Is There a Relation between Speech Recognition and the Position of Cochlear Implant Electrodes in the Inner Ear?

The anatomical position of the Nucleus-22 electrode array in the cochleae of 33 postlinguistically deaf adults has been estimated from spiral CT images.[1] [2] For 13 of these adults, the position of individual electrodes also has been derived from 2D radiographs[3] [4] with projection of the CT-derived array onto the radiographs.[5] Array insertion depth of the most apical electrode expressed as percentage of total cochlear canal length was significantly correlated with mean NU-6 word score for the 26 patients who had used the spectral peak (SPEAK) strategy for a year at the time of testing (r=0.452; p=0.020). This correlation is modest because the site of electrical stimulation in the cochlea is only one factor contributing to word recognition. It is well known that age at implantation and duration of deafness have a negative effect and number of years implant use has a positive effect on word recognition in postlinguistically deaf adults.[6] [7] Analyses of patients' scores suggest that etiologies that adversely affect the number of remaining auditory neurons (e.g., bacterial labyrinthitis, syphilis) or abnormal electrical current paths within the cochlea (e.g., otosclerosis) are associated with scores below the regression line. A number of patients with less than full insertion (i.e., between 52 and 64% of the total cochlear length) had excellent verbal skills and word scores substantially higher than the regression line. Because there was less than full insertion, the frequency bands of incoming sound assigned to the active electrodes were substantially lower than the estimated characteristic frequency of nearby neurons for these patients. For one patient, the frequency band assigned to the most apical electrode, 22 (paired with electrode 20), was 120 to 280 Hz in the speech processor program; the characteristic frequency estimated at this most apical electrode position was 1485 Hz. It is hypothesized that these patients' high scores were associated with their above average verbal skills, a year or more of implant experience, above average survival of auditory neurons, and sufficient central auditory nervous system plasticity to adapt to the shift in frequency of incoming sound to a higher pitch percept from stimulation with the implant than would normally be perceived acoustically. Documented improvement in these patients' word scores obtained at 3 months, 6 months, and yearly intervals thereafter with the multipeak (MPEAK) and SPEAK strategies lends support for the hypothesized neural plasticity.

REFERENCES

• 1 Ketten D R, Skinner M W, Wang G. In vivo measures of cochlear length and insertion depth of nucleus cochlear implant electrode arrays.  Ann Otol Rhinol Laryngol . 1998;  107(Suppl 175) 1-16
• 2 Skinner M W, Ketten D R, Holden L K. CT derived estimation of cochlear morphology and electrode array position in relation to word recognition in Nucleus-22 recipients.  (in press) J Assoc Res Otolaryn.
• 3 Cohen L T, Xu J, Xu S A, Clark G M. Improved and simplified methods for specifying positions of the electrode bands of a cochlear implant array.  Am J Otol . 1996;  17 859-865
• 4 Skinner M W, Cohen L T, Ketten D R. 2D radiograph and CT derived estimates of the position of cochlear implant electrodes in Nucleus-22 recipients.  (submitted for publication).
• 5 Whiting B R, Bae K T, Skinner M W. Three dimensional localization of cochlear implants by co-registration of CT and conventional radiographs.  Radiology . 2001;  221 543-549
• 6 Blamey P, Arndt P, Bergeron F. Factors affecting auditory performance of postlinguistically deaf adults using cochlear implants.  Audiol Neurootol . 1996;  1 293-306
• 7 Gantz B J, Woodworth G G, Abbas P J, Knutson J F, Tyler R S. Multivariate predictors of audiological success with multichannel cochlear implants.  Ann Otol Rhinol Laryngol . 1993;  102 909-916