Prediction of behavioral MCL using electrophysiological responses in children using MED-EL implant.

Abstract

The present study aimed to correlate the various electrophysiological tests of ECAP, EABR and ESRT with programming parameters. If there is a correlation between them, fitting formulae to be derived to predict programming parameters. Further this fitting formula was validated on a clinical population. 22 children between age range of 5-12 years using MED-EL implant participated study. Electrophysiological tests of Electrically evoked compound Action Potential (ECAP), Electrical Evoked Stapedial Reflex Threshold (ESRT) and Electrically Evoked Auditory Brain Stem Responses (EABR) were measured on electrodes no 1,4, 8, and 11. Based on Pearson correlation analysis, there was a moderate correlation observed between each of electrophysiological tests with MCL level. Fitting formulae of ECAP with either ESRT or EABR were found to be accurately predict the MCL level. These fitting formulae were clinically validated on 6 children using Sonata implant with OPUS 2 processor. Two new programs with MCL were predicted using combination of ECAP with EABR and ECAP with ESRT as parameters in the fitting formulae. These programs were given to the participants to use for two weeks. Predicted MCLs were found to slightly higher (about 2qu to 5qu) than original MCL level. Reliability analysis indicated that the formulae predicted MCL with good accuracy. Speech perception and sound field thresholds were measured in the participants' Everyday program and two predicted programs. When ECAP & EABR were the parameters, the predicted program had improved audibility as reflected in sound field thresholds as compared to those obtained with other two programs. Based on Freidman test, the results indicated that significantly lower thresholds were found for both ECAP & EABR, or ECAP & ESRT based programs when compared to Everyday program. However, speech perception scores were not significantly different among the program as per Freidman test. Thus, both the fitting models were clinically validated. The findings imply that it is not always advisable to run all three electrophysiological testing to predict the MCL levels in clinical population. It would save lot of time to run just two tests to predict the MCL in difficult to test population.