Effects of Genes, Lifestyles, and Noise Kurtosis on Noise-Induced Hearing Loss

Objective: To explore the association of lifestyles, caspase gene (CASP), and noise kurtosis with noise-induced hearing loss (NIHL). Design: Three hundred seven NIHL individuals and 307 matched controls from factories in Chinese factories participated in this case–control study. Age, sex, noise exposure, exfoliated oral mucosa cells, and lifestyles of participants were gathered by the authors. The single nucleotide polymorphisms (SNPs) were genotyped using the Kompetitive Allele Specific polymerase chain reaction (KASP) method. Results: The risk of NIHL was higher for people who worked in the complex noise environment than for people exposed to steady noise environment (adjusted: OR = 1.806, P = 0.002). Smoking and regular earphone use increased the risk of NIHL (adjusted: OR = 1.486, P = 0.038). The GG genotype of the recessive model and G allele in rs1049216, together with the TT genotype of the recessive model in rs6948 decreased the NIHL risk (adjusted: OR = 0.659, P = 0.017). Oppositely, the AA genotype of additive model in rs12415607 had a higher NIHL risk (adjusted: OR = 1.804, P = 0.024). In the additive models, there was a positive interaction between noise kurtosis and CASP3 polymorphisms (RERI = 1.294, P = 0.013; RERI = 1.198, P = 0.031). Conclusions: Noise kurtosis, three SNPs (rs1049216, rs6948, and rs12415607), smoking and earphone use were found to be related to NIHL, and there was a positive interaction between noise kurtosis and CASP3. Results from this study can be used to prevent and detect NIHL and for genetic testing.

Keywords: CASP, interaction, lifestyle, NIHL, noise kurtosis

How to cite this article:
Yin X, Li Z, Zhao T, Yang L. Effects of Genes, Lifestyles, and Noise Kurtosis on Noise-Induced Hearing Loss. Noise Health 2023;25:143-57
Introduction

Hearing impairment is one of the most common human disabilities, and presents great risk to everyday life. In 2005, the WHO estimated that 278 million people worldwide were living with disabling hearing impairment. In 2012, the WHO released new estimates of disabling hearing impairment based on 42 population-based studies.[1],[2] There are approximately 30 million employees in the Americas and Europe who are exposed to potentially hazardous noise levels, and there are about 400 million employees who are in danger of hearing loss.[3] Noise-induced hearing loss (NIHL) results from interactions between genetic and environmental factors.[4] The main contributing factors to hearing impairment are non-occupational and occupational noise exposure, and individual characteristic such as sex, age, smoking, medical problems, and hearing protection device (HPD) usage.[5],[6],[7]

The cumulative noise exposure (CNE) index combines the equivalent sound level and duration of noise reception to assess the risk of hearing loss. The current international standards for noise exposure rely entirely on one energy indicator (ISO-1999, 2013). The equal energy hypothesis is used to establish and enforce noise criteria. This hypothesis assumes that the effect of noise on the cochlea is proportional to the noise exposure energy intensity multiplied by the exposure duration. Noise in industries, such as the sort produced by machines, is complex, fluctuating, and unstable, so equal energy hypothesis does not work for this type of noise.[8] To address this discrepancy, Erdreich et al. proposed to classify complex noise according to peak value statistics. The kurtosis was defined as the ratio of fourth moment to the squared second moment of the distribution from the sampled amplitudes of the noise.[9] The larger the peak value, the higher the impulse property of complex noise. Thus, the time-domain variables (peak value, duration, and interval between pulses) that affect hearing are reduced to a simple, easy-to-calculate noise classification.[10] A study has shown that kurtosis has a dose–response relationship with the detection rate of high-frequency NIHL in noise exposed workers.[11]

The term NIHL refers to the death and destruction of cochlear hair cells. In apoptosis, a specific gene controls an orderly, cell-independent death program. Activation of the apoptosis cascade results in hair cell death.[12] Downstream of the apoptosis cascade, caspase-3, and caspase-7 are apoptotic effectors that are activated by the upstream promoter.[13],[14] A study has shown that there is a certain correlation between mutations in HSP70-1, HSP70-2, and HSP70-hom and susceptibility to NIHL.[15] However, an earlier meta-analysis on the relationship between HSP70 genes and susceptibility to NIHL found no direct correlation between these genes and NIHL in the Chinese population. Only rs1061581 and rs2227956 were directly associated with NIHL in male North Caucasian populations.[16] Several studies suggest that functional differences between polymorphic variants of apoptosis-related genes may alter transcription factor binding, thereby inducing apoptosis and repressing primary cell transformation.[17] Additionally, previous studies demonstrate that lifestyle habits such as smoking, drinking, and earphone use are associated with hearing loss.[18],[19] Many epidemiological studies have shown that a lack of physical activity is associated with hearing loss, possibly because less physical activity may result in a reduced flow of blood, oxygen, and nutrients to the cochlea, leading to degeneration of the stria vascularis (SV). The blood vessels in the SV are necessary for delivering oxygen and nutrients such as glucose to the cochlea.[20],[21]

Using a case–control study, we evaluated the relationship between noise (introducing noise kurtosis), lifestyle factors, CASP, and their interaction with NIHL in Chinese populations. As a result of these studies, genetic screens for hearing loss could be developed.

Materials And Methods

Participants

We developed a database of individuals exposed to high level occupational noise (environmental noise >85 dB) from three factories in Ningbo and Wenzhou, Zhejiang Province, from October 2017 to December 2017 and from two factories in Hangzhou, Zhejiang Province, from October 2018 to December 2018. These are participant standards: (1) people who have only worked in one factory and in the same working conditions; (2) test difference of hearing threshold less than 30 dB at each frequency between the left and right ears; (3) no history of army service; (4) no hearing loss family history; (5) no ototoxic drug use; (6) no ear disease history; and (7) no diabetes. In our study 307 hearing loss patients had an average hearing threshold >25 dB (A) at high frequencies in both ears (3000, 4000, and 6000 Hz). An additional 307 age (±3 years) and gender-matched individuals with normal hearing were selected from this database and used as controls.

Questionnaire survey

Study participants filled out the form in a survey about noise exposure. We gathered the following information: (1) demographic information (age, sex, etc.); (2) noise exposure information (working factory, working time, duration of daily noise exposure, etc.); (3) lifestyle information (tobacco using, drinking, and earphone use). Variables were defined as follows: (1) smoking: smoking one or more cigarettes every day and for at least 1 year; (2) drinking: average daily alcohol consumption, for example, more than 50 g liquor, more than 150 g red wine, or more than 500 g beer, and lasting for at least 1 year; (3) earphone use: average weekly earphone use over 1 year; (4) health of the ears (ear disease history, drugs that cause ototoxicity intake history). Hangzhou Normal University’s Science Ethics Committee approved the study (2017LL107). All study participants signed informed consent forms. Trained investigators assisted participants to complete study questionnaires in a face-to-face meeting.

Noise exposure measurement

We used ASV5910-R digital noise dosimeter (Hangzhou Aihua Instrument Co., LTD.) to measure the continuous equivalent A-weighted sound level (LAeq, 8h) at 48 khz as the sampling rate for 8 hours. The noise dosimeter is equipped with a 1/4 inch microphone and has a frequency response range of 10 Hz to 20 kHz and a measurement range of 40–141 dB (A). Before and after each sampling period, we calibrated the noise dosimeter with a sound level calibrator machine (AWA6221B, HangzhouAihua Instrument Company). The noise dosimeter was worn on the shoulder of the worker with the microphone facing up (Supplementary material 1). After collected noise exposure data, we loaded the data from the noise dosimeter to computer and finished the follow-up analysis.

CNE was used to quantify each worker’s noise exposure by using Formula 1:

As the environment study participants worked in was the same, n equals 1. Formula 1 could be transformed into Formula 2:

In Formulae 1 and 2, n is the total number of jobs exposed to noise, LAeq, 8h is the equivalent 8-hour working day continuous A level-weighted noise exposure level, and Ti is the noise exposure time (in years). The CNE unit is dB (A) per year.

MATLAB (Natick, MA) was used to calculate the sampling kurtosis of the continuous 40-second time window for the entire shift. The Formula 3 showed the formula of calculating kurtosis:

where χi is the i-th value, x̅ is the sample mean, and β is the noise kurtosis. In principle, when β = 3, the noise is Gaussian (steady) noise, while β > 3 indicates complex noise. Kurtosis is proportional to the noise impulse energy. We defined β ≥ 10 as complex noise and β < 10 as steady noise. Kurtosis depends on two factors: the sampling rate of the noise waveform and the length of the calculation window. The 40-second window not only takes into account the computational efficiency, but also reflects the dynamic characteristics of complex noise. In addition, the 40-second window was determined based on a previous animal study.[22]

Hearing test and hearing loss diagnosis

To assess NIHL, experienced audiologists performed pure-tone audiometry for the left and right ears of each participant, according to the requirements of the "Occupational Noise-induced Hearing Loss Diagnostic Criteria" (GBZ49-2014). In a soundproof room with background noise <25 dB (A), the test was performed on both ears with pure tone at different frequencies (Supplementary material 2). Before the test, all the subjects were asked to be outside the worksite for at least 16 hours. The results of pure sound measurements were adjusted to age and gender according to the requirements in Appendix A of the ISO1999:2013 standard. High-frequency NIHL (hNIHL) was diagnosed based on the binaural hearing threshold level at 3, 4, and 6 kHz (HTL 3, 4, 6 kHz) using the following formula:

Participants with HTL3, 4, 6kHz >25 dB HL were diagnosed with hearing loss of binaural high frequency. Patients with HTL3, 4, 6kHz ≤ 25 dB HL were diagnosed with non-binaural high frequency hearing loss.

SNP selection

SNPs were selected from previous studies reporting associated hearing loss and hearing loss-associated genes from the GeneCard databases (http://www.genecards.org) and PubMed (http://www.ncbi.nlm.nih.gov/snp). Five SNPs were selected in the CASP3 and CASP7 genes, including rs1049216, rs6948, and rs1127687 at the 3’ end, rs12415607 in the 2kb-upstream region, and rs2227310 in the exon region (Table S1).

DNA extraction and genotyping

All the participants‘ oral mucosa cells were collected by Yongming flocking swabs. We extracted DNA by using Tiangen Oral Mucosa Genomic DNA extraction kits (Tiangen Biotech, Beijing, China) (Supplementary material 3). The genotyping analysis was done using the Kompetitive allele-specific polymerase chain reaction (KASP) method by Hangzhou Hechuan Biological Technology Co., Ltd. (Table S2). To control the quality, we randomly selected 10% of the samples and re-classified the genes, and the concordance of five SNPs was 100%.

Statistical analysis

To perform statistical analysis, we used SPSS 24.0. Continuous variables for normal distribution were expressed as mean ± standard deviation (SD) and were analyzed by Student’s t-test, non-normally distributed continuous variables distribution were expressed as median (P25, P75) and were analyzed by Mann–Whitney U test. Categorical variables were described as percentage and analyzed by Chi-squared (χ2) test. We used binary logistic regression to conduct association analysis in noise exposure, gene, lifestyle choice, and NIHL. The association analysis was reported in the form of OR and 95% CI. We used crossover analysis to explore the interactions of noise, genes, and lifestyles, then determined by Microsoft Excel according to Knol et al.[23]P < 0.05 indicated that the difference was statistically significant (showed in bold in tables). The multiple comparison method was the Benjamini–Hochberg procedure.

Results

General characteristics

This research involved 307 NIHLs and 307 controls, with a median age of 36 and 34 years, respectively. The median kurtosis in the NIHL group [7.25 (4.63–14.30)] was significantly higher than that in the control group [5.85 (4.06–12.51); P = 0.006]. The proportion of individuals exposed to complex noise was significantly greater in the NIHLs than that in the control group. The chi-squared (χ2) test found a significant difference for smoking (P = 0.043). The median threshold shift in the control group was lower than that in the NIHL group ([Table 1]).

Table 1 General characteristics, noise exposure, and lifestyles distribution between NIHL and control groups

Click here to view

Different genetic models of three SNPs

We searched for the association between CASP polymorphisms and NIHL. There are four of the five SNPs in Hardy–Weinberg equilibrium in the control subjects (P > 0.05) as shown in Table S1. [Table 2] shows the three positive SNPs. The Chi-squared (χ2) test found the proportion of people carrying the AA+AG genotype of the rs1049216 recessive model as well as the A allele of the rs1049216 allele model in the NIHL group was significantly higher than in the control group (P < 0.05). But we did not find significant differences in the rs6948 and rs12415607 genetic models in this research. [Table 2]

Table 2 Different genetic models of three SNPs between NIHL and control groups

Click here to view

Associations of noise exposure–lifestyles–genetic models interaction with risk of NIHL

We examined the association between noise exposure, lifestyles, and genetic models with NIHL. In [Table 3], with adjustment of characters (sex, age, education background, and duration years), we found that kurtosis, smoking, the rs1049216 recessive and allele models, and the rs12415607 additive model were significantly associated with NIHL. The people who were exposed to the environment of steady noise (β < 10) had a lower risk of NIHL compared with those workers who were exposed to the complex noise environment (β ≥ 10). Smoking was more likely to cause NIHL in people than those without smoking. This means that the risk of people carrying the AA+AG genotype of rs1049216 developing NIHL is greater than that of those with the GG genotype of rs1049216. In addition to the recessive model, we found that the allele model of rs1049216 had a similar situation. In the rs1049216 allele model, people who carry the A allele have a high risk of developing NIHL than people with the G allele. In the rs12415607 additive model, compared with individuals who carry the CC genotype, those carrying the AA genotype were more likely to cause NIHL.

Table 3 Associations of noise exposure, lifestyles, and genetic models with risk of NIHL

Click here to view

After adjusting gender, age, education, and duration years (working years), we found significant differences between NIHL and the control group in earphone use and the rs6948 recessive model. Compared to the people who have no earphone use, regular earphone use posed a higher risk to be NIHL. In the rs6948 recessive model, individuals who carried the TT genotype were less likely to cause NIHL than those carrying the GG+GT genotype.

Interaction

NIHL is a complex disease caused by a combination of genetic and environmental factors. The interactions were analyzed in terms of noise kurtosis, lifestyle details (like smoking and earphone use), and three SNPs (rs1049216 recessive model, rs6948 recessive model, and rs12415607 additive model). The recessive model of SNP, which is interacting with noise and lifestyle, is 2 × 2 level. While interacting with other factors, the additive model is 2 × 3 level.

Association of noise-kurtosis-CASP-polymorphisms interaction for the risk of NIHL

[Table 4] shows the effects of the interaction between noise kurtosis and CASP3 rs1049216, rs6948 recessive models on the risk of NIHL. After we adjusted the sex, age, education, and duration years (working years), the workers exposed to complex noise environments who had the AA+AG genotype of rs1049216 or the GG+GT genotype of rs6948 were at a higher risk of NIHL compared with people exposed to steady noise who carried the GG genotype of rs1049216 or the TT genotype of rs6948. Within the strata of complex noise, workers carrying the AA+AG genotype of rs1049216 or the GG+GT genotype of rs6948 had a higher tendency of NIHL than those individuals who carrying the GG genotype of rs1049216 or the TT genotype of rs6948. At the level of risk genotype, the population in complex noise environment showed a greater risk of NIHL than the population in stable noise environment. We found a positive result from an interaction between noise kurtosis and CASP3 polymorphisms in the additive models.

Table 4 Interaction between noise kurtosis and CASP3 polymorphisms for the risk of NIHL

Click here to view

Table S3 shows the effects of the interaction between CASP7 rs12415607 addictive model and noise kurtosis for the risk of NIHL in the 2 × 3 level analysis. When we adjusted for sex, age, education background and duration years (working years), compared with individuals carrying CC genotype who worked in steady noise environment, those who worked in complex noise environment carrying CA genotype or AA genotype of rs12415607 had a higher tendency of NIHL. Compared with those working in a steady noise environment, individuals with the CA genotype who worked in complex noise environment had a higher NIHL risk. In the additive models, we did not find an interaction between the rs12415607 polymorphism and noise kurtosis (P-value of RERI >0.05).

Association of CASP polymorphisms–lifestyles interaction for the risk of NIHL

It shows the interaction effects between CASP3 rs6948 recessive model and lifestyle choice for the NIHL risk in [Table 5]. Compared with the TT genotype group who had no smoking or using earphone, the GG+GT genotype group performing smoking or regular earphone use were at a higher risk of NIHL. Within the strata of TT, individuals who had the smoking habit had a higher risk of NIHL than those without smoking. In the GG+GT genotype people using earphone regular was 1.396 times higher than those no earphone use. Within the strata of regular earphone use, the people carrying GG+GT genotype were at higher NIHL risk compared with those carrying the TT genotype. The interaction was not found between rs6948 polymorphism and lifestyle choice.

Table 5 Interaction between rs6948 polymorphism and lifestyles for the risk of NIHL

Click here to view

The effects of the interaction for NIHL tendency between CASP3 rs1049216 recessive model and lifestyle choice are shown in [Table S4. Compared with individuals carrying the GG genotype who have not tobacco using or earphone use, those carrying the AA+AG genotype who had smoking or regular earphone use had a greater NIHL risk. Within the strata of AA+AG, people who performed regular earphone use had a higher NIHL tendency than those never using earphone. Within the strata of regular earphone use, people carrying the AA+AG genotype were more likely to be NIHL than those GG genotype. We did not find the interaction between rs1049216 and lifestyles in the additive models.

[Table 6] shows the effects of the interaction between CASP7 rs12415607 additive model and lifestyles for the risk of NIHL in the 2×3 level analysis. When we adjusted for sex, age, education background, and duration years (working years) compared with the CC genotype workers who had no tobacco using or earphone use with those carrying the CA genotype who had smoking or regular earphone use had a higher NIHL tendency, and those having the AA genotype who used earphone regularly also had a higher NIHL risk. Within the strata of risk lifestyles (smoking or regular earphone use), the AA genotype workers were easier to cause NIHL than those workers with the CC genotype. The interaction was not found between rs12415607 and lifestyle choice in the addictive models.

Table 6 Interaction between rs12415607 polymorphism and lifestyles for the risk of NIHL

Click here to view

Association between noise-kurtosis and lifestyle choice interaction for the NIHL tendency

The effects which interacted by noise kurtosis and lifestyle choice for the NIHL tendency are shown in [Table 7]. Compared with individuals who worked in steady noise environment and also had no smoking or earphone use with those people who worked in a complex noise factory and had smoking or regular earphone use had a greater NIHL risk. In the complex noise group, the smoking people had a greater NIHL risk compared with the no smoking people. Within the strata of risk lifestyles (smoking or regular earphone use), people exposed to complex noise were more likely to cause NIHL compared with steady noise group. We did not find the interaction between lifestyles and noise kurtosis in the addictive models.

Table 7 Interaction between noise kurtosis and lifestyles for the risk of NIHL

Click here to view

Discussion

NIHL is a complex disease caused by genetic and environmental factors. However, NIHL is mainly determined by the level of biological damage, which is caused by noise. We investigated the relation between noise characteristics and hearing loss (HL). Steady and complex noise can be distinguished by kurtosis simply and effectively. We used kurtosis which can transform noise pulse peak, duration, and pulse interval distribution into simple variables. Kurtosis make noise classification greatly simple. We showed that the NIHL group‘s median noise kurtosis was higher than the control group‘s, and corresponded to the proportion of complex noise group. Univariate logistic regression analysis showed that workers exposed to complex noise have a 1.753 times higher risk of developing NIHL compared to workers exposed to steady noise. After adjusted to age, sex, education, and duration years (working years), additional multivariate analysis indicated the NIHL risk in the complex noise group have a 1.806 times higher compared to individuals in the steady noise group. These results are consistent with studies by Theiry et al. and Xie et al.[8],[24] Noise with a high kurtosis value damages the auditory system by direct mechanical force and disruption of metabolism.[25] However, we did not observe an association between CNE and NIHL.

Cochlear hair cell damage and death caused NIHL. Apoptosis is one mechanism underlying noise-induced hair cells death.[12] Caspase activation is a critical step leading to apoptosis. Proteins in the caspase family are cysteine proteases that target specific aspartic acid residues. After activation, caspase cleaves specific substrates, leading to apoptosis. The process of apoptosis is actually a cascade amplification of the caspase hydrolysis substrate. Here, we selected SNPs of the apoptotic effector genes CASP3 and CASP7 for population epidemiological studies. Our analysis demonstrates that the three SNPs rs1049216 (CASP3), rs6948 (CASP3), and rs12415607 (CASP7) are associated with noise, and are related to hearing loss.

The rs1049216 SNP is an A→G mutation located in the CASP3 3’ untranslated region (UTR). At present, there are few studies on the correlation between rs1049216 and NIHL, in Chinese populations. Most studies regarding this locus focus on cervical and prostate cancer. The A and G alleles in this study individuals were 18.2% and 81.8%, which were basically consistent with Chinese population A (19.4%) and T (80.6%) alleles. But they were not consistent with European individuals A (73.4%) and T (26.6%) alleles. The results indicated that there may be racial differences in the locus allele frequency of rs1049216. Further, our study indicates a polymorphism in the rs1049216 locus in Chinese individuals. The case group and the control group showed three genotypes (AA, AG, and GG). After adjusted to age, sex, education, and duration years (working years), multivariate analysis indicated individuals carrying the GG genotype have a 0.682 times lower risk of developing NIHL compared to individuals with the AA+AG genotypes. Similarly, the risk of developing NIHL in individuals carrying the G allele is 0.743 times lower than individuals carrying the A allele. Our data showed the GG genotype of the recessive model and the G genotype of the allele model have protective effect against NIHL. However, it is inconsistent with the findings of Wu et al.[26] The rs6948 SNP is a G→T mutation located in the CASP3 3’ UTR. At present, there are few studies which research East Asian or European people on the correlation between rs6948 and NIHL. In this study people, the G and T alleles were 16.4% and 83.6%, which were difference from European population which is G (54.4%) and T (45.6%), but were almost same with East Asian population G (18.8%) and T (81.2%) alleles. These data indicated that there might be ethnic differences in the locus allele frequency of rs6948. Further, our studies indicate a polymorphism in the rs6948 locus in Chinese individuals. The case group and the control group showed three genotypes (GG, GT, and TT). After adjusting to age, sex, education, and duration years (working years), multivariate analysis indicated individuals carrying the TT genotype have a 0.694 times lower risk of developing NIHL compared to individuals with the GG+GT genotypes. Our data illustrated that the TT genotype of the recessive model is protective against NIHL. Caspase-3 is encoded by CASP3, and plays a central role in apoptosis initiation. The rs1049216 locus is located in the CASP3 3’ UTR. Recent studies have found that the 3’ UTR regulates mRNA stability,[27],[28],[29] translation efficiency, and protein localization independent of RNA localization.[30] Thus, the 3’ UTR mutation caused by rs1049216 does not affect the encoded amino acid. However, the mutation regulates caspase number and location in cells of the inner ear, thereby playing a role in hearing loss.

The rs12415607 SNP is a C→A mutation located in the CASP7 2kb upstream region. At present, there are few studies on the correlation between rs12415607 and NIHL, in East Asian or European people. The main researches regarding this locus focus on lung cancer and cervical cancer.[30],[31] The C and A alleles in this study people were 61.1% and 38.9%. These were not similar with European population C (75.8%) and A (24.2%) alleles. These data were basically consistent with East Asian population C (59.1%) and A (40.9%) alleles. This showed that there maybe ethnic differences in the rs12415607 locus allele frequency. Further, our results indicate a polymorphism in the rs12415607 locus in Chinese individuals. The case group and the control group showed three genotypes (CC, CA, and AA). After adjusting to age, sex, education, and working years, multivariate analysis pointed out the NIHL risk in people carrying AA genotype was 1.804 times higher than those people carrying CC genotype. Our data illustrated that the AA genotype of the addictive model is risk against NIHL. CASP7 encodes Caspase-7, which plays a central role in the initiation of apoptosis. The rs12415607 locus is located in the 2 kb upstream region of CASP7. Although this locus does not encode amino acids, it may affect promoter function, thus affecting the CASP7 expression.

This study analyzed the associations between lifestyle factors, including smoking, drinking, and earphone use, and the incidence of NIHL. After adjusting by sex, age, education, and working years, NIHL was significantly associated with smoking and earphone use, which increases the NIHL risk, consistent with previous studies.[32],[33]. Smoking-induced HL is likely due to vascular changes, including capillary contraction, increased blood viscosity, and cochlear anoxia.[34] Previous studies have also shown that long-term earphone use may impair hearing function, and the prevalence of hearing loss may increase significantly over time.[35],[36]

In this study, the relative excess risk of interaction (RERI) is used. We reported the RERI value, 95% CI, and P-value. We selected positive multivariate analysis results, including interactions between noise kurtosis, earphone use, and the three SNPs (rs1049216 [recessive model], rs6948 [recessive model], and rs12415607 [additive model]) to analyze environmental and genetic interactions. We selected positive multivariate analysis results for interactions between each of the following factors: noise kurtosis, earphone use, and the three SNPs (rs1049216 [recessive model], rs6948 [recessive model], and rs12415607 [additive model]) to evaluate the combined effect of environmental and genetic factors.

We determined for the first time that there is interaction to affect NIHL between noise kurtosis, genetics, and lifestyle. We found a positive interaction between noise kurtosis and rs1049216 and rs6948 (CASP3). These results showed that the NIHL tendency in people worked in complex noise environment (β ≥ 10) carrying the rs1049216 AA+AG or rs6948 GG+GT genotype is higher than the combined risk in individuals with the rs1049216 AA+AG or rs6948 GG +GT genotype alone and complex noise alone group. Used the interaction measurement recommended by Knol, it showed some stratified analysis results. Among the complex noise group, those carrying the rs1049216 AA+AG genotype or the rs6948 GG+GT genotype had the higher risk of NIHL. Similarly, among people carrying the rs1049216 AA+AG genotype or rs6948 GG+GT genotype, those exposed to complex noise environment (β ≥ 10) were at greater risk of NIHL than exposed to steady noise (β < 10).

NIHL is a multifactorial disease influenced by the interaction of genetic and environmental factors. In our research, it showed that CASP3 SNPs, lifestyle choice and working noise environment are each correlated with NIHL. In addition, our results proved there are interactions in NIHL between genes, environment, and lifestyle. Unfavorable working factory and genetic susceptibility will likely step up the risk of NIHL. With larger sample sizes combined with laboratory experiments will clarify the interaction mechanisms between mutations and environmental factors for NIHL in the next studies.

There were some advantages in our study. We used kurtosis to describe the noise pulse pattern and distinguish noise model that is steady or complex. Using this indicator, we studied the interactions between noise pulse pattern, genetics, and lifestyle. Importantly, this is the first study to associate kurtosis with these interactions. Previous studies focused on CNE instead of kurtosis. We used the interaction method to report the interaction index RERI, OR, and 95% CI. Meanwhile, these tables also showed the number of people in NIHL group and control group together with the results of the stratified analysis. Thus, we can find a more complete information of the raw data characteristics by the table-form analysis. This study also had some limitations. First, we analyze the interaction between single genetic factors and working environmental factors by using crossover analysis which is helpful. However, this analysis requires both factors as two-category variables. Second, self-reported data may caused data deviation which may affect study effectiveness. At last, the sample size maybe not large enough. In the future we validate our results with large samples, also we will explore the molecular mechanisms of interactions using functional assays.

Conclusion

Noise kurtosis, CASP SNPs (rs1049216, rs6948, and rs12415607), smoking, and earphone use were related to NIHL. We also indicated the interactions between noise kurtosis and CASP3 polymorphisms. These results show a theoretical basis which is useful on the prevention and genetic testing of NIHL.

Acknowledgments

The authors thank all noise-exposed workers for participating in the study.

Funding

This study was funded by Zhejiang Key Research and Development Program of China (No. 2015C03050; No.2015C03039), Hangzhou Major Science and Technology Innovation Project (No. 20132015A01), General Research Project of Education Department of Zhejiang(Y201943082).

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.

References

1.Olusanya BO. Addressing the global neglect of childhood hearing impairment in developing countries. Plos Med 2007;4:626-30.

2.Stevens G et al. Global and regional hearing impairment prevalence: an analysis of 42 studies in 29 countries. Eur J Public Health 2013;23:146-52.

3.Jeffree MS, Ismail N, Lukman KA. Hearing impairment and contributing factors among fertilizer factory workers. J Occup Health 2016;58:434-43.

4.Yang Q et al. Genetic variation in EYA4 on the risk of noise-induced hearing loss in Chinese steelworks firm sample. Occup Environ Med 2016; 73:823-8.

5.Anton SD et al. Successful aging: advancing the science of physical independence in older adults. Ageing Res Rev 2015;24:304-27.

6.Shargorodsky J et al. Change in prevalence of hearing loss in US adolescents. JAMA J Am Med Assoc 2010;304(7):772-8.

7.Stanbury M, Rafferty AP, Rosenman K. Prevalence of hearing loss and work-related noise-induced hearing loss in Michigan. J Occup Environ Med 2008;50:72-9.

8.Xie H-W. et al. The use of the kurtosis-adjusted cumulative noise exposure metric in evaluating the hearing loss risk for complex noise. Ear Hear 2016;37:312-23.

9.Erdreich J. A distribution based definition of impulse noise. J Acoust Soc Am 1986;79:990-8.

10.Qiu Wei MZ, Weichao X. The application of the kurtosis metric in evaluating hearing trauma from complex noise exposures. J Otol. 2016;14:701-7.

11.Xie HW et al. Dose-response relationship between noise kurtosis and noise-induced hearing loss. Zhonghua lao dong wei sheng zhi ye bing za zhi (Chin J Ind Hyg Occup Dis) 2021;39(7):493-7.

12.Op de Beeck K, Schacht J, Van Camp G. Apoptosis in acquired and genetic hearing impairment: the programmed death of the hair cell. Hear Res, 2011;281:18-27.

13.Fetoni AR et al. Noise-induced hearing loss (NIHL) as a target of oxidative stress-mediated damage: cochlear and cortical responses after an increase in antioxidant defense. J Neurosci. 2013;33:4011-23.

14.Lee WK et al. Polymorphisms in the Caspase7 gene and the risk of lung cancer. Lung Cancer 2009;65:19-24.

15.Konings A et al. Variations in HSP70 genes associated with noise-induced hearing loss in two independent populations. Eur J Hum Genet 2009;17:329-35.

16.Lei S et al. Association between polymorphisms of heat shock protein 70 genes and noise-induced hearing loss: a meta-analysis. Plos One 2017;12.

17.Yan S et al. HuGE systematic review and meta-analysis demonstrate association of CASP-3 and CASP-7 genetic polymorphisms with cancer risk. Genet Mol Res 2013;12:1561-73.

18.Hussain T et al., Early indication of noise-induced hearing loss in young adult users of personal listening devices. Ann Otol Rhinol Laryngol 2018;127:703-9.

19.Wang X et al. Alterations in gray matter volume due to unilateral hearing loss. Sci Rep 2016;6:25811.

20.Joo Y-H., Han K-d, Park KH. Association of hearing loss and tinnitus with health-related quality of life: the Korea National Health and Nutrition Examination Survey. Plos One 2015;10:e0131247.

21.Han C et al. Effects of long-term exercise on age-related hearing loss in mice. J Neurosci 2016;36:11308-19.

22.Hamernik RP, Ahroon WA. Sound-induced priming of the chinchilla auditory system. Hear Res 1999;137:127-36.

23.Knol MJ, VanderWeele TJ. Recommendations for presenting analyses of effect modification and interaction. Int J Epidemiol. 2012;41:514-20.

24.Thiery L, Meyer-Bisch C. Hearing loss due to partly impulsive industrial noise exposure at levels between 87 and 90dB (A). J Acous Soc Am 1988;84:651-9.

25.Zhao YM et al. Application of the kurtosis statistic to the evaluation of the risk of hearing loss in workers exposed to high-level complex noise. Ear Hear 2010;31:527-32.

26.Wu Y et al. Associations of genetic variation in CASP3 gene with noise-induced hearing loss in a Chinese population: a case-control study. Environ Health, 2017;16:78.

27.An JJ et al. Distinct role of long 3 ’ UTR BDNF mRNA in spine morphology and synaptic plasticity in hippocampal neurons. Cell 2008;134:175-87.

28.Jambhekar A, Derisi JL. Cis-acting determinants of asymmetric, cytoplasmic RNA transport. RNA 2007;13:625-42.

29.Lau AG et al. Distinct 3 ’ UTRs differentially regulate activity-dependent translation of brain-derived neurotrophic factor (BDNF). Proc Natl Acad Sci U S A 2010;107:15945-50.

30.Berkovits BD, Mayr C. Alternative 3 ’ UTRs act as scaffolds to regulate membrane protein localization. Nature 2015;522:363.

31.Cao S. et al. Prognostic assessment of apoptotic gene polymorphisms in non-small cell lung cancer in Chinese. J Biomed Res 2013;27:231-8.

32.Mofateh M et al. Effect of smoking on hearing loss in refractory’s factory male worker with occupational noise exposure in Iran. J Pak Med Associat 2017;67:605-8.