Clinical Utilisation of High-frequency DPOAEs

Abstract

The value of assessing auditory function at frequencies above 8kHz to detect age-related changes and ototoxic damage in the cochlea is well established but not commonplace. Physiological changes in the auditory periphery due to age and ototoxicity are initially evident, and most prominent, at frequencies above 8kHz [1]. The most well investigated use of hearing thresholds and otoacoustic emissions above 8kHz is in monitoring auditory function in patients undergoing chemotherapy [2]. Ototoxic changes in hearing thresholds at frequencies between 10–14kHz prior to the manifestation of any changes at lower frequencies have been consistently documented in these patients. Age-related changes in hearing also appear at frequencies above 8kHz prior to any observable changes at regular audiometric frequencies [3]. The value of using hearing thresholds at frequencies above 8kHz to detect noise-induced hearing loss is debated in the literature with some reports of hearing thresholds at frequencies above 8kHz demonstrating more sensitivity to noise-induced damage than others [4].

The clinical utility of behavioural hearing thresholds has been largely explored; however, it is well established that these measures are influenced by factors such as attention, motivation, and patient compliance [1]. Objective measures, such as distortion product otoacoustic emissions (DPOAEs), hold much clinical promise and may be the only practical choice to evaluate children and adults unable to respond behaviourally [5]. DPOAEs offer a convenient and noninvasive means to detect ototoxic change in outer hair cell (OHC) function [6]. Distortion generated in the cochlea in response to the two simultaneous pure tones (ƒ₁ and ƒ₂, ƒ₁<ƒ₂) can be recorded in the ear canal at frequencies mathematically related to the stimulus frequencies [7]. The most extensively studied and clinically used is the DPOAE at the frequency 2ƒ₁-ƒ₂ [8]. Commonly, DPOAE levels as a function of frequency is used as an indicator of cochlear health. Compared to normal-hearing ears, DPOAE levels are reduced with mild to moderate hearing losses and are rarely present when hearing thresholds exceed 60dB HL [9]. Such a tool may be the only approach available to evaluate auditory function in young children and / or patients too sick to participate in ‘gold-standard’ behavioural audiometry.

The clinical utility of DPOAE measurements differentiating between normal and impaired cochlear function is well-established [9]. Several studies specifically support the use of DPOAEs as an objective measure of ototoxicity as it targets degradation of OHC function; however, these reports have been largely limited to measurements ≤ 6–8kHz, for example see Reavis et al. [10]. DPOAE measurement at these conventional frequencies has demonstrated earlier signs of ototoxic damage compared to behavioural audiometry at the same frequencies, for example see Arnold et al. [11]. As expected, DPOAEs at conventional frequencies lag in identifying damage confirmed with high-frequency audiometry (>8 kHz) [10]. It follows that high-frequency DPOAE measurements that specifically evaluate pre-neural peripheral function would combine the strengths of high-frequency audiometry (evaluation of the cochlear base) and conventional DPOAE (objective) measures. To this end, previous work has demonstrated the feasibility of measuring DPOAEs up to 16kHz, which has been found to be repeatable over time in young normal-hearing adults [12].

Delivering and recording acoustic signals at high frequencies is complicated by the need for specialised hardware and calibration techniques that are not readily available in most clinical and / or research settings. Measurements of behavioural hearing thresholds at high frequencies have largely been conducted using headphones calibrated after coupling via a flat-plate-coupler [13], which does not account for individual differences in ear canal geometry and acoustics. In contrast, in-the-ear calibration techniques attempt to overcome these individual differences through delivery of the signal at the desired level to each individual’s eardrum. One such method, accomplished in minutes, has been successfully employed by our group to make reliable measurements of hearing thresholds up to 20kHz (Figure 1) [1]. Calibration errors are known to arise when sound pressure level (SPL) measurements at the plane of the OAE microphone are used, particularly at higher frequencies, secondary to interactions between incident and reflected waves producing pressure nodes within the ear canal at the position of the microphone [14,15]. These pressure nodes lead to inaccurate estimations of the SPL at the surface of the tympanic membrane. Calibration procedures that measure sound intensity level or forward pressure level (FPL) of the stimulus rather than SPL greatly reduce the effects of standing waves [16,17] and decrease the overall variability in DPOAE measurements. Through translational research efforts in our laboratory, we have been able to overcome these complications using cutting-edge custom-built hardware and calibration procedures to obtain accurate and stable measurements of DPOAEs (and behavioural thresholds) up to 20kHz (Figure 2).

Figure 1.

Figure 2.

The feasibility of measuring hearing thresholds (Figure 1) and DPOAEs (Figure 2) up to 20kHz from individuals with normal hearing (≤ 20dB HL through 4kHz) is demonstrated here. While the full data set (n=352) was obtained from individuals between 10–65 years old for an ongoing project in the Dhar and Siegel laboratories, data from two age groups are presented here. Data from the younger group (10–21 years; n=84) serve as the benchmark for a relatively pristine auditory system, while data from the older group (46–55 years; n=62) demonstrate the sensitivity of these carefully executed measures. Clear differences at higher frequencies are observable between the groups in both measures. Critical for DPOAE measurement, the custom hardware allows the use of higher stimulus levels (for example, 75/75dB SPL) without significant system distortion. Specifically, DPOAE measurements extended beyond 8kHz were obtained with stimulus levels typically used in the clinic (open symbols) as well as using higher-level stimulus tones (filled symbols). These higher stimulus levels yield DPOAEs in the older group (filled squares) that can be compared with those obtained from the younger group using regularly employed stimuli (open circles). Findings suggest that using higher-level stimuli permits the opportunity to record DPOAEs at higher frequencies from a greater number of subjects. However, the impact of the use of these higher stimulus levels on the sensitivity of the measure is unknown at this point.

The important steps of suitable hardware, calibration, and signal delivery in examining the utility of high-frequency measures of auditory function have been accomplished. Along with a few other academic and industrial groups we are in the process of extending our techniques to make them available for clinical use. Routine measurement of auditory function over the entire audible frequency range using accurately calibrated stimuli is around the corner.