20Q: Otoacoustic Emissions - Clinical and Future Applications

From the Desk of Gus Mueller

In 1954, the musical group the Monotones asked, “Who wrote the book of love?” A reasonable question, which has yet to be answered. Sometimes we say that someone “wrote the book” on a given topic, when in fact that person has never written a book on anything. But, here is what we do know: Sumit Dhar wrote the book on otoacoustic emissions (along with his colleague Jay Hall), which is why he’s the right person to join us here at 20Q for a discussion on these measures.

Back in a 2014 20Q, Sumit provided a review of the clinical value of OAE testing. In summary, his statements went something like this: If OAEs are present, you can be fairly certain that one or more of the following is true:

• the patient has no major middle ear pathology
• the patient has a functional complement of outer hair cells
• the patient’s hearing is most likely within the normal range

If OAEs are absent, you can be fairly certain that one or more of the following is true:

• the patient has degradation of outer hair cell function
• the patient may have a middle ear condition preventing the forward transmission of stimuli and the reverse transmission of the OAEs
• the patient most likely has hearing thresholds worse than 20 dB HL

That’s a lot of information to be gained from a few minutes of testing. One could argue that OAEs should be performed routinely for most, if not all, patients—and some audiologists indeed have suggested this. Clearly, OAEs are a very useful objective diagnostic test. Like all tests, however, refinements and advancements continue to occur, and that is why we have brought back Sumit to fill us in on what has been happening the past 5 years.

Sumitrajit (Sumit) Dhar, PhD, is the Hugh Knowles Professor of Hearing Science at Northwestern University, Chair of the Roxelyn & Richard Pepper Department of Communication Sciences and Disorders, and Associate Dean for Research for the Northwestern School of Communication. Dr. Dhar’s involvement in the profession is extensive—he is the Editor in Chief of the American Journal of Audiology, Associate Editor of Audiology Today, and is on the Editorial Board of Plural Publishing. He has served on the Board of Directors of the Academy of Audiology, and is a former President of the Illinois Academy of Audiology.

Much of Sumit’s research has been associated with the early detection of age-related hearing loss and understanding the physiological bases, which often involves OAE measures. We’re fortunate to have him provide a valuable update on this important topic here at 20Q.

Gus Mueller, PhD
Contributing Editor

Browse the complete collection of 20Q with Gus Mueller CEU articles at www.audiologyonline.com/20Q

20Q: Otoacoustic Emissions - Clinical and Future Applications

Learning Outcomes 

After this course, readers will be able to:

• Describe the types and uses of OAEs in current clinical protocols
• Describe hidden hearing loss and the potential clinical role of OAEs in diagnosing this condition
• List advances in OAEs measurement and future uses of OAE measures 

               Sumit Dhar     

1. Last we talked, I think you were pitching lines about ears making music. Still at it?

You bet. Our last discussions on otoacoustic emissions (OAEs) were back in 2014: OAEs - Music to My Ears, and OAEs - Sound Clinical Tool. Since those articles were published, we know a bit more about how OAEs change over a lifetime across the length of the cochlea. We can get into the details later if you want. However, I'm hopeful I can get you excited by saying that we are finally beginning to get a good handle on the complexities of the physiological mechanisms responsible for OAEs. The even better news is that there are good indications that we will be able to leverage these complexities for clinical purposes rather than them simply confusing things for us. It is also exciting that significant advancements in instrumentation and calibration allow us to measure OAEs with greater precision over a wider range of frequencies. These technical developments along with the possibility of clinical exploitation of distortion and reflection emissions are the most exciting things that have happened since we last spoke on the topic.

2. Distortion and reflection emissions and not transients and distortion products? This is new territory for me.

The traditional way of thinking about different OAE types was based on the type of stimulus that was used to record them. In this nomenclature, there were transient or click-evoked and distortion product (evoked by two tones) OAEs. You could also evoke an OAE with a single tone to get stimulus frequency OAEs. But those have not caught on in clinics yet. Other than these evoked OAEs, there were also spontaneous OAEs that could be recorded without any external stimulation. As early as the late nineties, scientists were starting to imagine a different classification system for OAEs – one based on mechanistic differences in their production.

David Kemp introduced the framework of wave- and place-fixed emissions. Others refined this model further by defining distortion and reflection emissions (Shera & Guinan, 1999). There is a pretty extensive literature about what makes emissions fall in the distortion or reflection categories. But the business end of it all is that transient evoked (TE)OAEs when recorded using clinical levels are mostly reflection emissions and distortion product (DP)OAEs are mostly distortion emissions.

There are two reasons to be confident about this way of thinking about OAE types. First, the mechanism generating distortion and reflection emissions are predicted to lead to very different phase behavior of the OAE. And indeed, distortion and reflection emissions show very different phase behavior. Second, evidence is starting to emerge that these two emission types may behave in distinct ways for different kinds of insults to the cochlea. And this second observation is what makes for an exciting clinical future for OAEs.

3. What are the clinical implications of this way of thinking about OAEs?

I suspected that would be your next question. Let me just point to the evidence we already have. Rao and Long (2011) have shown that reflection OAEs are more vulnerable to aspirin-induced temporary ototoxicity compared to distortion OAEs. (Abdala & Kalluri, 2017) have proposed a framework to combine results from distortion and reflection emissions. This combined profile preliminarily shows clusters of ears with different kinds of pathology. And more indirectly, several groups have shown that reflection OAEs are modulated to a greater degree by the efferent auditory nervous system (Deeter, Abel, Calandruccio, & Dhar, 2009; Henin, Thompson, Abdelrazeq, & Long, 2011). And even more indirectly, Poling and colleagues (2014) have shown that the reflection component is gradually extinguished with age, as seen in the reduction and elimination of DPOAE fine structure. Several groups are actively investigating the differential effects of various pathologies on distortion and reflection OAEs. This is something our research team is studying, and hopefully, we’ll have more specific direction about clinical applications in the very near future.

4. Are you suggesting we start doing DPOAEs and TEOAEs on every patient?

Have you heard of a faster way to a CMS audit? I think adding both OAE types to the routine clinical protocol may be a bit premature at this time. Given what we know today, a more prudent approach would be to selectively apply this double-barrel approach on specific patient populations. A simple way of thinking about this would be to consider patients where hearing thresholds and OAEs are discordant. For example, if you have a patient with clinically normal hearing thresholds but absent TEOAEs or DPOAEs, it may be a good idea to try the other OAE test as well. Suspected neuropathy ears could also be good candidates for the double-barrel approach. In parallel, auditory researchers should develop this idea of divergence in OAE types in specific populations. Over time, the knowledge from clinicians and that from the laboratories will merge into a common pool and allow us to develop specific strategies for exploiting the mechanistic differences in OAE generation. Armed with this background knowledge and rock-solid evidence we can design specific clinical strategies to be applied for specific clinical populations.

So in brief, not yet!!

5. I'm hearing more and more about hidden hearing loss these days. Does OAE testing play into this?

Indeed, hidden hearing loss is quite the rage. Let’s trace its history a bit. In a seminal paper in 2009, Kujawa and Liberman demonstrated that after a brief exposure to high levels of noise, outer hair cell function recovered in mice but a cascade of progressive neural damage starting from the inner hair cell synapse continued in mice. In subsequent work, this sequence of events was found to also happen in various other lab animals. To dive a little deeper, what Kujawa and Liberman showed was the following: after brief noise exposure, DPOAE levels, as well as ABR wave amplitudes, were attenuated indicating a temporary threshold shift. Over time, DPOAE levels at many stimulus levels recovered to pre-exposure levels. That is, the DPOAE input-output function mimicked that from before the noise exposure. However, the ABR peak amplitudes continued to remain depressed. Through careful histology, Kujawa and Liberman documented that the OAE and ABR results reflected a recovery in outer hair cell function but continued degradation of neural tissue, starting with the synapse at the inner hair cell and progressing centrally. This combination of conditions was termed hidden hearing loss as hearing thresholds remain normal-like but supra-threshold function is affected as seen in the reduced amplitude of ABR peaks.

6. Does it happen in humans as well?

That is indeed the question of the day. Several groups are actively trying to not only answer the question whether hidden hearing loss exists in humans but also how to diagnose it. On the broadest scale, you could imagine everyone complaining of speech perception in noise with clinically normal hearing thresholds to have hidden hearing loss. However, we have now defined the clinical phenotype to be limited to neural degradation without any compromise in outer hair cell function. That may be too stringent a definition to satisfy in humans who are exposed to all kinds of toxins over a lifetime. Nonetheless, evidence of hidden hearing loss has been presented in those with a self-reported higher risk of noise exposure (Liberman, Epstein, Cleveland, Wang, & Maison, 2016). Others have found evidence of hidden hearing loss only in those exposed to very high levels of noise associated with events such as gunfire and blasts (Bramhall, Konrad-Martin, McMillan, & Griest, 2017).

7. If hidden hearing loss is related to the nerve and synapse, what do OAEs have to do with it?

That is a great question. OAEs come into play in two important ways. First, the classic definition of hidden hearing loss necessarily includes normal outer hair cell function. Thus, normal OAE levels would be a requisite condition for labeling the condition with confidence. It is a valid question as to whether such a pure segregation of hair cell and neural degeneration is realistic to expect in humans. The answer may be that patients walking into any clinic will have a mix of damage at various loci along the auditory pathway. In this case, our interest might shift to quantifying the contribution of each of these localized pathologies. To accurately quantify the “damage” at any neural level we would have to know the impact of any outer hair cell dysfunction and potentially compensate for it. Thus, OAEs could help define the classical version of hidden hearing loss and they could be used to independently quantify the degradation in the neural response after compensating for the changed input due to outer hair cell deficiencies.

8. What would be a good clinical test protocol if hidden hearing loss was suspected for a given patient?

This is an easy one to answer because I do not know the exact answer. It is clear that the clinical protocol would have to involve both behavioral and physiological/electrophysiological measures. Since normal hearing thresholds are a defining feature and we do not seem to be able to do anything unless we know audiometric thresholds, audiometry will likely be a part of the protocol. It also seems likely that we will want to include a measure of speech perception in noise as that is likely to be one of the complaints the patient walks in with. In the realm of physiology, OAEs to probe outer hair cell function, ABR measures including quantifying the SP/AP ratio, as well as acoustic reflex measures,  could be useful (Liberman, Epstein, Cleveland, Wang, & Mason, 2016; Valero, Hancock, Maison, & Liberman, 2018). While these measures seem to be the obvious choices, others are working on behavioral measures based on the predicted psychophysical consequences of the known physiological deficits in hidden hearing loss (Bharadwaj et al., 2019). I am sure the answer to this question will evolve dramatically in the coming years. So, stay tuned, but for now, refer to a recent consensus paper by Bramhall et al. (2019).

9. Is there any new OAE equipment or technology I should be aware of?

Yes, the world of clinical OAE technology does continue to evolve. Many of the changes that happen to clinical systems are under the hood and often improve performance without the user knowing anything has changed. This is typically achieved by software updates to the signal processing techniques used. But, every once in a while, there are big changes that significantly alter how we record OAEs and potentially what we use them for. For example, some of us already have OAE systems with two probes that allow simultaneous recording from the two ears. You can also purchase OAE systems that calibrate the stimuli in a way that account for ear canal acoustics and deliver much more accurately calibrated stimuli. This way of calibrating stimuli is the culmination of a decade or more of work. In the new method, rather than calibrating stimuli in the traditional sound pressure level (dB SPL), stimuli are delivered in what is being called forward pressure level (dB FPL).

10. Forward pressure? I know nothing about that approach.

Let’s start by defining the basic problem. We calibrate stimuli for different auditory tests using standardized couplers. Here the assumption is that the prescribed coupler represents a good approximation of the impedance properties of the ear canal as captured by a particular earphone. Thus, the level measured at the distal end of a coupler is believed to represent the sound pressure that would be presented to the typical eardrum. Of course, this ignores differences in ear anatomy between individuals. We could, however, account for individual ear canal acoustics in calibrating OAE stimuli as there is an active microphone in the ear canal. Great idea! Well . . . almost! Because there is an important complication. Calibrating using the levels recorded by the OAE microphone assumes that the sound pressure levels are identical at the OAE microphone and the ear drum. At some frequencies they are in fact very different. To make matters worse, the level at the OAE microphone at specific frequencies, e.g., where the distance between the OAE probe and the ear drum is a quarter of the wavelength, the sound waves emanating from the OAE probe and those returning to the probe after being reflected at the ear drum will be out of phase and the effective level seen by the microphone will be much lower than that at the ear drum where such phase cancellation is not occurring. Now, correcting the stimulus level to compensate for the dip in level would only result in an over-correction and over stimulation at that specific frequency. The solution to this problem is to carefully characterize the characteristics of the OAE probe and the ear canal to be able to compute the sound levels traveling forward and that being reflected back from the ear drum separately. The level of the signal moving forward towards the ear drum is labeled the forward pressure level. As I mentioned above, there is at least one clinical device that is doing this already and my guess is that we will see the entire field shift over to this approach the next few years.

11. But the OAE is not going forward. How do you account for that?

Very good point. Others recognize this subtlety as well. Once the ear canal has been characterized adequately, the effects of ear canal resonance can be subtracted from the overall OAE signal. This yields a purer estimate of the OAE as generated by the cochlea and filtered by the middle ear. This method of calibration was recently described by Charaziak and Shera (2017) and the corrected OAE level was presented as the Emission Pressure Level (EPL). Taken together, the calibration of stimuli in FPL and the representation of OAEs in EPL should improve test-retest reliability of OAE tests and possibly improve their diagnostic efficiency.

12. And how exactly is this process going to occur?

Well, here is the great news. Audiologists are already doing this. You probably have forgotten what an immittance audiometer looks like, but some of us still calibrate the probe by measuring responses in several cavities when we turn our immittance audiometers on in the morning. The process for calibrating the OAE probe will probably resemble this very closely. The measurement device will do all the computations and the audiologist can keep his or her focus on the patient and patient care.

13. Okay, so we now have better ways of calibrating signals back and forth. How does that change clinical practice?

For now, we are only talking about OAEs. It is not difficult to envision that better ways of calibrating and delivering stimuli to the ear drum will improve test performance for other audiological procedures as well. The main limitation, however, is that calibrating in FPL requires a microphone in the ear canal. This would increase the cost and complexity of conducting some basic audiological procedures. However, if this technological barrier could be overcome, one can imagine all acoustic stimuli used for audiological procedures being delivered in FPL.

14. You mean this forward pressure business can be used for other audiological tests as well?

This is a 20Q conversation on OAEs, and we are discussing a new calibration technique arguably for OAE testing. It is ironic, however, that the initial evidence supporting the use of FPL has come from experiments involving hearing thresholds. That is correct – good old behavioral audiometric thresholds. This is out of necessity as the relationship between stimulus and OAE level is not linear, so changes in OAE levels do not necessarily reflect equivalent changes in stimulus levels. This difficulty is easily overcome by using behavioral thresholds: thresholds should shift by a dB for every dB change in stimulus level. The basic test paradigm employed to evaluate test performance using FPL calibration is to evaluate hearing thresholds twice with some manipulation such as earphone placement or simply time between tests. Every time this has been attempted, FPL calibration has led to better test-retest variability in hearing thresholds (Miller, Reed, Robinson, & Perez, 2018; Souza, Dhar, Neely, & Siegel, 2014).

15. But we do not typically use OAEs in serial tests for most clinical applications, do we?

You are correct about the current state of affairs. In the US, the serial use of OAEs is mostly limited to monitoring cochlear status in individuals undergoing chemotherapy using a known ototoxin. However, other pockets of application exist. For example, OAEs are one of the tools used by some branches of the military to monitor hearing health. The optimist in me thinks that as prevention becomes more important, OAEs will start to play a greater role in detecting the first signs of change in cochlear function. Similarly, when (not if) hearing restoration through biological means becomes part of our practice, OAEs will be useful not only in differential diagnosis but also in determining and monitoring appropriate treatment. A likely example would be monitoring age-related changes at the base of the cochlea and initiating treatment in a timely fashion. In all of these applications, confidence in clinical decisions will critically depend on exact test outcomes with high test-retest reliability.

16. Does monitoring cochlear health at the base of the cochlea require different protocols?

You do have to use much higher frequencies, but there are clinical devices that already allow measurement of OAEs up to 12 kHz or even 16 kHz. The ability to measure at these frequencies with confidence is integrally tied to the use of the advanced calibration techniques that we discussed earlier. As these applications of OAEs become more popular, we could expect more devices to add these capabilities. In our experience, an upper measurement limit of 12 kHz or 16 kHz may be adequate to detect the earliest signs of aging (Poling et al., 2014). The demand for measurements at frequencies higher than 16 kHz may only be for ototoxicity monitoring in pediatric patients.

17. Now you are changing everything. Do the current measurement methods work at those higher frequencies?

This is again a great question. If by measurement methods you are referring to stimulus parameters such as level, frequency ratio, etc., the current methods will certainly work. However, there may well be more optimal conditions to use especially at the highest frequencies as the response properties of the cochlea are known to change with frequency. Work is underway to determine if changing stimulus parameters to match these expected changes in cochlear mechanics could improve test performance. If that turns out to be the case, then we can certainly opt for the modified parameters. However, there is no reason to wait for the modifications. The current stimulus parameters are good enough to get us going.

18. Will we then need new norms?

Yes, we will need new norms, because we have now changed the effective stimulus levels and the way we estimate the emission level. But, there already are some in the literature. See, for example, Poling and colleagues (2014). I am sure more are to come soon. One of the important differences between these and previous sets of norms is that we will probably need age-referenced norms with age divisions that are more detailed than just pediatric and adult. This is because OAEs are expected to change more dramatically with age at the highest frequencies and the improved calibration and measurement methods should give us data sets that are less varied.

19. Are there clinical OAE systems that already can do this fancy calibration?

As I have mentioned above already, there is at least one clinical device that can do FPL calibration and measure out to higher frequencies. There are others which do not use FPL calibration yet but can measure out to higher frequencies. I am not certain what calibration techniques are used in these machines. Audiologists using these machines should learn from the manufacturer about their approach to calibration.

20. This all sounds a little dry. Do you OAE people ever do anything fun?

Do we ever! How about measuring OAEs in space? Read all about it at nasa.gov.

If that is not enough fun, how about we finish where we started? The last time you and I talked, we had come up with the theme of ears making music and we started our current conversation remembering that thread. Well, check this out. Sonic artist Jakob Kirkegard has actually done what we were contemplating – made music out of OAEs recorded from 18 ears. These are spontaneous OAEs and he has strung them together into a composition. Check it out: https://youtu.be/9A6ms3yzBWk

References

Abdala, C., & Kalluri, R. (2017). Towards a joint reflection-distortion otoacoustic emission profile: Results in normal and impaired ears. Journal of the Acoustical Society of America, 142(2), 812. doi:10.1121/1.4996859

Bharadwaj, H.M., Mai, A.R., Simpson, J.M., Choi, I., Heinz, M.G., & Shinn-Cunningham, B.G. (2019). Non-Invasive assays of cochlear synaptopathy - candidates and considerations. Neuroscience, 407, 53-66. doi:10.1016/j.neuroscience.2019.02.031

Bramhall, N., Beach, E.F., Epp, B., Le Prell, C.G., Lopez-Poveda, E.A., Plack, C.J., . . . Canlon, B. (2019). The search for noise-induced cochlear synaptopathy in humans: Mission impossible? Hearing Research, 377, 88-103. doi:10.1016/j.heares.2019.02.016

Bramhall, N.F., Konrad-Martin, D., McMillan, G.P., & Griest, S.E. (2017). Auditory brainstem response altered in humans with noise exposure despite normal outer hair cell function. Ear and Hearing, 38(1), E1-E12. doi:10.1097/Aud.0000000000000370

Charaziak, K.K., & Shera, C.A. (2017). Compensating for ear-canal acoustics when measuring otoacoustic emissions. Journal of the Acoustical Society of America, 141(1), 515-531. doi:10.1121/1.4973618

Deeter, R., Abel, R., Calandruccio, L., & Dhar, S. (2009). Contralateral acoustic stimulation alters the magnitude and phase of distortion product otoacoustic emissions. Journal of the Acoustical Society of America, 126(5), 2413-2424. doi:10.1121/1.3224716

Henin, S., Thompson, S., Abdelrazeq, S., & Long, G.R. (2011). Changes in amplitude and phase of distortion-product otoacoustic emission fine-structure and separated components during efferent activation. Journal of the Acoustical Society of America, 129(4), 2068-2079. doi:10.1121/1.3543945

Kujawa, S.G., & Liberman, M.C. (2009). Adding insult to injury: Cochlear nerve degeneration after “temporary” noise-Induced hearing loss. The Journal of Neuroscience, 29(45), 14077-14085. doi:10.1523/ 

Liberman, M.C., Epstein, M.J., Cleveland, S.S., Wang, H.B., & Maison, S.F. (2016). Toward a differential diagnosis of hidden hearing loss in humans. Plos One, 11(9). doi:ARTN e016272610.1371/journal.pone.0162726

Miller, J.A.L., Reed, C.M., Robinson, S.R., & Perez, Z.D. (2018). Pure-tone audiometry with forward pressure level calibration leads to clinically-relevant improvements in test-retest reliability. Ear and Hearing, 39(5), 946-957. doi:10.1097/Aud.0000000000000555

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. Journal of the Acoustical Society of America, 135(1), 287-299. doi:10.1121/1.4845415

Rao, A., & Long, G.R. (2011). Effects of aspirin on distortion product fine structure: interpreted by the two-source model for distortion product otoacoustic emissions generation. Journal of the Acoustical Society of America, 129(2), 792-800. doi:10.1121/1.3523308

Shera, C.A., & Guinan, J.J.Jr. (1999). Evoked otoacoustic emissions arise by two fundamentally different mechanisms: a taxonomy for mammalian OAEs. Journal of the Acoustical Society of America, 105(2 Pt 1), 782-98.

Valero, M.D., Hancock, K.E., Maison, S.F., & Liberman, M.C. (2018). Effects of cochlear synaptopathy on middle-ear muscle reflexes in unanesthetized mice. Hearing Research, 363, 109-118. doi:10.1016/j.heares.2018.03.012