Tinnitus and cochlear implantation.

Tinnitus is a common experience, but there is very marked heterogeneity of aetiology, perception and the extent of distress among individuals who experience tinnitus. In view of this, a modern approach to tinnitus should consider homogeneous groups of individuals. This review considers tinnitus experiences in patients undergoing cochlear implantation, this being of interest because the prevalence of tinnitus in this patient group prior to surgery may shed some light on the link between cochlear dysfunction and tinnitus Second, any change in tinnitus experience as a result of electrode placement surgery or cochlear implant activation has relevance for patient counselling and informed consent. Finally, in recent tinnitus retraining therapy literature there has been the suggestion that unilateral sound therapy for tinnitus patients may set up an unhelpful asymmetry of input to the auditory system, with possible exacerbation of contralateral tinnitus. Unilateral cochlear implant use represents the most dramatic asymmetry possible and hence is a test of that hypothesis. Relevant papers (n = 32) were identified from literature databases. The standard of reporting tinnitus results was inconsistent. Tinnitus is experienced by up to 86% of adult cochlear implant candidates, but is not universal and is only troublesome in a small proportion (reported as 27% in one study). Electrode insertion may induce tinnitus in a small (up to 4%) number of patients, but this is rare. Cochlear implant device use is associated with reduction of tinnitus intensity and awareness in up to 86% of patients, and rarely with exacerbation (up to 9%). There are some indications in the literature that the more complex the simulation strategy, the larger that effect. Specifically, unilateral cochlear implant use was generally associated with reduction of contralateral tinnitus (in up to 67% of individuals) rather than exacerbation, and so the assertion that unilateral sound therapy for tinnitus is contraindicated is not proven.