Early Cochlear Implant Outcomes Predict Long-Term Speech Recognition

Area of Science:

• Audiology
• Otolaryngology
• Biostatistics

Background:

• Early identification of cochlear implant (CI) users at risk for poor audiological outcomes is crucial for timely intervention.
• Optimizing long-term CI benefit requires predicting speech recognition performance early in the post-implantation period.

Purpose of the Study:

• To develop and evaluate logistic regression models for predicting 12-month speech recognition performance in CI users.
• To assess the predictive value of early speech recognition scores and patient factors for long-term CI outcomes.

Main Methods:

• A retrospective cohort study of 625 postlingually deafened adult CI users.
• Logistic regressions were used to model the likelihood of significant improvement in consonant-nucleus-consonant (CNC) word scores at 12 months.
• Models incorporated baseline demographics, duration of deafness, pre-CI CNC scores, and early post-activation CNC improvement at 1 or 3 months.

Main Results:

• 82% of patients showed improved CNC scores at 12 months post-implantation.
• Early improvement (1 or 3 months) was a strong predictor of 12-month outcomes (ORs 45.72 and 22.22, respectively).
• The model using 3-month data achieved the highest predictive accuracy (AUC = 0.92) in the validation cohort.

Conclusions:

• Models incorporating early CNC score improvements demonstrate strong predictive discrimination for 12-month outcomes.
• A regression model utilizing 3-month post-CI speech recognition data can provide reliable early estimates of long-term performance.
• This tool can assist clinicians in identifying patients who may benefit from early intervention to optimize CI outcomes.