Extended Endoscopic-Assisted Kawase versus Endoscopic Endonasal Anterior Petrosectomy: A Cadaveric Study

Background: Less invasive techniques involving endoscopic assistance have been proposed for treatment of lesions of the cavernous sinus, petrous apex, medial jugular foramen, upper and lower clivus, and condylar regions. Performing a two-step CT-guided extended endoscope-assisted approach, we analyzed gains in visibility and surgical freedom and defined landmarks for surgical application.

Materials and Methods: In 5 cadaveric formalin-fixed, silicone-injected heads, bilateral subtemporal-extradural and endoscopic-endonasal CT-guided dissections were performed. First, through a subtemporal approach, the Kawase rhomboid area was drilled and the dura opened. Resections were completed for the postmeatal triangle, between cranial nerve (CN) VII, superior petrosal sinus, superior semicircular canal, and petrous apex covered by the mandibular nerve (V3). After each step, angled 4-mm endoscopes (0, 30, and 45 degree) were used to assess intradural visibility and maneuverability. Exposure depth was measured anteriorly and posteriorly at the level of the inferior petrosal sinus. Landmarks for bone resection were defined using stereotactic navigation. Subsequently, an endoscopic endonasal approach to reach the petrous apex extended laterally between Meckel’s cave and internal carotid artery (ICA), medially at the petroclival junction, and inferiorly to expose the lower cranial nerves.

Results: Transcranially, the postmeatal triangle covered the root entry zone (REZ) of CN V in all cases (10 sides). Average length of bone removal for this triangle at the level of the superior petrosal sinus was 11.4 mm and 3.8 mm anteriorly to the Kawase area, after mobilizing V3. Depth of exposure averaged 14 mm anteriorly and 20.5 mm posteriorly. Compared with the microscope, 0-degree endoscopes yielded no advantage in exposure whereas 30- and 45-degree endoscopes allowed visualization of CNs VI-XII and increased maneuverability. Specifically, the posterior extension of bone removal widened the corridor toward the ipsilateral retrochiasmatic area and perimesencephalic cistern (between CNs V and VI) while the anterior extension widened bilateral inferior exposure (CNs IX–XII). Combination of microscope, endoscope, and image guidance proved extremely effective for maximal bone removal, affording wide views: ipsilaterally, the upper clivus and medial condylar complex; and bilaterally, Dorello’s canal, lower clivus, medial jugular foramen, and hypoglossal canal. Maneuverability was achieved ipsilaterally at the level of basilar artery and bilaterally below the vertebrobasilar junction. Distances between brainstem, petroclival complex, and jugular tubercle were the main limitations to maneuverability. Visualization of CNs III to VI was achieved endonasally by lateral extension of the approach whereas visualization of CNs V to VIII was achieved by the transclival approach. The lower cranial nerves were exposed after extensive transpterygoid drilling. Maneuverability decreased with lateral extension of dissection.

Conclusions: The extended endoscope-assisted transcranial approach should be considered for intradural lesions that extend inferiorly and contralateral to the lower clivus and medial jugular foramen. The endonasal approach remains a better choice for extradural lesions located medially in the upper and middle clivus.