Noise Health Home 

[Download PDF]
Year : 1999  |  Volume : 1  |  Issue : 2  |  Page : 58--65

Absence of otoacoustic emissions in subjects with normal audiometric thresholds implies exposure to noise

Anjali Desai1, David Reed2, Alex Cheyne3, Scott Richards2, Deepak Prasher1,  
1 Institute of Laryngology and Otology, University College London, London, United Kingdom
2 Princess Margaret Hospital, Swindon, United Kingdom
3 SLE, Croydon, Surrey, United Kingdom

Correspondence Address:
Deepak Prasher
Institute of Laryngology and Otology, University College London, 330 Gray's Inn Road, London WC1X 8EE
United Kingdom


Otoacoustic emissions and contra-lateral sound activated efferent suppression of emissions were examined to determine whether they provide any early indication of auditory damage from exposure to noise. Three groups were studied: noise exposed workers (n=50, mean age 42 years), patients with Meniere's disease (n=24, mean age 48 years) and normal subjects (n=24, mean age 41 years). All subjects underwent routine pure tone audiometry, tympanometry and otoacoustic emission testing. As a number of studies have shown that with hearing threshold better than 30 dB HL, emissions are almost always present and are generally absent with hearing loss greater than 30 dB HL, subjects in this study were sub grouped into these two categories in order to examine the incidence of emissions. Absence of emissions in subjects with mean hearing thresholds better than 30 dB HL varied from 0% in normal controls, 8% in patients with Meniere's disease and a significantly high 56% in noise exposed workers despite similar mean hearing thresholds for all groups. The mean transient emission levels for the noise exposed workers was significantly lower than the controls and Meniere's groups. This study clearly indicates that in the noise-exposed group there is sub clinical and sub audiometric damage to the outer hair cells responsible for generation of otoacoustic emissions. Of those with normal otoacoustic emissions, the efferent suppression was absent in 60% of noise exposed workers but in only 3.8% of control subjects implying that the efferent control may also be affected in a significant proportion despite normal hearing thresholds and emissions.

How to cite this article:
Desai A, Reed D, Cheyne A, Richards S, Prasher D. Absence of otoacoustic emissions in subjects with normal audiometric thresholds implies exposure to noise.Noise Health 1999;1:58-65

How to cite this URL:
Desai A, Reed D, Cheyne A, Richards S, Prasher D. Absence of otoacoustic emissions in subjects with normal audiometric thresholds implies exposure to noise. Noise Health [serial online] 1999 [cited 2020 Feb 28 ];1:58-65
Available from:

Full Text


It has been shown that otoacoustic emissions (OAEs) can be recorded in almost all normal subjects. [Kemp, 1979; Probst et al, 1986] As the hearing threshold worsens the probability of recording normal otoacoustic emissions is significantly reduced. The hearing threshold level beyond which emissions may be absent has been variable across studies, from 15dB HL [Kemp, 1978], 25dB HL [Probst et al, 1987], 35dB HL [Bonfils and Uziel, 1989], to 40dB HL [Collet et al, 1989, Johnsen et al, 1993]. It is now generally accepted that 30 dB is a fence beyond which emissions may not always be recorded. It is considered that subjects exposed to noise may show an early indication of cochlear damage with absence of emissions prior to any observable change in auditory threshold. This study examined otoacoustic emissions and their suppression in three groups of subjects to determine whether the noise group shows any specific features that may be used to identify individuals within the group.

 Materials and Methods


Three groups were studied. The first was a control group (n=24) comprising of hospital staff and students. The second group consisted of people working in a tea packing factory (NIHL) (n=50) exposed to noise of 80-85dBA for a period ranging from 2 weeks to 10 years and the third group consisted of patients with Meniere's disease (MD) (n=24) visiting the hospital. Meniere's disease patients were clinically diagnosed on the basis of AAOO criteria [Alford, 1972]. Control subjects and patients with MD were tested in a sound proof booth in the Hospital whereas the noise exposed group were tested in a quiet room at the factory site.

All subjects underwent standard pure tone audiometry and tympanometry. Mean hearing was defined as an average of thresholds at 500, 1k, 2k and 4k Hz.

Normal middle ear function was defined as the ear drum compliance from 0.3 - 1.7 cm3 and the peak middle ear pressure of 50 daPa as the influence of the middle ear transmission properties on OAE is known [Hauser et al, 1993].

Otodynamics ILO88 analyser was used for Otoacoustic measurements.

TEOAE Procedure

The default setting of the ILO system was used to record the TEOAE with a stimulus comprising of "Non-linear" 80ms clicks presented at a repetition rate of 50Hz and an intensity of 80dB SPL. 260 sweeps were averaged for a complete response. Analysis was done in terms of total emission power, reproducibility, and spectral analysis as a function of hearing and age. TEOAEs were considered to be present if the amplitude was greater than 2.5 dB SPL and the reproducibility was greater than 50%.

Efferent Test Procedure

This test consists of recording of TEOAE with and without contralateral stimulation and the difference in response is considered. A dual channel OAE analyser was used, one channel (A) for ipsilateral and the other (B) for contralateral stimulation. For ipsilateral stimulation, a linear click at 603dB SPL and for contralateral, a continuous white noise (0.506kHz) at 40dB SL were used, applying an alternating technique, a 'Difference B on/off' mode, from the ILO88 software. This mode allows alternating recording of TEOAE responses with and without contralateral stimulation. A total of 600 sweeps were recorded, in 10 runs of 60 sweeps. The average responses were directly computed and the difference obtained by their subtraction represented the suppression effect.

SOAE Procedure

The standard ILO system SOAE recordings were made with 260 sweeps. Synchronised SOAE were recorded using a single 80ms click at approximately 75dB SPL, presented in 80 msec intervals. The presence of SOAE was observed as spectral peaks above the noise floor and their amplitude and frequency were measured using the cursor provided.

The prevalence of SOAE and the number of SOAE peaks per ear were determined.


195 ears of 98 subjects from the three groups (controls, NIHL and MD) were studied. The mean age and male/ female distribution of the three groups is shown in [Table 1]. The mean age was not significantly different between the groups.

Pure tone Audiometry

All the ears had normal compliance and middle ear pressure. The groups were subdivided into those having a mean (0.5,1,2,4 kHz) hearing level less than 30 dB and those greater than 30 dB. Further analysis was restricted to those with mean hearing level less than 30 dB. The mean hearing level for the three groups (127 ears) is shown in [Table 2]. The mean hearing level for the controls (15.45.1 dB), NIHL (14.5 8.3) and MD (12.2 7.7) was not significantly different.


Transient emissions were considered to be present if the response was greater than 2.5 dB SPL and the repeatability was greater than 50%. [Table 3] gives the range and mean TEOAE for the three groups. The mean TEOAE for NIHL (5.9 dB SPL) was significantly (p Efferent suppression of TEOAE

The ears with presence of TEOAEs were then subjected to a test for contralateral sound activated suppression of emissions. Suppression was considered to be present if the difference between the response with and without contralateral masking was greater than 1 dB SPL.

The percentage of ears with absence of suppression in [Table 5] for the NIHL group is significantly high (60%) compared to the controls (3.8%) despite the hearing level being similar for the two groups.

In [Figure 2] the relationship between duration of noise exposure and suppression is illustrated. The ratio of absence of suppression to presence in first decade of exposure to noise was 0.35, which increased to 0.6 for the second decade indicating further effect of continued exposure. SOAE

The results of the SOAE testing reveal that SOAEs were only present in six ears, 4 of which had mean hearing level less than 15 dB HL, but interestingly the other two had mean hearing level greater than 35 dB HL. The SOAEs were only observed in female subjects and none of them had tinnitus.


The salient feature of our study was that in a very high proportion (56%) of subjects with noise induced hearing loss transient evoked otoacoustic emissions (TEOAEs) were absent as compared to controls (0%) and subjects with Meniere's disease (8%) despite the mean hearing level being similar for the three groups and less than 30dB HL. In contrast to this there was one subject who, even with exposure to noise for more than 25 years, had TEOAEs present and his hearing was within normal audiometric limits. In an earlier study, Probst et. al. (1987) had also noted that in 10 ears with noise induced hearing loss and hearing less than 25 dB HL there were "fewer OAEs". Similarly Bicciolo et. al. (1993) found 48% of 46 subjects exposed to noise but having normal threshold had abnormal TEOAEs but none of their 15 normal subjects.

The absence of emissions seems to be specific to the noise exposed group, as patients with Meniere's disease and control do not show a similar pattern. In Meniere's disease, increase of the extracellular potassium concentration in the perilymph results in a progressive loss of cochlear hair cells. It has also been noted [Hauser et al, 1993] that in hydropic ears basilar membrane motion may be impeded along with uncoupling of the outer hair cell (OHC) stereocilia from the tectorial membrane. Thus it may be that in endolymphatic hydrops, outer hair cells are not damaged in the initial stage but made ineffective by the cochlear derived toxic activity allowing emissions to be recorded when the hearing thresholds are near normal. In the case of NIHL, two types of damage can be found : a pattern of hair cell degeneration in the first row of the outer hair cells (OHCs), then in the inner hair cells (IHCs), subsequently in the second and third row of OHCs; and a massive destruction of dendrites of the primary auditory neurones below the IHCs. Interestingly it has also been noted that after 14 days no dendritic damage was observed suggesting that a reconnection of the IHCs by the dendrites of the auditory neurones had occurred. Thus reduction in incidence of OAEs in the noise exposed group may be associated with sensory-cell damage to localised cochlear regions subserving specific frequencies that, according to audiometric thresholds, appear to be normal. This has been supported by the various animal studies [Braun, 1996; Bicciolo et al, 1993; Bohne and Clark, 1982; Davis et al, 1993].

Thus OAEs may be used as an objective means to assess at a preclinical stage, noise-induced, sensory-cell damage to cochlear frequency regions that appear to be audiometrically normal. Similar trend as seen in emissions was also noted in terms of suppression, which was absent in a large proportion of ears having hearing level better than 30 dB. Thus there were significant differences in the level of contralateral sound activated suppression of TEOAEs between normal subjects and those with noise induced hearing loss. The medial olivocochlear neurones are projected mainly contralaterally from the region around the medial nuclei of the superior olivary complex to the base of the outer hair cells of the organ of Corti; hence any dysfunction to the efferent mechanism may affect OHC activity. Patuzzi & Rajan (1992_ showed in guinea pigs that medial olivocochlear system produced significant threshold elevation of the compound action potential audiogram, which may be due to disruption of active processes through a reduction in either the electrical drive or the impedance of the basolateral membrane of OHCs resulting in significant hearing loss which they termed 'motor loss'. Auditory efferent system modulates afferent input at the cochlear level, suppressing any input that is not relevant or is undesirable. Therefore it is plausible that the efferent system in a noisy environment may be overactive in suppressing the unwanted signal and over a long period of time is no longer able to separate wanted from unwanted signal. In this instance OAE suppression would be absent or significantly reduced.

The study also provides an insight into individual susceptibility to noise. It shows that the subjects with decrease in their hearing level may be most susceptible to NIHL and subjects with TEOAEs absent with normal hearing level would be next susceptible. Subjects with absent suppression with normal TEOAEs and normal hearing level appears to be less susceptible to NIHL where as the least susceptible group appear to be the one with normal hearing level along with normal TEOAEs and presence of suppression. There are a few subjects who had a robust auditory system as their TEOAEs and suppression were present despite two decades of exposure and audiometric threshold remaining less than 30 dB.

Probst et al (1987) reported that SOAEs were found only at frequencies with interpolated thresholds of less than 15 dB HL and no SOAEs were detected in ears with PTAs greater than 25 dB HL. However our findings were not in agreement with the above. In our study we found SOAEs present in two ears with mean hearing loss greater than 35 dB HL.

In conclusion, early identification of cochlear damage from noise exposure is possible using otoacoustic emissions. The absence of emissions in the presence of normal pure tone audiometric threshold implies noise exposure as the factor of importance as other cochlear pathology with normal hearing retains presence of normal emissions.

Absence of efferent suppression may be an even earlier indicator of noise affecting the auditory system prior to any clinical or structural damage.[15]


1Alford B(1972) Meniere's disease: criteria for diagnosis and evluation of therapy for reporting results. Report of subcommittee on equilibrium and its management. Trans Am Ac Opthalmol Otolaryngol 76,1462-1464.
2Bicciolo G, Ruscito P, Rizzo S, Frenguelli A (1993) Evoked otoacoustic emissions in noise induced hearing loss. Acta Otorhinolaryngol (Ital) 13(6),505-515.
3Bohne B, Clark W (1982) Growth of hearing loss and cochlear lesion with increasing duration of noise exposure. In Perspectives on Noise-Induced Hearing Loss. Hamernik R, Henderson D, Salvi R, eds. New New York: Raven Press, 283-300.
4Bonfils P, Uziel A.(1989)Clinical applications of evoked acoustic emissions: results in normally hearing and hearing-impaired subjects. Ann Otol Rhinol Larygol 98,26-331.
5Braun M.(1996) Impediment of basilar membrane motion reduces overload protection but not threshold sensitivity: evidence from clinical and experimental hydrops. Hear Res 97(1-2), 1-10.
6Collet L, Gartner M, Moulin M, Kauffmann I, Disant F, Morgon A. (1989) Evoked otoacoustic emissions and sensorineural hearing loss. Arch Otolaryngol Head Neck Surg 115, 1060-1062.
7Davis R, Hamernik R, Ahroon W. (1993) Frequency selectivity in noise-damaged cochleas. Audiology 32, 110-131.
8Hauser R, Probst R, Harris F. (1993) Effect of atmospheric pressure on spontaneous, transiently evoked and distortion product otoacoustic emissions in normal human ears. Hear Res 69, 133-145.
9Hawkins J, Johnsson L. (1976) Patterns of sensorineural degeneration in human ears exposed to noise. In Effects of Noise on Hearing. Henderson D, Hamernik R, Dosanjh D, et al eds. New York: Raven Press, 91-100.
10Johnsen N, Parbo J, Elberling C. (1993) Evoked acoustic emissions from the human ear. Scand Audiol 22, 87-95.
11Kemp D. (1978) Stimulated acoustic emissions from within the human auditory system. J Acoust Soc Am 64, 1386-1391.
12Kemp D. (1979) Evidence of mechanical nonlinearity and frequency selective wave amplification in the cochlea. Arch Otorhinolaryngol 224, 37-45.
13Patuzzi R, Rajan R. (1992) Additivity of threshold elevations produced by disruption of outer hair cell function. Hear Res 60, 165-177.
14Probst R, Coats A, Martin G, Lonsbury-Martin B. (1986) Spontaneous, click and toneburst-evoked emissions from normal ears. Hear Res 21, 261-275.
15Probst R, Lonsbury-Martin B, Martin G, Coats A. (1987) Otoacoustic emissions in ears with hearing loss. Am J Otolaryngol 8, 73-81.