The aim of this retrospective cohort study was to assess speech perception outcomes of second-side cochlear implants (CI2) relative to first-side implants (CI1) in 160 participants who received their CI1 as a child. The predictive factors of CI2 speech perception outcomes were investigated. In addition, CI2 device use predictive models were assessed using the categorical variable of participant’s decision to use CI2 for a minimum of 5 years after surgery. Findings from a prospective study that evaluated the bilateral benefit for speech recognition in noise in a participant subgroup (n = 29) are also presented.
Participants received CI2 between 2003 and 2009 (and CI1 between 1988 and 2008), and were observed from surgery to a minimum of 5 years after sequential surgery. Group A (n = 110) comprised prelingually deaf children (severe to profound) with no or little acquired oral language before implantation, while group B (n = 50) comprised prelingually deaf children with acquired language before implantation, in addition to perilingually and postlingually deaf children. Speech perception outcomes included the monosyllable test score or the closed-set Early Speech Perception test score if the monosyllable test was too difficult. To evaluate bilateral benefit for speech recognition in noise, participants were tested with the Hearing in Noise test in bilateral and “best CI” test conditions with noise from the front and noise from either side. Bilateral advantage was calculated by subtracting the Hearing in Noise test speech reception thresholds in noise obtained in the bilateral listening mode from those obtained in the unilateral “best CI” mode.
On average, CI1 speech perception was 28% better than CI2 performance in group A, the same difference was 20% in group B. A small bilateral speech perception benefit of using CI2 was measured, 3% in group A and 7% in group B. Longer interimplant interval predicted poorer CI2 speech perception in group A, but only for those who did not use a hearing aid in the interimplant interval in group B. At least 5 years after surgery, 25% of group A and 10% of group B did not use CI2. In group A, prediction factors for nonuse of CI2 were longer interimplant intervals or CI2 age. Large difference in speech perception between the two sides was a predictor for CI2 nonuse in both groups. Bilateral advantage for speech recognition in noise was mainly obtained for the condition with noise near the “best CI”; the addition of a second CI offered a new head shadow benefit. A small mean disadvantage was measured when the noise was located opposite to the “best CI.” However, the latter was not significant.
Generally, in both groups, if CI2 did not become comparable with CI1, participants were more likely to choose not to use CI2 after some time. In group A, increased interimplant intervals predicted poorer CI2 speech perception results and increased the risk of not using CI2 at a later date. Bilateral benefit was mainly obtained when noise was opposite to CI2, introducing a new head shadow benefit.