The relationship between electrically evoked compound action potential and speech perception: a study in cochlear implant users with short electrode array

Abstract

Objectives: To determine the extent to which electrically evoked compound action potential (ECAP) measurements were related with speech perception performance in implant users with a short electrode array and to investigate the relationship between ECAP measures and performance according to specific devices.

Design: Prospective study.

Setting: Tertiary referral center.

Patients: Seventeen Hybrid cochlear implant users were tested in this study. Subjects were divided into 2 groups: 8 using the Nucleus Hybrid M and 9 using the Nucleus Hybrid RE. In addition, 21 Nucleus Freedom long electrode implant (CI24RE) users also were tested to compare with the results of the old device (CI24M).

Main outcome measures: ECAP growth functions were recorded using either an interphase gap (IPG) of 8 or 45 mus. We then calculated the slope of the growth function and changes in sensitivity with IPG. For each subject, these measures were compared with performance on tests of word recognition.

Results: The changes in sensitivity using 2 IPGs showed no correlation with the results of word recognition test in Hybrid cochlear implant users. In contrast, relatively strong correlations have been found between the slope of ECAP growth functions and performance on word recognition test. Additionally, when we separate the results of Hybrid M and RE, the slopes of ECAP growth functions from only Hybrid RE CI recipients were significantly correlated with speech performance. The slopes of ECAP growth function in CI24RE users with long electrode also were significantly correlated with performance. However, comparing between 2 independent correlations in RE devices, correlation was higher in Hybrid RE group.

Conclusion: The results presented in this article support the view that slope of the ECAP growth can show significant correlation to performance with a cochlear implant. Furthermore, these results suggest that the strength of the correlation may be related to the specific device. These results suggest that ECAP measures may be useful in developing a test to predict outcomes with the implant.

Figures

Figure 1

The change in sensitivity of the ECAP using two IPGs of 8 and 45 µs is shown from a single implant recipient (H-RE5) in response to stimulation on electrode 2. The change in sensitivity was defined as current change needed to equalize ECAP amplitude when IPG was increased from 8 to 45 µs. ECAP amplitudes were normalized to the largest amplitude of the 8 µs IPG growth function. The measures of shift in sensitivity at 0.2, 0.4, 0.6, and 0.8 normalized amplitude were used to calculate the mean change in sensitivity.

Figure 2

The growth functions using 8 µs IPG of electrode 4 measured from the 17 subjects. The left column shows the Hybrid RE growth functions and the right panel shows the Hybrid M growth functions. The slope of growth function was calculated using linear regression. The slope of each growth function in µV/CL is indicated in parentheses to the right of each subject number in the legend.

Figure 3

The relationship between the slopes of the ECAP amplitude intensity functions and the speech perception scores is shown for subjects with Hybrid RE and M CI. Left and right columns show correlations between slope of growth function and speech understanding in quiet (CNC words), and noise (BKB-SIN test), respectively. Open circles represent Hybrid M, and filled circles indicate Hybrid RE. Only Hybrid RE CI recipients show a significant correlation between the slope of the ECAP growth function and speech performance_(Left column : r=0.90, p

Figure 4

The changes in sensitivity with two IPGs obtained for subjects with Hybrid CIs are plotted as function of the scores of CNC words (Left column) and the scores of BKB-SIN test (Right column). This measure showed no correlation with results of speech performance (CNC test: r=0.02, p > 0.05; BKB-SIN test: r=−0.02, p > 0.05).

Figure 5

The relationship between the slope of the growth function and performance is shown for CI24M and CI24RE CI recipients. The left column plots the CNC word scores as a function of the slopes of the ECAP growth functions for 24 ears with CI24RE implants. The right column plots data from 37 CI24M CI recipients (Brown et al., 1999). In contrast to CI24M users, CI24RE recipients show a significant correlation between the slope of the ECAP growth function and CNC word scores (r=0.47, p