Near-total resection in sporadic vestibular schwannoma: is there a volumetric threshold for a win-win scenario?

Abstract

OBJECTIVE: Surgical management of large vestibular schwannoma (VS; Koos grades III and IV) requires a balance between the maximum extent of resection and the best functional preservation. The primary objective of this study was to determine the volumetric threshold of the VS residual tumor at risk of progression after incomplete resection. The secondary objective was to identify other risk factors of regrowth after incomplete resection. METHODS: This retrospective study included patients who underwent incomplete resection of sporadic VS at a single center from January 2008 to December 2018. The inclusion criteria were: adult age, large single sporadic VS, incomplete resection, and follow-up of > 5 years. Quantitative 3D volumetry was assessed on pre- and postoperative contrast-enhanced T1-weighted MRI using semiautomated segmentation. The volumetric criteria for residual tumor were < 250 mm3 for near-total resection (NTR) and < 2 cm3 for subtotal resection (STR). Univariate and multivariate logistic regression analyses were performed to assess predictors of regrowth after incomplete resection. A residual volume cutoff for risk of regrowth was determined using the Youden index via area under the curve analysis. RESULTS: The cohort included 119 patients (60 female, median age 58 years) who were categorized into 3 subgroups based on the residual VS according to 3D volumetry: NTR, STR, and partial resection (PR). NTR achieved the best long-term tumor control. Kaplan-Meier progression-free survival rates at 2, 5, and 10 years were 98%, 97%, and 95% for the NTR group; 69%, 56%, and 56% for the STR group; and 20%, 0%, and 0% for the PR group, respectively (p < 0.0001). The cutoff residual volume at risk of growth was 200 mm3, with sensitivity of 95% (95% CI 74%-99%) and specificity of 77% (95% CI 68%-85%, p < 0.001). Moreover, good facial nerve outcomes (House-Brackmann grades I and II) were best achieved with PR (100%), followed by STR (96%) and NTR (90%). In the univariate analysis, the risk factors for regrowth of residual tumor were cystic morphology, residual volume, and residual location (internal auditory canal, cisternal segment, and brainstem combined). The multivariate model identified the volume and location of residual as risk factors (p < 0.0001). CONCLUSIONS: These findings suggest that limited NTR (< 250 mm3) offered an excellent compromise, with long-term tumor control comparable to that of radical resection while preserving superior functional preservation. The authors hope to stimulate discussion toward a unified volumetrically established classification of incomplete resections, allowing for cooperation in future multicenter studies.