Transient evoked otoacoustic emissions
http://www.100md.com
《美国医学杂志》
Department of Otolayngology, Government Medical College and Hospital, Postgraduate Institute of Medical Education and Research, Chandigarh, India
Abstract
Objective: The purpose of this study was to collect parametric measures of TEOAEs in normal hearing children of various age-groups and to establish a normative baseline for Transient Evoked Otoacoustic Emissions (TEOAEs). Methods: Sixty subjects were investigated in three age-groups: neonates, 0-1 month; infants, 1 month-1 year; and children, 1-6 years. Each group comprised of 20 subjects. All the subjects underwent medical examination by a pediatrician and an ENT surgeon. Screening for hearing was done by immittance testing, behavior observation and conditioned play audiometry. The TEOAEs were analyzed for the parameters of amplitude, cross-correlation (wave reproducibility) and signal-to-noise ratio (SNR). Results: There was no difference between the mean amplitudes of the right and left ears in the groups. The females in the neonates group had higher emission amplitudes than the males. The mean amplitude of the subjects in the neonates group was significantly higher than the subjects in the infants or children groups. The cross correlation (wave reproducibility) was constant across the age. The mean SNR for all the subjects were well above 3 dB at frequencies 1.5 k, 2k, 3k and 4kHz. The neonates group showed the lowest SNR ranging between 3.47 to 9.62 dB. The infants group showed the highest SNR ranging between 6.13 to 13.11 dB. Conclusion: The TEOAEs response measures of SNR and cross correlation, at frequency bands 1.5, 2, 3 and 4 kHz, may provide more reliable outcomes than TEOAEs amplitude. Subjects in the age range of 0-1month show lower SNRs than those in higher age ranges. The values can be used as normative data for screening and diagnostic purposes in the pediatric population.
Keywords: Otoacoustic emissions; TEOAEs
The discovery of Otoacoustic emissions (OAE) by Kemp in 1978 has opened a new dimension in audiological assessment.[1] These are present in 98% of the ears of normal hearing individuals. The most promising application has been the use of evoked otoacoustic emissions as a screening device for the identification of hearing impairment, especially in neonates and infants. Transient evoked otoacoustic emissions (TEOAEs) have the advantage of being easy to use, rapid, objective, non-invasive and observed in almost all normal ears.[2]
It is not sufficient to obtain the emission values of the normal hearing population without considering the age of the subjects. There have been many studies[3],[4],[5],[6],[7],[8],[9],[10],[11],[12] reported in neonates, infants, children and adults using transient evoked otoacoustic emissions (TEOAEs). Children and adults exhibit a large difference in the size and resonance characteristics of ear canal and middle ear. These differences may produce a different TEOAE property, which restricts the applicability of wide data of adults on children. Several authors[6],[7],[10] have reported that the amplitude of TEOAEs reduces with age. Thus, one needs to study the characteristics of TEOAEs in various age groups in normals.
Therefore, the primary purpose of this study was to collect parametric measures of transient evoked otoacoustic emissions (TEOAEs) in normal hearing children of various age groups to provide a baseline against which otoacoustic activity in impaired ears can be compared. The present study was taken up with an aim of establishing a normative baseline for transient evoked otoacoustic emissions (TEOAEs) in neonates, infants and children and comparing the three groups, to establish the differences in these, if any.
Materials and Methods
Subjects: The subjects were investigated in three age groups: (1) neonates 0-1 month (mean age of 14 days), (2) infants 1 month - 1 year (mean age of 5 months), and (3) children 1-6 years (mean age of 3.5 years). Each group comprised of 20 subjects.
Equipment : A two channel clinical audiometer (Madsen Orbiter 922) and an immittance audiometer (Madsen Zodiac 901), calibrated prior to the study, were used to assess the pure tone thresholds and middle ear function of the subjects. The Smart OAE of Intelligent Hearing Systems (IHS), with software version 1.0 (Beta Version), and probe microphone ER-10D were used for TEOAE acquisition and analysis.
The frequency-domain plots include both the signal and noise. This is plotted together for signal-to-noise comparison. The signal-to-noise ratio (SNR) is the difference between the transient evoked emissions and level of the noise floor amplitudes in decibel (dB). The SNR level is typically used as the response criteria.
Procedure: The subjects adjudged by the pediatrician as normal were subjected to otoscopic examination by ENT surgeon to rule out any external or middle ear pathology and for immittance testing. Only those subjects with type 'A' tympanogram with reflexes present were included in the study.
All subjects below 3 years of age underwent behavior observation for hearing screening. The subjects above 3 years underwent conditioned play audiometry. These procedures were carried out in the sound treated rooms of the Speech and Hearing Unit, Department of Otorhinolaryngology, PGIMER, Chandigarh. The passing criteria for behavior observation audiometry was based on the observation of overt responses to controlled auditory signals and that for play audiometry was hearing thresholds < 25 dB HL.
Clicks were used as stimuli for obtaining TEOAEs with following settings: Sweeps 1024; Rate 20-40/s; Duration 100ms; Intensity 80 dB nHL and Gain 1000-4000 (automatic gain control).
Data Analysis : Mean of TEOAEs amplitude, correlation and signal-to-noise ratio (SNR) was computed in all the three groups. The obtained mean TEOAEs amplitude, correlation and signal-to-noise ratio (SNR) differences between right and left ears were subjected to unpaired t-test to find out the statistical significance between male and female ears in neonates, infants and children groups. In order to find out whether any statistical significant variation existed among neonates, infants and children TEOAEs SNR and amplitudes, one way Analysis of Variance (ANOVA) was applied.
Results
The TEOAEs amplitude, correlation and signal-to-noise ratio (SNR) of right and left ears were not statistically different; therefore the data for both ears were combined. TEOAEs amplitudes (in dB SPL) of male and female subjects in the three groups are shown for different frequencies in table1.
The mean TEOAEs amplitudes of female subjects in the neonate group are significantly larger than the male neonates. On an average the female subjects have 5-10 dB more TEOAE amplitudes than the male subjects. The TEOAE amplitude of the female subjects is highest in the neonates group. It is highest in the children group, for male subjects.
The mean TEOAEs cross correlation (or wave reproducibility) of male and female subjects in the three groups are shown for different frequencies in table2. It is observed that, except 1 kHz, the TEOAE cross correlation is in the range 0.675 (or 67.5%) to 0.965 (or 96.5%). It is constant across age.
The mean TEOAEs signal-to-noise ratios (SNR) of male and female subjects in the three groups are shown for different frequencies in table3.
The subjects in the neonate group show lowest SNRs, whereas the subjects in the infants group show highest SNRs, particularly at 2k, 3k and 4 kHz. The difference of the means of SNRs between the males and females of the same group is not statistically significant. The SNRs ranged from 4.02 to 13.11 dB at frequencies 1.5 kHz and above. The SNRs at 1 kHz are mostly below 3 dB.
Discussion
The mean minimum and maximum amplitudes in the three groups ranged from -29.22 to -7.18 dB SPL, respectively. In the neonate group there was no difference observed between the mean amplitudes of the right and left ears. These results are in accordance with the reports of Galattke and Robinette.[3] No significant differences were found in the amplitude between the right and left ears in the infants as well as children groups. Collet et al reported similar findings.[4] The data for both ears was, therefore, combined.
Female subjects in the neonate group were found to have a significantly larger mean amplitude (dB SPL) than the male neonates (dB SPL). Aidan, Avan and Bonfils and Kok et al similarly reported to have higher amplitude in females.[5],[6] No significant difference was observed between the mean amplitude of females and males in the infants as well as children groups. The mean amplitude of the subjects in the neonates group was significantly higher than the subjects in the infants as well as children groups. There was no significant difference between the mean amplitude in the children and infants groups. Norton and Widen found a similar reduction in amplitude with age.[7]
The cross correlation or the wave reproducibility in the present study was constant across age. Kon, Inagaki and Kaga reported similar findings in their study of normal subjects aged 1 month to 39 years.[10] The cross correlation at frequencies 2, 3 and 4 kHz was above 0.770 (77%). According to Aidan, Avan and Bonfils as well as Richardson et al, the parameter of cross correlation or reproducibility yields more reliable information at individual frequencies of 1.5, 2, 3 and 4 kHz.[5], [9]
The mean SNR for all the subjects in the present study was more than 3 dB at 1.5, 2, 3 and 4 kHz. It was less than 3 dB at 1 kHz. The subjects in the infants group showed significantly higher SNR as compared to the subjects in the other groups. The neonates showed the lowest SNR values. This may be because baby breathing and/or movements are sources of large noises. Norton et al have reported similar observations.[11] At frequencies 1.5, 2, 3 and 4 kHz, the subjects in the three groups showed SNR ranging between 3.97 to 13.11 dB.
Harrison and Norton (1999)[12] assessed whether the cut-off value for SNR, amplitude or reproducibility had potential diagnostic capabilities, and whether the response could be considered valid. Their study supported the use of a 3 dB SNR as the limiting value for response presence. They concluded that it might be desirable to use SNR as the response measure. They argued that, for cases in which both noise and amplitude are high, evaluation of response could be impossible if amplitude alone is considered. Further, reproducibility is essentially a signal to noise measure and it provides a redundant measure.
Studies such as those by Brass and Kemp and Brass, Watkins and Kemp and Dirckx et al have emphasized the use of >3 dB SNR criteria for reliable TEOAEs measurement.[13], [14], [15] The obtained values can be used as a normative data and thus can be used as a clinical tool for screening and diagnostic purposes in pediatric population.
References
1. Kemp DT. Stimulated acoustic emissions from within the human auditory system. J Acoust Soc Am 1978; 64 : 1386-1391.
2. Kemp DT. Otoacoustic emissions in perspective. In Robinette MS, Glattke TJ, Eds. Otoacoustic Emissions: Clinical Applications . New York; Thieme, 1997: 1-21.
3. Glattke TJ, Robinette MS. Transient evoked otoacoustic emissions. In Robinette MS, Glattke TJ, Eds. Otoacoustic emissions: Clinical applications . New York; Thieme, 1997: 63-82.
4. Collet L, Maulin A, Gartner M, and Morgan A. Age related change in evoked OAEs. Ann Oto Rhino Laryngol 1990; 99(12): 993-997.
5. Aidan D, Avan P, Bonfils P. Auditory screening in neonates by means of transient evoked otoacoustic emissions: A report of 2,842 recordings. Ann Oto Rhino Laryngol 1999; 108: 525-530.
6. Kok MR, van Zanten GA, Brocaar MP. Growth of evoked otoacoustic emissions during the first days post-partum: a preliminary report. Audiol 1992; 5 : 140-149.
7. Norton SJ, Widen JE. Evoked otoacoustic emissions in normal hearing infants and children. Ear Hear 1990; 11:121-127.
8. Elberling C, Parbo J, Johnsen NJ and Bagi P. Evoked acoustic emissions: Clinical application. Acta Otolaryngol 1985; 421: 77-85.
9. Richardson MP, Williamson TJ, Lenton SW, Tarlow MJ, Rudd PT. Otoacoustic emissions as a screening test for hearing impairment in children. Arch Dis Child 1995; 72: 294-297.
10. Kon K, Inagaki M, Kaga M. Developmental changes of distortion product and transient evoked otoacoustic emissions in different age groups. Brain Dev 2000; 22(1): 41-46.
11. Norton SJ, Gorga MP, Widen JE, Folsom RC, Sininger Y, Cone-Wesson B, Vohr BR, Fletcher KA. Identification of neonatal hearing impairment: Evaluation of transient evoked otoacoustic emission, distortion product emission, and auditory brain stem response test performance. Ear Hear 2000; 21: 508-528.
12. Harrison WA, Norton SJ. Characteristics of transient evoked otoacoustic emissions in normal hearing and hearing impaired children. Ear Hear 1999; 20: 75-86
13. Brass D, Kemp DT. The objective assessment of transient evoked otoacoustic emissions in neonates. Ear Hear 1994; 15: 371-377.
14. Brass D, Watkins P, Kemp DT. Assessment of an implementation of a narrow band, neonatal otoacoustic emission screening method. Ear Hear 1994; 15: 467-475.
15. Dirckx JJJ, Daemers TH, Somers F, Offeciers FE, Govaerts PJ. Numerical assessment of TOAE screening results: currently used criteria and their effect on TOAE prevalence figures. Acta Otolayngol 1996; 116: 672-679.(Kapoor Ravi, Panda Naresh)
Abstract
Objective: The purpose of this study was to collect parametric measures of TEOAEs in normal hearing children of various age-groups and to establish a normative baseline for Transient Evoked Otoacoustic Emissions (TEOAEs). Methods: Sixty subjects were investigated in three age-groups: neonates, 0-1 month; infants, 1 month-1 year; and children, 1-6 years. Each group comprised of 20 subjects. All the subjects underwent medical examination by a pediatrician and an ENT surgeon. Screening for hearing was done by immittance testing, behavior observation and conditioned play audiometry. The TEOAEs were analyzed for the parameters of amplitude, cross-correlation (wave reproducibility) and signal-to-noise ratio (SNR). Results: There was no difference between the mean amplitudes of the right and left ears in the groups. The females in the neonates group had higher emission amplitudes than the males. The mean amplitude of the subjects in the neonates group was significantly higher than the subjects in the infants or children groups. The cross correlation (wave reproducibility) was constant across the age. The mean SNR for all the subjects were well above 3 dB at frequencies 1.5 k, 2k, 3k and 4kHz. The neonates group showed the lowest SNR ranging between 3.47 to 9.62 dB. The infants group showed the highest SNR ranging between 6.13 to 13.11 dB. Conclusion: The TEOAEs response measures of SNR and cross correlation, at frequency bands 1.5, 2, 3 and 4 kHz, may provide more reliable outcomes than TEOAEs amplitude. Subjects in the age range of 0-1month show lower SNRs than those in higher age ranges. The values can be used as normative data for screening and diagnostic purposes in the pediatric population.
Keywords: Otoacoustic emissions; TEOAEs
The discovery of Otoacoustic emissions (OAE) by Kemp in 1978 has opened a new dimension in audiological assessment.[1] These are present in 98% of the ears of normal hearing individuals. The most promising application has been the use of evoked otoacoustic emissions as a screening device for the identification of hearing impairment, especially in neonates and infants. Transient evoked otoacoustic emissions (TEOAEs) have the advantage of being easy to use, rapid, objective, non-invasive and observed in almost all normal ears.[2]
It is not sufficient to obtain the emission values of the normal hearing population without considering the age of the subjects. There have been many studies[3],[4],[5],[6],[7],[8],[9],[10],[11],[12] reported in neonates, infants, children and adults using transient evoked otoacoustic emissions (TEOAEs). Children and adults exhibit a large difference in the size and resonance characteristics of ear canal and middle ear. These differences may produce a different TEOAE property, which restricts the applicability of wide data of adults on children. Several authors[6],[7],[10] have reported that the amplitude of TEOAEs reduces with age. Thus, one needs to study the characteristics of TEOAEs in various age groups in normals.
Therefore, the primary purpose of this study was to collect parametric measures of transient evoked otoacoustic emissions (TEOAEs) in normal hearing children of various age groups to provide a baseline against which otoacoustic activity in impaired ears can be compared. The present study was taken up with an aim of establishing a normative baseline for transient evoked otoacoustic emissions (TEOAEs) in neonates, infants and children and comparing the three groups, to establish the differences in these, if any.
Materials and Methods
Subjects: The subjects were investigated in three age groups: (1) neonates 0-1 month (mean age of 14 days), (2) infants 1 month - 1 year (mean age of 5 months), and (3) children 1-6 years (mean age of 3.5 years). Each group comprised of 20 subjects.
Equipment : A two channel clinical audiometer (Madsen Orbiter 922) and an immittance audiometer (Madsen Zodiac 901), calibrated prior to the study, were used to assess the pure tone thresholds and middle ear function of the subjects. The Smart OAE of Intelligent Hearing Systems (IHS), with software version 1.0 (Beta Version), and probe microphone ER-10D were used for TEOAE acquisition and analysis.
The frequency-domain plots include both the signal and noise. This is plotted together for signal-to-noise comparison. The signal-to-noise ratio (SNR) is the difference between the transient evoked emissions and level of the noise floor amplitudes in decibel (dB). The SNR level is typically used as the response criteria.
Procedure: The subjects adjudged by the pediatrician as normal were subjected to otoscopic examination by ENT surgeon to rule out any external or middle ear pathology and for immittance testing. Only those subjects with type 'A' tympanogram with reflexes present were included in the study.
All subjects below 3 years of age underwent behavior observation for hearing screening. The subjects above 3 years underwent conditioned play audiometry. These procedures were carried out in the sound treated rooms of the Speech and Hearing Unit, Department of Otorhinolaryngology, PGIMER, Chandigarh. The passing criteria for behavior observation audiometry was based on the observation of overt responses to controlled auditory signals and that for play audiometry was hearing thresholds < 25 dB HL.
Clicks were used as stimuli for obtaining TEOAEs with following settings: Sweeps 1024; Rate 20-40/s; Duration 100ms; Intensity 80 dB nHL and Gain 1000-4000 (automatic gain control).
Data Analysis : Mean of TEOAEs amplitude, correlation and signal-to-noise ratio (SNR) was computed in all the three groups. The obtained mean TEOAEs amplitude, correlation and signal-to-noise ratio (SNR) differences between right and left ears were subjected to unpaired t-test to find out the statistical significance between male and female ears in neonates, infants and children groups. In order to find out whether any statistical significant variation existed among neonates, infants and children TEOAEs SNR and amplitudes, one way Analysis of Variance (ANOVA) was applied.
Results
The TEOAEs amplitude, correlation and signal-to-noise ratio (SNR) of right and left ears were not statistically different; therefore the data for both ears were combined. TEOAEs amplitudes (in dB SPL) of male and female subjects in the three groups are shown for different frequencies in table1.
The mean TEOAEs amplitudes of female subjects in the neonate group are significantly larger than the male neonates. On an average the female subjects have 5-10 dB more TEOAE amplitudes than the male subjects. The TEOAE amplitude of the female subjects is highest in the neonates group. It is highest in the children group, for male subjects.
The mean TEOAEs cross correlation (or wave reproducibility) of male and female subjects in the three groups are shown for different frequencies in table2. It is observed that, except 1 kHz, the TEOAE cross correlation is in the range 0.675 (or 67.5%) to 0.965 (or 96.5%). It is constant across age.
The mean TEOAEs signal-to-noise ratios (SNR) of male and female subjects in the three groups are shown for different frequencies in table3.
The subjects in the neonate group show lowest SNRs, whereas the subjects in the infants group show highest SNRs, particularly at 2k, 3k and 4 kHz. The difference of the means of SNRs between the males and females of the same group is not statistically significant. The SNRs ranged from 4.02 to 13.11 dB at frequencies 1.5 kHz and above. The SNRs at 1 kHz are mostly below 3 dB.
Discussion
The mean minimum and maximum amplitudes in the three groups ranged from -29.22 to -7.18 dB SPL, respectively. In the neonate group there was no difference observed between the mean amplitudes of the right and left ears. These results are in accordance with the reports of Galattke and Robinette.[3] No significant differences were found in the amplitude between the right and left ears in the infants as well as children groups. Collet et al reported similar findings.[4] The data for both ears was, therefore, combined.
Female subjects in the neonate group were found to have a significantly larger mean amplitude (dB SPL) than the male neonates (dB SPL). Aidan, Avan and Bonfils and Kok et al similarly reported to have higher amplitude in females.[5],[6] No significant difference was observed between the mean amplitude of females and males in the infants as well as children groups. The mean amplitude of the subjects in the neonates group was significantly higher than the subjects in the infants as well as children groups. There was no significant difference between the mean amplitude in the children and infants groups. Norton and Widen found a similar reduction in amplitude with age.[7]
The cross correlation or the wave reproducibility in the present study was constant across age. Kon, Inagaki and Kaga reported similar findings in their study of normal subjects aged 1 month to 39 years.[10] The cross correlation at frequencies 2, 3 and 4 kHz was above 0.770 (77%). According to Aidan, Avan and Bonfils as well as Richardson et al, the parameter of cross correlation or reproducibility yields more reliable information at individual frequencies of 1.5, 2, 3 and 4 kHz.[5], [9]
The mean SNR for all the subjects in the present study was more than 3 dB at 1.5, 2, 3 and 4 kHz. It was less than 3 dB at 1 kHz. The subjects in the infants group showed significantly higher SNR as compared to the subjects in the other groups. The neonates showed the lowest SNR values. This may be because baby breathing and/or movements are sources of large noises. Norton et al have reported similar observations.[11] At frequencies 1.5, 2, 3 and 4 kHz, the subjects in the three groups showed SNR ranging between 3.97 to 13.11 dB.
Harrison and Norton (1999)[12] assessed whether the cut-off value for SNR, amplitude or reproducibility had potential diagnostic capabilities, and whether the response could be considered valid. Their study supported the use of a 3 dB SNR as the limiting value for response presence. They concluded that it might be desirable to use SNR as the response measure. They argued that, for cases in which both noise and amplitude are high, evaluation of response could be impossible if amplitude alone is considered. Further, reproducibility is essentially a signal to noise measure and it provides a redundant measure.
Studies such as those by Brass and Kemp and Brass, Watkins and Kemp and Dirckx et al have emphasized the use of >3 dB SNR criteria for reliable TEOAEs measurement.[13], [14], [15] The obtained values can be used as a normative data and thus can be used as a clinical tool for screening and diagnostic purposes in pediatric population.
References
1. Kemp DT. Stimulated acoustic emissions from within the human auditory system. J Acoust Soc Am 1978; 64 : 1386-1391.
2. Kemp DT. Otoacoustic emissions in perspective. In Robinette MS, Glattke TJ, Eds. Otoacoustic Emissions: Clinical Applications . New York; Thieme, 1997: 1-21.
3. Glattke TJ, Robinette MS. Transient evoked otoacoustic emissions. In Robinette MS, Glattke TJ, Eds. Otoacoustic emissions: Clinical applications . New York; Thieme, 1997: 63-82.
4. Collet L, Maulin A, Gartner M, and Morgan A. Age related change in evoked OAEs. Ann Oto Rhino Laryngol 1990; 99(12): 993-997.
5. Aidan D, Avan P, Bonfils P. Auditory screening in neonates by means of transient evoked otoacoustic emissions: A report of 2,842 recordings. Ann Oto Rhino Laryngol 1999; 108: 525-530.
6. Kok MR, van Zanten GA, Brocaar MP. Growth of evoked otoacoustic emissions during the first days post-partum: a preliminary report. Audiol 1992; 5 : 140-149.
7. Norton SJ, Widen JE. Evoked otoacoustic emissions in normal hearing infants and children. Ear Hear 1990; 11:121-127.
8. Elberling C, Parbo J, Johnsen NJ and Bagi P. Evoked acoustic emissions: Clinical application. Acta Otolaryngol 1985; 421: 77-85.
9. Richardson MP, Williamson TJ, Lenton SW, Tarlow MJ, Rudd PT. Otoacoustic emissions as a screening test for hearing impairment in children. Arch Dis Child 1995; 72: 294-297.
10. Kon K, Inagaki M, Kaga M. Developmental changes of distortion product and transient evoked otoacoustic emissions in different age groups. Brain Dev 2000; 22(1): 41-46.
11. Norton SJ, Gorga MP, Widen JE, Folsom RC, Sininger Y, Cone-Wesson B, Vohr BR, Fletcher KA. Identification of neonatal hearing impairment: Evaluation of transient evoked otoacoustic emission, distortion product emission, and auditory brain stem response test performance. Ear Hear 2000; 21: 508-528.
12. Harrison WA, Norton SJ. Characteristics of transient evoked otoacoustic emissions in normal hearing and hearing impaired children. Ear Hear 1999; 20: 75-86
13. Brass D, Kemp DT. The objective assessment of transient evoked otoacoustic emissions in neonates. Ear Hear 1994; 15: 371-377.
14. Brass D, Watkins P, Kemp DT. Assessment of an implementation of a narrow band, neonatal otoacoustic emission screening method. Ear Hear 1994; 15: 467-475.
15. Dirckx JJJ, Daemers TH, Somers F, Offeciers FE, Govaerts PJ. Numerical assessment of TOAE screening results: currently used criteria and their effect on TOAE prevalence figures. Acta Otolayngol 1996; 116: 672-679.(Kapoor Ravi, Panda Naresh)