Does the Presence of Spontaneous Components Affect the Reliability of Contralateral Suppression of Evoked Otoacoustic Emissions?

Abstract

Objectives: The function of the medial olivocochlear system can be evaluated by measuring the suppression of otoacoustic emissions (OAEs) by contralateral stimulation. One of the obstacles preventing the clinical use of the OAE suppression is that it has considerable variability across subjects. One feature that tends to differentiate subjects is the presence or absence of spontaneous OAEs (SOAEs). The purpose of the present study was to investigate the reliability of contralateral suppression of transiently evoked OAEs (TEOAEs) measured using a commercial device in ears with and without SOAEs. Design: OAEs were recorded in a group of 60 women with normal hearing. TEOAEs were recorded with a linear protocol (identical stimuli), a constant stimulus level of 65 dB peSPL, and contralateral broadband noise (60 dB SPL) as a suppressor. Each recording session consisted of three measurements: the first two were made consecutively without taking out the probe (the “no refit” condition); the third measurement was made after taking out and refitting the probe (a “refit” condition). Global (for the whole signal) and half-octave band values of TEOAE response levels, signal-to-noise ratios (SNRs), raw dB TEOAE suppression, and normalized TEOAE suppression, and latency were investigated. Each subject was tested for the presence of SOAEs using the synchronized SOAE (SSOAE) technique. Reliability was evaluated by calculating the intraclass correlation coefficient, standard error of measurement (SEM) and minimum detectable change. Results: The TEOAE suppression was higher in ears with SSOAEs in terms of normalized percentages. However, when calculated in terms of decibels, the effect was not significant. The reliability of the TEOAE suppression as assessed by SEM was similar for ears with and without SSOAEs. The SEM for the whole dataset (with and without SSOAEs) was 0.08 dB for the no-refit condition and 0.13 dB for the refit condition (equivalent to 1.6% and 2.2%, respectively). SEMs were higher for half-octave bands than for global values. TEOAE SNRs were higher in ears with SSOAEs. Conclusions: The effect of SSOAEs on reliability of the TEOAE suppression remains complicated. On the one hand, we found that higher SNRs generally provide lower variability of calculated suppressions, and that the presence of SSOAEs favors high SNRs. On the other hand, reliability estimates were not much different between ears with and without SSOAEs. Therefore, in a clinical setting, the presence of SOAEs does not seem to have an effect on suppression measures, at least when testing involves measuring global or half-octave band response levels.

The Reliability of Contralateral Suppression of Otoacoustic Emissions Is Greater in Women than in Men

“…The values of raw MOCR, as measured in decibels, are similar to those in previous studies on the same equipment (e.g., [14,17,24]). The present study also confirms the results of Stuart and Kerls [14] regarding the lack of difference in MOCR between men and women in terms of raw decibel.…”

“…This may simply reflect the smaller size of the dataset. It was similarly not present in an earlier study of small sample size [ 20 ], and was only discovered after larger datasets were employed [ 15 , 17 ]. For women, the results were similar to those of [ 17 ], with the MOCR being higher for ears with SSOAEs.…”

“…SEM was calculated as , where STD is the standard deviation of the combined test and retest MOCR estimate. These measures have been used in previous studies of MOCR reliability [ 16 , 17 , 30 ]. When the SEM value grows, the reliability is poorer.…”

“…Stuart and Kerls (2018) [ 14 ] reported that there was no laterality or gender effect in terms of raw decibel change to the MOCR, but they did see a gender effect in terms of percentage change, with men having a slightly higher effect. Some recent studies have also shown that the presence of SOAEs affects MOCRs when they are measured using evoked OAEs [ 15 , 16 , 17 ]. It has also been shown that the MOCR effect appears larger when using SOAEs rather than evoked OAEs, and that the presence of SOAEs increases MOCR detectability when evoked OAEs are studied [ 18 , 19 ].…”

“…Importantly, there seems to be little information on whether there are gender differences in MOCR reliability. Some studies have evaluated groups consisting of men and women combined [ 20 , 21 , 22 , 23 , 24 , 25 ], while others have considered only women [ 17 , 26 ]. Of all these studies, only Stuart and Cobb [ 21 ] investigated the effect of gender.…”