Speech Perception Outcomes in Children With Single-Sided Deafness Receiving Unilateral Cochlear Implantation Compared to Bimodal Device Users.

OBJECTIVES: This study aimed to (1) assess the effects of residual hearing in the acoustic hearing (AH) ear on speech perception outcomes with ongoing unilateral cochlear implant (CI) use in children and (2) investigate factors related to hearing history that might affect these outcomes. Children with single-sided deafness (SSD) were compared to peers with hearing loss (HL) in the AH ear contralateral to their CI. DESIGN: In total, 185 children who received a unilateral CI after limited durations of severe to profound HL {median (interquartile range [IQR]) = 1.9 (1.0 to 2.9) years} were included in this study. Of these, 43 (23.2%) had SSD and 142 (76.8%) used a contralateral hearing aid for mild to moderately-severe HL (PTA median [IQR] = 48.8 [42.5 to 63.8] dB HL, n = 39 [21.1%]) or severe to profound HL (PTA median [IQR] = 80.0 [72.1 to 85.5] dB HL, n = 103 [55.7%]) (bimodal device users). Median [IQR] age at implantation was 5.3 [3.1 to 10.6] years. Speech perception was measured in quiet (n = 182 children, average of 2.8 test sessions/child) and in co-located speech-weighted noise (+10 dB SNR) (n = 109 children, average of 2.0 test sessions/child) using word recognition tests. Spatial release of masking was measured to assess spatial separation benefits (n = 78 children, average of 1.6 tests/child). RESULTS: Speech perception in the AH ear was better in children with SSD than bimodal device users in quiet ( p < 0.001), but group differences were smaller in noise, reflecting significantly reduced scores in the SSD group in noise compared to quiet ( p < 0.001, mean [SD] difference = 27.3 [2.6] rationalized arcsine units [RAU]). Speech perception scores when using the CI-alone were similar between the three groups in quiet and noise ( p > 0.05). In quiet, speech perception was asymmetric with better scores for the AH ear for children with SSD (mean [SD] = 30.7 [29.1] RAU) and for the CI ear for bimodal device users (mean [±SD] = -5.25 [±21.7] RAU for mild to moderately-severe and -27.2 [±30.9] RAU for severe to profound HL), and the bilateral benefit was the smallest for children with SSD ( p < 0.01). In noise, however, these group differences were only significant between children with SSD and those with a severe to profound HL in the AH ear ( p = 0.02 for interaural asymmetry, p = 0.03 for bilateral benefit). In addition, asymmetry toward the AH ear decreased and bilateral benefit increased in quiet with shorter durations of deprivation ( p < 0.01) and longer CI experience ( p = 0.01), but these predictors had no effect in noise. Last, aural preference, measured by spatial release of masking, was consistent with interaural asymmetry in speech perception in quiet ( p = 0.01). Aural preference toward the AH ear was larger in the SSD than bimodal devices groups ( p < 0.01) and for children with post-lingual onset of deafness using a left rather than right CI ( p = 0.04). CONCLUSIONS: Benefits of cochlear implantation are evident both for children with SSD and bimodal device users, and scores were not affected by the degree of residual hearing in noise. These benefits of bilateral hearing are most pronounced for children with shorter durations of deprivation in either ear.