International Journal of Pediatric Otorhinolaryngology 79 (2015) 1310–1315
Contents lists available at ScienceDirect
International Journal of Pediatric Otorhinolaryngology journal homepage: www.elsevier.com/locate/ijporl
Tone burst evoked otoacoustic emissions in different age-groups of schoolchildren W. Wiktor Jedrzejczak a,b,*, Edyta Pilka a,b, Piotr H. Skarzynski a,b,c,d, Lukasz Olszewski a,b, Henryk Skarzynski a,b a
Institute of Physiology and Pathology of Hearing, ul. M. Mochnackiego 10, 02-042 Warsaw, Poland World Hearing Center, ul. Mokra 17, Kajetany, 05-830 Nadarzyn, Poland Heart Failure and Cardiac Rehabilitation Department of the Medical University of Warsaw, Warsaw, Poland d Institute of Sensory Organs, ul. Mokra 1, Kajetany, 05-830 Nadarzyn, Poland b c
A R T I C L E I N F O
A B S T R A C T
Article history: Received 12 April 2015 Received in revised form 29 May 2015 Accepted 31 May 2015 Available online 8 June 2015
Introduction: Otoacoustic emissions (OAEs) are believed to be good predictors of hearing status, particularly in the 1–4 kHz range. However both click evoked OAEs (CEOAEs) and distortion product OAEs (DPOAEs) perform poorly at 0.5 kHz. The present study investigates OAEs in the lower frequency range of 0.5–1 kHz evoked by 0.5 kHz tone bursts (TBOAEs) in schoolchildren and compares them with emissions evoked by clicks. Methods: Measurements were performed for two groups of normally hearing schoolchildren. Children from 1st grade (age 6–7 years) and children from 6th grade (age 11–12 years). Tympanometry, pure tone audiometry, and OAE measurements of CEAOEs, 0.5 kHz TBOAEs, and spontaneous OAEs (SOAEs) were performed. Additionally, analysis by the matching pursuit method was conducted on CEOAEs and TBOAEs to assess their time–frequency (TF) properties. Results: For all subjects OAEs response levels and signal to noise ratios (SNRs) were calculated. As expected, CEOAE magnitudes were greatest over the range 1–4 kHz, with a substantial decrease below 1 kHz. Responses from the 0.5 kHz TBOAEs were complementary in that the main components occurred between 0.5 and 1.4 kHz. In younger children, TBOAEs had SNRs 4–8 dB smaller in the 0.5–1.4 kHz range. In addition, CEOAEs had lower SNRs in the 0.7–1.4 kHz range, by 3–5 dB. TBOAEs in younger children had maximum SNRs shifted toward 1–1.4 kHz, whereas in older children it was more clearly around 1 kHz. The differences in response levels were less evident. The presence of SOAEs appreciably influenced both CEOAEs and TBOAEs, and TF properties of both OAEs did not differ significantly between grades. Conclusion: TBOAEs evoked at 0.5 kHz can provide additional information about frequencies below 1 kHz, a range over which CEOAEs usually have very low amplitudes. The main difference between the two age groups was that in older children CEOAEs and 0.5 kHz TBOAEs had higher SNRs at 0.5–1.4 kHz. Additionally, for ears with SOAEs, 0.5 kHz TBOAEs had higher response levels and SNRs similar to CEOAEs. ß 2015 Elsevier Ireland Ltd. All rights reserved.
Keywords: Tone burst evoked otoacoustic emissions Click evoked otoacoustic emissions Spontaneous otoacoustic emissions Transiently evoked otoacoustic emissions OAE Schoolchildren
1. Introduction Otoacoustic emissions (OAEs) [1] are believed to be good predictors of hearing status, particularly in the 1–4 kHz range. Click evoked OAEs (CEOAEs) are good indicators of cochlear function at 1 kHz, whereas distortion product OAEs (DPOAEs) perform better at higher frequencies. Both CEOAEs and DPOAEs
* Corresponding author at: World Hearing Center, ul. Mokra 17, Kajetany, 05-830 Nadarzyn, Poland. Tel.: +48 22 35 60 574; fax: +48 22 35 60 367. E-mail address: [email protected] (W.W. Jedrzejczak). http://dx.doi.org/10.1016/j.ijporl.2015.05.040 0165-5876/ß 2015 Elsevier Ireland Ltd. All rights reserved.
perform poorly at 0.5 kHz [2–4] where there is still room for improvement. There is still some debate on the value of testing the hearing of schoolchildren. Newborn hearing screening has been universally applied in many countries but doing the same for schoolchildren is not widely practiced. Some organizations recommend screening of preschool and school-age children, pointing out the risks associated with undetected hearing loss [5]. There is also some controversy on what screening method is the best. Some studies recommend OAEs [6,7], while others suggest pure tone audiometry [8]. If OAEs could be evoked over a broader frequency range then it
W.W. Jedrzejczak et al. / International Journal of Pediatric Otorhinolaryngology 79 (2015) 1310–1315
is likely that their usefulness could be improved. A basic problem with measuring OAEs in children is that measurements at lower frequencies are often contaminated by noise [9]. This noise susceptibility problem might be overcome by using low frequency tone bursts as a stimulus. Some studies (e.g. [4,10,11]) have shown that OAEs evoked by tone bursts (TBOAEs) may provide a better estimate of hearing status than those evoked by clicks. In particular, the 0.5-kHz TBOAE seems to be a more reliable measure than CEOAE when probing cochlear responses at low frequencies (e.g. [12]). Indications are that TBOAEs can be effectively measured in children, and there are commercially available devices that offer this measurement [13]. Screening of schoolchildren can be organized in a way that schoolchildren are tested at the beginning of school (i.e. at 1st grade) and later at 6th grade, a scheme implemented in rural areas in Poland [14,15]. When considering OAE-based screening, it should be kept in mind that OAEs tend to change with age [16]. Their amplitude decreases with age and there are shifts in dominant frequencies [17]. OAEs stabilize at around age 18, and thereafter remain quite stable if there is no hearing loss. But for young subjects, a caveat is that the results and indications for one age group might not readily translate to another. The least investigated type of OAEs are spontaneous OAEs (SOAEs) [18]. Because they are usually not detected in all ears [19], they are of little interest to clinicians. However, it is known that SOAEs generally coincide with minima in hearing thresholds [20], and several studies have shown that SOAEs tend to produce higher amplitude CEOAEs [17,21]. A similar effect has been detected in DPOAEs [22]. SOAEs are influenced by age: their presence is greatest in newborns and declines with age [17]. In terms of TBOAEs, the effect of SOAEs has so far only been seen as generating peaks in TBOAE spectra [4,23]. One analysis method that promises valuable insight into OAEs is time–frequency (TF) analysis. The method can reveal, for example, how the latency of CEOAEs depends on frequency, with high frequency components having shorter latency than low frequency components [24]. TF properties may be affected by some hearing deficits [25], although it is not clear if they change with age [26,27]. The TF properties of TBOAEs have already been the subject of a previous study [28] where it has been shown that sets of TBOAEs ranging in frequency from 1 to 4 kHz have the same main components as CEOAEs. However, the same work has also shown that the 0.5 kHz TBOAE may contain components not present in the CEOAE; in addition, the work also made clear that SOAEs show up in TF energy distributions of both CEOAEs and TBOAEs [28]. The present study investigates the usefulness of responses in the lower frequency range (0.5–1 kHz) evoked by 0.5 kHz tone bursts. Two age groups were tested, children from 1st grade and 6th grade of primary school, and the amplitude and SNR of 0.5 kHz TBOAEs recorded from them were compared with emissions evoked by clicks. The influence of SOAEs on evoked emissions was also investigated.
2. Material and methods Measurements were performed on two groups of normally hearing schoolchildren: 32 children from 1st grade (64 ears) of age 6–7 years and 32 children from 6th grade (64 ears) of age 11–12 years. All children underwent visual inspection of the ear canal and tympanic membrane of both ears followed by tympanometry, pure tone audiometry, and OAE measurement. All had pure tone thresholds better than 25 dB HL at 0.5–8 kHz, type A tympanograms, and no known history of otologic disease. The parents gave written informed consent prior to participation of their children in the study. The research procedures were approved by the Ethics Committee of the Institute of Physiology and Pathology of Hearing, Poland.
1311
OAEs were measured in low-noise ambient conditions using an ILO 292 system (Otodynamics Ltd). Standard click stimuli and 0.5 kHz tone bursts (average amplitude 80 3 peak dB SPL, nonlinear averaging protocol) were used to evoke a total of 260 OAE responses. The tone bursts were two cycles long with equal rise/fall times and no plateau. The initial part of the response was windowed automatically by the system to minimize stimuli artifacts. Window onset was 2.5 ms for clicks and 5 ms for 0.5 kHz tone bursts, and all recordings used an acquisition window of 20 ms. Half-octave-band values of OAE response levels and signal-to-noise ratios (SNRs) were used for analysis. SOAEs were acquired using the in-built routine provided by the ILO 292 equipment, resulting in a measurement of so-called synchronized SOAEs (SSOAEs). An ear was classified as ‘‘with SOAEs’’ when at least one long-lasting peak was found in the SSOAE spectrum that exceeded the noise floor by 5 dB SPL [29]. For all parameters the statistical significance of the mean difference between groups was evaluated using the Wilcoxon rank sum test. As a criterion of significance, a 95% confidence level (p < 0.05) was chosen. 2.1. Time–frequency (TF) analysis Time–frequency (TF) analysis of the recorded signals was done by decomposing the signals into their basic waveforms. The method of high-resolution adaptive approximation was used, a technique based on the matching pursuit (MP) algorithm [30]. A slight modification was used to account for the asymmetrical character of some components [31]. The modified MP method allowed the CEOAE and TBOAE signals to be decomposed into waveforms of defined frequency, latency, duration, and amplitude. The latency was taken to be the time taken from onset of the stimulus to the maximum point in the waveform envelope. Note that the presence of SOAEs can, when using some methods, cause a false shift in evoked OAE latency, whereas MP with an asymmetric dictionary provides estimates that are less prone to SOAEs [31]. 3. Results Global and half-octave band response levels and SNRs were calculated for OAEs from all subjects. Comparison of global parameters between grades revealed no significant differences in response level between CEOAEs (Fig. 1, left); a similar outcome can be seen for 0.5 kHz TBOAEs. However, Fig. 1 shows that in terms of SNRs, both CEOAEs and TBOAEs had significantly higher values in schoolchildren from the 6th grade. This probably relates to higher noise levels usually present in OAE measurements of younger children [32]. Fig. 2 shows average half-octave band values of OAE amplitudes evoked by clicks and 0.5 kHz tone bursts for children from the 1st grade. Fig. 3 shows comparable results for 6th grade children. As expected, CEOAE magnitudes were greatest over the range 1–4 kHz, decreasing substantially below 1 kHz. Responses from the 0.5-kHz TBOAEs were complementary in that the main components occurred between 0.5 and 1.4 kHz (top rows of Figs. 2 and 3). TBOAEs were larger than CEOAEs (on average more than 5 dB larger) at 0.5–1 kHz (the difference was significant at p < 0.05). At 1.4 kHz there was no statistically significant difference between CEAOEs and TBOAEs. As expected, at higher frequencies TBOAE response levels and SNRs declined, while CEOAEs were significantly higher. A notable feature is that the maximum level of the 0.5 kHz TBOAE occurs at 1 kHz, not at the center frequency of stimulation (0.5 kHz). The subsequent rows of Figs. 2 and 3 show levels of TBOAEs and CEOAEs when the dataset was divided into ears with and without SOAEs. There were 27% of ears with SOAEs in the 1st grade dataset and 33% of ears in the 6th grade dataset. Generally, both CEOAEs
W.W. Jedrzejczak et al. / International Journal of Pediatric Otorhinolaryngology 79 (2015) 1310–1315
15
CEOAE
15
TBOAE
CEOAE
TBOAE *
*
10
5
0
10
SNR [dB]
Response level [dB SPL]
1312
grade 1
grade 6
grade 1
5
0
grade 6
grade 1
grade 6
grade 1
grade 6
Fig. 1. Global average response level (left) and SNR (right) of CEOAEs and TBOAEs for ears of schoolchildren from grade 1 and grade 6. Whiskers indicate standard errors; asterisks at top denote significant differences.
and TBOAEs from the 6th grade ears had significantly higher response levels and SNRs over nearly all frequencies than the 1st grade ears. It is especially interesting that in 6th grade TBOAEs from ears with SOAEs (Fig. 3, middle row), TBOAE levels were always higher than in ears without SOAEs, even in the 0.7 kHz band, a frequency below that of any detected SOAE. There were some minor differences between OAEs of the two groups. In younger children, TBOAEs had SNRs 4–8 dB smaller in the 0.5–1.4 kHz range, and this difference was significant. In addition, CEOAEs had significantly lower SNRs in the 0.7–1.4 kHz
range, by 3–5 dB. TBOAEs in younger children had maximum SNRs shifted toward 1–1.4 kHz, whereas in older children it was more clearly around 1 kHz. The differences in response levels were less evident, with significant differences only for 0.7 kHz for TBOAEs and 0.5 kHz for CEOAEs. Fig. 4 shows TF properties of CEOAEs and TBOAEs, and a clear dependence of latency on frequency can be seen. For CEOAEs (top panel), latencies extend from around 5 ms for the 4 kHz band to 10 ms for the 1 kHz band. From 4 kHz to 1 kHz the dependence of latency on frequency is, on a log–log scale, almost a straight line.
TBOAE & CEOAE
TBOAE & CEOAE
20
30
10
20
0
10
−10
0
−20
0.5
0.7
1
1.4
2
2.8
−10
4
TBOAE CEOAE
0.5
0.7
1
TBOAE 30
10
20
0
10
−10
0
0.5
0.7
1
1.4
2
2.8
−10
4
4
SOAE−
0.5
0.7
1
1.4
2
2.8
4
CEOAE 30
10
20 SNR [dB]
Response level [dB SPL]
2.8
SOAE+
CEOAE 20
0 −10 −20
2
TBOAE
20
−20
1.4
10 SOAE+
0
0.5
0.7
1 1.4 2 2.8 frequency [kHz]
4
−10
SOAE−
0.5
0.7
1 1.4 2 2.8 frequency [kHz]
4
Fig. 2. Top row: Average half-octave band parameters for CEOAEs and TBOAEs for ears of schoolchildren from the 1st grade. Middle row: Parameters for TBOAEs for ears with SOAEs (27%) and without. Bottom row: Parameters for CEOAEs. Whiskers indicate standard errors, and asterisks denote significant differences.
W.W. Jedrzejczak et al. / International Journal of Pediatric Otorhinolaryngology 79 (2015) 1310–1315
TBOAE & CEOAE
TBOAE & CEOAE
20
30
10
20
0
10
−10
0
−20
0.5
0.7
1
1.4
2
2.8
−10
4
TBOAE CEOAE
0.5
0.7
1
TBOAE 30
10
20
0
10
−10
0
0.5
0.7
1
1.4
2
2.8
−10
4
2.8
4
SOAE−
0.5
0.7
1
1.4
2
2.8
4
CEOAE 30
10
20 SNR [dB]
Response level [dB SPL]
2
SOAE+
CEOAE 20
0 −10 −20
1.4
TBOAE
20
−20
1313
10 SOAE+
0
0.5
0.7
1 1.4 2 2.8 frequency [kHz]
4
−10
SOAE−
0.5
0.7
1 1.4 2 2.8 frequency [kHz]
4
Fig. 3. Top row: Average half-octave band parameters for CEOAEs and TBOAEs for ears of schoolchildren from the 6th grade. Middle row: Parameters for TBOAEs for ears with SOAEs (33%) and without. Bottom row: Parameters for CEOAEs. Whiskers indicate standard errors, and asterisks denote significant differences.
For lower frequencies it reaches 11 ms for 0.7 kHz and around 10 ms for 0.5 kHz. At these frequencies CEOAEs were weak and nearly absent. For TBOAEs (bottom panel) latency is only shown for the 0.5–2 kHz range since at higher frequencies there were practically no response (see Figs. 2 and 3). Latencies for TBOAEs were constant at around 10 ms. There were slight latency differences between the two age groups but none were significant. 4. Discussion This study is a continuation of a study comparing two systems of hardware – ILO 292 and HearID – and which opened up the possibility of usefully recording 0.5 kHz TBOAEs [13]. Based on a group of 6th grade schoolchildren, successful results were obtained from both devices. In the present study, the work with 0.5 kHz TBOAEs was extended to cover two groups of schoolchildren, those in the 1st and 6th grades (6–7 and 11–12 years old), and TBOAEs were compared with CEOAEs. The aim was to evaluate differences in TBOAEs between the two age groups and to see whether the presence of SOAEs had any effect. No significant differences in global response levels between age groups was found. However, global SNR for TBOAEs and CEOAEs was higher in schoolchildren from grade 6. In [33], schoolchildren of age around 12 years were examined using similar equipment,
and similar global responses and SNRs were obtained to 6th graders in the present study. In [34], several age groups ranging from 6 to 25 years were studied, and global response levels generally decreased with age. However, in [34] no statistically significant difference in global response levels was found between groups aged 6–10 and 11–15. This is similar to the present results (despite small differences in the age of subjects studied). The results here confirm that 0.5 kHz TBOAEs provide additional information to CEOAEs, particularly for the frequency range below 1 kHz—a range in which CEOAEs responses have very low amplitude. The amplitudes were similar for both age groups. For 6th grade children, CEOAE levels were 3–4 dB over the 1–4 kHz range. In [33] lower levels than this at 1 kHz were found. Possibly this is due to fact that they used a screening protocol that had a shorter recording window, causing some of the lower frequency responses to be lost. In [34] it was found that CEOAE levels (in third-octave frequency bands) decreased with age, although there was no statistically significant difference between responses in the 6–10 and 11–15 age-groups. There was no data on half-octave SNRs. In our work, the differences in response levels between the two age groups were similarly small. However, we did find a significant difference in SNRs for frequencies below 1.4 kHz. In this range, 6th graders had higher SNRs for both CEOAEs and 0.5 kHz TBOAEs.
W.W. Jedrzejczak et al. / International Journal of Pediatric Otorhinolaryngology 79 (2015) 1310–1315
1314
CEOAE 16 14 12 10 8 6
4
0.5
0.7
1
1.4
2
2.8
4
stimulus was close to 0.5 kHz. This curious behavior is probably related to spectral splatter from the short (2 cycle) tone burst stimulus and from the rapidly falling responses of the cochlea and the recording system at low frequencies. From the results of previous work using a 4 cycle burst, it is probably better to use 4 cycles to evoke a response with a narrower frequency spread and which has a peak response closer to 0.5 kHz. However such an option is not available with the commercial equipment used in this study. To conclude, 0.5 kHz TBOAEs may be recorded in groups of schoolchildren (either 7 or 12 years old) with similar effectiveness to CEOAEs. In the older group, CEOAEs and 0.5 kHz TBOAEs had best SNRs at 0.5–1.4 kHz. For ears with SOAEs, 0.5 kHz TBOAEs and CEOAEs had higher response levels and SNRs. TF properties did not differ significantly between grades for CEOAEs and TBOAEs. Acknowledgments
TBOAE 16 1st grade 6th grade
14
latency [ms]
12
Preliminary results of this study were presented at the ARO 2015 meeting. The authors wish to thank A. Piotrowska for help with organization of the measurements and A. Bell, B. Trzaskowski and K. Kochanek for comments on earlier versions of the manuscript.
10 References
8 6
4
0.5
0.7
1
1.4
2
2.8
4
frequency [kHz] Fig. 4. Average half-octave band latency for CEOAEs (top) and 0.5 kHz TBOAEs (bottom) for ears of schoolchildren from grade 1 and grade 6. Both axes are shown on a logarithmic scale. Whiskers indicate standard errors.
The presence of SOAEs significantly affected both CEOAEs and TBOAEs. Of special note, the level of TBOAEs in 6th graders was affected in the 0.7 kHz half-octave band (as well as in other bands), despite the fact that SOAEs are usually only observed above 1 kHz. This observation suggests that the presence of SOAEs at any frequency represents increased gain of the cochlear amplifier at all frequencies. Of all the ears we tested, the ears without SOAEs had lower OAE parameters than those with SOAEs, even though all ears were classified as having normal hearing. The implication here is that some ‘normal’ ears may not pass strict OAE screening criteria, leading to the suggestion that, in order to obtain robust OAEs in some ears, more averages should be made during measurements. TF properties did not differ significantly between grades for CEOAEs and TBOAEs. For CEOAEs, the dependence of latency on frequency was quite similar to that obtained using the same method for adults [31]. The lack of changes in TF properties may be due to our use of an asymmetric dictionary. If the effect of SOAEs is diminished in this way, then this may lead to no age effect in TF properties [26]. In the earlier study, latencies for 0.5 kHz TBOAEs were much longer (about 14 ms), although there a longer 8 ms stimulus was used [28]. Here, a 4 ms long stimulus was used and latency was found to be around 10 ms. It therefore seems that the difference in latencies is probably directly related to differences in stimulus length. An apparent anomaly in the 0.5 kHz TBOAE measurements is that they had a maximum level at around 1 kHz, even though the
[1] D.T. Kemp, Stimulated acoustic emissions from within the human auditory system, J. Acoust. Soc. Am. 64 (Nov (5)) (1978) 1386–1391. [2] M.P. Gorga, S.T. Neely, B.M. Bergman, K.L. Beauchaine, J.R. Kaminski, J. Peters, et al., A comparison of transient-evoked and distortion product otoacoustic emissions in normal-hearing and hearing-impaired subjects, J. Acoust. Soc. Am. 94 (Nov (5)) (1993) 2639–2648. [3] M. Suckfu¨ll, S. Schneeweiss, A. Dreher, K. Schorn, Evaluation of TEOAE and DPOAE measurements for the assessment of auditory thresholds in sensorineural hearing loss, Acta Otolaryngol. 116 (Jul (4)) (1996) 528–533. [4] W.A. Harrison, S.J. Norton, Characteristics of transient evoked otoacoustic emissions in normal-hearing and hearing-impaired children, Ear Hear. 20 (Feb (1)) (1999) 75–86. [5] H. Skarz˙yn´ski, A. Piotrowska, Screening for pre-school and school-age hearing problems: European consensus statement, Int. J. Pediatr. Otorhinolaryngol. 76 (1) (2012) 120–121. [6] C. Driscoll, J. Kei, B. McPherson, Outcomes of transient evoked otoacoustic emission testing in 6-year-old school children: a comparison with pure tone screening and tympanometry, Int. J. Pediatr. Otorhinolaryngol. 57 (1) (2001) 67–76. [7] D.G. Balatsouras, A. Kaberos, G. Kloutsos, N.C. Economou, V. Sakellariadis, A. Fassolis, et al., Correlation of transiently evoked to distortion-product otoacoustic emission measures in healthy children, Int. J. Pediatr. Otorhinolaryngol. 70 (1) (2006) 89–93. [8] L. S´liwa, S. Hatzopoulos, K. Kochanek, A. Piłka, A. Senderski, P.H. Skarz˙yn´ski, A comparison of audiometric and objective methods in hearing screening of school children. A preliminary study, Int. J. Pediatr. Otorhinolaryngol. 75 (4) (2011) 483–488. [9] P.J. Bray, D. Kemp, An advanced cochlear echo technique suitable for infant screening, Br. J. Audiol. 21 (1987) 191–204. [10] V. Lichtenstein, D.R. Stapells, Frequency-specific identification of hearing loss using transient-evoked otoacoustic emissions to clicks and tones, Hear. Res. 98 (Sep (1–2)) (1996) 125–136. [11] B. McPherson, S.F. Li, B.X. Shi, J.L. Tang, B.Y. Wong, Neonatal hearing screening: evaluation of tone-burst and click-evoked otoacoustic emission test criteria, Ear Hear. 27 (Jun (3)) (2006) 256–262. [12] W.W. Jedrzejczak, K. Kochanek, B. Trzaskowski, E. Pilka, P.H. Skarzynski, H. Skarzynski, Tone-burst and click-evoked otoacoustic emissions in subjects with hearing loss above 0.25, 0.5, and 1 kHz, Ear Hear. 33 (Nov–Dec (6)) (2012) 757–767. [13] W.W. Jedrzejczak, A. Piotrowska, K. Kochanek, L. Sliwa, H. Skarzynski, Lowfrequency otoacoustic emissions in schoolchildren measured by two commercial devices, Int. J. Pediatr. Otorhinolaryngol. 77 (10) (2013) 1724–1728. [14] H. Skarzynski, K. Kochanek, A. Senderski, P.H. Skarzynski, M. Ludwikowski, M. Kopaczewski, et al., Organization of the hearing screening examinations in Polish schools in rural areas and small towns, Cochlear Implants Int. 11 (Jun (Suppl. 1)) (2010) 143–147. [15] P.H. Skarz˙yn´ski, A. Piłka, M. Ludwikowski, M.B. Skarz˙yn´ska, Comparison of the frequency of positive hearing screening outcomes in schoolchildren from Poland and other countries of Europe: Central Asia, and Africa, J. Hear. Sci. 4 (4) (2014) OA51–OA58. [16] S.J. Norton, J.E. Widen, Evoked otoacoustic emissions in normal-hearing infants and children: emerging data and issues, Ear Hear. 11 (Apr (2)) (1990) 121–127.
W.W. Jedrzejczak et al. / International Journal of Pediatric Otorhinolaryngology 79 (2015) 1310–1315 [17] B.A. Prieve, T.S. Fitzgerald, L.E. Schulte, Basic characteristics of click-evoked otoacoustic emissions in infants and children, J. Acoust. Soc. Am. 102 (Nov (5 Pt 1)) (1997) 2860–2870. [18] D.T. Kemp, Evidence of mechanical nonlinearity and frequency selective wave amplification in the cochlea, Arch. Otorhinolaryngol. 224 (1–2) (1979) 37–45. [19] T. Kuroda, Clinical investigation on spontaneous otoacoustic emission (SOAE) in 447 ears, Auris Nasus Larynx 34 (Mar (1)) (2007) 29–38. [20] G.R. Long, A. Tubis, Investigations into the nature of the association between threshold microstructure and otoacoustic emissions, Hear. Res. 36 (Nov (2–3)) (1988) 125–138. [21] W.W. Jedrzejczak, S. Hatzopoulos, L. Sliwa, E. Pilka, K. Kochanek, H. Skarzynski, Otoacoustic emissions in neonates measured with different acquisition protocols, Int. J. Pediatr. Otorhinolaryngol. 76 (3) (2012) 382–387. [22] T. Kuroda, S. Fukuda, E. Chida, M. Kashiwamura, M. Matsumura, R. Ohwatari, et al., Effects of spontaneous otoacoustic emissions on distortion product otoacoustic emission, Auris Nasus Larynx 28 (May (Suppl)) (2001) S33–S38. [23] R. Probst, A.C. Coats, G.K. Martin, B.L. Lonsbury-Martin, Spontaneous, click-, and toneburst-evoked otoacoustic emissions from normal ears, Hear. Res. 21 (3) (1986) 261–275. [24] G. Tognola, F. Grandori, P. Ravazzani, Time–frequency distributions of clickevoked otoacoustic emissions, Hear. Res. 106 (Apr (1–2)) (1997) 112–122. [25] A. Moleti, R. Sisto, G. Tognola, M. Parazzini, P. Ravazzani, F. Grandori, Otoacoustic emission latency, cochlear tuning, and hearing functionality in neonates, J. Acoust. Soc. Am. 118 (Sep (3 Pt 1)) (2005) 1576–1584. [26] W.W. Jedrzejczak, S. Hatzopoulos, A. Martini, K.J. Blinowska, Otoacoustic emissions latency difference between full-term and preterm neonates, Hear. Res. 231 (Sep (1–2)) (2007) 54–62.
1315
[27] A. Moleti, R. Sisto, A. Paglialonga, F. Sibella, L. Anteunis, M. Parazzini, et al., Transient evoked otoacoustic emission latency and estimates of cochlear tuning in preterm neonates, J. Acoust. Soc. Am. 124 (Nov (5)) (2008) 2984–2994. [28] W.W. Jedrzejczak, A. Lorens, A. Piotrowska, K. Kochanek, H. Skarzynski, Otoacoustic emissions evoked by 0.5 kHz tone bursts, J. Acoust. Soc. Am. 125 (5) (2009) 3158–3165. [29] W.W. Jedrzejczak, K.J. Blinowska, K. Kochanek, H. Skarzynski, Synchronized spontaneous otoacoustic emissions analyzed in a time–frequency domain, J. Acoust. Soc. Am. 124 (Dec (6)) (2008) 3720–3729. [30] S.G. Mallat, Z. Zhang, Matching pursuit with time–frequency dictionaries, IEEE Trans. Signal Process. 41 (1993) 3397–3415. [31] W.W. Jedrzejczak, K. Kwaskiewicz, K.J. Blinowska, K. Kochanek, H. Skarzynski, Use of the matching pursuit algorithm with a dictionary of asymmetric waveforms in the analysis of transient evoked otoacoustic emissions, J. Acoust. Soc. Am. 126 (2009) 3137–3146. [32] V.W. Zhang, B. McPherson, B.X. Shi, J.L. Tang, B.Y. Wong, Neonatal hearing screening: a combined click evoked and tone burst otoacoustic emission approach, Int. J. Pediatr. Otorhinolaryngol. 72 (3) (2008) 351–360. [33] G. Pavlovcinova´, J. Jakubı´kova´, T. Trnovec, K. Lancz, S. Wimmerova´, E. Sovcı´kova´, et al., A normative study of otoacoustic emissions, ear asymmetry, and gender effect in healthy schoolchildren in Slovakia, Int. J. Pediatr. Otorhinolaryngol. 74 (2) (2010) 173–177. [34] D. Groh, J. Pelanova, M. Jilek, J. Popelar, Z. Kabelka, J. Syka, Changes in otoacoustic emissions and high-frequency hearing thresholds in children and adolescents, Hear. Res. 212 (Feb (1–2)) (2006) 90–98.
Contents lists available at ScienceDirect
International Journal of Pediatric Otorhinolaryngology journal homepage: www.elsevier.com/locate/ijporl
Tone burst evoked otoacoustic emissions in different age-groups of schoolchildren W. Wiktor Jedrzejczak a,b,*, Edyta Pilka a,b, Piotr H. Skarzynski a,b,c,d, Lukasz Olszewski a,b, Henryk Skarzynski a,b a
Institute of Physiology and Pathology of Hearing, ul. M. Mochnackiego 10, 02-042 Warsaw, Poland World Hearing Center, ul. Mokra 17, Kajetany, 05-830 Nadarzyn, Poland Heart Failure and Cardiac Rehabilitation Department of the Medical University of Warsaw, Warsaw, Poland d Institute of Sensory Organs, ul. Mokra 1, Kajetany, 05-830 Nadarzyn, Poland b c
A R T I C L E I N F O
A B S T R A C T
Article history: Received 12 April 2015 Received in revised form 29 May 2015 Accepted 31 May 2015 Available online 8 June 2015
Introduction: Otoacoustic emissions (OAEs) are believed to be good predictors of hearing status, particularly in the 1–4 kHz range. However both click evoked OAEs (CEOAEs) and distortion product OAEs (DPOAEs) perform poorly at 0.5 kHz. The present study investigates OAEs in the lower frequency range of 0.5–1 kHz evoked by 0.5 kHz tone bursts (TBOAEs) in schoolchildren and compares them with emissions evoked by clicks. Methods: Measurements were performed for two groups of normally hearing schoolchildren. Children from 1st grade (age 6–7 years) and children from 6th grade (age 11–12 years). Tympanometry, pure tone audiometry, and OAE measurements of CEAOEs, 0.5 kHz TBOAEs, and spontaneous OAEs (SOAEs) were performed. Additionally, analysis by the matching pursuit method was conducted on CEOAEs and TBOAEs to assess their time–frequency (TF) properties. Results: For all subjects OAEs response levels and signal to noise ratios (SNRs) were calculated. As expected, CEOAE magnitudes were greatest over the range 1–4 kHz, with a substantial decrease below 1 kHz. Responses from the 0.5 kHz TBOAEs were complementary in that the main components occurred between 0.5 and 1.4 kHz. In younger children, TBOAEs had SNRs 4–8 dB smaller in the 0.5–1.4 kHz range. In addition, CEOAEs had lower SNRs in the 0.7–1.4 kHz range, by 3–5 dB. TBOAEs in younger children had maximum SNRs shifted toward 1–1.4 kHz, whereas in older children it was more clearly around 1 kHz. The differences in response levels were less evident. The presence of SOAEs appreciably influenced both CEOAEs and TBOAEs, and TF properties of both OAEs did not differ significantly between grades. Conclusion: TBOAEs evoked at 0.5 kHz can provide additional information about frequencies below 1 kHz, a range over which CEOAEs usually have very low amplitudes. The main difference between the two age groups was that in older children CEOAEs and 0.5 kHz TBOAEs had higher SNRs at 0.5–1.4 kHz. Additionally, for ears with SOAEs, 0.5 kHz TBOAEs had higher response levels and SNRs similar to CEOAEs. ß 2015 Elsevier Ireland Ltd. All rights reserved.
Keywords: Tone burst evoked otoacoustic emissions Click evoked otoacoustic emissions Spontaneous otoacoustic emissions Transiently evoked otoacoustic emissions OAE Schoolchildren
1. Introduction Otoacoustic emissions (OAEs) [1] are believed to be good predictors of hearing status, particularly in the 1–4 kHz range. Click evoked OAEs (CEOAEs) are good indicators of cochlear function at 1 kHz, whereas distortion product OAEs (DPOAEs) perform better at higher frequencies. Both CEOAEs and DPOAEs
* Corresponding author at: World Hearing Center, ul. Mokra 17, Kajetany, 05-830 Nadarzyn, Poland. Tel.: +48 22 35 60 574; fax: +48 22 35 60 367. E-mail address: [email protected] (W.W. Jedrzejczak). http://dx.doi.org/10.1016/j.ijporl.2015.05.040 0165-5876/ß 2015 Elsevier Ireland Ltd. All rights reserved.
perform poorly at 0.5 kHz [2–4] where there is still room for improvement. There is still some debate on the value of testing the hearing of schoolchildren. Newborn hearing screening has been universally applied in many countries but doing the same for schoolchildren is not widely practiced. Some organizations recommend screening of preschool and school-age children, pointing out the risks associated with undetected hearing loss [5]. There is also some controversy on what screening method is the best. Some studies recommend OAEs [6,7], while others suggest pure tone audiometry [8]. If OAEs could be evoked over a broader frequency range then it
W.W. Jedrzejczak et al. / International Journal of Pediatric Otorhinolaryngology 79 (2015) 1310–1315
is likely that their usefulness could be improved. A basic problem with measuring OAEs in children is that measurements at lower frequencies are often contaminated by noise [9]. This noise susceptibility problem might be overcome by using low frequency tone bursts as a stimulus. Some studies (e.g. [4,10,11]) have shown that OAEs evoked by tone bursts (TBOAEs) may provide a better estimate of hearing status than those evoked by clicks. In particular, the 0.5-kHz TBOAE seems to be a more reliable measure than CEOAE when probing cochlear responses at low frequencies (e.g. [12]). Indications are that TBOAEs can be effectively measured in children, and there are commercially available devices that offer this measurement [13]. Screening of schoolchildren can be organized in a way that schoolchildren are tested at the beginning of school (i.e. at 1st grade) and later at 6th grade, a scheme implemented in rural areas in Poland [14,15]. When considering OAE-based screening, it should be kept in mind that OAEs tend to change with age [16]. Their amplitude decreases with age and there are shifts in dominant frequencies [17]. OAEs stabilize at around age 18, and thereafter remain quite stable if there is no hearing loss. But for young subjects, a caveat is that the results and indications for one age group might not readily translate to another. The least investigated type of OAEs are spontaneous OAEs (SOAEs) [18]. Because they are usually not detected in all ears [19], they are of little interest to clinicians. However, it is known that SOAEs generally coincide with minima in hearing thresholds [20], and several studies have shown that SOAEs tend to produce higher amplitude CEOAEs [17,21]. A similar effect has been detected in DPOAEs [22]. SOAEs are influenced by age: their presence is greatest in newborns and declines with age [17]. In terms of TBOAEs, the effect of SOAEs has so far only been seen as generating peaks in TBOAE spectra [4,23]. One analysis method that promises valuable insight into OAEs is time–frequency (TF) analysis. The method can reveal, for example, how the latency of CEOAEs depends on frequency, with high frequency components having shorter latency than low frequency components [24]. TF properties may be affected by some hearing deficits [25], although it is not clear if they change with age [26,27]. The TF properties of TBOAEs have already been the subject of a previous study [28] where it has been shown that sets of TBOAEs ranging in frequency from 1 to 4 kHz have the same main components as CEOAEs. However, the same work has also shown that the 0.5 kHz TBOAE may contain components not present in the CEOAE; in addition, the work also made clear that SOAEs show up in TF energy distributions of both CEOAEs and TBOAEs [28]. The present study investigates the usefulness of responses in the lower frequency range (0.5–1 kHz) evoked by 0.5 kHz tone bursts. Two age groups were tested, children from 1st grade and 6th grade of primary school, and the amplitude and SNR of 0.5 kHz TBOAEs recorded from them were compared with emissions evoked by clicks. The influence of SOAEs on evoked emissions was also investigated.
2. Material and methods Measurements were performed on two groups of normally hearing schoolchildren: 32 children from 1st grade (64 ears) of age 6–7 years and 32 children from 6th grade (64 ears) of age 11–12 years. All children underwent visual inspection of the ear canal and tympanic membrane of both ears followed by tympanometry, pure tone audiometry, and OAE measurement. All had pure tone thresholds better than 25 dB HL at 0.5–8 kHz, type A tympanograms, and no known history of otologic disease. The parents gave written informed consent prior to participation of their children in the study. The research procedures were approved by the Ethics Committee of the Institute of Physiology and Pathology of Hearing, Poland.
1311
OAEs were measured in low-noise ambient conditions using an ILO 292 system (Otodynamics Ltd). Standard click stimuli and 0.5 kHz tone bursts (average amplitude 80 3 peak dB SPL, nonlinear averaging protocol) were used to evoke a total of 260 OAE responses. The tone bursts were two cycles long with equal rise/fall times and no plateau. The initial part of the response was windowed automatically by the system to minimize stimuli artifacts. Window onset was 2.5 ms for clicks and 5 ms for 0.5 kHz tone bursts, and all recordings used an acquisition window of 20 ms. Half-octave-band values of OAE response levels and signal-to-noise ratios (SNRs) were used for analysis. SOAEs were acquired using the in-built routine provided by the ILO 292 equipment, resulting in a measurement of so-called synchronized SOAEs (SSOAEs). An ear was classified as ‘‘with SOAEs’’ when at least one long-lasting peak was found in the SSOAE spectrum that exceeded the noise floor by 5 dB SPL [29]. For all parameters the statistical significance of the mean difference between groups was evaluated using the Wilcoxon rank sum test. As a criterion of significance, a 95% confidence level (p < 0.05) was chosen. 2.1. Time–frequency (TF) analysis Time–frequency (TF) analysis of the recorded signals was done by decomposing the signals into their basic waveforms. The method of high-resolution adaptive approximation was used, a technique based on the matching pursuit (MP) algorithm [30]. A slight modification was used to account for the asymmetrical character of some components [31]. The modified MP method allowed the CEOAE and TBOAE signals to be decomposed into waveforms of defined frequency, latency, duration, and amplitude. The latency was taken to be the time taken from onset of the stimulus to the maximum point in the waveform envelope. Note that the presence of SOAEs can, when using some methods, cause a false shift in evoked OAE latency, whereas MP with an asymmetric dictionary provides estimates that are less prone to SOAEs [31]. 3. Results Global and half-octave band response levels and SNRs were calculated for OAEs from all subjects. Comparison of global parameters between grades revealed no significant differences in response level between CEOAEs (Fig. 1, left); a similar outcome can be seen for 0.5 kHz TBOAEs. However, Fig. 1 shows that in terms of SNRs, both CEOAEs and TBOAEs had significantly higher values in schoolchildren from the 6th grade. This probably relates to higher noise levels usually present in OAE measurements of younger children [32]. Fig. 2 shows average half-octave band values of OAE amplitudes evoked by clicks and 0.5 kHz tone bursts for children from the 1st grade. Fig. 3 shows comparable results for 6th grade children. As expected, CEOAE magnitudes were greatest over the range 1–4 kHz, decreasing substantially below 1 kHz. Responses from the 0.5-kHz TBOAEs were complementary in that the main components occurred between 0.5 and 1.4 kHz (top rows of Figs. 2 and 3). TBOAEs were larger than CEOAEs (on average more than 5 dB larger) at 0.5–1 kHz (the difference was significant at p < 0.05). At 1.4 kHz there was no statistically significant difference between CEAOEs and TBOAEs. As expected, at higher frequencies TBOAE response levels and SNRs declined, while CEOAEs were significantly higher. A notable feature is that the maximum level of the 0.5 kHz TBOAE occurs at 1 kHz, not at the center frequency of stimulation (0.5 kHz). The subsequent rows of Figs. 2 and 3 show levels of TBOAEs and CEOAEs when the dataset was divided into ears with and without SOAEs. There were 27% of ears with SOAEs in the 1st grade dataset and 33% of ears in the 6th grade dataset. Generally, both CEOAEs
W.W. Jedrzejczak et al. / International Journal of Pediatric Otorhinolaryngology 79 (2015) 1310–1315
15
CEOAE
15
TBOAE
CEOAE
TBOAE *
*
10
5
0
10
SNR [dB]
Response level [dB SPL]
1312
grade 1
grade 6
grade 1
5
0
grade 6
grade 1
grade 6
grade 1
grade 6
Fig. 1. Global average response level (left) and SNR (right) of CEOAEs and TBOAEs for ears of schoolchildren from grade 1 and grade 6. Whiskers indicate standard errors; asterisks at top denote significant differences.
and TBOAEs from the 6th grade ears had significantly higher response levels and SNRs over nearly all frequencies than the 1st grade ears. It is especially interesting that in 6th grade TBOAEs from ears with SOAEs (Fig. 3, middle row), TBOAE levels were always higher than in ears without SOAEs, even in the 0.7 kHz band, a frequency below that of any detected SOAE. There were some minor differences between OAEs of the two groups. In younger children, TBOAEs had SNRs 4–8 dB smaller in the 0.5–1.4 kHz range, and this difference was significant. In addition, CEOAEs had significantly lower SNRs in the 0.7–1.4 kHz
range, by 3–5 dB. TBOAEs in younger children had maximum SNRs shifted toward 1–1.4 kHz, whereas in older children it was more clearly around 1 kHz. The differences in response levels were less evident, with significant differences only for 0.7 kHz for TBOAEs and 0.5 kHz for CEOAEs. Fig. 4 shows TF properties of CEOAEs and TBOAEs, and a clear dependence of latency on frequency can be seen. For CEOAEs (top panel), latencies extend from around 5 ms for the 4 kHz band to 10 ms for the 1 kHz band. From 4 kHz to 1 kHz the dependence of latency on frequency is, on a log–log scale, almost a straight line.
TBOAE & CEOAE
TBOAE & CEOAE
20
30
10
20
0
10
−10
0
−20
0.5
0.7
1
1.4
2
2.8
−10
4
TBOAE CEOAE
0.5
0.7
1
TBOAE 30
10
20
0
10
−10
0
0.5
0.7
1
1.4
2
2.8
−10
4
4
SOAE−
0.5
0.7
1
1.4
2
2.8
4
CEOAE 30
10
20 SNR [dB]
Response level [dB SPL]
2.8
SOAE+
CEOAE 20
0 −10 −20
2
TBOAE
20
−20
1.4
10 SOAE+
0
0.5
0.7
1 1.4 2 2.8 frequency [kHz]
4
−10
SOAE−
0.5
0.7
1 1.4 2 2.8 frequency [kHz]
4
Fig. 2. Top row: Average half-octave band parameters for CEOAEs and TBOAEs for ears of schoolchildren from the 1st grade. Middle row: Parameters for TBOAEs for ears with SOAEs (27%) and without. Bottom row: Parameters for CEOAEs. Whiskers indicate standard errors, and asterisks denote significant differences.
W.W. Jedrzejczak et al. / International Journal of Pediatric Otorhinolaryngology 79 (2015) 1310–1315
TBOAE & CEOAE
TBOAE & CEOAE
20
30
10
20
0
10
−10
0
−20
0.5
0.7
1
1.4
2
2.8
−10
4
TBOAE CEOAE
0.5
0.7
1
TBOAE 30
10
20
0
10
−10
0
0.5
0.7
1
1.4
2
2.8
−10
4
2.8
4
SOAE−
0.5
0.7
1
1.4
2
2.8
4
CEOAE 30
10
20 SNR [dB]
Response level [dB SPL]
2
SOAE+
CEOAE 20
0 −10 −20
1.4
TBOAE
20
−20
1313
10 SOAE+
0
0.5
0.7
1 1.4 2 2.8 frequency [kHz]
4
−10
SOAE−
0.5
0.7
1 1.4 2 2.8 frequency [kHz]
4
Fig. 3. Top row: Average half-octave band parameters for CEOAEs and TBOAEs for ears of schoolchildren from the 6th grade. Middle row: Parameters for TBOAEs for ears with SOAEs (33%) and without. Bottom row: Parameters for CEOAEs. Whiskers indicate standard errors, and asterisks denote significant differences.
For lower frequencies it reaches 11 ms for 0.7 kHz and around 10 ms for 0.5 kHz. At these frequencies CEOAEs were weak and nearly absent. For TBOAEs (bottom panel) latency is only shown for the 0.5–2 kHz range since at higher frequencies there were practically no response (see Figs. 2 and 3). Latencies for TBOAEs were constant at around 10 ms. There were slight latency differences between the two age groups but none were significant. 4. Discussion This study is a continuation of a study comparing two systems of hardware – ILO 292 and HearID – and which opened up the possibility of usefully recording 0.5 kHz TBOAEs [13]. Based on a group of 6th grade schoolchildren, successful results were obtained from both devices. In the present study, the work with 0.5 kHz TBOAEs was extended to cover two groups of schoolchildren, those in the 1st and 6th grades (6–7 and 11–12 years old), and TBOAEs were compared with CEOAEs. The aim was to evaluate differences in TBOAEs between the two age groups and to see whether the presence of SOAEs had any effect. No significant differences in global response levels between age groups was found. However, global SNR for TBOAEs and CEOAEs was higher in schoolchildren from grade 6. In [33], schoolchildren of age around 12 years were examined using similar equipment,
and similar global responses and SNRs were obtained to 6th graders in the present study. In [34], several age groups ranging from 6 to 25 years were studied, and global response levels generally decreased with age. However, in [34] no statistically significant difference in global response levels was found between groups aged 6–10 and 11–15. This is similar to the present results (despite small differences in the age of subjects studied). The results here confirm that 0.5 kHz TBOAEs provide additional information to CEOAEs, particularly for the frequency range below 1 kHz—a range in which CEOAEs responses have very low amplitude. The amplitudes were similar for both age groups. For 6th grade children, CEOAE levels were 3–4 dB over the 1–4 kHz range. In [33] lower levels than this at 1 kHz were found. Possibly this is due to fact that they used a screening protocol that had a shorter recording window, causing some of the lower frequency responses to be lost. In [34] it was found that CEOAE levels (in third-octave frequency bands) decreased with age, although there was no statistically significant difference between responses in the 6–10 and 11–15 age-groups. There was no data on half-octave SNRs. In our work, the differences in response levels between the two age groups were similarly small. However, we did find a significant difference in SNRs for frequencies below 1.4 kHz. In this range, 6th graders had higher SNRs for both CEOAEs and 0.5 kHz TBOAEs.
W.W. Jedrzejczak et al. / International Journal of Pediatric Otorhinolaryngology 79 (2015) 1310–1315
1314
CEOAE 16 14 12 10 8 6
4
0.5
0.7
1
1.4
2
2.8
4
stimulus was close to 0.5 kHz. This curious behavior is probably related to spectral splatter from the short (2 cycle) tone burst stimulus and from the rapidly falling responses of the cochlea and the recording system at low frequencies. From the results of previous work using a 4 cycle burst, it is probably better to use 4 cycles to evoke a response with a narrower frequency spread and which has a peak response closer to 0.5 kHz. However such an option is not available with the commercial equipment used in this study. To conclude, 0.5 kHz TBOAEs may be recorded in groups of schoolchildren (either 7 or 12 years old) with similar effectiveness to CEOAEs. In the older group, CEOAEs and 0.5 kHz TBOAEs had best SNRs at 0.5–1.4 kHz. For ears with SOAEs, 0.5 kHz TBOAEs and CEOAEs had higher response levels and SNRs. TF properties did not differ significantly between grades for CEOAEs and TBOAEs. Acknowledgments
TBOAE 16 1st grade 6th grade
14
latency [ms]
12
Preliminary results of this study were presented at the ARO 2015 meeting. The authors wish to thank A. Piotrowska for help with organization of the measurements and A. Bell, B. Trzaskowski and K. Kochanek for comments on earlier versions of the manuscript.
10 References
8 6
4
0.5
0.7
1
1.4
2
2.8
4
frequency [kHz] Fig. 4. Average half-octave band latency for CEOAEs (top) and 0.5 kHz TBOAEs (bottom) for ears of schoolchildren from grade 1 and grade 6. Both axes are shown on a logarithmic scale. Whiskers indicate standard errors.
The presence of SOAEs significantly affected both CEOAEs and TBOAEs. Of special note, the level of TBOAEs in 6th graders was affected in the 0.7 kHz half-octave band (as well as in other bands), despite the fact that SOAEs are usually only observed above 1 kHz. This observation suggests that the presence of SOAEs at any frequency represents increased gain of the cochlear amplifier at all frequencies. Of all the ears we tested, the ears without SOAEs had lower OAE parameters than those with SOAEs, even though all ears were classified as having normal hearing. The implication here is that some ‘normal’ ears may not pass strict OAE screening criteria, leading to the suggestion that, in order to obtain robust OAEs in some ears, more averages should be made during measurements. TF properties did not differ significantly between grades for CEOAEs and TBOAEs. For CEOAEs, the dependence of latency on frequency was quite similar to that obtained using the same method for adults [31]. The lack of changes in TF properties may be due to our use of an asymmetric dictionary. If the effect of SOAEs is diminished in this way, then this may lead to no age effect in TF properties [26]. In the earlier study, latencies for 0.5 kHz TBOAEs were much longer (about 14 ms), although there a longer 8 ms stimulus was used [28]. Here, a 4 ms long stimulus was used and latency was found to be around 10 ms. It therefore seems that the difference in latencies is probably directly related to differences in stimulus length. An apparent anomaly in the 0.5 kHz TBOAE measurements is that they had a maximum level at around 1 kHz, even though the
[1] D.T. Kemp, Stimulated acoustic emissions from within the human auditory system, J. Acoust. Soc. Am. 64 (Nov (5)) (1978) 1386–1391. [2] M.P. Gorga, S.T. Neely, B.M. Bergman, K.L. Beauchaine, J.R. Kaminski, J. Peters, et al., A comparison of transient-evoked and distortion product otoacoustic emissions in normal-hearing and hearing-impaired subjects, J. Acoust. Soc. Am. 94 (Nov (5)) (1993) 2639–2648. [3] M. Suckfu¨ll, S. Schneeweiss, A. Dreher, K. Schorn, Evaluation of TEOAE and DPOAE measurements for the assessment of auditory thresholds in sensorineural hearing loss, Acta Otolaryngol. 116 (Jul (4)) (1996) 528–533. [4] W.A. Harrison, S.J. Norton, Characteristics of transient evoked otoacoustic emissions in normal-hearing and hearing-impaired children, Ear Hear. 20 (Feb (1)) (1999) 75–86. [5] H. Skarz˙yn´ski, A. Piotrowska, Screening for pre-school and school-age hearing problems: European consensus statement, Int. J. Pediatr. Otorhinolaryngol. 76 (1) (2012) 120–121. [6] C. Driscoll, J. Kei, B. McPherson, Outcomes of transient evoked otoacoustic emission testing in 6-year-old school children: a comparison with pure tone screening and tympanometry, Int. J. Pediatr. Otorhinolaryngol. 57 (1) (2001) 67–76. [7] D.G. Balatsouras, A. Kaberos, G. Kloutsos, N.C. Economou, V. Sakellariadis, A. Fassolis, et al., Correlation of transiently evoked to distortion-product otoacoustic emission measures in healthy children, Int. J. Pediatr. Otorhinolaryngol. 70 (1) (2006) 89–93. [8] L. S´liwa, S. Hatzopoulos, K. Kochanek, A. Piłka, A. Senderski, P.H. Skarz˙yn´ski, A comparison of audiometric and objective methods in hearing screening of school children. A preliminary study, Int. J. Pediatr. Otorhinolaryngol. 75 (4) (2011) 483–488. [9] P.J. Bray, D. Kemp, An advanced cochlear echo technique suitable for infant screening, Br. J. Audiol. 21 (1987) 191–204. [10] V. Lichtenstein, D.R. Stapells, Frequency-specific identification of hearing loss using transient-evoked otoacoustic emissions to clicks and tones, Hear. Res. 98 (Sep (1–2)) (1996) 125–136. [11] B. McPherson, S.F. Li, B.X. Shi, J.L. Tang, B.Y. Wong, Neonatal hearing screening: evaluation of tone-burst and click-evoked otoacoustic emission test criteria, Ear Hear. 27 (Jun (3)) (2006) 256–262. [12] W.W. Jedrzejczak, K. Kochanek, B. Trzaskowski, E. Pilka, P.H. Skarzynski, H. Skarzynski, Tone-burst and click-evoked otoacoustic emissions in subjects with hearing loss above 0.25, 0.5, and 1 kHz, Ear Hear. 33 (Nov–Dec (6)) (2012) 757–767. [13] W.W. Jedrzejczak, A. Piotrowska, K. Kochanek, L. Sliwa, H. Skarzynski, Lowfrequency otoacoustic emissions in schoolchildren measured by two commercial devices, Int. J. Pediatr. Otorhinolaryngol. 77 (10) (2013) 1724–1728. [14] H. Skarzynski, K. Kochanek, A. Senderski, P.H. Skarzynski, M. Ludwikowski, M. Kopaczewski, et al., Organization of the hearing screening examinations in Polish schools in rural areas and small towns, Cochlear Implants Int. 11 (Jun (Suppl. 1)) (2010) 143–147. [15] P.H. Skarz˙yn´ski, A. Piłka, M. Ludwikowski, M.B. Skarz˙yn´ska, Comparison of the frequency of positive hearing screening outcomes in schoolchildren from Poland and other countries of Europe: Central Asia, and Africa, J. Hear. Sci. 4 (4) (2014) OA51–OA58. [16] S.J. Norton, J.E. Widen, Evoked otoacoustic emissions in normal-hearing infants and children: emerging data and issues, Ear Hear. 11 (Apr (2)) (1990) 121–127.
W.W. Jedrzejczak et al. / International Journal of Pediatric Otorhinolaryngology 79 (2015) 1310–1315 [17] B.A. Prieve, T.S. Fitzgerald, L.E. Schulte, Basic characteristics of click-evoked otoacoustic emissions in infants and children, J. Acoust. Soc. Am. 102 (Nov (5 Pt 1)) (1997) 2860–2870. [18] D.T. Kemp, Evidence of mechanical nonlinearity and frequency selective wave amplification in the cochlea, Arch. Otorhinolaryngol. 224 (1–2) (1979) 37–45. [19] T. Kuroda, Clinical investigation on spontaneous otoacoustic emission (SOAE) in 447 ears, Auris Nasus Larynx 34 (Mar (1)) (2007) 29–38. [20] G.R. Long, A. Tubis, Investigations into the nature of the association between threshold microstructure and otoacoustic emissions, Hear. Res. 36 (Nov (2–3)) (1988) 125–138. [21] W.W. Jedrzejczak, S. Hatzopoulos, L. Sliwa, E. Pilka, K. Kochanek, H. Skarzynski, Otoacoustic emissions in neonates measured with different acquisition protocols, Int. J. Pediatr. Otorhinolaryngol. 76 (3) (2012) 382–387. [22] T. Kuroda, S. Fukuda, E. Chida, M. Kashiwamura, M. Matsumura, R. Ohwatari, et al., Effects of spontaneous otoacoustic emissions on distortion product otoacoustic emission, Auris Nasus Larynx 28 (May (Suppl)) (2001) S33–S38. [23] R. Probst, A.C. Coats, G.K. Martin, B.L. Lonsbury-Martin, Spontaneous, click-, and toneburst-evoked otoacoustic emissions from normal ears, Hear. Res. 21 (3) (1986) 261–275. [24] G. Tognola, F. Grandori, P. Ravazzani, Time–frequency distributions of clickevoked otoacoustic emissions, Hear. Res. 106 (Apr (1–2)) (1997) 112–122. [25] A. Moleti, R. Sisto, G. Tognola, M. Parazzini, P. Ravazzani, F. Grandori, Otoacoustic emission latency, cochlear tuning, and hearing functionality in neonates, J. Acoust. Soc. Am. 118 (Sep (3 Pt 1)) (2005) 1576–1584. [26] W.W. Jedrzejczak, S. Hatzopoulos, A. Martini, K.J. Blinowska, Otoacoustic emissions latency difference between full-term and preterm neonates, Hear. Res. 231 (Sep (1–2)) (2007) 54–62.
1315
[27] A. Moleti, R. Sisto, A. Paglialonga, F. Sibella, L. Anteunis, M. Parazzini, et al., Transient evoked otoacoustic emission latency and estimates of cochlear tuning in preterm neonates, J. Acoust. Soc. Am. 124 (Nov (5)) (2008) 2984–2994. [28] W.W. Jedrzejczak, A. Lorens, A. Piotrowska, K. Kochanek, H. Skarzynski, Otoacoustic emissions evoked by 0.5 kHz tone bursts, J. Acoust. Soc. Am. 125 (5) (2009) 3158–3165. [29] W.W. Jedrzejczak, K.J. Blinowska, K. Kochanek, H. Skarzynski, Synchronized spontaneous otoacoustic emissions analyzed in a time–frequency domain, J. Acoust. Soc. Am. 124 (Dec (6)) (2008) 3720–3729. [30] S.G. Mallat, Z. Zhang, Matching pursuit with time–frequency dictionaries, IEEE Trans. Signal Process. 41 (1993) 3397–3415. [31] W.W. Jedrzejczak, K. Kwaskiewicz, K.J. Blinowska, K. Kochanek, H. Skarzynski, Use of the matching pursuit algorithm with a dictionary of asymmetric waveforms in the analysis of transient evoked otoacoustic emissions, J. Acoust. Soc. Am. 126 (2009) 3137–3146. [32] V.W. Zhang, B. McPherson, B.X. Shi, J.L. Tang, B.Y. Wong, Neonatal hearing screening: a combined click evoked and tone burst otoacoustic emission approach, Int. J. Pediatr. Otorhinolaryngol. 72 (3) (2008) 351–360. [33] G. Pavlovcinova´, J. Jakubı´kova´, T. Trnovec, K. Lancz, S. Wimmerova´, E. Sovcı´kova´, et al., A normative study of otoacoustic emissions, ear asymmetry, and gender effect in healthy schoolchildren in Slovakia, Int. J. Pediatr. Otorhinolaryngol. 74 (2) (2010) 173–177. [34] D. Groh, J. Pelanova, M. Jilek, J. Popelar, Z. Kabelka, J. Syka, Changes in otoacoustic emissions and high-frequency hearing thresholds in children and adolescents, Hear. Res. 212 (Feb (1–2)) (2006) 90–98.
