Skip to main content
Download PDF

Relationship between hearing impairment and dementia and cognitive function: a Mendelian randomization study

Alzheimer's Research & Therapy volume 16, Article number: 215 (2024) Cite this article

Abstract

Background

There is a substantial body of observational research indicating an association between hearing impairment and dementia, yet the causal relationship and underlying mechanisms remain uncertain. This study aims to investigate the causal relationship between hearing impairment and its subtypes with dementia and cognitive function using two-sample Mendelian randomization (MR) analysis.

Methods

We performed two-sample MR analysis to examine the causal effects of hearing impairment and its subtypes, including conductive and sensorineural hearing loss (CSHL), conductive hearing loss (CHL), sensorineural hearing loss (SHL), and sudden sensorineural hearing loss (SIHL), on six dementia phenotypes (overall dementia, Alzheimer’s disease [AD], Lewy body dementia [DLB], frontotemporal dementia [FTD], Parkinson’s disease dementia, and vascular dementia) and four cognitive functions. Additionally, multivariable MR (MVMR) analysis was employed to investigate potential mediating mechanisms.

Results

Genetically determined CSHL was associated with an elevated risk of DLB (odds ratio [OR] 1.69; 95% CI 1.08 to 2.63; P = 0.021) and FTD (OR 1.66; 1.04 to 2.67; P = 0.035), but not AD (P = 0.958). Genetic predisposition to CHL was found to link with increased risks of AD (OR 1.07; 1.01 to 1.14; P = 0.031). Genetically determined SHL was causally associated with an elevated risk of semantic dementia (OR 3.81; 1.09 to 13.37; P = 0.037). Genetically predicted CHL and SIHL were both causally associated with lower general cognitive performance (β -0.015 and − 0.043; P = 0.007 and 0.013) and fluid intelligence score (β -0.045 and − 0.095; P = 0.037 and 0.040). In MVMR analysis, the causal relationship between hearing impairment and dementia was mediated by loneliness, depressed mood, and brain cortical volume, particularly the medial temporal lobe, but not by aging or ischemic stroke.

Conclusion

Overall, the study provides evidence supporting a causal relationship between hearing impairment and increased risks of different types of dementia (including AD, FTD, and DLB), as well as poorer general cognitive function. These findings underscore the importance of addressing hearing impairment as a modifiable risk factor for dementia.

Background

Hearing impairment is a common clinical issue among the elderly, with its prevalence increasing with age. Globally, approximately 10% of individuals aged 40–69 years experience hearing impairment (≥ 20 dB), with the prevalence increasing to 30% among those aged over 65 years, and reaching 70–90% among individuals aged 85 years or older [1,2,3]. Hearing impairment is categorized into conductive hearing impairment, sensorineural hearing impairment, and mixed hearing impairment. Age-related sensorineural hearing impairment, known as presbycusis, is the most prevalent type among adults [4]. Hearing impairment can lead to challenges in speech comprehension, impeding effective communication and social engagement, thereby profoundly affecting both physical and mental health [2].

Hearing impairment has been found to be closely associated with dementia, and been identified as an early biomarker as well as the most modifiable risk factor for dementia [5, 6]. A meta-analysis of observational cross-sectional and cohort studies has found that age-related hearing impairment is associated with declines in cognitive functions such as global cognition and executive functions, as well as an increased risk of all-cause dementia, but not specifically with Alzheimer’s disease (AD) [5]. Furthermore, another meta-analysis summarized the associations of hearing aids and cochlear implants with cognitive decline and dementia. The study found that the use of hearing aids among participants with hearing loss was linked to a 19% reduction in the risk of long-term cognitive decline and was notably correlated with a 3% enhancement in cognitive test scores assessing overall cognitive function [7]. The primary mechanisms thought to underlie the association between hearing impairment and dementia include three potential hypotheses. Firstly, the common cause hypothesis suggests that hearing impairment and cognitive decline are manifestations of different aspects of the same underlying factor. Common factors mainly include aging, vascular pathology, and dementia pathologies such as amyloid and tau [8,9,10]. Secondly, the information/sensory degradation hypothesis suggests that difficulty perceiving speech may lead to social withdrawal and depression, while hearing loss reduces stimulation in cognitive processes. Impoverished environment resulting from hearing loss leads to reduced cognitive reserve, thereby decreasing resilience to dementia [9, 10]. Thirdly, the hypothesis of interaction between auditory brain function and cognition proposes that auditory brain function is crucial in linking hearing impairment with cognitive decline, particularly in the medial temporal gyrus (MTG). Hearing impairment may alter the activity/structure of these cortical brain regions and interact with dementia pathology [11, 12]. While there is substantial evidence supporting the idea that hearing impairment could serve as an indicator of increased risk for cognitive decline and dementia, it is currently unclear whether this correlation is a causal relationship or merely driven by confounding factors upstream or reverse causality. Additionally, there is limited exploration into the relationship between different types of hearing impairment and various types of dementia and cognitive function.

Observational studies are prone to confounding and reverse causation, rendering them inadequate for establishing causal relationships. For example, age-related hearing impairment patients frequently present with concurrent cardiovascular risk factors, which are also risk factors for dementia. Mendelian randomization (MR) seeks to utilize genetic variation as an instrumental variable for the exposure of interest, offering evidence of causal relationships. Given that genetic variations are assumed to be randomly inherited and alleles are unaffected by the disease, this approach minimizes the influence of confounders and reverse causation biases to a significant degree [13].

In this study we aimed to utilize two-sample MR to investigate the causal relationship between hearing impairment and its subtypes with dementia (overall dementia, AD, Lewy body dementia [DLB], and frontotemporal dementia [FTD], Parkinson’s disease dementia [PDD], and vascular dementia [VaD]), and cognitive function. Furthermore, we employed multivariable Mendelian randomization (MVMR) to explore whether aging, ischemic stroke, loneliness, depression, and brain volume mediate the association between hearing impairment and dementia.

Methods

Study design

We inferred causal relationships between different subtypes of hearing impairment (exposures) and dementia and cognition (outcomes) using univariable 2-sample MR, in which the effects of the genetic instruments on the exposure and the outcome are obtained from two separate, non-overlapping datasets. Subsequently, we utilized a multivariable MR method to adjust for candidate covariates (aging, ischemic stroke, loneliness, depressed mood, and cortical volume), aiming to explore the mechanisms mediating the causal relationship between hearing impairment and dementia [14]. The reporting of this study adhered to the Strengthening the Reporting of Observational Studies in Epidemiology using Mendelian Randomisation (STROBE-MR) reporting guideline [15]. All summary data used in this study are publicly available and restricted to European ancestry. Additionally, all participants in previous studies included in this research obtained appropriate ethical approvals and informed patient consent.

Genome-wide association studies (GWAS) databases

The genetic variants associated with four subtypes of hearing impairment were obtained from the FinnGen repository. The dataset included 17,337 cases of conductive and sensorineural hearing loss (CSHL), 1,255 cases of conductive hearing loss (CHL), 15,952 cases of sensorineural hearing loss (SHL), and 1,491 cases of sudden sensorineural hearing loss (SIHL) [16].

We collected GWAS summary statistics for six dementia phenotypes, including overall dementia, AD, DLB, FTD, PDD, and VaD. Specifically, summary statistics for overall dementia (phenotype code 290.1), which encompasses all dementia types, consist of 1,974 cases and 410,833 controls, sourced from the UK biobank (UKB) [17]. The GWAS dataset for vascular dementia (phenotype code 290.16) comprises 444 cases and 410,833 controls, also sourced from the UKB [17]. The GWAS summary statistics for AD were sourced from the International Genomics of Alzheimer’s Project, which conducted the largest GWAS meta-analysis encompassing clinically diagnosed late-onset AD (onset age > 65 years; N = 21,982) [18]. For FTD and its subtypes, the GWAS results were obtained from the International Frontotemporal Dementia Genetics Consortium. The accessed data consisted of summary statistics derived from the discovery dataset, including 2,154 cases of all types of FTD, 1,377 cases of behavioral variant FTD (bvFTD), 308 cases of semantic dementia (SD), 269 cases of progressive non-fluent aphasia (PNFA), and 200 cases of FTD overlapping with motor neuron disease (FTD-MND) [19]. The GWAS summary statistics for dementia with Lewy bodies (DLB) were derived from a recent study involving 2,591 cases across 44 consortia [20]. The GWAS summary statistics for PDD were obtained from 12 longitudinal cohorts of PD patients, totaling 4,093 individuals. Cognitive impairment was defined based on binary outcomes (e.g., MMSE < 27 or MoCA < 24) assessed during participant visits [21].

For the cognitive function phenotypes, we primarily selected four cognitive function outcomes. GWAS data for the general cognitive performance were extracted from the Social Science Genetic Association Consortium GWAS meta-analyzing UK biobank fluid intelligence verbal-numerical reasoning scores and Cognitive Genomics Consortium (COGENT) neuropsychological test data (N = 257,841) [22]. GWAS results for fluid intelligence score, numeric memory, and executive function were all available from the UKB [23,24,25]. Fluid intelligence score was calculated as a simple unweighted sum of the number of correct answers given to the 13 logic/reasoning-type questions (N = 149,051). The numeric short-term memory was assessed by recording the maximum digit length recalled correctly (N = 106,162). Participants recalled a 2-digit number, with the length increasing by one digit per correct recall, up to 12 digits. Executive function was assessed using the common executive function (cEF) factor score, derived from the aggregation of performance across five executive function tasks (the trail-making task, symbol-digit substitution, backward digit span, prospective memory, and pairs matching), which were evaluated multiple times (N = 427,037) [25].

Based on previous research evidence, current understanding suggests possible mechanisms linking hearing impairment and dementia: the common cause hypothesis (involving aging and cerebrovascular pathology), the information/sensory degradation hypothesis (related to loneliness and depressive mood), and the hypothesis of interaction between auditory brain function and cognition (involving cortical atrophy) [9, 10, 12, 26]. Therefore, we collected GWAS data for five covariates. Biological aging was assessed using PhenoAge, which integrates 42 clinical measurements along with chronological age. The summary statistics were derived from a GWAS meta-analysis conducted on 28 cohorts consisting of 34,463 participants of European ancestry [27, 28]. Summary statistic for ischemic stroke were collected from the Multi-ancestry Genome-Wide Association Study of Stroke (MEGASTROKE) consortium, including 34,217 patients and 406,111 controls of European ancestry [29]. Summary statistics for loneliness and social isolation were derived from a multi-trait GWAS analysis using the UKB data. This assessment was based on self-reported responses to three related questions concerning perceived loneliness, frequency of social interactions, and ability to confide in someone [30]. Frequency of depressed mood in the last two weeks (N = 442,840) was assessed using a touchscreen questionnaire specifically designed for the UKB [31]. GWAS data for cortical volume were obtained from a recent study that primarily included individuals from the UKB (N = 31,797) and the Adolescent Brain Cognitive Development (ABCD) cohorts (N = 4,866) with European genetic ancestries [32]. Additionally, we collected GWAS data for cortical regional volume using the Desikan-Killiany parcellation method from the UKB [33]. Table 1 presents a comprehensive list of the GWAS studies utilized to obtain the summary statistics for each phenotype. We also provided additional clarifications for trait measurements in Additional file 1 (Table S1).

Table 1 Information of GWAS summary statistics for MR analyses
Full size table

Genetic instruments selection

To meet the strong association assumption, we selected single nucleotide polymorphisms (SNPs) associated with the exposure of hearing impairment at a genome-wide significance threshold (P < 5 × 10− 8) as instrumental variables (IVs). Since the phenotype of conductive hearing impairment had no SNPs at P < 5 × 10− 8, we used a relaxed significance threshold (P < 5 × 10− 6) to select IVs. A clumping cutoff R2 of 0.001 and a 10,000-kilobase window was used to avoid the effects of linkage disequilibrium based on the European-based 1000 Genome Projects reference panel. The strength of IVs was calculated by the F-statistic using the formula: \(\:F=\frac{{\beta\:}^{2}}{{se}^{2}}\) [34,35,36]. IVs with F-statistic < 10 are considered weak instruments [37]. Before the MR analysis, the alleles of both exposure and outcome SNPs were then harmonized to ensure that the effect alleles were identical. To meet the exclusivity assumption, SNPs directly associated with dementia or cognition outcomes (P < 5 × 10−8) were also removed in each MR analysis.

Statistical analysis

We performed univariable MR analyses to investigate the causal effects of hearing impairment and its subtypes on dementia diseases and cognitive function. For the relationship between hearing impairment and dementia diseases, we presented MR estimates as odds ratios (ORs) with 95% confidence intervals (CIs); while for cognitive function, β coefficients with CIs were used. We used the inverse-variance weighted (IVW) analysis as the primary method to evaluate the causal effects. The Wald ratio method was used to estimate the association containing only one IV [38]. To assess the robustness of primary analyses, we performed several complementary sensitivity analyses using weighted median, weighted mode, and MR-Egger regression with bootstrapped standard error method. According to Bowden et al. [39], the weighted-median approach provides consistent estimates as long as up to 50% of the information is derived from valid IVs. The robustness of estimates is enhanced by the weighted-mode method when the majority of similar individual-instrument causal effect estimates originate from valid instrumental variables, even if a significant portion of them is invalid [40]. We calculated Cochran’s IVW Q-values to identify potential heterogeneity among SNPs included in each analysis. To assess horizontal pleiotropy, we employed MR-Egger regression, wherein the deviation of the Egger intercept from zero was considered significant (P < 0.05) indicative of the presence of horizontal pleiotropy [41]. We also examined possible pleiotropy using the MR-PRESSO Global test with P values calculated according to 1,000 simulations. If substantial global heterogeneity was detected, a local outlier test was subsequently conducted to identify any outlier SNPs. Subsequently, the causal effect estimates were reassessed after removing outliers.

We performed MVMR through IVW method to investigate the potential mediating role of aging, ischemic stroke, loneliness, depressive mood, and cortical volume in the relationship between hearing impairment and dementia [14, 42]. We also calculated Cochran’s Q statistic to evaluate heterogeneity. When P value of Cochran’s Q value statistic was less than 0.05, a random effects IVW MR analysis would be used. In addition, we adjusted for each of the cortical regional volume using Desikan-Killiany parcellation in turn to estimate the direct causal effect of hearing impairment on dementia. After establishing a significance threshold (P < 5 × 10− 8) for IV selection, a total of 51 brain regions were included in the MVMR. All MR analyses in this study were performed using R software (version 4.1.2) and the “TwoSampleMR” package (version 0.5.6), “MRPRESSO” package (version 1.0), and “MVMR” package (version 0.4).

Results

After genetic IVs selection, we obtained a total of 28 genetic IVs for four hearing impairment traits (10 for CSHL; 6 for CHL; 11 for SHL; 1 for SIHL). No weak genetic IVs were identified as the F-statistics values ranged from 20.9 to 90.3. Additional file 1 (Table S2) provides detailed information about the genetic IVs used in the MR analyses for four hearing impairment traits.

Causal effect of hearing impairment on dementia

The causal effects of four types of hearing impairment on six dementia phenotypes are shown in Fig. 1. The results of the main IVW analysis suggested that genetic predisposition to CSHL was associated with an elevated risk of DLB (odds ratio [OR] 1.69; 95% CI 1.08 to 2.63; P = 0.021) and FTD (OR 1.66; 1.04 to 2.67; P = 0.035). These results also were supported by the sensitivity analyses based on weighted-median method (respectively P = 0.009 and P = 0.055). Cochran’s Q statistic indicated the absence of heterogeneity in these IVW analysis (respectively: Q = 8.49, P = 0.204; Q = 0.72, P = 0.948). No significant pleiotropy was observed in these analyses (respectively: MR-Egger intercept P = 0.821 and MR-PRESSO P = 0.089; P = 0.762 and P = 0.968). In the subgroup analysis, CSHL was only associated with the subtype of bvFTD (OR 1.86; 1.04 to 3.30; P = 0.035). Genetically determined CSHL was not associated with increased risks of AD (P = 0.958). However, genetic predisposition to CHL was found to link with increased risks of AD (OR 1.07; 1.01 to 1.14; P = 0.031). The complementary sensitivity analyses yielded comparable trends, although they did not reach statistical significance (all OR > 1; P range: 0.072 to 0.164). Similarly, no significant heterogeneity (Q = 1.25, P = 0.740) or pleiotropy (MR-Egger intercept P = 0.885 and MR-PRESSO P = 0.654) was observed in this analysis. In addition, genetically determined SHL was causally associated with an elevated risk of SD (OR 3.81; 1.09 to 13.37; P = 0.037). No significant association between SIHL and dementia risk was found. Furthermore, no causal link was established between hearing impairment and the risk of overall dementia, PDD, and VaD (P range: 0.210 to 0.911). Details of the MR estimates and sensitivity analyses can be found in Additional file 1 (Tables S3, S5, and S6).

Fig. 1
figure 1

Two-sample Mendelian randomization results of relationship between hearing impairment and its subtypes with dementia and cognitive function. Forest plot of the main inverse variance–weighted analyses for association of genetically predicted conductive and sensorineural hearing loss (CSHL), conductive hearing loss (CHL), and sensorineural hearing loss (SHL) with dementia and cognitive function, and Wald ratio analyses for association of genetically predicted sudden sensorineural hearing loss (SIHL) with dementia and cognitive function. CSHL, conductive and sensorineural hearing loss; CHL, conductive hearing loss; SHL, sensorineural hearing loss; SIHL, sudden sensorineural hearing loss. AD, Alzheimer’s disease; DLB, Lewy body dementia; FTD, frontotemporal dementia; SD, semantic dementia; PNFA, progressive non-fluent aphasia; MND, FTD overlapping with motor neuron disease; PDD, Parkinson’s disease dementia; VaD, vascular dementia

Full size image

Causal effect of hearing impairment on cognitive function

We also investigated the causal effects of four types of hearing impairment on cognitive function. The results of the main IVW analysis shown that genetically predicted CHL was causally associated with lower general cognitive performance (β -0.015; -0.025 to -0.004; P = 0.007) and fluid intelligence score (β -0.045; -0.087 to -0.003; P = 0.037). These results were confirmed by the weighted-median method (respectively: P = 0.005 and P = 0.007). No significant heterogeneity was detected in the association between CHL and general cognitive performance (Q = 4.98, P = 0.418); however, there was evidence of significant heterogeneity in relation to fluid intelligence score (Q = 11.18, P = 0.048). No significant pleiotropy was observed in these analyses (respectively: MR-Egger intercept P = 0.270 and MR-PRESSO P = 0.437; P = 0.422 and P = 0.103). In addition, the Wald ratio analysis suggested that SIHL was also causally associated with lower general cognitive performance (β -0.043; -0.076 to -0.009; P = 0.013) and fluid intelligence score (β -0.095; -0.185 to -0.004; P = 0.040). There were no causal effects of CSHL and SHL on cognition function (P range: 0.405 to 0.978). No causal association was found between hearing impairment and the decline in numeric memory and executive function (P range: 0.216 to 0.918; Fig. 1). Details of the MR estimates and sensitivity analyses can be found in Additional file 1 (Tables S4–S6).

Causal effect of hearing impairment and dementia with adjustment for covariates

To investigate various hypotheses regarding how hearing impairment leads to dementia, we conducted MVMR analysis to examine whether the impact of hearing impairment on dementia depends on specific covariates. In MVMR, based on the IVW method, adjusting for aging (P range: <0.001 to 0.058) and ischemic stroke (P range: 0.004 to 0.028) did not significantly alter the causal effects of CHL on AD, as well as CSHL on DLB and FTD. The causal association of CHL and AD was substantially changed with adjustment for genetically predicted loneliness (OR 0.74; 0.51 to 1.09; P = 0.131) but not depressed mood (OR 1.07; 1.02 to 1.14; P = 0.010). The causal effect of CSHL on DLB and FTD were attenuated to nonsignificant when adjusted for genetically predicted depressed mood (respectively: P = 0.187 and P = 0.096) and loneliness (respectively: P = 0.638 and P = 0.390). It is worth noting that the causal effect of CHL on AD (OR 1.04; 0.96 to 1.12; P = 0.363), and CSHL on DLB (OR 1.44; 0.92 to 2.25; P = 0.111) and FTD (OR 1.23; 0.75 to 2.01; P = 0.413) were all changed with adjustment for genetically predicted cortical volume (all P > 0.10; Table 2; Additional file 1: Table S7). There was evidence of heterogeneity between CHL and AD (Q = 54.13, P = 0.004), as well as CHSL and DLB (Q = 51.75, P = 0.015). We further adjusted for 51 cortical regional volume to investigate which specific brain areas exhibited significantly modified causal relationships between hearing impairment and dementia. For the association between CHL and AD, the potential mediators contained nine brain regions (including left middletemporal, superiorfrontal, lateralorbitofrontal, inferiorparietal, parsorbitalis, and precuneus, and right medialorbitofrontal and postcentral lobe). For the association between CSHL and FTD, three brain regions were involved (including left posteriorcingulate, and right middletemporal and transversetemporal lobe). For the association between CSHL and DLB, 40 brain regions were involved (including bilateral middletemporal lobe, inferiorparietal lobe and so on). Among them, the middletemporal lobe is the only brain region involved in all three types of dementia (all P > 0.10; Fig. 2; Additional file 1: Table S8).

Table 2 MVMR results of the causal association between genetically predicted hearing loss and dementia
Full size table
Fig. 2
figure 2

MVMR results of causal effect of hearing impairment and dementia with adjustment for cortical regional volume. The orange brain areas represent those regions that significantly alter the causal link between hearing impairment and dementia. The red brain areas indicate mediator regions (medial temporal gyrus) involved in all three types of dementia. CSHL, conductive and sensorineural hearing loss; CHL, conductive hearing loss; AD, Alzheimer’s disease; DLB, Lewy body dementia; FTD, frontotemporal dementia

Full size image

Discussion

In this study, we utilized the two-sample MR analysis to investigate the causal effects of hearing impairment and its subtypes on six dementia phenotypes and four cognitive functions. Additionally, we explored potential mechanisms through MVMR analysis. Our study revealed that genetically determined hearing impairment is causally associated with increased risks of dementia and poorer general cognitive function. Specifically, CSHL was associated with increased risks of dementia with DLB and FTD. CHL was associated with increased risk of AD, whereas SHL was linked to heightened risk of SD. CHL and SIHL were both associated with lower general cognitive performance and fluid intelligence scores. Furthermore, our results suggest that the causal relationship between hearing impairment and dementia may be mediated by loneliness, depressed mood, and brain cortical volume, particularly MTG, but not by aging or ischemic stroke.

The evidence supporting the relationship between hearing impairment and dementia primarily derives from observational studies, including evidence on the cognitive benefits of hearing restorative devices [5, 7, 43]. While some studies have rigorously controlled for potential confounding factors, establishing causal inference remains challenging. A meta-analysis of observational studies indicates that hearing impairment is associated with all-cause dementia but not specifically with AD [5]. Furthermore, previous MR studies using GWAS data on self-reported hearing difficulty did not find a causal link between hearing impairment and AD risk [44, 45]. However, our study, employing GWAS data on objectively defined hearing impairment and its subtypes, identified a positive causal association between hearing impairment and various types of dementia, while demonstrating a negative causal relationship with cognitive function. Unlike previous MR studies, our research distinguishes subtypes of hearing impairment and uses non-self-reported data. Compared to subjective and memory-biased self-reported hearing impairment, objectively defined subtypes of hearing impairment (CHL and SHL) may more accurately reflect the actual extent of auditory damage. A meta-analysis has summarized the relationship between hearing impairment and multiple cognitive functions, finding a negative correlation between hearing impairment and global cognition, executive functions, episodic/semantic memory, processing speed, and visuospatial ability [5]. Our results expanded on this evidence and confirmed the causal relationship between hearing impairment and general cognitive dysfunction. The results of this study provide two significant insights: firstly, hearing impairment has the potential to induce different types of dementia and cognitive dysfunction; and secondly, distinct subtypes of hearing impairment may be associated with specific subtypes of dementia.

Of note, there is a lack of observational studies exploring the relationship between different subtypes of peripheral hearing impairment and various types of dementia [46]. It is reported that different subtypes of dementia may exhibit distinct central auditory dysfunction [12]. DLB is characterized by auditory hallucinations and impairments of auditory scene analysis, tone, and rhythm processing have been observed. AD is associated with deficits in auditory scene processing, while FTD is characterized by deficits in rhythm, pitch, timbre perception, and sound detection. Furthermore, pathological studies provide evidence that hearing impairment can lead to different types of dementia, which is consistent with our findings. Firstly, numerous studies have demonstrated a complex relationship between hearing impairment and AD pathology, although the results are inconsistent. Based on positron emission tomography (PET) measurements, increased burdens of β-amyloid (Aβ) and tau are associated with objective measurements of hearing worsening in AD patients [47]. In comparison to AD patients without hearing loss, those with hearing loss exhibit elevated levels of P-tau in cerebrospinal fluid [48]. However, in asymptomatic pre-clinical subjects, no association was found between hearing loss and brain Aβ load [49]. Secondly, there is research indicating that compared to AD patients, patients with DLB have a higher risk of hearing impairment based on pure-tone average assessment [46]. One study (N = 2,755) investigated hearing impairment is associated with neurofibrillary tangle pathology and the deposition of cortical Lewy bodies [50]. However, other research indicates no such association [51]. Recently, another study (N = 442) found that age-related hearing impairment is associated with neuritic plaques and Lewy body pathology, suggesting that hearing impairment may predict increased risk for both DLB and AD, which supports our research [52]. Thirdly, there is limited research on the connection between hearing impairment and FTD. Impaired hearing was found to have an inverse association with FTD-tau, but there was no observed association between impaired hearing and nontauopathy FTD [50]. Overall, our study confirms causal relationships between hearing impairment and various neurodegenerative dementias rather than vascular dementia. Interestingly, we first revealed causal links between CHL and AD, as well as SHL and SD. The mechanisms underlying these subtype differences remain unclear. An animal study found that CHL induced by chronic perforation of the tympanic membrane can aggravate memory impairment and reduce glucose metabolism in the hippocampus in AD mice [53]. Similarly, SHL induced by cochlear ablation may render hippocampal synapses more vulnerable to Aβ-induced damage [54].There is research indicating that both CHL and SHL can cause hippocampal neurodegeneration [55, 56]. Furthermore, SHL can also lead to decreased neurogenesis and axonal growth as assessed by doublecortin, increased tau protein phosphorylation, increased neuroinflammation in the hippocampus, and delayed responses of hippocampal neurons [56]. It’s interesting that patients with primary progressive aphasia exhibit poorer performance on pure-tone audiometry compared to patients with AD [57]. However, patients with AD exhibit more pronounced difficulties in perceiving acoustically degraded speech compared to those with SD [58]. We speculate that primary progressive aphasia may involve more central auditory pathways, which are more susceptible to the effects of abnormal proteins such as TAR DNA binding protein 43 or tau proteins [57]. Future research needs to explore the relationships between different types of hearing loss and risks/pathologies of various dementia, as well as their diagnostic value.

Our study further reveals a causal link between hearing impairment and cognitive function. It’s worth noting that we only found associations between hearing impairment and general cognition as well as fluid intelligence, rather than numerical memory and executive function. General cognitive and fluid intelligence phenotypes may partially contribute to and influence cognitive reserve. Enhancing cognitive reserve is a crucial strategy of preventing dementia [59]. Based on our findings, we hypothesize that hearing impairment may accelerate the depletion of cognitive reserve, which could be evidence supporting the information/sensory degradation hypothesis. In other words, hearing impairment may lead to a decline in cognitive reserve, thereby making patients more susceptible to neurodegenerative dementia, while having weaker connections to the specific/key symptoms of different types of dementia. Previous studies support our findings. One study indicated that a more pronounced rise in auditory thresholds was associated with a swifter deterioration in fluid reasoning [60]. Additionally, observational studies have found an association between hearing impairment and global cognition. However, other specific cognitive symptoms, such as executive functions and episodic/semantic memory, have been also found to be associated with hearing impairment [5]. We do not exclude the possibility that due to the omission of some important instrumental variables in the GWAS of various hearing impairment subtypes we included, no sufficient statistical power was used to discover causal links between hearing and other cognitive symptoms. Further research is still needed to confirm our results.

The exact mechanisms underlying the relationship between hearing impairment and dementia have not been fully elucidated, although several theoretical hypotheses have been extensively discussed. One potential causal pathway explaining the association between hearing impairment and dementia involves social isolation and loneliness. Hearing impairment is associated with increased social isolation among older adults, possibly due to communication barriers. Research has found that these psychosocial issues are associated with an increased risk of cognitive decline [61]. Our research findings support this hypothesis, as we found that social isolation and depressed mood influenced the causal relationship between hearing impairment and dementia. A longitudinal study (N = 3,670) revealed that after adjusting for social isolation and depressive mood, the observed association of self-reported hearing loss with decreased MMSE was no longer statistically significant [62]. Additionnally, a UK Biobank study (N = 82,039) provided some limited, yet not statistically significant, evidence that depressive symptoms and social isolation together mediate the association between hearing impairment and dementia [63]. However, evidence from the National Alzheimer’s Coordinating Center Uniform Data set (N = 8529) suggests that depression accounted for only 6% of the relationship between hearing loss and dementia in a model including hearing loss and time-varying depression, indicating that depression is an independent risk factor rather than a mediator [64]. Another frequently discussed hypothesis is the association between hearing impairment and brain atrophy in relevant brain regions. Research has shown that older adults with hearing impairment experience an accelerated rate of brain volume decline compared to those with normal hearing, particularly in whole-brain volume, temporal lobe or auditory cortex volume, and hippocampal volume [65]. Interestingly, research has proposed the hypothesis that neurodegenerative pathologies target the auditory brain [12]. This theory suggests that the spread of pathogenic proteins in neurodegenerative dementias may selectively affect the auditory network, involving brain regions such as the prefrontal cortex, superior temporal gyrus, MTG, inferior parietal lobe, temporo-parietal junctional cortex, orbitofrontal cortex, Heschl’s gyrus, and medial geniculate body. These regions are associated with precognitive auditory processing, auditory cognition, and general cognitive processes. In our study, we utilized MVMR to indirectly infer the brain regions influencing the relationship between hearing impairment and dementia, and found that many of these regions are implicated in the auditory brain network (Fig. 2). Among them, the MTG may play a crucial role. Recent findings suggest that cognitive reserve deficits observed in individuals with hearing impairment may involve alterations in MTG activation [66]. Additionally, functional connectivity of MTG in patients with hearing impairment may be altered [9, 67]. Further exploration of the mechanisms involving MTG in mediating the relationship between hearing impairment and dementia is warranted. Furthermore, it is noteworthy that we observed that aging and vascular factors did not significantly alter the causal relationship between hearing impairment and dementia. This indirectly support the notion that the association between hearing impairment and dementia may be not merely coincidental but causal.

One of the main strengths of our study is that, compared to previous observational studies, our MR analysis is based on causal inference using instrumental variables to assess the association between hearing impairment and dementia and cognitive function, thereby minimizing the potential bias due to unmeasured confounding. However, there are several limitations to our study. Firstly, some of the cognitive phenotypes we selected may not fully capture the functional domains of interest. For example, only using the maximum digit length recalled correctly to reflect numeric memory has certain limitations, so caution is needed when interpreting our results. Secondly, in our instrumental variable selection, we relaxed the P-value threshold for the CHL phenotype to obtain a sufficient number of instrumental variables, which may raise concerns about the presence of weak instrument bias, although the F-values for the instrumental variables were all greater than 10. Thirdly, we did not explore the mechanisms mediating the association between hearing impairment and cognitive function, considering that covariates and cognitive function GWAS data were both mainly from the UKB cohort. Fourthly, our study primarily relies on publicly available GWAS rather than individual subject data. This means we may not capture individual-specific effects and limits our ability to explore and control for broader confounding factors. Future research will require large-scale cohort studies to further investigate and confirm our findings. Finally, our analysis mainly included GWAS data from participants of European ancestry, which limits the generalizability of our findings to other racial groups.

Conclusions

In conclusion, our MR study provides strong evidence that genetically determined hearing impairment is causally linked to an increased risk of various types of dementia as well as poorer general cognitive function. Our findings suggest that specific subtypes of hearing impairment are associated with increased risks of different types of dementia, and that the relationship between hearing impairment and dementia may be mediated by loneliness, depressed mood, and brain cortical volume (particularly the medial temporal gyrus). These results highlight the importance of early identification and intervention for hearing impairment as a potential preventive measure for dementia and cognitive decline.

Data availability

Genome-wide summary statistics used in this study are available through https://gwas.mrcieu.ac.uk/ (Hearing impairment, Alzheimer’s disease, dementia with Lewy bodies, cognitive function phenotype), https://rdr.ucl.ac.uk/articles/dataset/IFGC_Summary-statistics_Data-sharing/13042166 (Frontotemporal dementia and its subtypes phenotype), https://portal.ide-cam.org.uk/overview/483 (Cortical volume phenotype), and https://open.win.ox.ac.uk/ukbiobank/big40/ (Cortical regional volume phenotype using Desikan-Killiany parcellation).

Abbreviations

AD:

Alzheimer’s disease

DLB:

Lewy body dementia

FTD:

Frontotemporal dementia

PDD:

Parkinson’s disease dementia

VaD:

Vascular dementia

MR:

Mendelian randomization

MVMR:

Multivariable Mendelian randomization

GWAS:

Genome-wide association study

CSHL:

Conductive and sensorineural hearing loss

CHL:

Conductive hearing loss

SHL:

Sensorineural hearing loss

SIHL:

Sudden sensorineural hearing loss

bvFTD:

Behavioral variant frontotemporal dementia

SD:

Semantic dementia

PNFA:

Progressive non-fluent aphasia

FTD-MND:

FTD overlapping with motor neuron disease

UKB:

UK biobank

COGENT:

Cognitive Genomics Consortium

ABCD:

Adolescent Brain Cognitive Development

IVs:

Instrumental variables

SNPs:

Single nucleotide polymorphisms

IVW:

Inverse-variance weighted method

References

  1. Chadha S, Kamenov K, Cieza A. The world report on hearing, 2021. Bull World Health Organ. 2021;99(4):242–A242.

    Article  PubMed  PubMed Central  Google Scholar 

  2. The Lancet Global Health null. Amplifying the global issue of hearing loss. Lancet Glob Health. 2022;10(10):e1360.

    Article  Google Scholar 

  3. Deafness. and hearing loss [Internet]. [cited 2023 Dec 11]. https://www.who.int/news-room/fact-sheets/detail/deafness-and-hearing-loss

  4. Michels TC, Duffy MT, Rogers DJ. Hearing loss in adults: Differential diagnosis and treatment. Am Fam Physician. 2019;100(2):98–108.

    PubMed  Google Scholar 

  5. Loughrey DG, Kelly ME, Kelley GA, Brennan S, Lawlor BA. Association of Age-related hearing loss with cognitive function, cognitive impairment, and dementia: a systematic review and Meta-analysis. JAMA Otolaryngol Head Neck Surg. 2018;144(2):115–26.

    Article  PubMed  Google Scholar 

  6. Livingston G, Huntley J, Sommerlad A, Ames D, Ballard C, Banerjee S, et al. Dementia prevention, intervention, and care: 2020 report of the Lancet Commission. Lancet. 2020;396(10248):413–46.

    Article  PubMed  PubMed Central  Google Scholar 

  7. Bsy Y, Hjjmd S, Ems T, Ls N, Csh H. R H, Association of Hearing Aids and Cochlear Implants With Cognitive Decline and Dementia: A Systematic Review and Meta-analysis. JAMA neurology [Internet]. 2023 Jan 2 [cited 2023 Dec 3];80(2). https://pubmed.ncbi.nlm.nih.gov/36469314/

  8. Wayne RV, Johnsrude IS. A review of causal mechanisms underlying the link between age-related hearing loss and cognitive decline. Ageing Res Rev. 2015;23:154–66.

    Article  PubMed  Google Scholar 

  9. Slade K, Plack CJ, Nuttall HE. The effects of Age-related hearing loss on the brain and cognitive function. Trends Neurosci. 2020;43(10):810–21.

    Article  CAS  PubMed  Google Scholar 

  10. Griffiths TD, Lad M, Kumar S, Holmes E, McMurray B, Maguire EA, et al. How Can Hear Loss Cause Dementia? Neuron. 2020;108(3):401–12.

    CAS  Google Scholar 

  11. Huang AR, Lin FR. Hearing loss and dementia in older adults: a narrative review. J Chin Med Association. 2024;87(3):252.

    Article  Google Scholar 

  12. Johnson JCS, Marshall CR, Weil RS, Bamiou DE, Hardy CJD, Warren JD. Hearing and dementia: from ears to brain. Brain. 2021;144(2):391–401.

    Article  PubMed  Google Scholar 

  13. Larsson SC, Traylor M, Malik R, Dichgans M, Burgess S, Markus HS, et al. Modifiable pathways in Alzheimer’s disease: mendelian randomisation analysis. BMJ. 2017;359:j5375.

    Article  PubMed  PubMed Central  Google Scholar 

  14. Li M, Jiang C, Lai Y, Wang Y, Zhao M, Li S, et al. Genetic Evidence for Causal Association between Atrial Fibrillation and Dementia: a mendelian randomization study. J Am Heart Assoc. 2023;12(16):e029623.

    Article  PubMed  PubMed Central  Google Scholar 

  15. Skrivankova VW, Richmond RC, Woolf BAR, Davies NM, Swanson SA, VanderWeele TJ, et al. Strengthening the reporting of observational studies in epidemiology using mendelian randomisation (STROBE-MR): explanation and elaboration. BMJ. 2021;375:n2233.

    Article  PubMed  PubMed Central  Google Scholar 

  16. Ran S, Wu Y, Liu B. Sarcopenia and Risk of hearing loss: a 2-Sample mendelian randomization study. Am J Med. 2023;136(5):e102–3.

    Article  CAS  PubMed  Google Scholar 

  17. Karczewski KJ, Gupta R, Kanai M, Lu W, Tsuo K, Wang Y et al. Pan-UK Biobank GWAS improves discovery, analysis of genetic architecture, and resolution into ancestry-enriched effects [Internet]. medRxiv; 2024 [cited 2024 Jul 14]. p. 2024.03.13.24303864. https://www.medrxiv.org/content/https://doiorg.publicaciones.saludcastillayleon.es/10.1101/2024.03.13.24303864v1

  18. Kunkle BW, Grenier-Boley B, Sims R, Bis JC, Damotte V, Naj AC, et al. Genetic meta-analysis of diagnosed Alzheimer’s disease identifies new risk loci and implicates Aβ, tau, immunity and lipid processing. Nat Genet. 2019;51(3):414–30.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  19. Ferrari R, Hernandez DG, Nalls MA, Rohrer JD, Ramasamy A, Kwok JBJ, et al. Frontotemporal dementia and its subtypes: a genome-wide association study. Lancet Neurol. 2014;13(7):686–99.

    Article  PubMed  PubMed Central  Google Scholar 

  20. Chia R, Sabir MS, Bandres-Ciga S, Saez-Atienzar S, Reynolds RH, Gustavsson E, et al. Genome sequencing analysis identifies new loci associated with Lewy body dementia and provides insights into its genetic architecture. Nat Genet. 2021;53(3):294–303.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  21. Iwaki H, Blauwendraat C, Leonard HL, Kim JJ, Liu G, Maple-Grødem J, et al. Genomewide association study of Parkinson’s disease clinical biomarkers in 12 longitudinal patients’ cohorts. Mov Disord. 2019;34(12):1839–50.

    Article  PubMed  PubMed Central  Google Scholar 

  22. Lee JJ, Wedow R, Okbay A, Kong E, Maghzian O, Zacher M, et al. Gene discovery and polygenic prediction from a genome-wide association study of educational attainment in 1.1 million individuals. Nat Genet. 2018;50(8):1112–21.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  23. Richard EL, McEvoy LK, Cao SY, Oren E, Alcaraz JE, LaCroix AZ, et al. Biomarkers of kidney function and cognitive ability: a mendelian randomization study. J Neurol Sci. 2021;430:118071.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  24. Cullen B, Nicholl BI, Mackay DF, Martin D, Ul-Haq Z, McIntosh A, et al. Cognitive function and lifetime features of depression and bipolar disorder in a large population sample: cross-sectional study of 143,828 UK Biobank participants. Eur Psychiatry. 2015;30(8):950–8.

    Article  CAS  PubMed  Google Scholar 

  25. Hatoum AS, Morrison CL, Mitchell EC, Lam M, Benca-Bachman CE, Reineberg AE, et al. Genome-Wide Association Study Shows that executive functioning is influenced by GABAergic processes and is a Neurocognitive Genetic Correlate of Psychiatric disorders. Biol Psychiatry. 2023;93(1):59–70.

    Article  CAS  PubMed  Google Scholar 

  26. Wang HF, Zhang W, Rolls ET, Alzheimer’s Disease Neuroimaging Initiative, Li Y, Wang L, et al. Hearing impairment is associated with cognitive decline, brain atrophy and tau pathology. EBioMedicine. 2022;86:104336.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  27. McCartney DL, Min JL, Richmond RC, Lu AT, Sobczyk MK, Davies G, et al. Genome-wide association studies identify 137 genetic loci for DNA methylation biomarkers of aging. Genome Biol. 2021;22(1):194.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  28. Wang M, Yang M, Liang S, Wang N, Wang Y, Sambou ML, et al. Association between sleep traits and biological aging risk: a mendelian randomization study based on 157 227 cases and 179 332 controls. Sleep. 2024;47(3):zsad299.

    Article  PubMed  Google Scholar 

  29. Malik R, Chauhan G, Traylor M, Sargurupremraj M, Okada Y, Mishra A, et al. Multiancestry genome-wide association study of 520,000 subjects identifies 32 loci associated with stroke and stroke subtypes. Nat Genet. 2018;50(4):524–37.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  30. Day FR, Ong KK, Perry JRB. Elucidating the genetic basis of social interaction and isolation. Nat Commun. 2018;9(1):2457.

    Article  PubMed  PubMed Central  Google Scholar 

  31. Ruth Mitchell, Elsworth BL, Mitchell R, Raistrick CA, Paternoster L, Hemani G, Gaunt TR. (2019): MRC IEU UK Biobank GWAS pipeline version 2. https://doiorg.publicaciones.saludcastillayleon.es/10.5523/bris.pnoat8cxo0u52p6ynfaekeigi

  32. Warrier V, Stauffer EM, Huang QQ, Wigdor EM, Slob EAW, Seidlitz J, et al. Genetic insights into human cortical organization and development through genome-wide analyses of 2,347 neuroimaging phenotypes. Nat Genet. 2023;55(9):1483–93.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  33. Smith SM, Douaud G, Chen W, Hanayik T, Alfaro-Almagro F, Sharp K, et al. An expanded set of genome-wide association studies of brain imaging phenotypes in UK Biobank. Nat Neurosci. 2021;24(5):737–45.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  34. Emdin CA, Khera AV, Kathiresan S, Mendelian Randomization. JAMA. 2017;318(19):1925–6.

    Article  PubMed  Google Scholar 

  35. Pierce BL, Ahsan H, Vanderweele TJ. Power and instrument strength requirements for mendelian randomization studies using multiple genetic variants. Int J Epidemiol. 2011;40(3):740–52.

    Article  PubMed  Google Scholar 

  36. Guan B, Wang A, Xu H. Causal associations of remnant cholesterol with cardiometabolic diseases and risk factors: a mendelian randomization analysis. Cardiovasc Diabetol. 2023;22(1):207.

    Article  PubMed  PubMed Central  Google Scholar 

  37. Burgess S, Thompson SG, CRP CHD Genetics Collaboration. Avoiding bias from weak instruments in mendelian randomization studies. Int J Epidemiol. 2011;40(3):755–64.

    Article  PubMed  Google Scholar 

  38. Burgess S, Small DS, Thompson SG. A review of instrumental variable estimators for mendelian randomization. Stat Methods Med Res. 2017;26(5):2333–55.

    Article  PubMed  Google Scholar 

  39. Bowden J, Davey Smith G, Haycock PC, Burgess S. Consistent estimation in mendelian randomization with some Invalid instruments using a weighted median estimator. Genet Epidemiol. 2016;40(4):304–14.

    Article  PubMed  PubMed Central  Google Scholar 

  40. Ooi BNS, Loh H, Ho PJ, Milne RL, Giles G, Gao C, et al. The genetic interplay between body mass index, breast size and breast cancer risk: a mendelian randomization analysis. Int J Epidemiol. 2019;48(3):781–94.

    Article  PubMed  PubMed Central  Google Scholar 

  41. Bowden J, Davey Smith G, Burgess S. Mendelian randomization with invalid instruments: effect estimation and bias detection through Egger regression. Int J Epidemiol. 2015;44(2):512–25.

    Article  PubMed  PubMed Central  Google Scholar 

  42. Seyedsalehi A, Warrier V, Bethlehem RAI, Perry BI, Burgess S, Murray GK. Educational attainment, structural brain reserve and Alzheimer’s disease: a mendelian randomization analysis. Brain. 2023;146(5):2059–74.

    Article  PubMed  Google Scholar 

  43. Ying G, Zhao G, Xu X, Su S, Xie X. Association of age-related hearing loss with cognitive impairment and dementia: an umbrella review. Front Aging Neurosci. 2023;15:1241224.

    Article  PubMed  PubMed Central  Google Scholar 

  44. Mitchell BL, Thorp JG, Evans DM, Nyholt DR, Martin NG, Lupton MK. Exploring the genetic relationship between hearing impairment and Alzheimer’s disease. Alzheimers Dement (Amst). 2020;12(1):e12108.

    PubMed  Google Scholar 

  45. Andrews SJ, Fulton-Howard B, O’Reilly P, Marcora E, Goate AM. Consortium collaborators of the ADG. Causal associations between modifiable risk factors and the Alzheimer’s Phenome. Ann Neurol. 2021;89(1):54–65.

    Article  CAS  PubMed  Google Scholar 

  46. Jung J, Bae SH, Han JH, Kwak SH, Nam GS, Lee PH, et al. Relationship between hearing loss and dementia differs according to the underlying mechanism. J Clin Neurol. 2021;17(2):290–9.

    Article  PubMed  PubMed Central  Google Scholar 

  47. Zheng M, Yan J, Hao W, Ren Y, Zhou M, Wang Y, et al. Worsening hearing was associated with higher β-amyloid and tau burden in age-related hearing loss. Sci Rep. 2022;12(1):10493.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  48. Zhang WJ, Li DN, Lian TH, Guo P, Zhang YN, Li JH, et al. Clinical features and potential mechanisms relating neuropathological biomarkers and blood-brain barrier in patients with Alzheimer’s Disease and hearing loss. Front Aging Neurosci. 2022;14:911028.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  49. Sarant JZ, Harris DC, Busby PA, Fowler C, Fripp J, Masters CL, et al. No influence of age-related hearing loss on brain Amyloid-β. J Alzheimers Dis. 2022;85(1):359–67.

    Article  PubMed  PubMed Central  Google Scholar 

  50. Brenowitz WD, Besser LM, Kukull WA, Keene CD, Glymour MM, Yaffe K. Clinician-judged hearing impairment and associations with neuropathologic burden. Neurology. 2020;95(12):e1640–9.

    Article  PubMed  PubMed Central  Google Scholar 

  51. Neff RM, Jicha G, Westgate PM, Hawk GS, Bush ML, McNulty B. Neuropathological findings of Dementia Associated with subjective hearing loss. Otol Neurotol. 2019;40(9):e883–93.

    Article  PubMed  Google Scholar 

  52. Katanga JA, Hamilton CA, Walker L, Attems J, Thomas AJ. Age-related hearing loss and dementia-related neuropathology: an analysis of the United Kingdom brains for dementia research cohort. Brain Pathol. 2023;33(6):e13188.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  53. Kim JS, Lee HJ, Lee S, Lee HS, Jeong YJ, Son Y, et al. Conductive hearing loss aggravates memory decline in Alzheimer Model mice. Front Neurosci. 2020;14:843.

    Article  PubMed  PubMed Central  Google Scholar 

  54. Chang M, Kim HJ, Mook-Jung I, Oh SH. Hearing loss as a risk factor for cognitive impairment and loss of synapses in the hippocampus. Behav Brain Res. 2019;372:112069.

    Article  PubMed  Google Scholar 

  55. Kurioka T, Mogi S, Yamashita T. Decreasing auditory input induces neurogenesis impairment in the hippocampus. Sci Rep. 2021;11(1):423.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  56. Shen Y, Hu H, Fan C, Wang Q, Zou T, Ye B, et al. Sensorineural hearing loss may lead to dementia-related pathological changes in hippocampal neurons. Neurobiol Dis. 2021;156:105408.

    Article  CAS  PubMed  Google Scholar 

  57. Hardy CJD, Frost C, Sivasathiaseelan H, Johnson JCS, Agustus JL, Bond RL, et al. Findings of impaired hearing in patients with Nonfluent/Agrammatic variant Primary Progressive Aphasia. JAMA Neurol. 2019;76(5):607–11.

    Article  PubMed  PubMed Central  Google Scholar 

  58. Jiang J, Johnson JCS, Requena-Komuro MC, Benhamou E, Sivasathiaseelan H, Chokesuwattanaskul A, et al. Comprehension of acoustically degraded speech in Alzheimer’s disease and primary progressive aphasia. Brain. 2023;146(10):4065–76.

    Article  PubMed  PubMed Central  Google Scholar 

  59. Scherrer JF, Morley JE. Lowering dementia risk and slowing progression of disease: the role of cognitive reserve and cognitive training. Br J Psychiatry. 2021;218(5):252–3.

    Article  PubMed  Google Scholar 

  60. Ghisletta P, Dahle CL, Raz N. Age-related hearing loss, cognitive performance, and metabolic risk in healthy adults: a seven-year longitudinal study. J Gerontol B Psychol Sci Soc Sci. 2022;78(3):409–20.

    Article  PubMed Central  Google Scholar 

  61. Mick P, Kawachi I, Lin FR. The association between hearing loss and social isolation in older adults. Otolaryngol Head Neck Surg. 2014;150(3):378–84.

    Article  PubMed  Google Scholar 

  62. Amieva H, Ouvrard C, Giulioli C, Meillon C, Rullier L, Dartigues JF. Self-reported hearing loss, hearing aids, and Cognitive decline in Elderly adults: a 25-Year study. J Am Geriatr Soc. 2015;63(10):2099–104.

    Article  PubMed  Google Scholar 

  63. Stevenson JS, Clifton L, Kuźma E, Littlejohns TJ. Speech-in-noise hearing impairment is associated with an increased risk of incident dementia in 82,039 UK Biobank participants. Alzheimers Dement. 2022;18(3):445–56.

    Article  PubMed  Google Scholar 

  64. Brewster KK, Hu MC, Zilcha-Mano S, Stein A, Brown PJ, Wall MM, et al. Age-related hearing loss, late-life Depression, and risk for Incident Dementia in older adults. Journals Gerontology: Ser A. 2021;76(5):827–34.

    CAS  Google Scholar 

  65. Lin FR, Ferrucci L, An Y, Goh JO, Doshi J, Metter EJ, et al. Association of hearing impairment with brain volume changes in older adults. NeuroImage. 2014;90:84–92.

    Article  CAS  PubMed  Google Scholar 

  66. Gao M, Feng T, Zhao F, Shen J, Zheng Y, Liang J, et al. Cognitive reserve disorder in age-related hearing loss: cognitive cortical compensatory to auditory perceptual processing. Cereb Cortex. 2023;33(16):9616–26.

    Article  PubMed  Google Scholar 

  67. Li Z, Zhu Q, Geng Z, Song Z, Wang L, Wang Y. Study of functional connectivity in patients with sensorineural hearing loss by using resting-state fMRI. Int J Clin Exp Med. 2015;8(1):569–78.

    PubMed  PubMed Central  Google Scholar 

Download references

Acknowledgements

The authors thank the UK Biobank, FinnGen, the International Genomics of Alzheimer’s Project, the International FTD-Genetics Consortium, the Cognitive Genomics consortium, the Adolescent Brain Cognitive Development, and the Multi-ancestry Genome-Wide Association Study of Stroke for providing GWAS summary data.

Funding

This work was supported by the National Natural Science Foundation of China (82271464).

Author information

Author notes

  1. Deming Jiang and Jiahui Hou contributed equally to this work.

Authors and Affiliations

  1. Department of Neurology, Xuanwu Hospital, Capital Medical University, Changchun Street 45, Beijing, China

    Deming Jiang, Jiahui Hou, Haitian Nan, Ailing Yue, Min Chu, Yihao Wang, Yingtao Wang & Liyong Wu

Authors
  1. Deming Jiang
    View author publications

    You can also search for this author inPubMed Google Scholar

  2. Jiahui Hou
    View author publications

    You can also search for this author inPubMed Google Scholar

  3. Haitian Nan
    View author publications

    You can also search for this author inPubMed Google Scholar

  4. Ailing Yue
    View author publications

    You can also search for this author inPubMed Google Scholar

  5. Min Chu
    View author publications

    You can also search for this author inPubMed Google Scholar

  6. Yihao Wang
    View author publications

    You can also search for this author inPubMed Google Scholar

  7. Yingtao Wang
    View author publications

    You can also search for this author inPubMed Google Scholar

  8. Liyong Wu
    View author publications

    You can also search for this author inPubMed Google Scholar

Contributions

D.M.J. and L.Y.W. contributed to the conception and design of the study; D.M.J. and J.H.H. contributed to the acquisition and analysis of data, drafting the text, and preparing the figures; H.T.N., A.L.Y., M.C., Y.H.W., Y.T.W., and L.Y.W. contributed to the revision of the manuscript. All authors read and approved the final manuscript.

Corresponding author

Correspondence to Liyong Wu.

Ethics declarations

Ethics approval and consent to participate

No individual-level data was utilized in this study. Therefore, no new ethical review board approval was required. The research utilized published studies and consortia that offer publicly accessible summary statistics. The original studies included in this research have obtained approval from their respective ethical review boards, and participants have provided informed consent.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Additional file 1: Table S1:

Additional information on GWAS included in the study. Table S2: The instrumental variables for hearing impairment in MR analysis; Table S3: Associations between hearing loss and dementia from Mendelian randomization analyses; Table S4: Associations between hearing loss and cognition from Mendelian randomization analyses; Table S5: Results of Cochrane’s Q test between hearing loss and dementia and cognition; Table S6: Results of MR-Egger intercept test between hearing loss and dementia and cognition; Table S7: The results of the MVMR analysis (adjusting for aging, ischemic stroke, loneliness, depressed mood, or cortical volume); Table S8: The results of the MVMR analysis (adjusting regional cortical volume).

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Jiang, D., Hou, J., Nan, H. et al. Relationship between hearing impairment and dementia and cognitive function: a Mendelian randomization study. Alz Res Therapy 16, 215 (2024). https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s13195-024-01586-6

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s13195-024-01586-6

Keywords

Alzheimer's Research & Therapy

ISSN: 1758-9193