Long-Term Outcomes of Subtotal Resection for Large Vestibular Schwannomas: Results From the Acoustic Neuroma Subtotal Resection Study, Part I-Facial Nerve Outcomes and Degree of Resection.

Abstract

Complete removal of large vestibular schwannomas (VS) poses a significant risk of facial nerve [cranial nerve VII (CNVII)] paralysis. However, less-than-total removal may be associated with an increased risk of regrowth, which could require further treatment with additional risks. The objective was to assess long-term outcomes of postoperative CNVII function in relation to the extent of resection in patients who underwent removal of large VS. A multicenter, prospective, nonrandomized cohort study of patients with large VS (≥2.5 cm) who underwent gross total resection (GTR), near-total resection (NTR), or subtotal resection (STR). GTR was defined as no visible remnant, NTR as a remnant <0.5 cm3 on postoperative MRI, and STR as any larger remnant. CNVII function was evaluated using binary variables: House-Brackmann grades I-II represented excellent-to-good function, whereas House-Brackmann grades III-VI represented fair-to-poor function. A total of 126 patients were included with a mean (SD) follow-up of 60 (±31) months. The mean preoperative tumor diameter was 3.3 ± 0.7 cm. Thirty-five received GTR, 39 received NTR, and 52 received STR based on postoperative MRI. Smaller preoperative tumor size (P = .03) and volume (P = .03) were associated with good immediate CNVII function. Neither immediate nor late facial nerve function was related to the extent of resection. STR was associated with a 3-fold higher surgical failure rate compared with GTR (P = .02), and patients who experienced surgical failure requiring stereotactic radiosurgery or revision surgery had increased odds of fair-to-poor long-term CNVII function (odds ratio 4.6, 95% CI, 1.7-12.5, P = .002). The degree of resection did not predict immediate or late good CNVII function. Surgical failure was associated with worse long-term function. GTR or NTR should be attempted whenever feasible to reduce the likelihood of repeat surgery and poor facial nerve outcomes.