Sudden Traumatic Noise Exposure Induces an Altered Wave I in the Auditory Brainstem Response in Human

Abstract

Introduction

Noise trauma may induce auditory symptoms and temporary or permanent shifts in hearing thresholds, assessed with a tone audiogram. Until recently it has been assumed that primary changes after noise trauma exclusively represent damage to the sensory hair cells. Contrary, animal studies have now shown early alterations in auditory brainstem response (ABR) characteristic of auditory neuropathy (AN), as reduction or absence of wave I, following noise trauma. AN involves a variety of disease mechanisms in the cochlea and auditory nerve, affecting the synaptic encoding and/or neural transmission of auditory information. The aim of this study was to evaluate whether patients exposed to noise trauma exhibit click ABR alterations characteristic of AN.

Method

Thirty-one adult patients who attended an ear-nose and throat clinic, after a noise trauma, were included and an age- and gender-matched control group of 31 adults was constructed. To evaluate audiologic hearing impairment pure tone audiometry, speech intelligibility, click ABR and transient evoked otoacoustic emissions (TEOAE)/cochlear microphonics (CM) were performed. A full linear mixed model including the interaction between sex and group was tested.

Results

Our study showed that individuals exposed to sudden and distinct traumatic noise exhibit a statistically significant reduction (81.1 nV, p < 0.001) in the amplitude of wave I in ABR.

Conclusion

We suggest that the hearing investigation after sudden traumatic noise exposure should therefore be expanded with ABR to confirm or exclude AN.

Introduction

Noise exposure may cause both immediate as well as delayed hearing impairment, which has historically been interpreted as signs of hair cell damage. Recent observations have convincingly expanded our knowledge regarding this matter.

A widespread loss of spiral ganglion cells, which contact the inner hair cells, was observed in mice after noise exposure [1]. The neuronal loss was not associated with hair cell loss and was therefore considered to be primary neural degeneration [1]. Furthermore, noise exposure can permanently damage auditory nerve fibers even when the threshold shift in the audiogram is temporary, indicating hair cell recovery [2, 3].

Clinical reports have identified patients with hearing impairment despite having normal tone audiograms, in which expanded audiologic investigation indicates pathology in the neural transmission of the auditory signal [4–6]. This neural dysfunction has been named in various ways focusing on different aspects of the pathology or impairment [7]. We will refer to auditory neuropathy (AN) when discussing dysfunctions in the transmission pathway from the first afferent neurons in the cochlea to the auditory cortex, and cochlear synaptopathy (CS) to specify dysfunctions localized to the first synapse, between the hair cell and the auditory nerve, in the cochlea.

AN involves a variety of disease mechanisms in the cochlea and auditory nerve, affecting the synaptic encoding and/or neural transmission of auditory information [8]. This dysfunction often leads to hearing difficulties, impacting the processing of acoustic time signals essential for speech understanding, sound localization, and speech discrimination in background noise [9, 10].

Auditory brainstem response (ABR) is an objective measurement, ideal for detecting auditory dysfunction [11]. The ABR response consists of five peaks which may decrease, become abnormal or be absent if there are disorders that cause desynchronization of the neural activity [11–15]. The height of the peak, down to the following trough, defines the amplitude and a larger amplitude implies a greater number of active neural fibers [16]. The time from the initial auditory stimulus to the peak, the latency, will be prolonged if the intensity of the stimulus decreases but also by age and male gender [16].

Several animal studies have shown a reduction in the amplitude of wave I in the ABR after noise exposure and that this reduction is associated with CS [2, 3, 17–19]. Some studies indicate a correlation between noise exposure and a reduction in the amplitude of wave I in humans [5, 6, 20].

Contrary, there are studies on humans that show that wave I is not affected after noise exposure [21–23]. Despite these findings a significant gap in understanding remains regarding the precise impact of sudden traumatic noise exposure on the ABR in humans. Addressing this gap is crucial for developing more accurate diagnostic tools and effective interventions. Our aim in this study was to evaluate whether individuals exposed to sudden and traumatic noise exhibit changes in their ABR.

Method

Study Design and Ethical Approval

This was a clinical case-control study with extended audiologic evaluation after noise trauma. The study was approved by Swedish Ethical Review Authority Sweden (Approval No. 2020-03913).

Study Participants and Data Collection

Primary sampling method was undertaken with a search in the data charts for patients exposed to a sudden and traumatic noise. Thirty-one adult patients with a history of a sudden and traumatic noise exposure who had a doctor visit at an ear-nose and throat department in three Regions in Sweden, Västerbotten, Västernorrland, and Jämtland-Härjedalen, were asked to provide informed consent to participate. Relevant clinical parameters were taken from patient medical records, including audiometric data, type of noise exposure and medical reports. An age- and gender-matched control group, of 31 adults, was constructed by asking siblings, relatives of the participants in the case group and unrelated participants for informed consent to participate. All participants in both case and control groups spoke Swedish fluently.

Audiometric Testing

To evaluate audiologic hearing impairment and AN, pure tone audiometry, speech intelligibility in noise, click ABR and transient evoked otoacoustic emissions (TEOAE)/cochlear microphonics (CM) were performed. All audiologic testing was executed according to clinical routine and performed during the same day. All audiometric equipment was calibrated and is used in routine health care. All tests were performed by licensed audiologists.

Pure Tone Audiometry

Pure tone audiometry was performed using Otometrics Astera2 (Natus Medical, Middelton, WI, USA), Primus PFU Pro HW (Auditdata A/S, Copenhagen, Denmark), or Interacoustics Equinox (Interacoustics A/S, Middelfart, Denmark) in a soundproof room with TDH-39 headphones and bone vibrator B-71. Frequency ranges from 125 to 8,000 Hz for air conduction and 500–4,000 Hz for bone conduction. Pure tone average for air and bone conduction was calculated at 500, 1,000, 2,000, and 4,000 Hz (PTA4). A pure tone average for the high frequencies (HFPTA) were calculated at 3,000, 4,000 and 6,000 Hz. If necessary, masking was performed as a standard procedure.

Speech Intelligibility in Noise

Speech intelligibility in noise was assessed using the word recognition score (WRS) of the “speech in noise” monosyllabic word test, which is a test for Swedish-speaking patients. A list of 50 monosyllabic phonetically balanced words was presented with speech level in dB SPL where the patient experiences the loudness as a normal conversational level, usually 30–35 dB louder sound level than PTA4, and with fixed 4 dB signal-to-noise ratios in the same ear, according to Swedish routine [24, 25]. The speech audiometry was performed with the same audiometers as in pure tone audiometry with TDH-39 headphones and each ear were tested separately. The WRS index was calculated as the percentage of correctly repeated words. The result is assessed based on expected values of the speech intelligibility index and considers hearing thresholds and age. If the result differs more than 10% from the predicted value, it is a significant deviation [26, 27].

Auditory Brainstem Response

The setup for the ABR recordings was Interacoustics Eclipse (Interacoustics A/S, Middelfart, Denmark) with high and low pass filters of 100 Hz and 3,000 Hz. The click stimuli were used with an alternating polarity present with 11.1 click/s at 80 dB HL through E-A-RTONE insert earphones for ABR. A minimum of 2,000 accepted sweeps was collected in each measurement. Each ear was measured twice.

The mastoid electrodes (Ambu Neuroline 720) were placed on the vertex, the forehead or cheek as a ground electrode and on the right and left mastoid as active electrodes. In one participant in the case group, the electrodes were placed on the earlobes instead of the mastoid because of psoriasis skin problems behind the ears. All participants rested comfortably in an armchair under dimmed lightning. A licensed audiologist analyzed the results regarding presence, amplitude, and latency of wave I, III, and V.

TEOAE and Cochlear Microphonics

The TEOAE were recorded using Interacoustics Titan (in 24 participants) and Eclipse (in seven participants) (Interacoustics A/S, Middelfart, Denmark). Acoustic stimulation was delivered at 80 ± 3 dB peak Sound Pressure Level using nonlinear clicks with duration 80.0 click/s. The frequency range was 840–4,759 Hz. Probe fit was confirmed before initiating the test.

The diagnostic criteria for the TEOAE to be considered present was an overall signal-to-noise ratio ≥6 dB SPL, and response reproducibility and stimulus stability of the waveforms ≥80% [28]. The participants were instructed to remain quiet and keep their heads still during the measurement. When recording the CM, the same electrodes and measuring equipment were used as in ABR. Acoustic stimuli were clicks at 80 dB HL with a rate of 88.8 Hz, filtered from 100 Hz to 5,000 Hz. Presented with alternating polarity, the stimuli were separated into A (condensation) and B (rarefaction) curves.

Statistical Analysis

All statistical analyses were performed using R version 4.5.1 and RStudio (version 2024.12.1). Linear mixed-effects models (LMMs) were fitted using the lme4 package (version 1.1-37). The primary objective was to evaluate the impact of sound trauma on ABR metrics, specifically focusing on the amplitude and latency of waves I and V. The dataset included both case and control groups, with additional variables such as sex, age, ear, PTA4 and HFPTA. Additionally, separate analyses were conducted for PTA4 air using Welch’s two-sample t test to compare the case and control groups, stratified by ear (right and left). To assess the robustness of these findings, the same comparisons were also performed using the nonparametric Wilcoxon rank-sum test.

Preprocessing

ABR amplitudes for waves I and V were scaled by a factor of 1,000 to enhance interpretability, resulting in output values expressed in nanovolts. Categorical variables including group (case/control), ear (left/right), and sex (men/women) were converted to factors.

Modelling Strategy

For each outcome variable (amplitude and latency of waves I and V), an initial LMM was fitted including the interaction between group and sex, along with age and PTA4 as covariates. The same model was also tested with HFPTA replacing PTA4 as a covariate. To account for the correlation due to repeated measurements on the same participants (i.e., measurements of the right and left ear), a random intercept for each participant was included. This approach allowed for the adjustment of individual variability and provided a robust analysis of the impact of sound trauma on auditory function. If the interaction term was statistically significant, the full model was retained for post hoc analysis. Otherwise, a reduced model excluding the interaction was used to assess main effects. Post hoc comparisons of estimated marginal means were conducted using the emmeans package (version 1.11.2) to explore differences between groups and sexes.

Wave I Amplitude: Stratified Analysis

In addition to the general model, separate LMMs were fitted for men and women to explore potential sex-specific effects on wave I amplitude. This stratified analysis was motivated by previous findings suggesting gender-related differences in this ABR component [6, 29].

Model Fit and Robustness

For all tests, the alpha level was set to 0.05 for statistical significance. Additionally, the resulting residuals were analyzed to ensure that the assumptions for the linear mixed model were met. Model fit was evaluated using marginal and conditional R² values from the MuMIn package (version 1.11.48). To ensure robustness of the findings, aligned rank transform (ART) models were also fitted using the ARTool package (version 0.11.2). These nonparametric models served as a sensitivity analysis to validate results in cases where assumptions of the LMMs may not be fully met.

Results

Participants/Demographics

Thirty-one participants, 13 females and 18 males, previously exposed to a sudden and distinct traumatic noise event, were included in the study after informed consent. The average age at the time of the noise trauma was 54 years, ranging from 24 to 75 years.

All participants completed all hearing measurements. One participant did not accept the 80 dB level in ABR because of hyperacusis and was therefore tested at 60 dB.

The most common sudden noise exposure was sudden bangs in factory premises (n = 8), explosions (n = 6), gunshots when hunting (n = 4), screams (n = 3), and other (n = 10) (Table 1). The time from trauma to the hearing measurement in the study varied from 71 days up to 2,788 days (Table 1). Analysis did not reveal any statistically significant association between the elapsed time since noise trauma and the responses in ABR. Time from the trauma to the first medical examination in an ear-nose and throat department varied between examinations the same day and up to 555 days after the trauma. Eight participants had a medical examination the same or the next day, ten participants had a medical examination within a month and nine had a medical examination after more than a month’s time. In four participants, it is unclear exactly when the trauma happened (Table 1).

Table 1.

Overview of the noise trauma for the participants in the case group

Sex	Age, years	Date of trauma	Date of studya	Time from trauma to study, month	Type of noise trauma
M	49	12-2017	02-2022	50	Sudden loud metal noise in factory
M	62	01-2020	01-2024	49	Sudden loud metal noise in bakery
M	28	12-2020	09-2023	33	Sudden loud metal noise in factory
M	41	10-2021	06-2023	20	Sudden loud metal noise in factory
F	83	11-2014	03-2022	89	Sudden loud metal noise in factory
F	82	06-2015	02-2022	81	Sudden loud metal noise in factory
F	56	09-2016	04-2022	67	Sudden loud metal noise in factory
F	63	06-2021	01-2024	31	Sudden loud metal noise in commercial kitchen
M	77	11-2016	03-2022	64	Explosion car tire
M	75	01-2017	09-2021	56	Explosion car battery
M	59	09-2019	02-2024	52	Explosion truck battery
M	56	01-2020	01-2024	48	Explosion bike tire
F	79	05-2016	08-2022	75	Explosion bike tire
F	48	07-2022	12-2023	16	Explosion of bottle
M	77	09-2019	03-2022	30	Gun shot
M	69	08-2021	10-2023	25	Gun shot
M	65	03-2022	06-2023	15	Gun shot
F	50	08-2021	11-2023	28	Gun shot
M	57	11-2016	05-2022	66	Scream
F	40	08-2017	09-2021	50	Scream
F	46	08-2021	01-2024	30	Scream
M	50	09-2019	08-2023	48	Concert
F	66	01-2015	09-2021	81	Concert
F	37	04-2020	08-2023	41	Sudden loud music in headphones
F	27	07-2020	02-2024	44	Sudden loud music in speaker
M	55	06-2017	09-2021	51	Wax cleaning with ear suction
M	64	04-2019	08-2022	40	Wax cleaning with ear suction
M	67	04-2017	05-2022	61	Cutting steel pipe
M	52	01-2022	06-2023	18	Fireworks
M	48	12-2023	03-2024	2	Store alarm
F	55	01-2015	08-2022	92	Towel slap over ear

^aDate of study is the time when the hearing measurements, included in the study, were performed.

The control group consisted of 31 gender- and age-matched participants. None of the controls had been exposed to a sudden noise trauma. Ten had a mild to moderate sensorineural hearing loss and the rest had normal hearing. All participants in the control group completed the same hearing measurements as the case group.

Pure Tone Audiometry

In the noise exposed group, the PTA for the four frequencies 500, 1,000, 2,000, and 4,000 Hz (PTA4) was 26 dB HL (±SD 19) in the right ear and 25 dB HL (±SD 16) in the left ear, compared to the control group where the PTA4 was 13 dB HL (±SD 13) in the right ear and 12 dB HL (±SD 10) in the left ear. The HFPTA for 3, 4 and 6 kHz in the noise exposed group was 41 dB HL (±SD 24) on the right ear and 40 dB HL (±SD 22) on the left ear, and in the control group the corresponding numbers were 20 dB HL (±SD 17) on the right ear and 19 dB HL (±SD 15) on the left ear.

There was a statistically significant difference between the case and control groups for both the right ear (Welch’s t test, p = 0.002) and the left ear (p = 0.0002). These results were confirmed using the nonparametric Wilcoxon rank-sum test, which yielded consistent findings (Fig. 1). None of the participants, neither in the noise exposed nor in the control groups had conductive hearing loss.

Fig. 1.

Pure tone audiometry thresholds for each frequency and ear in the case and control group.

Speech Intelligibility in Noise

The mean WRS in noise in the noise exposed group was 57% (range 0–88) in the right ear and 59% (range 4–88) in the left ear, and respective expected values were 63% (range 32–80) for the right ear and 63% (range 40–82) for the left ear. Eleven of the participants in the noise exposed group performed more than 10% lower values than expected value, five binaurally and 6 monaurally. Seven of these individuals also exhibited a reduced or absent wave I amplitude in the ABR.

The mean WRS in noise for the control group was 80% (range 48–100) for the right ear and 74% (range 56–90) for the left ear and the expected value was 77% (range 54–84) for both ears. Five among the controls performed more than 10% lower value than expected value, 2 binaurally and 3 monaurally.

TEOAE and Cochlear Microphonics

In the noise exposed group, the total TEOAE mean was 12.1 dB SPL (±SD 7.5) for the right ear and 10.9 dB SPL (±SD 6.6) for the left ear compared to the control group which present total TEOAE mean of 15.7 dB SPL (±SD 4.5) for the right ear and 15.3 dB SPL (±SD 5.1) for the left ear. Participants who did not present any TEOAE:s were examined with normal CM.

Auditory Brainstem Response

All 62 participants (31 cases and 31 controls) underwent the ABR measurement at the level of 80 dB nHL except for one participant in the case group who experienced hyperacusis and tinnitus and completed the measurement at the level of 60 dB nHL. Despite the lower level of stimuli intensity, waves I and V could be detected. Statistical analysis excluding this participant showed no difference in the outcome.

Amplitude Wave I

Fifteen out of 31 participants in the case group showed absence of wave I, with six binaurally and nine monaural. One participant in the control group showed absence of wave I monaurally. We have chosen to consider the absence of a wave as an observation where the amplitude is zero and those observations are accordingly included in the statistical analyses. The total means of the amplitude for wave I were 101 nV (±SD 0) the right ear and 98 nV (±SD 96) in the left ear for the case group, and 187 nV (±SD 107) in the right ear and 217 nV (±SD 134) in the left ear for the control group. The mean amplitude for women was 124 nV in right ear and 139 nV in the left ear compared to men who had an amplitude of 85 nV in the right ear and 70 nV in the left ear in the case group. The corresponding numbers for the control group were 236 nV in the right ear and 289 nV in the left ear for women compared to 155 nV in the right ear and 165 nV in the left ear for men (online suppl. Tabel S1; for all online suppl. material, see https://doi.org/10.1159/000553367). One participant in the control group showed absence of wave I monaural, but neither showed hearing loss in the other hearing measurements nor reported tinnitus.

A full linear mixed model including the interaction between sex and group was initially tested. However, the interaction term was not statistically significant and did not improve model fit. Therefore, a simpler model excluding the interaction was used. This linear mixed model analysis showed that the amplitude of wave I was significantly lower (p < 0.001) in the case group than in the control group (81.1 nV). The significance remained when adjusting for age, sex, and differences in PTA4. The significance remained when PTA4 was replaced with HFPTA in the model. Age and sex had significant impact on the amplitude of wave I, where higher age (2.2 nV/year; p = 0.021) and male gender (73.5 nV; p = 0.001) generate a lower amplitude. The linear mixed model explained 34% of the variance in the dataset (R²), with an adjusted R² of 61%. There was no significant difference in the magnitude of the reduction of the amplitude of wave I for the case group between the sexes (Fig. 2).

Fig. 2.

ABR amplitude wave I. ABR comparison between case and controls and between sexes for amplitude of wave I. ***p < 0.001; **p < 0.01; ^†p < 0.05.

Latency Wave I

The total mean of the latency of wave I in the case group was 1.46 ms (±SD 0.22) in the right ear and 1.35 ms (±SD 0.16) in the left ear. The mean latencies for the women were 1.45 ms in the right ear and 1.35 ms in the left ear compared to men who had a latency of 1.47 ms in the right ear and 1.32 ms in the left ear. Participants with lack of amplitude on wave I are not included.

The total mean of the latency of wave I in the control group was 1.44 ms (±SD 0.15) in the right ear and 1.38 ms (±SD 0.14) in the left ear. The mean latency of wave I for women was 1.40 ms in the right ear and 1.33 ms in the left ear compared to men who had a latency of 1.47 ms in the right ear and 1.42 ms in the left ear (online suppl. Tabel S1). A full linear mixed model including the interaction between sex and group was initially tested. However, the interaction term was not statistically significant and did not improve model fit. Therefore, a simpler model excluding the interaction was used. This linear mixed model analysis was performed and showed that age had a significant impact on the latency which increases with increasing age (0.003 ms/year; p = 0.026). There was no significant difference between the case group and the control group or between men and women regarding the latency. The latency of wave I was not affected by PTA4 or by HFPTA. The linear mixed model explained 16% of the variance in the dataset (R²), with an adjusted R² of 33% (Fig. 3).

Fig. 3.

ABR latency wave I. ABR comparison between case and controls and between sexes for latency of wave I.

Amplitude Wave V

Three participants in the case group showed absence of wave V monaural. None showed absence binaurally.

The total mean of the amplitude of wave V in the right ear was 309 nV (±SD 163) and 328 nV (±SD 144) in the left ear. The mean amplitude for the women was 261 nV in right ear and 362 nV in the left ear compared to men who had an amplitude of 338 nV in the right ear and 305 nV in the left ear. One participant in the control group showed absence of wave V binaurally. The total mean of the amplitude of wave V in the control group was 439 nV (±SD 149) in the right ear and 400 nV (±SD 159) in the left ear. The mean amplitude for women was 528 nV in the right ear and 465 nV in the left ear compared to men who had an amplitude of 361 nV in the right ear and 353 nV in the left ear (online suppl. Tabel S1). A linear mixed model analysis was conducted to examine the effects of group (case vs. control), sex, and age on ABR amplitudes. The model revealed a significant main effect of group (p = 0.003) and a significant interaction between sex and group (p = 0.02), indicating that the effect of group membership on ABR amplitude differed by sex. No significant effect of age was observed. Given the significant interaction, post hoc simple effects analyses were performed to further explore the nature of this relationship. These analyses showed that among women, ABR amplitudes were significantly lower in the case group compared to the control group (164 nV; p = 0.004), whereas no significant difference between groups was found among men. The linear mixed model accounted for 18% of the variance in the dataset (R²), with an adjusted R² of 52% (Fig. 4).

Fig. 4.

ABR amplitude wave V. ABR comparison between case and controls and between sexes for amplitude of wave V. ***p < 0.001.

Latency Wave V

The total mean of the latency of wave V in the case group was 5.73 ms (±SD 0.43) in the right ear and 5.59 ms (±SD 0.35) in the left ear. The mean latency for the women was 5.58 ms in the right ear and 5.40 ms in the left ear compared to men who had a latency of 5.85 ms in the right ear and 5.73 ms in the left ear. The total mean of the latencies of wave V in the control group were 5.42 ms (±SD 0.28) in the right ear and 5.48 ms (±SD 0.25) in the left ear. The mean latency of wave V for women was 5.27 ms in the right ear and 5.29 ms in the left ear compared to men who had a latency of 5.53 ms in the right ear and 5.44 ms in the left ear (online suppl. Tabel S1). A full linear mixed model including the interaction between sex and group was initially tested. However, the interaction term was not statistically significant and did not improve model fit. Therefore, a simpler model excluding the interaction was used. This linear mixed model analysis was conducted to investigate the latency of wave V in relation to group (case vs. control), sex, age, and PTA4 or HFPTA. The model revealed a significant main effect of group, with the case group exhibiting longer latencies compared to the control group (mean difference = 0.2 ms; p = 0.010). A significant sex effect was also observed, where men showed longer latencies than women (mean difference = 0.23 ms; p = 0.002). Additionally, age had a significant impact on latency, with an increase of 0.006 ms per year (p = 0.034), indicating that latency increases with age. In contrast, neither PTA4 nor HFPTA showed a significant association with latency. The linear mixed model explained 35% of the variance in the dataset (R²), with an adjusted R² of 81% (Fig. 5).

Fig. 5.

ABR latency wave V. ABR comparison between case and controls and between sexes for latency of wave V. ***p < 0.001.

Discussion

We show in this study that the amplitude of wave I in ABR is significantly reduced for participants with a history of a sudden and distinct traumatic noise event. The significance remained when adjusting for age, sex, differences in PTA4, HFPTA, and the interaction between group and sex.

Analysis of ABR Wave I – Auditory Neural Transmission

Our study demonstrates that individuals who have experienced a sudden and distinct noise trauma show a statistically significant reduction or lack in the amplitude of wave I in the ABR. Noise trauma induces a reduction in auditory neural fibers in the very first cochlear synapse detectable through ABR by a reduction in the amplitude in wave I in rodents [2, 3, 17–19, 30].

Regarding humans, corresponding evidence is difficult to obtain as a direct comparison of the count of neural fibers or cochlear synaptic ribbons and ABR wave I amplitude is not an option. Although the more diffuse diagnosis AN is used for patients, it seems likely that reduction of wave I in ABR after sudden and distinct noise exposure indicate CS also in human.

Earlier reports have been inconclusive. Some studies have shown a significant reduction of the amplitude in wave I [5, 6, 20, 29] correlated with noise exposure while other studies could not show any significant reduction at all [21–23].

The relationship between temporary threshold shift and sudden and distinct noise exposure has not been well investigated [31], and to the best of our knowledge, the relationship between sudden and distinct noise exposure and ABR has not been reported for humans previously.

In our study, we have chosen to include the participants who show lack of amplitude in ABR wave I in the analysis. Absence of wave I in humans is considered to be an indication of AN together with normal OAE/CM [11, 12, 32, 33]. Some studies have chosen to exclude participants with absence of wave I and others seem to include them according to tables [5, 6, 21–23].

Neurophysiological Consequences of Sudden and Distinct Noise Trauma

In our study, we have several participants who have a missing wave I, 42% (13 participants, four women and nine men) in the case group and 3% (one male participant) in the control group. The participant in the control group with a missing wave I also have an abnormal wave V. In other studies, they have very few cases with absence of wave I [5, 6, 21–23]. One reason for these differences may be that the cases in our study have been exposed to a powerful sudden traumatic noise compared to participants in other studies who have estimated their noise exposure over a long period of time. Previous studies in humans have used self-reported noise exposure for example in an interview [5, 20] or a questionnaire [6, 21, 22, 34]. The participants in our case group have been exposed to a sudden noise trauma, contrary to other studies where the exposure over a long period of time was estimated by the participants themselves. Noise trauma can be difficult to control through a questionnaire where the participants estimate their noise exposure, sometimes over a long period of time. Additionally, it seems to be a lack of standardization of questionnaires [22]. Our aim with this study was to reduce the inconsistencies regarding earlier studies by using a cohort with a more precisely defined short time period of a sudden noise trauma, rather than self-reported noise exposure over long time-periods. This approach could potentially lessen recall bias related to long-term exposure questionnaires.

Correlation between WRS and ABR Amplitude Wave I

Affected word recognition scores (WRS) in noise are considered indicative of AN [35]. In the present study, 11 participants showed WRS in noise that were more than 10% lower than the expected values, and notably, 7 of these individuals also exhibited a reduced or absent wave I amplitude in the ABR. This convergence of impaired speech-in-noise performance and diminished peripheral neural responses supports the presence of disrupted auditory nerve function and is in line with previous descriptions of AN. Several studies have demonstrated that speech perception in noise is disproportionately poor in individuals with AN relative to what would be expected from the audiogram, a deficit commonly attributed to impaired encoding of fine temporal structure and amplitude modulations [10, 35].

Sex Differences in ABR following Noise Exposure

We can show a robust statistically significant reduction in wave I for both men and women after sudden noise exposure that remains when adjusting for age, sex and differences in PTA4. Bramhall et al. [5] showed a reduced ABR wave I in young veterans exposed to high level military self-reported noise compared to veterans and non-veterans with lower levels of self-reported noise exposure. Mekki et al. [20] included a case group, with a history of being exposed to significant loudness levels, compared with a control group and showed a significantly lower amplitude in wave I for the case group. Stamper & Johnson presented in their study on humans a weak correlation between a reduced ABR wave I and self-reported noise exposure when they reported the results for all participants. However, when analyzing men and women separately, the reduction was only significant for women [6, 29]. Men were only representing a third of the participants, which may have influenced the results [6, 29]. Women have a significantly larger amplitude of wave I than men [36], and it is therefore important to consider sexes in the analyses. Mekki et al. [20] did not consider sex differences even though the sex distribution in their study was 29 women and 11 men. Bramhall et al. [5] observed only weak differences between sexes, where women had greater amplitudes of wave I compared to men. However, they concluded that their study was not designed to compare sex differences. Bramhall et al. [5] stated that the weak difference between men and women in their study is an order of magnitude smaller than the reduction between veterans in the high noise group and the non-veteran group.

Electrode Placement and Methodological Considerations in ABR Recordings

In our study, we used the mastoid electrode, which is highly reliable for waves I, and V [37]. Different electrode types and placement may render different results and, for example, a TM-electrode provides a higher amplitude with a greater variation, twice the standard deviation, compared to a mastoid electrode [6]. Several studies reported no correlation between self-reported noise exposure and wave I amplitude in ABR using TIPtrode in the ear canal [21, 22, 37]. Bramhall et al. [5] on the other hand, were able to demonstrate a correlation between noise exposure and the amplitude of wave I using TIPtrode in the ear canal. Valderama et al. [22] could not show any correlation between noise exposure and amplitude wave I in ABR; however, the variation in self-reported noise exposure was very large.

Age-Related Changes in ABR Measures

In our study, we can show that the amplitude of wave I decreases with increasing age, but wave I remained significantly reduced in the noise exposed group even when we adjusted for age. The observation that wave I decreases with increasing age is consistent with previous studies, both for rodents [1, 2, 17, 18, 30] and humans [38]. In our study, the analysis shows that age also influences the latency of wave I. Konrad-Martin et al. [38] also observed a correlation between the reduction in all amplitudes and a delay of the latencies in waves I and III with age, and they interpreted the results as evidence that aging reduces the number and/or synchrony of fibers in cochlear nucleus or auditory nerve. It is worth mentioning that almost all participants in their study were men [38].

Fernandez et al. [18] in their study on mice showed that noise exposure accelerates age-related synaptopathy. Prendergast et al. [23] discuss whether if humans exposed to noise at an early age, are more sensitive to synaptopathy later in life when age-related changes affect the auditory system.

In several studies on humans, all participants were young, from 18 up to 40 years old, and had normal hearing thresholds in their PTAs. Therefore, it has not been considered whether age affects the amplitude [5, 6, 20, 23]. The degree of hearing thresholds does not seem to affect the amplitude of neither wave I nor wave V in our study. These results are in accordance with other studies [5, 6, 20–23] that included only participants with normal hearing. Despite a mild to moderate increase in PTA4 in our study, this does not seem to affect the click ABR amplitudes, that also examines the frequencies 1–4 kHz [39].

Analysis of ABR Wave V – Central Auditory Processing following Noise Exposure

In our study, the amplitude of wave V showed significant difference between case and control groups for women but not for men. Mekki et al. [20] show a significant difference in wave V amplitude between groups where the case group presents a larger amplitude compared to the control group. However, they report a case group that consists mainly of women, who have a larger amplitude compared to men [36], and they do not report how the distribution between men and women is in the control group [20]. Stamper & Johnson did not found any differences between men and women regarding neither amplitude nor latency of wave V [29].

Further, we found that the latency of wave V was significantly longer in the case group compared to the control group. The statistical analysis showed that this may be explained by differences in PTA4 between groups were higher PTA4 render longer latencies. Prendergast et al. [23] initially found a delay of wave V correlated to noise exposure. However, when they added age as a predictor, the correlation between noise exposure and wave V was no longer significant. The analysis of wave V is essential in relation to wave I because wave I reflects the peripheral and earliest neural activity in the auditory system, whereas wave V represents neural processing at a more central level within the auditory brainstem. Against this background, the analysis of wave V becomes crucial for understanding the downstream consequences of peripheral neural damage. While wave I primarily reflects the summed activity of the auditory nerve, wave V originates from more central generators in the auditory brainstem, such as the inferior colliculus [40, 41]. Therefore, comparing wave I and wave V allows assessment of whether reduced peripheral input is accompanied by alterations in central neural processing [42, 43]. Conversely, concurrent alterations in both wave I and wave V would suggest that peripheral synaptic loss has broader consequences for neural transmission along the auditory pathway. Thus, the inclusion of wave V analysis is critical for interpreting the functional impact of noise-induced peripheral neural damage and for distinguishing between purely peripheral deficits and combined peripheral-central auditory dysfunction.

Limitations

Despite the robust findings, our study has some methodological limitations that should be considered. First, the assessment of noise exposure was based on self-reported data, which can introduce recall bias and variability in the accuracy of the exposure estimates. While objective measures such as dosimeters or sound level meters could provide more precise data, these methods may not be feasible for retrospective studies. Second, the choice of electrode type might influence the amplitude measurements. While we used mastoid electrodes, which are reliable for waves I and V, some other studies have used TIPtrodes, leading to different results. Standardizing the electrode type across studies could help reduce variability and improve comparability. In our study, we used TEOAE to investigate the function of the outer hair cells and in case of absence of TEOAE we continued with CM. Our aim was to exclude other causes with potential to affect ABR such as conductive obstruction. DPOAE could be considered for a more frequency specific analyze. Finally, our study included participants with a wide age range, and while we adjusted for age in our analyses, age-related changes in auditory function could still influence the results. Future research should consider stratifying participants by age to better understand the impact of age on ABR wave amplitudes.

Clinical Significance

The significant reduction in wave I amplitude observed in our study underscores the need for comprehensive diagnostic measures beyond pure tone audiograms to detect AN in patients with a history of noise trauma. Early detection and intervention are crucial for improving outcomes for these individuals. Patients exposed to sudden and distinct noise should undergo thorough hearing evaluations, including ABR testing, to confirm or rule out AN. This approach can lead to more accurate diagnoses and better-targeted treatments, potentially mitigating the long-term impact of noise trauma on auditory function and quality of life. Moreover, our findings highlight the importance of considering sex differences in clinical assessments and research studies. Tailoring diagnostic protocols to account for these differences can improve the accuracy of ABR measurements and the interpretation of results.

Conclusion

Individuals exposed to sudden and distinct noise trauma exhibit a statistically significant reduction in the amplitude of wave I in ABR that may indicate AN. This finding underscores the importance of expanding hearing investigations after exposure to include comprehensive diagnostic measures for AN. Such expanded diagnostics are crucial for early detection and intervention, potentially improving outcomes for those affected by sudden and distinct traumatic noise.

Acknowledgments

We would like to thank all the participants for taking part in this study. We also thank the experienced audiologists Caroline Eckhardt and Marit Samstad Gerhardsen, who performed some of the hearing measurements.

Statement of Ethics

The study was approved by Swedish Ethical Review Authority, Sweden (Approval No. 2020-03913). Written informed consent was obtained from participants in this study.

Conflict of Interest Statement

No conflict of interest, financial, or otherwise are declared by the authors.

Funding Sources

This work was supported by grants from Umeå University and the FoU-department in the county of Jämtland Härjedalen. The funding supported the primary author’s (E.L.) research time. The funder had no role in study design, execution, analysis or manuscript conception, planning, writing or decision to publish.

Author Contributions

E.L. performed the data collection, analyzed the data and wrote the manuscript under supervision of M.W. E.K. and Å.K. contributed to analyzing data and reviewing the final version of the manuscript. F.Ö. contributed to statistical analysis and reviewing the manuscript. S.H. contributed to reviewing the final version of the manuscript and supervision. M.W. contributed with conceptualization, methodology, data curation, analysis of the data, funding acquisition, writhing the manuscript, and main supervision of the project.

Funding Statement

This work was supported by grants from Umeå University and the FoU-department in the county of Jämtland Härjedalen. The funding supported the primary author’s (E.L.) research time. The funder had no role in study design, execution, analysis or manuscript conception, planning, writing or decision to publish.

Data Availability Statement

The data that support the findings of this study are not publicly available due to privacy reasons but are available from corresponding author upon reasonable request.