Effect of exceeding compliance voltage on speech perception in cochlear implants.

OBJECTIVES

In cochlear implants, the maximum current I (Amperes) that can be delivered on a cochlear implant electrode is determined by V = I * R, where V = compliance voltage (Volts) and R = electrode impedance (Ohms). Generally, electrode impedances are measured during each clinical visit and are used to set electrical stimulation parameters in cochlear implants. However, electrode impedances can rise during the course of cochlear implant use and lead to electrical stimulation voltage requirements exceeding the maximum compliance voltage of the medical device. Electric stimulation requirements that exceed the compliance voltage lead to clipping of the biphasic electrical pulse (current going into the cochlea) and are known to adversely affect cochlear implant outcomes.

DESIGN

Thirteen (11 unilateral and 2 bilateral) Advanced Bionics cochlear implant patients with a HiRes 90k™ cochlear implant participated in this study. Speech perception scores were measured using the patient's baseline clinical program with the most comfortable loudness levels (M-levels) and the following four test programs: (1) stimulation clipped at 15% below clinical M-levels (15%C) (2) stimulation clipped at 30% below clinical M-levels (30%C) (3) M-levels decreased by 15% (15%M) and (4) M-levels decreased by 30% (30%C). Speech perception scores were measured using AzBio sentences presented at 60 dB SPL in quiet and in the presence of multi-talker babble (+10 dB SNR).

RESULTS

Relative to the clinical baseline program, speech perception scores with the four test programs decreased in both quiet and noisy listening conditions. In quiet, speech perception scores measured with the 30%M and 30%C programs were significantly (p < 0.001) poorer than the baseline program. No significant differences in speech perception scores were measured between the baseline and the 15%C or 15%M programs. In the noisy listening condition, speech perception scores were significantly poorer than the baseline program for the 15%C (p = 0.008), 30%C (p < 0.001), and 30%M (p < 0.001) programs. No significant differences in speech perception scores were obtained between the baseline and the 15%M program in the noisy listening condition. Speech perception scores measured with the 30%C program were significantly (p < 0.001) poorer than those with the 30%M program, suggesting that clipping was more detrimental than reducing electrical stimulation levels.

CONCLUSION

Small amounts (15%) of clipping can significantly decrease speech perception in the presence of background noise. Large amounts (30%) of both clipping and M-level reduction may lead to significantly poorer speech perception in quiet and in background noise. The decrease in speech perception scores can most likely be attributed to reduced volume and poorer spectro-temporal representation. Therefore, it is important to establish comfortably loud electrical stimulation levels without exceeding the compliance voltage to maximize cochlear implant outcomes.