 Hi, my name is Ken Grant and I'm here from Malta Reed National Military Medical Center. I'd like to thank the organizers for giving us the chance to tell our story today entitled Functional Hearing and Communication Deficits in blast exposed service members with normal to near normal hearing thresholds. Here's our mandatory disclaimer slide. Now, like all stories, we need some context. Between 2007 and 2009, the US was at the height of its wars in Iraq and Afghanistan and service members were returning home with horrific injuries, mostly due to blast explosions. In addition, traumatic brain injuries labeled the signature wound of these conflicts was also prevalent with roughly 400,000 confirmed cases between 2000 and 2018. Now, because most of these service members had clinically normal hearing thresholds, they were sometimes assessed for a central auditory processing disorder. And since their audiogram looked normal, they were not considered at this time for as a hearing aid candidate. Well, this brings us to our question. What about all the service members that do not see clinical service? They're exposed to loud noise and blasts as part of the regular training and deployments and wouldn't they be expected to have hearing difficulties and noise like their clinical counterparts? Now, given that annual hearing physicals rarely include anything more than an audiogram, what are we missing and how large a problem is this? So given this context, our first of three challenges was to identify service members with normal to the normal audiograms who at risk for having functional hearing and communication deficits. Specifically, we wanted to know the prevalence of FHCD in this population and the risk factors. Once we're able to identify an at risk individual with functional hearing problems, what do we do to further assess potential sources of FHCD? We can't assume the peripheral encoding of sound is okay just because the audiogram is normal. We also must consider the likely possibility that there may be some cognitive impairment due to the blast even in service members who would never diagnose with a TBI, traumatic brain injury. Our third challenge is to take the results from our hearing and communication assessments and turn them into recommendations for treatment plans for our clinicians. We tested roughly 3,400 service members with normal to near normal hearing thresholds. We measured two tests of speech and noise and two tests of binaural integration as well as a subjective survey. And we calculated the failure rate for each of these tests as the proportion of subjects with performance scores less than or equal to the bottom 5th percentile performance of a control group, reporting no history of blast exposure and having audiometric thresholds of 20 dB or less. We subdivided our subjects into six groups, three blast categories and two levels of hearing threshold. Here we look at the risk factors associated with blast severity and hearing threshold. Last exposure and its severity tended to increase the probability of performing poorly on one or more tests, especially the survey responses. Here, the probability of a blast close subject performing like one of the bottom 5th percentile subjects of the controls was increased by a factor of four. Even slight elevations in hearing threshold led to a substantial increase in failure rates. Especially the NOS PI condition and the subjective survey. For blast exposed subjects with slightly elevated hearing thresholds, failure rates increased for all tests dramatically, especially when the blast was close enough to feel. In this follow on study, we tested three groups of subjects. A control group just like the previous one. We also tested two groups of blast exposed service members. Both blast groups had clinically normal hearing thresholds. One blast group was a low performing group on a composite score of our time compressed speech and NOS PI tests or on our subjective survey test. The other blast group was a high performing group performing within the normal limits on every test that we gave in this part of our screen. And all their subject will mail and all blast exposures except for one per group were blast close. The assessment battery was designed to test both sensory bottom up coding as well as cognitive top down processing. The sensory bottom up hypothesis was motivated by the groundbreaking work of Kejawin Lee-Min-Hoo who demonstrated that even with normal hearing thresholds, the neural encoding of super threshold sounds can be greatly impaired following noise exposure. The anticipated effects of super threshold distortion on speech recognition are shown in the following cartoons. In simple backgrounds with one or more speakers control subjects are expected to do pretty well without much difficulty. However, service members with degraded super threshold processing might be expected to do fair with a single speaker, although the task might be difficult. As the number of interfering speakers increase, service members will likely experience a sense of a greater effort as well as less confidence that they actually heard what the message was intended, the intended message, sorry. In contrast to a sensory bottom up hypothesis, high intensity shock waves associated with blast can cause brain injury, the cortical networks and their connections with other networks without any visible signs of injury. This cartoon service members with blast-induced cognitive deficits may find any number of competing speakers problematic regardless of whether hearing thresholds were compromised or not. Within our three test groups, control subjects tended to be significantly younger. To account for age differences across the group, we systematically removed subjects from the control group until we had an age match control sample with a cumulative distribution of ages and cohort size that best matched the two blast groups. And even with age matching, we again note that the three groups had dimension, demonstratively different hearing threshold profiles where control subjects are better than our high-performing subjects and our high-performing subjects have thresholds that are better than a low-performing subject. The same pattern was seen in the DPLAE results. Now, our tests of peripheral and auditory central auditory processing included behavioral and physiological measures. And poor performance on these tests would indicate a likely sensory coding problem. Test of cognitive capacity would divide into four clusters including attention, processing speed, working memory and memory. And many of these tests were administered in the visual modality and that's indicated by our red font. And we did this deliberately to avoid confusion when interpreting cognitive performance as every one of these tests can be negatively impacted by signal distortion caused by poor sensory encoding. This figure shows the results from a selected set of tests that were used to first assign subjects to their respective groups on the upper left. And again, for this talk, we consider only those tests with a clear difference between our low-performing subjects and our control subjects or when the failure rate of a given test was much worse than controls. Surviving tests were analyzed and results converted to Z-scores. Their sign changed so that up going is always worse and the Z-scores were converted based on the mean and standard deviation for the larger control group. Test highlighted within these red boxes indicate tests where the average blast low performance is much worse than either the blast high or the blast or the control subjects. And the blast high performance look just like the controls. Deficits observed in clinical audiometry tests, upper right and frequency following metrics, the lower right are consistent with reduced peripheral and brainstem encoding of auditory input. Tests within this blue box show a different pattern where blast high and blast low performance were similarly impaired and both worse than the control subjects. Within this grouping, the speed of capacity of language processing where the scope test stands out supporting an additional, which supports suggesting, sorry, an additional cognitive deficit. Now the results of the frequency following results are particularly intriguing in that the response to the DAW syllable was slightly reduced compared to the blast low and blast high group. Their internal noise when there's no signal was slightly elevated compared to the control and blast high performing group. And when you put these two together, you end up with a internal signal to noise ratio that's porous for the blast low group and equal for the blast high and control groups. The bottom panel, like the middle, shows an auditory processing deficit only for the blast low group. And specifically this measure shows the consistency of response consistency of response to repeated presentations of the exact same stimulus. This last challenge has to do with recommendations for treatment options for reducing FHC data. DoD and VA audiologists haven't had many options in this area. Auditory and cognitive training is a good bet, but patients have to show an ability to comply with training regimens that can often be quite onerous. Another option growing in popularity are low gain hearing aids because these service members have normal to new normal hearing thresholds. The amount of gain is very light, is very slight. The average, sorry, the challenge facing research is to uncover why these low gain hearing aids might be working given that audibility is not a problem. Now, for research community to provide information of real value to the clinical audiologist, we have to either reduce the burden of assessing CAPD or we have to provide a rational basis for choosing among specific therapies that are available. Recent surveys administered to DoD and VA audiologists reveal that the CAPD diagnosis is rarely given to service service even after hours and hours of testing. And I'm sorry, we're wondering what if a quick, like what we've done, what if a quick screening test to inform the clinician of whether or not to go into a deeper dive into auditory function, whether that would yield benefit for the patient. It's possible that a quick screening measure of speech and noise and binaural integration, as well as something like our survey may provide value and save provider and patient time. Excuse me. Although more research is needed in this area, we feel that there are interesting results in the current study to guide clinicians. For example, test pointing to peripheral encoding problems would suggest the signal processing solution to reduce FHCD and listener effort. The FFR results indicate a poor internal SNR for the blast low subjects providing gain in this case would amplify the speech signal without affecting the internal noise, thus improving the internal signal to noise ratio and the encoding quality. We have some preliminary data to support the use of low gain hearing aids for this population. This figure from a recent dissertation from Melcox Ryan shows a clear benefit of low gain hearing aids for recognizing both standard speech and noise tests as well as time compressed and reverberant speech and noise tests. Now as a last cautionary note, the service members encountered as part of our studies that come in for their annual hearing physicals and patients seen at audiology and speech language pathology clinics and military hospitals and patients seen at VA clinics may be very different populations. In particular, this last slide contrasts the population seen as part of our prevalence study and patients seen in our audiology clinic that were there for a CAPD evaluation. In the clinic, there's a greater percentage of patients who present with elevated hearing thresholds. They also present with higher levels of blast exposures and higher levels of traumatic brain injury and higher levels of PTSD. Managing these patients compared to service members appearing for their annual health exams may lead to very different solutions to this challenging problem. Naturally to carry out this work, there's a lot of people to thank from these different locations that helped us with this multi-site study and I've listed our funding sources here. And thank you and I'll take questions if there's any time left.