal cavity to the nasopharyngeal cavity, to the SS or from the nasal cavity to the maxillary sinus, to the pterygopalatine fossa and then to the target of the petrous apex; Fig. 1). A thorough understanding of the relationships and distances between the landmarks used by the endoscopic surgeon to navigate the petrous apex is critical. For surgical orientation, the ideal anatomic reference point must be consistent, and both easy to find and use. In contrast to a previous report,2 the present study emphasises that the
nasopharyngeal cavity is the starting point for all EEA (Fig. 1, Table 1). The Eustachian tube is located inferior to the ICA. Its medial end lies in the nasopharynx, where it protrudes at the TT. The TT is easily identifiable and is a consistent anatomic landmark as it is seldom affected by pathology or previous surgery. During the EEA, by referring to the distances between the TT and other landmarks in different phases, such as the optic nerve protuberance in the SS, the rotundum foramen in the pterygopalatine fossa, and the ovale foramen around the petrous apex, surgeons can find other landmarks easily (Table 2, Supplementary Fig. 1). The SS is a window to the petrous apex that has been described in previous studies.3 In a well-pneumatised SS, the sellar floor, the carotico–optic recesses and the clival indentation are viewed as bony landmarks with the endoscope. Cavallo et al. revealed that the carotico–optic recess is a consistent bony depression bordering the optic nerve and the ICA.9 It has a triangular configuration. Even when the bony structure is damaged, the carotico–optic recess still serves as a consistent anatomic landmark. The average distances from the optic nerve protuberance in the SS to the foramen rotundum and to the foramen ovale could be used to guide surgery from the SS to the following phase (Table 2, Supplementary Fig. 2). The most critical structure at the petrous apex is the ICA. Special attention should be given to the course of the ICA to the petrous apex. An endoscopic endonasal study has shown that the ICA can be divided into several segments.9 The cavernous segment of the ICA is easily recognised during endoscopic endonasal surgery as the carotid bony protuberance lateral to the sellar floor or clival
1698
K. Feng et al. / Journal of Clinical Neuroscience 19 (2012) 1695–1698
indentation.10 Nevertheless, the lacerum segment of the ICA is difficult to identify due to its deep location. Except for the vidian nerve canal, which has been emphasised for tracing this segment in previous reports,6,11 we have identified multiple landmarks in this study, such as the TT, the carotico–optic recess, the foramen rotundum and the foramen ovale, to localise the lacerum segment of the ICA. The average distances from the medial or the lateral border of the lacerum segment of the ICA to the TT in the nasopharyngeal phase and to the carotico–optic recess in the sinus phase are highly variable. These data not only indicate that the course of the ICA as it passes the foramen lacerum is extremely variable but also emphasise the potential advantages of preoperative 3D CT scans. According to the data presented here and the location of the TT and the carotico–optic recess, the medial and lateral border of the lacerum segment of the ICA will be easy to identify even in the nasopharyngeal or SS phase. The distance from the foramen ovale to the petrous carotid artery was consistent with the results of previous reports.12 The average distance from the foramen rotundum or the foramen ovale to the lateral border of the lacerum segment of the ICA can guide surgeons to locate the ICA after exposure of the foramen rotundum and the foramen ovale. Although the use of skulls or cadavers allows direct measurement for obtaining anatomic data, many studies have been unable to reliably calculate average distances between the endoscopic endonasal landmarks because of the limited number of specimens. CT scans have been used recently to measure the distances between landmarks in the skull base,12 but the 2D CT images do not precisely reflect the complicated 3D structure of this region. To overcome the disadvantages of conventional methods, some studies have applied 3D-reconstructed CT scans of the skull base.13 In the CTA images, both the bony structures and the vessels can be located with accuracy.14 However, the utility of a CTA scan combined with 3D-reconstruction software as a tool for anatomic study of the EEA has not been fully appreciated. To our knowledge, this study is the first to measure the distances between the relevant landmarks using 3D-CTA images, with the intent of accessing the petrous apex through the EEA. Based on the precise distances between the structures obtained from the 3D-CTA obtained preoperatively, surgeons can identify the landmarks step-by-step using a 4-mm external diameter suction tip to measure distances. 5. Conclusion The endoscopic endonasal approach to the petrous apex can be divided into several phases according to the cavities utilised. A
cadaveric anatomic study demonstrates that multiple landmarks, rather than a single landmark, can be identified in each phase. Using these landmarks to perform the endoscopic endonasal approach from one phase to the next, surgeons can reduce the invasiveness of the approach and minimise the associated morbidity. The 3D-CTA image can provide specific anatomic information on both the bony and vascular landmarks and the spatial relationships prior to surgery. Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.jocn.2011.09.042. References 1. Kassam AB, Gardner P, Snyderman C, et al. Expanded endonasal approach: fully endoscopic, completely transnasal approach to the middle third of the clivus, petrous bone, middle cranial fossa, and infratemporal fossa. Neurosurg Focus 2005;19:E6. 2. Georgalas C, Kania R, Guichard JP, et al. Endoscopic transsphenoidal surgery for cholesterol granulomas involving the petrous apex. Clin Otolaryngol 2008;33: 38–42. 3. Zanation AM, Snyderman CH, Carrau RL, et al. Endoscopic endonasal surgery for petrous apex lesions. Laryngoscope 2009;119:19–25. 4. Hofstetter CP, Singh A, Anand VK, et al. The endoscopic, endonasal, transmaxillary transpterygoid approach to the pterygopalatine fossa, infratemporal fossa, petrous apex, and the Meckel cave. J Neurosurg 2010; 113:967–74. 5. Goravalingappa R, Han JC, Mangiardi J, et al. Endoscopic/microscopic approach to sphenopetroclival complex: an anatomical study. Skull Base Surg 1999;9: 33–9. 6. Osawa S, Rhoton Jr AL, Seker A, et al. Microsurgical and endoscopic anatomy of the vidian canal. Neurosurgery 2009;64:385–411 [discussion 411–2]. 7. Chatrath P, Nouraei SA, De Cordova J, et al. Endonasal endoscopic approach to the petrous apex: an image-guided quantitative anatomical study. Clin Otolaryngol 2007;32:255–60. 8. Schwartz TH, Fraser JF, Brown S, et al. Endoscopic cranial base surgery: classification of operative approaches. Neurosurgery 2008;62:911–1005. 9. Cavallo LM, Cappabianca P, Galzio R, et al. Endoscopic transnasal approach to the cavernous sinus versus transcranial route: anatomic study. Neurosurgery 2005;56:379–89. 10. Herzallah IR, Casiano RR. Endoscopic endonasal study of the internal carotid artery course and variations. Am J Rhinol 2007;21:262–70. 11. Vescan AD, Snyderman CH, Carrau RL, et al. Vidian canal: analysis and relationship to the internal carotid artery. Laryngoscope 2007;117:1338–42. 12. Maina R, Ducati A, Lanzino G. The middle cranial fossa: morphometric study and surgical considerations. Skull Base 2007;17:395–403. 13. Villavicencio AT, Leveque JC, Bulsara KR, et al. Three-dimensional computed tomographic cranial base measurements for improvement of surgical approaches to the petrous carotid artery and apex regions. Neurosurgery 2001;49:342–53. 14. Sennaroglu L, Slattery 3rd WH. Petrous anatomy for middle fossa approach. Laryngoscope 2003;113:332–42.
Contents lists available at SciVerse ScienceDirect
Journal of Clinical Neuroscience journal homepage: www.elsevier.com/locate/jocn
Neuroanatomical study
Anatomy of the petrous apex as related to the endoscopic endonasal approach Kong Feng a, Zhang Qiuhang a,⇑, Zhang Wei b, Liu Jiabin d, Wei Yukui a, Chen Ge a, Zhang Zhiping c a
Skull Base Surgery Center, Capital Medical University Xuanwu Hospital, 45 Changchunjie Street, Xicheng District, Beijing 100053, China Department of Otorhinolaryngology Head and Neck Surgery, Capital Medical University Xuanwu Hospital, Beijing, China c Department of Neurosurgery, Capital Medical University Xuanwu Hospital, Beijing, China d Department of Radiology, Capital Medical University Xuanwu Hospital, Beijing, China b
a r t i c l e
i n f o
Article history: Received 14 June 2011 Accepted 10 September 2011
Keywords: Anatomy Endoscope Internal carotid artery Meckel’s cave Skull base
a b s t r a c t The endoscopic endonasal approach (EEA) has been reported to be an efficient approach for treating lesions of the petrous apex. However, there have been only limited anatomic studies for the EEA. Furthermore, most of the relevant distances for EEA cannot be measured easily on a cadaveric skull. Two fresh adult cadaver heads and five formalin-fixed adult cadaver heads were dissected using the EEA to identify groups of landmarks for safe guidance during this approach. The distances between these landmarks were then measured by CT angiography by using three-dimensional software. The EEA to the petrous apex can be divided into five phases. In each phase, a group of landmarks, rather than a single landmark, can be identified easily for guiding the next phase of the approach. There was no significant difference between males and females in any of the distances reported in the present study. The EEA can be performed to manage a petrous apex lesion more safely by referring to multiple landmarks and the distances between them. Ó 2012 Published by Elsevier Ltd.
1. Introduction The petrous apex is a challenge for the surgeon due to its deep location and the proximity of critical structures, such as the internal carotid artery (ICA), cavernous sinus and Meckel’s cave. It is considered one of the most difficult areas to approach surgically. In recent years, the endoscopic endonasal approach (EEA) has been reported to be an efficient approach for treating petrous apex lesions, such as cholesterol granuloma, cholesteatoma and schwannoma.1–4 As opposed to the three-dimensional (3D) images provided by a microscope, the image provided by an endoscope is two dimensional (2D). Thus, during the endoscopic procedure, surgeons must refer to multiple sequential landmarks by moving the endoscope to gain 3D information. Some cadaveric studies have described the view of the petrous apex region under the endoscope,5,6 and a few landmarks, such as the vidian canal, have been selected as guides for the EEA. However, the previously described landmarks are so limited that if some are destroyed by the lesion before surgery, or are removed for the purpose of extending the surgical corridor, there will not be enough to guide the procedure. In addition, the relevant distances for the EEA should be measured from the ventral side of the skull base. However, it is usually difficult to measure these distances on a dry skull or cadaveric head. Although Chatrath et al. have attempted to ⇑ Corresponding author: Tel.: +86 13701267977; fax: +86 010 83198836. E-mail address: [email protected] (Z. Qiuhang). 0967-5868/$ - see front matter Ó 2012 Published by Elsevier Ltd. http://dx.doi.org/10.1016/j.jocn.2011.09.042
identify some surgical landmarks, and have measured some distances between the landmarks using a navigation system to gain access to the petrous apex,7 detailed anatomic information related to the petrous apex for the EEA requires further study. To our knowledge, the use of CT angiography (CTA) in conjunction with the EEA has not been described previously. The goals of the present study were to identify additional landmarks (and define groups of consistent anatomical landmarks rather than a single landmark) in each phase of the EEA to the petrous apex, and to quantify the distances between the landmarks by CTA using 3D software. These data would provide the surgeon with practical information so that the petrous apex can be approached safely and the vital structures around it can be protected during endoscopic endonasal surgery. 2. Materials and methods 2.1. Cadaveric anatomic study Four male and three three female adult cadaver heads were dissected. All seven specimens (two fresh and five formalin-fixed) had well-pneumatised sphenoid sinuses (SS). The arterial and venous systems were injected with red and blue latex, respectively. Rigid endoscopes (Karl Storz and Co., Tuttlingen, Germany), 4 mm in diameter, 18 cm in length, with 0° and 30° lenses, were used as appropriate for different steps of the dissection. The endoscope was connected to a xenon light source (Cold Light Fountain
1696
K. Feng et al. / Journal of Clinical Neuroscience 19 (2012) 1695–1698
Xenon 300; Karl Storz) through a fiber-optic cable and to a digital camera fitted with three charge-couple-device sensors (Image 1; Karl Storze). A digital videorecorder system was used. The endoscope was advanced along one side of the nasal cavity. The unilateral inferior turbinate and middle turbinate were removed to enlarge the surgical corridor. Anatomic landmarks within the nasal cavity and the nasopharyngeal cavity were identified and captured by the video camera. After the uncinate process was removed, the ethmoid sinus was opened until the lamina papyracea was exposed. A posterior nasal septectomy and bilateral sphenoidotomies were performed. Anatomic landmarks within the SS were identified and recorded. Then, the bone of the lateral posterior wall of the SS was drilled away. The ostium of the maxillary sinus was extended posterior to the posterior wall of the sinus. The sphenopalatine foramen and the sphenopalatine artery were identified and transected. The contents of the pterygopalatine fossa were then displaced laterally, and the anterior opening of the vidian canal was exposed. A high-speed drill was then used to reduce the root of the pterygoid process inferior, medial to the vidian artery and superior to the Eustachian tube. The dura mater of the middle cranial fossa was exposed to the level of the foramen ovale and foramen rotundum. 2.2. Measurement using CTA From February 2007 to September 2008, CTA images of 100 adult patients (69 male, 31 female; 20–79 years of age [mean age, 55.29 years]) (200 sides) with suspected cardiac vascular disease were selected and reconstructed into 3D images of the head, using reconstruction software. Patients with head or cranial disorders were excluded. CTA was performed using a GE Light speed V CT-XT scanner (GE Healthcare, Chalfont St. Giles, UK). Contrast (60 mL UltravistÒ [iopromide], Bayer HealthCare Pharmaceuticals, Wayne, NJ, USA) and saline (40 mL) were injected intravenously (i.v.) at 4 mL/s. The images were acquired at 120 kV and 500 mA with a slice thickness of 0.625 mm and a spacing of 0.969 mm. Image data were transferred to the workstation with 3D visualisation software (GE Advantage Workstation 4.3) to be reconstructed by the volumerendering method. The distances between the anatomic landmarks were measured on both sides. In an effort to eliminate observer error and bias, all measurements were taken twice. The mean of the two values was used to calculate the final results. 2.3. Statistical analysis The mean distances and standard deviation (SD) were calculated for each measurement by the Statistical Package for the Social Sciences for Windows (version 17.0) (SPSS, Chicago, IL, USA) and are presented as the mean ± SD.
3. Results In the EEA to the petrous apex, the access stage could be divided into five phases. Each phase was defined by the cavities or spaces along the surgical corridor: (i) nasal, (ii) nasopharyngeal, (iii) sinus, (iv) pterygopalatine fossa, and (v) petrous apex (Fig. 1). In each phase, there was a group of landmarks, rather than a single landmark, that could be identified easily; these groups were then used for guiding the next phase of the approach. The landmarks in each group are listed in Table 1. In the nasopharyngeal phase the torus tubalis (TT) was a consistent landmark that provided the orientation required to define the positions of many surgically important structures in the subse-
Fig. 1. The access stage of the endoscopic endonasal approach to the petrous apex could be divided into five phases: (i) nasal, (ii) nasopharyngeal, (iii) sinus, (iv) pterygopalatine fossa, and (v) petrous apex phase. (1 = nasal cavity, 2 = nasopharyngeal cavity, 3 = sphenoidal sinus, 4 = maxillary sinus, 5 = pterygopalatine fossa, 6 = petrous apex) (Copyright: Capital Medical University, Xuan Wu Hospital, Beijing, China).
quent phases of the EEA to the petrous apex (Supplementary Fig. 1). The distances from the TT to the landmarks of the sinus phase, the pterygopalatine fossa phase, and the petrous apex phase, as measured by CTA (Supplementary Fig. 1) are shown in Table 2. The distance from the TT to the optic nerve protuberance in the SS, the medial border of the petroclival ICA, and to the lateral border of the petroclival ICA are shown in Table 2. In the sinus phase, the SS was the central cavity of the EEA. This group of landmarks was composed of bony depressions in the posterior wall and the lateral wall of the SS, which included the bulge of the sellar turcica in the SS, called the sellar floor; the bony depression at the upper clivus region, called the clival indentation; the carotid artery protuberance, the optic nerve protuberance and the carotico–optic recess (Supplementary Fig. 2). The carotico–optic recess is a depressed area at the base of the optic strut where it joins the sphenoid body. The distances from the carotico–optic recess to other landmarks (Supplementary Fig. 2) are listed in Table 2. In the maxillary sinus, the infraorbital nerve, the posterior wall of the maxillary sinus and the sphenopalatine foramen comprised the group of landmarks with which to reach the pterygopalatine fossa. In the pterygopalatine fossa phase, the landmarks were the vidian canal, the root of the pterygoid process, the foramen rotundum and the maxillary division of the trigeminal nerve. In the petrous apex phase, the course of the ICA was an important landmark. The other landmarks around the ICA were the foramen ovale, the mandibular division of the trigeminal nerve and the fibrocartilaginous tissue of the foramen lacerum (Supplementary Fig. 3). The average distance from the foramen rotundum to other landmarks is listed in Table 2. There was no significant difference between males and females in any distance reported.
1697
K. Feng et al. / Journal of Clinical Neuroscience 19 (2012) 1695–1698 Table 1 Groups of landmarks used for guidance in the endoscopic endonasal approach (EEA) to the petrous apex
Table 2 Distance between the landmarks related to the endoscopic endonasal approach to the petrous apex Distance (mm) Left
TT–OP TT–FR TT–FO COR–FR COR–FO TT–mICA TT–lICA COR–mICA COR–lICA FO–aICA FO–lICA FR–lICA
Right
Total
Minimum
Maximum
Mean ± SD
Minimum
Maximum
Mean ± SD
Mean ± SD
25.70 16.70 13.10 12.00 24.30 4.90 6.30 19.90 19.10 2.90 4.40 17.50
41.9 32.40 27.10 26.30 32.90 19.90 25.40 32.90 37.00 9.00 15.20 26.50
34.76 ± 3.14 24.44 ± 3.26 18.73 ± 3.06 20.80 ± 2.75 28.14 ± 1.76 9.68 ± 2.66 11.96 ± 2.79 25.74 ± 1.92 29.64 ± 2.69 4.72 ± 1.13 8.98 ± 2.21 21.50 ± 1.94
26.20 17.80 13.00 14.80 24.50 4.6 6.90 19.40 17.3 2.40 3.90 15.60
41.70 31.90 26.60 28.30 31.90 18.60 23.40 31.70 37.40 9.60 13.00 26.20
34.81 ± 3.12 24.54 ± 3.18 18.83 ± 3.12 21.07 ± 2.74 28.05 ± 1.60 9.97 ± 2.68 12.68 ± 2.91 25.60 ± 2.02 30.18 ± 2.63 4.55 ± 1.30 8.56 ± 2.12 21.58 ± 2.00
34.79 ± 3.12 24.49 ± 3.21 18.78 ± 3.08 20.94 ± 2.74 28.09 ± 1.68 9.82 ± 2.67 12.32 ± 2.86 25.67 ± 1.97 29.91 ± 2.67 4.63 ± 1.22 8.77 ± 2.17 21.54 ± 1.97
aICA = anterior border of horizontal segment of petrous ICA, COR = carotico–opic recess, FO = foramen ovale, FR = foramen rotundum, ICA = internal carotid artery, lICA = lateral border of the lacerum segment of ICA, mICA = medial border of the lacerum segment of the ICA, OP = intracranial opening of the optic canal, SD = standard deviation, TT = torus tubalis.
4. Discussion Since the early 2000s, EEA has gained popularity because of advances in illumination techniques and optics. Based on the successful management of sellar region lesions, endoscopic endonasal surgery has been performed for lesions of the petrous apex.1–4 Understanding the relevant anatomy is a prerequisite for the EEA to the petrous apex. The key to the EEA to the petrous apex is to locate the ICA and the second and the third divisions of the trigeminal nerve. Other approaches to the petrous apex during endoscopic endonasal surgery have been described. Zanation et al. divided it into three approaches: a medial approach, a medial approach with ICA lateralisation, and a transpterygoid infrapetrous approach.3 In Theodore’s categorisation scheme, the transmaxillary transpterygoidal approach with opening of the ethmoid and SS can be used to reach the petrous apex.8 According to the results of the present study, the EEA to the petrous apex can be understood as an approach from a certain cavity to another cavity or space (that is, from the nasal cavity to the nasopharyngeal cavity, to the SS or from the nasal cavity to the maxillary sinus, to the pterygopalatine fossa and then to the target of the petrous apex; Fig. 1). A thorough understanding of the relationships and distances between the landmarks used by the endoscopic surgeon to navigate the petrous apex is critical. For surgical orientation, the ideal anatomic reference point must be consistent, and both easy to find and use. In contrast to a previous report,2 the present study emphasises that the
nasopharyngeal cavity is the starting point for all EEA (Fig. 1, Table 1). The Eustachian tube is located inferior to the ICA. Its medial end lies in the nasopharynx, where it protrudes at the TT. The TT is easily identifiable and is a consistent anatomic landmark as it is seldom affected by pathology or previous surgery. During the EEA, by referring to the distances between the TT and other landmarks in different phases, such as the optic nerve protuberance in the SS, the rotundum foramen in the pterygopalatine fossa, and the ovale foramen around the petrous apex, surgeons can find other landmarks easily (Table 2, Supplementary Fig. 1). The SS is a window to the petrous apex that has been described in previous studies.3 In a well-pneumatised SS, the sellar floor, the carotico–optic recesses and the clival indentation are viewed as bony landmarks with the endoscope. Cavallo et al. revealed that the carotico–optic recess is a consistent bony depression bordering the optic nerve and the ICA.9 It has a triangular configuration. Even when the bony structure is damaged, the carotico–optic recess still serves as a consistent anatomic landmark. The average distances from the optic nerve protuberance in the SS to the foramen rotundum and to the foramen ovale could be used to guide surgery from the SS to the following phase (Table 2, Supplementary Fig. 2). The most critical structure at the petrous apex is the ICA. Special attention should be given to the course of the ICA to the petrous apex. An endoscopic endonasal study has shown that the ICA can be divided into several segments.9 The cavernous segment of the ICA is easily recognised during endoscopic endonasal surgery as the carotid bony protuberance lateral to the sellar floor or clival
1698
K. Feng et al. / Journal of Clinical Neuroscience 19 (2012) 1695–1698
indentation.10 Nevertheless, the lacerum segment of the ICA is difficult to identify due to its deep location. Except for the vidian nerve canal, which has been emphasised for tracing this segment in previous reports,6,11 we have identified multiple landmarks in this study, such as the TT, the carotico–optic recess, the foramen rotundum and the foramen ovale, to localise the lacerum segment of the ICA. The average distances from the medial or the lateral border of the lacerum segment of the ICA to the TT in the nasopharyngeal phase and to the carotico–optic recess in the sinus phase are highly variable. These data not only indicate that the course of the ICA as it passes the foramen lacerum is extremely variable but also emphasise the potential advantages of preoperative 3D CT scans. According to the data presented here and the location of the TT and the carotico–optic recess, the medial and lateral border of the lacerum segment of the ICA will be easy to identify even in the nasopharyngeal or SS phase. The distance from the foramen ovale to the petrous carotid artery was consistent with the results of previous reports.12 The average distance from the foramen rotundum or the foramen ovale to the lateral border of the lacerum segment of the ICA can guide surgeons to locate the ICA after exposure of the foramen rotundum and the foramen ovale. Although the use of skulls or cadavers allows direct measurement for obtaining anatomic data, many studies have been unable to reliably calculate average distances between the endoscopic endonasal landmarks because of the limited number of specimens. CT scans have been used recently to measure the distances between landmarks in the skull base,12 but the 2D CT images do not precisely reflect the complicated 3D structure of this region. To overcome the disadvantages of conventional methods, some studies have applied 3D-reconstructed CT scans of the skull base.13 In the CTA images, both the bony structures and the vessels can be located with accuracy.14 However, the utility of a CTA scan combined with 3D-reconstruction software as a tool for anatomic study of the EEA has not been fully appreciated. To our knowledge, this study is the first to measure the distances between the relevant landmarks using 3D-CTA images, with the intent of accessing the petrous apex through the EEA. Based on the precise distances between the structures obtained from the 3D-CTA obtained preoperatively, surgeons can identify the landmarks step-by-step using a 4-mm external diameter suction tip to measure distances. 5. Conclusion The endoscopic endonasal approach to the petrous apex can be divided into several phases according to the cavities utilised. A
cadaveric anatomic study demonstrates that multiple landmarks, rather than a single landmark, can be identified in each phase. Using these landmarks to perform the endoscopic endonasal approach from one phase to the next, surgeons can reduce the invasiveness of the approach and minimise the associated morbidity. The 3D-CTA image can provide specific anatomic information on both the bony and vascular landmarks and the spatial relationships prior to surgery. Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.jocn.2011.09.042. References 1. Kassam AB, Gardner P, Snyderman C, et al. Expanded endonasal approach: fully endoscopic, completely transnasal approach to the middle third of the clivus, petrous bone, middle cranial fossa, and infratemporal fossa. Neurosurg Focus 2005;19:E6. 2. Georgalas C, Kania R, Guichard JP, et al. Endoscopic transsphenoidal surgery for cholesterol granulomas involving the petrous apex. Clin Otolaryngol 2008;33: 38–42. 3. Zanation AM, Snyderman CH, Carrau RL, et al. Endoscopic endonasal surgery for petrous apex lesions. Laryngoscope 2009;119:19–25. 4. Hofstetter CP, Singh A, Anand VK, et al. The endoscopic, endonasal, transmaxillary transpterygoid approach to the pterygopalatine fossa, infratemporal fossa, petrous apex, and the Meckel cave. J Neurosurg 2010; 113:967–74. 5. Goravalingappa R, Han JC, Mangiardi J, et al. Endoscopic/microscopic approach to sphenopetroclival complex: an anatomical study. Skull Base Surg 1999;9: 33–9. 6. Osawa S, Rhoton Jr AL, Seker A, et al. Microsurgical and endoscopic anatomy of the vidian canal. Neurosurgery 2009;64:385–411 [discussion 411–2]. 7. Chatrath P, Nouraei SA, De Cordova J, et al. Endonasal endoscopic approach to the petrous apex: an image-guided quantitative anatomical study. Clin Otolaryngol 2007;32:255–60. 8. Schwartz TH, Fraser JF, Brown S, et al. Endoscopic cranial base surgery: classification of operative approaches. Neurosurgery 2008;62:911–1005. 9. Cavallo LM, Cappabianca P, Galzio R, et al. Endoscopic transnasal approach to the cavernous sinus versus transcranial route: anatomic study. Neurosurgery 2005;56:379–89. 10. Herzallah IR, Casiano RR. Endoscopic endonasal study of the internal carotid artery course and variations. Am J Rhinol 2007;21:262–70. 11. Vescan AD, Snyderman CH, Carrau RL, et al. Vidian canal: analysis and relationship to the internal carotid artery. Laryngoscope 2007;117:1338–42. 12. Maina R, Ducati A, Lanzino G. The middle cranial fossa: morphometric study and surgical considerations. Skull Base 2007;17:395–403. 13. Villavicencio AT, Leveque JC, Bulsara KR, et al. Three-dimensional computed tomographic cranial base measurements for improvement of surgical approaches to the petrous carotid artery and apex regions. Neurosurgery 2001;49:342–53. 14. Sennaroglu L, Slattery 3rd WH. Petrous anatomy for middle fossa approach. Laryngoscope 2003;113:332–42.