Cochlear-facial dehiscence - the most common cause of facial nerve stimulation from a cochlear implant? A case-control study.

Objectives: To investigate the prevalence of cochlear-facial dehiscence (CFD) and other radiographical pathologies in ears with facial nerve stimulation (FNS) from a cochlear implant (CI). Methods: Retrospective case-control study of 27 patients with CI and FNS on either ear (study group) and 27 patients without FNS, matched for age, sex and type of electrode array (control group). Preoperative CT scans of all 108 ears were re-evaluated. Subanalyses included comparisons between the study and control groups and associations between FNS and radiographic pathologies. Results: CFDs were detected in 20 of 54 ears (37%) in the study group and in 3 of 54 ears (6%) in the control group (P < 0.001). The corresponding numbers of otosclerosis were 10 (18%) and 0 (P = 0.011) and of developmental anomalies 16 (30%) and 8 (15%) (not significant). FNS was present in 33 ears in the study group, of which 14 (42%) had a CFD. FNS was absent in six ears with CFD and CI, four of which contralateral to an ear with FNS. Eight of 14 ears with FNS and CFD had a lateral electrode array and six had a perimodiolar electrode array. We found no association between the presence of CFD and stimulation thresholds for FNS. The adjusted odds ratio for developing FNS in the presence of a CFD was 9.9 (95% CI 2.7-36.0). Conclusions: CFD was the most common radiographic pathology in ears with FNS, with a 10-fold increased risk of FNS. To avoid CI-related FNS, preoperative CT scan and awareness of typical dehiscence symptoms are strongly recommended.