Discussion

Following the development of endoscopic middle-ear surgery, the number of reports of a direct approach to the internal auditory canal to remove vestibular schwannoma has increased. However, the disadvantages include limited access, difficult haemostasis, and difficulty removing the tumour when it extends through the porus and cerebellopontine angle. Therefore, the endoscopic transcanal transpromontorial approach technique was clinically proposed to remove vestibular schwannomas limited to less than 1 cm in diameter involving the most lateral portion of the internal auditory canal and tumours more than 1 cm but less than 2 cm involving the cerebellopontine angle, classified as Koos grading scale stage I and II, respectively.^{Reference Marchioni, Carner, Soloperto, Bianconi, Sacchetto and Sacchetto2}

The disadvantages and limitations of surgery have been improved following various trials and previous studies. For example, Marchioni et al. removed tumours equal to or less than 3 cm in size with linear progression into the cerebellopontine angle using an expanded transcanal transpromontorial approach technique with a microscope and endoscope.^{Reference Marchioni, Carner, Soloperto, Bianconi, Sacchetto and Sacchetto2} They reported preservation of facial nerve function in 95.9 per cent of cases using this technique (House–Brackmann grade I–II) with stable results at follow up; facial nerve function was reduced in 4.1 per cent of cases but was no worse than grade III.^{Reference Marchioni, Soloperto, Masotto, Fabbris, Rossi and Villari3}

Another criticism of this approach is that it requires the removal of the cochlea to access the internal auditory canal and therefore precludes possible simultaneous or subsequent cochlear implantation. A transcanal infrapromontorial route with near-total cochlear preservation was performed by Rubini et al., who removed only the most posterior portion of the basal turn of the cochlea in cadaveric dissections. This approach may be considered an option for removing small intracanalicular schwannomas (less than 0.5 cm cerebellopontine angle spread) when concurrent cochlear implantation is indicated.^{Reference Rubini, Bianconi, Patel and Marchioni4} However, to reach the internal auditory canal without damaging the cochlea, the infracochlear corridor must have a specific anatomical geometry to allow adequate visualisation while avoiding the jugular bulb and internal carotid artery. First, to allow for safe passage past the jugular bulb and internal carotid artery, the diameter of the infracochlear tunnel must be more than 3 mm to accommodate both an endoscope and drill.^{Reference Kempflea, Fiorilloa, Kanumuria, Barbera, Edgea and Cunnanea5} An aberrantly high jugular bulb can limit the surgical approach (Figure 4).

Fig. 4. The three-dimensionally reconstructed images (a) and axial (b), sagittal (c), and coronal (d) plane images of temporal bone computed tomography with a high jugular bulb (arrow). Endoscopic transcanal transpromontorial approach is limited by the high jugular bulb occupying the space (white arrow in a–d) in the middle-ear cavity. A = anterior; R = right; L = left; P = posterior; H = highest; F = floor

In the conventional transcanal transpromontorial approach procedure, the following surgical landmarks limit dissection: the jugular bulb is located inferiorly, the internal carotid artery is located anteriorly, the geniculated ganglion and tympanic segments of the facial nerve lie superiorly, and the mastoid segments of the facial nerve lie posteriorly. After fully exploring the medial wall of the vestibule, this approach allowed identification of an anatomical triangle among the geniculate ganglion superiorly, the basal turn of the cochlea anteriorly and the spherical recess of the internal auditory canal posteroinferiorly. Drilling into this anatomical triangle enabled identification of the intralabyrinthine tract of the facial nerve from the internal auditory canal to the geniculate ganglion, and the anatomical position of the intralabyrinthine facial nerve was noted.^{Reference Marchioni, Alicandri-Ciufelli, Mattioli, Nogeira, Tarabichi and Villari6}

Identification of the labyrinthine segment of the facial nerve among the four nerve segments coming from the internal auditory canal fundus is essential. The directions of the four nerve segments and the bony crest serve as guides because the labyrinthine segment is located superior-anterior to the internal auditory canal and runs toward the geniculate ganglion (point 5).

A radiological study using six radiological points to measure the maximum area of exposure in the transcanal transpromontorial approach using 3D temporal bone reconstructions showed a mean lateral window in the middle ear of 152.9 mm² and a similar medial window at the internal auditory canal porus of 151.9 mm².^{Reference Yacoub, Wimmer, Molinari, Alicandri-Ciufelli, Presutti and Caversaccio7} The six points were applied in this study.

The image transmitted to the monitor through the endoscope is not stereoscopic. The critical structures of the surgical landmarks are located under the medial wall of the middle-ear cavity (Figure 5). Therefore, knowledge of the surgical landmarks is mandatory to preserve the facial nerve.^{Reference Marchioni, Carner, Rubini, Nogueira, Masotto and Alicandri-Ciufelli8} At each stage of an operation, the surgeon must check for the facial nerve using an endoscope and decide where to drill. If the distance and angle between the structures are known, especially in a narrow space, this will help the surgeon understand and proceed with the surgery.

Fig. 5. Composite and endoscopic images of the three-dimensional reconstruction of the temporal bone computed tomography. The structures of critical surgical landmarks are located underneath the medial wall of the middle-ear cavity. Point 1: the highest level of the jugular bulb; point 2: the posterior wall of the ascending petrous internal auditory canal at the level of the hypotympanum; point 3: the level of the Eustachian tube; point 4: cochlear apex; point 5: facial nerve at the centre of the geniculate ganglion; point 6: centre of the second genu; and point 7: the branch of the chorda tympani nerve.

In the cadaveric study by Presutti et al., an imaginary line passing from the geniculate ganglion to the spherical recess just above the apical turn of the cochlea indicated the facial nerve route through the inner ear.^{Reference Presutti, Bonali, Marchioni, Pavesi, Feletti and Anschuetz9} We obtained similar results in this study. The value of aP_fOR (101.81 ± 8.86°) was between those of aP₄OR (77.32 ± 6.31°) and aP₅OR (111.77 ± 7.87°) but closer to aP₅OR, which is similar to the imaginary line reported by Presutti et al.^{Reference Savic and Djeric10} The distances dP_fR and dP_fO were 6.18 ± 0.34 and 5.30 ± 0.41 mm, respectively (Table 1).

The internal carotid artery and jugular bulb are the two largest blood vessels neighbouring the middle-ear cavity. As the internal carotid artery travels through the petrous bone, it first passes the middle-ear cavity and cochlea anteriorly and then bends to run medially to the Eustachian tube.

An aberrant high jugular bulb is characterised by a high-positioned jugular fossa, either medial or lateral to the cochlea. Histopathological studies show an estimated 25 per cent incidence of a high jugular bulb protruding into the tympanic cavity in the temporal bone.^{Reference Savic and Djeric10} Therefore, pre-operative assessment of the vascular structures within the temporal bone is of prime importance.^{Reference Wang, Shi, Liu, Wang, Huang and Chen11} In this study, P₁–P₃ were localised to the vascular structure. In particular, the standard deviation of the dP₁O and dP₁R values, which are the distances from the oval window and round window to the jugular bulb, were larger than the distances to P₂ and P₃ (Table 1).

The point P₃ could be identified by drilling underneath the Eustachian tube, and P₅ could be traced using the cochleariform process during surgery because the cochleariform process is also positioned just inferolateral to the geniculate ganglion.^{Reference Komune, Matsuo, Miki and Rhoton1} The cochlear apex (P₄) traces along the modiolus and could be identified between P₃ and P₅. The distance and angle to P₄ based on the oval window and round window could be affected by various cochlear morphologies.^{Reference Erixon, Högstorp, Wadin and Rask-Andersen12}

Removing the triangular bone landmark exposes the dura lining the internal acoustic canal. Opening the dura of the internal acoustic canal exposes the facial, cochlear, and superior and inferior vestibular nerves. The upper portion of the middle turn of the cochlea serves as a landmark for the labyrinthine part of the facial nerve (P_f) because it is located just superior to the middle turn of the cochlea.^{Reference Komune, Matsuo, Miki and Rhoton1}

The expanded transcanal transpromontorial approach is a combined microscopic and endoscopic technique proposed for large vestibular schwannomas up to 3 cm in diameter.^{Reference Marchioni, Carner, Soloperto, Bianconi, Sacchetto and Sacchetto2} In the expanded transcanal transpromontorial approach, the mastoid segment of the facial nerve becomes the boundary for surgery, marking the related P₆ and P₇ surgical indicators. The facial nerve gives rise to the chorda tympani in the vertical mastoid segment, which lies near the lower margin of the bony tympanic ring and 4–5 mm proximal to the stylomastoid foramen. However, this branching point is variable.^{Reference McManus, Dawes and Stringer13}

Radiological evaluation is less accurate than microscopic anatomical studies by cadaver dissection of the temporal bone structures. Moreover, the anatomical landmarks found in actual surgery are more critical. Nevertheless, the advantage of an imaging study is that it can be performed pre-operatively for patients in order to customise their surgery.

In this study, the measured distance to the centre point of the oval window and round window was 2.65 ± 0.36 mm. This value is similar to that of 2.02 ± 0.56 mm from the study by Shraddha et al., considering that it was measured with the shortest distance between the edge of the oval window and round window in cadaver studies.^{Reference Shraddha, Gaurkar, Deshmukh, Khatri, Kalambe and Lakhotia14} In this study, to increase the accuracy of the designation of this position on 3D reconstructed images, a 3D cursor was activated on the axial, sagittal and coronal plane images. Despite the limitations of this imaging study, it has the advantage of 3D confirmation to check the location of important structures and prepare surgeons for any likely problems they may encounter during surgery.