For many forms of sensorineural hearing loss, permanent threshold elevations are often associated with hair cell damage or loss. These losses have received much experimental attention and are a primary focus of prevention and treatment efforts (Rubel et al. 2013, Hear Res. 2013, 297: 42–51). Recent work in noise and aging, however, has revealed a much more insidious process that progressively interrupts communication between sensory hair cells and auditory neurons, leading ultimately to death of the neurons themselves. These neurodegenerative changes are likely very common, occurring even in ears with normal threshold sensitivity and a full complement of hair cells; thus, they present challenges to our traditional approaches to diagnosis and management.
The inner hair cell-cochlear nerve fiber synapse is the primary conduit through which information about the acoustic environment is transmitted to the auditory nervous system. In ears that age normally, e.g., without noise exposure, synapses are lost gradually, throughout life, and are seen throughout the cochlea long before age-related loss of threshold sensitivity or hair cells (Sergeyenko et al 2013, J Neurosci. 33(34):13686-94). Cochlear nerve cell bodies (spiral ganglion cells, SGC) show proportional declines, with losses recorded in aging mice consistent with those observed in age-graded human temporal bones (Makary et al 2011, J Assoc Res Otolaryngol. 12(6):711-17).
Noise produces similar synaptic losses, but immediately, and then accelerates aging, even for exposures that produce reversible threshold shifts and no hair cell loss (Kujawa and Liberman 2006, J Neurosci. 26(7):2115-23; Kujawa and Liberman 2009, J Neurosci. 29(45):14077-85). Losses at short post-exposure times are restricted to cochlear frequency regions with maximum acute threshold shift, and are followed by proportional SGC loss in the same cochlear regions. As animals age, losses spread to cochlear regions that initially appear uninvolved in the noise insult. Noise-induced cochlear neurodegeneration has now been observed in several mammalian species; there is no reason to suspect that the human will provide an exception to this general finding.
This widespread ‘primary’ neurodegeneration has remained hidden for many years. Although thresholds are sensitive metrics of hair cell damage, they are relatively insensitive to diffuse loss of cochlear synapses and cochlear neurons: a) DPOAEs are unaffected because only pre-synaptic processes are required for their generation; b) neural response thresholds (e.g., ABR) are unaffected, because the noise targets cochlear neurons with high thresholds (Furman et al 2013, J Neurophysiol. 110(3):577-86) and c) behavioral audiometric thresholds are unaffected, for the same reason as (b) and because stimulus detection requires less neural information than stimulus discrimination. Although thresholds fail to capture the communication failure, Kujawa and colleagues have identified key indicators evident in the suprathreshold neural response.
This primary cochlear neurodegeneration is a likely contributor to a variety of auditory perceptual abnormalities common with aging and after noise, including speech-in-noise difficulties (Bharadwaj et al 2014, Front Syst Neurosci. 8(26), tinnitus and hyperacusis (Gu et al 2010, J Neurophysiol. 104(6): 3361-70 Schaette et al 2011, J Neurosci. 31(38):13452-57).
These sobering findings have important implications for public health. Once an ear has been exposed to noise, a question that may be asked is whether the noise insult can influence future changes in the ear and hearing; for example, those that accrue with age. Traditionally, the focus has been on thresholds, and lack of delayed threshold shifts after noise has been taken as evidence that delayed effects of noise do not occur. Recent work, using powerful new tools, provides clear evidence that it can. How do we now think about noise risk? Federal noise exposure guidelines aim to protect against permanent threshold shifts, a protection goal that assumes that reversible threshold shifts are associated with cochlear recovery and a safe exposure. Accumulating evidence suggests that this assumption is unwarranted.