J Neurosurg 145:1128–1140, 2026
Choosing an operative corridor for vestibular schwannoma requires balancing tumor removal with hearing, facial function and the demands of safe neurovascular dissection. Erginoglu and colleagues compare single-stage retrosigmoid and translabyrinthine resections in a consistent surgical practice, examining tumor volume alongside anatomical grade. Their findings associate the translabyrinthine route with more complete removal and less postoperative edema in larger tumors, while the retrosigmoid route retains the possibility of hearing preservation. Rather than establishing a universally superior approach, the study illustrates how audiometry, internal auditory canal involvement and the tumor-brainstem relationship should inform individualized planning.
Objective
To compare resection, cranial nerve function, radiological changes and recovery after retrosigmoid (RS) and translabyrinthine (TL) surgery, and assess how preoperative tumor volume influences outcomes beyond Koos grade.
Methods
This retrospective study included 283 patients undergoing single-stage resection at one tertiary center between 2006 and 2023: 207 through TL and 76 through RS. All procedures involved one neurosurgeon and a consistent neurotology team. Previous surgery or radiotherapy, neurofibromatosis types 1 or 2, and staged, combined or alternative approaches were excluded. Follow-up was at least 12 months.
Serviceable hearing, defined by audiometric class A or B, generally favored RS; nonserviceable hearing generally favored TL. However, anatomical or audiological considerations led to deviations in approximately 30% of cases. The groups therefore differed substantially before surgery: serviceable hearing was present in 30.9% of TL and 72.4% of RS patients.
Tumor volume was estimated from three MRI diameters using a modified ellipsoid formula. Outcomes included extent of resection, House-Brackmann facial function, hearing, MRI abnormalities and modified Rankin Scale scores. Multivariable models and two separate propensity-matched analyses, each with 52 patients per approach, addressed measured baseline differences.
Main results
For Koos III-IV tumors, Table 2 reports gross-total resection in 106/125 TL cases (84.8%) versus 39/56 RS cases (69.6%). Edema occurred in 0.8% versus 30.4%, respectively. Across all grades, favorable one-year facial function among patients starting with House-Brackmann I-II was 93.9% after TL and 94.7% after RS, without significant separation. Hearing preservation occurred only after RS. Median hospitalization was five days after TL versus four after RS in both Koos strata, although rehabilitation discharge was more frequent after RS in larger tumors. Greater volume reduced the likelihood of complete resection.
Interpretation
The operative decision should integrate hearing potential with the required exposure. TL offers direct internal auditory canal access and early facial nerve identification, while sacrificing hearing; RS provides broad cerebellopontine angle exposure with a possibility of hearing preservation. The authors describe CSF release and avoidance of fixed retractors during RS, yet intermittent cerebellar manipulation remained necessary. Less manipulation is a plausible explanation for the TL edema findings, not a proven causal mechanism. Dense adherence to the facial nerve or brainstem should still limit resection rather than be overcome to achieve a radiographic endpoint.
Limitations
This was not a randomized comparison. Hearing status and anatomy actively determined the approach, and adjustment cannot eliminate confounding by indication. Single-surgeon consistency improves procedural uniformity but limits generalizability. The 18-year period also encompassed changing practice: RS use declined over time. These results do not compare surgery with observation or radiosurgery, nor do they apply directly to previously treated or staged cases.
Sparse events made several adjusted models unstable. The very large estimated association between TL and favorable discharge function had a wide confidence interval and should not be interpreted as a reliable measure of treatment effect. Multiple comparisons were performed without correction. Gliosis, unlike edema, did not remain significantly different in the matched radiological analysis. Radiological abnormalities also did not consistently correspond to persistent disability.
Reporting inconsistencies warrant caution. The abstract labels the 84.8% and 69.6% figures as combined gross-total/near-total resection, whereas Table 2 assigns them to gross-total resection alone. The reported hearing-preservation percentages use the entire RS Koos subgroups as denominators, rather than clearly restricting them to patients with serviceable preoperative hearing; they should not be presented as individualized preservation probabilities. Patient-reported quality of life and systematic headache outcomes were not collected, and the low recurrence counts do not establish long-term oncological equivalence.
Clinical takeaway
Use tumor volume alongside Koos grade, detailed audiometry and canal-brainstem anatomy to select the corridor. TL may offer useful exposure when hearing preservation is not a realistic priority; RS remains relevant when it is. This series supports tailored planning and function-limited resection, not routine TL selection for every large tumor or an expectation of shorter hospitalization.