Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

Brain Volume Differences Associated With Hearing Impairment in Adults

Brain Volume Differences Associated With Hearing Impairment in Adults Speech comprehension depends on the successful operation of a network of brain regions. Processing of degraded speech is associated with different patterns of brain activity in comparison with that of high-quality speech. In this exploratory study, we studied whether processing degraded auditory input in daily life because of hearing impairment is associated with differences in brain volume. We compared T1-weighted structural magnetic resonance images of 17 hearing-impaired (HI) adults with those of 17 normal-hearing (NH) controls using a voxel-based morphometry analysis. HI adults were individually matched with NH adults based on age and educational level. Gray and white matter brain volumes were compared between the groups by region-of-interest analyses in structures associated with speech processing, and by whole-brain analyses. The results suggest increased gray matter volume in the right angular gyrus and decreased white matter volume in the left fusiform gyrus in HI listeners as compared with NH ones. In the HI group, there was a significant correlation between hearing acuity and cluster volume of the gray matter cluster in the right angular gyrus. This correlation supports the link between partial hearing loss and altered brain volume. The alterations in volume may reflect the operation of compensatory mechanisms that are related to decoding meaning from degraded auditory input. Keywords hearing loss, structural plasticity, gray matter, white matter, angular gyrus, voxel-based morphometry Date received: 25 January 2017; revised: 8 January 2018; accepted: 22 January 2018 resemblance to the processing of degraded speech by Introduction normally hearing listeners. Speech comprehension in Brain plasticity following early-onset deafness is well documented (see Glick & Sharma, 2017; Merabet & Department of Otolaryngology—Head and Neck Surgery, Section Ear Pascual-Leone, 2010 for reviews). In individuals with & Hearing, VU University Medical Center, Amsterdam, the Netherlands complete hearing loss, cortical brain areas that are nor- Amsterdam Public Health Research Institute, VU University Medical mally responsible for processing auditory input, such as Center, the Netherlands Department of Anatomy & Neurosciences, VU University Medical Center, the primary auditory cortex (Heschl’s gyrus) and the sec- Amsterdam, the Netherlands ondary auditory cortex (planum temporale), are taken Department of Psychiatry, VU University Medical Center, Amsterdam, over by the remaining intact senses. These areas then the Netherlands respond to visual, tactile, and sign-language input (Glick Amsterdam Neuroscience, Amsterdam, the Netherlands Department of Psychology, VU University, Amsterdam, the Netherlands & Sharma, 2017). Particularly, regions in the superior Department of Behavioural Sciences and Learning, Linnaeus Centre temporal cortex that process auditory speech input in HEAD, The Swedish Institute for Disability Research, Linko¨ping University, normal-hearing (NH) individuals respond to visual Sweden speech input in deaf individuals (Merabet & Pascual- Corresponding author: Leone, 2010). Alterations in the brain that underlie the Adriana A. Zekveld, Department of Otolaryngology-Head and Neck adaptive strategies used by individuals with partial hear- Surgery, Section Ear & Hearing and Amsterdam Public Health Research ing loss to understand speech are less studied. Institute, VU University Medical Center, P.O. Box 7057, 1007 MB, One way of studying the processing of speech in hear- Amsterdam, the Netherlands. ing-impaired (HI) listeners is to assume that it bears Email: [email protected] Creative Commons Non Commercial CC BY-NC: This article is distributed under the terms of the Creative Commons Attribution-NonCommercial 4.0 License (http://www.creativecommons.org/licenses/by-nc/4.0/) which permits non-commercial use, reproduction and distribution of the work without further permission provided the original work is attributed as specified on the SAGE and Open Access pages (https://us.sagepub.com/en-us/nam/open-access-at-sage). 2 Trends in Hearing NH adults is thought to rely on a hierarchically orga- associated with gray matter volume in the left auditory nized network of brain regions that minimally includes cortex (Eckert, Cute, Vaden, Kuchinsky, & Dubno, bilateral superior and middle temporal gyri, the left pre- 2012). In contrast, hearing impairment in middle-aged frontal and premotor cortex, and the left inferior tem- adults has been linked to increased gray matter volume poral cortex (Hickok and Poeppel, 2007; Peelle, in the secondary auditory cortex (Brodmann area 22; Johnrude, & Davis, 2010; Rauschecker, 2011). The pro- Boyen, Langers, De Kleine, & Van Dijk, 2013). In add- cessing of degraded speech by NH individuals has been ition, changes in the integrity of white matter tracts in linked to increased activity in bilateral superior temporal pathways leading to the auditory cortex, including the cortices (Binder, Liebenthal, Possing, Medler, & Ward, lateral lemniscus and the anterior thalamic radiation 2004; Davis, Ford, Kherif, & Johnsrude, 2011; Scott, have been reported (Chang et al., 2004; Husain et al., Rosen, Wickham, & Wise, 2004; Wong, Uppunda, 2011; Lin et al., 2008). In sum, these studies mostly Parrish, & Dhar, 2008; Zekveld, Heslenfeld, Festen, & point to altered gray matter volume in the superior tem- Schoonhoven, 2006). Attempts to understand speech in poral cortices, and altered white matter integrity below low signal-to-noise conditions has shown to evoke activ- the primary auditory areas (although, see Profant et al., ity in bilateral anterior insulae and in the opercular part 2014); however, the extent to which the reported alter- of the left inferior frontal gyrus (Adank, Davis, & ations can be attributed to hearing impairment alone is Hagoort, 2012; Binder et al., 2004; Zekveld et al., difficult to determine because of the lack of age-matched 2006). These areas are thought to be engaged in effortful, and NH or HI and younger control groups. articulatory strategies that support the comprehension of Studies in the domain of language and audition sug- severely distorted speech. Similarly, the premotor area gest that structural plasticity is often observed within the and bilateral anterior superior temporal sulci, which same brain regions that functionally underlie the behav- are known to be involved in speech production, are ior at hand (see Golestani, 2014, for a review). Therefore, recruited during the correct perception of distorted a reasonable hypothesis would be that partial hearing speech (Adank, 2012). Lastly, bilateral angular gyri loss is accompanied by gray and white matter volume and left supplementary motor area are associated with alterations in the brain regions that are involved in the listening to distorted speech in an effortful manner comprehension of degraded speech. In this study, we (Kuchinsky et al., 2011; Wild et al., 2012). These findings explored the relationship between brain volume and par- raise the possibility that HI listeners may differentially tial hearing loss. For this, we compared the magnetic recruit regions in the inferior frontal, temporal, or par- resonance images of HI adults with those of age- and ietal areas while listening to speech. educational-level matched NH controls in a voxel- A number of studies have explored functional plasti- based morphometry (VBM) analysis. We hypothesized city following hearing impairment (see Mudar & Husain, that hearing impairment would be associated with alter- 2016 for a review). In elderly listeners, hearing impair- ations in gray and white matter volume in bilateral ment has been linked to decreased processing in the superior temporal sulci, bilateral superior and middle bilateral superior temporal gyri, thalamus, and brain- temporal gyri, the left inferior frontal gyrus, the left pre- stem during the comprehension of linguistically complex central gyrus, insula, angular gyri, and the premotor sentences (Peelle, Troiani, Grossman, & Wingfield, area, and that these alterations would be correlated 2011). In addition, altered processing has been reported with the severity and duration of the participants’ hear- in the default mode and dorsal saliency networks of HI ing impairment. participants in comparison with age-matched controls (Husain, Carpenter-Thompson, & Schmidt, 2014). These studies support the idea that hearing impairment Method may be related to altered functional processing beyond Participants the primary auditory areas, such as the attentional, emo- tional, and cognitive control networks. In total, 17 adults with NH (5 men, 12 women; age Studies exploring structural neuroplasticity following range: 20–62, M¼ 45.88, SD¼ 15.56 years) and 17 mild to moderate hearing loss have mostly focused on adults with hearing impairment (5 men, 12 women; age middle-aged to elderly populations (see Cardin, 2016; range: 20–63, M¼ 45.65, SD¼ 15.66 years) participated Mudar & Husain, 2016 for reviews). Among elderly in the study. HI participants were individually matched people, hearing impairment has been associated with with NH ones based on age and educational level. Of the reduced gray matter volume in the primary auditory 17 pairs, 14 were individually matched based on sex, and cortex (Husain et al., 2011; Peelle et al., 2011) and accel- the ratio of sex was matched between the groups. erated rates of gray matter volume decline in the right Participants with NH were recruited from among the temporal lobe (Lin et al., 2014). Particularly, hearing loss employees and students of the VU University medical in the higher frequency range has been shown to be center (VUmc) and the VU University Amsterdam, the Alfandari et al. 3 Netherlands. They had pure-tone thresholds of maximal severity of hearing impairment did not correlate signifi- 20 dB HL at the octave frequencies between 500 cantly (r¼ 0.42, p¼ .09). Neither duration nor severity and 4000 Hz. The mean pure-tone average (PTA; mean correlated significantly with age (r¼0.46, p¼ .06; hearing-threshold at 1000, 2000, and 4000 Hz, averaged r¼0.19, p¼ .46, respectively). over both ears) of the participants with normal hearing All air-bone gaps were smaller than 10 dB and all was 5.5 dB HL (SD¼ 5.5, range: 5 to 18.3 dB HL). participants had normal tympanograms. All partici- Thresholds at 8000 Hz were on average 17.94 dB HL pants scored better than 80% on each ear on a (SD¼ 16.45, range: 2.5 to 45 dB HL; see Figure 1 for speech audiogram with standard monosyllabic Dutch the average hearing thresholds at the octave frequencies consonant–vowel–consonant word lists (Bosman and between 250 and 8000 Hz). Smoorenburg, 1995). Furthermore, all participants Participants with hearing impairment were recruited were native Dutch speakers who used only spoken from among the patients of the outpatient clinic of the language and no sign language. All were classified as Ear & Hearing section of the Department of right-handed by the Dutch ‘‘Classification of left and Otolaryngology-Head and Neck Surgery of the VUmc. right-handed subjects’’ (van Strien, 1992). They had All participants with hearing impairment had symmet- normal or corrected-to-normal vision, and were screened rical sensorineural hearing loss. For inclusion in the cur- by a near-vision test that is equivalent to the visual acuity rent study, the mean PTA of each ear had to be between Snellen chart (Bailey and Lovie, 1980). Exclusion criteria 35 and 65 dB HL. Also, the asymmetry in the pure-tone were the use of psychotropic medication, a history of a thresholds between both ears had to be at most 20 dB at neurological/psychiatric disease, reading problems (e.g., one, 15 dB at two, or 10 dB at three of the octave fre- dyslexia), claustrophobia, epilepsy, pregnancy, or metal quencies between 250 and 4000 Hz. The mean PTA of in the body contraindicating MRI scanning. All partici- the group with hearing impairment was 49.8 dB HL pants provided written informed consent, and the study (SD¼ 7.3, range: 40–61.6 dB HL). Thresholds at was approved by the Ethics Committee of VUmc. 8000 Hz were on average 50.88 dB HL (SD¼ 22.32, range: 12.5–97.5 dB HL). The etiologies of the impair- MRI Acquisition ments included combinations of congenital, familial, noise-induced, and age-related hearing loss. One partici- T1-weighted MRI images were obtained using a 3T GE pant reported perinatal asphyxia as the suspected eti- Signa scanner (General Electric Company, Fairfield, CT, ology, and four participants reported unknown causes. USA), equipped with an eight-channel phased array The average duration of hearing impairment was 17 head coil, using a fast spoiled gradient-recalled echo years (range: 1–43 years, SD¼ 12 years). Duration and sequence, with the following parameters: repetition time- ¼ 8,236 ms, echo time¼ 3.248 ms, inversion time- ¼ 450 ms, flip angle¼ 12 , field of view¼ 220 mm , 166 sagittal slices, resolution¼1mm 0.9 mm 0.9 mm. Voxel-Based Morphometry Analysis Image preprocessing was performed using Statistical Parametric Mapping 8 (SPM8; http://www.fil.ion.ucl. ac.uk/spm, Wellcome Department of Cognitive Neurology, London, UK, 2008) and VBM8-toolbox (http://dbm.neuro.uni-jena.de/vbm.html) that ran on Mathworks Matrix Laboratory 8.0 (MATLAB; MathWorks, Natrick, MA, USA). The VBM8-toolbox was used in default settings. To reduce between-subject variability, structural images were oriented to the anter- ior/posterior commissure line. Thereafter, they were bias-corrected with a cutoff of 30 mm full-width-at- half-maximum and segmented into gray matter, white matter, and cerebrospinal fluid. White and gray matter images were warped to a standard stereotactic space (152 T1 MNI template, Montreal Neurological Institute) Figure 1. Pure-tone hearing thresholds (averaged over both using linear affine transformation and high-dimensional ears) of hearing-impaired and normal-hearing participants at the DARTEL normalization (Ashburner & Friston, 2000). octave frequencies between 250 and 8000 Hz. Error bars denote the standard error of the mean. In this step, the normalized images were modulated using 4 Trends in Hearing Table 1. Clusters That Differed in Gray or White Matter Volume Between Hearing-Impaired and Normal-Hearing Participants Revealed by the Whole-Brain Analyses. Anatomical region Contrast L/R Tissue (Brodmann area) k TZ p . MNI (x, y, z) e uncorr NH< HI R GM Angular gyrus (39/40) 76 4.67 4.01 <.001 36, 57, 36 NH> HI L WM Fusiform gyrus (19/37) 28 4.13 3.65 <.001 38, 76, 11 k ¼ cluster size; R¼ right hemisphere; L¼ left hemisphere; GM¼ gray matter; WM¼ white matter; HI¼ hearing-impaired; NH¼ normal-hearing; MNI (x, y, z)¼ Montreal Neurological Institute stereotactic space coordinates. nonlinear deformation. This is a correction for individ- significantly differed between the groups using ual differences in brain size that enables the comparison MarsBaR (http://marsbar.sourceforge.net). Among the of brain volume rather than tissue density (Ashburner, HI group, we calculated correlations between cluster 2007). All images were visually inspected for quality. volume (adjusted for total gray or white matter Covariances between the volumes were calculated to volume), duration of hearing impairment (in years), identify outliers. Finally, volumes were spatially and hearing acuity (mean PTA in dB HL). smoothed with a 10-mm full-width-half-maximum Gaussian kernel. To investigate differences between HI and NH adults Results in gray and white matter volume, we constructed general ROI Analyses linear models (GLMs) separately for white and gray matter images using SPM 8. In these models, group The GLMs revealed larger gray matter volume in the (HI or NH) was the independent variable, and gray right angular gyrus in the participants with hearing (or white) matter volume was the dependent variable. impairment as compared with the participants with Total gray (or white) matter volume and age were normal hearing: p < 0.05; T¼ 4.58; Z¼ 3.96; FWE included in the models as nuisance covariates. To restrict k ¼ 52; MNI(x,y,z)¼ 36, 57, 36. We observed no stat- our analyses to the brain regions that are thought to be istically significant (p < 0.05) differences in gray or FWE involved in listening to degraded speech, we selected the white matter volume between the groups in the other bilateral pars orbitalis, pars triangularis, pars opercu- predefined regions. laris, superior temporal gyri, Heschl’s gyri, supplemen- tary motor area, and angular gyri as regions-of-interests Whole-Brain Analyses (ROIs). We used the Automated Anatomical Labeling (AAL) atlas (Tzourio-Mazoyer et al., 2002) to define The GLMs revealed a cluster in the right angular gyrus these regions. In addition, we exclusively selected and with larger gray matter volume in HI listeners in com- separately considered (cf. Peelle et al., 2011) the primary parison with NH listeners (see Figure 2 and Table 1). auditory cortices using the bilateral TE1.0 and TE1.1 The coordinates of the peak voxel of this cluster were masks (Morosan et al., 2001) within the SPM Anatomy the same as those of the cluster resulting from the ROI Toolbox (Eickhoff et al., 2005). Analyses for each ROI analysis; however, this cluster was larger and exceeded were separately conducted with a statistical threshold of the boundaries of the AAL mask. In addition, the p< .05, family-wise error rate (FWE) corrected. Last, models revealed that the participants with hearing exploratory whole-brain analyses were conducted at a impairment had smaller white matter volume in a cluster more lenient threshold of p< .001, uncorrected, with in the left fusiform gyrus compared with the participants an extent threshold of k > 25. The rationale for these with NH (see Figure 3 and Table 1). whole-brain analyses was to explore hearing status– related volume changes in regions outside of the a Relationship Between Volume, Hearing Acuity, and priori hypothesized ones. The results of these analyses Hearing Impairment Duration may be utilized in future studies (e.g., meta-analyses) that focus on related research questions. Among the HI participants, gray matter cluster volume in the right angular gyrus (adjusted for total gray matter volume) correlated positively with severity of hearing Correlation Analyses impairment (Pearson’s r¼ 0.5, p¼ .04; Figure 2c). In order to assess the link between hearing impairment There was no significant association between this and cluster volume, we extracted the estimated gray or volume and duration of hearing impairment (Pearson’s white matter volumes within the clusters that r¼ 0.45, p¼ .07). White matter cluster volume in the left Alfandari et al. 5 Figure 2. (a) Cluster of gray matter volume (in red) in the right angular gyrus that is larger in the hearing-impaired group as compared with the listeners with normal hearing, overlayed on the Automated Anatomical Labeling right angular gyrus mask (in blue). (b) Same region as in Figure 2(a), sagittal view. x and z are the slice coordinates in MNI space. (c) Relationship between gray matter cluster volume in the right angular gyrus and hearing acuity (mean pure-tone average at 1000, 2000, and 4000 Hz averaged over both ears) in the hearing- impaired group. Plotted are standardized gray matter residuals, adjusted for the effects of total gray matter volume. fusiform gyrus was not statistically significantly asso- ciated with either hearing acuity or duration of hearing impairment (Pearson’s r¼ 0.11, p¼ .66; Pearson’s r¼ 0.1, p¼ .69, respectively). Discussion In this study, we explored the association between partial hearing loss and brain volume. The comparison between structural magnetic resonance images of HI and NH adults in gray and white matter volume in the areas involved in speech perception revealed larger gray matter volume in the right angular gyrus in HI listeners as compared with the NH ones. Furthermore, among the HI listeners, we observed a positive relationship between severity of hearing impairment and gray matter volume Figure 3. Cluster of white matter volume in the left fusiform in the right angular gyrus. This relationship supports the gyrus that is smaller in the hearing-impaired listeners as compared association between altered brain volume and partial with the normal-hearing ones. x is the slice coordinate in MNI hearing loss. space. 6 Trends in Hearing The angular gyrus is considered to be an interface for auditory area or the white matter underneath it. the integration and transfer of information from differ- This outcome supports the idea that hearing impairment ent modalities and processing subsystems (see Seghier, alone may not be sufficient for reduced volume in the 2013 for a review). Studies on structural plasticity sug- auditory cortex (Profant et al., 2014). The discrepancy gest that volume increases in the angular gyrus are asso- between the current findings and the previous reports ciated with learning a new skill that requires the may be related to the complex interaction between employment of multiple modalities (see Draganski & aging and hearing loss (Wayne & Johnsrude, 2015). May, 2008 for a review). Functional neuroimaging stu- Whereas previous studies have focused on age-related dies with NH adults suggest the involvement of the hearing loss in mostly middle-aged to older adults (see angular gyrus in the adaptation to degraded speech Mudar & Husain, 2016 for a review), our participants (Guediche, Blumstein, Fiez, & Holt, 2014), and learning comprised adults with variable etiologies of impairment. of new speech sounds (Golestani & Zattore, 2004). In the Because aging is associated with impaired functional light of the above, our results may reflect mechanisms connectivity in the salience network, and impaired con- related to learning to understand distorted speech. nectivity between the salience and auditory networks Differences in angular gyrus volume may also reflect (Onoda, Ishihara, & Yamaguchi, 2012), older adults the compensatory use of brain networks. Hearing with hearing impairment may perhaps benefit from impairment not only results in less sound input, but is adaptive strategies less compared with younger adults also associated with temporal and frequency distortions with hearing impairment. For this reason, decreased that reduce the fidelity of the signal (Plomp, 1978). These gray matter volume in the primary auditory cortex may distortions cannot be compensated for by hearing aids be more evident in aging populations; increased gray (Plomp, 1978). Functional neuroimaging studies suggest matter volume in the right angular gyrus may be evident that hearing impairment may be associated with in individuals who benefit sufficiently from compensa- increased use of cognitive control and attentional net- tory mechanisms. works that operate in concert with the angular gyrus The relatively small sample in the current study might (see Cardin, 2016 for a review). Moreover, activity in have lowered the statistical sensitivity of the analyses to the right angular gyrus has previously been associated detect additional differences between the groups. This may with the comprehension of degraded speech in the pres- particularly be the case for the left Heschl’s gyrus, as this ence of visual speech cues (e.g., facial cues and lip move- region is known to have high macro-anatomical variability ments; McGettigan et al., 2012). Thus, increased volume between individuals (Marie et al., 2015). Furthermore, the in this area may be related to the larger dependency of heterogeneity in the etiologies of the hearing impairments HI listeners on visual speech cues (Erber, 1975; Pelson & in our sample may have limited the ability of our analyses Prather, 1974). Last, larger volume in the right angular to detect additional anatomical correlates of partial hear- gyrus is in line with the report of increased and possibly ing loss. This study examined gray and white matter compensatory activity in the right-hemisphere networks volume associated with partial hearing loss in a cross-sec- when listening to degraded speech (Liikkanen et al., tional design. Although it is plausible that impaired sen- 2007). sory information may have led to alterations in brain In addition to larger gray matter volume in the angu- volume, longitudinal studies with larger groups of moder- lar gyrus, our whole-brain analysis revealed smaller ately HI adults are needed to confirm this interpretation. white matter volume in the left fusiform gyrus of the In conclusion, in this exploratory study, we investi- HI listeners as compared with that of the NH ones. gated gray and white matter volume differences between This result goes together with reports of altered white HI and NH listeners in a VBM analysis. We observed matter integrity in the inferior fronto-occipital fasciculus larger gray matter volume in the right angular gyrus in in HI populations (Husain et al., 2011). However, the the HI group as compared to the NH group. Supporting altered white matter volume in the fusiform gyrus the link between volume and hearing ability, there was should be interpreted with caution because it results an association between poorer hearing acuity and larger from a statistically uncorrected multiple-comparison right angular gyrus cluster volume. Together, these analysis that runs the risk of showing false positives. results suggest that residual hearing impairment may Previous (VBM) studies with HI participants report be associated with volumetric changes in higher order smaller gray matter volume in the primary auditory area brain regions that are involved decoding meaning from (Eckert et al., 2012; Peelle et al., 2011) and altered white degraded speech input. matter volume beneath the primary auditory area (Mudar & Husain, 2016). The current comparison of Declaration of Conflicting Interests HI adults with age- and educational-level matched NH The authors declared no potential conflicts of interest with listeners did not reveal any relationship between hearing respect to the research, authorship, and/or publication of this impairment and brain volume in either the primary article. Alfandari et al. 7 hearing loss. Journal of the Association for Research in Funding Otolaryngology, 13(5), 703–713. doi: 10.1007/s10162-012- The authors disclosed receipt of the following financial support 0332-5. for the research, authorship, and/or publication of this article: Eickhoff, S. B., Stephan, K. E., Mohlberg, H., Grefkes, C., This work was supported by the Netherlands Organization for Fink, G. R., Amunts, K., ... Zilles, K. (2005). A new SPM Scientific Research (Veni grant 45110031). toolbox for combining probabilistic cytoarchitectonic maps and func tional imaging data. Neuroimage, 25(4), ORCID iDs 1325–1335. doi: 10.1016/j.neuroimage.2004.12.034. Defne Alfandari http://orcid.org/0000-0003-2882-0185 Erber, N. P. (1975). Auditory-visual perception of speech. Adriana A. Zekveld http://orcid.org/0000-0003-1320-6908 Journal of Speech and Hearing Disorders, 40(4), 481–492. doi: 10.1044/jshd.4004.481. References Glick, H., & Sharma, A. (2017). Cross-modal plasticity in Adank, P. (2012). The neural bases of difficult speech compre- developmental and age-related hearing loss: Clinical impli- hension and speech production: Two activation likelihood cations. Hearing Research, 343, 191–201. doi: 10.1016/ estimation (ALE) meta-analyses. Brain and Language, j.heares.2016.08.012. 122(1), 42–54. doi: 10.1016/j.bandl.2012.04.014. Golestani, N. (2014). Brain structural correlates of individual Adank, P., Davis, M. H., & Hagoort, P. (2012). Neural dis- differences at low-to high-levels of the language processing sociation in processing noise and accent in spoken language hierarchy: A review of new approaches to imaging research. comprehension. Neuropsychologia, 50(1), 77–84. doi: International Journal of Bilingualism, 18(1), 6–34. doi: 10.1016/j.neuropsychologia.2011.10.024. 10.1177/1367006912456585. Ashburner, J. (2007). A fast diffeomorphic image registration Golestani, N., & Zatorre, R. J. (2004). Learning new sounds of algorithm. Neuroimage, 38, 95–113. doi: 10.1016/ speech: Reallocation of neural substrates. Neuroimage, j.neuroimage.2007.07.007. 21(2), 494–506. doi: 10.1016/j.neuroimage.2003.09.071. Ashburner, J., & Friston, K. J. (2000). Voxel-based Guediche, S., Blumstein, S., Fiez, J., & Holt, L. L. (2014). morphometry—The methods. Neuroimage, 11, 805–821. Speech perception under adverse conditions: Insights from doi: 10.1006/nimg.2000.0582. behavioral, computational, and neuroscience research. Bailey, I. L., & Lovie, J. E. (1980). The design and use of a new Frontiers in Systems Neuroscience, 7, 126. doi: 10.3389/ near-vision chart. American Journal of Optometry and fnsys.2013.00126. Physiological Optics, 57(6), 378–387. Hickok, G., & Poeppel, D. (2007). The cortical organization of Binder, J. R., Liebenthal, E., Possing, E. T., Medler, D. A., & speech processing. Nature Reviews Neuroscience, 8(5), Ward, B. D. (2004). Neural correlates of sensory and deci- 393–402. doi: 10.1038/nrn2113. sion processes in auditory object identification. Nature Husain, F. T., Carpenter-Thompson, J. R., & Schmidt, S. A. Neuroscience, 7(3), 295–301. doi: 10.1038/nn1198. (2014). The effect of mild-to-moderate hearing loss on audi- Bosman, A. J., & Smoorenburg, G. F. (1995). Intelligibility tory and emotion processing networks. Frontiers in Systems of Dutch CVC syllables and sentences for listeners Neuroscience, 8, 10. doi: 10.3389/fnsys.2014.00010. with normal hearing and with three types of hearing Husain, F. T., Medina, R. E., Davis, C. W., Szymko-Bennett, impairment. International Journal of Audiology, 34(5), Y., Simonyan, K., Pajor, N. M., .. . Horwitz, B. (2011). 260–284. Neuroanatomical changes due to hearing loss and chronic Boyen, K., Langers, D. R., de Kleine, E., & van Dijk, P. (2013). tinnitus: A combined VBM and DTI study. Brain Research, Gray matter in the brain: Differences associated with tin- 1369, 74–88. doi: 10.1016/j.brainres.2010.10.095. nitus and hearing loss. Hearing Research, 295, 67–78. doi: Kuchinsky, S. E., Vaden, K. I., Keren, N. I., Harris, K. C., 10.1016/j.heares.2012.02.010. Ahlstrom, J. B., Dubno, J. R., ... Eckert, M. A. (2011). Cardin, V. (2016). Effects of aging and adult-onset hearing Word intelligibility and age predict visual cortex activ- loss on cortical auditory regions. Frontiers in ity during word listening. Cerebral Cortex, 22(6), Neuroscience, 10, 199. doi: 10.3389/fnins.2016.00199. 1360–1371. doi: 10.1093/cercor/bhr211. Chang, Y., Lee, S. H., Lee, Y. J., Hwang, M. J., Bae, S. J., Liikkanen, L. A., Tiitinen, H., Alku, P., Leino, S., Yrttiaho, S., Kim, M. N., ... Kang, D. S. (2004). Auditory neural path- & May, P. J. (2007). The right-hemispheric auditory cortex way evaluation on sensorineural hearing loss using diffusion in humans is sensitive to degraded speech sounds. tensor imaging. Neuroreport, 15(11), 1699–1703. doi: Neuroreport, 18(6), 601–605. doi: 10.1097/ 10.1097/01.wnr.0000134584.10207.1a. WNR.0b013e3280b07bde. Davis, M. H., Ford, M. A., Kherif, F., & Johnsrude, I. S. Lin, F. R., Ferrucci, L., An, Y., Goh, J. O., Doshi, J., Metter, (2011). Does semantic context benefit speech understanding E. J., ... Resnick, S. M. (2014). Association of hearing through ‘‘top–down’’ processes? Evidence from time- impairment with brain volume changes in older adults. resolved sparse fMRI. Journal of Cognitive Neuroscience, Neuroimage, 90, 84–92. doi: 10.1016/ 23(12), 3914–3932. doi: 10.1162/jocn_a_00084. j.neuroimage.2013.12.059. Draganski, B., & May, A. (2008). Training-induced structural Lin, Y., Wang, J., Wu, C., Wai, Y., Yu, J., & Ng, S. (2008). changes in the adult human brain. Behavioural Brain Diffusion tensor imaging of the auditory pathway in sen- Research, 192(1), 137–142. doi: 10.1016/j.bbr.2008.02.015. sorineural hearing loss: Changes in radial diffusivity and Eckert, M. A., Cute, S. L., Vaden, K. I., Kuchinsky, S. E., & diffusion anisotropy. Journal of Magnetic Resonance Dubno, J. R. (2012). Auditory cortex signs of age-related Imaging, 28(3), 598–603. doi: 10.1002/jmri.21464. 8 Trends in Hearing Marie, D., Jobard, G., Crivello, F., Perchey, G., Petit, L., morphometry of the central auditory pathway and auditory Mellet, E., ... Tzourio-Mazoyer, N. (2015). Descriptive cortex in aging. Neuroscience, 260, 87–97. doi: 10.1016/ anatomy of Heschl’s gyri in 430 healthy volunteers, includ- j.neuroscience.2013.12.010. ing 198 left-handers. Brain Structure and Function, 220(2), Rauschecker, J. P. (2011). An expanded role for the dorsal 729–743. doi: 10.1007/s00429-013-0680-x. auditory pathway in sensorimotor control and integration. McGettigan, C., Faulkner, A., Altarelli, I., Obleser, J., Hearing Research, 271(1), 16–25. doi: 10.1016/ Baverstock, H., & Scott, S. K. (2012). Speech comprehen- j.heares.2010.09.001. sion aided by multiple modalities: Behavioural and neural Scott, S. K., Rosen, S., Wickham, L., & Wise, R. J. (2004). interactions. Neuropsychologia, 50(5), 762–776. doi: A positron emission tomography study of the neural basis 10.1016/j.neuropsychologia.2012.01.010. of informational and energetic masking effects in speech Merabet, L. B., & Pascual-Leone, A. (2010). Neural reorgan- perception. Journal of the Acoustical Society of America, ization following sensory loss: The opportunity of change. 115(2), 813–821. doi: 10.1121/1.1639336. Nature Reviews Neuroscience, 11(1), 44–52. doi: 10.1038/ Seghier, M. L. (2013). The angular gyrus multiple functions nrn2758. and multiple subdivisions. The Neuroscientist, 19(1), Morosan, P., Rademacher, J., Schleicher, A., Amunts, K., 43–61. doi: 10.1177/1073858412440596. Schormann, T., & Zilles, K. (2001). Human primary audi- Tzourio-Mazoyer, N., Landeau, B., Papathanassiou, D., tory cortex: Cytoarchitectonic subdivisions and mapping Crivello, F., Etard, O., Delcroix, N., .. . Joliot, M. into a spatial reference system. Neuroimage, 13(4), (2002). Automated anatomical labeling of activations in 684–701. doi: 10.1006/nimg.2000.0715. SPM using a macroscopic anatomical parcellation of the Mudar, R. A., & Husain, F. T. (2016). Neural alterations in MNI MRI single-subject brain. Neuroimage, 15(1), acquired age-related hearing loss. Frontiers in Psychology, 7, 273–289. doi: 10.1006/nimg.2001.0978. 828. doi: 10.3389/fpsyg.2016.00828. van Strien, J. W. (1992). Classificatie van links- en rechtshan- Onoda, K., Ishihara, M., & Yamaguchi, S. (2012). Decreased dige proefpersonen [Classification of left- and right-handed functional connectivity by aging is associated with cognitive subjects.]. Nederlands Tijdschrift voorde Psychologie, 47, decline. Journal of Cognitive Neuroscience, 24(11), 2186–2198. 88–92. doi: 10.1162/jocn_a_00269. Wayne, R. V., & Johnsrude, I. S. (2015). A review of causal Peelle, J. E., Johnsrude, I. S., & Davis, M. H. (2010). mechanisms underlying the link between age-related hearing Hierarchical processing for speech in human auditory loss and cognitive decline. Ageing Research Reviews, 23, cortex and beyond. FrontiersinHuman Neuroscience, 4,51. 154–166. doi: 10.1016/j.arr.2015.06.002. doi: 10.3389/fnhum.2010.00051. Wild, C. J., Yusuf, A., Wilson, D. E., Peelle, J. E., Davis, M. Peelle, J. E., Troiani, V., Grossman, M., & Wingfield, A. H., & Johnsrude, I. S. (2012). Effortful listening: The pro- (2011). Hearing loss in older adults affects neural systems cessing of degraded speech depends critically on attention. supporting speech comprehension. The Journal of The Journal of Neuroscience, 32(40), 14010–14021. doi: Neuroscience, 31(35), 12638–12643. doi: 10.1523/ 10.1523/JNEUROSCI.1528-12.2012. JNEUROSCI.2559-11.2011. Wong, P. C., Uppunda, A. K., Parrish, T. B., & Dhar, S. Pelson, R. O., & Prather, W. F. (1974). Effects of visual mes- (2008). Cortical mechanisms of speech perception in noise. sage-related cues, age, and hearing impairment on spee- Journal of Speech, Language, and Hearing Research, 51(4), chreading performance. Journal of Speech, Language, and 1026–1041. doi: 10.1044/1092-4388(2008/075). Hearing Research, 17(3), 518–525. doi: 10.1044/ Zekveld, A. A., Heslenfeld, D. J., Festen, J. M., & Schoonhoven, jshr.1703.518. R. (2006). Top–down and bottom–up processes in speech Plomp, R. (1978). Auditory handicap of hearing impairment comprehension. NeuroImage, 32(4), 1826–1836. doi: and the limited benefit of hearing aids. The Journal of the 10.1016/j.neuroimage.2006.04.199. Acoustical Society of America, 63(2), 533–549. Profant, O., Skoch, A., Balogova´ , Z., Tinteˇ ra, J., Hlinka, J., & Syka, J. (2014). Diffusion tensor imaging and MR http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Trends in Hearing Pubmed Central

Brain Volume Differences Associated With Hearing Impairment in Adults

Download PDF
8 pages

Loading next page...
 
/lp/pubmed-central/brain-volume-differences-associated-with-hearing-impairment-in-adults-KMVUPOOUhI

References (49)

Publisher
Pubmed Central
Copyright
© The Author(s) 2018
ISSN
2331-2165
eISSN
2331-2165
DOI
10.1177/2331216518763689
Publisher site
See Article on Publisher Site

Abstract

Speech comprehension depends on the successful operation of a network of brain regions. Processing of degraded speech is associated with different patterns of brain activity in comparison with that of high-quality speech. In this exploratory study, we studied whether processing degraded auditory input in daily life because of hearing impairment is associated with differences in brain volume. We compared T1-weighted structural magnetic resonance images of 17 hearing-impaired (HI) adults with those of 17 normal-hearing (NH) controls using a voxel-based morphometry analysis. HI adults were individually matched with NH adults based on age and educational level. Gray and white matter brain volumes were compared between the groups by region-of-interest analyses in structures associated with speech processing, and by whole-brain analyses. The results suggest increased gray matter volume in the right angular gyrus and decreased white matter volume in the left fusiform gyrus in HI listeners as compared with NH ones. In the HI group, there was a significant correlation between hearing acuity and cluster volume of the gray matter cluster in the right angular gyrus. This correlation supports the link between partial hearing loss and altered brain volume. The alterations in volume may reflect the operation of compensatory mechanisms that are related to decoding meaning from degraded auditory input. Keywords hearing loss, structural plasticity, gray matter, white matter, angular gyrus, voxel-based morphometry Date received: 25 January 2017; revised: 8 January 2018; accepted: 22 January 2018 resemblance to the processing of degraded speech by Introduction normally hearing listeners. Speech comprehension in Brain plasticity following early-onset deafness is well documented (see Glick & Sharma, 2017; Merabet & Department of Otolaryngology—Head and Neck Surgery, Section Ear Pascual-Leone, 2010 for reviews). In individuals with & Hearing, VU University Medical Center, Amsterdam, the Netherlands complete hearing loss, cortical brain areas that are nor- Amsterdam Public Health Research Institute, VU University Medical mally responsible for processing auditory input, such as Center, the Netherlands Department of Anatomy & Neurosciences, VU University Medical Center, the primary auditory cortex (Heschl’s gyrus) and the sec- Amsterdam, the Netherlands ondary auditory cortex (planum temporale), are taken Department of Psychiatry, VU University Medical Center, Amsterdam, over by the remaining intact senses. These areas then the Netherlands respond to visual, tactile, and sign-language input (Glick Amsterdam Neuroscience, Amsterdam, the Netherlands Department of Psychology, VU University, Amsterdam, the Netherlands & Sharma, 2017). Particularly, regions in the superior Department of Behavioural Sciences and Learning, Linnaeus Centre temporal cortex that process auditory speech input in HEAD, The Swedish Institute for Disability Research, Linko¨ping University, normal-hearing (NH) individuals respond to visual Sweden speech input in deaf individuals (Merabet & Pascual- Corresponding author: Leone, 2010). Alterations in the brain that underlie the Adriana A. Zekveld, Department of Otolaryngology-Head and Neck adaptive strategies used by individuals with partial hear- Surgery, Section Ear & Hearing and Amsterdam Public Health Research ing loss to understand speech are less studied. Institute, VU University Medical Center, P.O. Box 7057, 1007 MB, One way of studying the processing of speech in hear- Amsterdam, the Netherlands. ing-impaired (HI) listeners is to assume that it bears Email: [email protected] Creative Commons Non Commercial CC BY-NC: This article is distributed under the terms of the Creative Commons Attribution-NonCommercial 4.0 License (http://www.creativecommons.org/licenses/by-nc/4.0/) which permits non-commercial use, reproduction and distribution of the work without further permission provided the original work is attributed as specified on the SAGE and Open Access pages (https://us.sagepub.com/en-us/nam/open-access-at-sage). 2 Trends in Hearing NH adults is thought to rely on a hierarchically orga- associated with gray matter volume in the left auditory nized network of brain regions that minimally includes cortex (Eckert, Cute, Vaden, Kuchinsky, & Dubno, bilateral superior and middle temporal gyri, the left pre- 2012). In contrast, hearing impairment in middle-aged frontal and premotor cortex, and the left inferior tem- adults has been linked to increased gray matter volume poral cortex (Hickok and Poeppel, 2007; Peelle, in the secondary auditory cortex (Brodmann area 22; Johnrude, & Davis, 2010; Rauschecker, 2011). The pro- Boyen, Langers, De Kleine, & Van Dijk, 2013). In add- cessing of degraded speech by NH individuals has been ition, changes in the integrity of white matter tracts in linked to increased activity in bilateral superior temporal pathways leading to the auditory cortex, including the cortices (Binder, Liebenthal, Possing, Medler, & Ward, lateral lemniscus and the anterior thalamic radiation 2004; Davis, Ford, Kherif, & Johnsrude, 2011; Scott, have been reported (Chang et al., 2004; Husain et al., Rosen, Wickham, & Wise, 2004; Wong, Uppunda, 2011; Lin et al., 2008). In sum, these studies mostly Parrish, & Dhar, 2008; Zekveld, Heslenfeld, Festen, & point to altered gray matter volume in the superior tem- Schoonhoven, 2006). Attempts to understand speech in poral cortices, and altered white matter integrity below low signal-to-noise conditions has shown to evoke activ- the primary auditory areas (although, see Profant et al., ity in bilateral anterior insulae and in the opercular part 2014); however, the extent to which the reported alter- of the left inferior frontal gyrus (Adank, Davis, & ations can be attributed to hearing impairment alone is Hagoort, 2012; Binder et al., 2004; Zekveld et al., difficult to determine because of the lack of age-matched 2006). These areas are thought to be engaged in effortful, and NH or HI and younger control groups. articulatory strategies that support the comprehension of Studies in the domain of language and audition sug- severely distorted speech. Similarly, the premotor area gest that structural plasticity is often observed within the and bilateral anterior superior temporal sulci, which same brain regions that functionally underlie the behav- are known to be involved in speech production, are ior at hand (see Golestani, 2014, for a review). Therefore, recruited during the correct perception of distorted a reasonable hypothesis would be that partial hearing speech (Adank, 2012). Lastly, bilateral angular gyri loss is accompanied by gray and white matter volume and left supplementary motor area are associated with alterations in the brain regions that are involved in the listening to distorted speech in an effortful manner comprehension of degraded speech. In this study, we (Kuchinsky et al., 2011; Wild et al., 2012). These findings explored the relationship between brain volume and par- raise the possibility that HI listeners may differentially tial hearing loss. For this, we compared the magnetic recruit regions in the inferior frontal, temporal, or par- resonance images of HI adults with those of age- and ietal areas while listening to speech. educational-level matched NH controls in a voxel- A number of studies have explored functional plasti- based morphometry (VBM) analysis. We hypothesized city following hearing impairment (see Mudar & Husain, that hearing impairment would be associated with alter- 2016 for a review). In elderly listeners, hearing impair- ations in gray and white matter volume in bilateral ment has been linked to decreased processing in the superior temporal sulci, bilateral superior and middle bilateral superior temporal gyri, thalamus, and brain- temporal gyri, the left inferior frontal gyrus, the left pre- stem during the comprehension of linguistically complex central gyrus, insula, angular gyri, and the premotor sentences (Peelle, Troiani, Grossman, & Wingfield, area, and that these alterations would be correlated 2011). In addition, altered processing has been reported with the severity and duration of the participants’ hear- in the default mode and dorsal saliency networks of HI ing impairment. participants in comparison with age-matched controls (Husain, Carpenter-Thompson, & Schmidt, 2014). These studies support the idea that hearing impairment Method may be related to altered functional processing beyond Participants the primary auditory areas, such as the attentional, emo- tional, and cognitive control networks. In total, 17 adults with NH (5 men, 12 women; age Studies exploring structural neuroplasticity following range: 20–62, M¼ 45.88, SD¼ 15.56 years) and 17 mild to moderate hearing loss have mostly focused on adults with hearing impairment (5 men, 12 women; age middle-aged to elderly populations (see Cardin, 2016; range: 20–63, M¼ 45.65, SD¼ 15.66 years) participated Mudar & Husain, 2016 for reviews). Among elderly in the study. HI participants were individually matched people, hearing impairment has been associated with with NH ones based on age and educational level. Of the reduced gray matter volume in the primary auditory 17 pairs, 14 were individually matched based on sex, and cortex (Husain et al., 2011; Peelle et al., 2011) and accel- the ratio of sex was matched between the groups. erated rates of gray matter volume decline in the right Participants with NH were recruited from among the temporal lobe (Lin et al., 2014). Particularly, hearing loss employees and students of the VU University medical in the higher frequency range has been shown to be center (VUmc) and the VU University Amsterdam, the Alfandari et al. 3 Netherlands. They had pure-tone thresholds of maximal severity of hearing impairment did not correlate signifi- 20 dB HL at the octave frequencies between 500 cantly (r¼ 0.42, p¼ .09). Neither duration nor severity and 4000 Hz. The mean pure-tone average (PTA; mean correlated significantly with age (r¼0.46, p¼ .06; hearing-threshold at 1000, 2000, and 4000 Hz, averaged r¼0.19, p¼ .46, respectively). over both ears) of the participants with normal hearing All air-bone gaps were smaller than 10 dB and all was 5.5 dB HL (SD¼ 5.5, range: 5 to 18.3 dB HL). participants had normal tympanograms. All partici- Thresholds at 8000 Hz were on average 17.94 dB HL pants scored better than 80% on each ear on a (SD¼ 16.45, range: 2.5 to 45 dB HL; see Figure 1 for speech audiogram with standard monosyllabic Dutch the average hearing thresholds at the octave frequencies consonant–vowel–consonant word lists (Bosman and between 250 and 8000 Hz). Smoorenburg, 1995). Furthermore, all participants Participants with hearing impairment were recruited were native Dutch speakers who used only spoken from among the patients of the outpatient clinic of the language and no sign language. All were classified as Ear & Hearing section of the Department of right-handed by the Dutch ‘‘Classification of left and Otolaryngology-Head and Neck Surgery of the VUmc. right-handed subjects’’ (van Strien, 1992). They had All participants with hearing impairment had symmet- normal or corrected-to-normal vision, and were screened rical sensorineural hearing loss. For inclusion in the cur- by a near-vision test that is equivalent to the visual acuity rent study, the mean PTA of each ear had to be between Snellen chart (Bailey and Lovie, 1980). Exclusion criteria 35 and 65 dB HL. Also, the asymmetry in the pure-tone were the use of psychotropic medication, a history of a thresholds between both ears had to be at most 20 dB at neurological/psychiatric disease, reading problems (e.g., one, 15 dB at two, or 10 dB at three of the octave fre- dyslexia), claustrophobia, epilepsy, pregnancy, or metal quencies between 250 and 4000 Hz. The mean PTA of in the body contraindicating MRI scanning. All partici- the group with hearing impairment was 49.8 dB HL pants provided written informed consent, and the study (SD¼ 7.3, range: 40–61.6 dB HL). Thresholds at was approved by the Ethics Committee of VUmc. 8000 Hz were on average 50.88 dB HL (SD¼ 22.32, range: 12.5–97.5 dB HL). The etiologies of the impair- MRI Acquisition ments included combinations of congenital, familial, noise-induced, and age-related hearing loss. One partici- T1-weighted MRI images were obtained using a 3T GE pant reported perinatal asphyxia as the suspected eti- Signa scanner (General Electric Company, Fairfield, CT, ology, and four participants reported unknown causes. USA), equipped with an eight-channel phased array The average duration of hearing impairment was 17 head coil, using a fast spoiled gradient-recalled echo years (range: 1–43 years, SD¼ 12 years). Duration and sequence, with the following parameters: repetition time- ¼ 8,236 ms, echo time¼ 3.248 ms, inversion time- ¼ 450 ms, flip angle¼ 12 , field of view¼ 220 mm , 166 sagittal slices, resolution¼1mm 0.9 mm 0.9 mm. Voxel-Based Morphometry Analysis Image preprocessing was performed using Statistical Parametric Mapping 8 (SPM8; http://www.fil.ion.ucl. ac.uk/spm, Wellcome Department of Cognitive Neurology, London, UK, 2008) and VBM8-toolbox (http://dbm.neuro.uni-jena.de/vbm.html) that ran on Mathworks Matrix Laboratory 8.0 (MATLAB; MathWorks, Natrick, MA, USA). The VBM8-toolbox was used in default settings. To reduce between-subject variability, structural images were oriented to the anter- ior/posterior commissure line. Thereafter, they were bias-corrected with a cutoff of 30 mm full-width-at- half-maximum and segmented into gray matter, white matter, and cerebrospinal fluid. White and gray matter images were warped to a standard stereotactic space (152 T1 MNI template, Montreal Neurological Institute) Figure 1. Pure-tone hearing thresholds (averaged over both using linear affine transformation and high-dimensional ears) of hearing-impaired and normal-hearing participants at the DARTEL normalization (Ashburner & Friston, 2000). octave frequencies between 250 and 8000 Hz. Error bars denote the standard error of the mean. In this step, the normalized images were modulated using 4 Trends in Hearing Table 1. Clusters That Differed in Gray or White Matter Volume Between Hearing-Impaired and Normal-Hearing Participants Revealed by the Whole-Brain Analyses. Anatomical region Contrast L/R Tissue (Brodmann area) k TZ p . MNI (x, y, z) e uncorr NH< HI R GM Angular gyrus (39/40) 76 4.67 4.01 <.001 36, 57, 36 NH> HI L WM Fusiform gyrus (19/37) 28 4.13 3.65 <.001 38, 76, 11 k ¼ cluster size; R¼ right hemisphere; L¼ left hemisphere; GM¼ gray matter; WM¼ white matter; HI¼ hearing-impaired; NH¼ normal-hearing; MNI (x, y, z)¼ Montreal Neurological Institute stereotactic space coordinates. nonlinear deformation. This is a correction for individ- significantly differed between the groups using ual differences in brain size that enables the comparison MarsBaR (http://marsbar.sourceforge.net). Among the of brain volume rather than tissue density (Ashburner, HI group, we calculated correlations between cluster 2007). All images were visually inspected for quality. volume (adjusted for total gray or white matter Covariances between the volumes were calculated to volume), duration of hearing impairment (in years), identify outliers. Finally, volumes were spatially and hearing acuity (mean PTA in dB HL). smoothed with a 10-mm full-width-half-maximum Gaussian kernel. To investigate differences between HI and NH adults Results in gray and white matter volume, we constructed general ROI Analyses linear models (GLMs) separately for white and gray matter images using SPM 8. In these models, group The GLMs revealed larger gray matter volume in the (HI or NH) was the independent variable, and gray right angular gyrus in the participants with hearing (or white) matter volume was the dependent variable. impairment as compared with the participants with Total gray (or white) matter volume and age were normal hearing: p < 0.05; T¼ 4.58; Z¼ 3.96; FWE included in the models as nuisance covariates. To restrict k ¼ 52; MNI(x,y,z)¼ 36, 57, 36. We observed no stat- our analyses to the brain regions that are thought to be istically significant (p < 0.05) differences in gray or FWE involved in listening to degraded speech, we selected the white matter volume between the groups in the other bilateral pars orbitalis, pars triangularis, pars opercu- predefined regions. laris, superior temporal gyri, Heschl’s gyri, supplemen- tary motor area, and angular gyri as regions-of-interests Whole-Brain Analyses (ROIs). We used the Automated Anatomical Labeling (AAL) atlas (Tzourio-Mazoyer et al., 2002) to define The GLMs revealed a cluster in the right angular gyrus these regions. In addition, we exclusively selected and with larger gray matter volume in HI listeners in com- separately considered (cf. Peelle et al., 2011) the primary parison with NH listeners (see Figure 2 and Table 1). auditory cortices using the bilateral TE1.0 and TE1.1 The coordinates of the peak voxel of this cluster were masks (Morosan et al., 2001) within the SPM Anatomy the same as those of the cluster resulting from the ROI Toolbox (Eickhoff et al., 2005). Analyses for each ROI analysis; however, this cluster was larger and exceeded were separately conducted with a statistical threshold of the boundaries of the AAL mask. In addition, the p< .05, family-wise error rate (FWE) corrected. Last, models revealed that the participants with hearing exploratory whole-brain analyses were conducted at a impairment had smaller white matter volume in a cluster more lenient threshold of p< .001, uncorrected, with in the left fusiform gyrus compared with the participants an extent threshold of k > 25. The rationale for these with NH (see Figure 3 and Table 1). whole-brain analyses was to explore hearing status– related volume changes in regions outside of the a Relationship Between Volume, Hearing Acuity, and priori hypothesized ones. The results of these analyses Hearing Impairment Duration may be utilized in future studies (e.g., meta-analyses) that focus on related research questions. Among the HI participants, gray matter cluster volume in the right angular gyrus (adjusted for total gray matter volume) correlated positively with severity of hearing Correlation Analyses impairment (Pearson’s r¼ 0.5, p¼ .04; Figure 2c). In order to assess the link between hearing impairment There was no significant association between this and cluster volume, we extracted the estimated gray or volume and duration of hearing impairment (Pearson’s white matter volumes within the clusters that r¼ 0.45, p¼ .07). White matter cluster volume in the left Alfandari et al. 5 Figure 2. (a) Cluster of gray matter volume (in red) in the right angular gyrus that is larger in the hearing-impaired group as compared with the listeners with normal hearing, overlayed on the Automated Anatomical Labeling right angular gyrus mask (in blue). (b) Same region as in Figure 2(a), sagittal view. x and z are the slice coordinates in MNI space. (c) Relationship between gray matter cluster volume in the right angular gyrus and hearing acuity (mean pure-tone average at 1000, 2000, and 4000 Hz averaged over both ears) in the hearing- impaired group. Plotted are standardized gray matter residuals, adjusted for the effects of total gray matter volume. fusiform gyrus was not statistically significantly asso- ciated with either hearing acuity or duration of hearing impairment (Pearson’s r¼ 0.11, p¼ .66; Pearson’s r¼ 0.1, p¼ .69, respectively). Discussion In this study, we explored the association between partial hearing loss and brain volume. The comparison between structural magnetic resonance images of HI and NH adults in gray and white matter volume in the areas involved in speech perception revealed larger gray matter volume in the right angular gyrus in HI listeners as compared with the NH ones. Furthermore, among the HI listeners, we observed a positive relationship between severity of hearing impairment and gray matter volume Figure 3. Cluster of white matter volume in the left fusiform in the right angular gyrus. This relationship supports the gyrus that is smaller in the hearing-impaired listeners as compared association between altered brain volume and partial with the normal-hearing ones. x is the slice coordinate in MNI hearing loss. space. 6 Trends in Hearing The angular gyrus is considered to be an interface for auditory area or the white matter underneath it. the integration and transfer of information from differ- This outcome supports the idea that hearing impairment ent modalities and processing subsystems (see Seghier, alone may not be sufficient for reduced volume in the 2013 for a review). Studies on structural plasticity sug- auditory cortex (Profant et al., 2014). The discrepancy gest that volume increases in the angular gyrus are asso- between the current findings and the previous reports ciated with learning a new skill that requires the may be related to the complex interaction between employment of multiple modalities (see Draganski & aging and hearing loss (Wayne & Johnsrude, 2015). May, 2008 for a review). Functional neuroimaging stu- Whereas previous studies have focused on age-related dies with NH adults suggest the involvement of the hearing loss in mostly middle-aged to older adults (see angular gyrus in the adaptation to degraded speech Mudar & Husain, 2016 for a review), our participants (Guediche, Blumstein, Fiez, & Holt, 2014), and learning comprised adults with variable etiologies of impairment. of new speech sounds (Golestani & Zattore, 2004). In the Because aging is associated with impaired functional light of the above, our results may reflect mechanisms connectivity in the salience network, and impaired con- related to learning to understand distorted speech. nectivity between the salience and auditory networks Differences in angular gyrus volume may also reflect (Onoda, Ishihara, & Yamaguchi, 2012), older adults the compensatory use of brain networks. Hearing with hearing impairment may perhaps benefit from impairment not only results in less sound input, but is adaptive strategies less compared with younger adults also associated with temporal and frequency distortions with hearing impairment. For this reason, decreased that reduce the fidelity of the signal (Plomp, 1978). These gray matter volume in the primary auditory cortex may distortions cannot be compensated for by hearing aids be more evident in aging populations; increased gray (Plomp, 1978). Functional neuroimaging studies suggest matter volume in the right angular gyrus may be evident that hearing impairment may be associated with in individuals who benefit sufficiently from compensa- increased use of cognitive control and attentional net- tory mechanisms. works that operate in concert with the angular gyrus The relatively small sample in the current study might (see Cardin, 2016 for a review). Moreover, activity in have lowered the statistical sensitivity of the analyses to the right angular gyrus has previously been associated detect additional differences between the groups. This may with the comprehension of degraded speech in the pres- particularly be the case for the left Heschl’s gyrus, as this ence of visual speech cues (e.g., facial cues and lip move- region is known to have high macro-anatomical variability ments; McGettigan et al., 2012). Thus, increased volume between individuals (Marie et al., 2015). Furthermore, the in this area may be related to the larger dependency of heterogeneity in the etiologies of the hearing impairments HI listeners on visual speech cues (Erber, 1975; Pelson & in our sample may have limited the ability of our analyses Prather, 1974). Last, larger volume in the right angular to detect additional anatomical correlates of partial hear- gyrus is in line with the report of increased and possibly ing loss. This study examined gray and white matter compensatory activity in the right-hemisphere networks volume associated with partial hearing loss in a cross-sec- when listening to degraded speech (Liikkanen et al., tional design. Although it is plausible that impaired sen- 2007). sory information may have led to alterations in brain In addition to larger gray matter volume in the angu- volume, longitudinal studies with larger groups of moder- lar gyrus, our whole-brain analysis revealed smaller ately HI adults are needed to confirm this interpretation. white matter volume in the left fusiform gyrus of the In conclusion, in this exploratory study, we investi- HI listeners as compared with that of the NH ones. gated gray and white matter volume differences between This result goes together with reports of altered white HI and NH listeners in a VBM analysis. We observed matter integrity in the inferior fronto-occipital fasciculus larger gray matter volume in the right angular gyrus in in HI populations (Husain et al., 2011). However, the the HI group as compared to the NH group. Supporting altered white matter volume in the fusiform gyrus the link between volume and hearing ability, there was should be interpreted with caution because it results an association between poorer hearing acuity and larger from a statistically uncorrected multiple-comparison right angular gyrus cluster volume. Together, these analysis that runs the risk of showing false positives. results suggest that residual hearing impairment may Previous (VBM) studies with HI participants report be associated with volumetric changes in higher order smaller gray matter volume in the primary auditory area brain regions that are involved decoding meaning from (Eckert et al., 2012; Peelle et al., 2011) and altered white degraded speech input. matter volume beneath the primary auditory area (Mudar & Husain, 2016). The current comparison of Declaration of Conflicting Interests HI adults with age- and educational-level matched NH The authors declared no potential conflicts of interest with listeners did not reveal any relationship between hearing respect to the research, authorship, and/or publication of this impairment and brain volume in either the primary article. Alfandari et al. 7 hearing loss. Journal of the Association for Research in Funding Otolaryngology, 13(5), 703–713. doi: 10.1007/s10162-012- The authors disclosed receipt of the following financial support 0332-5. for the research, authorship, and/or publication of this article: Eickhoff, S. B., Stephan, K. E., Mohlberg, H., Grefkes, C., This work was supported by the Netherlands Organization for Fink, G. R., Amunts, K., ... Zilles, K. (2005). A new SPM Scientific Research (Veni grant 45110031). toolbox for combining probabilistic cytoarchitectonic maps and func tional imaging data. Neuroimage, 25(4), ORCID iDs 1325–1335. doi: 10.1016/j.neuroimage.2004.12.034. Defne Alfandari http://orcid.org/0000-0003-2882-0185 Erber, N. P. (1975). Auditory-visual perception of speech. Adriana A. Zekveld http://orcid.org/0000-0003-1320-6908 Journal of Speech and Hearing Disorders, 40(4), 481–492. doi: 10.1044/jshd.4004.481. References Glick, H., & Sharma, A. (2017). Cross-modal plasticity in Adank, P. (2012). The neural bases of difficult speech compre- developmental and age-related hearing loss: Clinical impli- hension and speech production: Two activation likelihood cations. Hearing Research, 343, 191–201. doi: 10.1016/ estimation (ALE) meta-analyses. Brain and Language, j.heares.2016.08.012. 122(1), 42–54. doi: 10.1016/j.bandl.2012.04.014. Golestani, N. (2014). Brain structural correlates of individual Adank, P., Davis, M. H., & Hagoort, P. (2012). Neural dis- differences at low-to high-levels of the language processing sociation in processing noise and accent in spoken language hierarchy: A review of new approaches to imaging research. comprehension. Neuropsychologia, 50(1), 77–84. doi: International Journal of Bilingualism, 18(1), 6–34. doi: 10.1016/j.neuropsychologia.2011.10.024. 10.1177/1367006912456585. Ashburner, J. (2007). A fast diffeomorphic image registration Golestani, N., & Zatorre, R. J. (2004). Learning new sounds of algorithm. Neuroimage, 38, 95–113. doi: 10.1016/ speech: Reallocation of neural substrates. Neuroimage, j.neuroimage.2007.07.007. 21(2), 494–506. doi: 10.1016/j.neuroimage.2003.09.071. Ashburner, J., & Friston, K. J. (2000). Voxel-based Guediche, S., Blumstein, S., Fiez, J., & Holt, L. L. (2014). morphometry—The methods. Neuroimage, 11, 805–821. Speech perception under adverse conditions: Insights from doi: 10.1006/nimg.2000.0582. behavioral, computational, and neuroscience research. Bailey, I. L., & Lovie, J. E. (1980). The design and use of a new Frontiers in Systems Neuroscience, 7, 126. doi: 10.3389/ near-vision chart. American Journal of Optometry and fnsys.2013.00126. Physiological Optics, 57(6), 378–387. Hickok, G., & Poeppel, D. (2007). The cortical organization of Binder, J. R., Liebenthal, E., Possing, E. T., Medler, D. A., & speech processing. Nature Reviews Neuroscience, 8(5), Ward, B. D. (2004). Neural correlates of sensory and deci- 393–402. doi: 10.1038/nrn2113. sion processes in auditory object identification. Nature Husain, F. T., Carpenter-Thompson, J. R., & Schmidt, S. A. Neuroscience, 7(3), 295–301. doi: 10.1038/nn1198. (2014). The effect of mild-to-moderate hearing loss on audi- Bosman, A. J., & Smoorenburg, G. F. (1995). Intelligibility tory and emotion processing networks. Frontiers in Systems of Dutch CVC syllables and sentences for listeners Neuroscience, 8, 10. doi: 10.3389/fnsys.2014.00010. with normal hearing and with three types of hearing Husain, F. T., Medina, R. E., Davis, C. W., Szymko-Bennett, impairment. International Journal of Audiology, 34(5), Y., Simonyan, K., Pajor, N. M., .. . Horwitz, B. (2011). 260–284. Neuroanatomical changes due to hearing loss and chronic Boyen, K., Langers, D. R., de Kleine, E., & van Dijk, P. (2013). tinnitus: A combined VBM and DTI study. Brain Research, Gray matter in the brain: Differences associated with tin- 1369, 74–88. doi: 10.1016/j.brainres.2010.10.095. nitus and hearing loss. Hearing Research, 295, 67–78. doi: Kuchinsky, S. E., Vaden, K. I., Keren, N. I., Harris, K. C., 10.1016/j.heares.2012.02.010. Ahlstrom, J. B., Dubno, J. R., ... Eckert, M. A. (2011). Cardin, V. (2016). Effects of aging and adult-onset hearing Word intelligibility and age predict visual cortex activ- loss on cortical auditory regions. Frontiers in ity during word listening. Cerebral Cortex, 22(6), Neuroscience, 10, 199. doi: 10.3389/fnins.2016.00199. 1360–1371. doi: 10.1093/cercor/bhr211. Chang, Y., Lee, S. H., Lee, Y. J., Hwang, M. J., Bae, S. J., Liikkanen, L. A., Tiitinen, H., Alku, P., Leino, S., Yrttiaho, S., Kim, M. N., ... Kang, D. S. (2004). Auditory neural path- & May, P. J. (2007). The right-hemispheric auditory cortex way evaluation on sensorineural hearing loss using diffusion in humans is sensitive to degraded speech sounds. tensor imaging. Neuroreport, 15(11), 1699–1703. doi: Neuroreport, 18(6), 601–605. doi: 10.1097/ 10.1097/01.wnr.0000134584.10207.1a. WNR.0b013e3280b07bde. Davis, M. H., Ford, M. A., Kherif, F., & Johnsrude, I. S. Lin, F. R., Ferrucci, L., An, Y., Goh, J. O., Doshi, J., Metter, (2011). Does semantic context benefit speech understanding E. J., ... Resnick, S. M. (2014). Association of hearing through ‘‘top–down’’ processes? Evidence from time- impairment with brain volume changes in older adults. resolved sparse fMRI. Journal of Cognitive Neuroscience, Neuroimage, 90, 84–92. doi: 10.1016/ 23(12), 3914–3932. doi: 10.1162/jocn_a_00084. j.neuroimage.2013.12.059. Draganski, B., & May, A. (2008). Training-induced structural Lin, Y., Wang, J., Wu, C., Wai, Y., Yu, J., & Ng, S. (2008). changes in the adult human brain. Behavioural Brain Diffusion tensor imaging of the auditory pathway in sen- Research, 192(1), 137–142. doi: 10.1016/j.bbr.2008.02.015. sorineural hearing loss: Changes in radial diffusivity and Eckert, M. A., Cute, S. L., Vaden, K. I., Kuchinsky, S. E., & diffusion anisotropy. Journal of Magnetic Resonance Dubno, J. R. (2012). Auditory cortex signs of age-related Imaging, 28(3), 598–603. doi: 10.1002/jmri.21464. 8 Trends in Hearing Marie, D., Jobard, G., Crivello, F., Perchey, G., Petit, L., morphometry of the central auditory pathway and auditory Mellet, E., ... Tzourio-Mazoyer, N. (2015). Descriptive cortex in aging. Neuroscience, 260, 87–97. doi: 10.1016/ anatomy of Heschl’s gyri in 430 healthy volunteers, includ- j.neuroscience.2013.12.010. ing 198 left-handers. Brain Structure and Function, 220(2), Rauschecker, J. P. (2011). An expanded role for the dorsal 729–743. doi: 10.1007/s00429-013-0680-x. auditory pathway in sensorimotor control and integration. McGettigan, C., Faulkner, A., Altarelli, I., Obleser, J., Hearing Research, 271(1), 16–25. doi: 10.1016/ Baverstock, H., & Scott, S. K. (2012). Speech comprehen- j.heares.2010.09.001. sion aided by multiple modalities: Behavioural and neural Scott, S. K., Rosen, S., Wickham, L., & Wise, R. J. (2004). interactions. Neuropsychologia, 50(5), 762–776. doi: A positron emission tomography study of the neural basis 10.1016/j.neuropsychologia.2012.01.010. of informational and energetic masking effects in speech Merabet, L. B., & Pascual-Leone, A. (2010). Neural reorgan- perception. Journal of the Acoustical Society of America, ization following sensory loss: The opportunity of change. 115(2), 813–821. doi: 10.1121/1.1639336. Nature Reviews Neuroscience, 11(1), 44–52. doi: 10.1038/ Seghier, M. L. (2013). The angular gyrus multiple functions nrn2758. and multiple subdivisions. The Neuroscientist, 19(1), Morosan, P., Rademacher, J., Schleicher, A., Amunts, K., 43–61. doi: 10.1177/1073858412440596. Schormann, T., & Zilles, K. (2001). Human primary audi- Tzourio-Mazoyer, N., Landeau, B., Papathanassiou, D., tory cortex: Cytoarchitectonic subdivisions and mapping Crivello, F., Etard, O., Delcroix, N., .. . Joliot, M. into a spatial reference system. Neuroimage, 13(4), (2002). Automated anatomical labeling of activations in 684–701. doi: 10.1006/nimg.2000.0715. SPM using a macroscopic anatomical parcellation of the Mudar, R. A., & Husain, F. T. (2016). Neural alterations in MNI MRI single-subject brain. Neuroimage, 15(1), acquired age-related hearing loss. Frontiers in Psychology, 7, 273–289. doi: 10.1006/nimg.2001.0978. 828. doi: 10.3389/fpsyg.2016.00828. van Strien, J. W. (1992). Classificatie van links- en rechtshan- Onoda, K., Ishihara, M., & Yamaguchi, S. (2012). Decreased dige proefpersonen [Classification of left- and right-handed functional connectivity by aging is associated with cognitive subjects.]. Nederlands Tijdschrift voorde Psychologie, 47, decline. Journal of Cognitive Neuroscience, 24(11), 2186–2198. 88–92. doi: 10.1162/jocn_a_00269. Wayne, R. V., & Johnsrude, I. S. (2015). A review of causal Peelle, J. E., Johnsrude, I. S., & Davis, M. H. (2010). mechanisms underlying the link between age-related hearing Hierarchical processing for speech in human auditory loss and cognitive decline. Ageing Research Reviews, 23, cortex and beyond. FrontiersinHuman Neuroscience, 4,51. 154–166. doi: 10.1016/j.arr.2015.06.002. doi: 10.3389/fnhum.2010.00051. Wild, C. J., Yusuf, A., Wilson, D. E., Peelle, J. E., Davis, M. Peelle, J. E., Troiani, V., Grossman, M., & Wingfield, A. H., & Johnsrude, I. S. (2012). Effortful listening: The pro- (2011). Hearing loss in older adults affects neural systems cessing of degraded speech depends critically on attention. supporting speech comprehension. The Journal of The Journal of Neuroscience, 32(40), 14010–14021. doi: Neuroscience, 31(35), 12638–12643. doi: 10.1523/ 10.1523/JNEUROSCI.1528-12.2012. JNEUROSCI.2559-11.2011. Wong, P. C., Uppunda, A. K., Parrish, T. B., & Dhar, S. Pelson, R. O., & Prather, W. F. (1974). Effects of visual mes- (2008). Cortical mechanisms of speech perception in noise. sage-related cues, age, and hearing impairment on spee- Journal of Speech, Language, and Hearing Research, 51(4), chreading performance. Journal of Speech, Language, and 1026–1041. doi: 10.1044/1092-4388(2008/075). Hearing Research, 17(3), 518–525. doi: 10.1044/ Zekveld, A. A., Heslenfeld, D. J., Festen, J. M., & Schoonhoven, jshr.1703.518. R. (2006). Top–down and bottom–up processes in speech Plomp, R. (1978). Auditory handicap of hearing impairment comprehension. NeuroImage, 32(4), 1826–1836. doi: and the limited benefit of hearing aids. The Journal of the 10.1016/j.neuroimage.2006.04.199. Acoustical Society of America, 63(2), 533–549. Profant, O., Skoch, A., Balogova´ , Z., Tinteˇ ra, J., Hlinka, J., & Syka, J. (2014). Diffusion tensor imaging and MR

Journal

Trends in Hearing – Pubmed Central

Published: Mar 20, 2018

There are no references for this article.