When to Refer a Hearing-impaired Patient for a Cochlear Implant Evaluation

Jacob B. Hunter, Anthony M. Tolisano

Research output: Contribution to journalArticlepeer-review

Abstract

OBJECTIVES: To explore the predictive value of utilizing routine audiometry to best determine cochlear implant (CI) candidacy using AzBio sentences. METHODS: A retrospective chart review was performed between 2011 and 2018 for 206 adult patients who underwent CI evaluation assessed with AzBio sentences. Better hearing ear word recognition score (WRS) using Northwestern University-6 word lists presented at decibel hearing level from a standard audiogram was used to determine when best to refer a patient for CI evaluation. Predicted AzBio scores from multivariate regression models were calculated and compared with the actual CI candidacy to assess accuracy of the regression models. RESULTS: Race, marital status, hearing aid type, better hearing ear WRS, and HL were all independently and significantly associated with AzBio testing in quiet on univariate analyses. Better hearing ear WRS and better hearing ear decibel hearing level predicted AzBio Quiet on multivariate regression analysis. For AzBio +10 dB signal-to-noise ratio (SNR), sex, and better hearing ear WRS each significantly predicted speech perception testing. Predicted CI candidacy was based on AzBio sentence testing of ≤60% for the ease of statistical analysis. Regression models for AzBio sentence testing in quiet and +10 dB SNR agreed with the actual testing most of the time (85.0 and 87.9%, respectively). A generalized linear model was built for both AzBio testing in quiet and +10 dB SNR. CONCLUSION: A WRS of <60% in the better hearing ear derived from a routine audiogram will identify 83.1% of CI candidates while appropriately excluding 63.8% of patients.

ASJC Scopus subject areas

  • Otorhinolaryngology
  • Sensory Systems
  • Clinical Neurology

Fingerprint

Dive into the research topics of 'When to Refer a Hearing-impaired Patient for a Cochlear Implant Evaluation'. Together they form a unique fingerprint.

Cite this