Age-related hearing loss and dementia

Abstract

Dementia is a devastating disease and global health challenge that is highly prevalent worldwide. A growing body of research has shown an independent association between age-related hearing loss (ARHL) and dementia, identifying ARHL as a compelling potential target in preventive strategies for dementia. However, a causal linkage between ARHL and dementia needs to be investigated before making definitive clinical guidelines and treatment recommendations regarding ARHL as a modifiable risk factor. In this review, we discuss the association between ARHL and dementia, the importance of addressing this finding, as well as common mechanisms (e.g., microvascular disease) and causal mechanisms (e.g., depletion of cognitive reserve and social isolation) that may explain the nature of this relationship. We also highlight future directions for research, including randomized controlled trials, developing high-resolution microvascular imaging, and further refining audiometric testing.

Introduction

Dementia is a debilitating disease and leading public health concern that affects approximately 50 million people worldwide, a number that is projected to increase to 152 million in 2050. The economic burden of dementia to society is staggering, with an estimated worldwide cost of $818 billion in 2015.¹ Given the limited efficacy of targeted pharmacologic treatment modalities,² it is important to focus on reducing modifiable risk factors through preventative measures; evidence shows that successful management of modifiable risk factors can be effective in both delaying or preventing the disease and reducing healthcare costs.³ Moreover, a growing body of research has shown an association between age-related hearing loss (ARHL) and dementia,^4–11 thus identifying ARHL as both a potential modifiable risk factor and possible means to improve clinical outcomes in patients with dementia. This review highlights the link between ARHL and dementia, as well as why further investigation into this relationship is crucial.

What is age-related hearing loss?

Age-related hearing loss, otherwise known as presbycusis (from Greek presbys “old” and akousis “hearing”), is the gradual, progressive hearing loss that occurs in individuals as they grow older.¹² Though the term encompasses all conditions that lead to hearing loss in the elderly, ARHL most commonly presents as a bilateral, symmetrical sensorineural hearing loss that arises from idiopathic degeneration of inner ear structures as a person enters the sixth decade of life.^12,13 The hearing loss first affects higher frequencies (pitches), and is characterized by decreased hearing sensitivity (sounds are quieter) and impaired speech perception (words are distorted). Related to this, individuals have difficulty with noisy environments, conversations, and music appreciation. This can affect participation in social activities.¹³

Hearing loss is highly prevalent in the elderly, and severely undertreated. Approximately two-third of adults older than 70 years are affected by hearing loss¹⁴, yet fewer than 20% of adults with hearing loss obtain treatment (e.g. hearing aids).¹⁵ As the global population is growing and aging, the number of people with hearing loss is growing rapidly. In fact, the World Health Organization (WHO) projects a corresponding rise in the prevalence of hearing loss around the world. Currently, there are approximately 466 million people with disabling hearing loss globally; unless action is taken, this number is projected to increase to 630 million by 2030, and over 900 million by 2050.¹⁶ There is also a significant economic burden that comes with untreated hearing loss – an estimated annual global cost of US$ 750 billion.¹⁷

Age-related hearing loss and dementia

Recognition of hearing loss as a potential risk factor for dementia is a relatively new development, and has not been previously prioritized in the management of patients with or at risk for dementia.¹⁸ Recent results of prospective cohort studies have shown that hearing loss confers an independent risk of incident dementia (i.e. new cases of dementia) in individuals that are cognitively intact, but hearing impaired at baseline.^5–9,11 In fact, the population attributable fraction (PAF) of hearing loss for dementia (i.e., the percentage reduction in incident dementia over a given time if an individual risk factor, such as hearing loss, were completely eliminated) has been found to be 23.0%, which is higher than the PAF of any other individual modifiable risk factor, including depression (10.1%), social isolation (5.9%), smoking (13.9%), hypertension (5.1%), and diabetes (3.2%).¹⁸ The high relative risk and high prevalence of ARHL (both of which account for its high PAF), as well as the fact that it is severely undertreated, easily diagnosed, and treatable (compared to other risk factors such as age and family history) makes it a compelling target in preventive strategies for dementia. However, a causal linkage between ARHL and dementia needs to be established before making definitive recommendations regarding ARHL as a modifiable risk factor.

Potential Mechanisms

The mechanisms underlying the association between ARHL and dementia are unclear. Making causal inferences is difficult for several reasons: the correlational nature of the existing research, the wide variability and subjectivity in how cognition and dementia are assessed, and the limited way hearing is typically assessed in epidemiologic studies (e.g., pure-tone audiometry vs. more extensive measures such as word recognition, temporal processing, and frequency selectivity).¹⁹ There are several plausible mechanisms explaining the hearing-cognition relationship. These can be divided into causal mechanisms (i.e., mediators through which ARHL causes dementia) and common mechanisms (i.e., confounders that affect both ARHL and dementia).²⁰

Causal Mechanisms

Hearing loss may be causally related to dementia. Below are some potential causal pathways that explain why ARHL may mechanistically contribute to dementia.

Depletion of Cognitive Reserve

Research has shown that under conditions where auditory perception is difficult, such as hearing loss, individuals are burdened with a greater cognitive load,^21–23 and thus may more readily exhaust their cognitive reserve. The concept of cognitive reserve postulates that individual differences in the ability to optimize cognitive performance through differential recruitment of brain networks and alternative strategies enable some to better cope with neuropathology (e.g., in Alzheimer’s disease or traumatic brain injury) compared to others.²⁴ This concept of cognitive reserve has been used to explain why the extent of neuropathology seen at autopsy does not always correlate with the clinical manifestations of dementia: individuals with increased cognitive reserve may be more resilient to brain pathology and better suited to stave off clinical expression of the disease.^25,26 Hearing loss appears to have a potential effect on cognitive reserve. Studies have demonstrated that individuals with hearing loss dedicate more neural resources to facilitate auditory processing at the expense of other cognitive processes, such as working memory.^22,23 Loughrey et al. (2018) demonstrated in a meta-analysis that though hearing loss was associated with decline in all cognitive domains of interest, it was more predictive of decline in delayed and semantic memory compared to short-term memory and executive functions (e.g., attention, fluency, reasoning, and working memory).²⁷ This finding dovetails with research demonstrating increased activation of frontal and pre-frontal cortical regions on speech-evoked EEG in post-lingually hearing-impaired adults, reflecting increased recruitment of short-term memory and executive function to aid speech perception; these regions have traditionally been associated with working memory and executive function tasks.^27,28 Loughrey et al. (2018) also found a stronger association of low- to middle-frequency hearing loss with immediate recall and processing speed, which may reflect the natural course of ARHL as it progresses from high to low frequencies as one ages.²⁹ Thus, cognitive resources do not appear to be evenly reallocated in individuals with hearing loss, especially as the disease progresses. Moreover, reallocation of neural resources to facilitate auditory processing and exhaustion of cognitive reserves may lead to earlier clinical presentation of dementia.³⁰ When such depletion of neural resources in adults with hearing loss is accompanied by age-related declines in attention,³¹ working memory,³² and processing speed,³³ one can imagine the difficulties older adults experience on a day-to-day basis as they try to comprehend and retain speech that they struggle to hear.

Social isolation

Another potential causal pathway explaining the association between ARHL and dementia involves social isolation. ARHL confers an increased risk of social isolation in the elderly, which may in turn increase the risk of dementia. Hearing loss is associated with increased social isolation in older adults, likely through impaired communication.^34–36 Promoting a strong social environment for older individuals has been long thought of as an important aspect of healthful aging, and has been shown to potentially reduce risk of mortality and comorbidities such as cardiovascular disease, diabetes, and depression.^37–39 Longitudinal epidemiological^40–42 and neuropathological⁴³ studies have demonstrated that poor social networks, reduced social support, and loneliness increase the risk of incident dementia in individuals who were initially disease-free, even after adjusting for potentially confounding variables. Of note, this risk of dementia associated with hearing loss seemed to only increase at hearing thresholds of 25 dB or more (i.e., a mild degree of hearing loss or beyond); this is considered the threshold at which hearing loss begins to impact verbal communication.^7,44,45 It has been hypothesized that an extensive social network can provide a protective effect from cognitive decline and dementia through emotional and intellectual stimulation, as well as practical support.⁴⁶ Thus, if this causative mechanism is validated, preventing social isolation must be a goal in aural rehabilitation to promote healthy aging and prevent cognitive decline and dementia.

Social isolation is unlikely to be a confounder of the relationship between hearing loss and dementia because in addition to causing cognitive impairment, social isolation would also have to cause hearing loss. If anything, social isolation would result in less sound exposure, and a lower chance of a noise-induced contribution to age-related hearing loss.

Changes in Brain Structure

Another hypothesis is that ARHL may induce detectable changes in brain structure, which may in turn increase the risk of dementia. The most straightforward method to evaluate whether ARHL is associated with structural changes in the brain is by measuring brain volumes on MRI.²⁰ Several cross-sectional studies have previously noted that older adults with hearing loss have decreased volumes in the primary auditory cortex on brain MRI, possibly due to chronically reduced stimulation of this region by degraded auditory signals.^47,48 This relationship between ARHL and reduced brain volumes was strengthened through radiographic evidence from a study of 126 dementia-free patients in the Baltimore Longitudinal Study of Aging.⁴⁹ In this cohort, participants with audiometric hearing loss were found to have an increased rate of brain volume reduction over a 6.4-year follow-up period compared to normal hearing peers, after analyses were adjusted for potential confounders. The size of this effect was equivalent to the difference in rate of brain atrophy seen between individuals with normal cognition and those who develop incident mild cognitive impairment.^49,50 Of note, this accelerated rate of brain volume decline in older adults with ARHL compared to those with normal hearing was observed for the entire brain as well as the right temporal lobe, which houses areas crucial for spoken language processing (i.e., superior temporal, middle temporal, and inferior temporal gyri).⁴⁹ Such structures in the temporal lobe are not only important for spoken language processing, but also for semantic memory, sensory integration, and in early stages of cognitive impairment or early Alzheimer’s disease.^51–54 Thus, it is possible that volume losses in regions responsible for auditory processing may have unforeseen effects on other such cognitive processes dependent on these same regions and potentially predispose individuals to an increased risk of dementia.

Common Mechanisms

It is also possible that ARHL itself does not cause dementia; instead, a common mechanism underlies both dementia and ARHL (i.e., a confounder). Several examples are listed below.

Microvascular disease

There is a strongly held hypothesis that vascular degeneration can play a role in ARHL.^13,55,56 Microvascular disease may induce ischemic injury to inner ear structures, i.e., through insufficient cochlear blood flow,⁵⁷ just as cerebral small vessel disease may contribute to decreased cognitive function and even fully developed dementia by means of reduced blood supply to the brain and inadequate distribution of oxygen, glucose and other nutrients.^58,59 Previous studies have implicated insufficient cochlear blood flow with hearing loss, both in the acute (e.g., cochlear ischemia, subsequent reperfusion and reactive oxygen species formation^60,61 resulting in sudden sensorineural hearing loss⁶²) and in the chronic (i.e., ARHL^57,63) setting. Moreover, studies in animals and in post-mortem human temporal bones have specifically pointed towards abnormalities of the stria vascularis as a possible cause of ARHL.^64–67 The stria vascularis is a highly specialized, complex epithelial structure in the cochlea containing multiple capillary networks that help produce and regulate endolymph (an inner ear fluid) through secretion of potassium ions. This helps generates the endolymphatic potential (a positive voltage of approximately 80 mV across the cochlear endolymphatic spaces) that is essential for normal cochlear function.⁶⁸ Outer and inner hair cells rely on potassium influx for depolarization, mechanical shortening (in outer hair cells), as well as auditory transduction (in inner hair cells) necessary for normal hearing.⁶⁹ If these capillary networks are not supplied with blood, the endolymphatic potential drops and potassium inflow is disrupted; thus, cochlear ischemia can potentially result in dramatic impairment of cochlear function and loss of hearing.⁵⁷ Given that normal blood supply to the cochlea is essential for normal hearing function (i.e., through sustaining the endolymphatic potential and ensuring the delivery of oxygen, glucose, and other nutrients), developing novel tools to measure cochlear blood flow to further investigate pathological mechanisms in which microvascular disease may contribute towards ARHL, can help further elucidate the nature of the relationship between ARHL and dementia.

However, several of the studies investigating the relationship between ARHL and dementia have found an association exists even after adjusting for vascular disease risk factors, such as diabetes, smoking, and hypertension.^7–9,11 Previous studies, including those of temporal bone histology, have classically shown that “pure” strial presbycusis (ARHL due to degeneration of the stria vascularis) is characterized by progressive hearing loss across all frequencies and/or low-frequency hearing loss.^13,56,70 Moreover, cardiovascular disease has been found to be associated with low-frequency hearing loss.^63,71 These data seemingly contradicts the audiometric pattern seen most commonly in ARHL: a progressive, bilateral high-frequency hearing loss. That said, the validity of classifying ARHL hearing loss into different subtypes (e.g., strial, sensory, neural), has been questioned, since no single histopathologic finding (e.g., degeneration of the stria vascularis, spiral ganglion cells, hair cells) can wholly account for the variable clinical presentation of ARHL.^63,72,73

Other common neuropathologic processes

It is also possible that another common neuropathologic process (e.g., similar to neurofibrillary tangles or amyloid plaques in Alzheimer’s disease) affects both the brain and peripheral structures necessary for hearing (i.e., the cochlea), causing both dementia and ARHL. However, no such process has ever been identified.^74,75

Additional considerations

We must consider the possibility that the mechanism explaining the association between ARHL and dementia is multifactorial and potentially involving several of the aforementioned mechanisms. It is also possible that there may be an overdiagnosis of dementia in individuals with hearing loss, or, conversely, an overdiagnosis of hearing loss in individuals with dementia. Overdiagnosis of dementia in individuals with hearing loss is unlikely, as a consensus panel established the diagnostic criteria for incident dementia in these studies.^4–11 In addition, cognitive testing typically occurs in a quiet environment face-to-face with a trained examiner. Unless an individual had severe hearing loss, this situation would not result in, for example, incorrectly heard instructions on cognitive testing.⁷⁶ Moreover, individuals participating in prospective studies have had normal cognition at baseline.^5–11

Overdiagnosis of hearing loss in those with dementia is also unlikely. There is no substantial evidence suggesting that cognitive impairment or dementia affects the reliability of routine audiometric testing, as pure-tone audiometry (the standard measure for hearing in these studies) relies on cochlear transduction and neuronal afferents to brainstem nuclei and primary auditory cortex; it does not require higher level cortical processing typically affected by dementias such as Alzheimer’s Disease.⁷⁷ For example, even a 5 year old can accurately undergo pure tone audiometry.

Furthermore, ARHL is not the only sensory impairment shown to be associated with cognitive decline. Both visual and olfactory impairment have also both been independently associated with cognitive impairment.^78,79 Visual impairment may be analogous to hearing impairment in the way it is related to cognitive impairment. Both hearing and visual impairment may lead to social isolation and reduction of cognitively stimulating information, thus contributing to cognitive impairment and possibly eventual dementia. On the other hand, olfactory loss may have a relationship with cognitive decline and dementia in the opposite direction. Although studies have demonstrated evidence for an association of olfactory dysfunction with cognitive decline and dementia, olfactory loss may be a non-causal marker or a sequela of such neurodegenerative conditions.^80–83 In fact, studies have demonstrated pathology such as deposition of tau neurofibrillary tangles in the olfactory bulb and throughout the olfactory system in individuals with Alzheimer disease and other neurodegenerative disorders; the levels of tau deposition in such regions have been found to be associated with the level of olfactory loss.^84–86 In contrast, Alzheimer’s pathology has never been seen in the peripheral auditory system.

Future Direction

As more and more research points towards an association between ARHL and dementia, we must investigate a causal link between ARHL and dementia prior to making any definitive treatment recommendations. Though preliminary prospective studies have demonstrated beneficial effects of hearing loss treatment (e.g., hearing aids or cochlear implants) on cognitive function (e.g., a lower rate of progression to dementia⁸⁷ or an improvement in cognitive abilities^87–89) randomized clinical trials are needed to support causation. The Aging and Cognitive Health Evaluation in Elders Pilot (ACHIEVE-P) (n=40) demonstrated feasibility for the full scale ACHIEVE trial (n=850, recruitment starting November 2017) as the first randomized controlled trial to determine the effectiveness of a best practices hearing intervention (compared to a successful aging control) on reducing cognitive decline and preventing dementia.⁹⁰ Moreover, employing audiometric testing measures other than pure-tone audiometry, such as word recognition scores (which have predictive value on how well one can potentially benefit from hearing aids) and measures that assess central auditory processing (which pure-tone audiometry fails to assess), may also help refine future studies by identifying who are appropriate candidates for such randomized controlled trials that employ a hearing intervention. Developing neuroimaging modalities (e.g., more powerful MR imaging) that can better delineate inner ear vasculature with higher resolution may aid us in understanding if microvascular changes in the ear (i.e., similar to those known to contribute to cognitive deficits and dementia) play a role in hearing loss. More sophisticated neuroimaging modalities investigating cortical changes and neurocognitive compensation before and after hearing loss treatment (i.e., hearing aid use) in speech tasks may also help shed light on a causal mechanism.

Conclusion

Dementia has been described as the greatest global challenge for health and social care in the 21^st century. Engaging in preventive strategies through reduction of modifiable risk factors can be effective in delaying or preventing the disease, reducing healthcare costs, and stemming the immense burden on relatives and supporters of disease sufferers.¹⁸ A growing body of research has shown that hearing loss confers an independent risk of dementia^5–11 and the highest population attributable fraction for dementia compared to any other individual modifiable risk factor (e.g., physical inactivity, depression).¹⁸ Given that ARHL is highly prevalent, severely undertreated, easily diagnosed, treatable, and precedes the onset of dementia by 5 to 10 years, it is an ideal modifiable risk factor that can be targeted as a preventative strategy for managing dementia.^15,91` However, further research is necessary to investigate potential causal mechanisms underlying the relationship between ARHL and dementia, as well as the cognitive benefits of a hearing intervention in elderly adults with hearing loss.

Figure 1.

Age-related hearing loss and dementia: a conceptual model