Indian Journal of Otology

: 2020  |  Volume : 26  |  Issue : 3  |  Page : 127--131

Auditory brainstem response to level-specific CE-CHIRP® threshold estimation in normal-hearing adults

Ahmad Aidil Arafat Dzulkarnain, Suhaila Ahmad Shuckri, Noraidah Ismail 
 Department of Audiology and Speech-language Pathology, Kulliyyah of Allied Health Sciences, International Islamic University Malaysia, Kuantan, Pahang, Malaysia

Correspondence Address:
Dr. Ahmad Aidil Arafat Dzulkarnain
Department of Audiology and Speech-Language Pathology, Kulliyyah of Allied Health Sciences, International Islamic University Malaysia, Jalan Sultan Ahmad Shah, 25200, Kuantan, Pahang


Background: The aim of the present study was to compare the hearing thresholds between pure tone audiometry (PTA) and auditory brainstem response (ABR) from level-specific (LS) CE-Chirp® and click stimuli in normal adult subjects. Materials and Methods: Twenty-four adults with normal audiometric thresholds participated in the study. The ABR was recorded from the study participants at 80 dBnHL until their respective auditory thresholds using both the LS CE-Chirp® and click stimuli. Study Design and Statistical Analysis: A quasi-experimental study design was used. Audiometric thresholds (low frequencies [LFs], mid frequencies [MFs], and high frequencies [HFs]) and the ABR thresholds from both stimuli were compared using the Friedman test with Wilcoxon signed-rank test as the post hoc analysis. Results: No statistically significant difference was identified between the PTA and the ABR to LS CE-Chirp® thresholds at LFs and only small differences (<6 dB) median thresholds differences were identified at the MFs and HF. The amplitudes of wave III and V were larger for ABR to LS CE-Chirp® as compared to the ABR from the click stimulus. Conclusion: This study concluded that the ABR to LS CE-Chirp® has closer thresholds than the audiogram as compared to the ABR from click in normal-hearing adult subjects. At the suprathreshold (80 dBnHL), the ABR amplitudes of wave III, and V were larger in LS CE-Chirp® than the click stimulus.

How to cite this article:
Dzulkarnain AA, Shuckri SA, Ismail N. Auditory brainstem response to level-specific CE-CHIRP® threshold estimation in normal-hearing adults.Indian J Otol 2020;26:127-131

How to cite this URL:
Dzulkarnain AA, Shuckri SA, Ismail N. Auditory brainstem response to level-specific CE-CHIRP® threshold estimation in normal-hearing adults. Indian J Otol [serial online] 2020 [cited 2021 Apr 10 ];26:127-131
Available from:

Full Text


In recent years, level-specific (LS) CE-Chirp® stimulus continuously attracts considerable attention to elicit auditory brainstem response (ABR) due to its abilities to improve the spectral neural synchrony along the basilar membrane and the auditory nerve fibers. LS CE-Chirps stimulus is different from other versions of chirp stimuli, in which the duration of the stimulus varies according to the intensity levels.[1] The adjustment of the stimulus duration according to the intensity level is aimed to minimize spectral distortion from the upward spread of excitation of the neural responses from different frequency regions. The upward spread of excitation of the ABR from the previous versions of chirps stimuli at the high intensity levels was reported to cause poor representation of the ABR earlier waves (I and III) and minimal increase in the ABRs to chirp stimulus wave V amplitude in comparisons with those ABRs at lower or mid-intensity levels[2] and with the ABR from click stimulus.[3],[4]

While the literature on LS CE-Chirp® was limited, several positive findings have been documented, for example, the report on 100% presence of the earlier waves of I and III in the ABR to LS CE-Chirp® in normal-hearing adult subjects as compared to the traditional or CE Chirps[2],[5] and the significant increase in wave I, III, and V amplitudes at high-intensity levels in normal-hearing subjects.[5],[6] Despite these positive findings, it is apparent that there is scant evidence of studies that was investigating the differences in the ABR to broadband LS CE-Chirp® thresholds estimation in comparison to the pure tone audiometry (PTA) thresholds. This is in contrast with the previous reports that compared ABR to click auditory thresholds with audiogram[7],[8],[9] and the only single report that compared the traditional chirp auditory thresholds with the audiogram.[10] Based on the literature on the ABR from click stimulus, it was generally reported that the click ABR auditory thresholds were best correlated with the audiogram at frequencies between 2 and 4 kHz and the differences with the audiogram ranging between 11 and 20 dB.[7],[9] The only study to our knowledge that investigated the relationship of the broadband chirp thresholds with the audiogram was the one from Xu et al.[10] The authors did not used the typical broadband chirp stimulus that sweeps from high frequencies (HFs) to low frequencies (LFs) but divided the spectrum of the broadband chirp to LF (0.1–0.85 kHz) and HFs (1–10 kHz) in 68 infants with mild–to-severe sensorineural hearing losses. The results showed an excellent correlation of both LF and HF behavioral thresholds with the LFs and HF broadband chirp, respectively.

From the summary of the above literature, the knowledge on the relationship of the broadband chirp with audiogram in general and LS CE-Chirp® specifically is still unclear unlike to the ABR to click stimulus. In addition, although the knowledge of the relationship of click ABR thresholds and PTA is well established, most of the studies either used subject with sensorineural hearing loss or combined the data of the subject with hearing losses with those with normal hearing. The purpose of the data combination was to obtain strong statistical analysis to evaluate the relationship between the ABR thresholds and PTA thresholds rather than to measure the differences between those two thresholds. To our knowledge, there were no known studies that specifically compared the differences in the ABR thresholds using broadband stimulus and PTA thresholds only in the normal-hearing adult subjects.

The results from previous studies using both click ABR and the traditional chirp ABR may not necessarily applicable to the ABR to the broadband LS CE-Chirp® because of the different sound properties between both stimuli and different neural responses encoding from the stimulation across the basilar membrane. Specifically, while click stimulus is a broadband frequency acoustic signal, it has a rapid onset that only effectively stimulates the basal region of the cochlear.[11] On the other hand, for ABR to the LS CE-Chirp®, while it is similar in nature with the click stimulus in term of its broad frequency spectrum, the stimulus manipulated the time versus frequency arrival at the cochlear, therefore hypothetically being able to stimulate the whole parts of the basilar membrane and provides an optimum neural synchrony.[12] Given to these factors, there is a strong need to understand the relationship between the PTA thresholds and LS CE-Chirp® ABR thresholds. The aims of the present study therefore were to investigate the LS CE-Chirp® thresholds in normal-hearing adults' subjects and to determine the differences in the PTA thresholds estimation between ABRs to LS CE-Chirp® and click stimulus.

 Materials and Methods


Twenty-four normal-hearing adult subjects consist of 9 males and 15 females between 23 and 26-year-old participated. All of the study participants have (i) no significant history related to a hearing loss, (ii) normal-hearing thresholds at octave frequencies of 250–8000 Hz in both ears, (iii) Type A tympanogram in both that ears that suggests normal middle ear functions, and (iv) normal acoustic reflexes thresholds at 500, 1000, and 2000 Hz on both the ears and mode of stimulations.


The study protocol received an unconditional approval from the Institution Research Ethics Committee with the reference approval of IREC 2018-098. The study was conducted at one of the electrophysiology testing room at the hearing and speech clinic. The ABR test was conducted using Interacoustics Eclipse ABR system.

Vertical electrode montage was used in the present study as previous literature reported larger wave V amplitude as compared to the ipsilateral montage with minimal residual noise levels, in particular for adult subjects.[5] Before placing the electrodes, Nuprep Skin Preparation Gel was used to clean subjects' skin at the high forehead, nape of the neck, and right shoulder. Ambu Neuroline 720 Disposable Electrodes were then placed on these prepared areas, where the non-inverting (+ve) electrode was placed at the high forehead, the ground electrode was placed on the right shoulder, and the inverting electrode (−ve) on the nape of the neck. An optimal individual electrode impedance (below 3000 Ω) and inter-electrode impedance (below 2000 Ω) were always maintained by the researcher throughout the ABR testing.

The ABRs were elicited using both the 0.1 ms rarefaction click (0.2–11 kHz) and the commercially available LS CE-Chirp® (0.2–11 kHz) at a stimulus repetition rate of 33 Hz. Testing begun at 80 dBnHL with the stimulus presented through Eclipse ER-3A insert phone and only the right ear was tested. To seek for the ABR thresholds, the test continues by reducing the intensity levels in 10–20 dB steps until no ABRs were identified by two experience audiologists. The intensity level then was increased in 5 dB steps until the ABR was re-identified again. A repeat recording was performed at the lowest level that the ABR was identified, to be taken as the threshold of the patient. The nontest ear (left ear) was presented with a masker of 40 dBnHL (offset masker) to avoid the participation of the non-test cochlea. The acoustic calibration was performed by the manufacturers in accordance with the reference equivalent threshold sound pressure level from ISO 389-6:2007 for click and Physikalisch-Technische Bundesanstalt (Braunschweig, Germany) for the LS CE-Chirp®. The reference levels of 0 dBnHL were 35.5 dB peSPL for the click and 31.5 dB peSPL for the LS CE-Chirp®.

The ABRs were averaged using the Bayesian weighted averaging technique. To ensure a high quality of the ABR recording and to save time, different strategies were employed to stop the test according to the intensity levels. At intensity above 40 dBnHL, the ABR was averaged until the residual noise level reached 40 nV (regardless of its signal-to-noise ratio (SNR)). Below 40 dBnHL and until the individual ABR thresholds, the averaging was stopped if the F-test at multiple points (F mp) reached 3.1 or at a maximum of four thousand sweeps (99% confidence that the ABR was present). The ABR was only considered to be present only if the SNR is at 3:1 or more.

The averaged ABR responses were filtered at 100–3000 Hz. The filter slope setting was set at 12 dB/octave in a 16 ms time window. For both stimuli (LS CE-Chirp® and click) at 80 dBnHL and threshold, the ABR recording was recorded twice to check for the repeatability of the waveforms. Only the first original waveform was used for the final data analysis. The artifact rejection level was set at ± 40 μV.

Waveform analysis

The main variables of the present study are the absolute amplitudes and latencies of wave I, III, and V at 80 dBnHL and the ABR thresholds. The amplitudes in microvolts of waves I, III, and V were computed from the respective peak of the waves to their following troughs. The absolute latency was measured from the 0 msec reference time window to the peak of each respective wave (waves I, III, and V). Note that the chirp component in the LS CE-Chirp® is always delayed for 1.5 msec from the reference stimulus onset to align with the click ABR latencies from normal-hearing subjects.[2] ABR thresholds in the present study are defined as the lowest levels that the ABR can be identified by two examiners. The ABR waveform identification (e.g., absence/presence and peak selection) including the thresholds was decided by a consensus decision between two examiners. Both examiners have experience more than 10 years in analyzing the ABR waveforms.

Statistical analysis

The differences in the ABR to LS CE-Chirp® and click and PTA thresholds were compared using Friedman test at 95% confident levels. The PTA thresholds were classified into three frequencies range: the LFs, MFs, and HFs. The LF PTA thresholds were the average of PTA thresholds for 250 and 500 Hz, the MFs were the average PTA thresholds of 1000 and 2000 Hz, and the HFs were the average PTA thresholds for 4000 and 8000 Hz. Wilcoxon signed-rank test was further used to identify pairs of thresholds that may have a statistically significant difference. Wilcoxon signed-rank test was also used to compare the mean differences in the ABR absolute latencies and amplitudes for wave I, III, and V at 80 dBnHL between the ABR from LS CE-Chirp® and click stimuli. The data have several breaches of the parametric assumptions that include nonnormal distribution of the data and the nonhomogeneity of variance that could not be solved even with data transformation. Therefore, nonparametric statistical was used for the statistical analysis. All statistical analyses were performed using SPSS version 20.


Comparison between the pure tone audiometric threshold and the auditory brainstem response from level-specific CE-chirp and click stimuli

Friedman's test identified significant median differences among the auditory thresholds from LS CE-Chirp, click, and PTA (average of LF, MF, and HF) (P < 0.001). [Table 1] summarizes the post hoc Wilcoxon signed-rank test and the median differences (with interquartile range) that compare all of the possible pairs of the ABR thresholds (LS CE-Chirp® and click) and the PTA thresholds (LF, MF and HF). According to the [Table 1], the median thresholds of ABR from both LS CE-Chirp® and click were significantly higher (P < 0.05) than the average PTA thresholds in all frequencies (low, mid, and high) except for the ABR to LS CE-Chirp® and the average PTA at LF (P > 0.05). The median differences were approximately 5 dB and 6 dB higher than the average PTA thresholds for ABR to LS CE-Chirp® at MFs and HF, respectively. For ABR to click stimulus, the thresholds were approximately 3, 5, and 10 dB higher than the average PTA thresholds for LFs, MFs, and HF, respectively. In addition, the ABR to click thresholds was significantly higher than the ABR to LS CE-Chirp® (P < 0.001) with the median differences of approximately 5 dB. [Figure 1] shows the example of ABR waveforms elicited from both LS CE-Chirp and click at suprathresholds and thresholds in one of the study participants.{Table 1}{Figure 1}

Auditory brainstem response level specific CE-chirp versus click for wave I, III, and V latencies and amplitudes at 80 dBnHL

[Table 2] summarizes the median (interquartile range) of the absolute latencies for I, III, and V and the statistical analysis to compare the results between the ABR elicited from click and LS CE-Chirp stimuli. The Wilcoxon signed-rank test identified a statistically significant difference in the absolute latencies of wave I, III, and V between LS CE-Chirp and click (P < 0.05). In general, the ABR wave I, III, and V absolute latencies were significantly longer from the LS CE-Chirp than the ABR from the click stimulus.{Table 2}

Wilcoxon signed-rank test further identified statistically significant differences of the ABR wave III and V amplitudes between the ABR to LS CE-Chirp and click stimuli (P < 0.05), as highlighted in [Table 3]. No statistically significant difference was identified in the ABR wave I amplitudes. The results from [Table 3] also show the ABR to LS CE-Chirp® wave III and V amplitudes that were significantly larger than the ABR from the click stimulus.{Table 3}


The aims of the present study were to investigate the estimation of the ABR LS CE-Chirp® thresholds in normal-hearing adults' subjects and to determine the differences in the PTA threshold estimation between both LS CE-Chirp® and click stimuli.

In general, this study found that the ABR to LS CE-Chirp® thresholds in normal-hearing adult subjects were the same with PTA thresholds at LF and has only small differences (<6 dB) at other frequencies region. The present study findings concur with the previous report by Xu et al.[10] who found that LS CE-Chirp® thresholds were <5 dB difference than the audiogram when tested among pediatric population with different ranges of degree of sensorineural hearing loss. In addition, the ABR to LS CE-Chirp® stimulus has lower thresholds (average of 5 dB) than the ABR to click stimulus, indicating that the ABR to LS CE-Chirp® stimulus provides better threshold estimation than the ABR to click stimulus. Overall, the present study suggests that the LS CE-Chirp® ABR thresholds have good prediction for LF thresholds and only overestimate approximately 6 dB for the higher frequency in normal-hearing subjects. The small differences with the audiogram and the lower threshold than the ABR from click stimulus – further support the ABR to LS CE-Chirp® as a tool to accurately predicts the behavioral thresholds. This in addition with the fact that the ABR to LS CE-Chirp® is able to increase the ABR wave V amplitude. This benefit arises from the abilities of this stimulus to manipulate the onset of difference frequencies signal based on the cochlear tonotopicity to achieve an optimal neural synchrony.

In addition, the present study also found an increase in the ABR waves III and V amplitudes when using LS CE-Chirp® stimulus over the click stimulus at suprathresholds of 80 dBnHL. This finding was consistent with the earlier studies that suggested the adjustment of the stimulus duration of LS CE-Chirp® stimulus in accordance to the intensity levels that could avoid upward spread of excitation thus increase the possibility of the presence of wave I, III, and V and at the same time can increase the overall wave amplitudes.[2],[5]

The result of the present study suggests further research to understand the estimation of ABR to LS CE-Chirp®' thresholds in other populations such as pediatric and those patient with various types and degree of hearing loss. The present study found that the LS CE-Chirp® has better estimation at lower frequencies; therefore, testing subject with sloping or reverse sloping audiogram could help of better understanding if ABR to LS CE-Chirp® are more reflecting hearing at LF unlike the click stimulus (at HF) in future studies.


This study concluded that the ABR to LS CE-Chirp® has smaller thresholds differences with the audiogram than the ABR from click in normal-hearing adult subjects. At 80 dB nHL, the wave amplitudes of III and V were larger in ABR to LS CE-Chirp® than the ABR to click stimulus. The study's conclusion is applied only to the participants, equipment, stimulus and recording parameters used in this study. Caution must be taken before applying this study's conclusion beyond the study participants and all of the factors mentioned earlier.

Financial support and sponsorship

This work was supported by the Trans-disciplinary Research Grant Scheme (TRGS19-02-001-0004) from Ministry of Higher Education of Malaysia.

Conflicts of interest

There are no conflicts of interest.


1Elberling C, Callø J, Don M. Evaluating auditory brainstem responses to different chirp stimuli at three levels of stimulation. J Acoust Soc Am 2010;128:215-23.
2Kristensen SG, Elberling C. Auditory brainstem responses to level-specific chirps in normal-hearing adults. J Am Acad Audiol 2012;23:712-21.
3Rodrigues GRI, Lewis DR. Comparison of click and CE-chirp® stimuli on Brainstem Auditory Evoked Potential recording. Revista da Sociedade Brasileira de Fonoaudiologia 2012;17:412-6.
4Petoe MA, Bradley AP, Wilson WJ. On chirp stimuli and neural synchrony in the suprathreshold auditory brainstem response. J Acoust Soc Am 2010;128:235-46.
5Dzulkarnain AA, Noor Ibrahim SH, Anuar NFA, Abdullah SA, Tengku Zam Zam TZ, Rahmat S, et al. Influence of two-electrode montages on the level-specific (LS) CE-Chirp auditory brainstem response (ABR) at multiple intensity levels. Int J Audiol 2017;56:723-32.
6Cargnelutti M, Cóser PL, Biaggio EP. LS CE-Chirp® vs. Click in the neuroaudiological diagnosis by ABR. Braz J Otorhinolaryngol 2017;83:313-7.
7Baldwin M, Watkin P. Predicting the degree of hearing loss using click auditory brainstem response in babies referred from newborn hearing screening. Ear Hear 2013;34:361-9.
8Lu TM, Wu FW, Chang H, Lin HC. Using click-evoked auditory brainstem response thresholds in infants to estimate the corresponding pure-tone audiometry thresholds in children referred from UNHS. Int J Pediatr Otorhinolaryngol 2017;95:57-62.
9van der Drift JF, Brocaar MP, van Zanten GA. The relation between the pure-tone audiogram and the click auditory brainstem response threshold in cochlear hearing loss. Audiology 1987;26:1.
10Xu ZM, Cheng WX, Yao ZH. Prediction of frequency-specific hearing threshold using chirp auditory brainstem response in infants with hearing losses. Int J Pediatr Otorhinolaryngol 2014;78:812-6.
11Canale A, Dagna F, Lacilla M, Piumetto E, Albera R. Relationship between pure tone audiometry and tone burst auditory brainstem response at low frequencies gated with Blackman window. Eur Arch Otorhinolaryngol 2012;269:781-5.
12Dau T, Wegner O, Mellert V, Kollmeier B. Auditory brainstem responses with optimized chirp signals compensating basilar-membrane dispersion. J Acoust Soc Am 2000;107:1530-40.