Mapping tinnitus-related brain activation: an activation-likelihood estimation metaanalysis of PET studies

In tinnitus, PET and other functional imaging modalities have shown functional changes not only in the auditory cortex but also in nonauditory regions such as the limbic, frontal, and parietal areas. Nonetheless, disparities in task dimension among studies, low statistical power due to small sample size, and the intrinsic uncertainty of a modality that measures activity indirectly limit the comprehensive understanding of the results from PET studies. These difficulties prompted us to undertake a metaanalysis of PET studies on tinnitus using a coordinate-based technique (activation-likelihood estimation) to retrieve the most consistent activation areas across different task dimensions and to compare the results with those from other imaging modalities. We performed 2 activation-likelihood estimation metaanalyses on data from 10 studies with 56 foci in which we examined the contrast between tinnitus individuals and controls and the difference in activation between sound stimuli and resting state in tinnitus individuals. The studies show that the most consistently activated regions in tinnitus subjects, compared with controls, were the left primary and bilateral secondary auditory cortices, left middle and bilateral inferior temporal gyri, left parahippocampal area, left geniculum body, left precuneus, right anterior cingulate cortex, right claustrum, right middle and inferior frontal gyri, and right angular gyrus. The relatively activated area under sound stimuli, compared with resting state, in tinnitus subjects was the secondary auditory cortex. Our study reconfirms the findings of previous quantitative electroencephalography or magnetoencephalography studies because most of the 14 brain areas with significant activation found in our metaanalysis replicate these earlier data. Our results suggest that the areas described in the tinnitus network are solidly replicable regardless of the applied functional imaging technique. This study proves that PET is a useful modality for tinnitus research and solidifies human tinnitus research itself by confirming previously described brain areas involved in the generation and maintenance of tinnitus.