Endoscopic Transnasal Transmaxillary Approach to Orbital Apex through the Meningo-Orbital Band: A Cadaveric Feasibility Study
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0028-3886.355088
Source of Support: None, Conflict of Interest: None
Keywords: Cavernous sinus, expanded endonasal surgery, meningo-orbital band, middle cranial fossa, optic nerve decompression, superior orbital fissure, trans maxillary, transpterygoid, transsphenoid
The meningo-orbital band (MOB) is a fibrous dural fold that tethers the frontotemporal dura to the lateral part of the superior orbital fissure (SOF) and the periorbita. The MOB separates the anterior and the middle cranial fossa and occupies the lateral part of SOF. Medially, the MOB blends with the periorbita at the medial compartment of the SOF., The MOB and the lateral part of the SOF contain a few veins and the meningo-orbital artery but are devoid of cranial nerves; therefore, the MOB can be cut without any neurological deficits. The medial compartment of the SOF at the orbital apex contains the cranial nerves where the periorbita is continuous with the dura of the cavernous sinus. An extradural transcranial approach needs to cut the MOB to expose the middle cranial fossa (MCF). Release of the MOB is essential to perform anterior clinoidectomy for the unroofing of the optic canal, exposure and or mobilization of the clinoidal segment of the internal carotid artery (ICA), and splitting the outer dural cavernous wall to expose the middle cranial base furthermore medially., The transcranial route is lengthy, requires extensive bone drilling and brain retraction for medial and central skull base pathologies. More rationale was considered by developing less-invasive approaches for procedures such as biopsy, orbital apex decompression, and excision of small-sized tumors. Endoscopic subtemporal and transorbital approaches for release and or bypass of the MOB have been described in anatomical and surgical studies for medially situated pathologies in the anterior and middle cranial fossa,, orbital apex can be approached superomedially through the endonasal transsphenoidal transethmoidal corridor. The superolateral part of the orbital apex is approached through transcranial and transorbital corridors. The endonasal transpterygoid approach is a relatively broader minimally invasive route to the medial part of MCF but needs to be expanded to reach the orbital apex superolaterally.,Limitations of this approach are due to the narrow surgical corridor and attachments of temporal lobe dura to the periorbita, lateral wall of the cavernous sinus, and cranial base. Endonasal mobilization of MCF dura from the lateral wall of the cavernous sinus has been addressed at the maxillary nerve (V2), then the SOF hindered further superior mobilization of MCF dura., Theoretically, the endonasal release of the MOB from the periorbita and cavernous sinus will secure the medial compartment of the SOF containing cranial nerves. This will increase the bony window to MCF superiorly until the frontotemporal dural junction by allowing drilling of the greater and lesser wings of the sphenoid bone around the lateral part of the SOF. In the same context, endonasal exposure and decompression of the SOF and optic apex can be achieved. To the best of our knowledge, an endonasal approach to releasing the MOB has not yet been described. The authors, therefore, aimed to test the feasibility of exposure and release of the MOB through an endonasal endoscopic approach.
Five adult latex injected cadaveric specimens underwent a helical computed tomography (CT) scan followed by the creation of an axial dataset (Siemens Emotion 16), 1 mm or thinner slice thickness, 1 mm slice gap axial to occlusion plane, and including the whole head. Digital data were imported to a DICOM viewer (RadiAnt, Medixant, and Poznan, Poland). We divided the superior orbital fissure into medial and lateral compartments in relation to the bony projection of the lower border, which represents the attachment of the annular tendon. The SOF shape, dimensions, and angulation were assessed. [Figure 1]a shows different points on the SOF used for measurements in a dry skull. [Figure 1]b, [Figure 1]c, and [Figure 1]d show the corresponding points and angles of the SOF in three-dimensional (3D) reconstructed and coronal views of the paranasal sinuses, respectively.
The ethics committee of Fayoum University, Egypt, approved this research protocol (Code number: EC 2021). Ten sides of latex-injected embalmed cadaveric heads were dissected endonasally. A nasal endoscope (Karl Storz, Tuttlingen, Germany) 4 mm in diameter, 18 cm in length with 0 and 30 degrees angles was used. Digital pictures were reproduced by coupling the endoscope to a video camera and a computer capture system.
Uncinectomy, wide middle meatal antrostomy, complete anterior and posterior ethmoidectomy, wide sphenoidotomy, and middle turbinate resection were done. The sphenopalatine foramen and sphenopalatine artery (SPA) were exposed, and the perpendicular plate of the palatine bone was drilled to expose the upper medial quadrant of the pterygopalatine fossa (PPF). The posterior wall of the maxillary sinus was removed. The SPA was cut, and vascular structures and PPF fat were dissected and retracted inferiorly. The vidian canal, the foramen rotundum (FR), and the maxillary nerve (V2) were identified in the upper medial part of the posterior wall of the PPF [Figure 2]a and [Figure 2]b. This was followed by the decompression of the SOF medially and inferiorly. This step aimed to allow later exposure and manipulation of the MOB without traction or compression on any neurovascular structure in the medial compartment of the SOF.
Medial decompression of SOF and optic canal
The lamina papyracea was thinned out with a diamond drill and removed using a septal dissector, exposing the periorbita from the frontal process of the maxilla anteriorly until the orbital apex posteriorly, and from the frontal part of fovea ethmoidal superiorly until the orbital floor inferiorly. The opticocarotid recess (OCR) was identified in the sphenoid sinus. It demarcates the optic strut, which forms the floor of the optic canal. The bone overlying the carotid protuberance of the para sellar ICA and medial wall of the cavernous sinus was thinned out with a diamond burr and removed with a 1 mm Kerrison punch. Bone removal extended from the parasellar ICA and planum sphenoidale until the orbital apex and posterior part of the periorbita. Bone covering the orbital apex was usually tough and required drilling with a diamond burr. The optic canal was drilled out with diamond burr starting proximally at the planum sphenoidale, tuberculum sellae, and ending anteriorly at the annulus of Zinn. Optic nerve decompression started first medially then superiorly. Anteriorly, the optic nerve declines downward away from the skull base, allowing more bone removal from the roof of the optic canal before it enters the annulus of Zinn. Inferior optic nerve decompression could be added by partial removal of the optic strut, if possible. This step was done through the OCR after identification of the para sellar ICA to avoid its injury. Partial removal of the optic strut reveals the paraclinoid ICA and the proximal dural ring blending with the periorbita and the annulus of Zinn. The annulus of Zinn is exposed by a diamond drill from the distal part of the optic nerve and skull base superiorly until V2 inferiorly. The maxillary strut was finally the only remaining bone at the SOF medially [Figure 3]a, [Figure 3]b, [Figure 3]c, and [Figure 3]d.
Inferior decompression of SOF and cavernous sinus apex
The periorbita was dissected laterally and superiorly from the GWS. The MOB was early encountered owing to the anterior orientation of the lateral part of SOF. The pterygoid base was drilled to expose the intracanalicular V2 segment in the maxillary canal between the FR at the PPF and MCF. V2 canal was drilled medially and inferiorly through the pterygoid base. The remaining part of the pterygoid base was drilled until the periosteum adjacent to the medial pterygoid muscle to give space for downward transposition of V2. The upper part of the posterior wall of the maxillary sinus was removed to expose the inferior orbital fissure and infraorbital nerve. A fibromuscular band (Muller's muscle) appeared entangling V2 and was dissected cautiously to release and avoid injury of V2. The whole course of V2 and the proximal part of the infraorbital nerve were exposed. V2 was delivered out of its canal and transposed inferiorly. The GWS and remaining part of the V2 canal were thinned out by a diamond burr and removed by a Kerrison punch. The maxillary strut was finally removed by a Kerrison punch. V2 was transposed superiorly and or inferiorly to allow removal of GWS. The corridor through GWS extended from CS medially to temporalis and lateral pterygoid muscles laterally. Inferiorly, the corridor extended until the mandibular nerve (V3) and MCF base. Superiorly, the medial compartment of SOF was fully decompressed. A small rim of GWS was left below the lateral part of SOF obscured by the MOB [Figure 4]a, [Figure 4]b, [Figure 4]c, and [Figure 4]d.
Exposure and release of MOB
In the sagittal plane, the lateral part of SOF is situated more anterior than its medial part. From the ventral endonasal perspective, the lateral part of MOB was encountered early during the dissection of the periorbita from GWS. A distinct plane between MOB, periorbita, and the medial compartment of SOF appeared with more medial and posterior dissection. The MOB appeared as a whitish fibrous band that blends with the periorbita and dura propria. The MOB was separated by sharp dissection with a curved micro-scissor from the periorbita of the medial compartment of SOF, which was essential to avoid compression or traction on the medial compartment of SOF. After its separation, the MOB floated a few mm below the lesser wing of the sphenoid bone, obscuring the upper remnant rim of GWS. The MOB was removed by piecemeal excision to visualize the remnant rim of GWS (lower margin of the lateral part of SOF). From this step onward, a slight upward orbital retraction was needed to continue dissection. Orbital retraction could be minimized using a low lateral transmaxillary window through which the endoscope was directed upward. The upper remnant rim of GWS was removed with a Kerrison punch, preferably with an upward curved end. Exposure of the meningo-orbital artery passing through the cranio-orbital foramen was a good landmark of reaching the vicinity of the frontotemporal dural junction and or the level of the edge of the lesser wing of sphenoid (LWS). The frontotemporal dural junction and the upper rim of the lateral part of SOF were exposed. The edge of LWS was drilled from lateral to medial toward the AZ and optic canal. The edge of LWS ended medially into the roof of the optic canal and optic strut and was used as a landmark to identify the optic nerve. This led to 360-degree decompression of SOF [Figure 5]a, [Figure 5]b, [Figure 5]c, [Figure 5]d, [Figure 5]e, and [Figure 5]f.
Separation of temporal lobe dura from periorbita and lateral cavernous dural wall
Medially, the temporal lobe dura is attached to the lateral wall of the cavernous sinus. A plane is encountered at the junction of the cavernous sinus with periorbita and medial compartment of SOF. Here, the ophthalmic nerve is identified and helps to peel off the temporal dura from the inner layer of the lateral wall cavernous sinus, exposing the inner lateral cavernous membrane. The plane of dural splitting at V1 can be obscured by fat in the anteromedial triangle. The level of dural separation from the cavernous sinus can be appreciated by the position and medial retraction of V2. Further, at V2, a distinct interdural plane is easily identified. The connection of interdural plane at V1 and V2 avoided the disruption of the inner cavernous dural membrane and any cranial nerve injury. Further dissection upward and laterally of the temporal lobe dura will separate it from the cavernous sinus. Care must be taken during dissection above the V1–V2 anteromedial triangle (sensory triangle) to avoid disruption of the inner lateral cavernous membrane and or injury of nerves in the motor triangle. These are the trochlear and oculomotor nerves in the inner membrane of the lateral wall of the cavernous sinus. An endoscopic transmaxillary approach was done for better visualization of cranial nerves in the lateral wall of the cavernous sinus. The endoscopic transmaxillary approach allowed angled lateral to medial visualization of the sagittal oriented lateral wall of the cavernous sinus, orbital apex, and medial part of SOF. This allowed complete peeling of the lateral wall of the cavernous sinus while visualizing the motor nerves (trochlear and oculomotor) and avoided a blind procedure. Medially, Meckel's cave became more apparent, joining V1 and V2. The temporal lobe dura was mobilized inferiorly from the MCF base. This was achieved by the removal of the inferior remnant of GWS and pterygoid wedge. Lateral dissection reached the mandibular nerve (V3), lateral pterygoid, and temporalis muscle. An interdural plane was obtained between temporal lobe dura and V3 and led to better visualization of Meckel's cave and the Gasserian ganglion. Dissection ended posteriorly at the superior petrosal sinus [Figure 6]a, [Figure 6]b, [Figure 6]c, [Figure 6]d, [Figure 6]e, [Figure 6]f, and [Figure 6]g.
A video showing steps of cadaveric dissection.
[Table 1] shows different measurements and angles of the SOF. The SOF is oblique (mean: 39 ± 2.75 degrees) to the mid-sagittal plane. The medial and lateral compartments are almost nearly horizontal to each other, showing minimal angulation (range: 156 to 217, mean: 179 ± 23.34 degrees). The maximum length and width of SOF were 1.35 to 1.79 cm (mean: 1.6 ± 0.16 cm) and 5.54 to 9.8 mm (mean: 7.5 ± 1.4 mm), respectively, whereas the length of the lateral compartment corresponding to MOB ranged from 5.47 to 10.2 mm (mean 6.08 ± 2.58 mm). The distance between the mid-sagittal plane to the medial and lateral ends of SOF ranged from 1.36 to 1.75 cm (mean: 1.56 ± 0.13 cm) and 2.16 to 2.94 cm (mean: 2.97 ± 0.11 cm) respectively. The distance between the medial and lateral ends of SOF in the horizontal plane was 1.09 to 1.48 cm (mean: 1.23 ± 0.15 cm).
The MOB was exposed in 10 cadaveric sides as a whitish fibrous dural band with a distinct plane with dura propria and periorbita. Its release led to the exposure of V1 at the medial compartment of the SOF. The MCF dura was mobilized at V1, V2, and V3 from the inner membranous layer of the lateral wall of the cavernous sinus, orbital apex, and MCF bony base. The surgical corridor to MCF was bounded superiorly by the frontotemporal dural junction, inferiorly by MCF bony base, medially by cavernous sinus and orbital apex, and laterally by temporalis and lateral pterygoid muscles. The SOF was decompressed 360 degrees. The MOB was exposed by both endonasal and transmaxillary approaches. The endoscopic transmaxillary approach was associated with less orbital retraction and allowed better visualization and avoided blind dural splitting of the lateral wall of the cavernous sinus. The overall time of dissection for each side ranged from 95 to 120 min. No breach of the integrity of neurovascular structures was encountered.
A critical step in MCF approach is the proper mobilization of the temporal lobe dura from the cranial base, lateral wall of the cavernous sinus, and the periorbita. This allows isolation and protection of neurovascular structures. Cutting the MOB leads classically to interdural dissection of the lateral wall of the cavernous sinus at V1, whereas following its remnant more medially leads to the anterior clinoid process. This, if done blindly, can lead to the injury of cranial nerves and needs a high degree of expertise.,,
The endonasal MCF approach is a short route to the central skull base. The main limitations are the narrow surgical corridor and difficult vascular control. The SOF and MOB present an obstacle for widening the surgical corridor superiorly and mobilizing MCF dura lateral to the orbital apex and cavernous sinus. The concept of the endonasal approach to the SOF, mainly its lateral part, sounds difficult owing to the general posterior relation of SOF to the orbit, Proper anatomical knowledge revealed the feasibility of endonasal exposure of the SOF up to its lateral end. This can be achieved by the dissection of inferior orbital fissure and transposition of V2, In this study, radiological dimensions of SOF were similar to previous similar anatomical and radiological studies., The oblique vertical orientation of SOF and length of its lateral part, coinciding with MOB were similar to other previous studies. The short distance of lateral and medial ends of SOF from midline was in favor of the endonasal approach to the MOB. Furthermore, the distance between the medial and lateral ends of SOF in the horizontal plane was shorter than the actual length of SOF owing to the small oblique sagittal angle. The width of the MOB depends upon the shape and length of the lateral part of SOF and can be appreciated from CT.
The anterolateral orientation of SOF and early identification of plane between MOB and periorbita and dura propria also favor the ventral endonasal approach of MOB. In comparison with the transcranial approach, MOB prevents temporal lobe retraction, and further dissection must be directed posteriorly in a blind manner to obtain a plane with the periorbita. In lateral approaches, this step is bothersome and many techniques have been developed to avoid injury of cranial nerves in the medial compartment of SOF.,,,
Early decompression of SOF medially is an essential step before dissection of the periorbita to expose MOB. This minimizes the risk of retraction or compression on the medial compartment of SOF. A similar step is performed in lateral extradural approaches. Early drilling of the lateral part of SOF avoids its compression and makes the MOB more evident, [Figure 7]a, [Figure 7]b, [Figure 7]c. In transorbital approaches, the MOB readily floats in the field; however, extensive orbital retraction is required. The endonasal approach proposed in this study also requires retraction of periorbita; in comparison to the transorbital external approach, orbital retraction and or compression is minimal. This is probably due to early medial orbital decompression as an initial step of the endonasal approach to MOB. The transorbital approach provides better simultaneous access to the anterior cranial fossa. However, the current study aims to improve mainly access and overcome the limitation of the endonasal approach to MCF. The endonasal transmaxillary approach provides a 360-degree decompression of the orbital apex and superior orbital fissure and small-sized tumors in the most medial part of the MCF in close relation to the superolateral part of the SOF can be either biopsied or removed via this relatively minimal invasive approach. [Table 2] shows a comparison between different approaches to the orbital apex.
The endoscopic transmaxillary approach is commonly added to the endonasal approach. Its lateral to medial angulation, compared to the usual transnasal approach, which is limited by the pyriform aperture, allows endoscopic visualization of hidden areas of the skull base. In the current study, the endoscopic transmaxillary approach allowed lateral to medial visualization of the lateral wall of the cavernous sinus. Inferior to superior angulation was associated with less orbital retraction and more effortless drilling of the edge of the lesser wing of the sphenoid, of the lateral part SOF. The MOB was, however, adequately exposed by both transnasal and transmaxillary endoscopic approaches.
Another advantage of the endonasal route is the early exposure of V2, which helps identify the MOB. V2 points out the location of the foramen rotundum to confirm the location of the medial compartment of SOF. Easy identification of the interdural plane at V2 also allows for easier separation of temporal lobe dura at V1. Finally, the procedure of release of the MOB allows for better interdural separation of the lateral cavernous wall. This was an obstacle when the peeling procedure was done at V2 only without releasing the MOB. Additionally, endonasal exposure and release of MOB will allow for better decompression of SOF.
Neuronavigation would have facilitated the cadaveric dissection but was not used in this study. A comparative cadaveric study to contrast the degree of exposure of MCF, the lateral wall of cavernous and petrous apex between transcranial and endoscopic trans maxillary approaches needs to be conducted to convincingly comment on the relative superiority of operative exposure and working angles on one technique over the other. The middle meningeal artery and greater superficial petrosal nerve were encountered in this study during the dissection of the dura from the middle cranial fossa. Details of exposure of these structures are out of the scope of this study but are needed for further enhancing lateral exposure via this approach as a large number of MCF lesions extend posterolaterally to the petrous apex.
The current study provides detailed endonasal anatomy of the upper limit of the endonasal approach to MCF and adds to better endonasal 360-degree exposure of the orbital apex and in particular the SOF. The MCF dura was mobilized at V1 and MOB from the cavernous sinus apex and medial compartment of SOF. Further mobilization of MCF was achieved at V2 and V3 from the lateral wall of the cavernous sinus and MCF base.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
[Figure 1], [Figure 2], [Figure 3], [Figure 4], [Figure 5], [Figure 6], [Figure 7]
[Table 1], [Table 2]