Share this article on:

Diffusion Tensor Imaging of the ‘Auditory Connectome’

Kozin, Elliott D. MD; Lee, Daniel J. MD

doi: 10.1097/01.HJ.0000525525.25937.b0
Journal Club

Dr. Kozin, left, is a resident in the Harvard Combined Program in Otolaryngology. Dr. Lee, right, is the director of pediatric otology and neurotology at the Massachusetts Eye and Ear Infirmary and an associate professor in the department of otology and laryngology at Harvard Medical School.

Sensorineural hearing loss (SNHL) is the most commonly diagnosed birth defect, with about three out of every 1,000 children in the United States born deaf or hard of hearing (MMWR. 2010;59:220 Children with hearing loss can face significant challenges in psycho-social well-being, quality of life, and economic independence (Int J Pediatr Otorhinolaryngol. 2004;68[3]:287 Auditory neuroprosthetics, such as cochlear implants (CIs), have revolutionized the management of pediatric hearing loss with improved audiometric and quality of life outcomes (Eur Ann Otorhinolaryngol Head Neck Dis. 2016;133[1]:31

To date, over 300,000 individuals have received CIs (F1000Prime Rep. 2015;7:45; Fig. 1). Success with CI requires a normal to near normal bony cochlea, intact cochlear nerve, and a functional central auditory system to enable propagation of the electrical signal and appropriate sound processing in the brain. Physiologic measures such as the audiogram and the auditory brainstem response (ABR) provide indirect evidence of neural connectivity, but patients with profound hearing loss require imaging techniques to assess the integrity of these pathways. The bony anatomy of the inner ear can be appreciated with high-resolution computed tomography (CT), but the anatomy of the auditory pathways beyond the cochlea represents a complex neural network that, until recently, has been difficult to resolve.

Back to Top | Article Outline


The majority of children who have CIs have meaningful sound and speech awareness, but outcomes among similar cohorts remain highly variable (Acta Otorhinolaryngol Ital. 2011;31[5]:281 Studies suggest that duration of deafness and anatomic differences (including the size of the cochlear nerve, which connects the inner ear to the brain, and central auditory pathway development) may influence hearing outcomes (Ann Otol Rhinol Laryngol. 1992;101[4]:342; Int J Pediatr Otorhinolaryngol. 2014;78[6]:912 Improved imaging techniques of the internal auditory canal (IAC), such as advanced magnetic resonance imaging (MRI), result in greater understanding of pediatric auditory neuroanatomy (Laryngoscope. 2015;125[10]:2382; Fig. 2) However, there remain cases in which the integrity of these structures, including the cochlear nerve, are not well-defined. Further, while functional magnetic resonance (fMRI) imaging allows for an individualized and nuanced understanding of auditory pathway anatomy and physiology, it is often expensive, time-consuming, and challenging to perform in children (Hear Res. 2014;307:4

The treatment algorithm for pediatric patients with severe to profound hearing loss and ambiguous anatomy of the inner ear is an initial trial of a hearing aid followed by consideration of a cochlear implant (Otolaryngol Head Neck Surg. 2014;151[2]:308 However, this clinical approach may be suboptimal as CI outcomes improve with implantation at younger ages, and delay in intervention may result in missing the critical window for speech and language development (Otolaryngol Head Neck Surg. 2016;154[2]:247; J Cogn Neurosci. 2004;16[8]:1412 Furthermore, unsuccessful CI surgery can lead to significant financial and emotional costs and hamper the development of proficiency in sign language (Laryngoscope. 2011;121[12]:2604 Taken together, disorders of the cochlear nerve and hearing pathways can stand to benefit from emerging advanced imaging modalities that may improve prognostication of hearing outcomes and patient expectations, as well as determination of whether a patient should receive a cochlear implant or auditory brainstem implant (ABI; Fig. 1).

Back to Top | Article Outline


Diffusion tensor imaging (DTI) is a new advanced radiologic technique that can map and quantify individual neuronal networks, also called fiber tracks (Handb Clin Neurol. 2015; 129:277 Neuronal networks are being heavily investigated in the Human Connectome Project, which is funded by the National Institutes of Health BRAIN Initiative. As one example of this technology, neurosurgeons use DTI to see how neuronal networks surround brain tumors to reduce damage to normal structures during tumor surgery.

DTI technology offers many ways to indirectly determine the integrity of neuronal networks. One approach is the creation of fiber maps or tractography, which provides a qualitative understanding of how neurons are connected in the brain. Another measurement, fractional anisotropy (FA), allows for a quantitative understanding of discrete areas in the brain. Generally speaking, a decrease in FA indicates a decrease in the integrity of a neuronal network.

Back to Top | Article Outline


Pilot studies in our laboratory and by other groups have demonstrated that DTI is a powerful tool in the investigation of central auditory pathways or the auditory connectome (PLoS One. 2015;10[10]:e0140643; Hear Res. 2014;318:1; Brain Behav. 2014;4[4]:531; Front Syst Neurosci. 2014;8:87; Hear Res. 2015;323:1; Clin Neuroradiol. 2015;26[4]:439; Figs. 3 and 4). Two recent studies, in particular, have demonstrated its potential to further our understanding of this highly complex central auditory network. Knowledge from these studies will lead to more accurate predictors of outcomes following the management of children and adults with hearing loss.

In a groundbreaking paper by Huang, et al., investigators examined the utility of DTI for CI recipients (PLoS One. 2015 The objective of this study was to determine the influence of congenital hearing in children on the integrity of the auditory connectome. More specifically, the authors aimed to determine whether DTI could be used to predict post-operative CI outcomes.

Specifically, 24 children with congenital hearing loss and 20 normal-hearing controls underwent conventional MRI and DTI examination. Six regions along the central auditory pathways were evaluated—the trapezoid body, superior olivary nucleus, inferior colliculus, medial geniculate body, auditory radiation, and white matter of Heschl's gyrus. The category of auditory performance (CAP) score was also assessed before and after cochlear implantation. Among the 24 patients, eight patients with a CAP score over six were classified into the good outcome group, and 16 patients with a CAP score below six were classified into the poor outcome group.

The authors found a significant decrease was observed in FA values at the six points along the auditory pathway in SNHL patients compared with normal-hearing controls. Compared with good-outcome subjects, poor-outcome subjects displayed decreased FA values. The authors explained that correlation analyses revealed strong correlations between FA values and CAP scores. Strong correlations between CAP scores and age at implantation were also found. The authors concluded that preoperative DTI could be used to evaluate changes in neuronal networks in the central auditory pathways that are not readily detectable by standard MRI. In addition, they concluded that DTI might play an important role in evaluating the outcome of cochlear implantation.

In a pilot study by Vos, et al., investigators aimed to reconstruct the auditory nerve using DTI (Hear Res. 2015 Authors reassembled the cochlear nerve in five normal hearing subjects and five patients with long-term profound single-sided and bilateral SNHL. A specialized imaging protocol was designed to image the cochlear nerve; the nerve was reconstructed using fiber tractography and DTI metrics, e.g., FA values.

After comparing various DTI metrics, there was a small but significant reduction in FA values in the auditory nerves bilaterally in patients with SNHL compared with normal hearing controls. These results are early evidence of structural changes of both auditory nerves because of single-sided deafness. Notably, the authors did not find any differences in DTI metrics when comparing the normal and abnormal sides in patients with single-sided deafness.

The authors also noted several technical limitations of current DTI technology. Specifically, the cochlear nerve, after leaving the cochlea, joins with the vestibular nerve to form the vestibulocochlear nerve (Cranial Nerve VIII [CN 8]). CN 8 runs with the facial nerve (CN 7) in the IAC. Given the proximity of these cranial nerves, it is challenging for DTI to distinguish between them and measure individual nerves. The authors noted that further refinements in DTI technology may enable it to distinguish subdivisions of CN 8, as well as differences between CN 7 and 8.

DTI is a promising tool for hearing specialists that can be used to characterize the detailed structure of the central auditory pathways and help predict outcomes following implant surgery. Future studies will likely refine imaging acquisition techniques to better visualize and quantify changes in the brainstem anatomy associated with sound deprivation and determine clinical applicability.

Copyright © 2017 Wolters Kluwer Health, Inc. All rights reserved.