Effects of Composite Air Pollution and Genetic Susceptibility on Tinnitus Risk: A Large Population-Based Study

Abstract

Importance

Tinnitus, affecting 10% to 15% of the global population, is a debilitating condition often linked to hearing loss and neurological disorders. While air pollution is a known risk factor for respiratory and cardiovascular diseases, its association with tinnitus remains underexplored.

Objective

This study investigates the association between air pollution, polygenic risk score (PRS), and tinnitus prevalence.

Design

Cross-sectional analysis.

Setting

UK Biobank (UKB) population-based cohort.

Participants

Seventy-nine thousand two hundred seventy-seven individuals from the UKB with available tinnitus, air pollution, and genetic data.

Exposures

Air pollution measures (PM_2.5, PM_2.5–10, PM₁₀, NO₂, and NO_x). A composite air pollution score was calculated, and PRS was derived from 6 tinnitus-associated single-nucleotide polymorphisms.

Main outcome measures

Prevalence, frequency, and severity of tinnitus. Logistic regression models were used, adjusting for sociodemographic, health, and hearing-related covariates.

Results

Each interquartile range increase in air pollution score was associated with higher odds of current tinnitus (OR = 1.06, 95% CI: 1.03-1.08, P < .001), particularly transient tinnitus (OR = 1.04, P = .008). PRS was independently associated with prevalent tinnitus (P < .001), and higher OR were observed among individuals with both high air pollution exposure and high PRS (OR = 1.34, 95% CI: 1.18-1.52, P < .001). Individual pollutants showed weaker associations.

Conclusion

A joint association of air pollution exposure and genetic susceptibility with tinnitus prevalence was observed in this cross-sectional analysis, highlighting the importance of integrated environmental and genetic assessments in auditory health.

Relevance

Our findings emphasize the potential value of integrating environmental and genetic factors for risk stratification of tinnitus in populations.

Graphical Abstract

Keywords

tinnitus air pollution polygenic risk score UK Biobank gene-environment interaction

Key Messages

Composite air pollution exposure was associated with tinnitus prevalence.

A joint association of air pollution exposure and genetic susceptibility with tinnitus prevalence was observed.

This study suggests the potential value of considering both environmental and genetic factors for tinnitus risk stratification.

Introduction

Tinnitus, characterized by the perception of sound without an external source, is a widespread and often debilitating condition, impacting an estimated 10% to 15% of people worldwide.^1,2 It is frequently linked to hearing loss, psychological stress, and neurological disorders, significantly reducing the quality of life for affected individuals.^3
-5 Although the underlying causes of tinnitus remain poorly understood, environmental factors, particularly air pollution, have gained attention as potential contributors.

Air pollution, such as particulate matter (PM), and nitrogen dioxide (NO₂), is a major public health concern due to its well-documented effects on the respiratory and cardiovascular systems.^6
-8 However, its impact on auditory health, specifically tinnitus, has not been thoroughly investigated. Emerging evidence suggests that air pollutants may disrupt auditory function through mechanisms such as redox imbalance, chronic inflammatory responses, and neuronal damage.^9
-11 For example, PM_2.5 can cross the blood-brain barrier, triggering neuroinflammation that may interfere with neural circuits involved in hearing.^12,13 Similarly, NO₂ is known to generate reactive oxygen species (ROS),^14,15 which may cause oxidative damage to cochlear hair cells and auditory neurons, and contribute to hearing impairment. The diverse manifestations of tinnitus, ranging from transient to constant in frequency, and varying in severity, highlight the need for a detailed exploration of how different air pollutants may influence this condition.

Polygenic risk scores (PRS) have emerged as a powerful tool to quantify an individual’s genetic susceptibility to complex diseases by aggregating the effects of numerous genetic variants identified through genome-wide association studies.^16,17 Unlike single-gene mutations, PRS captures the cumulative contribution of multiple low-penetrance alleles, enabling risk stratification for conditions with polygenic architectures.¹⁸ Given the growing evidence supporting the heritable nature of tinnitus,¹⁹ and the increasing value of PRS as a tool for quantifying genetic susceptibility, integrating PRS into tinnitus research, either as a predictor or as an effect modifier, represents a necessary approach. A genome-wide association study (GWAS) meta-analysis by Clifford et al identified 6 loci associated with tinnitus,²⁰ which underscores tinnitus’s polygenic nature and highlights PRS as a potential biomarker for identifying high-risk individuals. Emerging evidence highlights the dual role of PRS in tinnitus, demonstrating both direct genetic effects and mediation of environmental influences including green space exposure.²¹ Air pollution, a pervasive environmental exposure, together with genetic susceptibility, may be associated with tinnitus prevalence jointly. While PM_2.5/NO₂ induce oxidative stress, tinnitus-associated single nucleotide polymorphisms (SNPs) in methionine sulfoxide reductase (MSRA, rs11249981) and collagen type XI alpha 1 chain (COL11A1, rs143424888) genes impair oxidative damage repair.²⁰ Thus, we hypothesize that PRS could possibly exhibit a joint association with air pollution and the prevalence of tinnitus.

By leveraging the rich and diverse dataset from the UK Biobank (UKB), this study aims to comprehensively examine the association between air pollution and tinnitus. We investigate the associations of overall air pollution exposure and individual pollutants with tinnitus prevalence, severity, and frequency. Furthermore, we explore the role of PRS in the relationship between air pollution and tinnitus. Our findings may provide valuable insights into the environmental and genetic factors of tinnitus and hopefully inform public health strategies to reduce its burden.

Methods

Recruitment and Study Design

The UKB is a large-scale prospective cohort study that has recruited over 500,000 participants aged 40–69 years from 22 research centers across the United Kingdom since 2006.²² It collected genetic, biological, and health-related data through questionnaires, interviews, physical measurements, and biological samples. Further details about the UKB resource are available at its official website: https://www.ukbiobank.ac.uk/.

We conducted a cross-sectional study to investigate the relationship between air pollution and tinnitus. There were 502,401 participants included in the baseline survey. First, we excluded individuals with missing answers to the tinnitus questionnaire and those with diseases of the middle ear, mastoid, inner ear, or other ear disorders based on the ICD-10 classification (n = 347,850, Table S1). Additionally, participants with missing data on air pollution, covariates and genetic information were excluded (n = 75,274). A total of 79,277 participants were included in the cross-sectional analysis to explore whether air pollution was associated with the occurrence, frequency, and severity of tinnitus. Furthermore, we conducted a sensitive analysis by excluding participants who reported any kinship relationships (Figure 1).

Figure 1.

Participant selection and study design.

Air Pollution

The annual average concentrations of air pollutants, including PM with aerodynamic diameter ≤2.5 μm (PM_2.5), PM with an aerodynamic diameter between 2.5 and 10 μm (PM_2.5-10), PM with an aerodynamic diameter ≤10 μm (PM₁₀), nitrogen dioxide (NO₂), and nitrogen oxides (NO_x), were obtained from a Land Use Regression (LUR) model.²³ The LUR model was based on the European Study of Cohorts for Air Pollution Effects (ESCAPE) monitoring done between January 26, 2010, and January 18, 2011, and air pollution estimates are representative for the year 2010. Studies have confirmed similar patterns of air pollution exposure in air pollution exposures of the UK Biobank cohort and the spatial distributions of air pollutants from the pubic UK Air Information Resources.^24,25

To assess the joint exposure to multiple air pollutants, we constructed an air pollution score by summing the concentrations of 5 air pollutants (PM_2.5, PM_2.5-10, PM₁₀, NO₂, and NO_x), weighted by the multivariable-adjusted regression coefficients (β coefficients) for tinnitus in the cross-sectional analysis (Table S2). The β coefficients were derived from Model 3, where each pollutant was included as an independent variable separately. The equation was: Air pollution score = (βPM_2.5 * PM_2.5 + βPM_2.5-10 * PM_2.5-10 + βPM₁₀ * PM₁₀ + βNO₂ * NO₂ + βNO_x * NO_x) * (5/sum of the β coefficients). The air pollution score ranged from −712.74 to 525.84, with higher values indicating higher joint air pollution exposure.

Tinnitus

Tinnitus data were obtained from the self-reported tinnitus questionnaire (Table S3). Participants were asked to complete the questionnaire on whether they had suffered from tinnitus, and the extent to which they were affected by tinnitus. The answer “current/present tinnitus” was defined as “current tinnitus,” while the answer “past tinnitus” or “no tinnitus” was defined as “no current tinnitus.” The responses identified as “current tinnitus” were classified as “constant tinnitus” (answered as “frequent tinnitus”) and “transient tinnitus” (answered as “infrequent tinnitus”). Additionally, the severity of tinnitus was classified as “upsetting tinnitus” (answered as “severe” or “moderate” tinnitus) and “not upsetting tinnitus” (answered as “slight” or “not at all”). Previous studies have provided detailed definitions regarding tinnitus.^26,27

Genetic Data

The genetic data were obtained from the UKB, and 6 independent loci associated with tinnitus has been identified by a previous study according to the GWAS,²⁰ which represents the first large-scale GWAS to report loci for tinnitus using UKB cohort, with subsequent studies demonstrating its applicability.²¹ Details of the SNPs information regarding the genotyping process, imputation, and quality control have been described in previous literature.²⁸ Specific information about the SNPs was presented in Table S4. We constructed a PRS based on the 6 SNPs at genome-wide significance levels for tinnitus in the UKB cohort. The calculation of weighted PRS has been performed in previous studies with the following formula: PRS = (β1 × SNP1 + β2 × SNP2 + β3 × SNP3 + β4 × SNP4 + β5 × SNP5 + β6 × SNP6), where βi (i = 1, 2, . . ., 6) was the estimated risk for each SNP in the genome-wide association study, and SNPi (i = 1, 2, . . ., 6) was the number of risk alleles for each SNP. Participants were stratified into low (Tertile 1), intermediate (Tertile 2), and high (Tertile 3) genetic risk according to the PRS.

Covariates

The covariates were obtained from touchscreen questionnaires, verbal interviews, and physical measurements at baseline, encompassing: age, sex, race, education, employment, Townsend deprivation index (TDI), smoking, alcohol consumption, body mass index (BMI), physical activity, healthy status, diabetes, cardiovascular disease, aspirin, ibuprofen, neuroticism score, noisy workplace, difficulty in hearing, speech recognition threshold (SRT) left, SRT right, hearing aid (HA). The neuroticism score was categorized into low (score 0-2), low-medium (score 3-5), medium-high (score 6-9), and high (score 10-13) levels.²⁹ The SRT was categorized into normal (SRT < −5.5 dB), insufficient (−5.5 dB to −3.5 dB), and poor (SRT > −3.5 dB).^30,31 Detailed information is shown in Table S5.

Statistical Analysis

The baseline characteristics were reported as the median (interquartile range, IQR) or mean (standard deviation, SD) for continuous variables, and frequencies (percentage) for categorical variables. Significant differences were calculated using the Wilcoxon-Mann-Whitney test, or Student’s t-test for continuous variables, and the Chi-square test for categorical variables.

In the cross-sectional analysis, logistic regression was used to estimate odds ratios (ORs) with 95% confidence intervals (CIs) for the association between air pollution and genetic risk with tinnitus, respectively. We first performed a crude model with unadjusted covariates, and then fitted 3 models with adjusted covariates, as follows: Model 1 was adjusted for age, sex, race, education, employment, TDI, smoking, alcohol, BMI, and physical activity; Model 2 was adjusted for age, sex, race, education, employment, TDI, smoking, alcohol, BMI, physical activity, healthy status, diabetes, cardiovascular disease, aspirin, ibuprofen, and neuroticism score; Model 3 was adjusted for age, sex, race, education, employment, TDI, smoking, alcohol, BMI, physical activity, healthy status, diabetes, cardiovascular disease, aspirin, ibuprofen, neuroticism score, noisy workplace, difficulty in hearing, SRT right, SRT left, and HA. Moreover, we explored the joint association of air pollution exposure and PRS with prevalent tinnitus.

Furthermore, the sensitivity analysis was conducted to verify the stability of previous results. First, we explored the association between the air pollution score and tinnitus, considering the interaction effect between air pollution score and variables. Second, we computed the Pearson correlation coefficients of air pollution concentrations to assess the correlation between different pollutants (Table S6), and then calculated the Variance Inflation Factor (VIF) values of the independent variables included in the air pollution score model to examine the presence of multicollinearity among the variables. High VIF values (typically greater than 10) suggest that a particular variable is highly collinear with other variables, which can lead to unreliable estimates of the model coefficients. By calculating the VIF, we can identify variables with multicollinearity to improve the stability and interpretability of the model (Table S7). Lastly, we constructed an alternative air pollution score using principal component analysis (PCA). PCA was applied to the concentrations of the air pollutants. The first 2 principal components (PC1 and PC2) were selected to represent joint exposure to multiple air pollutants, explaining a cumulative 91.63% of the total variance. Individual PC scores were calculated and examined in cross-sectional analyses, showing associations comparable to those of the primary weighted air pollution score (Table S8).

All analyses were conducted using R version 4.4.2 (R Studio Inc., Boston, MA, USA). The statistical tests were 2-sided, and a P-value <.05 was set as statistically significant.

Results

Baseline Characteristics of the Participants

Baseline characteristics of the study population are presented in Table 1. The study included 79,277 participants from the UK Biobank, of whom 10,501 (13.2%) reported current tinnitus. Compared to non-tinnitus participants, those with tinnitus were older (mean age 57.47 vs 55.18 years, P < .001), more likely to be male (56.1% vs 46.7%, P < .001), and had a higher prevalence of cardiovascular disease (30.5% vs 25.1%, P < .001). Notable differences existed across socioeconomic, occupational, and hearing-related factors.

Table 1.

Baseline Characteristics of Participants in the UK Biobank Study.

Characteristics		Tinnitus (N = 79,277)
Variable	Level	No current tinnitus (n = 68,776)	Current tinnitus (n = 10,501)	P-value
Age, years	/	55.18 (8.15)	57.47 (7.65)	<.001
Sex, n (%)	Female	36,633 (53.3)	4615 (43.9)	<.001
Sex, n (%)	Male	32,143 (46.7)	5886 (56.1)
Race, n (%)	Other	4813 (7.0)	588 (5.6)	<.001
Race, n (%)	White	63,963 (93.0)	9913 (94.4)
Education, n (%)	Low	12,324 (17.9)	2286 (21.8)	<.001
	Intermediate	26,173 (38.1)	4058 (38.6)
	High	30,279 (44.0)	4157 (39.6)
Employment, n (%)	Other	4770 (6.9)	789 (7.5)	<.001
	Retired	19,274 (28.0)	3786 (36.1)
	Working	44,732 (65.0)	5926 (56.4)
TDI^a	/	−1.43 (2.76)	−1.44 (2.79)	.859
Smoking, n (%)	Never	39,681 (57.7)	5600 (53.3)	<.001
	Previous	22,949 (33.4)	4010 (38.2)
	Current	6146 (8.9)	891 (8.5)
Alcohol, n (%)	Never	2339 (3.4)	329 (3.1)	<.001
	Previous	2018 (2.9)	391 (3.7)
	Current	64,419 (93.7)	9781 (93.1)
BMI^a (kg/m²)	Normal or underweight	24,796 (36.1)	3530 (33.6)	<.001
	Over weight	29,050 (42.2)	4559 (43.4)
	Obese	14,930 (21.7)	2412 (23.0)
Physical activity, n (%)	Low	11,418 (16.6)	1906 (18.2)	<.001
	Moderate	28,537 (41.5)	4251 (40.5)
	High	28,821 (41.9)	4344 (41.4)
Healthy status, n (%)	Other	14,710 (21.4)	2089 (19.9)	.001
Healthy status, n (%)	Good/Fair	54,066 (78.6)	8412 (80.1)
Diabetes, n (%)	No	65,818 (95.7)	9956 (94.8)	<.001
Diabetes, n (%)	Yes	2958 (4.3)	545 (5.2)
Cardiovascular disease,^a n (%)	No	51,533 (74.9)	7299 (69.5)	<.001
Cardiovascular disease,^a n (%)	Yes	17,243 (25.1)	3202 (30.5)
Aspirin, n (%)	No	61,801 (89.9)	9049 (86.2)	<.001
Aspirin, n (%)	Yes	6975 (10.1)	1452 (13.8)
Ibuprofen, n (%)	No	58,371 (84.9)	8797 (83.8)	.004
Ibuprofen, n (%)	Yes	10,405 (15.1)	1704 (16.2)
Neuroticism score,^a n (%)	Low	28,431 (41.3)	3668 (34.9)	<.001
	Low-medium	20,652 (30.0)	3109 (29.6)
	Medium-high	15,349 (22.3)	2749 (26.2)
	High	4344 (6.3)	975 (9.3)
Noisy workplace, n (%)	No	56,582 (82.3)	7501 (71.4)	<.001
Noisy workplace, n (%)	Yes	12,194 (17.7)	3000 (28.6)
Difficulty in hearing, n (%)	No	55,841 (81.2)	5252 (50.0)	<.001
Difficulty in hearing, n (%)	Yes	12,935 (18.8)	5249 (50.0)
SRT right,^a n (%)	Normal	54,133 (78.7)	7659 (72.9)	<.001
	Insufficient	12,737 (18.5)	2309 (22.0)
	Poor	1906 (2.8)	533 (5.1)
SRT left,^a n (%)	Normal	54,666 (79.5)	7757 (73.9)	<.001
	Insufficient	12,295 (17.9)	2228 (21.2)
	Poor	1815 (2.6)	516 (4.9)
HA,^a n (%)	No	68,032 (98.9)	10,041 (95.6)	<.001
HA,^a n (%)	Yes	744 (1.1)	460 (4.4)
PM_2.5^a (μg/m³)	/	9.86 (0.92)	9.84 (0.91)	.025
PM_2.5-10^a (μg/m³)	/	6.42 (0.84)	6.42 (0.85)	.648
PM₁₀^a (μg/m³)	/	16.26 (1.82)	16.23 (1.84)	.226
NO₂^a (μg/m³)	/	26.85 (7.31)	26.50 (7.19)	<.001
NO_x^a (μg/m³)	/	43.29 (14.38)	42.81 (14.10)	.001
Air pollution score^a	/	199.88 (115.60)	206.13 (114.27)	<.001
Genetic risk category, n (%)	Low genetic risk	16,840 (24.5)	2498 (23.8)	<.001
	Intermediate genetic risk	34,942 (50.8)	5177 (49.3)
	High genetic risk	16,994 (24.7)	2826 (26.9)

Descriptive statistics are provided as the mean (SD) or median (range) for continuous factors, and frequencies and percentages for categorical factors.

TDI: Townsend deprivation index; BMI: body mass index—the classification of BMI referred to WHO standard categories and was classified into obese (>30 kg/m²), over weight (25-29.9 kg/m²), normal or underweight (<24.9 kg/m²); Cardiovascular disease: cardiovascular disease including heart attack, angina, stroke and high blood pressure; Neuroticism score: low (0-2), low-medium (3-5), medium-high (6-9), high (10-13); SRT: speech reception threshold; HA: hearing aid; PM_2.5: particular matter with aerodynamic diameter ≤2.5 μm; PM_2.5-10: particular matter with an aerodynamic diameter between 2.5 and 10 μm; PM₁₀: particular matter with an aerodynamic diameter ≤10 μm; NO₂: nitrogen dioxide; NO_X: nitrogen oxides; Air pollution score: derived by the weighted average score of 5 air pollutants.

Air Pollution and Tinnitus

Our analysis revealed significant associations between composite air pollution and tinnitus outcomes in cross-sectional study (Table 2; Table S9). Each IQR increase in air pollution score corresponded to 6% higher odds of current tinnitus (OR = 1.06, 95% CI: 1.03-1.08, P < .001), exhibiting a clear dose-response relationship (P-trend < .001). This association persisted through full adjustment (Model 3 OR = 1.03, 95% CI: 1.01-1.06, P = .008). Frequency analysis showed differential associations: transient tinnitus maintained significant associations across all models (Model 3 OR = 1.04, P = .008), while constant tinnitus associations attenuated after adjustment. Air pollution showed limited associations with tinnitus severity, with non-upsetting tinnitus demonstrating borderline significance in crude model (OR = 1.02, 95% CI: 1.01-1.04; P = .012) that attenuated in adjusted models, while no associations were observed for upsetting tinnitus across all models. Moreover, analysis of individual air pollutants revealed minor and inconsistent associations with tinnitus prevalence that failed to maintain statistical significance after full covariate adjustment (Figure S1). Also, we found that all the variables showed very limited multiplicative interaction with air pollution score (Table S10), thus subgroup analysis were not further performed. Overall, we found that the composite air pollution score showed stronger associations than any single pollutants. While our study included longitudinal analyses (Table S11) among 1227 participants with current tinnitus, the results were largely null, and did not clearly reproduce the cross-sectional associations.

Table 2.

Logistic Multivariable Regression Analysis of the Association Between Air Pollution Score and the Occurrence, Frequency, and Severity of Tinnitus Reported for the Cross-Sectional Study in Crude Model.

Air pollution score	Current tinnitus^a (n = 10,501)		The frequency of tinnitus (n = 10,501)				The severity of tinnitus (n = 18,666)
	Current tinnitus^a (n = 10,501)		Constant tinnitus^a (n = 4,287)		Transient tinnitus^a (n = 6,214)		Upsetting tinnitus^a (n = 3,134)		Not upsetting tinnitus^a (n = 15,532)
	Odds ratio (95% CI)	P-value	Odds ratio (95% CI)	P-value	Odds ratio (95% CI)	P-value	Odds ratio (95% CI)	P-value	Odds ratio (95% CI)	P-value
Each IQR increment	1.06 (1.03, 1.08)	<.001	1.07 (1.03, 1.10)	<.001	1.05 (1.02, 1.08)	<.001	0.98 (0.95, 1.02)	.297	1.02 (1.01, 1.04)	.012
Q1	1.00 (ref)	—	1.00 (ref)	—	1.00 (ref)	—	1.00 (ref)	—	1.00 (ref)	—
Q2	1.05 (0.98, 1.12)	.162	1.17 (1.06, 1.29)	.002	0.97 (0.90, 1.06)	.538	1.04 (0.93, 1.16)	.535	0.99 (0.93, 1.05)	.703
Q3	1.10 (1.03, 1.17)	.005	1.15 (1.04, 1.27)	.007	1.07 (0.98, 1.16)	.120	1.00 (0.89, 1.12)	.994	1.02 (0.97, 1.08)	.471
Q4	1.13 (1.05, 1.20)	<.001	1.22 (1.11, 1.35)	<.001	1.07 (0.98, 1.16)	.132	0.97 (0.87, 1.09)	.651	1.07 (1.01, 1.13)	.016
Q5	1.19 (1.11, 1.27)	<.001	1.27 (1.15, 1.40)	<.001	1.14 (1.05, 1.23)	.002	0.99 (0.88, 1.10)	.796	1.06 (1.00, 1.12)	.047
P-trend		<.001		<.001		<.001		.485		.002

Crude model was unadjusted.

Abbreviations: CI, confidence interval; IQR, interquartile range.

Current tinnitus: participants responded: “Yes, now most or all of the time” or “Yes, now a lot of the time” or “Yes, now some of the time”; Constant tinnitus: participants responded: “Yes, now most or all of the time” or “Yes, now a lot of the time”; Transient tinnitus: participants responded: “Yes, now some of the time”; Upsetting tinnitus: participants responded: “severely” or “moderately”; Not upsetting tinnitus: participants responded: “slightly” or “not at all.”

Joint Association of Air Pollution and PRS With Tinnitus

We first observed that PRS has a significantly positive association with the OR of tinnitus, which remained significant in the multivariable-adjusted models (Table S12). In the further joint analyses, participants with high PRS demonstrated consistently elevated tinnitus prevalence across all models for current tinnitus as well as constant and transient tinnitus, whereas low or intermediate PRS showed less significant association (Figure 2; Figure S2). Moreover, compared to those with the low pollution exposure and genetic risk, individuals with the high air pollution quintile and PRS exhibiting substantially increased current tinnitus (OR = 1.34, 95% CI: 1.18-1.52, P < .001), with comparable associations observed for constant tinnitus (OR = 1.55, 95% CI: 1.27-1.89, P < .001) and transient tinnitus (OR = 1.21, 95% CI: 1.03-1.23, P = .019) but not for upsetting or not upsetting tinnitus. This joint association was statistically significant across all the adjusted models. Moreover, only negligible impacts were observed after excluding participants with at least one relative in the UKB study (Figure S3).

Figure 2.

Joint associations of air pollution exposure and genetic susceptibility with tinnitus by occurrence, frequency, and severity in the crude model.

Discussion

This study represents one of the first large-scale investigations into the relationship between air pollution and tinnitus, utilizing the extensive data from the UKB cohort. Cross-sectional analyses indicate that cumulative exposure to multiple air pollutants, as captured by the composite air pollution score, was associated with higher odds of prevalent tinnitus, whereas individual pollutants demonstrate minimal independent associations. Notably, we identified a joint association of air pollutants and genetic factors with tinnitus prevalence, with individuals exhibiting both high polygenic risk and elevated air pollution exposure showing higher odds of tinnitus compared to those with low exposure and genetic predisposition. These findings highlight the potential value of jointly considering environmental and genetic factors for understanding heterogeneity in tinnitus.

The association between composite air pollution scores and tinnitus observed in cross-sectional study builds upon emerging evidence linking environmental exposures to auditory dysfunction. Previous study revealed that higher levels of PM₁₀, NO_x, and NO₂ were linked to an increased risk of hearing impairment (PM₁₀: OR = 1.15, 95% CI: 1.02-1.30; NO_X: OR = 1.02, 95% CI: 1.00-1.03; NO₂: OR = 1.03, 95% CI: 1.01-1.06), and further identified an interactive effect between air pollution exposure and hearing impairment.³² Moreover, an analysis based on the UKB reported an inverse association between residential greenness and hearing impairment,³³ while it is suggested greenness may beneficially influence the relationship between air pollution and health outcomes.³⁴ These findings align with previous epidemiological studies on the detrimental impacts of air pollution on neurological and sensory health.^35
-37 While previous research has primarily focused on hearing loss, our findings expand this paradigm by revealing that exposure to multiple air pollutants is associated with tinnitus prevalence. As demonstrated in previous studies, oxidative stress and neuroinflammation are increasingly recognized as key pathogenic mechanism in tinnitus.³⁸ In this context, air pollution may contribute to auditory pathway damage by inducing oxidative stress–related mitochondrial dysfunction³⁹ or by promoting inflammatory responses,⁹ providing a plausible explanation for the associations between air pollution and tinnitus observed in our study. However, the prospective study did not consistently reproduce the cross-sectional results, which may be attributable to several considerations: (a) the relatively short follow-up period for assessing air pollution exposure and tinnitus outcomes, as the initial assessment visit occurred between 2006 and 2010, with the most recent follow-up extending to 2019 and beyond; (b) cross-sectional associations may be influenced by reverse causation; (c) assessment of residential-based air pollution exposure or self-reported tinnitus outcomes may have attenuated prospective associations; (d) the limited number of participants in the longitudinal analyses may have reduced statistical power.

The stronger associations seen with combined pollutants than with individual components likely reflect the joint distribution of various exposures. Previous studies have shown that PM_2.5 and NO₂ co-exposure induces significantly greater tissue injury than either pollutant alone.⁴⁰ This aligns with our observation that the air pollution score showed consistent associations, while individual pollutants exhibited only limited associations. This observation may be explained by multiple pathways: (a) PM_2.5 can induce the air-blood barrier disruption in lung tissues,⁴¹ suggesting that PM_2.5-induced systemic inflammation may increase cochlear vulnerability to NO₂-mediated oxidative damage; (b) combined exposures may overwhelm cellular antioxidant defenses; and (c) different pollutants may target distinct components of the auditory pathway (eg, PM_2.5 more likely affecting cochlear blood flow while NO₂ damages hair cells).⁴² In addition, while our results might suggest that individual components have limited biological impact, they could likely reflect the relatively modest impact expected for single pollutants, the high correlation between pollutants, or the potential non-linear exposure-response relationships that our models were unable to fully capture. Moreover, the specificity of associations for tinnitus presence rather than severity suggests that air pollution may primarily influence tinnitus prevalence rather than its perceptual significance. This aligns with emerging studies distinguishing between peripheral auditory damage (which may initiate tinnitus) and central neural plasticity mechanisms (which determine tinnitus severity and bothersomeness).^43,44 Our finding that the associations with air pollution were strongest for constant tinnitus supports this framework, as persistent tinnitus is more strongly linked to peripheral auditory damage than transient forms.⁴⁵

A novel aspect of our study is the exploration of PRS as a potential factor jointly associated with air pollution in relation to tinnitus prevalence. We constructed the PRS using SNPs identified in the first large-scale GWAS of tinnitus, obtained from the UKB cohort population, and these loci demonstrated stable predictive performance for tinnitus in our multivariable-adjusted models. The PRS showed robust associations independent of exposures to air pollutants, and the stronger associations observed in the joint analyses highlight the potential relevance of considering both genetic and environmental factors. For example, individuals with high PRS and high pollution exposure exhibited higher odds of constant tinnitus (OR = 1.55). Previous studies have suggested that air pollution might alter DNA methylation patterns in auditory neural pathway genes,^46
-48 potentially unmasking genetic predispositions. This is particularly relevant for our finding that the association between PRS and prevalent tinnitus differed across levels of air pollution exposure. The PRS included multiple genes that have been annotated in relation to oxidative stress and neural processes described in prior literature.²⁰ Also, these results highlight the importance of integrating genetic and environmental data to better understand the heterogeneity of complex conditions like tinnitus. However, as the SNPs used to construct the PRS were derived from UKB, the score may be influenced by genetic characteristics. Validation from larger, independent GWAS will be essential in future research.

The strengths of this study include its large sample size, comprehensive assessment of multiple pollutants, detailed characterization of tinnitus outcomes, and the incorporation of genetic data through PRS. However, our study has several limitations that should be acknowledged. First, as with all studies utilizing the UKB, the generalizability of our findings may be limited by the cohort’s demographic characteristics. The UKB participants are predominantly of European ancestry, middle-aged to older adults, and tend to be healthier and more socioeconomically advantaged than the general population, which may introduce selection bias and limit the generalizability of our findings to diverse populations. Second, as this study is primarily cross-sectional and lacks replication in the prospective analysis, causal inference is limited. Therefore, our findings should be interpreted as exploratory, and future studies with extended follow-up periods are needed to better capture the long-term impact of air pollution on tinnitus. Third, the assessment of air pollution exposure was based on LUR modeled estimates rather than direct measurements, which may introduce exposure misclassification, particularly among highly mobile individuals. Although these models are widely used and validated, they may not fully capture individual-level variations in exposure. Such nondifferential misclassification may bias the association toward the null, leading to conservative effect estimates. Fourth, tinnitus was self-reported, which may lead to recall bias and result in misclassification of tinnitus status. However, the UKB’s robust validation procedures and large sample size help mitigate this concern to some extent. Finally, while we adjusted for a range of confounders, residual confounding by unmeasured factors (eg, occupational noise exposure, dietary habits, or other environmental stressors) cannot be ruled out. Future studies should incorporate more detailed exposure assessments and consider additional confounding variables to strengthen the validity of the findings.

Conclusions

This study observed associations between air pollution exposure and tinnitus prevalence that differed across levels of genetic susceptibility. Our findings highlight the importance of considering both environmental and genetic factors in tinnitus risk stratification. These hypothesis-generating findings provide a foundation for future longitudinal studies to examine tinnitus risk in relation to air pollution exposure and genetic susceptibility.

Supplemental Material

sj-docx-1-ohn-10.1177_19160216261442718 – Supplemental material for Effects of Composite Air Pollution and Genetic Susceptibility on Tinnitus Risk: A Large Population-Based Study

Supplemental material, sj-docx-1-ohn-10.1177_19160216261442718 for Effects of Composite Air Pollution and Genetic Susceptibility on Tinnitus Risk: A Large Population-Based Study by Ding Yang, Zi-Xuan Huang, Lin-Qiu Li, Hang Li, Jie Deng, Yi Wei, Kai-Tian Chen, Guan-Xia Xiong, Wen-Bin Lei, Lin Chen and Shu-Bin Fang in Journal of Otolaryngology - Head & Neck Surgery

Footnotes

Author Contributions

Ding Yang: Conceptualization, Formal analysis, Data curation. Zi-Xuan Huang: Formal analysis, Data curation. Lin-Qiu Li, Hang Li, Jie Deng, Yi Wei, Kai-Tian Chen, Guan-Xia Xiong: Methodology, Visualization. Wen-Bin Lei, Chen Lin: Writing—review & editing, Investigation. Shu-Bin Fang: Writing—original draft, Supervision, Project administration.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by grants from National Natural Science Foundation of China (grant no. 82101185), Natural Science Foundation of Guangdong Province (grant no. 2024A1515010530, 2025A1515011837), and Basic Research Program of Guangzhou Municipal Science and Technology Bureau (no. 2024A04J4695).

Ethics Statement

The UKB study received ethical approval from the North West Multi-Centre Research Ethics Committee (REC reference: 21/NW/0157). Informed consent was obtained electronically from all participants. Further details about the UKB resource are available at its official website: .

ORCID iDs

Zi-Xuan Huang

Shu-Bin Fang

Supplemental Material

Additional supporting information is available in the online version of the article.

References

Baguley

McFerran

Hall

Tinnitus Lancet 2013;382:1600-1607.

McCormack

Edmondson-Jones

Somerset

Hall

A systematic review of the reporting of tinnitus prevalence and severity. Hearing Res 2016;337:70-79.

Langguth

Kreuzer

Kleinjung

De Ridder

Tinnitus: causes and clinical management. Lancet Neurol 2013;12:920-930.

Shore

Roberts

Langguth

Maladaptive plasticity in tinnitus—triggers, mechanisms and treatment. Nat Rev Neurol 2016;12:150-160.

Bartels

Staal

Albers

FW.

Tinnitus and neural plasticity of the brain. Otol Neurotol 2007;28:178-184.

Brook

Rajagopalan

Pope

3rd , et al. Particulate matter air pollution and cardiovascular disease: an update to the scientific statement from the American Heart Association. Circulation 2010;121:2331-2378.

Cai

Pei

, et al. Association between airborne fine particulate matter and residents’ cardiovascular diseases, ischemic heart disease and cerebral vascular disease mortality in areas with lighter air pollution in China. Int J Env Res Public Health 2018;15:1918.

Dabass

Talbott

Rager

, et al. Systemic inflammatory markers associated with cardiovascular disease and acute and chronic exposure to fine particulate matter air pollution (PM(2.5)) among US NHANES adults with metabolic syndrome. Environ Res 2018;161:485-491.

Block

Calderón-Garcidueñas

Air pollution: mechanisms of neuroinflammation and CNS disease. Trends Neurosci 2009;32:506-516.

10.

Park

Kim

Choi

YH.

Long-term exposure to ambient air pollutants and hearing loss in Korean adults. Sci Total Environ 2022;820:153124.

11.

Alhussaini

Aljabri

Al-Harbi

Abdulrahman Almohammadi

Al-Harbi

Bashir

Air pollution and its adverse effects on the central nervous system. Cureus 2023;15:e38927.

12.

Kang

Tan

Lee

Cho

An air particulate pollutant induces neuroinflammation and neurodegeneration in human brain models. Adv Sci (Weinh) 2021;8:e2101251.

13.

You

Chang

RC.

The pathogenic effects of particulate matter on neurodegeneration: a review. J Biomed Sci 2022;29:15.

14.

Tejero

Shiva

Gladwin

MT.

Sources of vascular nitric oxide and reactive oxygen species and their regulation. Physiol Rev 2019;99:311-379.

15.

Maniaci

La Via

Lechien

, et al. Hearing loss and oxidative stress: a comprehensive review. Antioxidants (Basel) 2024;13:842.

16.

Zeggini

Ioannidis

JP.

Meta-analysis in genome-wide association studies. Pharmacogenomics 2009;10:191-201.

17.

Wray

Ripke

Mattheisen

, et al. Genome-wide association analyses identify 44 risk variants and refine the genetic architecture of major depression. Nat Genet 2018;50:668-681.

18.

Torkamani

Wineinger

Topol

EJ.

The personal and clinical utility of polygenic risk scores. Nat Rev Genet 2018;19:581-590.

19.

Vona

The road traveled and journey ahead for the genetics and genomics of tinnitus. Mol Diagn Ther 2022;26:129-136.

20.

Clifford

Maihofer

Stein

Ryan

Nievergelt

CM.

Novel risk loci in tinnitus and causal inference with neuropsychiatric disorders among adults of European ancestry. JAMA Otolaryngol Head Neck Surg 2020;146:1015-1025.

21.

Yuan

Tian

Sun

Greenness, genetic predisposition, and tinnitus. Adv Sci (Weinh) 2024;11:e2306706.

22.

Sudlow

Gallacher

Allen

, et al. UK Biobank: an open access resource for identifying the causes of a wide range of complex diseases of middle and old age. PLoS Med 2015;12:e1001779.

23.

Eeftens

Beelen

de Hoogh

, et al. Development of Land Use Regression models for PM(2.5), PM(2.5) absorbance, PM(10) and PM(coarse) in 20 European study areas; results of the ESCAPE project. Environ Sci Technol 2012;46:11195-11205.

24.

Wang

Zhou

Song

, et al. Joint exposure to various ambient air pollutants and incident heart failure: a prospective analysis in UK Biobank. Eur Heart J 2021;42:1582-1591.

25.

Wang

Song

, et al. Obesity and the relation between joint exposure to ambient air pollutants and incident type 2 diabetes: a cohort study in UK Biobank. PLoS Med 2021;18:e1003767.

26.

Pierzycki

Edmondson-Jones

Dawes

Munro

Moore

Kitterick

PT.

Tinnitus and sleep difficulties after cochlear implantation. Ear Hear 2016;37:e402-e408.

27.

Pierzycki

Kitterick

PT.

Insomnia, anxiety and depression in adult cochlear implant users with tinnitus. Ear Hear 2021;42:235-243.

28.

Bycroft

Freeman

Petkova

, et al. The UK Biobank resource with deep phenotyping and genomic data. Nature 2018;562:203-209.

29.

Wang

Zhou

, et al. Baseline vitamin D status, sleep patterns, and the risk of incident type 2 diabetes in data from the UK Biobank study. Diabetes Care 2020;43:2776-2784.

30.

Brenowitz

Filshtein

Yaffe

, et al. Association of genetic risk for Alzheimer disease and hearing impairment. Neurology 2020;95:e2225-e2234.

31.

Yévenes-Briones

Caballero

Struijk

, et al. Diet quality and the risk of impaired speech reception threshold in noise: the UK Biobank cohort. Ear Hear 2022;43:361-369.

32.

Yuan

Tian

Sun

The risk of hearing impairment from ambient air pollution and the moderating effect of a healthy diet: findings from the United Kingdom Biobank. Front Cell Neurosci 2022;16:856124.

33.

Yuan

Tian

Sun

The association between residential greenness and hearing impairment. Environ Sci Pollut Res Int 2023;30:51113-51124.

34.

Son

Choi

Fong

Heo

Lim

Bell

ML.

The roles of residential greenness in the association between air pollution and health: a systematic review. Environ Res Lett 2021;16:10.1088/1748-9326/ac0e61.

35.

Calderón-Garcidueñas

Kulesza

Mansour

Aiello-Mora

Mukherjee

González-González

LO.

Increased gain in the auditory pathway, Alzheimer’s disease continuum, and air pollution: peripheral and central auditory system dysfunction evolves across pediatric and adult urbanites. J Alzheimers Dis 2019;70:1275-1286.

36.

Chang

Tsai

Lee

, et al. Increased risk of sensorineural hearing loss as a result of exposure to air pollution. Int J Env Res Public Health 2020;17:1969.

37.

Tsai

Hsu

Lai

, et al. Long-term exposure to air pollution and the risk of developing sudden sensorineural hearing loss. J Transl Med 2021;19:424.

38.

Yang

Xie

, et al. Nrf2 deficiency enhances oxidative stress and promotes susceptibility to tinnitus in mice. Sci Rep 2025;15:16474.

39.

Rodríguez-Martínez

Martínez

Espinosa-García

Maldonado

Rivas-Arancibia

Mitochondrial dysfunction in the hippocampus of rats caused by chronic oxidative stress. Neuroscience 2013;252:384-395.

40.

Mainka

Zak

Synergistic or antagonistic health effects of long- and short-term exposure to ambient NO(2) and PM(2.5): a review. Int J Env Res Public Health 2022;19:14079.

41.

Jia

Wei

Zhang

, et al. Pyroptosis participates in PM(2.5)-induced air-blood barrier dysfunction. Environ Sci Pollut Res Int 2022;29:60987-60997.

42.

Kuang

Zhang

Ren

Chen

Different components of air pollutants and neurological disorders. Front Public Health 2022;10:959921.

43.

De Ridder

Schlee

Vanneste

, et al. Tinnitus and tinnitus disorder: theoretical and operational definitions (an international multidisciplinary proposal). Prog Brain Res 2021;260:1-25.

44.

Henry

Roberts

Caspary

Theodoroff

Salvi

RJ.

Underlying mechanisms of tinnitus: review and clinical implications. J Am Acad Audiol 2014;25:5-22; quiz 126.

45.

Vijayakumar

Cho

Maharajan

Jang

CH.

A review on peripheral tinnitus, causes, and treatments from the perspective of autophagy. Exp Neurobiol 2022;31:232-242.

46.

Guo

, et al. Effects of environmental noise exposure on DNA methylation in the brain and metabolic health. Environ Res 2017;153:73-82.

47.

Patil

Perez-Carpena

Lopez-Escamez

JA.

A systematic review on the contribution of DNA methylation to hearing loss. Clin Epigenetics 2024;16:88.

48.

Rider

Carlsten

Air pollution and DNA methylation: effects of exposure in humans. Clin Epigenetics 2019;11:131.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

6.13 MB