Introduction: Acoustic neuromas (ANs) are typically treated with microsurgery or stereotactic radiosurgery (SRS). Our tertiary care center offers teams experienced in both surgery (neurosurgery, otology) and radiosurgery (neurosurgery, radiation oncology). AN cases are discussed at a multidisciplinary skull base conferences. Although available evidence suggests that SRS is associated with fewer periprocedural complications and better facial nerve outcomes for small tumors, patients may still be interested in surgical resection. Given that both options are available at our institution and patients are typically discussed with all treating specialties, we sought to identify patient factors associated with each treatment and compare outcomes in contemporaneous cohorts (matched and unmatched) of microsurgical and SRS patients.
Methods: Retrospective chart review of consecutive acoustic neuromas treated from 2011 to 2016 was performed. Patients with acoustic neuroma ≤ 3 cm and undergoing surgery (a single neurosurgeon [J.J.M.] with the help of neurootologist) or SRS were included. Multivariate logistic regression was performed to identify factors associated with treatment modality. A matched cohort of surgical and SRS patients was created by propensity-matching on age, tumor size, and pretreatment growth, identifying 39 matches of 78 patients. Clinical outcomes were compared between matched surgical and SRS cohorts. Multivariate propensity matching performed with logistic regression and greedy-match model with 0.75 caliper. Statistics performed with S.A.S 9.4 (Cary, North Carolina, United States).
Results: A total of 140 ANs (<3 cm) were identified, 58 SRS and 82 OR. Average follow-up was similar between both cohorts (SRS 22 vs. OR 21 months); as were rates of prior treatment. The SRS cohort had fewer females (36 vs. 60%, p = 0.006), older patients (59 vs. 51 years, p = 0.0004), smaller tumors (1.7 vs. 2.1 cm, p < 0.0001), and were more likely to present with vestibular dysfunction (54 vs. 26%, p = 0.0006) and documented growth (43 vs. 25%, p = 0.029). Subjective hearing improvement was reported more frequently by SRS patients (77 vs. 18.9%, p < 0.001) and tinnitus improvement was more common in surgical patients (69 vs. 42%, p = 0.043). Posttreatment tinnitus developed more commonly in SRS patients (21 vs. 2%, p = 0.015), while facial nerve palsy was more common in surgical patients (HB > 2; 0 vs, 12.4% p = 0.004). Tumor progression was similar between both cohorts.
Matched cohorts were statistically similar in average age, tumor size, length of follow-up, and documented growth ([Fig. 1]). No difference in tumor progression between matched cohorts was observed (p = 0.43). Patients who presented with hearing loss were more likely to report improvements after SRS compared with surgical patients (76 vs. 14%, p < 0.0001). No significant differences in tinnitus improvement (p = 0.1), facial nerve palsy (p = 0.11), or tumor progression (p = 0.43) were observed.
Conclusion: Both patients and providers have biases regarding treatment modality for acoustic neuromas. In our contemporaneous cohort, older males were more likely to undergo SRS. SRS patients were more likely to have smaller tumors with documented growth and present with vestibular dysfunction. These differences may be due to patient or institutional biases. While outcomes in our matched cohort were not significantly different the length of follow-up was limited.
