Subjective tinnitus ("tinnitus") is the perception of sound in the ears or head in the absence of an external sound and difficult to treat. Individuals with tinnitus can experience severe emotional distress, depression, anxiety, and insomnia [1
]. A recent study in 14,278 adults reported an overall prevalence of 25.3% for any experience of tinnitus in the previous year and 7.9% for frequent or constant (at least once a day) tinnitus [6
]. Prevalence increases with age, peaking at 31.4% and 14.3% from age 60 to 69 years for these two tinnitus frequencies, respectively [6
]. The increasing prevalence with age is not surprising, because hearing loss is known to be an associated risk factor for tinnitus [7
]. With increasing life expectancy, and because hearing loss and noise exposure are increasingly affecting military personnel [8
] and youth [10
], tinnitus has become a significant public health issue.
Hearing loss predicts tinnitus presence, but not severity [11
]. Conversely, individuals with hearing loss do not necessarily experience tinnitus. There is therefore a need to determine other factors for this debilitating hearing disorder and its consequences for health in order to better prevent and treat it. One likely candidate is stress. Because stress has long been identified as a trigger or co-morbidity of tinnitus, based mainly on anecdotal and retrospective reports, this idea has been taken for granted in classical teachings on tinnitus [13
]. In addition, recent large population studies have established that emotional exhaustion and long-term stress are predictors of hearing disorders, including tinnitus [14
]. Functional and electroencephalographic brain imaging studies have also shown aberrant links between limbic (involved in emotions) and auditory system structures [16
]. Structural brain differences (i.e., grey matter decrease) in tinnitus involving parts of the limbic system have also been reported. More specifically, less grey matter in the nucleus accumbens [18
] and the left hippocampus [20
] suggests a depletion that could be related to long-term exposure to stress, among other factors.
Another line of research has focused on the hypothalamus-pituitary-adrenal (HPA) axis functioning responsible for the stress response via the stress hormone cortisol. In a first study, overall or chronic basal cortisol levels (secreted naturally in a circadian pattern) were higher in a subsample of tinnitus participants when levels were considered over a one-week period, although diurnal levels were similar to those of age-matched controls [21
]. In a further study [23
], tinnitus participants were submitted to the Trier Social Stress Test [24
]. They showed delayed and blunted cortisol response to the stressor despite similar psychological stress levels to age-matched controls. This response is similar to that of patients with chronic fatigue syndrome [25
], suggesting an exhausted stress response due to long-term stress in tinnitus participants. The apparent contradiction between these two studies could be explained by the fact that basal cortisol levels and stress responsiveness are modulated by two distinct feedback systems. Circulating glucocorticoids are released by the HPA axis and bind with two kinds of receptors: the high-affinity mineralocorticoid receptor (MR) and the lower-affinity glucocorticoid receptor (GR). The HPA axis is a closed-loop system that is subjected to a tight negative feedback control mediated by these two receptor types. HPA axis tone, assessed in basal cortisol levels, is regulated by the MR receptors [26
]. Stress responsiveness is determined by the GR receptors, which are more critical for terminating the HPA axis stress response, and are located in many brain areas such as the hypothalamus, brain stem, hippocampus, amygdala, and pituitary gland, as well as the inner ear.
A noninvasive way to test for exhausted HPA axis hypothesis in tinnitus participants is to examine the sensitivity of the HPA axis negative feedback response to glucocorticoids. The Dexamethasone (DEX) suppression test is a pharmacological challenge that is widely used to test for HPA axis dysregulation in clinical populations such as patients with depression or post-traumatic stress disorder. Dexamethasone is a synthetic glucocorticoid with high GR receptor affinity that does not cross the blood-brain barrier [27
]. Because the pituitary gland is located outside the blood-brain barrier, DEX selectively activates the pituitary GR, leaving the pituitary MR and the MR and GR in other brain tissues unaffected [30
]. Once the pituitary GRs are activated, they downregulate cortisol production further down the HPA axis in the adrenal cortex. The DEX suppression test is therefore a direct test for an altered effect of GR activation in the pituitary on cortisol secretion [32
], and it indicates the sensitivity of the HPA axis negative feedback response to glucocorticoids. Depressed patients often show HPA axis hyperactivity and nonsuppression of HPA axis cortisol secretion after DEX administration [33
]. In contrast, patients suffering from post-traumatic stress disorder often display cortisol hyper
suppression. Hypersuppression is detected by using a lower dose of DEX (0.5 mg instead of 1 mg) to better discriminate HPA axis feedback sensitivity between patients and controls [34
In the present study, both basal cortisol and HPA axis response to the low-dose DEX test were measured in tinnitus participants and controls comparable in age, education, and overall health status. By assessing MR-mediated (basal) as well as GR-mediated (cortisol suppression after DEX administration) feedback in the same participants, both feedback systems were assessed simultaneously to gain a more global insight into HPA axis anomalies in tinnitus participants. If tinnitus participants display greater sensitivity to HPA axis negative feedback (GR-mediated), they should display hypersuppression after DEX administration compared to age-matched controls, despite normal basal (MR-mediated) cortisol levels.
In addition, hearing thresholds were assessed before and after pharmacological challenge to examine the effects of cortisol manipulation on both detection and discomfort thresholds. Glucocorticoid receptors (GR) have been found in abundance in the human inner ear [35
], but their function remains unclear. Although no studies have examined the effects of experimental manipulation of cortisol suppression
on hearing detection thresholds in humans, there is some evidence that cortisol increase
exerts a direct influence on hearing. For instance, patients with adrenal cortical insufficiency (a quasi-total absence of cortisol secretion, such as in Addison's disease) had more acute auditory detection sensitivity and lower discomfort threshold than matched controls [36
]. When corticosteroid levels were restored to normal via administration of exogenous glucocorticoids, auditory measures reverted to normal. This effect has been replicated in rats [37
]. Experimentally increased cortisol concentrations in normal adults have resulted in reduced auditory sensitivity at high frequencies [38
]. The opposite effect was recently reported in rats, however, although the cortisol increase was induced by a stressful stimulus and not cortisol administration: rats exposed to a rodent acoustic repellent showed higher cortisol levels but lower hearing thresholds [39
]. To our knowledge, the effects of cortisol manipulation on hearing discomfort thresholds have never been assessed in human participants with tinnitus. Yet, it is estimated that increased hearing sensitivity is present in 80% of patients with tinnitus [40
]. Discomfort thresholds have also been found to predict tinnitus prevalence and severity in the general population [12
]. Based on human studies, it was thus hypothesized that detection and discomfort thresholds in both tinnitus and control participants would be lower after cortisol suppression, and possibly to a greater extent in tinnitus than in control ears due to their greater sensitivity to cortisol manipulation.