A Predictive Model for Cochlear Implant Outcome in Children with Cochlear Nerve Deficiency

The outcome of cochlear implantation (CI) in patients with cochlear nerve deficiency (CND) is variable, resulting in a wide range of speech perception performance, from degrees of environmental sound perception to conversation without lip-reading. Twenty-five cochlear implantees with CND were enrolled retrospectively to determine the factors correlated with CI outcome in patients with CND and to develop a predictive model for CI outcome. CI outcome was evaluated using the Categories of Auditory Performance (CAP) score at 2 years after CI. Patients with negative auditory brainstem response (ABR) showed a significantly lower CAP score than those with positive ABR (2.5 +/- 1.7, 4.8 +/- 0.7; p = 0.001). The area ratio of vestibulocochlear nerve (VCN) to facial nerve (FN) at the cerebellopontine angle on magnetic resonance images was positively correlated with CI outcome (p < 0.001). With multiple regression analysis, a predictive equation accounting for 66% of variance of CAP score at 2 years after CI was deduced: CAP score = 0.7 + 1.9 * (ABR) + 1.2 * (VCN/FN). We found that preoperative ABR and area ratio of VCN to FN at the cerebellopontine angle could predict CI outcome in patients with CND. Preoperative counselling based on our predictive model might be helpful to determine treatment modality for auditory rehabilitation and which ear to implant.