Introduction
Tinnitus, a chronic condition manifesting as ringing or buzzing sounds, persistently disrupts daily life. It affects approximately 20% of the general population aged 12 and older (Park et al., 2014). Its prevalence is notable among children and adolescents, increasing with age (Park et al., 2014). Tinnitus is divided into subjective, which is undetectable by physicians, and objective, which may originate from adjacent structures such as blood vessels or muscle tremors (Crummer and Hassan, 2004; Henry et al., 2014; Han et al., 2009). Unlike objective tinnitus, the causes of subjective tinnitus (Crummer and Hassan, 2004; Han et al., 2009; Henry et al., 2014) are highly varied. Hearing loss is a primary factor influencing tinnitus (Crummer and Hassan, 2004; Han et al., 2009; Henry et al., 2014). Additionally, subjective tinnitus is associated with depressive moods, physical pain, and sleep quality (Ausland et al., 2021; Hackenberg et al., 2023; Alster et al., 1993).
Recent studies have established a significant association between otologic symptoms and degenerative changes in the brain (Zheng et al., 2022; Bisogno et al., 2021; Cheng et al., 2021; Shulman et al., 2007). Hearing loss can accelerate cognitive impairment by promoting the degeneration of brain function (Zheng et al., 2022; Bisogno et al., 2021). Moreover, these changes can be mitigated by the use of hearing aids (Lin et al., 2023). Lin et al. demonstrated that hearing aids could partially offset cognitive deterioration in a long-term, large-scale cohort study (Lin et al., 2023). Furthermore, a link between degenerative brain changes and tinnitus has also been established by several studies (Chu et al., 2020; Cheng et al., 2021). These studies disclosed that dementia, Parkinsonism, and mild cognitive impairment (MCI) are associated with tinnitus (Chu et al., 2020; Cheng et al., 2021). However, the mechanisms underlying this association remain to be fully elucidated.
Prior research aimed to clarify the identification of tinnitus-associated brain networks. Lee et al. and De Ridder et al. discovered that the default mode network (DMN), central executive network (CEN), and salience network (SN) are implicated in tinnitus, and alterations in these networks can trigger tinnitus (Lee et al., 2022; De Ridder et al., 2022). In addition, activation of the noise-canceling pathway can also provoke tinnitus (Lee et al., 2022; De Ridder et al., 2022). Furthermore, several studies on brain changes in tinnitus patients have shown that tinnitus is linked with alterations in brain structures and metabolism (Lee et al., 2022; De Ridder et al., 2022; Mirz et al., 2000). However, the relationship with other brain degenerative changes, such as β-amyloid accumulation, in tinnitus patients remains unexplored. Although some studies have demonstrated an association between β-amyloid accumulation and hearing loss (Pan et al., 2024; Zheng et al., 2022), they did not evaluate the β-amyloid accumulation pattern in tinnitus patients.
Herein, we aimed to identify the structural and functional brain changes in MCI patients with tinnitus, particularly focusing on degenerative changes including β-amyloid accumulation patterns among MCI patient groups with and without tinnitus in individuals with age-normative hearing levels.
Discussion
We assessed β-amyloid plaque accumulation, brain volume, and metabolism in relation to the presence of tinnitus in MCI patients. Our findings indicated distinct differences in brain pathology, structure, and function between the MCI-T and MCI-NT groups. The MCI-T group exhibited higher β-amyloid plaque deposition particularly in the auditory cortex regions and the temporal lobe. Furthermore, this group demonstrated elevated metabolism in the inferior frontal gyrus, insula, and ACC. Similarly, metabolic activity was higher in the anterior and posterior DMN but reduced in the ECN when compared to the MCI-NT group. The THI scores correlated significantly with the gray matter volume in the insula, ACC, superior frontal gyrus, supplementary motor area, white matter adjacent to the hippocampus, and precentral gyrus. These results highlight structural and functional brain alterations in MCI-T patients, including amyloid plaque deposition in the auditory cortex, increased volume and activity in the DMN, enhanced volume in the SN including the insula, ACC, and parts of the inferior frontal gyrus, and diminished activity in the ECN compared to the MCI-NT group.
β-amyloid plaque contributes to degenerative changes in the brain including neuronal loss (DeTure and Dickson, 2019), with our research demonstrating its deposition in the superior and middle temporal gyrus as well as in the temporal pole regions in the MCI-T group. The affected superior temporal gyrus areas are Brodmann areas 41 and 42, which include the primary auditory cortex, while parts of the Rolandic operculum also belong to the auditory cortex (Khalighinejad et al., 2021; Moerel et al., 2014). Since tinnitus is linked to aberrant auditory cortex activity (Galazyuk et al., 2012; Lee et al., 2022) and β-amyloid plaque deposition can induce abnormal neuronal activity in adjacent cells (Zott et al., 2019; Wood, 2019), the presence of β-amyloid plaque in parts of the auditory cortex may contribute to the onset of tinnitus in these cortical areas.
Additionally, the middle temporal gyrus is associated with semantic memory, visual perception, and multi-sensory integration (Onitsuka et al., 2004; Mesulam, 1998), which previous studies have demonstrated to be diminished in patients with tinnitus (Karimi Boroujeni et al., 2020; Mahoney et al., 2011; Dehmel et al., 2008; Lanting et al., 2010; Spiegel et al., 2015). Moreover, the temporal pole is implicated in a range of functions including language, semantic and auditory processing, visual processing, multimodal sensory integration, and emotion (Herlin et al., 2021). The β-amyloid plaque-deposited areas in the MCI-T group were predominantly in regions associated with semantic processing and visual or auditory sensory functions. Consequently, the pattern and progression of cognitive impairment in the MCI-T group might differ from that in MCI-NT patients, characterized initially by multi-sensory deficits coupled with semantic impairments. Thus, tinnitus in the MCI group might be considered a biomarker indicating a predisposition towards dementia with semantic deficits.
Higher metabolic activity of glucose in the SN, encompassing the ACC, insula, and parts of the inferior frontal gyrus, was evident in the MCI-T group, as demonstrated by our findings (Seeley et al., 2007). Additionally, the volume of these areas exhibited a significant correlation with THI. The SN is linked to rewards, motivation, emotion, and pain, and facilitates the switch between the ECN and (Schimmelpfennig et al., 2023). Abnormally activated SN has been identified as one of the chronic tinnitus-related brain network changes (De Ridder et al., 2011; Lee et al., 2022). Our results support the involvement of SN in tinnitus, consistent with previous studies (De Ridder et al., 2011; Lee et al., 2022).
The volume of the white matter adjacent to the hippocampus and superior frontal gyrus was significantly negatively associated with THI. These regions are crucial for learning and memory, functions that are compromised in Alzheimer’s disease (Boisgueheneuc et al., 2006; Yang et al., 2019; Bird and Burgess, 2008). Moreover, the volume of the precentral gyrus and supplementary motor area correlated with the THI score. Earlier studies have shown that motor functions decline as Alzheimer’s disease progresses (Buchman and Bennett, 2011). Another study indicated that the supplementary motor area is also affected in the early stages of Alzheimer’s disease (Vidoni et al., 2012). Given that the THI scores of the MCI-NT group were considered to be 0, the associations between the volumes of these regions and THI might be attributed to the impairments in these areas within the MCI-NT group, suggesting further advanced cognitive and motor function deterioration.
The regions with heightened metabolic activity in the MCI-T group included the DMN. Comprising the dorsal medial prefrontal cortex, posterior cingulate cortex, precuneus, and angular gyrus, as well as the medial orbitofrontal cortex and anterior cingulate cortex (Yeshurun et al., 2021), the DMN is active during rest, exploration, and unfocused states (Yeshurun et al., 2021). Increased activity of the DMN has been documented in prior studies (Chen et al., 2018a; Chen et al., 2018b). The heightened activity of the DMN accounts for the predominance of tinnitus perception during rest (Chen et al., 2018a; Chen et al., 2018b). Our results align with these earlier findings.
Additionally, our metabolic network analysis showed that the MCI-T group exhibited higher DMN activity and lower ECN activity compared to the MCI-NT group. The ECN, also associated with tinnitus (De Ridder et al., 2022) governs executive functions critical for goal-oriented actions, such as working memory and attention (De Ridder et al., 2022; Vincent et al., 2008). This network comprises the lateral prefrontal cortex, anterior cingulate cortex, and inferior parietal lobule (Vincent et al., 2008). Typically, disruptions involving increased metabolic activity and functional connectivity in these networks have been identified as changes linked to tinnitus in the general population, findings that differ from those of our study (De Ridder et al., 2022; Lee et al., 2022). These discrepancies may result from impairments in and alterations of the connections among the SN, DMN, and ECN in MCI (Xue et al., 2021; Chand et al., 2017). Earlier research indicated that the SN is unable to modulate the DMN or ECN effectively in MCI patients (Xue et al., 2021; Chand et al., 2017). Moreover, the DMN shows elevated activity in conditions like semantic memory decline and tinnitus, (Chen et al., 2018a; Chen et al., 2018b; Gardini et al., 2015) while ECN activity may diminish in MCI patients with impaired executive functions (Liu et al., 2021). Furthermore, the DMN is typically anti-correlated with the ECN (Fox et al., 2005). Thus, the heightened DMN function in the MCI-T group could suppress the ECN, resulting from inadequate modulation by the SN and the strong anti-correlation with the highly active DMN.
In addition, the ICA results indicated that the DMN was negatively correlated with the ECN in the MCI-T group, whereas it showed a positive correlation in the MCI-NT group. These inter-network modifications could stem from alterations in brain networks related to MCI. A preceding study found that increased functional connectivity between the posterior DMN and ECN correlates with greater β-amyloid deposition and reduced working memory (Zhukovsky et al., 2023). In this study, the MCI-NT group, which had a lower CERAD-K score indicating further progression of MCI, exhibited increased connectivity between the posterior DMN and ECN.
The ICA results highlighted that the posterior DMN and posterior visual network were positively correlated in the MCI-T group. Previous research suggested that a positive relationship between the DMN and the visual network reflects prolonged emotional experiences (Xu et al., 2023). Given that some tinnitus patients suffer from emotion-related disorders, such as depression or anxiety (Hackenberg et al., 2023), this positive correlation between the DMN and visual networks might also elucidate the emotional alterations linked to tinnitus.
In addition, the negative correlation between the DMN and SMN appeared to be weaker in the MCI-T group compared to the MCI-NT group. The DMN is typically anti-correlated with the SMN in healthy individuals (Schwarz et al., 2013). The DMN and SMN are important for perception and internal mental activities, respectively (Xie et al., 2024). A previous study reported that a weakened network between the DMN and SMN is present in tinnitus patients, indicating decreased cognitive control and reduced attention (Xie et al., 2024). This study also supports those findings. Consequently, MCI patients with tinnitus may have less attention and cognitive control compared to those without tinnitus.
This study has some limitations. First, the small number of subjects in the experiment group (MCI-T) was a significant constraint. We were unable to enroll enough participants during the COVID-19 era. Additionally, to minimize the effect of covariates such as other coexisting neurocognitive diseases and hearing-related factors, which could influence the brain function, structures, and tinnitus (Noble, 2008; Shinden et al., 2021; Zheng et al., 2022), only tinnitus patients without other neurocognitive diseases and with hearing levels less than 40 dB were included. This resulted in more homogeneous subjects, although the small number of patients was included. To overcome this limitation, we applied nonparametric analysis and used a strict cut-off for significant differences in the various imaging tests. Despite this limitation, our findings consistently demonstrated changes related to tinnitus and THI, which might provide valuable insights into structural and metabolic alterations in the brain cortex and networks.
Additionally, the CERAD-K scores of each group were not equally distributed. Although Alzheimer’s disease was excluded, the degree of MCI might differ between the groups. To adjust for the differences in CERAD-K scores between the groups, we employed multivariable analysis, controlling for CERAD-K as a covariate. This adjustment is an important consideration for future research.
Another limitation of our study was the evaluation of the THI score with a small number of MCI-T patients. The THI score-associated changes of brain volume may reflect the differences between the MCI-T group and the MCI-NT group, rather than gradual changes in each brain region according to the THI score, due to the limited number of subjects. Further studies with a larger MCI-T group may provide a clearer understanding of the THI-associated brain changes.
Since patients with hearing levels greater than 40 dB were not included, this study may not represent the functional or structural changes in chronic tinnitus patients with hearing loss. Further studies evaluating brain changes in chronic tinnitus patients with hearing loss may be necessary.