Internal Auditory Canal Variability and Facial Nerve Outcome after Translabyrinthine Resection of Vestibular Schwannomas

Background: The internal auditory canal (IAC) contains the vestibulocochlear and facial nerves and serves as an important landmark during resection of vestibular schwannomas (VSs). We have observed anatomical variability of the IAC and differences in surgical exposure depending on individual anatomy. Based on these observations, we analyzed whether facial nerve outcome after a translabyrinthine approach for resection of a VS is correlated with IAC anatomic variability.

Methods: We retrospectively identified patients with pathologically confirmed VSs treated using the translabyrinthine approach between May 2014 and March 2019. Patients <18 years, with tumors found intraoperatively to be arising from the facial nerve, or who underwent nonelective procedures were excluded. To determine the size and variability of the IAC, we assessed preoperative axial thin-slice T2-weighted MRI sequences. We measured the anterior (APD) and posterior (PPD) petrous distances, the porus dilation, and the internal auditory angle (IAA). We also estimated tumor volume. Facial nerve outcomes were quantified using the House–Brackman (HB) score recorded in the medical record on postoperative day (POD) 1, at discharge, and at 1-month follow-up.

Results: The study population included 65 consecutive patients (33 female) with a mean age of 50.5 years (18–85 years). All patients presented with moderate to severe sensorineural hearing loss. Tumor volume ranged from 0.03 to 52.8 cm³ (mean, 8.3 cm³). APD ranged from 7.3 to 34.9 mm (mean, 13.29 mm). PPD ranged from 7.8 to 33.3 mm (mean, 25.19 mm). IAA ranged from 0 to 28 degrees (mean 12.8 degrees). PD ranged from 0 to 11.3 mm (mean 3.17 mm). On univariate and multivariate linear regression, there was weak correlation between tumor volume and facial nerve outcomes at POD 1 (p = 0.03), discharge (p = 0.01), and follow-up (p = 0.1). IAA was an independent predictor of facial nerve outcomes at POD 1 (p = 0.0001), discharge (p = 0.0001), and follow-up (p = 0.0001). Using an ROC curve, an IAA cutoff value of 14.5 degrees predicting poor HB grade (≤2) was identified. This cutoff yielded a sensitivity and specificity, respectively, of 0.72 and 0.66 at POD1, 0.76 and 0.64 at discharge, and 0.76 and 0.64 at follow-up.

Conclusion: We have shown that the IAA is an independent predictor of short-term facial nerve outcome in patients being treated via a translabyrinthine approach for VSs. We hypothesize that the angle of the facial nerve toward the IAC affects its visualization during the translabyrinthine approach and that a smaller IAA (anteriorly angled IAC) is likely to direct tumor growth anteriorly. Therefore, the IAA is a critical measurement that can be used to determine the relative risk of facial nerve palsy using the TL approach. Our results here indicate that the translabyrinthine approach will likely provide suboptimal facial nerve visualization with an IAA <14.5 degree and a retrosigmoid approach should be considered because of the better visualization of the facial nerve and worse facial nerve outcome in the group using a translabyrinthine approach.