Age-related hearing impairment and frailty in Alzheimer's disease: interconnected associations and mechanisms

Among potentially modifiable age-related conditions linked to dementia, Alzheimer's disease (AD), and late-life cognitive disorders, age-related hearing impairment (ARHI) or presbycusis is the most widely diffused sensory disorder and one of the principal causes of chronic disability in older adults (Gates and Mills, 2005). The impairments of peripheral (sensory or strial) and central (predominantly neural) auditory pathways, diagnosed with different procedures, are often variously imbricated in determining ARHI, with mixed clinical findings (Gates and Mills, 2005). A growing body of epidemiological evidence linking ARHI with late-life cognitive disorders (Panza et al., 2015a) suggested the potential for correcting hearing loss so that elders can function better also from a cognitive point of view with appropriate treatment. 
 
ARHI is also a substantial marker for frailty in older age, another age-related clinical condition for identifying older persons at elevated risk for numerous adverse health outcomes such as falls, institutionalization, hospitalization, disability, and death (Rodriguez-Manas, 2013). Frailty is as a multidimensional syndrome characterized by a nonspecific state of vulnerability, reduced multisystem physiological reserve, and decreased resistance to stressors (Rodriguez-Manas, 2013). Although there is no consensus regarding the operational definition of frailty, in general, two are the most frequently used approaches: the first is the physical or “phenotypic” model of frailty, while the second is based on deficit accumulation, measured with the so called frailty indexes, and defined as an accumulation of health-related deficits and disorders (Rodriguez-Manas, 2013). However, also psychological, cognitive and social factors are part of this multidimensional syndrome, with great influence on its definition and treatment. Cognition has already been suggested as a possible component of frailty with increased risk of adverse outcomes. Therefore, the prevention of cognitive-related adverse outcomes including delirium (Eeles et al., 2012) and late-life cognitive disorders (Robertson et al., 2013; Panza et al., 2015b) may be possible also through frailty prevention.

Among potentially modifiable age-related conditions linked to dementia, Alzheimer's disease (AD), and late-life cognitive disorders, age-related hearing impairment (ARHI) or presbycusis is the most widely diffused sensory disorder and one of the principal causes of chronic disability in older adults (Gates and Mills, 2005). The impairments of peripheral (sensory or strial) and central (predominantly neural) auditory pathways, diagnosed with different procedures, are often variously imbricated in determining ARHI, with mixed clinical findings (Gates and Mills, 2005). A growing body of epidemiological evidence linking ARHI with late-life cognitive disorders (Panza et al., 2015a) suggested the potential for correcting hearing loss so that elders can function better also from a cognitive point of view with appropriate treatment.
ARHI is also a substantial marker for frailty in older age, another age-related clinical condition for identifying older persons at elevated risk for numerous adverse health outcomes such as falls, institutionalization, hospitalization, disability, and death (Rodríguez-Mañas, 2013). Frailty is as a multidimensional syndrome characterized by a nonspecific state of vulnerability, reduced multisystem physiological reserve, and decreased resistance to stressors (Rodríguez-Mañas, 2013). Although there is no consensus regarding the operational definition of frailty, in general, two are the most frequently used approaches: the first is the physical or "phenotypic" model of frailty, while the second is based on deficit accumulation, measured with the so called frailty indexes, and defined as an accumulation of health-related deficits and disorders (Rodríguez-Mañas, 2013). However, also psychological, cognitive and social factors are part of this multidimensional syndrome, with great influence on its definition and treatment. Cognition has already been suggested as a possible component of frailty with increased risk of adverse outcomes. Therefore, the prevention of cognitive-related adverse outcomes including delirium (Eeles et al., 2012) and latelife cognitive disorders (Robertson et al., 2013;Panza et al., 2015b) may be possible also through frailty prevention.

Peripheral Age-related Hearing Impairment, Alzheimer's Disease, and Cognition
In the USA, ARHI is even more prevalent among older adults than previously reported, with the ARHI prevalence that approximately doubles every decade of life from the second through to the seventh decade (Quaranta et al., 2015). At early stage, ARHI typically affects audibility of the higher frequencies (6000 and 8000 Hz), interfering with regular speech comprehension in both quiet and noise, and spreading to the mid and low frequencies over time. Part of the hearing problems are not related to the peripheral deficit of the auditory system but to the central auditory processing (CAP) dysfunction. Subjects with this condition have considerable difficulty in understanding speech in presence of a background noise or in reverberant rooms, but no problem in a quiet environment. Both peripheral and central auditory dysfunctions are therefore relevant to assess a possible influence of ARHI on late-life cognitive disorders.
Over two decades ago, a review article with also original findings summarized the first 25 years of research on possible associations between hearing impairment and cognitive dysfunction reviewing a relevant number of reports (Gennis et al., 1991). The cumulative evidence coming from this first series of studies suggested a strong link between peripheral ARHI and cognitive impairment or decline in demented or institutionalized patients, while there was no association in nondemented older subjects (Gennis et al., 1991). More recently, a case-control report on AD patients from a tertiary center (Gold et al., 1996) and a number of cross-sectional (Gussekloo et al., 2005;Lin, 2011) and longitudinal population-based studies (Lin et al., 2004(Lin et al., , 2013Valentijn et al., 2005;Wallhagen et al., 2008;Gallacher et al., 2012;Kiely et al., 2012) (Table 1) confirmed the interplay among peripheral auditory dysfunction and AD, dementia, and cognitive disorders in older age, with a single exception in a 2-year follow-up of the Australian Longitudinal Study of Aging (Anstey et al., 2001). However, a weak association between hearing loss and memory decline was found in a re-evaluation of the same sample in a longer follow-up of 8 years (Anstey et al., 2003) (Table 1).

Central Auditory Processing Dysfunction, Mild Cognitive Impairment, and Alzheimer's Disease
Age-related decline in CAP appeared not to be an isolated condition, but an entity with a multifactorial nature associated to age-and/or disease-related brain and auditory changes (Gates and Mills, 2005). In fact, CAP dysfunction increases with age (Gates and Mills, 2005), but given the increased incidence of both peripheral ARHI and cognitive impairment in late life, the interpretation of central auditory tests may be difficult. In particular, in the early phase of cognitive decline, the differential impact of the two types of auditory deficits on late-life cognitive disorders may be not easy to determine. Several existing studies mostly focused on the impact of peripheral auditory deficit on late-life cognition, while studies on the link between CAP dysfunction and mild cognitive impairment (MCI) or AD are sparse (Gennis et al., 1991;Kurylo et al., 1993;Gates et al., 1996Gates et al., , 2002Gates et al., , 2008Gates et al., , 2010Gates et al., , 2011Idrizbegovic et al., 2011Idrizbegovic et al., , 2013. Several cross-sectional case-control (Kurylo et al., 1993;Idrizbegovic et al., 2011) and two population-based studies (Gates et al., 2008(Gates et al., , 2010 suggested a strong involvement of CAP dysfunction in MCI (Idrizbegovic et al., 2011), dementia (Gates et al., 2010), and AD (Kurylo et al., 1993;Gates et al., 2008;Idrizbegovic et al., 2011). In particular, findings from the Adult Changes in Thought (ACT) Study suggested that CAP disorders may be associated with executive dysfunction in older subject with and without memory loss and dementia (Gates et al., 2010). Therefore, executive control function may be a key factor in the speech-based behavioral tasks evaluating CAP dysfunction because lesions of the central auditory pathway are infrequent in early AD (Kurylo et al., 1993).
A few longitudinal case-control (Idrizbegovic et al., 2013) and population-based studies (Gates et al., 1996(Gates et al., , 2002(Gates et al., , 2011 suggested that CAP dysfunction in ARHI may be central in determining an increased risk of cognitive decline and incident dementia or AD (Gates et al., 1996), and AD (Gates et al., 2002(Gates et al., , 2011 (Table 1). Therefore, deficit in CAP could be an early marker of MCI or AD, with a "gradient" existing in CAP disorders among subjects with subjective memory complaints, MCI, and early AD (Idrizbegovic et al., 2011).

Age-related Hearing Impairment-dementia Link and Frailty
Some factors may be involved in causal mechanistic pathways linking ARHI and cognition, while other factors may constitute shared pathological processes or etiological pathways underlying both ARHI and cognitive disorders in late life. Cognitive testing may be confounded by ARHI in association with poor verbal communication. However, some studies used also nonverbal cognitive tests or were insensitive to the exclusion of subjects with serious hearing loss from the analyses (Anstey et al., 2003;Lin et al., 2013), so not supporting the hypothesis that miscommunication in hearing loss should impair cognitive testing. Furthermore, in older subjects with subclinical cognitive impairment may occur an overdiagnosis of peripheral ARHI, although audiometric testing appeared to be reliable also in patients with early dementia.
Evidence coming from epidemiological (Fratiglioni et al., 2000;Wilson et al., 2007) and neuropathological studies (Bennett et al., 2006) suggested that social isolation and loneliness, caused by communication impairments in older subjects with ARHI, may lead to cognitive decline and AD. Moreover, also the cognitive reserve concept, often conditioned by communication defects, may account for the link between ARHI and cognition. Cognitive/brain reserve appears to be a buffer against the functional impairments caused by accumulating age-related brain changes or AD-related pathology, so acting as a modulator of the interplay between neuropathology and cognitive outcomes. Increased cognitive load to help compensate auditory processing Frontiers in Aging Neuroscience | www.frontiersin.org may reduce the neural resources available to other cognitive processes such as working memory and perceptual speed, increasing the deleterious effects of AD pathology and revealing the earlier clinical symptoms of dementia (Boyle et al., 2008). Furthermore, a key link between communication difficulties, social isolation, and cognitive decline is the reduced capacity to participate in mentally stimulating activities. In fact, cognitive reserve acts as a buffer via engaging in cognitively stimulating behaviors within an enriched environment, so enhancing neuroplasticity, Furthermore, while AD-related neuropathology is absent in the peripheral auditory pathways (Sinha et al., 1993), peripheral ARHI may contribute to loss of gray matter volume in primary auditory cortex (Peelle et al., 2011), accelerated rates of decline in regional brain volumes in the right temporal lobe and whole brain volume (Lin et al., 2014), and variation in the integrity of central auditory white matter tracks (Chang et al., 2004). AD-related neurodegeneration may be also involved in peculiar damage of central auditory nuclei required for higherorder auditory processing (Parvizi et al., 2001). However, more serious effects may be caused by damage to higher-order cortical areas involved in language processing (Kurylo et al., 1993), so not excluding a shared neuropathological origin underlying both ARHI and AD/dementia. Among possible confounders, common pathological processes, or shared etiological pathways linking ARHI and cognition, frailty syndrome could have a central role. In fact, ARHI may be also a strong indicator for both the most widely diffused models of frailty. In some operational definitions and frailty indexes, ARHI is one of the suggested components of frailty (Frailty Index-Comprehensive Geriatric Assessment, Groningen Frailty Indicator, and Puts model) (Panza et al., 2015a) and it may predict functional decline or incident falls in older adults (Lin and Ferrucci, 2012), some of the healthrelated adverse outcomes linked to frailty. Among potentially modifiable risk factors, there is a growing body of evidence about the impact of several operational definitions of frailty on late-life cognitive disorders (Robertson et al., 2013;Panza et al., 2015b). In particular, an international consensus group has recently proposed the clinical label "cognitive frailty" for describing the simultaneous presence of both physical frailty and cognitive impairment in nondemented older individuals (Kelaiditi et al., 2013), representing also a possible precursor of neurodegenerative processes and AD. Epidemiological studies strongly suggested that physical frailty may be associated with incident AD and MCI, nonAD dementias, vascular dementia (VaD), AD-related neuropathology, and cognitive impairment and decline in late life (Robertson et al., 2013;Solfrizzi et al., 2013;Panza et al., 2015b), so also validating cognitive frailty as a new clinical condition. Several factors and diseases associated with physical frailty are also related to cognitive impairment, including nutritional factors, metabolic disorders, inflammatory markers, hormones, diabetes mellitus, congestive heart failure, and stroke (Robertson et al., 2013;Panza et al., 2015b), suggesting an underlying and shared pathogenesis probably linked to vascular determinants. In fact, in 2006, the term "cognitive frailty" was firstly used to indicate a specific state of cognitive vulnerability in MCI and related entities exposed to vascular risk with a consequent elevated progression to dementia (Panza et al., 2006). Physical frailty has been proposed also as a prodromal stage of VaD (Robertson et al., 2013;Solfrizzi et al., 2013). This could be therefore a further common pathway explaining the ARHI-frailty-cognition interplay given that another neurobiological process such as vascular disease or shared vascular factors may cause ARHI, frailty, and dementia.

Conclusions and Future Directions
In recent years, there has been growing attention on the possible correlations between sensorial abnormalities and latelife cognitive disorders. Epidemiological evidence is mainly focused on peripheral auditory disorders and cognition, but in longitudinal population-based studies both peripheral and CAP dysfunctions appear to be associated with incident cognitive impairment and AD and accelerated cognitive decline. While some randomized controlled trials (RCTs) showed improvement in cognitive function or global measures of change in hearingaid users not cognitively impaired (Mulrow et al., 1990) or with dementia (Allen et al., 2003), determining whether treating hearing loss could delay cognitive decline and dementia remains an open issue. In fact, RCTs with more representative cohorts and technology (i.e., digital hearing aids or cochlear implants), longer follow-up periods, and estimating the effects of hearing rehabilitative interventions on cognitive and global functioning have never been performed (Lin, 2012). At present, use of amplification can be an effective tool for minimizing the perceived disability of older adults and reducing the AD caregiver burden by enhancing the communication abilities of the patients (Palmer et al., 1998). However, hearing aids alone could be not enough to properly manage ARHI, and interventions should be broader, incorporating also concerted counseling, environmental accommodations, and rehabilitative hearing training. Cognition and dementia are causally linked to frailty, and ARHI is also a component of frailty included in several operational definitions. Therefore, frailty could have an important impact in the prevention of late-life cognitive disorders, with nutrition and physical exercise as factors potentially affecting frailty status in advanced age (Clegg et al., 2013). Further investigation on the role of vascular risk on the ARHI-frailty-cognition interplay is warranted to better understand causal mechanisms. Overlaps or interactions among the contributing factors could be investigated supplementing speech-based behavioral measures of CAP with nonbehavioral measures based on electrophysiological studies, and structural, spectroscopic, and functional neuroimaging to detect shared neurobiological markers of ARHI and cognitive decline.