Abstract
OBJECTIVES: To estimate the prevalence of cochlear-facial nerve dehiscence (CFD) in cochlear implant (CI) recipients, measure cochlear-facial nerve partition width (CFPW), and analyze the relationship of CFD and CFPW for the presence or absence of facial nerve stimulation (FNS). METHODS: A retrospective review of 314 ears from 294 adult CI recipients was conducted. Two blinded reviewers measured CFPW and identified cases with CFD from the preoperative or postoperative high-resolution computed tomography (CT) scan. Analyses included the associations of CFPW and CFD with FNS, defined as facial movement elicited by CI stimulation. Audiologic and surgical data were reviewed for electrode characteristics and management of FNS. RESULTS: Reviewers had excellent inter-rater reliability for CFPW measurements (intraclass correlation coefficient = 0.974). CFPWs were significantly smaller in ears with FNS (median: 0.40 mm) compared to those without FNS (median: 0.48 mm; p = 0.003). The prevalence of CFD was 3.2% (n = 10) and 70% of cases with CFD experienced FNS (relative risk [RR] = 14.2, 95% CI: 6.88-25.50, p < 0.001). Cases with CFD demonstrated a higher rate of FNS regardless of electrode type (lateral wall versus precurved). CONCLUSION: CFD is a risk factor of FNS in CI recipients. Preoperative high-resolution CT can identify patients with CFD that are at high risk for FNS; however, use of a precurved electrode array may not reduce the risk for FNS in this population. Investigation is needed of interventions to mitigate FNS risk in patients with CFD. LEVEL OF EVIDENCE: 3.