The relationship between AutoNRT thresholds and subjective programming levels revisited.

Objectives: The aim of this study wasto systemically examine the relationship between AutoNRT thresholds and subjective programming levels by controlling for individual subject variation, and further, to propose an alternative way to calculate the AutoNRT threshold profile for programming purposes.

Methods: The study was a prospective observational study. AutoNRT thresholds, T- and C-levels were recorded at six and twelve months after activation. All subjects had received a CI24RE implant.

Results: Forty-one adults participated in the study. A linear mixed-effects model analysis of the relationship between the AutoNRT thresholds and the T-and C-levels resulted in a large unexplainable variation between subjects, the standard deviation was between 15.42 and 18.89 CL. However, the study showed that the profiles for the T- and C-levels could be predicted with an acceptable accuracy from AutoNRT. A model for calculating profiles from AutoNRT thresholds based on linear regression resulted in the lowest deviation from the subjective programming thresholds, with somewhat better results for C-levels than T-levels.

Conclusion: The results from this study showed that the predictability was reasonable when calculating the profiles for the T- and C-levels based on AutoNRT. And, further, that AutoNRT thresholds can be used when programming cochlear implants, to make global adjustments without predicting actual T- or C-levels.