20Q: OAEs - Music to My Ears

From the Desk of Gus Mueller

Yes, the ear does indeed create its own sounds, and we call them otoacoustic emissions or OAEs.  Thomas Gold suggested their existence in the late 1940s, David Kemp explained them more completely in 1978, and the measurement of OAEs has been part of our clinical diagnostic battery since the 1980s.

You would think that by now, we’d know just about all there is to know about OAEs, but there continues to be new observations every year. We’re fortunate to have an international expert on this topic back with us this month at 20Q.  Sumit Dhar, PhD, is Professor at the Roxelyn and Richard Pepper Department of Communication Sciences and Disorders at Northwestern University.

As you might recall, Dr. Dhar provided an excellent review of OAEs and their origins here at 20Q last month.  As time was running out, our Question Man realized that 20 Questions just wasn’t enough to cover this intriguing topic, and Dr. Dhar graciously agreed to return.

When we left their conversation last month, Sumit had just provided a practical clinical summary of what is “safe to assume” if OAEs are either present or absent.  I know you’ll find several more clinical pearls this month as the conversation continues.  
 

Gus Mueller, PhD 
Contributing Editor
August 2014

To browse the complete collection of 20Q with Gus Mueller CEU articles, please visit www.audiologyonline.com/20Q

20Q: OAEs – Music to My Ears

Sumitrajit Dhar

1.  Thanks for returning to answer the rest of my questions on OAEs. Do you remember where we left off?

Yes, I had just finished summarizing how we can use OAEs in the clinic, and what we might expect when they are present or absent.  Is it all coming back?

2.  Yes, you indicated that those were the general rules.

Right.  There can be (and probably will be) exceptions.  The one thing I had briefly mentioned, but we never got back to, is that OAEs also can be used as a predictive measure.

3.  Predictive of what?

There are a few things, but here is one example.  Let’s take a group of soldiers leaving for combat training and we measure their OAEs before they leave.  We know that they probably are going to be exposed to a lot of noise during training.  We could measure their DPOAEs or TEOAEs—for this example it probably wouldn’t matter. Now let’s line them up, with those with the largest OAEs to the left and those with the smallest OAEs to the right. What if I told you that those to the right end of the line are four to eight times more likely to return from combat training with cochlear damage?

4.  I’d say that’s a fairly good predictive measure of noise-induced hearing loss.  Is there solid research to support this?

Certainly.  This is exactly what Lynne Marshall and her colleagues (2006) found in one of the most interesting applications of OAEs in recent times. What is even more striking is that a relationship between pre-exposure OAE level and post-exposure change in OAEs was detectable even in those who did not exhibit any changes in hearing thresholds.

5.  Previously you mentioned the auditory efferent system.  Is that in play here?

Possibly, but in ways that we are not certain about yet.   As I discussed last time, there is significant scientific as well as clinical interest in the auditory efferents. From the scientific standpoint, we want to understand their functional role in humans. And from the clinical standpoint, we would like to reliably measure the strength of the efferent modulation of cochlear mechanics as this seems to be related to how vulnerable an animal might be to noise induced cochlear damage (Maison & Liberman, 2000).  It also appears to be related to how well humans can localize in noise (Andeol et al., 2011), and how well they might perceive speech in a noisy background (Kumar & Vanaja, 2004). Most recently, Liberman and colleagues (2014) demonstrated that de-efferented animals age faster than those with an intact efferent system, even when they are not exposed to hazardous levels of noise.

6.  And you’re saying that OAEs are a good way to study all this?

OAEs are a natural choice in evaluating the efferent “reflex” in humans as they are non-invasive and directly assess the impact of the auditory efferents on outer hair cells. The auditory efferent nervous system originates in the auditory cortex and terminates in the cochlea after traversing through several waypoints. Specifically, the medial olivocochlear (MOC) bundle, the last leg of this anatomical pathway, originates in the medial superior olivary complex with its fibers terminating at the base of the outer hair cells. The activation of the MOC system results in acetylcholine mediated reduction in the gain of the cochlear amplifier, resulting in reduced cochlear output observable in compound action potentials and OAEs, among other things.

7.  Why don’t we use OAEs routinely to assess efferent function in our patients?

We probably will do so someday soon. However, several complications have to be resolved before a routine clinical test can be recommended. First, every once in a while we observe an enhancement in the OAE rather than a reduction. It appears that we now understand this enhancement to be artificial and a result of a change in the phase relationship between multiple OAE components. Second, the MOC reflex (MOCR) is typically elicited using noise or tones in the contralateral ear. These elicitors may also activate the stapedial reflex in the middle ear. Activating the stapedial reflex in both ears creates all kinds of complications such as reducing the effective level of the stimulus tones in the cochlea. Third, the MOCR appears to be rather small in magnitude when measured using evoked emissions (TEOAEs, DPOAEs, etc.). This in itself is not a problem as long as the change is dependable and repeatable. Which brings us to the final complication, the repeatability or stability of the MOCR has not been established yet. Certainly, there are these complications but work is underway to solve them and I am fairly confident that we will have a usable clinical test protocol soon.

8.  Any advice for the clinician on what to do in the meantime?

I am glad you asked.  If you’d like to measure the MOCR in one of your patients, here is a step-by-step procedure you can try out:

• First, make sure you know their middle ear reflex threshold for broadband noise.
• Once you have that measure, pull out a coin from your purse or pocket and flip it. Make sure to catch it before it hits the ground, and flop it on top of your opposite hand.
• If it’s heads, measure the patients DPOAEs between 1 and 2.5 kHz at five or six points using low stimulus levels (e.g., L₁=55 and L₂=40 dB SPL). If it is tails, measure the patients TEOAEs using the standard clinical protocol that you always use.
• Once you have this baseline, repeat the measurement while playing broadband noise through the immittance audiometer that you just used to measure the acoustic reflex at a level 5 to 10 dB lower than the reflex threshold for broadband noise into the contralateral ear.
• Finally repeat the measurement without the noise.

So now you have three measures, two without and one with contralateral noise. The difference between the two measures without noise should give you an estimate of test-retest variability in this subject and the difference between either of the no-noise measures and the noise measure should give you an estimate of the MOCR. Hopefully the MOCR is bigger than the test variability.

9.  Flip a coin?  You do not seem to have a preference between DPOAEs and TEOAEs?

In this case, that is in assessing the MOCR, we do not know enough to claim that one type of OAE might be better suited than the other. But you bring up a good point. The traditional view has been that OAEs are OAEs. They can be spontaneous or evoked. We further classified evoked OAEs based on the type of stimulus used to evoke them – TEOAEs evoked by transient stimuli, SFOAEs evoked by single tonal stimuli, and DPOAEs evoked by two simultaneous tones.

10.  That works for me. How else do you classify them? 

Well, there is a second school of thought that classifies OAEs into two groups based on the generation mechanisms responsible for them. According to this classification system there are two major mechanisms that can produce OAEs – distortion and reflection (Shera & Guinan, 1999). The distortion mechanism is tied to the nonlinear amplification process in the cochlea. The reflection mechanism is tied to a process of back scattering of inward traveling energy due to non specific irregularities or roughness in the cochlea. This roughness could be anatomical, such as irregular spacing or numbers of outer hair cells; or physiological, such as fluctuations in gain along the length of the cochlea.

11.  “Back scattering of inward traveling energy?”  Are you really expecting me to follow all of this?

Stick with me, there’s not too much more.  It’s important to point out that the OAEs generated by these two mechanisms have distinct phase signatures with the phase of reflection OAEs varying rapidly and that of distortion OAEs hardly varying at all with frequency. This dichotomy does hold experimentally when the phase behavior of SFOAEs (evoked by stimulus levels lower than 40 dB SPL) and that of DPOAEs is examined. The possibility of these two different generation mechanisms and their exact nature is a source of current debate. As a result, much work has been devoted to understanding the relationship of this type of classification to the traditional classification of OAEs. It turns out that most OAEs measured in humans, with the exception of spontaneous OAEs, probably contain a mixture of portions generated by each mechanism.

12.  After 35 years of research, I would have thought that we at least would know how OAEs are produced?

Our persistent uncertainties about OAEs are a testament to their complex nature. Not only are we still debating their generation mechanisms, we are also arguing (in a nice way) about how they travel out of the cochlea. The question at the core of the propagation debate is whether you can have backward traveling waves. If you can have a traveling wave propagate against the mechanical gradient of the basilar membrane, OAEs could travel from the cochlea to the ear canal via that mechanism. If however, a backward traveling wave is physically improbable, there has to be another mode of backward propagation that carries OAEs out to the ear canal. One probability is a compression wave through the cochlear fluids, which would carry the OAEs to the middle ear boundary with the speed of sound in fluids – much faster than a traveling wave.

13.  Are there other areas of uncertainty?

I told you it was complex.  Yet another area of emerging knowledge concerns the area of the cochlea that contributes to OAEs for a given stimulus frequency. We have taken for granted that OAEs are a local phenomenon. That is OAEs at 2000 Hz tell us something about the health of the cochlear region with the same characteristic frequency. Recent evidence, however, suggests that in some species at higher input levels there may be significant contributions from regions of the cochlea basal to the characteristic frequency region of the stimulus frequency (Martin, Stagner, & Lonsbury-Martin, 2010). This of course forces us to rethink how place specific OAE test interpretation can be. Fortunately, the contribution of these basal sources appears to be minimal at regular test levels (~ 65 dB SPL) in human ears. However, we do need to be aware that the cochlear region contributing to the OAEs measured in the ear canal expands with increasing stimulus level.

14.  With all that is unknown, I’m now wondering if we even should be using OAEs clinically?

Looks like I have been somewhat successful in our conversation. My students claim that all I do is make them question and become uncertain about everything we do clinically. That turns out only to be one of my secondary missions. My primary mission is to get every student to find the evidence that supports everything they do clinically. To that end, there is plenty of solid evidence to suggest that OAEs are a reliable marker of cochlear, especially outer hair cell, health. That knowledge can then be readily extended to suggest that OAEs should be an excellent tool for detecting pathologies where the outer hair cells are particularly compromised. Turns out that we know of a few such pathologies. Noise induced cochlear damage may be the most obvious one. And as we discussed earlier, there is a truckload of evidence suggesting that noise damage is evident in OAEs before any other test of the auditory system. Turns out though, that the return of OAEs to normal levels after noise exposure should not be taken to mean that everything is back to pristine condition. Recent evidence suggests that the initial noise induced temporary pathology could lead to a cascade of neural degenerative events that continue even after OAEs have returned to normal baseline levels (Kujawa & Liberman, 2009).

15.  You’ve convinced me on the noise thing.  Are there other applications?

There really are many more applications. Lets take the most ubiquitous of all auditory pathologies – age-related hearing loss. We have just published a pair of papers that together demonstrate the sensitivity of OAEs (DPOAEs in this case) to age related changes in the cochlea. In the first paper we documented hearing thresholds in over 350 individuals between the ages of 10 and 65 years (Lee et al., 2012). We picked these individuals to have as pristine an auditory history as possible. No noise exposure, no ototoxic medications, no family history of hearing loss, no complaint of any hearing difficulty – you get the picture. We showed that these individuals have identical hearing thresholds through 2000 Hz, but signs of aging are increasingly evident at higher frequencies, with the greatest disparities in thresholds as a function of age evident at frequencies higher than 12,000 Hz. Nothing controversial or surprising so far, correct?

16.  Not yet.  So the second paper must have been a bit more exciting?

It was.  In the second paper, we demonstrate that signs of aging are visible in DPOAE levels at frequencies as low as 1000 Hz for individuals as young as 20 years old (Poling, Siegel, Lee, Lee, & Dhar, 2014).

17.  Wow, when we talk about aging, we are not thinking about 20-year-olds.

Right! Here is an example of how OAEs might force us to alter our thinking about auditory aging and when might be the right time to start an awareness campaign.  I am certainly no advocate of using OAEs for any and all clinical situations. That in fact will lead to overuse of the CPT codes sparking yet another review by CMS, and these reviews rarely end well. I do believe, however, that as we learn more about the biology and physics of OAEs we will find more appropriate and powerful clinical uses for these measures. We have already talked about noise and aging. But detection of certain types of chemotoxicity, diagnosis of auditory neuropathy, and many other applications readily come to mind.

18.  We’re just about finished, so how about some predictions.  What are three things about the clinical application of OAEs that might change in the next five years?

I am not very good at predicting the future. In fact, I specialize in buying high and selling low. But I could tell you the three things that I would like to see in the clinical practice of OAEs in the near future.

19.  That’s good enough for me.  I’d like to hear them.

• First, I would like to see standardization of calibration techniques. There has been very significant advancement in the last decade on better ways to deliver controlled stimuli to the ear canal. I would like to see all clinical devices use these modern techniques. One of the frustrations with OAEs is their variability in some individuals. Errant calibration is certainly one source of this variability and we are now able to eliminate this one source of variability. Let’s do it.
• Second, partly because we are able to calibrate more accurately, we should see more routine and reliable measurements at frequencies above 6 or 8 kHz. I think we will see this extension first in DPOAEs. But being able to reliably record DPOAEs from the true base of the cochlea will lead to sensitive tests of changes due to ototoxic drugs, aging, etc.
• Third, I think we are at the threshold of understanding how to differentially apply different OAE types. This is the area that I am not as certain about but am seeing emerging evidence to suggest that TEOAEs, for example, may be more sensitive to certain pathologies than DPOAEs. Once further developed, this understanding could allow us to use the correct OAE type for any specific condition or pathology, significantly boosting their clinical utility. Stay tuned!!

20.  We’ve now gone through a total of 40 questions in our two visits, and I just have to say one final thing:  You sound way too excited about OAEs. What makes you so excited?

The music in your ears! Think about what makes music so pleasing to you. The dynamic range in the mixture of soft and loud notes and the fine differences in pitches of the notes that make music so enjoyable to you are all a consequence of what we now call the active processes in the cochlea. And there is no other marker of these active processes that is as easily accessible to us as OAEs. This tie between the essential function of the cochlea and OAEs makes them an incredible tool in the hands of the scientist trying to understand how the cochlea works. The same tie makes OAEs a powerful clinical probe into cochlear conditions that impact the cochlear amplifier. It might sound like an odd thing to say thirty five years after the discovery of OAEs, but the party is just getting started.

References

Andeol, G., Guillaume, A., Micheyl, C., Savel, S., Pellieux, L., & Moulin, A. (2011). Auditory efferents facilitate sound localization in noise in humans. J Neurosci, 31(18), 6759-6763.

Kujawa, S. G., & Liberman, M. C. (2009). Adding insult to injury: cochlear nerve degeneration after "temporary" noise-induced hearing loss. J Neurosci, 29(45), 14077-14085. doi: 10.1523/JNEUROSCI.2845-09.2009

Kumar, U. A., & Vanaja, C. S. (2004). Functioning of olivocochlear bundle and speech perception in noise. Ear Hear, 25(2), 142-146.

Lapsley Miller, J. A., Marshall, L., Heller, L. M., & Hughes, L. M. (2006). Low-level otoacoustic emissions may predict susceptibility to noise-induced hearing loss. J Acoust Soc Am, 120(1), 280-296.

Lee, J., Dhar, S., Abel, R., Banakis, R., Grolley, E., Lee, J., . . . Siegel, J. (2012). Behavioral hearing thresholds between 0.125 and 20 kHz using depth-compensated ear simulator calibration. Ear Hear, 33(3), 315-329. doi: 10.1097/AUD.0b013e31823d7917

Maison, S. F., & Liberman, M. C. (2000). Predicting vulnerability to acoustic injury with a noninvasive assay of olivocochlear reflex strength. J Neurosci, 20(12), 4701-4707.

Liberman, M. C., Liberman, L. D., Maison, S. F. (2014) "Efferent feedback slows cochlear aging", J Neurosci, 34(13), 4599-4607.

Martin, G. K., Stagner, B. B., & Lonsbury-Martin, B. L. (2010). Evidence for basal distortion-product otoacoustic emission components. J Acoust Soc Am, 127(5), 2955-2972. doi: 10.1121/1.3353121

Poling, G. L., Siegel, J. H., Lee, J., Lee, J., & Dhar, S. (2014). Characteristics of the 2f(1)-f(2) distortion product otoacoustic emission in a normal hearing population. J Acoust Soc Am, 135(1), 287-299. doi: 10.1121/1.4845415

Shera, C. A., & Guinan, J. J., Jr. (1999). Evoked otoacoustic emissions arise by two fundamentally different mechanisms: a taxonomy for mammalian OAEs. J Acoust Soc Am, 105(2 Pt 1), 782-798.