Age-related hearing loss (ARHL) is an important and prevalent disease in the elderly population. According to the World Health Organization (WHO), about 18% of the population over 65 years are affected by disabling hearing loss [1
]. The WHO defines ARHL as a loss of hearing ability of more than 40 dB [1
] in the better ear, measured by pure-tone audiometry.
There is still ongoing debate about the origin and the context of ARHL. In the last two decades a link between age-related hearing loss (ARHL) and dementia and other age-related outcomes has been well established [2
]. Thus, ARHL could be considered as a good marker of the aging process. The context makes it difficult to establish risk factors for this specific outcome in view of the many facets of the aging process and the factors driving or slowing down the process. Therefore, it is not surprising that only a few population studies in the literature have described the factors related to the incidence or prevalence of ARHL [2
]. The establishment of such factors - particularly if modifiable—with sufficient evidence could be an important step forward towards the prevention of ARHL and could potentially allow the repair of early damage to the hearing before it can reach an irreversible state [3
]. Such investigations also need to be seen as a general public health approach not only to understanding and preventing hearing loss but also to tracking the long-term trajectories of age-related chronic diseases.
The etiology of age-related hearing impairment is highly heterogeneous, multifactorial and much debated in modern science. A major hypothesis about aging, and subsequently hearing loss, regards inflammation [4
]. It had already been speculated that inflammation could have a fundamental role in the onset and degeneration of cochlear damage [5
]. In recent years there has been accumulating evidence linking age-related hearing loss with numerous aging disorders such as cerebrovascular diseases (CVDs) [6
] and physical frailty [7
]. The prevalence of both age-related hearing loss and CVD is increased in older adults [8
] and these conditions often coexist, suggesting that there may be some common underlying factors [6
]. It is well known that serum markers of systemic inflammation increase with age and have been associated with cardiovascular outcomes and all-cause mortality [10
]. Preventing systemic inflammation and microcirculation abnormalities could therefore be a public health strategy to lower the incidence of ARHL [2
]. It is well known that systemic inflammation could be decreased or increased by the consumption of peculiar foods [14
] and by specific dietary patterns [15
]. Anti-inflammatory foods rich in Vitamin A, C, and E like vegetables, fresh fruits and nuts has been considered to be the most important foods protective of ARHL [16
]. Likewise, pro-inflammatory foods, like sugar-rich juices, desserts and alcoholic drinks have been shown to increase the risk of onset of ARHL [17
]. The relationship between inflammation processes and age-related hearing loss is well depicted in the biology of inner ear damage both in animal models and in population studies [5
]. Despite this evidence, the association between particular foods or diet is still unclear [16
The aim of this study was therefore to investigate which foods were associated with a presence of ARHL cross-sectionally and retrospectively in a population-based study on elderly subjects located in Southern Italy.
In the present study, we could link pro-inflammatory food groups and some alcoholic beverages with ARHL but not anti-inflamatory food groups. Further, ARHL clearly appeared to be age related but was not linked to classical age related medical and clinical-chemical blood conditions in this study.
Sugary foods, alcohol, and caloric drinks are typically considered paradigms of an unhealthy lifestyle, not only related to diet, but also to smoking and sedentary behavior [29
]. Unhealthy foods may interact with hearing thresholds in many altered pathways, previously well studied in animal models, involving increased inflammation linked to impairments of insulin signaling, abnormal proteostasis, oxidative stress and alterations of the intermediary metabolism in the inner ear [30
In the present study, there was no significant relationship between smoking habit and ARHL, even though several studies have reported an association between hearing loss and smoking [31
]. In particular, a meta-analysis suggested moderate to large associations between smoking and hearing loss [33
]. Our finding could probably be related to the assessment method we used for smoking habit, taking into account only the current smoking status.
Furthermore, in our study sample, the ARHL group had higher, but not significant, mean blood glucose levels, with a slightly higher prevalence of diabetes mellitus. Angiopathy, neuropathy, oxidative stress, and remnants of glycation end products may be the underlying mechanisms linking type 2 diabetes mellitus to ARHL [34
]. A relationship between these two chronic conditions has been confirmed in two different animal model studies [35
ARHL pathophysiology is classically caused by a combination of microvascular and neurodegenerative processes. Early evidence of ARHL physiopathology was suggested by the autoptic studies of Schuknecht and Gacek, who described findings in the human inner ear and compared these to premortem hearing tests [37
]. Although the histological technique was crude compared with today’s standards, these studies provided the opportunity to characterize the hearing in temporal bones with a single pattern of cellular loss. After these seminal studies, two principal pathways have been suggested to explain the development of ARHL: abnormalities in the metabolism of K+ Na+ regulation in the inner ear and the breakdown of the blood–labyrinth barrier due to inflammatory processes [38
]. The first hypothesis has been confirmed especially in animal model studies of induced hearing loss, showing degeneration of the stria vascularis, the highest density area of Na-K pumps in the inner ear [39
]. The stria vascularis is heavily vascularised and has an extremely high metabolic rate. Histopathological studies of aging gerbils have provided strong evidence for vascular involvement in ARHL [41
]. Morphometric analyses of lateral wall preparations contrast-stained to highlight blood vessels have shown losses in the strial capillary area in aged animals [41
]. These findings clearly support the role of microcirculation alterations in ARHL pathways. Moreover, diabetes mellitus and glycemic abnormalities could be related to alterations of microvessel metabolism and may be considered the prominent risk factor for age-related conditions such as ARHL or macular abnormalities [43
]. However, we could not find such relations in our study.
The inflammation hypothesis motivated us to start a systematic analysis regarding the role of food intake in ARHL development. Anti-inflammatory foods are connected with a high antioxidant content and dietary vitamin intake [4
]. Free radical formation in the inner ear is a key mechanism of hearing loss [44
], causing vasoconstriction and therefore death of the inner ear cells. Subsequent reperfusion of cochlear cells further contributes to free radical formation and further cell death, similar to stroke mechanisms. Inflammation is a normal adaptive response aimed at restoring tissue functionality and homeostasis after infection or mechanical tissue injury that could have unintended negative consequences. Previous authors recognized the potentially important role of inflammation in causing age related hearing loss, but there is still not a well defined mechanistic hypothesis on the pathophysiological pathway [45
]. The most recognized one could be related to the mixed effect of vasoactive function (that leads to micro-ischaemic events) and endothelial dysfunction (cell apoptosis induction) of peripheral inflammatory cytokines [46
]. Those pathological events could affect the complex micro-circulation of cochlea, accelerating contextually the senescence of the cytoneural structures (inner hair cells) and the dysfunction of stria vascularis metabolism [46
]. Generally, pro-inflammatory foods, especially sugars, are typically associated not only with an increase of systemic inflammation, but particularly with micro-vessels damage in terms of micro-ischemic events [47
]. The intake of those harmful foods could catalyze those pathological vascular effects, shared with the pathophysiology of age related hearing loss.
Antioxidants such as vitamins, which inhibit the formation of free radicals, may play a specific role in preventing and treating ARHL [4
]. Therefore, several studies have reported a relationship between ARHL and vitamins A [49
], C [51
], and E in humans [52
]. In the present study, ARHL subjects used on average a lower amount of foods that naturally contain more vitamins A, C, and E, i.e., vegetables, fruits and nuts, but the difference was not significant. Lately, the focus on anti-inflammatory diets has been increasing. These diets, of which the Mediterranean diet is the best example, are characterised by a higher intake of vegetables, fruits, whole grains, legumes, nuts, fish, lean meat, dairy, olive oil, and moderate alcohol consumption (red wine) and a very low intake of processed foods with a rich sugar, salt and saturated fat content. This makes the diet a source of high-quality FAs (i.e., omega-3 and omega-9), fibers and complex carbohydrates, vitamins and minerals [53
]. One of the most powerful effects of these diets is to help in the prevention and treatment of non-communicable diseases, such as cardiovascular diseases, type 2 diabetes, hypertension and cancer [54
]. For example, the protective mechanisms of increasing vegetable and fruit intake against cardiovascular disease include decreasing blood pressure, regulation of lipid metabolism and reducing oxidative stress and low-grade inflammation. The high content of antioxidants (flavonoids, vitamin C, Vitamin E, ß-carotene) reduces DNA damaging. The protective effects of fruit and vegetables intake toward type 2 diabetes mainly depend on their rich fiber content, that improves insulin sensitivity. A favorable effect of nuts on cardiovascular health is due to the unique composition of these in monounsaturated fatty acids and polyunsaturated fatty acids, fiber, magnesium, arginine and polyphenols. Possible effects include a reduction of low-grade inflammation, oxidative stress, endothelial dysfunction and an improvement of the lipid profile and insulin resistance [55
]. In Figure 1
are listed the pro-inflammatory and anti-inflammatory foods, specified in accordance with the food groups used for the statistical analysis.
To make a more focused analysis of the effect of food components on age-related hearing loss, we estimated the average amount of micronutrients intake obtained through the diet. We found a significant decrease in the risk of ARHL in subjects with a higher intake of vitamin A but not of the other vitamins. This highlights two major points: (1) the micronutrients assessment method using the FFQ is not the most appropriate to evaluate this kind of information since it cannot directly estimate the presence of retinoic acid (Vitamin A); (2) retinoic acid could have several functions in protection against age-related diseases (including hearing loss), mainly due to its delaying effect on apoptotic processes [56
] and role in the renewal of the inner hair cells [57
Overall, to better understand the impact of pro-inflammatory foods on ARHL, further epidemiological findings are needed.
4.1. Sugar as a Risk Factor for Age-Related Hearing Loss
Sugars, juices and caloric drinks (such as soda or coke) are among the foods with the highest glycemic index (GI) and glycemic load (GL). Dietary GI is commonly used to characterize the postprandial blood glucose response to the consumption of carbohydrates. GL is the product of a food’s GI and total carbohydrate content and represents both the quantity and quality of carbohydrates. High GI and GL diets have been shown to be associated with an increased risk of coronary heart disease, stroke, and type 2 diabetes mellitus [38
]. Cardiovascular diseases have been proposed as a potential risk factor for ARHL, and a few studies have analyzed the association between cardiovascular-related potential risk factors and hearing capacity, including hypertension, diabetes mellitus, smoking history, and coronary heart disease [38
]. Therefore, it is reasonable to hypothesize that diets rich in sugar, sugary fruit juices and caloric drinks could also increase the risk of developing ARHL [17
]. On the other hand, the relationship between carbohydrate consumption and ARHL is thought to be related not to carbohydrates themselves, but to serum triglyceride levels. Consequently, diets rich in carbohydrates, especially sugars and drinks with highly concentrated fructose, such as juices, may lead to high serum triglyceride levels, and hence, be expected to affect auditory function [58
4.2. Alcohol and Age-Related Hearing Loss
ARHL could be influenced by alcohol consumption, involving several underlying mechanisms such as an impairment of the cochlear blood supply, with resulting hypoxia and ischemic damage, oxidative stress and associated mitochondrial dysfunction, a loss of neurosensory cochlear cells, and neurodegeneration of central auditory pathways [59
]. However, a moderate alcohol intake may be protective of cochlear blood flow [61
], promoting cytoprotective and anti-inflammatory mechanisms that strengthen cellular survival pathways, and directly enhance neuroprotective mechanisms that preserve hearing [62
]. By contrast, alcohol intake may also adversely alter central processing of auditory information [59
Otherwise, an increasing body of evidence suggests that long-term moderate alcohol intake may protect against ARHL [64
]. Some cross-sectional analyses reported an inverse association between alcohol consumption and ARHL [66
], although other reports did not confirm this protective effect [63
]. In a prospective study of 870 men and women aged 49 and older, no association was observed between alcohol consumption and the 5-year incidence of measured hearing loss, although cross-sectional analysis demonstrated a significant protective association between the moderate consumption of alcohol and ARHL [64
]. In a prospective study of 26,809 older men, no association between total alcohol consumption and the risk of self-reported hearing loss was found [68
]. It is worth recalling that a high alcohol intake is associated with elevated plasma triglycerides, linked to cardiovascular disease [69
] and, as already pointed out, to auditory function [58
]. Hence, the influence of long-term alcohol consumption on ARHL remains unclear. Given that the relation between alcohol consumption and other health outcomes varies according to the kind of beverage drunk [70
], in the present study we explored the association of wine, beer and spirits consumption with ARHL and our analysis showed that a higher consumption of beer and spirits significantly increased the ARHL risk. Wine may not be hazardous due to the presence of polyphenols, in particular resveratrol in red wine, known for their antioxidant properties [72
4.3. Strengths and Limitations
To the best of our knowledge, the present is the first population-based study to investigate the relationship between inflammatory food intakes and ARHL, in particular using instrumental methods to assess hearing loss [17
]. A particular strength of the study was the utilization of past data that allow to asses relations that could induce ARHL. However, the use of past data is not exactly identical with a longitudinal observation, because the subjects examined at M3 did not include a hearing examination. Otherwise, fortunately, we could use data of the dietary and the clinical laboratory assessments from the past M3 examination. A further limitation is the FFQ as assessment method of diet, since it is memory-based and considered to be prone to measurement error limiting the ability to study the diet-disease relationships [73
]. Despite the reported limitation of this assessment method, FFQs remain the dietary assessment method most commonly used to study dietary patterns and population habits [74
]. Moreover, the results and significance do not agree between the two populations (GreatAGE and M3) (despite are the same osbserverd subjects) because between the two observation periods the eating habits of the surveyed subjects have changed, probably due both to the advancement of the age and to the change in lifestyle and taste of the population in general.