Abstract
Objective The absence of precise landmarks in the middle fossa floor and frequent anatomical variations make it difficult to localize the internal acoustic canal (IAC) during the middle fossa approach (MFA). We aimed to investigate the reliability and utility of the neuronavigation system (NNS) in the MFA and to delineate specific technical considerations regarding NNS during the approach.
Method One-millimeter-thin section computed tomography scans were performed on five formalin-fixed human cadavers (10 sides). During the MFA, structures, such as the IAC, vestibule and cochlea hidden in the temporal bone were investigated under NNS guidance.
Results All the superficial landmarks, such as the foramen spinosum and ovale were correctly localized by NNS. Deeper landmarks, such as the central part of the IAC lying beneath the surface of the petrous apex could not be localized via NNS. The exact area of bone removal along roof of IAC was determined by using the orientation provided by the probe placed between the basal turn of cochlea and the vestibule. We were able to validate the location of the IAC via a medial to lateral drilling by using the navigation this reference point.
Conclusion The NNS can be used effectively during the MFA, and localizing superficial landmarks on the middle fossa floor with a higher accuracy may prove helpful in identifying the IAC from above. By referring to the cochlea–vestibule junctional area, the exact location of the trace of the IAC can be revealed.
Keywords: middle fossa approach, internal acoustic canal, neuronavigation, petrous
Introduction
The middle fossa approach (MFA) is utilized in surgeries for various pathologies such as resection of small intracanalicular acoustic schwannomas, decompression of the facial nerve in Bell's palsy or traumatic facial paralysis, and vestibular nerve sectioning. 1 2 3 4 Additionally, resection of upper and middle petroclival tumors, clipping of basilar artery aneurysms, and accessing brainstem lesions are feasible through anterior petrosectomy and extended MFA. 5 6 7 8 Complex anatomy of petrous bone including important vital neurovascular structures such as the petrous internal carotid artery (ICA), cochlea, semicircular canals, internal acoustic canal (IAC), and its anatomic variations pose challenges for the MFA. Localizing the IAC is arguably the most important step in these approaches. 9 The absence of precise landmarks due to the target structures being covered in the bone along the middle fossa floor and their frequent anatomical variations make it difficult to localize the IAC from a superolateral surgical perspective. In this study, we aimed to investigate the reliability and utility of the neuronavigation system (NNS) during the MFA. Our additional purpose was to revisit specific technical considerations for localizing the IAC with NNS during the MFA.
Materials and Methods
Five formalin-fixed human cadavers (10 sides) were used in this study. Before to laboratory phase, computed tomography scans with 1-mm thin sections were procured for all heads. These data were uploaded to a frameless NNS (Stealth Station S7, Medtronic Navigation Inc., Louisville, KY, United States) to produce axial, coronal, and sagittal views in bone window and 3D reconstructed view of the head.
Following the exposure of the middle fossa floor, the accuracy of the localization of the IAC through NNS was investigated ( Fig. 1 ). During the initial phase of the approach, the NNS probe was initially positioned on the coronal suture to assess the accuracy of the NNS and furthermore, the identification of the squamous suture, the foramen spinosum and ovale through navigation were checked following the middle fossa exposure ( Fig. 2A,B ). The structures hidden along the middle fossa floor such as the IAC, the superior semicircular canal, vestibule, and the cochlea were also evaluated with guidance of the NNS. The surface markers were detected to localize the IAC, and the orientation obtained by these markers was marked with an ink pen. Then the marked area was drilled with 1- and 3-mm diamond tips from medial to lateral and the accuracy of the localization of the IAC in the drilled trace was investigated.
Fig. 1.

Cadaveric dissection showing right-side middle cranial fossa and internal acoustic canal through the middle fossa approach. (A) The dura has been elevated from the middle fossa floor to initially expose the foramen spinosum and the middle meningeal artery has been transected. Next, the dissection proceeded anteriorly and medially to identify the foramen ovale and mandibular nerve (V3). By elevating the dura in a lateral to medial direction over the V3 toward Meckel's cave, the greater superficial petrosal nerve (GSPN) was localized at the point of its intersection with the lateral edge of the V3. The elevation of middle fossa dura continued medially along the petrous apex until reaching the petrous ridge. Subsequently, the GSPN has been followed posteriorly and laterally, exposing the geniculate ganglion, arcuate eminence, and tegmen tympani. The estimated IAC line ( red dashed line ) and IAC porus ( green arrow ), directed at a 60-degree angle from the long axis of the GSPN toward the petrous ridge, were investigated with NNS. Importantly, the trajectory of the IAC could not be detected by NNS medially along the petrous ridge. However, IAC trajectory could be traced with the projection feature of NNS and the exact area of bone removal along its roof was determined by using the orientation provided by the probe placed between the basal turn of cochlea and the vestibule ( blue arrow ). We were able to validate the location of the IAC via a medial to lateral drilling during the MFA by using this navigation reference point, specifically located in between the cochlea and vestibule. (B) The blue arrow shows the point between the basal turn of cochlea and the vestibule, determined by NNS. Drilling the IAC trace determined by using this point and the NNS projection from a medial to lateral direction revealed the meatal and labyrinthine segments of the facial nerve and lead to the exposure of the the geniculate ganglion. Following the exposure of the IAC, the tympanic segment and middle ear ossicles were exposed by removing the tegmen tympani. Note that the basal turn of the cochlea is located anteromedially of the fundus of the IAC, at the angle between the GSPN and the labyrinthine segment. The vestibule is located anteromedially to the semicircular canals and posterolateral to the fundus of the IAC. The vestibule and cochlea connect with each other just inferolateral to the fundus of the IAC. (C) The arcuate eminence and semicircular canals have been removed to expose the vestibule. The vestibule connects the semicircular canals and the cochlea and locates posterolateral to the fundus of the IAC. In this study, the area between the basal turn of cochlea and the vestibule is the NNS reference point to locate the IAC during the MFA. In the light of IAC trajectory obtained via the projection feature of NNS using this reference point, all perimeatal facial nerve segments and the vestibular nerve have been exposed. Bill's bar is located between the facial nerve and the superior vestibular nerve. Arc., arcuate; CN, cranial nerve; Coch., cochlea; Emin., eminence; Eust., eustachian; For., foramen; GG, geniculate ganglion; IAC, internal acoustic canal; ICA, internal carotid artery; Laby., labyrinthine; Mal., malleus; Mand., mandibular; Max., maxillary; MFA, middle fossa approach; MMA, middle meningeal artery; N., nerve; NNS, neuronavigation system; Oph., ophthalmic; Pet., petrous; Proc., process; Segm., segment; Sup., superior; SV, superior vestibular nerve; Tegm., tegmen; Tent., tentorium; Tymp., tympani; V1, ophthalmic nerve; V2, maxillary nerve; V3, mandibular nerve.
Fig. 2.

Step-by-step middle fossa dissection and revealing the IAC with the guidance neuronavigation system. (A) After craniotomy and dissection of the temporobasal dura, the foramen spinosum was exposed and demonstrated with the NNS. (B) The foramen ovale was revealed and demonstrated by the NNS. (C) After demonstrating the foramen spinosum and foramen ovale and confirming the accuracy of the NNS, the area between the basal turn of cochlea and the vestibule (or approximately the most lateral part of the meatal foramen or labyrinthine segment) was found with the guidance of the NNS. (D) Using the projection feature of the NNS, the navigation probe was oriented according to the porus of the IAC and the trace of the IAC was revealed. (E) After drilling the IAC trace determined with the guidance of the NNS, the IAC was revealed and the navigation probe was placed inside the IAC and the IAC was displayed in the NNS. IAC, internal acoustic canal; NNS, neuronavigation system.
Results
The cadavers were fixed in the surgical position with a three-pin head holder for a subtemporal extradural approach to expose the whole length of the middle fossa floor and the NNS was introduced. A frontotemporal curvilinear skin incision was made, starting from front of the tragus and extending upward, turning posteriorly from above the pinna, and finally advancing anteriorly. The temporalis muscle was elevated from the squamous portion of temporal bone and retracted anteroinferiorly. A temporal craniotomy, centered at the root of the zygoma, was then performed and the temporal bone was drilled until the middle fossa surface was flattened.
The basal dura of temporal lobe was then elevated from the middle fossa floor. First, the foramen spinosum and the middle meningeal artery were identified, and the middle meningeal artery was transected. The dissection proceeded anteriorly and medially to identify the foramen ovale and mandibular nerve (V3). By elevating the dura in a lateral to medial direction over the V3 toward Meckel's cave, the greater superficial petrosal nerve (GSPN) was localized at the point of its intersection with the lateral edge of the V3. The elevation of middle fossa dura continued medially along the petrous apex until reaching the petrous ridge. Subsequently, the GSPN was traced posteriorly and laterally, exposing the arcuate eminence, and tegmen tympani ( Fig. 1A ). 7
After proper introduction, it was observed that the coronal suture, the squamous suture, the foramen spinosum (average radius: 3 mm, min: 1.8 mm, max: 4.7 mm) and the foramen ovale (average radius: 6.87 mm, min: 1.87 mm, max: 8.57 mm) were correctly localized by the NNS in all cadavers, thus confirming the accuracy of the NNS ( Table 1 and Fig. 2A,B ). Moreover, this offers evidence of navigation accuracy within the average range. The part of the IAC close to the petrous ridge was next targeted along the posterior edge of the petrous apex, however, due to the thickness of the superior wall of the canal, the IAC could not be reliably localized along the petrous apex with the NNS. At this point, the NNS only showed the probe contact on the petrous apex without clearly pinpointing the the center of the IAC porus.
Table 1. Foramen ovale, foraman rotundum and internal acoustic canal measurements.
| FO | FS | IAC lateral | IAC medial | |
|---|---|---|---|---|
| Head 1 R | 6.9 | 3.5 | 3.99 | 5.24 |
| Head 1 L | 7.4 | 4.18 | 3.51 | 4.85 |
| Head 2 R | 8.57 | 4.73 | 2.82 | 5.15 |
| Head 2 L | 7.81 | 2.35 | 3.11 | 5.18 |
| Head 3 R | 7.36 | 1.76 | 3.68 | 5.9 |
| Head 3 L | 1.48 | 2.15 | 4.08 | 6.08 |
| Head 4 R | 7.15 | 1.89 | 2.99 | 3.3 |
| Head 4 L | 6.02 | 2.98 | 3 | 3.68 |
| Head 5 R | 8.39 | 3.05 | 4.14 | 4.61 |
| Head 5 L | 7.64 | 3.33 | 3.97 | 5.33 |
Abbreviations: FO, foramen ovale; FS, foramen spinosum; IAC, internal auditory canal; L, left; R, right.
Next, more superficial landmarks were investigated to localize the IAC. The area between the vestibule and the basal turn of the cochlea below the superior semicircular canal was located with the assistance of the navigation probe ( Figs. 1A,B , 2C , and 3A ). Importantly, this interval between the vestibule and the cochlea approximately corresponds to the most superficial and lateral part of the IAC, consistent with the location of the labyrinthine segment ( Fig. 1C ). When the probe was positioned over this area, the projection feature of the NNS was activated to show the exact orientation of the medial part (porus) of the IAC ( Figs. 2D and 3B ).
Fig. 3.

Different from the specimen in Fig. 1 , a specimen with a higher bone thickness in the perigeniculate area. (A) The navigation probe was placed between the basal turn of the cochlea and the vestibule, approximately coinciding with the projection of the meatal foramen and the labyrinthine segment. (B) Using the projection feature of the NNS, the navigation probe was directed toward the porus of the IAC and the trace of the IAC was revealed. (C) After drilling the IAC trace determined with the help of the NNS, the IAC was revealed and the navigation probe was placed inside the IAC and the IAC was displayed in the NNS. IAC, internal acoustic canal; NNS, neuronavigation system.
The presumed IAC area based on the NNS data and orientation was marked with an ink pen. The presumed area of interest was then opened in a medial to lateral direction along the middle fossa floor—beginning from the petrous apex, thus aiming to leave the removal of the more superficial lateral parts of IAC later in the drilling process. The accurate localization of the IAC using the abovementioned surface markers through the NNS could be achieved in all specimens ( Fig. 1A,B ). After the localization of the IAC with this method, the navigation probe was placed inside the IAC, confirming once again the accuracy of the navigation system (average radius: 4.9 mm, min: 3.3 mm, max: 6.08 mm; Figs. 2E and 3C and Table 1 ). Upon comparison of the IAC radius values with those of the foramen ovale and spinosum, it was noted that they fall within the same range and the error margin was minimal. Greater bone thickness over the perigeniculate area did not prevent precisely localizing the IAC trace by using the NNS ( Fig. 3 ).
Discussion
Any approach through the middle fossa floor clearly has the potential risk of misorientation due to the lack of reliable landmarks, thus, hindering the visualization of the surgical target area. It has been reported that the MFA and its extensions are complicated skull base routes to brainstem and the cerebellopontine angle due to the complex and variational anatomy of the middle fossa floor.
Early and precise localization of the IAC, the main step of the MFA, is usually difficult due to the lack of evident, reliable superficial landmarks, narrow working area, and its close relationships with important neurovascular structures. Furthermore, the IAC is curtained by the temporal lobe and located at the deepest point of the surgical field, thus not recognized with ease during the MFA. These factors obligate adopting several surgical landmarks for localizing the IAC, however, most of these neurovascular structures—such as GSPN and petrous segment of the ICA—are also covered with bone, making the surgical approach more challenging.
Several techniques to localize the IAC during the MFA have been proposed based on the anatomical relationships of several neurovascular structures. The methods of House, Fisch, and Garcia-Ibanez and Garcia-Ibanez, are the most widely used techniques. 4 10 11 12 13 14 Of these, House et al's method proposes following the GSPN posteriorly to locate the geniculate ganglion (GG) and a subsequent medial dissection to identify the labyrinthine segment of the facial nerve within the IAC. 10 11 Fisch's technique suggests localizing the blue line of the superior semicircular canal by drilling the arcuate eminence and then localizing the IAC by marking a 60-degree angle from the superior semicircular canal toward the petrous ridge. 13 The Garcia-Ibanez and Garcia-Ibanez's method is based on localizing the IAC by an imaginary line that divides the long axis of the arcuate eminence and the GSPN into two equal angles. 12 In addition to these methods, a previous study from our group described a reliable 61-degree angle between the long axis of the GSPN and the IAC toward the petrous ridge. 7 This easily appreciated angle may be used to open the IAC from above, beginning drilling the meatus medially at the petrous ridge for a safer removal, where the facial and vestibulocochlear nerves are more deeply located. During the drilling, the cochlea anteromedially, vestibule posterolaterally, and superior semicircular canal posteriorly significantly limit the bone removal at the IAC.
All these methods are solely based on constant anatomical landmarks and may not be reliable in case of anatomical variability. One such example is the dehiscence of GG, which may leave the GG vulnerable to damage during elevation of the dura mater. 15 Additionally, the anatomical relationship between the arcuate eminence and the superior semicircular canal may also be variable. Therefore, localizing the superior semicircular canal with reference to the arcuate eminence may not be reliable in some cases. 9 16 17 18 19
The use of NNS and its reliability specifically for the MFA has not been studied yet. The NNS during the MFA has the potential to promote the validation of an already exposed anatomical structure and to guide the direction of the surgical dissection. In this study, we confirmed that superficially located middle fossa key landmark structures can constantly be precisely appreciated and validated via the NNS and IAC can be localized with the guidance of correctly defined surface landmarks with the help of navigation. 20 The NNS indicates the point of probe contact, which is extremely useful for superficial points of interest. Contrarily, NNS cannot detect more deeply located part of medial IAC without drilling on the petrous apex. The depth of the IAC medially at the petrous ridge is greater and the NNS cannot reliably localize the meatus due to the increased thickness of the superior wall of the IAC. However, IAC trajectory could be traced and the exact area of bone removal along its roof was determined by using the orientation provided by the probe placed between the basal turn of cochlea and the vestibule. We were able to validate the location of the IAC via a medial to lateral drilling during the MFA by using this navigation reference point, specifically located in between the cochlea and vestibule deep to the superior semicircular canal.
Theoretically, localization of deep structures via NNS can be performed using the projection feature or instead by freezing the current image and then playing the image externally. In this study, we preferred the projection feature to localize the IAC. The probe of the NNS placed over the petrous ridge to locate the deeper medial part of the IAC does not facilitate revealing the acoustic canal. Navigation technology still needs to evolve for unexposed, deep-lying target structures that are covered by bone. Adding an imaginary line through the “probe's eye view” in these circumstances may further complicate an already complex perception of the targeted deep structure, since the direction of the IAC does not overlap with only one of the axial, coronal, or sagittal planes.
Conclusion
In essence, the NNS seems to be an indispensable tool in MFA to overcome difficulties regarding limited exposure of targeted structures and absence of sound anatomical landmarks. Navigation guidance leads to reliable identification of superficial landmarks along the middle fossa floor. The method of choice to pinpoint the IAC during the MFA can also be confirmed via NNS. We still advocate the necessity of detailed anatomical awareness through laboratory dissections for a safe and effective MFA, and combining this knowledge with the NNS when needed.
Acknowledgments
The authors are grateful to the Department of Anatomy at Koç University School of Medicine for providing the anatomical specimens used in this study. Additionally, the authors sincerely thank those who donated their bodies to science and their families so that this anatomical research could be performed.
Conflict of Interest None declared.
Authors' Contributions
T.A.K. contributed to the investigation, resources, and writing of the original draft.
B.K. was responsible for conceptualization, methodology, and the reviewing and editing of the writing.
A.K. assisted with investigation and resources, while L.A. also contributed to investigation and resources.
N.T. provided supervision and participated in the reviewing and editing of the writing.
G.S.Z. was involved in conceptualization, methodology, and supervision.
Corrigendum: The article has been updated as per corrigendum published on 10 April 2024 (DOI: https://doi.org/10.1055/s-0044-1786156).
References
- 1.Sanus G Z, Tanriverdi T, Tanriover N, Ulu M O, Uzan M.Hearing preserved traumatic delayed facial nerve paralysis without temporal bone fracture: neurosurgical perspective and experience in the management of 25 cases Surg Neurol 20097103304–310., discussion 310 [DOI] [PubMed] [Google Scholar]
- 2.Sanuş G Z, Tanriöver N, Tanriverdi T, Uzan M, Akar Z. Late decompression in patients with acute facial nerve paralysis after temporal bone fracture. Turk Neurosurg. 2007;17(01):7–12. [PubMed] [Google Scholar]
- 3.Sanna M, Zini C, Mazzoni A et al. Hearing preservation in acoustic neuroma surgery. Middle fossa versus suboccipital approach. Am J Otol. 1987;8(06):500–506. [PubMed] [Google Scholar]
- 4.House W F, Shelton C.Middle fossa approach for acoustic tumor removal. 1992 Neurosurg Clin N Am 20081902279–288., vi [DOI] [PubMed] [Google Scholar]
- 5.Kawase T, Shiobara R, Toya S.Anterior transpetrosal-transtentorial approach for sphenopetroclival meningiomas: surgical method and results in 10 patients Neurosurgery 19912806869–875., discussion 875–876 [PubMed] [Google Scholar]
- 6.Kawase T, Toya S, Shiobara R, Mine T. Transpetrosal approach for aneurysms of the lower basilar artery. J Neurosurg. 1985;63(06):857–861. doi: 10.3171/jns.1985.63.6.0857. [DOI] [PubMed] [Google Scholar]
- 7.Tanriover N, Sanus G Z, Ulu M Oet al. Middle fossa approach: microsurgical anatomy and surgical technique from the neurosurgical perspective Surg Neurol 20097105586–596., discussion 596 [DOI] [PubMed] [Google Scholar]
- 8.MacDonald J D, Antonelli P, Day A L. The anterior subtemporal, medial transpetrosal approach to the upper basilar artery and ponto-mesencephalic junction. Neurosurgery. 1998;43(01):84–89. doi: 10.1097/00006123-199807000-00054. [DOI] [PubMed] [Google Scholar]
- 9.Arìstegui M, Cokkeser Y, Saleh E et al. Surgical anatomy of the extended middle cranial fossa approach. Skull Base Surg. 1994;4(04):181–188. doi: 10.1055/s-2008-1058953. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.House W F. Surgical exposure of the internal auditory canal and its contents through the middle, cranial fossa. Laryngoscope. 1961;71:1363–1385. doi: 10.1288/00005537-196111000-00004. [DOI] [PubMed] [Google Scholar]
- 11.Catalano P J, Eden A R. An external reference to identify the internal auditory canal in middle fossa surgery. Otolaryngol Head Neck Surg. 1993;108(02):111–116. doi: 10.1177/019459989310800202. [DOI] [PubMed] [Google Scholar]
- 12.Garcia-Ibanez E, Garcia-Ibanez J L. Middle fossa vestibular neurectomy: a report of 373 cases. Otolaryngol Head Neck Surg. 1980;88(04):486–490. [PubMed] [Google Scholar]
- 13.Fisch U. Transtemporal surgery of the internal auditory canal. Report of 92 cases, technique, indications and results. Adv Otorhinolaryngol. 1970;17:203–240. [PubMed] [Google Scholar]
- 14.Fisch U, Mattox D. New York: Thieme; 1988. Microsurgery of the Skull Base. [Google Scholar]
- 15.Hamzaoğlu V, Beger O, Erdoğan O et al. Radioanatomic assessment of the geniculate ganglion dehiscence and dimension: a cadaveric study. World Neurosurg. 2020;134:e913–e919. doi: 10.1016/j.wneu.2019.11.036. [DOI] [PubMed] [Google Scholar]
- 16.Bulsara K R, Leveque J-C, Gray L, Fukushima T, Friedman A H, Villavicencio A T.Three-dimensional computed tomographic analysis of the relationship between the arcuate eminence and the superior semicircular canal Neurosurgery 200659(1, Suppl 1)ONS7–ONS12., discussion ONS7–ONS12 [DOI] [PubMed] [Google Scholar]
- 17.Kartush J M, Kemink J L, Graham M D.The arcuate eminence. Topographic orientation in middle cranial fossa surgery Ann Otol Rhinol Laryngol 198594(1 Pt 1):25–28. [DOI] [PubMed] [Google Scholar]
- 18.Seo Y, Ito T, Sasaki T, Nakagawara J, Nakamura H.Assessment of the anatomical relationship between the arcuate eminence and superior semicircular canal by computed tomography Neurol Med Chir (Tokyo) 20074708335–339., discussion 339–340 [DOI] [PubMed] [Google Scholar]
- 19.Peris-Celda M, Perry A, Carlstrom L P, Graffeo C S, Driscoll C LW, Link M J.Key anatomical landmarks for middle fossa surgery: a surgical anatomy study J Neurosurg 2018(November):1–10. [DOI] [PubMed] [Google Scholar]
- 20.Komune N, Matsushima K, Matsuo S, Safavi-Abbasi S, Matsumoto N, Rhoton A LJ., Jr The accuracy of an electromagnetic navigation system in lateral skull base approaches. Laryngoscope. 2017;127(02):450–459. doi: 10.1002/lary.25998. [DOI] [PubMed] [Google Scholar]
