Increased Risk of Sensorineural Hearing Loss as a Result of Exposure to Air Pollution

Whether exposure to air pollution is associated with developing sensorineural hearing loss (SHL) remains controversial. Using data from the National Health Insurance Research Database, we recruited a total of 75,767 subjects aged older than 20 years with no history of SHL from 1998 to 2010, and they were followed up until SHL was observed, they withdrew from the National Health Insurance program, or the study ended. The subjects were evenly exposed to low-level, mid-level, and high-level carbon monoxide (CO) and nitrogen dioxide (NO2). The incidence rate ratio of SHL for patients exposed to high-level CO was 1.24 (95% confidence interval (CI) = 1.14–1.36). The NO2 pollutants increased the incidence rate ratios of SHL in mid-level NO2 and high-level NO2 exposures by 1.10 (95% CI = 1.10–1.32) and 1.36 (95% CI = 1.24–1.49) times, respectively. The adjusted hazard ratio (adj. HR) of SHL in patients exposed to high-level CO was 1.45 (95% CI = 1.31–1.59), relative to that of patients exposed to low-level CO. Compared to patients exposed to low-level NO2, patients exposed to mid-level NO2 (adj. HR = 1.40, 95% CI = 1.27–1.54) and high-level NO2 (adj. HR = 1.63, 95% CI = 1.48–1.81) had a higher risk of developing SHL. The increased risk of SHL following the increased concentrations of air pollutants (CO and NO2) was statistically significant in this study. In conclusion, the subjects’ exposure to air pollution exhibited a significantly higher risk of developing SHL in Taiwan.

Health Insurance Database which is the health database of a million beneficiaries randomly selected from the NHIRD. The identification information was encrypted to protect the patients' personal privacy. The disease codes were identified according to the International Classification of Disease, Ninth Revision, Clinical Modification (ICD-9-CM). We recruited subjects aged older than 20 years with no history of SHL from 1998 to 2010, and they were followed up until SHL was observed, they withdrew from the NHI program, and the study ended (December 2011).

Exposure Measurement
The data regarding the air pollutants were collected from 74 ambient air quality monitoring stations across Taiwan. The air quality data are managed by Taiwan Environmental Protection Administration. The annual concentrations of NO 2 and CO from 1998 to 2010 were grouped into the following three levels according to tertiles: the low-, mid-, and high-level groups. The NO 2 concentrations of the low-, mid-, and high-level groups were <19.53 (ppb), 19.53-25.66 (ppb), and >25.66 (ppb), respectively. Moreover, the CO concentrations of the low-, mid-, and high-level groups were <0.61 (ppm), 0.61-0.76 (ppm), and >0.76 (ppm), respectively. To analyze the exposures across a 13 year period (from 1998 to 2010), we calculated the annual average of pollutants from baseline to the date of hearing loss occurrence.

Statistical Analysis
We show the demographic distribution as a percentage. The incidence density rates of hearing loss were calculated, and the incidence rate ratio (IRR) was assessed by Poisson regression. We estimated the hazard ratios using the multivariate Cox proportional hazard model, adjusting for age, sex, insurance fee, urbanization, and comorbidities.

Results
We recruited 75,767 subjects in the present study. Table 1 shows the distribution of the demographic characteristics of the study sample. The subjects' mean age was 54.1 (±11.1) years with a mean age of 11.1 (±2.2) years the following years. Moreover, 45.3% of the subjects were men. The subjects were evenly exposed to low-, mid-, and high-level CO and NO 2 . Most of the subjects lived in moderately urbanized areas (33.1%) and highly urbanized areas (32.5%). Table 2 shows that the top three comorbidities with high incidence rates were HT (57.5%), DM (20.6%), stroke (9.1%), head injury (9.8%), CKD (5.3%), IHD (34.6%), alcoholism (1.4%), nicotine dependence (2.2%), asthma (17.5%), COPD (34.5%), and RA (0.5%).  The IRRs of SHL in different air pollutant exposure levels are presented in Table 3. Subjects exposed to low-level air pollutants were assigned to the reference group. The IRR of SHL in subjects exposed to high-level CO was 1.24 (95% confidence interval (CI) = 1.14-1.36). The NO 2 pollutants increased the incidence of SHL in subjects exposed to mid-level NO 2 and high-level NO 2 by 1.10 (95% CI = 1.10-1.32) and 1.36 (95% CI = 1.24-1.49) times, respectively. From our analyses, the CO levels in highly urbanized region, moderately urbanized region, boomtown, and others were 0.82 ± 0.31, 0.69 ± 0.24, 0.71 ± 0.22, and 0.58 ± 0.21, respectively (Table 3). Furthermore, the NO 2 levels in highly urbanized region, moderately urbanized region, boomtown, and others were 24.69 ± 6.65, 21.83 ± 6.45, 23.54 ± 5.53, 17.87 ± 6.37, respectively (Table 3). Our results suggest that the distribution of CO and NO 2 levels are highly correlated with the urbanization. According to Table 4, the adjusted hazard ratio (adj. HR) of SHL in subjects exposed to high-level CO was 1.45 (95% CI = 1.31-1.59), relative to that of subjects exposed to low-level CO. Compared to subjects exposed to low-level NO 2 , subjects exposed to mid-level NO 2 (adj. HR = 1.40, 95% CI = 1.27-1.54) and high-level NO 2 (adj. HR = 1.63, 95% CI = 1.48-1.81) had a higher risk of developing SHL. Moreover, the increased risk of SHL following the increased concentrations of air pollutants (CO and NO 2 ) was statistically significant in this study. In the present study, we found that the incidence rate ratio of SHL in the highest level of CO (IRR = 1.24, 95%CI = 1.14-1.36) was significantly higher than that in low level of CO (Table 4). Furthermore, we found that the incidence rate ratio of SHL in the mid-level (IRR = 1.21, 95% CI = 1.10-1.32) and highest level of NO 2 (IRR = 1.36, 95% CI = 1.24-1.49) was significantly higher than that in low level of NO 2 (Table 4). According to Table 5, the aHR of hearing loss in the high CO concentration group relative to the low CO concentration group was 1.45 (95% CI, 1.31-1.59). Patients in the mid-NO 2 level (aHR = 1.40, 95% CI = 1.27-1.54) and high NO 2 level (aHR = 1.63, 95% CI = 1.48-1.81) groups had a higher risk of hearing loss compared to patients in the low NO 2 level group. The increased concentrations of the air pollutants (i.e., CO and NO2) enhancing the risk of hearing loss was statistically significant. Moreover, continuous analyses based on the subjects' sex and age (subjects aged 50 years) to assess the association between air pollution and SHL were performed. Our data suggest that both male (aHR = 1.25, 95% CI = 1.03-1.52) and female (aHR = 1.49, 95% CI = 1.20-1.85) subjects have a significantly high incidence of SHL when they are exposed to CO pollution. Moreover, female subjects had a significantly high incidence of SHL when exposed to NO2 (aHR = 1.01, 95% CI = 1.001-1.02). Furthermore, elderly subjects aged greater than 50 years had a significantly higher risk of SHL (CO: aHR = 1.37, 95% CI = 1.17-1.62) when they were exposed to CO pollution. Additionally, we provided the probability-free prediction of SHL among the three levels of CO and NO2 concentrations (Figure 1).

Discussion
Although air pollution has been considered a risk factor for several diseases, its association with hearing disorders has been less studied. In this study, we hypothesized that air pollution may also be associated with SHL development. We focused on investigating the two major traffic-related air pollutants: nitrogen dioxide (NO2) and carbon monoxide (CO). We analyzed the data from a

Discussion
Although air pollution has been considered a risk factor for several diseases, its association with hearing disorders has been less studied. In this study, we hypothesized that air pollution may also be associated with SHL development. We focused on investigating the two major traffic-related air pollutants: nitrogen dioxide (NO2) and carbon monoxide (CO). We analyzed the data from a longitudinal cohort study using Taiwan National Health Insurance (NHI) data to assess the significant effects of these traffic-related air pollutants to SHL. Our results revealed that the subjects who were chronically exposed to air pollution were at a significantly increased risk of developing SHL. Based on the longitudinal human data, our results suggest that NO 2 and CO could have potential adverse effects on SHL. Furthermore, we showed that long-term exposure to the highest quartile of NO 2 significantly increased the risk of SHL by almost 1.63 fold, even after adjusting the confounding factors. Additionally, exposure to the highest quartile of CO also increased the risk of SHL by 1.45 fold. Nitrogen dioxide and CO result in cellular damages through different mechanisms. Nitrogen dioxide is a reactive nitrogen species (RNS). Reactive nitrogen species and reactive oxygen species cause cell apoptosis. On the contrary, CO is considered toxic, as it prevents the blood system from effectively carrying oxygen around the body, specifically to the vital organs such as the heart and brain. However, any report showing the association between CO and SHL does not exist yet. The present study suggests that exposure to CO and NO 2 has several adverse effects on the auditory system.
Several studies investigating the clinical effects of air pollution on different organ systems in the human body have been conducted [4,[7][8][9]11,53]. Using similar datasets, we have previously shown that CO and NO 2 were associated with AMD [11], dementia [53], and PD [9]. Compared with the lowest exposure group, the adj. HRs for AMD and dementia were 1.91 and 1.5 and 1.84 and 1.61 for the highest quartile of NO 2 exposure and the highest quartile of CO exposure, respectively. Based on the adj. HRs, our data indicated that these two pollutants have similar adverse effects on SHL, AMD, and dementia. On the contrary, either NO 2 or CO is not associated with the incidence of PD. However, particulate matter 10 (PM 10 ) exposure can have significant adverse effects on PD development. Similar to AMD and dementia, SHL is also an age-related disease that primarily affects the cochlear hair cells and spiral ganglion neurons.
The onset of SHL has been associated with several risk factors including inflammation and oxidative stress. Our data further suggest that RNS might also contribute to SHL progression. Nitrogen dioxide is one of the most important components of traffic-related air pollution. It can be hydrolyzed to nitrous and nitric acid, both of which can induce inflammation by oxidative stress and lipid peroxidation. It has been suggested that gentamicin application intratympanally can result in a hearing threshold shift beginning on the second day after gentamicin application. This hearing impairment is simultaneously associated with an increased NO 2 − (a stable oxidation product of NO 2 ) concentration in the lateral wall, indicating that NO 2 − contributes to gentamicin-induced hearing impairment [54].
Although the primary route of exposure to NO 2 is via inhalation, studies assessed not only the association between the respiratory system and NO 2 exposure but also the association between the cardiovascular [55,56] and neurological systems and NO 2 exposure [57]. Previously, air pollution has been reported to be a risk factor for sudden SHL [52]. The major differences between our results and the reports of Lee et al. [52] are as follows: (1) our study was the first study to use a large longitudinal cohort from 1998 to 2010 to demonstrate the significant association between SHL and high levels of ambient NO 2 and CO. (2) According to the report of Lee et al. [52], they utilized the data from the daily patient numbers admitted to the hospital with sudden SHL in 2015 and demonstrated that in Busan, Korea, there is a statistically significant weak association between the daily numbers of patients admitted to the hospital with sudden SHL and the exposure of mean daily PM 10 and PM 2.5 concentrations [52].
It has been suggested that the risk of SHL is also significantly associated with factory workers who are co-exposed to noise and heavy metals or organic solvents [58]. Co-exposure to noise and heavy metal or toxic pollution results in a worse SHL compared with isolated exposures [59]. In the present study, both CO and NO 2 levels are relatively higher in highly urbanized region and boomtown than in other areas (Table 3). Notably, both noise and air pollution may damage the cochlear HCs via oxidative stress [60], hypoxia-induced reactive oxygen species, and HC damage [48]. Although we found a statistically significant association between air pollution and subsequent SHL, the present study had the following limitations that need to be considered: (1) The noise levels in the studied area and the genetic variants in the studied subjects, which are considered as important risk factors for SHL, were not clarified in the present study, and without the noise-level data, the scope and significance of the study are limited. (2) The study conducted in the assigned residential area was according to the subject's condition which was associated with the treatment of acute upper respiratory infections in a clinic or hospital, resulting in the underestimation of every subject's exposure level to each air pollutant in the studied area. However, our findings are still significant and require further investigation. (3) In the present study, we used the ICD-9-CM codes 389.10-389.12 and 388.01 for data acquisition of age-related hearing loss and SHL. However, we did not include the ICD-9-CM codes of sudden hearing loss (388.02) and conductive hearing loss (389.00 and 389.06) considering that only 3.6% (2724 SHL subjects/75,767 subjects) of the subjects were identified as SHL patients in this study (Table 1).

Conclusions
In conclusion, we linked the national health database and air quality database to report NO 2 and CO as risk factors for SHL. Our results indicate that the highest quartile of each pollutant could increase the risk of SHL by almost 1.63 and 1.45 fold, respectively. Notably, a mid-level exposure to NO 2 , but not CO, could also significantly contribute to SHL development by 1.40 fold.

Conflicts of Interest:
The authors declare no conflict of interest Data Availability: The authors confirm that all data underlying the findings are fully available without restriction. The dataset is owned by the Taiwan National Health Research Institutes (NHRI). Requests for the dataset may be sent via an e-mail to the NHRI at nhird@nhri.org.tw or call at +886-037-246166 ext. 33603 for immediate service.