- Email: [email protected]
Prediction of frequency-specific hearing threshold using chirp auditory brainstem response in infants with hearing losses
International Journal of Pediatric Otorhinolaryngology 78 (2014) 812–816
Contents lists available at ScienceDirect
International Journal of Pediatric Otorhinolaryngology journal homepage: www.elsevier.com/locate/ijporl
Prediction of frequency-specific hearing threshold using chirp auditory brainstem response in infants with hearing losses Zheng-min Xu *, Wen-xia Cheng, Zhi-hong Yao Department of Otolaryngology-Head and Neck Surgery, Children’s Hospital of Fudan University, 399 Wan Yuan Road, Shanghai 201102, PR China
A R T I C L E I N F O
A B S T R A C T
Article history: Received 19 July 2013 Received in revised form 11 February 2014 Accepted 17 February 2014 Available online 26 February 2014
Objectives: To investigate the clinical usefulness of the LS-chirp auditory brainstem response for estimation of behavioral thresholds in young children with mild to severe hearing losses. Methods: 68 infants (136 ears) aged 6–12 months (mean age = 9.2 months) with bilateral mild to severe hearing losses were studied at Children’s Hospital of Fudan University. In all cases, the children were referred for LS-chirp ABR and visual reinforcement audiometric (VRA) measurements. The lowfrequency band chirp (LF-chirp) thresholds (frequency band = 0.1–0.85 kHz) were compared to the average VRA thresholds (frequency band = 0.25–0.5 kHz), whereas the high-frequency band chirp (HFchirp) thresholds (frequency band = 1–10 kHz) were compared to the average VRA thresholds (frequency band = 1–4 kHz) using statistical correlation coefficient values. Results: The LS-chirp ABR thresholds are very close to behavioral hearing levels. The mean differences between chirp-ABR and VRA thresholds were within 5 dB HL for all measurements. The smallest mean threshold difference (<3 dB HL) was obtained for the severe hearing loss group. The correlation coefficient values (r) were 0.97 at low-frequency and high-frequency bands. For each carrier frequency, the best correlations between chirp-ABR thresholds and VRA thresholds were obtained at VRA frequency of 0.25 kHz/LF-chirp (r = 0.98) and VRA frequency of 1 kHz/HF-chirp (r = 0.98). Conclusions: This study demonstrates the effectiveness using chirp-ABR predicted frequency-specific thresholds, especially of low and middle frequencies. LS-chirp ABR thresholds determined behavioral thresholds in patients with severe hearing losses were better than for mild hearing losses. The use of a chirp-ABR testing ensures higher sensitivity and accuracy than that of auditory stead-state evoked response (ASSR) for measuring frequency-specific thresholds in young children. ß 2014 Elsevier Ireland Ltd. All rights reserved.
Keywords: Chirp-ABR threshold VRA threshold Infants
1. Introduction Approximately one child in every 1000 new-borns has a hearing impairment in the well-baby nursery population. The incidence of such a hearing loss in babies admitted to the neonatal intensive care unit (NICU) is 20–40 times higher [1,2]. Early detection of the high incidence of hearing loss is very important for early diagnosis of hearing loss, and then in the benefits of early intervention focusing on the development of language. A study by Christine et al. [3] has shown that hearing-impaired children who receive intervention within 6 months post-conception age have significantly greater potential for speech and language development than do children receiving intervention at a later age.
* Corresponding author. Tel.: +86 21 64931926; fax: +86 21 64931901. E-mail addresses: [email protected], [email protected] (Z.-m. Xu). http://dx.doi.org/10.1016/j.ijporl.2014.02.020 0165-5876/ß 2014 Elsevier Ireland Ltd. All rights reserved.
It is a challenge for clinicians and researchers with current objective measuring methods to determine hearing thresholds precisely for infants in their first 6 months of life and consequently prescribing adequate levels of intervention. In recent years, the auditory stead-state evoked response (ASSR) has emerged as an alternative electrophysiological technique for measurement of objective thresholds in young children. However, several studies have documented high variability in relationship between ASSR and behavioral thresholds for the normal hearing and hearing-impaired infants [4–6]. ASSR thresholds in these subjects have been differently correlated to behavioral hearing level at low frequencies and high frequencies. In all evoked potentials, synchronization of the nerve fiber firing is important to generate a large amplitude auditory brain-stem response (ABR) waveform. The tonepips-ABR evoked potential is frequency specific since all but the frequency area of the tone pulse are desynchronized on the cochlea, especially of the low frequency. The traditional click-evoked ABR is difficult due to the well-know
Z.-m. Xu et al. / International Journal of Pediatric Otorhinolaryngology 78 (2014) 812–816
cochlear travel delay whereby frequency areas of the cochlea are stimulated one after the other [7]. The concept of the chirp was first applied to auditory electrophysiology by Shore and Nutall [8], and has since been studied intensively for its use within the auditory field by Elberling et al. [9]. The CE-chirp is a family of stimuli designed to compensate for this cochlear travel delay and provide enhanced neural synchronicity [10]. The stimuli family includes a broadband CE-chirp as well as narrow band CE-chirp stimuli for frequency specific testing. In limitation of the CE-chirp application, it is adequate at lower levels of stimulation (<60 dB nHL). In 2010, Elberling and Don [11] have developed chirp stimulus that mean the level-specific chirp-stimulus (LS-chirp). LS-chirp is a broadband chirp that was used with a magnitude spectrum corresponding to that of the click. The LS-chirp will produce larger ABR amplitudes and better resolution. It should be applied at higher levels (>60 dB nHL). So, LS-chirp evoked auditory brainstem response (chirp-ABR) has been regarded as a more synchronous response than the tonepips-ABR and the click-evoked ABR, referring to the belief that the chirp stimulates lower-, mid-, and higher-frequency regions of the cochlea simultaneously [12,13,9]. The chirp-stimulus generates an optimized synchronous excitation of neurons in all regions of the basilar membrane. The stimulus structure allows modification to excite predefined regions in the cochlear [14]. However, the chirp-ABRs have been investigated in few studies involving young children with varying degrees of hearing loss. Only some researchers [10] reported that in this study a variety of tools were used to analyze the synchronicity of ABRs evoked by CEchirp- and click-stimuli at 40 dB HL in 32 normal hearing subjects aged 18–55 years. Our recent study examines the clinical usefulness of the upward LS-chirp stimulus developed by Elberling and Don [11] in young patients. We investigate the relationship LS-chirp-ABR and behavioral visual reinforcement audiometric (VRA) hearing thresholds for frequency-specific information in a group of infants with varying degrees of sensorineural hearing loss. In particular, we examine how accurately the prediction formulae (which were based on chirp-ABR/VRA threshold correlation data) estimated hearing threshold in order to find out if it is possible to be used as the clinical application with regard to diagnostic audiology in young children. 2. Subjects and methods 2.1. Subjects 68 infants aged 6–12 months (mean age = 9.2 months) with bilateral mild to severe hearing losses were studied at Children’s Hospital of Fudan University. In all cases, the children were referred for chirp-ABR and VRA measurements following failure on a click-evoked ABR screen assessment (ABR threshold >40 dB nHL–90 dB nHL in both ears). Subjects were included only if they showed no evidence of external and middle ear pathology (normal multiple probe tone tympanometric results (GSI TympStar) and normal otoscopy in both ears). Children with deteriorating hearing loss (confirmed by repeat evoked potential and behavioral assessments) were excluded. In all subjects LS-chirp ABR and VRA thresholds were measured binaurally. VRA data (one ear) was correlated with same ear-specific chirp-ABR data. 2.2. Visual reinforcement audiometric (VRA) measuring VRA measurements were established for each ear in a soundfield room. The child sat in a chair, and then the test assessed
813
hearing using earphones in each ear separately. When a sound at a specific frequency was presented, the child’s eye-shift or head-turn response toward the sound source was rewarded by activation of a lighted mechanical toy. The behavioral thresholds were obtained using conditioned audiometric techniques (visual response audiometry) with earphones and the 10 dB L down, 5 dB HL up threshold each procedure. Our clinical audiometer (Dianostic Audiometer AD229e) was used to generate warble tones at the frequency range of 0.25-, 0.5-, 1-, 2-, and 4-kHz. 2.3. Chirp auditory brain-stem response measuring The chirp-ABR was measured with ERA-System Corona Equipment using a pair of ER-3A insert earphones. The level-specific chirpstimulus (LS-chirp) was a broadband chirp that was used with a magnitude spectrum corresponding to that of the click. Its nominal edge frequencies were 0.1 and 10 kHz. LS-chirp was modified in such a way that an equalized acoustical frequency spectrum is obtained from the output of the ER-3A earphone to excite predefine regions in the cochlea. The equations defining the chirp were calculated to be the inverse of the delay-line characteristic of cochlear partition on the basis of the linear cochlea model. The (acoustic) magnitude spectrum decreases continuously with increasing frequency. Therefore the LS-chirp was given two frequency-spectrum bands that corresponded to the inverse of a model of the amplitude– frequency response of the ABR. The application of the low-frequency bands chirp (LF-chirp = 0.1–0.85 kHz) and the high-frequency band chirp (HF-chirp = 1–10 kHz) provided the possibility of an objective assessment of the hearing threshold in the respective frequency range. The chirp-stimuli were presented using maximum level of 100 dB HL with a 21.1/s rate. Responses were recorded for 20 ms following chirp onset. The chirp-ABR thresholds were obtained using a 10 dB down, 5 dB up search procedure with a starting level of approximately 60 dB HL. Threshold was defined as the lowest stimulus level at which peak wave V could be visualized in the two frequency bands (the LF-chirp threshold found at wave V peaklatency about 13–14 ms and the HF-chirp threshold found at wave V peak-latency about 6–7 ms). 2.4. Prediction of hearing threshold using statistic The LF-chirp ABR thresholds (frequency band = 0.1–0.85 kHz) were compared to the average VRA thresholds (frequency band = 0.25–0.5 kHz), whereas the HF-chirp thresholds (frequency band = 1–10 kHz) were compared to the average VRA thresholds (frequency band = 1–4 kHz) using statistical correlation coefficient values. The Pearson r value for each of the test frequency band was calculated. 3. Results 68 infants (136 ears) with bilateral mild to severe hearing losses were tested with two different measuring procedures (VRA and LSchirp ABR). The VRA thresholds were measured at 0.25, 0.5, 1, 2, and 4 kHz in 136 ears of 68 infants, whereas the LS-chirp ABR thresholds were also performed in 136 ears of 68 infants at the LFchirp (0.1–0.85 kHz) and the HF-chirp (1–10 kHz). The mean chirpABR thresholds at the LF-chirp and HF-chirp bands were compared with mean VRA thresholds at the frequency bands of the 0.25– 0.5 kHz and the 1–4 kHz. The LS-chirp ABR thresholds overestimated the VRA thresholds in all instances. The mean differences between chirp-ABR and VRA thresholds were within 5 dB HL for all measurements. Overall, the data show a strong relationship between LS-chirp ABR thresholds and VRA thresholds. The correlation coefficient values (r) were 0.97 at low-frequency and high-frequency bands.
Z.-m. Xu et al. / International Journal of Pediatric Otorhinolaryngology 78 (2014) 812–816
814
Table 1 Mean VRA thresholds and LS-chirp ABR estimates for category 1 (45–65 dB HL) and category 2 (70–90 dB HL) groups (no. = 136 ears). Category
Low frequency band
High frequency band
Frequency (kHz)
(0.25–0.5 kHz vs 0.1–0.85 kHz)
(1–4 kHz vs 1–10 kHz)
L-chirp-freq. (0.1–0.85)
Mean
No.
Mean threshold dB SD
No.
threshold dB SD Category 1 VRA Chirp-ABR Difference
50.5 5.5 55.1 6.3 4.6
58 58 58
50.3 5.4 55.1 6.3 4.8
58 58 58
Category 2 VRA Chirp-ABR Difference
78.3 6.1 80.5 6.5 2.2
88 88 88
78.1 6.4 80.5 6.5 2.4
88 88 88
To evaluate further the mean difference between chirp-ABR and VRA thresholds, the data were divided into two categories based on chirp-ABR mean threshold-level. The groups were defined as follows: category 1, mean 45–65 dB HL and category 2, mean 70– 90 dB HL. Table 1 gives the mean thresholds for both procedures as obtained from each test frequency band and each chirp-ABR category. The smallest mean threshold difference (<3 dB HL) was obtained for category 2. In contrast, the infants with chirp-ABR thresholds less than 70 dB (category 1) were quite different about 5 dB. The results indicated better LS-chirp ABR threshold estimations for the severe group (category 2) at all frequencies (Table 1). Fig. 1 shows a high degree of correlation between Chirp-ABR thresholds in both low- and high-frequency audiometry bands in young patient with severe hearing loss. Fig. 2 shows the relationship between the chirp-ABR and VRA hearing thresholds obtained at each frequency band (LF-chirp-ABR threshold vs mean VRA threshold for 0.25 and 0.5 kHz and HFchirp-ABR threshold vs mean VRA threshold for 1, 2, and 4 kHz) for all of the ears tested (n = 136). As with the scatter plot, the data fitted to the line that could mean the great linear regression, which reflect the fact that chirp-ABR thresholds are very close to behavioral hearing levels. To further investigate the relationship between the LS-chirp ABR and VRA thresholds for frequency-specific information, the Pearson r values were calculate for each test frequency and each chirp-ABR category. For each carrier frequency, the best correlations between chirp-ABR thresholds and VRA thresholds were obtained at VRA frequency of 0.25 kHz/LF-chirp-ABR (r = 0.98) and VRA frequency of 1 kHz/HF-chirp-ABR (r = 0.98). The results indicated better LS-chirp ABR threshold estimations for all infants
[(Fig._1)TD$IG]
Table 2 The frequency distribution of the obtained behavioral and LS-chirp-ABR thresholds for correlation coefficient values in categories 1 and 2.
VRA freq. Category 1 Pearson r Category 2 Pearson r
H-chirp-freq. (1–10)
0.25
0.5
1
2
4
0.87
0.71
0.88
0.72
0.66
0.91
0.88
0.91
0.88
0.90
with hearing losses at frequencies of 0.25 kHz and 1 kHz that were within the speech frequency band. For infants with chirp-ABR thresholds in category 2 (severe hearing losses), the behavioral/ chirp-ABR threshold relationship was stronger than that in category 1 (mild hearing losses) at 0.25 kHz and 1 kHz (Table 2). In the LS-chirp ABR thresholds corresponding to the configuration of the behavioral audiogram, F test analyses for all ears showed no significant difference between chirp-ABR and VRA thresholds (p > 0.05) in category 1 (mild hearing losses) and category 2 (severe hearing losses). This finding further revealed a clear relationship still existed between chirp-ABR and behavioral level. 4. Discussion The results of our study compare LS-chirp ABR thresholds with behavioral thresholds obtained by means of visual reinforce audiometric (VRA) for infants with mild to severe hearing loss. All LS-chirp ABR thresholds were overestimating behavioral thresholds. The overall, the data show a strong relationship between LSchirp ABR thresholds and VRA thresholds. Correlation coefficient value (r) was 0.97. The mean difference between chirp-ABR and behavioral thresholds was 3–5 dB. The majority of chirp-ABR thresholds (65%) were recorded between 70 and 90 dB (category 2) from the corresponding behavioral thresholds, and the smallest threshold difference was within 3 dB. In contrast, the infants with chirp-ABR thresholds less than 70 dB (category 1) differed with about 5 dB. This result indicates a closer correlation between chirpABR and severe behavioral thresholds than for mild thresholds (less than 70 dB HL). These improved correlations for more severe degrees of hearing loss were in agreement with ASSR/behavioral threshold relationship in Rance’s study [5]. This increased sensitivity of the LS-chirp ABR to more severe degrees of hearing loss may be related to recruitment associated with cochlear hearing impairment. Elberling et al. [9] explain that the chirpstimulus generates an optimized synchronous excitation of
Fig. 1. A high degree of correlation between chirp-ABR thresholds in both low- and high-frequency audiometry bands in young patient with severe hearing loss.
[(Fig._2)TD$IG]
Z.-m. Xu et al. / International Journal of Pediatric Otorhinolaryngology 78 (2014) 812–816 100
Mean VRA threshold(dB HL) at 1-4kHz
B 90
80
70
60
50
40 30
40
50
60
70
80
90
100
Mean VRA threshold(dB HL) at 0.25-0.5kHz
H-Chirp-ABR threshold(dB HL) 100
A 90
80
70
60
50
40 40
50
60
70
80
90
100
L-Chirp-ABR threshold(dB HL) Fig. 2. The distribution of LS-chirp-ABR/behavioral threshold comparisons at each frequency band for all of the ears tested (n = 136).
neurons in all regions of the basilar membrane. The stimulus structure allows modification to excite predefined regions in the cochlea. At high intensity LS-chirp stimulus may be optimally suited as more sensitive ABR measurements in infants with severe hearing losses. Our data showed that LS-chirp ABR thresholds determined behavioral thresholds (mean threshold difference between 3 and 5 dB) were better than ASSR predicted by Swanepoel et al. data [6]. This may be that the LS-chirp produce larger ABR amplitudes and better resolution [9]. We interpret these results as a compatible with the concept of better hair-cell synchronization. As with the frequency-specific information, the data with the scatter plot (Fig. 2) fit to the linear regression, which reflect the fact that LS-chirp ABR thresholds are close to behavioral hearing levels at LF-chirp/VRA (0.25–0.5 kHz) and HF-chirp/VRA (1– 4 kHz). Pearson correlation coefficients were calculated to assess the relationship between chirp-ABR frequency-band (LF-chirp and HF-chirp) and behavioral thresholds at each frequency. Our data showed that the best correlations between LS-chirp ABR thresholds and VRA thresholds were obtained at VRA frequency of 0.25 kHz/LF-chirp-ABR (r = 0.98) and VRA frequency of 1 kHz/HFchirp-ABR (r = 0.98). These results imply better LS-chirp ABR threshold estimations for infants with hearing losses at frequencies of 0.25 kHz and 1 kHz. They are very important low-middle
815
speech frequency bands for clinical application. However, Swanepoel et al. [6] reported that the largest mean difference between ASSR and behavioral thresholds was at 0.5 kHz, and it was the only frequency indicating a statistically significant difference between ASSR and behavioral thresholds. Problems in estimating 0.5 kHz ASSR have been reported previously [15,16], and could result from the fact that the low-frequency-evoked response has a greater intrinsic jitter, due to neural asynchrony. Another reason may be that stimulus protocols for amplitudemodulated tone at 0.5 kHz are not yet optimal. So, the ASSR predicted thresholds are not suitable in estimating low frequency thresholds, especially of 0.5 kHz. Our result revealed more sensitive LS-chirp ABR measurement in estimating low-middle frequency thresholds (0.25–1 kHz). There may be factors which explain the higher correlation between the LS-chirp ABR and behavioral thresholds at low-middle frequencies. The LS-chirp evoked ABR has been regarded as a more synchronous response than the click-evoked ABR, referring to the belief that the chirp stimulates lower-, mid-, and higher-frequency regions of the cochlea simultaneously, especially of lower-, middle-frequency. Don et al. [17] found that the low-chirp-stimulus evoked a significantly larger amplitude of the ABR than that of a click-ABR. Table 2 show the frequency distribution of the obtained behavioral and chirp-ABR thresholds in categories 1 and 2. These data further confirmed the best chirp-ABR threshold estimations for infants with sever hearing losses at frequencies of 0.25 kHz and 1 kHz. In this study, there are some limitations of using the LS-chirp ABR that are considered in the following. The chirp stimulus designed from two bands (LF-chirp = 0.1–0.85 kHz and HFchirp = 1–10 kHz) that could not be precisely correspond to VRA frequency bands (0.25–0.5 kHz and 1–4 kHz). Thus, the chirp may represent the frequencies of 0.25 kHz, 0.5 kHz, 1 kHz, 2 kHz, and 4 kHz that are correspond to VRA frequencies. The problem is also that chirp stimulus fixed by instrument is split between 0.85 kHz and 1 kHz in this study. When the instrument (stimulus parameters) is improved, we will further study in this issue for relationship between VRA and LS-chirp ABR thresholds. In conclusion, this study demonstrates the effective implementation of LS-chirp ABR predicted frequency-specific thresholds, especially of low-middle-frequency. LS-chirp ABR thresholds determined behavioral thresholds in patients with severe hearing losses were better than for mild hearing losses. The use of a chirpABR testing ensures higher sensitivity and accuracy than that of ASSR for measuring frequency-specific thresholds. It is possible to be used as the clinical application with regard to diagnostic audiology in young children. References [1] American Academy of Pediatrics, Newborn and infant hearing loss: detection and intervention (RE9846), Pediatrics 6 (1999) 527–530. [2] L.E. Dalzell, M.S. Orlando, C. Seeger, 3-Stage newborn hearing screening reduces failures at discharge, in: Poster presented at American Academy of Audiology, Salt Lake City, UT, 1996. [3] Y.I. Christine, L.S. Allison, K.C. Diane, L.M. Albert, Language of early- and lateridentified children with hearing loss, Pediatrics 102 (1998) 1161–1170. [4] F.M. Riberiro, R.M. Carallo, A.M. Marcoux, Auditory steady-state evoked responses for preterm and term neonates, Audiol. Neurotol. 15 (2010) 97–110. [5] G. Rance, F. Rickards, Prediction of hearing threshold in infants using auditory steady-state evoked potentials, J. Am. Acad. Audiol. 13 (2002) 236–245. [6] D.W. Swanepoel, R. Hugo, R. Roode, Auditory steady-state responses for children with severe to profound hearing loss, Arch. Otolaryngol. Head Neck Surg. 130 (2004) 531–535. [7] M. Cebulla, E. Sturzebecher, C. Elbering, J. Muller, New click like stimuli for hearing testing, J. Am. Acad. Audiol. 18 (2007) 725–738. [8] S.E. Shore, A.L. Nuttal, High-synchrony cochlear compound action evoked by rising frequency-swept tone bursts, J. Acoust. Soc. Am. 78 (1985) 1283–1295. [9] C. Elberling, M. Don, M. Cebulla, E. Sturzebecher, Auditory steady-state responses to chirp stimuli based on cochlear traveling wave delay, J. Acoust. Soc. Am. 122 (2007) 2772–2785.
816
Z.-m. Xu et al. / International Journal of Pediatric Otorhinolaryngology 78 (2014) 812–816
[10] C. Elberling, M. Don, Auditory brainstem responses to a chirp stimulus designed from derived-band latencies in normal-hearing subjects, J. Acoust. Soc. Am. 124 (2008) 3022–3037. [11] C. Elberling, M. Don, A direct approach for the design of chirp stimuli used for the recording of auditory brainstem responses, J. Acoust. Soc. Am. 128 (2010) 2955– 2964. [12] M. Cebulla, C. Elberling, Auditory brain stem responses evoked by different chirps based on different delay models, J. Am. Acad. Audiol. 21 (2010) 452–460. [13] T. Dau, O. Wegner, V. Mellert, B. Kollmeier, Auditory brainstem responses with optimized chirp signals compensating basilar-membrane dispersion, J. Acoust. Soc. Am. 107 (2000) 1530–1540.
[14] M.A. Petoe, A.P. Bradley, W.J. Wilson, On chirp stimuli and neural synchrony in the suprathreshold auditory brainstem response, J. Acoust. Soc. Am. 128 (2010) 235– 246. [15] O.G. Lins, T.W. Piton, B.L. Boucher, Frequency-specific audiometry using steadystate responses, Ear Hear. 17 (1996) 81–96. [16] T.W. Piton, M.S. John, A. Dimitrijevic, D. Purcell, Human auditory stead-state responses, Int. J. Audiol. 42 (2003) 177–221. [17] M. Don, C. Elberling, E. Maloff, Input and output compensation for the cochlear travelling wave delay in wide-band ABR recordings: implications for small acoustic tumor detection, J. Am. Acad. Audiol. 20 (2009) 99–108.
Contents lists available at ScienceDirect
International Journal of Pediatric Otorhinolaryngology journal homepage: www.elsevier.com/locate/ijporl
Prediction of frequency-specific hearing threshold using chirp auditory brainstem response in infants with hearing losses Zheng-min Xu *, Wen-xia Cheng, Zhi-hong Yao Department of Otolaryngology-Head and Neck Surgery, Children’s Hospital of Fudan University, 399 Wan Yuan Road, Shanghai 201102, PR China
A R T I C L E I N F O
A B S T R A C T
Article history: Received 19 July 2013 Received in revised form 11 February 2014 Accepted 17 February 2014 Available online 26 February 2014
Objectives: To investigate the clinical usefulness of the LS-chirp auditory brainstem response for estimation of behavioral thresholds in young children with mild to severe hearing losses. Methods: 68 infants (136 ears) aged 6–12 months (mean age = 9.2 months) with bilateral mild to severe hearing losses were studied at Children’s Hospital of Fudan University. In all cases, the children were referred for LS-chirp ABR and visual reinforcement audiometric (VRA) measurements. The lowfrequency band chirp (LF-chirp) thresholds (frequency band = 0.1–0.85 kHz) were compared to the average VRA thresholds (frequency band = 0.25–0.5 kHz), whereas the high-frequency band chirp (HFchirp) thresholds (frequency band = 1–10 kHz) were compared to the average VRA thresholds (frequency band = 1–4 kHz) using statistical correlation coefficient values. Results: The LS-chirp ABR thresholds are very close to behavioral hearing levels. The mean differences between chirp-ABR and VRA thresholds were within 5 dB HL for all measurements. The smallest mean threshold difference (<3 dB HL) was obtained for the severe hearing loss group. The correlation coefficient values (r) were 0.97 at low-frequency and high-frequency bands. For each carrier frequency, the best correlations between chirp-ABR thresholds and VRA thresholds were obtained at VRA frequency of 0.25 kHz/LF-chirp (r = 0.98) and VRA frequency of 1 kHz/HF-chirp (r = 0.98). Conclusions: This study demonstrates the effectiveness using chirp-ABR predicted frequency-specific thresholds, especially of low and middle frequencies. LS-chirp ABR thresholds determined behavioral thresholds in patients with severe hearing losses were better than for mild hearing losses. The use of a chirp-ABR testing ensures higher sensitivity and accuracy than that of auditory stead-state evoked response (ASSR) for measuring frequency-specific thresholds in young children. ß 2014 Elsevier Ireland Ltd. All rights reserved.
Keywords: Chirp-ABR threshold VRA threshold Infants
1. Introduction Approximately one child in every 1000 new-borns has a hearing impairment in the well-baby nursery population. The incidence of such a hearing loss in babies admitted to the neonatal intensive care unit (NICU) is 20–40 times higher [1,2]. Early detection of the high incidence of hearing loss is very important for early diagnosis of hearing loss, and then in the benefits of early intervention focusing on the development of language. A study by Christine et al. [3] has shown that hearing-impaired children who receive intervention within 6 months post-conception age have significantly greater potential for speech and language development than do children receiving intervention at a later age.
* Corresponding author. Tel.: +86 21 64931926; fax: +86 21 64931901. E-mail addresses: [email protected], [email protected] (Z.-m. Xu). http://dx.doi.org/10.1016/j.ijporl.2014.02.020 0165-5876/ß 2014 Elsevier Ireland Ltd. All rights reserved.
It is a challenge for clinicians and researchers with current objective measuring methods to determine hearing thresholds precisely for infants in their first 6 months of life and consequently prescribing adequate levels of intervention. In recent years, the auditory stead-state evoked response (ASSR) has emerged as an alternative electrophysiological technique for measurement of objective thresholds in young children. However, several studies have documented high variability in relationship between ASSR and behavioral thresholds for the normal hearing and hearing-impaired infants [4–6]. ASSR thresholds in these subjects have been differently correlated to behavioral hearing level at low frequencies and high frequencies. In all evoked potentials, synchronization of the nerve fiber firing is important to generate a large amplitude auditory brain-stem response (ABR) waveform. The tonepips-ABR evoked potential is frequency specific since all but the frequency area of the tone pulse are desynchronized on the cochlea, especially of the low frequency. The traditional click-evoked ABR is difficult due to the well-know
Z.-m. Xu et al. / International Journal of Pediatric Otorhinolaryngology 78 (2014) 812–816
cochlear travel delay whereby frequency areas of the cochlea are stimulated one after the other [7]. The concept of the chirp was first applied to auditory electrophysiology by Shore and Nutall [8], and has since been studied intensively for its use within the auditory field by Elberling et al. [9]. The CE-chirp is a family of stimuli designed to compensate for this cochlear travel delay and provide enhanced neural synchronicity [10]. The stimuli family includes a broadband CE-chirp as well as narrow band CE-chirp stimuli for frequency specific testing. In limitation of the CE-chirp application, it is adequate at lower levels of stimulation (<60 dB nHL). In 2010, Elberling and Don [11] have developed chirp stimulus that mean the level-specific chirp-stimulus (LS-chirp). LS-chirp is a broadband chirp that was used with a magnitude spectrum corresponding to that of the click. The LS-chirp will produce larger ABR amplitudes and better resolution. It should be applied at higher levels (>60 dB nHL). So, LS-chirp evoked auditory brainstem response (chirp-ABR) has been regarded as a more synchronous response than the tonepips-ABR and the click-evoked ABR, referring to the belief that the chirp stimulates lower-, mid-, and higher-frequency regions of the cochlea simultaneously [12,13,9]. The chirp-stimulus generates an optimized synchronous excitation of neurons in all regions of the basilar membrane. The stimulus structure allows modification to excite predefined regions in the cochlear [14]. However, the chirp-ABRs have been investigated in few studies involving young children with varying degrees of hearing loss. Only some researchers [10] reported that in this study a variety of tools were used to analyze the synchronicity of ABRs evoked by CEchirp- and click-stimuli at 40 dB HL in 32 normal hearing subjects aged 18–55 years. Our recent study examines the clinical usefulness of the upward LS-chirp stimulus developed by Elberling and Don [11] in young patients. We investigate the relationship LS-chirp-ABR and behavioral visual reinforcement audiometric (VRA) hearing thresholds for frequency-specific information in a group of infants with varying degrees of sensorineural hearing loss. In particular, we examine how accurately the prediction formulae (which were based on chirp-ABR/VRA threshold correlation data) estimated hearing threshold in order to find out if it is possible to be used as the clinical application with regard to diagnostic audiology in young children. 2. Subjects and methods 2.1. Subjects 68 infants aged 6–12 months (mean age = 9.2 months) with bilateral mild to severe hearing losses were studied at Children’s Hospital of Fudan University. In all cases, the children were referred for chirp-ABR and VRA measurements following failure on a click-evoked ABR screen assessment (ABR threshold >40 dB nHL–90 dB nHL in both ears). Subjects were included only if they showed no evidence of external and middle ear pathology (normal multiple probe tone tympanometric results (GSI TympStar) and normal otoscopy in both ears). Children with deteriorating hearing loss (confirmed by repeat evoked potential and behavioral assessments) were excluded. In all subjects LS-chirp ABR and VRA thresholds were measured binaurally. VRA data (one ear) was correlated with same ear-specific chirp-ABR data. 2.2. Visual reinforcement audiometric (VRA) measuring VRA measurements were established for each ear in a soundfield room. The child sat in a chair, and then the test assessed
813
hearing using earphones in each ear separately. When a sound at a specific frequency was presented, the child’s eye-shift or head-turn response toward the sound source was rewarded by activation of a lighted mechanical toy. The behavioral thresholds were obtained using conditioned audiometric techniques (visual response audiometry) with earphones and the 10 dB L down, 5 dB HL up threshold each procedure. Our clinical audiometer (Dianostic Audiometer AD229e) was used to generate warble tones at the frequency range of 0.25-, 0.5-, 1-, 2-, and 4-kHz. 2.3. Chirp auditory brain-stem response measuring The chirp-ABR was measured with ERA-System Corona Equipment using a pair of ER-3A insert earphones. The level-specific chirpstimulus (LS-chirp) was a broadband chirp that was used with a magnitude spectrum corresponding to that of the click. Its nominal edge frequencies were 0.1 and 10 kHz. LS-chirp was modified in such a way that an equalized acoustical frequency spectrum is obtained from the output of the ER-3A earphone to excite predefine regions in the cochlea. The equations defining the chirp were calculated to be the inverse of the delay-line characteristic of cochlear partition on the basis of the linear cochlea model. The (acoustic) magnitude spectrum decreases continuously with increasing frequency. Therefore the LS-chirp was given two frequency-spectrum bands that corresponded to the inverse of a model of the amplitude– frequency response of the ABR. The application of the low-frequency bands chirp (LF-chirp = 0.1–0.85 kHz) and the high-frequency band chirp (HF-chirp = 1–10 kHz) provided the possibility of an objective assessment of the hearing threshold in the respective frequency range. The chirp-stimuli were presented using maximum level of 100 dB HL with a 21.1/s rate. Responses were recorded for 20 ms following chirp onset. The chirp-ABR thresholds were obtained using a 10 dB down, 5 dB up search procedure with a starting level of approximately 60 dB HL. Threshold was defined as the lowest stimulus level at which peak wave V could be visualized in the two frequency bands (the LF-chirp threshold found at wave V peaklatency about 13–14 ms and the HF-chirp threshold found at wave V peak-latency about 6–7 ms). 2.4. Prediction of hearing threshold using statistic The LF-chirp ABR thresholds (frequency band = 0.1–0.85 kHz) were compared to the average VRA thresholds (frequency band = 0.25–0.5 kHz), whereas the HF-chirp thresholds (frequency band = 1–10 kHz) were compared to the average VRA thresholds (frequency band = 1–4 kHz) using statistical correlation coefficient values. The Pearson r value for each of the test frequency band was calculated. 3. Results 68 infants (136 ears) with bilateral mild to severe hearing losses were tested with two different measuring procedures (VRA and LSchirp ABR). The VRA thresholds were measured at 0.25, 0.5, 1, 2, and 4 kHz in 136 ears of 68 infants, whereas the LS-chirp ABR thresholds were also performed in 136 ears of 68 infants at the LFchirp (0.1–0.85 kHz) and the HF-chirp (1–10 kHz). The mean chirpABR thresholds at the LF-chirp and HF-chirp bands were compared with mean VRA thresholds at the frequency bands of the 0.25– 0.5 kHz and the 1–4 kHz. The LS-chirp ABR thresholds overestimated the VRA thresholds in all instances. The mean differences between chirp-ABR and VRA thresholds were within 5 dB HL for all measurements. Overall, the data show a strong relationship between LS-chirp ABR thresholds and VRA thresholds. The correlation coefficient values (r) were 0.97 at low-frequency and high-frequency bands.
Z.-m. Xu et al. / International Journal of Pediatric Otorhinolaryngology 78 (2014) 812–816
814
Table 1 Mean VRA thresholds and LS-chirp ABR estimates for category 1 (45–65 dB HL) and category 2 (70–90 dB HL) groups (no. = 136 ears). Category
Low frequency band
High frequency band
Frequency (kHz)
(0.25–0.5 kHz vs 0.1–0.85 kHz)
(1–4 kHz vs 1–10 kHz)
L-chirp-freq. (0.1–0.85)
Mean
No.
Mean threshold dB SD
No.
threshold dB SD Category 1 VRA Chirp-ABR Difference
50.5 5.5 55.1 6.3 4.6
58 58 58
50.3 5.4 55.1 6.3 4.8
58 58 58
Category 2 VRA Chirp-ABR Difference
78.3 6.1 80.5 6.5 2.2
88 88 88
78.1 6.4 80.5 6.5 2.4
88 88 88
To evaluate further the mean difference between chirp-ABR and VRA thresholds, the data were divided into two categories based on chirp-ABR mean threshold-level. The groups were defined as follows: category 1, mean 45–65 dB HL and category 2, mean 70– 90 dB HL. Table 1 gives the mean thresholds for both procedures as obtained from each test frequency band and each chirp-ABR category. The smallest mean threshold difference (<3 dB HL) was obtained for category 2. In contrast, the infants with chirp-ABR thresholds less than 70 dB (category 1) were quite different about 5 dB. The results indicated better LS-chirp ABR threshold estimations for the severe group (category 2) at all frequencies (Table 1). Fig. 1 shows a high degree of correlation between Chirp-ABR thresholds in both low- and high-frequency audiometry bands in young patient with severe hearing loss. Fig. 2 shows the relationship between the chirp-ABR and VRA hearing thresholds obtained at each frequency band (LF-chirp-ABR threshold vs mean VRA threshold for 0.25 and 0.5 kHz and HFchirp-ABR threshold vs mean VRA threshold for 1, 2, and 4 kHz) for all of the ears tested (n = 136). As with the scatter plot, the data fitted to the line that could mean the great linear regression, which reflect the fact that chirp-ABR thresholds are very close to behavioral hearing levels. To further investigate the relationship between the LS-chirp ABR and VRA thresholds for frequency-specific information, the Pearson r values were calculate for each test frequency and each chirp-ABR category. For each carrier frequency, the best correlations between chirp-ABR thresholds and VRA thresholds were obtained at VRA frequency of 0.25 kHz/LF-chirp-ABR (r = 0.98) and VRA frequency of 1 kHz/HF-chirp-ABR (r = 0.98). The results indicated better LS-chirp ABR threshold estimations for all infants
[(Fig._1)TD$IG]
Table 2 The frequency distribution of the obtained behavioral and LS-chirp-ABR thresholds for correlation coefficient values in categories 1 and 2.
VRA freq. Category 1 Pearson r Category 2 Pearson r
H-chirp-freq. (1–10)
0.25
0.5
1
2
4
0.87
0.71
0.88
0.72
0.66
0.91
0.88
0.91
0.88
0.90
with hearing losses at frequencies of 0.25 kHz and 1 kHz that were within the speech frequency band. For infants with chirp-ABR thresholds in category 2 (severe hearing losses), the behavioral/ chirp-ABR threshold relationship was stronger than that in category 1 (mild hearing losses) at 0.25 kHz and 1 kHz (Table 2). In the LS-chirp ABR thresholds corresponding to the configuration of the behavioral audiogram, F test analyses for all ears showed no significant difference between chirp-ABR and VRA thresholds (p > 0.05) in category 1 (mild hearing losses) and category 2 (severe hearing losses). This finding further revealed a clear relationship still existed between chirp-ABR and behavioral level. 4. Discussion The results of our study compare LS-chirp ABR thresholds with behavioral thresholds obtained by means of visual reinforce audiometric (VRA) for infants with mild to severe hearing loss. All LS-chirp ABR thresholds were overestimating behavioral thresholds. The overall, the data show a strong relationship between LSchirp ABR thresholds and VRA thresholds. Correlation coefficient value (r) was 0.97. The mean difference between chirp-ABR and behavioral thresholds was 3–5 dB. The majority of chirp-ABR thresholds (65%) were recorded between 70 and 90 dB (category 2) from the corresponding behavioral thresholds, and the smallest threshold difference was within 3 dB. In contrast, the infants with chirp-ABR thresholds less than 70 dB (category 1) differed with about 5 dB. This result indicates a closer correlation between chirpABR and severe behavioral thresholds than for mild thresholds (less than 70 dB HL). These improved correlations for more severe degrees of hearing loss were in agreement with ASSR/behavioral threshold relationship in Rance’s study [5]. This increased sensitivity of the LS-chirp ABR to more severe degrees of hearing loss may be related to recruitment associated with cochlear hearing impairment. Elberling et al. [9] explain that the chirpstimulus generates an optimized synchronous excitation of
Fig. 1. A high degree of correlation between chirp-ABR thresholds in both low- and high-frequency audiometry bands in young patient with severe hearing loss.
[(Fig._2)TD$IG]
Z.-m. Xu et al. / International Journal of Pediatric Otorhinolaryngology 78 (2014) 812–816 100
Mean VRA threshold(dB HL) at 1-4kHz
B 90
80
70
60
50
40 30
40
50
60
70
80
90
100
Mean VRA threshold(dB HL) at 0.25-0.5kHz
H-Chirp-ABR threshold(dB HL) 100
A 90
80
70
60
50
40 40
50
60
70
80
90
100
L-Chirp-ABR threshold(dB HL) Fig. 2. The distribution of LS-chirp-ABR/behavioral threshold comparisons at each frequency band for all of the ears tested (n = 136).
neurons in all regions of the basilar membrane. The stimulus structure allows modification to excite predefined regions in the cochlea. At high intensity LS-chirp stimulus may be optimally suited as more sensitive ABR measurements in infants with severe hearing losses. Our data showed that LS-chirp ABR thresholds determined behavioral thresholds (mean threshold difference between 3 and 5 dB) were better than ASSR predicted by Swanepoel et al. data [6]. This may be that the LS-chirp produce larger ABR amplitudes and better resolution [9]. We interpret these results as a compatible with the concept of better hair-cell synchronization. As with the frequency-specific information, the data with the scatter plot (Fig. 2) fit to the linear regression, which reflect the fact that LS-chirp ABR thresholds are close to behavioral hearing levels at LF-chirp/VRA (0.25–0.5 kHz) and HF-chirp/VRA (1– 4 kHz). Pearson correlation coefficients were calculated to assess the relationship between chirp-ABR frequency-band (LF-chirp and HF-chirp) and behavioral thresholds at each frequency. Our data showed that the best correlations between LS-chirp ABR thresholds and VRA thresholds were obtained at VRA frequency of 0.25 kHz/LF-chirp-ABR (r = 0.98) and VRA frequency of 1 kHz/HFchirp-ABR (r = 0.98). These results imply better LS-chirp ABR threshold estimations for infants with hearing losses at frequencies of 0.25 kHz and 1 kHz. They are very important low-middle
815
speech frequency bands for clinical application. However, Swanepoel et al. [6] reported that the largest mean difference between ASSR and behavioral thresholds was at 0.5 kHz, and it was the only frequency indicating a statistically significant difference between ASSR and behavioral thresholds. Problems in estimating 0.5 kHz ASSR have been reported previously [15,16], and could result from the fact that the low-frequency-evoked response has a greater intrinsic jitter, due to neural asynchrony. Another reason may be that stimulus protocols for amplitudemodulated tone at 0.5 kHz are not yet optimal. So, the ASSR predicted thresholds are not suitable in estimating low frequency thresholds, especially of 0.5 kHz. Our result revealed more sensitive LS-chirp ABR measurement in estimating low-middle frequency thresholds (0.25–1 kHz). There may be factors which explain the higher correlation between the LS-chirp ABR and behavioral thresholds at low-middle frequencies. The LS-chirp evoked ABR has been regarded as a more synchronous response than the click-evoked ABR, referring to the belief that the chirp stimulates lower-, mid-, and higher-frequency regions of the cochlea simultaneously, especially of lower-, middle-frequency. Don et al. [17] found that the low-chirp-stimulus evoked a significantly larger amplitude of the ABR than that of a click-ABR. Table 2 show the frequency distribution of the obtained behavioral and chirp-ABR thresholds in categories 1 and 2. These data further confirmed the best chirp-ABR threshold estimations for infants with sever hearing losses at frequencies of 0.25 kHz and 1 kHz. In this study, there are some limitations of using the LS-chirp ABR that are considered in the following. The chirp stimulus designed from two bands (LF-chirp = 0.1–0.85 kHz and HFchirp = 1–10 kHz) that could not be precisely correspond to VRA frequency bands (0.25–0.5 kHz and 1–4 kHz). Thus, the chirp may represent the frequencies of 0.25 kHz, 0.5 kHz, 1 kHz, 2 kHz, and 4 kHz that are correspond to VRA frequencies. The problem is also that chirp stimulus fixed by instrument is split between 0.85 kHz and 1 kHz in this study. When the instrument (stimulus parameters) is improved, we will further study in this issue for relationship between VRA and LS-chirp ABR thresholds. In conclusion, this study demonstrates the effective implementation of LS-chirp ABR predicted frequency-specific thresholds, especially of low-middle-frequency. LS-chirp ABR thresholds determined behavioral thresholds in patients with severe hearing losses were better than for mild hearing losses. The use of a chirpABR testing ensures higher sensitivity and accuracy than that of ASSR for measuring frequency-specific thresholds. It is possible to be used as the clinical application with regard to diagnostic audiology in young children. References [1] American Academy of Pediatrics, Newborn and infant hearing loss: detection and intervention (RE9846), Pediatrics 6 (1999) 527–530. [2] L.E. Dalzell, M.S. Orlando, C. Seeger, 3-Stage newborn hearing screening reduces failures at discharge, in: Poster presented at American Academy of Audiology, Salt Lake City, UT, 1996. [3] Y.I. Christine, L.S. Allison, K.C. Diane, L.M. Albert, Language of early- and lateridentified children with hearing loss, Pediatrics 102 (1998) 1161–1170. [4] F.M. Riberiro, R.M. Carallo, A.M. Marcoux, Auditory steady-state evoked responses for preterm and term neonates, Audiol. Neurotol. 15 (2010) 97–110. [5] G. Rance, F. Rickards, Prediction of hearing threshold in infants using auditory steady-state evoked potentials, J. Am. Acad. Audiol. 13 (2002) 236–245. [6] D.W. Swanepoel, R. Hugo, R. Roode, Auditory steady-state responses for children with severe to profound hearing loss, Arch. Otolaryngol. Head Neck Surg. 130 (2004) 531–535. [7] M. Cebulla, E. Sturzebecher, C. Elbering, J. Muller, New click like stimuli for hearing testing, J. Am. Acad. Audiol. 18 (2007) 725–738. [8] S.E. Shore, A.L. Nuttal, High-synchrony cochlear compound action evoked by rising frequency-swept tone bursts, J. Acoust. Soc. Am. 78 (1985) 1283–1295. [9] C. Elberling, M. Don, M. Cebulla, E. Sturzebecher, Auditory steady-state responses to chirp stimuli based on cochlear traveling wave delay, J. Acoust. Soc. Am. 122 (2007) 2772–2785.
816
Z.-m. Xu et al. / International Journal of Pediatric Otorhinolaryngology 78 (2014) 812–816
[10] C. Elberling, M. Don, Auditory brainstem responses to a chirp stimulus designed from derived-band latencies in normal-hearing subjects, J. Acoust. Soc. Am. 124 (2008) 3022–3037. [11] C. Elberling, M. Don, A direct approach for the design of chirp stimuli used for the recording of auditory brainstem responses, J. Acoust. Soc. Am. 128 (2010) 2955– 2964. [12] M. Cebulla, C. Elberling, Auditory brain stem responses evoked by different chirps based on different delay models, J. Am. Acad. Audiol. 21 (2010) 452–460. [13] T. Dau, O. Wegner, V. Mellert, B. Kollmeier, Auditory brainstem responses with optimized chirp signals compensating basilar-membrane dispersion, J. Acoust. Soc. Am. 107 (2000) 1530–1540.
[14] M.A. Petoe, A.P. Bradley, W.J. Wilson, On chirp stimuli and neural synchrony in the suprathreshold auditory brainstem response, J. Acoust. Soc. Am. 128 (2010) 235– 246. [15] O.G. Lins, T.W. Piton, B.L. Boucher, Frequency-specific audiometry using steadystate responses, Ear Hear. 17 (1996) 81–96. [16] T.W. Piton, M.S. John, A. Dimitrijevic, D. Purcell, Human auditory stead-state responses, Int. J. Audiol. 42 (2003) 177–221. [17] M. Don, C. Elberling, E. Maloff, Input and output compensation for the cochlear travelling wave delay in wide-band ABR recordings: implications for small acoustic tumor detection, J. Am. Acad. Audiol. 20 (2009) 99–108.