Abstract
Background
Emerging evidence in animal models and humans suggests that statins may provide partial protection against noise-induced and drug-induced hearing loss. However, evidence for protection against age-related hearing loss (ARHL) is mixed, and longitudinal studies in large adult cohorts from the general population are limited. This study examined the association between statin use and ARHL in a diverse, community-based longitudinal cohort of adults across the lifespan.
Methods
The 1239 participants, aged ≥ 55 years, are from the MUSC Longitudinal Cohort Study of Age-Related Hearing Loss (1988–present). Following a comprehensive baseline examination, participants attended annual examinations during which audiometric and statin use data were collected. Multivariable linear regression models were performed on baseline data for cross-sectional analyses, and general linear mixed models were performed on longitudinal data to assess the association between statin use, two types of statins, and hearing over time. Hearing outcomes include four pure-tone averages (PTA): narrow (0.5, 1.0, 2.0, and 4.0 kHz), low frequency (0.25, 0.5, and 1.0 kHz), broad (0.5, 1.0, 2.0, 3.0, 4.0, 6.0, and 8.0 kHz), and high frequency (2.0, 3.0, 4.0, 6.0, and 8.0 kHz). Missing covariate data were imputed with multiple imputation using chain equations (MICE).
Results
Statin use (versus no use) was associated with better hearing at baseline for the broad and high-frequency PTAs for adults aged 55–64 and 65–74 years. However, this statin benefit was not observed for the 75+-year age group and for all participants when hearing was examined over time. Without age stratification, similar results were observed for atorvastatin use but not for simvastatin use.
Conclusions
Overall, the findings suggest statin benefits for mid-to-high frequency hearing for middle-aged to older adults at early stages of ARHL. On the basis of evidence suggesting that some statins have beneficial effects on hearing, well-controlled clinical trials are needed to examine effects of statin use and statin types on ARHL in the general population.
Supplementary Information
The online version contains supplementary material available at 10.1007/s40266-025-01268-4.
Key Points
| Question: What is the association between statin use and hearing loss in aging? |
| Findings: This study included 1239 participants from the Medical University of South Carolina Longitudinal Cohort Study of Age-Related Hearing Loss. At baseline, participants aged 55–64 and 65–74 years (but not 75 years and older) who used statins had better hearing in mid-to-high frequencies as compared with participants who did not use statins. Atorvastatin use contributed to this association, while simvastatin use did not. However, in longitudinal analysis, a statin benefit was not observed for any participant group when hearing was examined during the participants’ time in the study, suggesting that statin use does not slow the normally observed declines in hearing as individuals age. |
| Meaning: Well-controlled clinical trials are needed to examine the effects of statin use and statin types on hearing in the general population. |
Introduction
By 2050, at least 700 million people worldwide will need rehabilitation for some degree of hearing loss [1]. Age-related hearing loss (ARHL), for which prevalence increases with age [2, 3], is associated with poorer communication and health-related quality of life and several comorbid health and psychosocial conditions [4–6]. However, despite public health approaches for improving prevention and increasing treatment of hearing loss among older adults, hearing loss remains one of the most common chronic conditions of aging, and uptake of hearing technologies is low [7]. Therefore, other methods for the prevention, protection, and treatment of ARHL are being studied, including the extent to which commonly prescribed medications can reduce or delay the onset of ARHL. These investigations include statins, which are widely prescribed for the treatment of hypercholesterolemia to prevent and/or slow cardiovascular disease (CVD), with statin benefits outweighing the risk of side effects [8, 9]. One rationale for an indirect benefit of statin use for ARHL is that controlling cholesterol may reduce damage to cochlear vasculature, which could preserve cochlear blood supply and thus slow the progression of hearing loss in the aging ear and/or reduce susceptibility to effects of noise. Additional direct effects of statins on auditory function could occur that are independent of their cholesterol-lowering effects, for example, statins have antioxidant properties that could reduce oxidative stress and inflammatory effects in the inner ear.
Research in animal models and humans on effects of statins on auditory function provides support for examining effects of statins on ARHL [10]. In preclinical studies, specific statins have shown protective effects against hearing loss associated with ototoxic drugs such as gentamicin [11], cisplatin [12], and kanamycin [13]. In patients receiving cisplatin therapy for head and neck cancer, concomitant use of atorvastatin resulted in reduced (versus no statin use) cisplatin-induced hearing loss with no adverse effect on survival rates [14]. Similar protective effects against noise-induced hearing loss (NIHL) have been reported in animal models [15, 16], and statins have been suggested as a possible treatment for sudden sensorineural hearing loss [17].
The evidence in animal models for an effect of statins on slowing or preventing ARHL is mixed [18, 19], and effects in humans may vary by age, sex, duration of use, and statin type. In one epidemiological study based in Australia, adults aged ≥ 50 years followed for over 5 years showed that high dietary intake of cholesterol increased odds of hearing loss, whereas self-reported statin use decreased odds of hearing loss both cross-sectionally and longitudinally [20]. In cross-sectional analyses of audiometry and self-reported medication use obtained from two large population-based data sets in Korea, drugs to treat hyperlipidemia had a protective effect on hearing in older (≥ 60 years) but not middle-aged adults [21]. In other population-based cross-sectional analyses, positive effects of statins on hearing loss were reported only in male individuals ≥ 70 years with longer duration of use [22] and in male individuals with hyperlipidemia using hydrophilic statins (pravastatin and rosuvastatin) [23]. Another cross-sectional study reported lower odds of poorer hearing for statin users (versus non-statin users) among patients with Meniere’s Disease [24]. One limitation of these studies is that they relied on a single pure-tone average (PTA) of hearing thresholds at selected audiometric frequencies or on hearing loss categories to define hearing loss prevalence or incidence. Given that aging effects on hearing vary by frequency, analyses that are limited to one PTA cannot be used to demonstrate the effects of statins on thresholds across the frequency range. Other limitations of some previous reports are restricted samples and cross-sectional studies conducted in racially homogenous individuals consisting primarily of white or Asian individuals [20, 21]. Thus, there is inconsistent evidence regarding the effects of statin use on ARHL in humans and a paucity of high-quality longitudinal studies with adequate hearing data to evaluate these effects.
The purpose of this study, conducted in a diverse, community-based longitudinal cohort of adults across the lifespan, was to determine the association between statin use and ARHL. The study used cross-sectional data to examine differences in hearing at baseline between statin and non-statin users, and between users of two types of statins (atorvastatin and simvastatin) and non-statin users. Using longitudinal data, we also report changes in hearing over time for statin and non-statin users, and for users of two types of statins and non-statin users.
Methods
Participants
The Medical University of South Carolina (MUSC) Longitudinal Cohort Study of Age-Related Hearing Loss is an ongoing (1988–present) community-based cohort study based in Charleston, SC, USA [25–28]. Participants are continuously enrolled and are recruited through community events, online and print advertisements, and participant referral. Recruitment, consenting, and test protocols were approved by the Institutional Review Board for Human Research at MUSC. Inclusion criteria were the following: ≥ 18 years of age and general health sufficient to attend multiple laboratory visits. Exclusion criteria were inability to provide informed consent, evidence of conductive hearing loss (air–bone gaps ≥ 15 dB at frequencies 0.5–4.0 kHz and/or abnormal tympanograms), active otologic/neurologic disease, or poor test reliability. Two to six visits were scheduled to complete a baseline protocol that included, but was not limited to, pure-tone air-conduction thresholds (0.25, 0.5, 1.0, 2.0, 3.0, 4.0, 6.0, and 8.0 kHz) and bone-conduction thresholds (0.25, 0.5, 1.0, 2.0, 3.0, and 4.0 kHz) at conventional frequencies; tympanograms; and responses to the following self-report questionnaires: demographics (age, sex at birth, ethnicity/race, occupation, and native language), medical history (general health conditions, smoking status, prescription and over-the-counter medications [dose information not obtained]), and noise exposure history. Following completion of the initial test battery, participants returned annually to obtain an updated audiogram, health history, and hearing health history. Positive noise history at baseline was defined as having answered yes to occupational noise exposure and/or firearm use ≥ 1 year. Participants were classified as having cardiovascular disease (CVD) risk if they answered yes to a history of one or more medical conditions: chest pain or angina, diabetes, heart disease, hypertension, or stroke. Participants self-reported being a never, current, or past smoker.
At the time of analyses, 1704 participants were enrolled in the study, of which 1239 participants were ≥ 55 years of age and had baseline data, including pure-tone audiometry and statin use status. To maximize the number of audiograms in the longitudinal analyses, drug records of statin use or nonuse were matched to all audiograms obtained within 6 months of the self-reported drug use date using a Greedy algorithm. That is, a single drug record reporting statin use or nonuse may have been matched to more than one audiogram, if the audiogram(s) occurred within 6 months of when the drug record was collected. Table 1 presents the demographic, health, and hearing information for the baseline cohort by statin use, which was used for the cross-sectional analyses. A similar table reporting characteristics of the longitudinal cohort is presented as Online Resource 1.
Table 1.
Cross-sectional cohort: demographics and health characteristics, and audiometric characteristics for the non-statin and statin users
| Demographic characteristics | Non-statin N = 906 |
Statin N = 333 |
p-value |
|---|---|---|---|
| Age (years) | |||
| Mean ± SD | 68.7 ± 7.4 | 70.4 ± 7.2 | 0.0003* |
| N (%) | N (%) | ||
| Sex | |||
| Female | 530 (58.5) | 190 (57.1) | 0.6484 |
| Male | 376 (41.5) | 143 (42.9) | |
| Racea | |||
| Racial minority | 131 (14.5) | 71 (21.3) | 0.0036** |
| White | 775 (85.5) | 261 (78.4) | |
| Unknown | – | 1 (0.3) | |
| Positive noise historyb | |||
| Yes | 385 (42.5) | 146 (43.8) | 0.0343** |
| No | 426 (47.0) | 168 (50.5) | |
| Unknown | 95 (10.5) | 19 (5.7) | |
| Smoking statusc | |||
| Current | 55 (6.1) | 23 (6.9) | < 0.0001** |
| Never | 248 (27.4) | 123 (36.9) | |
| Past | 289 (31.9) | 152 (45.6) | |
| Unknown | 314 (34.7) | 35 (10.5) | |
| Cardiovascular disease riskd | |||
| Yes | 481 (53.1) | 272 (81.7) | < 0.0001** |
| No | 396 (43.7) | 61 (18.3) | |
| Unknown | 29 (3.2) | – | |
| Audiometric characteristics (dB HL) | Mean ± SD | Mean ± SD | p-value |
|---|---|---|---|
| Narrow PTA (0.5, 1.0, 2.0, 4.0 kHz) | 29.7 ± 16.0 | 29.6 ± 15.0 | 0.8917 |
| Broad PTA (0.5, 1.0, 2.0, 3.0, 4.0, 6.0, 8.0 kHz) | 36.9 ± 17.2 | 36.4 ± 16.5 | 0.7988 |
| Low-frequency PTA (0.25, 0.5, 1.0 kHz) | 18.0 ± 12.0 | 18.8 ± 11.2 | 0.1444 |
| High-frequency PTA (2.0, 3.0, 4.0, 6.0, 8.0 kHz) | 44.3 ± 20.6 | 43.1 ± 19.7 | 0.4784 |
PTA results unadjusted for age, sex, race, smoking status, cardiovascular disease (CVD) risk, and noise exposure. Age, sex, race, smoking status, CVD risk factors, and noise history were based on self-report. Total N = 1239
dB HL decibels hearing level, kHz kilohertz, PTA pure-tone average, SD standard deviation
Wilcoxon–Mann Whitney test; age and PTAs, *p ≤ 0.05; chi-squared test; sex, race, noise exposure, smoking status, and CVD risk, **p ≤ 0.05
aRacial minorities: ~89% Black or African American, 7.92% two or more races, 2.48% Asian, and 0.5% American Indian or Alaska Native
bPositive noise history: self-reported yes to occupational noise exposure and/or firearm use > 1 year
cPositive smoking history: self-reported as being a current or past smoker
dPositive CVD risk: self-reported yes to a history of one or more of the following: chest pain or angina, diabetes, heart disease, hypertension, or stroke
Cross-Sectional and Longitudinal Analyses of Statin Use
For cross-sectional analyses, the full baseline cohort of 1239 participants was classified into two statin-use groups: non-statin users (N = 906), defined as those not taking any statin at baseline, and statin users (N = 333), defined as taking any statin at baseline. Those taking only non-statin antihyperlipidemic drug(s) were included in the non-statin user group and those taking any type of statin drug, either solely or paired with another statin, with a non-statin antihyperlipidemic drug, or with a non-statin antihypertensive/antianginal drug were included in the statin user group. Of the 333 participants reporting statin use at baseline, 128 were atorvastatin users, and 115 were simvastatin users, with the remainder reporting another statin type.
For longitudinal analyses, participants were required to have had at least two visits and no more than one change from statin user to non-statin user (or vice versa) throughout their time in the study; a change only to another statin was not considered a change in statin use. Of the 1239 participants from the cross-sectional cohort, 189 participants had a single visit (baseline), and 78 participants had multiple changes in statin use, leaving 972 participants who met the criteria for longitudinal analyses. Data for each participant included in the longitudinal cohort were evaluated for statin use at each measured timepoint in the study period. Of the 972 participants, 570 participants did not take a statin at any measured timepoint in the study period (“never-statin users”), 234 participants were taking a statin at all measured timepoints in the study period (“always-statin users”), and 168 participants either started as statin users at baseline and changed to non-statin users at some point in the study or vice versa. Among the longitudinal cohort, 137 participants reported atorvastatin use, and 112 participants reported simvastatin use at some point during the study.
In the longitudinal analyses, the number of visits and time between visits varied by participant. There were a total of 7105 visits across all participants in the longitudinal analyses. Throughout the study period, the mean number of visits for the 972 participants in the longitudinal cohort was 7.3 ± 5.3 (range 2–30 visits). The mean time from baseline to final visit was 2.5 ± 2.9 (range < 1–26.5 years).
Statistical Analyses
Outcome Measures
Pure-tone air-conduction thresholds from left and right ears of each participant were averaged across ears at each frequency following descriptive analyses that showed similar mean thresholds in both ears at all frequencies [29]. Missing (not measured) thresholds for 3.0 and 6.0 kHz were minimal and, therefore, thresholds for adjacent frequencies (2.0 and 4.0 kHz; 4.0 and 8.0 kHz, respectively) were averaged for each ear. Four PTAs were calculated from averaged right and left ear thresholds. These outcomes included: narrow (0.5, 1.0, 2.0, and 4.0 kHz), low frequency (0.25, 0.5, and 1.0 kHz), broad (0.5, 1.0, 2.0, 3.0, 4.0, 6.0, and 8.0 kHz), and high frequency (2.0, 3.0, 4.0, 6.0, and 8.0 kHz).
Additional Covariates
Statin use was the primary independent variable. Baseline age, sex, race, smoking status, CVD risk, and noise history were controlled for as covariates and were selected as factors known to affect hearing in older adults [4, 30]. These covariates were measured at baseline for the cross-sectional analysis. For the longitudinal analyses, covariates that may have changed from the start of the study were measured as of the most recent visit.
Statistical Approach
Analyses were completed with SAS 9.4 statistical software (Cary, NC, USA). Statistical significance was determined at the 0.05 level, and all hypothesis tests were two-sided. The proportion of complete cases with no missing baseline data was 67.4% (n = 835). Two preliminary analyses (not reported) were performed: (1) cases with complete data in a model adjusted for baseline age, sex, race, smoking status, CVD risk, and noise history found no statistically significant associations between statin use and PTAs, and (2) a simpler model adjusting only for demographic characteristics with the most complete data (baseline age, sex, and race) found statin users (versus nonusers) had significantly better narrow, broad, and high-frequency PTAs. Both results highlight the impact missing data had on our analyses. Therefore, to retain the full sample, missing baseline covariate (age, sex, race, smoking status, CVD risk, and noise history) data were imputed using the multiple imputation using chain equations (MICE) procedure [31]. Additional details about this approach may be found in Online Resource 6.
For cross-sectional analysis of the association between statin use and PTA outcomes, multivariable linear regression models were performed on baseline data (N = 1239). General linear models were fit using PROC GENMOD while adjusting for covariates at baseline. To determine whether the statin association differed by age group, models were repeated by stratifying age into three groups as follows: 55–64 years (statin users, 78; non-statin users, 308), 65–74 years (statin users, 167; non-statin users, 411), 75+ years (statin users, 88; non-statin users, 187). Age ranges were selected on the basis of the age distribution of the cross-sectional cohort and divided into similar sample sizes. To rule out potential effects of selection bias in participants who did not return for follow-up visits, a sensitivity analysis was conducted using baseline data from the longitudinal cohort (n = 972) and compared with results from the cross-sectional analyses (N = 1239) (reported in Online Resource 5).
To assess the association between statin use and PTAs over time, general linear mixed models (GLMMs) were performed on the longitudinal data using PROC MIXED with an unstructured covariance matrix. Each model specified a random intercept and random time effect, where the time variable was defined as years since baseline age, to account for length of participation in the study, and a unique participant identifier was specified as the subject effect. GLMMs were used owing to the unbalanced structure of the repeated measures as timing between participant visits is irregular and number of visits varies by participant. Furthermore, GLMMs are suitable for repeated measures within participants and allow for estimation of both fixed (population-average) and random (subject-specific) effects. The unstructured covariance matrix was selected as appropriate because it is the most flexible, such that no two pairs of observations are assumed to be equally correlated, and allows for correlation between random effects. In addition, the unstructured covariance provided the best fit when comparing across the Akaike information criterion (AIC) and Bayesian information criterion (BIC) values using other covariance structures for the same models. Baseline age was included in the longitudinal analysis as an independent risk factor. CVD risk, smoking status, and noise history were used as covariates in the model and measured as of the most recent visit for the longitudinal analyses owing to participants starting and stopping at different times. These characteristics were not time-varying covariates, and the most recent visit information was used to account for a participant’s full history. To examine potential cumulative effects of long-term statin use and to compare with the cross-sectional results, a subset analysis was performed on the longitudinal data comparing the 234 always-statin users to the 570 never-statin users.
Associations between use of specific statin types and PTA outcomes included the two most used statins at baseline, atorvastatin (n = 128, 38.4%) and simvastatin (n = 115, 34.5%). Similar multivariable linear regression models were performed for cross-sectional analysis of atorvastatin or simvastatin users at baseline compared with non-statin users at baseline. GLMMs were used for longitudinal analysis of statin type. Atorvastatin and simvastatin use, reported at any time throughout the study period, was assessed in separate models. Both atorvastatin (n = 137) and simvastatin (n = 112) users were compared with never-statin users (n = 570).
For all PTAs, cross-sectional analyses (model 1) show the overall mean difference in PTAs for statin users as compared with non-statin users after controlling for baseline covariates (age, sex, race, smoking status, CVD risk, and noise history). Two models are reported for all longitudinal analyses. In model 2a, the statin association is reported using all repeated measures for all participants, answering the question, “Is there an overall difference in PTA outcomes between statin users and non-statin users when using all available data?” Model 2b extends this to additionally examine whether the rate of change in PTA outcomes differs by statin use by adding an interaction term between statin use and years since baseline age. This answers the question, “Is statin use associated with slower declines in hearing over time as individuals age?” The interaction term tests whether threshold differences by statin use change over time, defined as years since baseline age, which is reported as a rate of change. Both models 2a and 2b adjust for baseline age, years since baseline age (time), sex, race, most recent smoking status, CVD risk, and noise history.
Results
Baseline Characteristics
Demographic, health, and audiometric characteristics by statin use among the cross-sectional cohort of 1239 participants are presented in Table 1. At baseline, statin users were slightly older (mean age of 70.4 versus 68.7 years, p = 0.0003), had a higher percentage of racial minority participants (21.3 versus 14.5%, p = 0.0036), and a similar percentage of female participants (57.1 versus 58.5%, p = 0.6484) as compared with non-statin users. A significantly greater percentage of statin users reported having at least one CVD risk factor compared with non-statin users (81.7 versus 53.1%, p < 0.0001). Noise history (p = 0.0343) and smoking history (p < 0.0001) distributions were both statistically significantly different between statin and non-statin users, with statin users having slightly higher positive noise history (43.7 versus 42.5%) and higher rates of current (6.9 versus 6.1%) and past (45.6 versus 31.9%) smokers. Unadjusted PTAs at baseline were found to be similar between statin users and non-statin users for all PTA outcomes. A similar Table 1, limited to participants meeting the criteria for longitudinal analysis, is shown in Online Resource 1.
Online Resource 2 shows participant demographic characteristics and health information at baseline among atorvastatin, simvastatin, and non-statin users. Similar statistically significant results were found in the overall cohort at baseline, except for noise history, where no differences were found by statin type (p = 0.5596). Mean age at baseline for simvastatin users was 70.9 years, atorvastatin users 70.2 years, and non-statin users 68.7 years (p = 0.0014). Atorvastatin users had the highest percentage of racial minority participants (25.8%), followed by simvastatin (20.0%) and non-statin users (14.5%). Similar proportions of atorvastatin, simvastatin, and non-statin users were female (p = 0.1225). As expected, 85.2% of atorvastatin users and 82.6% of simvastatin users reported having at least one CVD risk factor compared with only 53.1% of non-statin users (p < 0.0001). Smoking history (p < 0.0001) distribution was statistically significantly different, with atorvastatin users having the highest percentage of current (7.8%) and never (45.3%) smokers and simvastatin users having the highest percentage of past smokers (51.3%). Similar to the full cohort at baseline, no unadjusted statistically significant differences by statin type were found for any PTA outcomes.
Cross-Sectional Analyses
Results from the cross-sectional analyses (model 1) indicated that statin users had significantly better hearing thresholds, as defined by narrow, broad, and high-frequency PTAs, after controlling for baseline age, sex, race, smoking status, CVD risk, and noise history (Fig. 1A; Table 2). Statin users had, on average, better thresholds by 2.00 dB (p = 0.0223) for narrow PTA and 2.54 dB (p = 0.0054) for broad PTA. The difference in hearing sensitivity was greatest for the high-frequency PTA, with an average 3.36 dB better threshold (p = 0.0017) for the statin than non-statin users. There was no significant relationship between statin use and low-frequency PTA.
Fig. 1.
Mean (+1 standard deviation) pure-tone averages (PTAs) at baseline for statin users (N = 333) and non-statin users (N = 906) (A); for atorvastatin users (N = 128) and non-statin users (B); and for simvastatin users (N = 115) and non-statin users (C). In Table 1, see the “Audiometric characteristics (dB HL)” section for definitions of PTAs. Thresholds were adjusted for baseline age, sex, race, smoking status, cardiovascular risk, and noise history (*p ≤ 0.05)
Table 2.
Cross-sectional and longitudinal multivariable models of associations between PTA and statin and non-statin users
| Variable | Cross-sectional (N = 1239) | Longitudinal (N = 972) | |||||
|---|---|---|---|---|---|---|---|
| Model 1: mean statin difference | Model 2a: mean statin difference | Model 2b: mean statin difference with interaction term | |||||
| Dependent PTAs (dB HL) | Independent | Coefficient | p-value* | Coefficient | p-value** | Coefficient | p-value** |
| Narrow | Statin versus non-statin | −2.00 | 0.0223* | −0.01 | 0.9505 | −0.33 | 0.0994 |
| Statin × years since baseline age (RC) | 0.07 | 0.0051** | |||||
| Broad | Statin versus non-statin | −2.54 | 0.0054* | 0.01 | 0.9483 | −0.23 | 0.2488 |
| Statin × years since baseline age (RC) | 0.05 | 0.0365** | |||||
| Low-frequency | Statin versus non-statin | −0.72 | 0.3086 | 0.03 | 0.8683 | −0.28 | 0.2112 |
| Statin × years since baseline age (RC) | 0.07 | 0.0128** | |||||
| High-frequency | Statin versus non-statin | −3.36 | 0.0017* | −0.01 | 0.9408 | −0.16 | 0.4926 |
| Statin × years since baseline age (RC) | 0.03 | 0.2809 | |||||
Model 1: adjusted for age, sex, race, smoking status, cardiovascular disease (CVD) risk, and noise exposure. Total N = 1239 participants with baseline data. Models 2a and 2b: adjusted for years since baseline age, sex, race, smoking status, CVD risk, and noise exposure that were within 6 months of last audiogram or last record. Total N = 972 (234 always-statin; 570 never-statin; 144 changed from non-statin to statin; 24 changed from statin to non-statin) participants meeting criteria for analysis totaling 7105 visits. Models 1, 2a, and 2b: mean statin difference estimates and mean PTA difference between statin and non-statin users; interaction term represents rate of change in PTA outcomes by statin use over time; all coefficients represent change in dB HL
See Table 1, “Audiometric characteristics (dB HL)” section for definitions of PTAs. Longitudinal data defined as at least two visits and no more than one change in statin use
dB HL decibels hearing level, PTA pure-tone average, RC rate of change
Linear regression model, *p ≤ 0.05; generalized linear mixed-effects model, **p ≤ 0.05
Age-stratified analyses on the cross-sectional data show that the statin use association depended on both PTA and age group, after controlling for baseline covariates (age, sex, race, smoking status, CVD risk, and noise history). Among the 55–64-year age group, statin users showed better thresholds, on average, than non-statin users for both broad PTA (4.34 dB, p = 0.0097) and high-frequency PTA (5.64 dB; p = 0.0056). A smaller (but still significant) difference was observed for the 65–74-year age group, with better average thresholds by 2.78 dB (p = 0.0490) for the broad PTA and 3.66 dB (p = 0.0249) for the high-frequency PTA. No significant difference for the 75+-year age group was seen for either broad or high-frequency PTAs (Fig. 2; Table 3). As in the overall analyses, the association between statin use and low-frequency PTA was not statistically significant across any age group. Contrary to the overall analyses, the relationship between statin use and narrow PTA was no longer significant in the age-stratified analysis.
Fig. 2.
Mean (+1 standard deviation) pure-tone averages (PTAs) at baseline for statin users and non-statin users stratified by age. Statin versus non-statin user: 55–64 years (N = 78 versus N = 308); 65–74 years (N = 167 versus N = 411); 75+ years (N = 88 versus N = 187). In Table 1, see the “Audiometric characteristics (dB HL)” section for definitions of PTAs. Thresholds were adjusted for baseline age, sex, race, noise exposure history, smoking status, and cardiovascular risk (*p ≤ 0.05)
Table 3.
Cross-sectional multivariable model of associations between PTA and statin and non-statin users stratified by age group
| Variable | Cross-sectional (N = 1239) | ||
|---|---|---|---|
| Model 1: mean statin difference | |||
| Dependent PTAs (dB HL) | Independent statin versus non-statin (age, years) | Coefficient | p-value* |
| Narrow | 55–64 | −2.98 | 0.0528 |
| 65–74 | −2.24 | 0.0975 | |
| 75+ | −0.21 | 0.9002 | |
| Broad | 55–64 | −4.34 | 0.0097* |
| 65–74 | −2.78 | 0.0490* | |
| 75+ | 0.07 | 0.9664 | |
| Low-frequency | 55–64 | −1.28 | 0.2721 |
| 65–74 | −0.63 | 0.5655 | |
| 75+ | −0.18 | 0.9029 | |
| High-frequency | 55–64 | −5.64 | 0.0056* |
| 65–74 | −3.66 | 0.0249* | |
| 75+ | −0.17 | 0.9289 | |
Model 1: adjusted for age, sex, race, smoking status, cardiovascular disease risk, and noise exposure. Mean statin difference estimates and mean PTA differences between statin and non-statin users; coefficients represent change in dB HL. See Table 1, “Audiometric characteristics (dB HL)” section for definitions of PTAs. Total N = 1239 participants with baseline data. Statin versus non-statin users: 55–64 years, N = 78 versus N = 308; 65–74 years, N = 167 versus N = 411; 75+ years, N = 88 versus N = 187
dB HL decibels hearing level, PTA pure-tone average
Linear regression model, *p ≤ 0.05
Analysis investigating the impact of statin type on hearing sensitivity provides evidence suggesting that atorvastatin contributes to significantly better hearing thresholds as measured by most PTAs at baseline (Online Resource 3, Fig. 1B; Online Resource 4, Fig. 1C). After controlling for age, race, sex, smoking status, CVD risk, and noise history, atorvastatin users had significantly better narrow PTAs (by 2.70 dB, p = 0.0340) and broad PTAs (by 3.00 dB, p = 0.0240) compared with non-statin users. Consistent with the overall statin association results, atorvastatin users also had significantly better high-frequency PTA thresholds by 3.49 dB (p = 0.0251) compared with non-statin users; differences between low-frequency PTA thresholds were not statistically significant between atorvastatin users and non-statin users (Online Resource 3, Fig. 1B). When comparing simvastatin users and non-statin users, no statistically significant differences were found for any PTA outcomes after controlling for all covariates (Online Resource 4, Fig. 1C).
Longitudinal Analyses
Longitudinal analyses were performed to assess statin association on hearing using all repeated measures without the influence of time (model 2a) and to consider the overall statin association on each PTA outcome to answer the question, “Is there an overall difference in PTA outcomes between statin users and non-statin users when using all available data?” Model 2b additionally examines whether the rate of change in hearing differs by statin use using the same variables as in model 2a while adding an interaction term between statin use and time (i.e., the effect modification or rate of change in PTA outcomes by statin use over time to answer the question, “Is statin use associated with slower declines in hearing over time as individuals age?”). Thus, both models 2a and 2b also consider age-related declines in hearing (not reported), which were found to be statistically significant across all longitudinal models and consistent with reports of ARHL [e.g., 26, 33–37].
Results from the longitudinal analysis of statin and non-statin users indicate no statistically significant differences in overall mean difference in PTA outcomes for statin use when using all available data (Table 2, model 2a,b, row 1). However, model 2b (Table 2, row 2) suggests a statistically significant association between statin use and time (rate of change) for narrow PTA, broad PTA, and low-frequency PTA. Among statin users, average hearing thresholds increase by an additional 0.07 dB per year compared with non-statin users (p = 0.0051) for narrow PTA. Similarly, average hearing thresholds among statin users increased by an additional 0.05 dB per year (p = 0.0365) for broad PTA and an additional 0.07 dB per year (p = 0.0128) for low-frequency PTA. Although these results contradict the direction of the overall mean difference in PTAs among statin users seen in the cross-sectional analyses where statin users show better thresholds (model 1), the additional rate of decline for statin users (versus non-statin users) may be too small to be clinically meaningful.
Results from the longitudinal analysis by statin type differ from the cross-sectional results for atorvastatin users (Online Resource 3). Although atorvastatin users were found to have significantly better average thresholds for narrow PTA, broad PTA, and high-frequency PTA in the cross-sectional cohort (Online Resource 3, model 1), statistically significant differences were only found in the longitudinal cohort for broad and high-frequency PTAs when using all available data (Online Resource 3, model 2a). However, these longitudinal results suggest that atorvastatin users have poorer hearing, on average, than non-statin users for broad PTA by 0.72 dB (p = 0.0158) and for high-frequency PTA by 0.81 dB (p = 0.0191). The rate of change of hearing among atorvastatin users increased by an additional 0.08 dB per year (p = 0.0414) for narrow PTA (Online Resource 3, model 2b, row 2). Among simvastatin users, no statistically significant mean differences in PTAs by statin use were observed (Online Resource 4, model 2a), which is similar to the findings from the cross-sectional analysis (Online Resource 4, model 1). Interestingly, simvastatin use appears to have a small beneficial association on high-frequency PTA over time (Online Resource 4, model 2b, row 2) with hearing declining by 0.12 dB less per year compared with non-statin users (p = 0.0353). The differences observed in the longitudinal results when using all available data may be too small to be clinically meaningful. Furthermore, statistical significance for that association does not hold when also considering statin effect modification over time.
The differences in the results of the models between the cross-sectional and longitudinal analyses led us to compare individuals who had consistently used a statin throughout their participation in the study (always-statin users) versus those who had consistently not used a statin (never-statin users). A subset analysis (Table 4) comparing these two groups shows similar results to the cross-sectional analyses (Table 2). Both longitudinal models 2a and 2b suggest overall better average PTA thresholds for narrow PTA, broad PTA, and high-frequency PTA for always-statin users (versus never). Always-statin users show an average of 1.98 dB better narrow PTA (p = 0.0361) compared with never-statin users, similar to the 2.00-dB difference in the cross-sectional cohort. Average broad PTA thresholds were better among always-statin users by 2.66 dB (p = 0.0095), similar to the 2.54-dB difference observed in the cross-sectional cohort. Always-statin users had an average high-frequency PTA 3.25 dB better (p = 0.0066) than never-statin users (Table 4, model 2a). This suggests that duration or consistency of statin use may influence the association. However, rate of change (model 2b, row 2) was not statistically significant for any PTAs, indicating that, although always-statin users have overall better hearing than never-statin users when using all available data without the influence of time, consistent statin use does not appear to have a durable effect on rate of change of hearing over time.
Table 4.
Longitudinal multivariable models of associations between PTA and always-statin and never-statin users
| Variable | Longitudinal (N = 804) | ||||
|---|---|---|---|---|---|
| Model 2a: mean statin difference | Model 2b: mean statin difference with interaction term | ||||
| Dependent PTAs (dB HL) | Independent | Coefficient | p-value** | Coefficient | p-value** |
| Narrow | Always-statin versus never-statin user | −1.98 | 0.0361** | −1.99 | 0.0351** |
| Statin × years since baseline age (RC) | −0.01 | 0.8505 | |||
| Broad | Always-statin versus never-statin user | −2.66 | 0.0095** | −2.70 | 0.0085** |
| Statin × years since baseline age (RC) | 0.07 | 0.3031 | |||
| Low-frequency | Always-statin versus never-statin user | −1.01 | 0.2101 | −1.03 | 0.2028 |
| Statin × years since baseline age (RC) | −0.02 | 0.7919 | |||
| High-frequency | Always-statin versus never-statin user | −3.25 | 0.0066** | −3.44 | 0.0042** |
| Statin × years since baseline age (RC) | 0.12 | 0.1621 | |||
Models 2a and 2b: adjusted for years since baseline age, sex, race, smoking status, cardiovascular disease risk, and noise exposure that were within 6 months of last audiogram or last record. Total N = 804 (234 always statin; 570 never statin) participants meeting criteria for analysis. Models 2a and 2b: mean statin difference estimates and mean PTA differences between always-statin users and never-statin users; interaction term represents rate of change in PTA outcomes by always-statin use over time; all coefficients represent change in dB HL. See Table 1, “Audiometric characteristics (dB HL)” section for definitions of PTAs and footnotes for definition of longitudinal data
dB HL decibels hearing level, PTA pure-tone average, RC rate of change
Generalized linear mixed-effects model, **p ≤ 0.05
A sensitivity analysis of baseline data from the longitudinal cohort (Online Resource 5) suggests that statin users have, on average, better high-frequency PTA by 2.68 dB (p = 0.0222) than non-statin users, similar to the cross-sectional results that showed better high-frequency PTA by 3.36 dB (p = 0.0017) for statin users. As with the cross-sectional analysis, there is no statistically significant difference in low-frequency PTA (p = 0.7025) between statin and non-statin users. Thus, although the sensitivity analysis replicated the statistically significant differences found in the cross-sectional results for high- and low-frequency PTAs, threshold differences for narrow and broad PTA followed the same general trends but did not reach statistical significance. These differences may relate to the smaller sample size for the baseline data from the longitudinal cohort (Online Resource 5).
Discussion
The primary purpose of this study was to determine the association between statin use and ARHL in a diverse, community-based sample of the general population. Cross-sectional results suggest that statin users have better thresholds at baseline than non-statin users for narrow, broad, and high-frequency PTAs, but no statin benefit was observed for low-frequency PTA. Further, using cross-sectional data and comparing age-stratified baseline thresholds for groups of participants 55–64, 65–74, and 75+ years of age, larger threshold differences were observed between statin and non-statin users for the 55–64-year age group, reduced or non-significant threshold differences for the 65–74-year-aged group, and no significant differences for the 75+-year age group. Analyses by statin type without age stratification showed that atorvastatin contributed to the overall statin association, but simvastatin did not. When including all available longitudinal data without considering the influence of time, a statin benefit was observed when comparing participants who used statins throughout their time in the study (versus those who never used statins). Finally, results of longitudinal analyses that assessed the association of statin use over time suggest that statin use did not slow the observed declines in hearing as participants aged. Overall, these results are consistent with a statin benefit on thresholds at mid-to-high frequencies for middle-aged to older adults at the early stages of ARHL.
The observed pattern of threshold benefits favoring mid-to-high frequencies is consistent with some studies that evaluated associations between statin use and hearing but not with others. One study in mice reported greater protection with statin use against cisplatin-induced hearing loss at higher than lower frequencies [12], which may occur because cochlear damage due to cisplatin was greatest at higher frequencies. However, no differences by frequency were observed in a mouse study of statin protection against NIHL [16], and only minimal differences by frequency were observed in a study of statin protection against kanamycin-induced hearing loss in rats [13]. A higher frequency trend was observed in statin protection when assessed by amplitudes of distortion-product otoacoustic emissions [18], but a lower frequency trend in statin protection in the early stages of ARHL was observed when hearing was assessed by auditory brainstem responses [19]. Among human studies, one study reported significantly reduced cisplatin-induced threshold shifts at frequencies ≥ 4.0 kHz among statin users [14], and another study focused on ARHL showed no such frequency trend [20]. Taken together, frequency-related patterns in statin association on hearing in laboratory animals are inconsistent, and most human studies of statins and hearing report only one PTA, so frequency-related patterns in statin associations cannot be determined.
In the current study, baseline thresholds were higher at mid-to-higher frequencies than at lower frequencies (Fig. 1; Table 1), which is consistent with hearing thresholds reported previously for this cohort [27, 38, 39] and other large cohorts of older adults [2, 40]. This raises the question of whether the typical audiometric shape of ARHL, with more hearing loss at higher than lower frequencies, may explain the frequency differences observed in the statin benefit. The mechanisms underlying ARHL that give rise to its audiometric phenotype include degeneration of the stria vascularis microvasculature [41], which reduces power for outer hair cell motility throughout the cochlea. This loss in amplification of low-level sounds across the cochlea reduces hearing sensitivity more for higher than lower frequency sounds, resulting in the typical gradually sloping threshold configuration of ARHL [42, 43]. Thus, a statin association favoring the higher frequencies in ARHL could relate to preservation of cochlear blood supply in the highly vascularized stria, rather than mitigating hair cell damage solely affecting the higher frequency portion of the cochlea.
Other evidence from the current study does not support an association between statin use and magnitude of hearing loss. First, the age-stratified cross-sectional results demonstrated that statin benefits at baseline were largest in the middle-aged group (55–64 years), smaller statin benefits but still significant in the older group (65–74 years), and not significant in the oldest group (75+ years); Fig. 2 shows that baseline thresholds increased systematically with age group. Second, statin use and statin type differences were observed at baseline when participants were at their youngest ages and had better thresholds. In contrast, these associations were not observed when using all available data throughout participants’ time in the study when participants were older, and thresholds were higher. Therefore, benefits of statin use appear to diminish with increasing age and hearing loss.
Results of the current study related to difference in statin type were consistent with another human study showing a statin benefit attributable to atorvastatin but not to simvastatin [14], although that study was not sufficiently powered to detect statin benefits due to simvastatin. While all statins function as HMG-CoA reductase inhibitors, they differ in derivation, potency, and bioavailability. Differential effects of atorvastatin and simvastatin may reflect differences in pharmacokinetic and pharmacodynamic profiles. Atorvastatin is more potent and hydrophilic than simvastatin, with higher bioavailability and a longer half-life. It relies less on first-pass hepatic metabolism and has increased extrahepatic tissue distribution [44]; thus, atorvastatin may be either more available to the cochlea or available for more sustained periods of time compared with simvastatin (for review [10]). This, in turn, may support better cochlear blood flow and decreased oxidative stress within the stria vascularis, a hypothesis supported by previous animal studies that show preservation of cochlear microvascular and auditory function with statin therapy [18, 41]. In contrast, simvastatin is a prodrug that is more extensively metabolized [45], leading to less systemic bioavailability, possibly reducing its efficacy in cochlear protection [8, 14]. Variations in observed hearing outcomes may be explained, in part, by some of these pharmacologic differences. Although all statins function as HMG-CoA reductase inhibitors, these differences in metabolism and tissue distribution merit a deeper investigation into the differential effects of statins on the auditory system. The current evidence from animal models and human studies is insufficient to explain the statin type differences on thresholds observed in the current study.
Finally, statin benefits observed at baseline did not persist in longitudinal analyses that assessed statin associations over time, suggesting that thresholds increased with age at rates that overcame the relatively small threshold benefit attributable to statins at baseline. Thus, the results did not identify a durable longitudinal association between hearing thresholds and statin use. This could be indicative of adaptive cellular responses, cumulative metabolic effects, or progressive age-related cochlear degradation that counter any early protective influences of statins. These findings are similar to those from a previous study of hearing and blood lipid levels from the same cohort, which showed that elevated triglyceride levels were associated with small increases in PTAs at baseline, but no significant association remained following longitudinal analyses [46]. As in this earlier study, it is possible that longer-term benefits of statins as users age are modulated by other factors detrimental to hearing that contribute to declines in hearing with increasing age, such as dietary and environmental factors.
This study is the first to assess the relationship between ARHL and statin use in a cohort where hearing and statin use were measured repeatedly over time, which strengthens the validity of the findings. Other strengths of this community-based cohort study include its large and racially diverse sample and comprehensive measures of hearing and health, including CVD risk factors. This cohort is similar to cohorts in other epidemiological studies of hearing in terms of age and ARHL [2, 27, 40], which increases generalizability. Some study limitations exist. This observational study did not include randomized assignment to statin and non-statin groups. In addition, because we did not control for all demographic factors and health conditions that may affect hearing, or for interactions with other drugs taken, threshold differences between statin user groups could partially be explained by these unmeasured factors. We do not have dosage information for statin users; for statin users at baseline, we could not determine precisely when statin use began or the total duration of statin use. Outcomes of this study were limited to detection thresholds; we did not evaluate benefit of statins for suprathreshold or other aspects of auditory function.
Conclusions
This study, conducted in a diverse, community-based sample of the general population, showed that statin use (versus nonuse) was significantly associated with better hearing at baseline in mid-to-high frequencies. Atorvastatin contributed to this association, while simvastatin did not. The overall statin benefit observed at baseline was largest for 55–64-year olds, diminished for 65–74-year olds, and not observed for 75+-year olds. When using all available longitudinal data and not considering the influence of time, a statin benefit was observed only with consistent statin use. However, a statin benefit was not observed for any participant group when hearing was examined longitudinally over time, suggesting that statin use does not slow the normally observed declines in hearing over time as individuals age. Taken together, the results indicate that statin use is associated with beneficial effects on hearing, and they highlight the need for well-controlled clinical trials with the sole purpose of evaluating the effects of statin use and specific statin types on ARHL in the general population.
Supplementary Information
Below is the link to the electronic supplementary material.
Acknowledgements
We thank the participants of the MUSC Longitudinal Cohort Study of Age-Related Hearing Loss for their commitment and dedication to the study.
Funding
Open access funding provided by the Carolinas Consortium. This work was supported (in part) by the National Institute on Deafness and Other Communication Disorders (NIDCD) at the National Institutes of Health (grant no. P50 DC 000422) awarded to the Medical University of South Carolina and by the South Carolina Clinical and Translational Research (SCTR) Institute, with an academic home at the Medical University of South Carolina, National Center for Advancing Translational Sciences at the National Institutes of Health (grant no. UL1 TR001450) (L.G.M., L.J.M., A.N.S., and J.R.D.). This investigation was conducted in a facility constructed with support from Research Facilities Improvement Program grant no. C06 RR14516 from the National Center for Research Resources at the National Institutes of Health. This work was also supported by the Division of Intramural Research at the NIDCD (ZIA DC000079) (K.A.F. and L.L.C.). The content is solely the responsibility of the authors and does not necessarily represent the views of the National Institutes of Health.
Declarations
Conflict of interest
Lizmarie G. Maldonado, Lois J. Matthews, Katharine A. Fernandez, Annie N. Simpson, Lisa L. Cunningham, and Judy R. Dubno declare that they have no potential conflicts of interest that might be relevant to the contents of this manuscript.
Availability of data and material
Please contact the corresponding author to request data used in this study, which can be made available with a data use agreement and institutional approvals according to guidelines of the Medical University of South Carolina.
Ethics approval
Recruitment, consenting, and test protocols were approved by the Institutional Review Board for Human Research at the Medical University of South Carolina (approval no. Pro00143394). The study was performed in accordance with the ethical standards as laid down in the 1964 Declaration of Helsinki and its later amendments.
Consent to participate
All participants provided written informed consent.
Consent for publication
Not applicable.
Code availability
Not applicable.
Author contributions
Conceptualization: JRD, LLC, KAF, LJM, and ANS; methodology: JRD, LLC, KAF, LJM, ANS, and LGM; acquisition of data: LJM and JRD; formal analysis: LGM and ANS; writing—original draft: LGM and JRD; writing—review and editing: LLC, KAF, LJM, and ANS; administrative, technical, and material support: JRD and LJM; acquisition of funding: JRD. All authors read and approved the final version.
References
- 1.World Health Organization. World report on hearing. Geneva: World Health Organization; 2021. License: CC BY-NC-SA 3.0 IGO. Available from: https://www.who.int/publications/i/item/9789240020481.
- 2.Cruickshanks KJ, Wiley TL, Tweed TS, Klein BEK, Klein R, Mares-Perlman JA, et al. Prevalence of hearing loss in older adults in Beaver Dam, Wisconsin: the epidemiology of hearing loss study. Am J Epidemiol. 1998;148(9):879–86. 10.1093/oxfordjournals.aje.a009713. [DOI] [PubMed] [Google Scholar]
- 3.Lin FR, Thorpe R, Gordon-Salant S, Ferrucci L. Hearing loss prevalence and risk factors among older adults in the United States. J Gerontol A Biol Sci Med Sci. 2011;66(5):582–90. 10.1093/gerona/glr002. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Nash SD, Cruickshanks KJ, Klein R, et al. The prevalence of hearing impairment and associated risk factors: the Beaver Dam Offspring Study. Arch Otolaryngol Head Neck Surg. 2011;137(5):432–9. 10.1001/archoto.2011.15. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Strawbridge WJ, Wallhagen MI, Shema SJ, Kaplan GA. Negative consequences of hearing impairment in old age: a longitudinal analysis. Gerontologist. 2000;40(3):320–6. 10.1093/geront/40.3.320. [DOI] [PubMed] [Google Scholar]
- 6.Dalton DS, Cruickshanks KJ, Klein BE, Klein R, Wiley TL, Nondahl DM. The impact of hearing loss on quality of life in older adults. Gerontologist. 2003;43(5):661–8. 10.1093/geront/43.5.661. [DOI] [PubMed] [Google Scholar]
- 7.Bainbridge KE, Ramachandran V. Hearing aid use among older U.S. adults; the National Health and Nutrition Examination Survey, 2005–2006 and 2009–2010. Ear Hear. 2014;35(3):289–94. 10.1097/01.aud.0000441036.40169.29. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Ramkumar S, Raghunath A, Raghunath S. Statin therapy: review of safety and potential side effects. Acta Cardiol Sin. 2016;32(6):631–9. 10.6515/acs20160611a. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Newman CB, Preiss D, Tobert JA, et al. American Heart Association Clinical Lipidology, Lipoprotein, Metabolism and Thrombosis Committee, a Joint Committee of the Council on Atherosclerosis, Thrombosis and Vascular Biology and Council on Lifestyle and Cardiometabolic Health; Council on Cardiovascular Disease in the Young; Council on Clinical Cardiology; and Stroke Council. Statin safety and associated adverse events: a scientific statement from the American Heart Association. Arterioscler Thromb Vasc Biol. 2019;39(2):e38–e81. 10.1161/ATV.0000000000000073. [Erratum in: Arterioscler Thromb Vasc Biol. 2019;39(5):e158. 10.1161/ATV.0000000000000073]. [DOI] [PubMed]
- 10.Whitlon DS. Statins and hearing. Hear Res. 2022;425:108453. 10.1016/j.heares.2022.108453. [DOI] [PubMed] [Google Scholar]
- 11.Brand Y, Setz C, Levano S, et al. Simvastatin protects auditory hair cells from gentamicin-induced toxicity and activates Akt signaling in vitro. BMC Neurosci. 2011;12:114. 10.1186/1471-2202-12-114. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Fernandez K, Spielbauer KK, Rusheen A, Wang L, Baker TG, Eyles S, et al. Lovastatin protects against cisplatin-induced hearing loss in mice. Hear Res. 2020;389:107905. 10.1016/j.heares.2020.107905. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Lee CH, Jeon J, Lee SM, Kim SY. Pravastatin administration alleviates kanamycin-induced cochlear injury and hearing loss. Int J Mol Sci. 2022;23(9):4524. 10.3390/ijms23094524. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Fernandez KA, Allen P, Campbell M, Page B, Townes T, Li C-M, et al. Atorvastatin is associated with reduced cisplatin-induced hearing loss. J Clin Invest. 2021;131(1):e142616. 10.1172/JCI142616. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Richter CP, Young H, Richter SV, et al. Fluvastatin protects cochleae from damage by high-level noise. Sci Rep. 2018;8(1):3033. 10.1038/s41598-018-21336-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Dépreux F, Czech L, Young H, Richter C-P, Zhou Y, Whitlon DS. Statins protect mice from high-decibel noise-induced hearing loss. Biomed Pharmacother. 2023;163:114674. 10.1016/j.biopha.2023.114674. [DOI] [PubMed] [Google Scholar]
- 17.Borghi C, Modugno GC, Pirodda A. Possible role of HMG-CoA reductase inhibitors for the treatment of sudden sensorineural hearing loss (SSHL). Med Hypotheses. 2002;58(5):399–402. 10.1054/mehy.2001.1535. [DOI] [PubMed] [Google Scholar]
- 18.Syka J, Ouda L, Nachtigal P, Solichová D, Semecký V. Atorvastatin slows down the deterioration of inner ear function with age in mice. Neurosci Lett. 2007;411(2):112–6. 10.1016/j.neulet.2006.10.032. [DOI] [PubMed] [Google Scholar]
- 19.Choo OS, Lee YY, Kim YS, Kim YJ, Lee DH, Kim H, et al. Effect of statin on age-related hearing loss via drug repurposing. Biochimica et Biophysica Acta (BBA) Mol Cell Res. 2022;1869(11):119331. 10.1016/j.bbamcr.2022.119331. [DOI] [PubMed] [Google Scholar]
- 20.Gopinath B, Flood VM, Teber E, McMahon CM, Mitchell P. Dietary intake of cholesterol is positively associated and use of cholesterol-lowering medication is negatively associated with prevalent age-related hearing loss. J Nutr. 2011;141(7):1355–61. 10.3945/jn.111.138610. [DOI] [PubMed] [Google Scholar]
- 21.Choo OS, Yoon D, Choi Y, Choo O-S, Jo S, Jung H-M, et al. Drugs for hyperlipidaemia may slow down the progression of hearing loss in the elderly: a drug repurposing study. Basic Clin Pharmacol Toxicol. 2019;124(4):423–30. 10.1111/bcpt.13150. [DOI] [PubMed] [Google Scholar]
- 22.Kim SY, Lee CH, Min C, Yoo DM, Choi HG. Association between statin medication and hearing impairment in a national health screening cohort. Sci Rep. 2021;11(1):14388. 10.1038/s41598-021-93916-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Song I, Kim M, Choi H, et al. Hydrophilic and lipophilic statin use and risk of hearing loss in hyperlipidemia using a common data model: multicenter cohort study. Sci Rep. 2023;13(1):12373. 10.1038/s41598-023-39316-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Bulbul MG, Wanstreet T, Kais A, Jones GH, Kellermeyer BM. Meniere’s disease and statins: is there an association? A single institution study. OTO Open. 2024;8(3):e177. 10.1002/oto2.177. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Matthews LJ, Lee FS, Mills JH, Dubno JR. Extended high-frequency thresholds in older adults. J Speech Lang Hear Res. 1997;40(1):208–14. 10.1044/jslhr.4001.208. [DOI] [PubMed] [Google Scholar]
- 26.Lee FS, Matthews LJ, Dubno JR, Mills JH. Longitudinal study of pure-tone thresholds in older persons. Ear Hear. 2005;26(1):1–11. 10.1097/00003446-200502000-00001. [DOI] [PubMed] [Google Scholar]
- 27.Dubno JR, Lee FS, Matthews LJ, Ahlstrom JB, Horwitz AR, Mills JH. Longitudinal changes in speech recognition in older persons. J Acoust Soc Am. 2008;123(1):462–75. 10.1121/1.2817362. [DOI] [PubMed] [Google Scholar]
- 28.Simpson AN, Matthews LJ, Cassarly C, Dubno JR. Time from hearing aid candidacy to hearing aid adoption: a longitudinal cohort study. Ear Hear. 2019;40(3):468–76. 10.1097/AUD.0000000000000641. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Dillard LK, Matthews LJ, Dubno JR. Prevalence of self-reported hearing difficulty on the Revised Hearing Handicap Inventory and associated factors. BMC Geriatr. 2024;24(1):510. 10.1186/s12877-024-04901-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Cruickshanks KJ, Nondahl DM, Tweed TS, Wiley TL, Klein BEK, Klein R, et al. Education, occupation, noise exposure history and the 10-yr cumulative incidence of hearing impairment in older adults. Hear Res. 2010;264(1–2):3–9. 10.1016/j.heares.2009.10.008. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.White IR, Royston P, Wood AM. Multiple imputation using chained equations: issues and guidance for practice. Stat Med. 2011;30(4):377–99. 10.1002/sim.4067. [DOI] [PubMed] [Google Scholar]
- 32.Graham JW. Missing data analysis: making it work in the real world. Annu Rev Psychol. 2009;60:549–76. 10.1146/annurev.psych.58.110405.085530. [DOI] [PubMed] [Google Scholar]
- 33.Wiley TL, Chappell R, Carmichael L, Nondahl DM, Cruickshanks KJ. Changes in hearing thresholds over 10 years in older adults. J Am Acad Audiol. 2008;19(4):281–92. 10.3766/jaaa.19.4.2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Brant LJ, Fozard JL. Age changes in pure-tone hearing thresholds in a longitudinal study of normal human aging. J Acoust Soc Am. 1990;88(2):813–20. 10.1121/1.399731. [DOI] [PubMed] [Google Scholar]
- 35.Pearson JD, Morrell CH, Gordon-Salant S, Brant LJ, Metter EJ, Klein LL, et al. Gender differences in a longitudinal study of age-associated hearing loss. J Acoust Soc Am. 1995;97(2):1196–205. 10.1121/1.412231. [DOI] [PubMed] [Google Scholar]
- 36.Gates GA, Cooper JC. Incidence of hearing decline in the elderly. Acta Otolaryngol. 1991;111(2):240–8. 10.3109/00016489109137382. [DOI] [PubMed] [Google Scholar]
- 37.Dillard LK, Matthews LJ, Maldonado L, Simpson AN, Dubno JR. Demographic factors impact the rate of hearing decline across the adult lifespan. Commun Med (Lond). 2024;4(1):171. 10.1038/s43856-024-00593-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Vaden KI Jr, Neely ST, Harris SE, Dubno JR. Metabolic and sensory components of age-related hearing loss: associations with distortion- and reflection-based otoacoustic emissions. Trends Hear. 2023;27:1–18. 10.1177/23312165231213776. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Helfer KS, Maldonado L, Matthews LJ, Simpson AN, Dubno JR. Extended high-frequency thresholds: associations with demographic and risk factors, cognitive ability, and hearing outcomes in middle-aged and older adults. Ear Hear. 2024;45(6):1427–43. 10.1097/AUD.0000000000001531. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Gates GA, Cooper JC Jr, Kannel WB, Miller NJ. Hearing in the elderly: the Framingham cohort, 1983-1985. Part I. Basic audiometric test results. Ear Hear. 1990;11(4):247–56. [PubMed] [Google Scholar]
- 41.Gratton MA, Schmiedt RA, Schulte BA. Age-related decreases in endocochlear potential are associated with vascular abnormalities in the stria vascularis. Hear Res. 1996;102(1–2):181–90. 10.1016/s0378-5955(96)90017-9. [DOI] [PubMed] [Google Scholar]
- 42.Schmiedt RA, Lang H, Okamura HO, Schulte BA. Effects of furosemide applied chronically to the round window: a model of metabolic presbyacusis. J Neurosci. 2002;22(21):9643–50. 10.1523/JNEUROSCI.22-21-09643.2002. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Eckert MA, Harris KC, Lang H, Lewis MA, Schmiedt RA, Schulte BA, et al. Translational and interdisciplinary insights into presbyacusis: a multidimensional disease. Hear Res. 2021;402:108109. 10.1016/j.heares.2020.108109. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Lennernäs H. Clinical pharmacokinetics of atorvastatin. Clin Pharmacokinet. 2003;42:1141–50. 10.2165/00003088-200342130-00005. [DOI] [PubMed] [Google Scholar]
- 45.Mauro VF. Clinical pharmacokinetics and practical applications of simvastatin. Clin Pharmacokinet. 1993;24(3):195–202. 10.2165/00003088-199324030-00002. [DOI] [PubMed] [Google Scholar]
- 46.Simpson AN, Matthews LJ, Dubno JR. Lipid and C-reactive protein levels as risk factors for hearing loss in older adults. Otolaryngol Head Neck Surg. 2013;148(4):664–70. 10.1177/0194599812473936. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.