Evaluation of Auditory Function in a Population of Clinically Healthy Cats Using Evoked Otoacoustic Emissions

Abstract

Cats may demonstrate deafness due to a variety of aetiologies and the current preferred method for assessing auditory function is the brainstem auditory evoked response (BAER). The BAER has largely been replaced by otoacoustic emission (OAE) testing in human neonatal deafness screening as the equipment is more readily available, is cheaper and the test is less invasive and simpler. This is the first study to demonstrate that transient evoked OAEs (TEOAE) and distortion product OAEs (DPOAE) can be recorded in cats using commercially available equipment. Protocols for recording the emissions and analysing the results are given. DPOAE testing is suggested to be quicker in this population of healthy cats and shows promise in rapidly providing detailed information about auditory function at a variety of different frequencies.

Cats suffer from a variety of causes of sensorineural deafness, including congenital sensorineural deafness, presbycusis and ototoxicity, as well as deafness due to the disturbed conduction of sound caused by otitis media and nasopharyngeal polyps.^1–3 Pigment-associated congenital deafness in white cats is recognised in both pure and mixed breed cats with a prevalence in pure breed white cats of up to 20%, ⁴ and an association between congenital deafness and blue irises in these cats has long been recognised. ^5–7 Inherited congenital deafness has not been reported in non-white cats and the deafness prevalence in the wider cat population is not known. Despite extensive study, the genetics of congenital sensorineural deafness remains incompletely understood. ^5,6,8 Deafness in cats may results in substantial welfare considerations with complete bilateral deafness increasing the risk of death or injury as deaf cats are unable to hear approaching danger.^9,10 The inability of kittens to hear their mother or vice versa may result in kittens becoming separated from or neglected by the queen. Deaf animals are easily startled resulting in an increased risk of bite injuries to humans. ¹ Unilaterally deaf cats cope better but may have difficulty localising sounds.^1,9 Despite these problems, deaf cats can make good pets, although generally should be kept indoors. ⁹

The current test of choice for identifying cats with congenital sensorineural deafness, and evaluation of other causes of deafness, is the brainstem auditory evoked response (BAER).^1,6,9 The BAER is performed in sedated or anaesthetised cats and involves the placement of subdermal needle electrodes to measure electrical brain activity in response to auditory stimuli.^1,6,9,11 However, this test requires expensive equipment, the availability of which is restricted to specialist institutions. ⁹ Unlike in dogs, congenital pigment-associated deafness in cats may be partial,^12,13 making BAER screening more complex and time-consuming. In human audiology, otoacoustic emission (OAE) tests have largely replaced the BAER as the first choice screening test in neonates. ¹⁴ The OAE tests are also widely used for diagnostic hearing testing in human patients. The OAE tests are inexpensive, rapid, non-invasive tests of auditory function over a range of frequencies.

OAEs are low amplitude sounds produced by the outer hair cells of the cochlea that can be measured by placing a probe containing a microphone in the external ear canal. OAEs can be divided into several types depending on the stimuli used to elicit them. The two forms of OAEs most commonly used for hearing testing are click-evoked transient evoked OAEs (TEOAEs) and distortion product OAEs (DPOAEs). For TEOAEs, broadband click stimuli activate the cochlea simultaneously over a wide frequency region and following amplification and averaging result in recordable emissions, actively generated by the outer hair cells, with frequencies between 1 and 4 kHz in a normal adult human ear. ¹⁵ During TEOAE testing, time-locked responses to multiple broadband click stimuli are recorded after stimulus presentation and are alternately placed in one of two recording bins. Differences between the two generated waveforms represent noise within the recording. To determine whether a response is present, the sound-to-noise ratio (the difference between the mean noise and the mean response recorded) at different frequencies or in different frequency bands is calculated. The larger the sound-to-noise ratio, the more likely a response is present. In addition, the whole response reproducibility is derived from the correlation between the two overlaid waveforms and gives a measure of the reproducibility of the response during the recording period. High reproducibility can be used as further evidence of adequate cochlear function. Fast fourier transformation of the recorded waveform enables the OAE and noise energy to be displayed in the frequency spectrum. Emissions of different frequencies reflect activity of outer hair cells at different distances from the base of the cochlea and, therefore, despite the use of broadband click stimuli, information on the integrity of the cochlea in different frequency regions can be obtained simultaneously allowing the diagnosis of frequency specific hearing loss. ¹⁵ TEOAEs have high test—retest reliability, ¹⁶ can be recorded in essentially all normally hearing human ears,^17,18 and are typically absent at frequencies at which audiometric hearing thresholds exceed 20–30 dB HL.^15,19

DPOAEs are elicited by the simultaneous, continuous presentation of pairs of closely spaced frequencies (denoted f₁ and f₂, where f₂ > f₁₎ that evoke recordable emissions from the normal cochlea at other predictable frequencies. In people, the most robust emissions are generated at 2f₁ — f₂ and an f₂ to f₁ ratio of 1.2 is commonly used to generate these emissions. ¹⁶ By varying the frequencies of the paired stimuli, the integrity of the cochlea in different frequency regions can be tested. Like TEOAEs, DPOAEs have good tester—test reliability, ¹⁶ and can be recorded in the vast majority of normally hearing human ears, however, their amplitude is reduced in patients with hearing loss of between 15 and 45–55 dB HL,^18,20 and they are absent at audiometric hearing thresholds greater than this level.^19,20 These properties make DPOAE testing suitable for documenting partial hearing loss and for monitoring changes in outer hair cell function over time as well as for hearing screening.

Despite their extensive use in people, little work has been done on OAE testing in companion animals. There are no studies reporting the use of click-evoked TEOAEs in cats and although experimental work has been performed in cats using DPOAE testing for physiological studies,^21–25 and to look at the pathological effect of noise on the cochlea,^26–28 the majority of these studies involved surgical manipulation of the ear canal, tympanic bullae or brain. No clinical studies of evoked OAEs have been performed in cats with a view to their use in the diagnosis of deafness in this species.

The aims of this study were to investigate whether (1) click-evoked TEOAEs and (2) DPOAEs can be elicited and recorded in cats within a clinical situation, and (3) to assess the clinical utility of these tests in the assessment of cochlear function in a population of cats with apparently normal hearing.

Materials and methods

This study was approved by the Ethics and Welfare Committee of the School of Veterinary Medicine, University of Glasgow.

Animals

Eighteen cats undergoing routine surgical procedures at two Scottish Society for the Prevention of Cruelty to Animals (SSPCA) clinics were included in this study. All animals demonstrated a normal response to behavioural assessment of hearing (ear movement [Preyer's reflex] or a head turn in response to a sound stimulus), no evidence of neurological disease and no history of ear, or apparent hearing, problems. OAE testing was performed in the cat kennel room, during the anaesthetic recovery, after extubation. The time available for OAE testing was dependent on the time the administered anaesthetic agents took to wear off and the demeanour of the cat as head shaking and excessive movement could result in displacement of the probe from the ear or excessive noise. All cats were tested in sternal recumbency. The approximate age, gender, neuter status, eye colour, coat colour, weight, surgical procedure performed and sedative, anaesthetic and analgesic agents administered were recorded. Otoscopic examination was performed to ensure normal patency of the external ear canals. Ears were cleaned with dry cotton buds to remove excess cerumen.

Equipment

OAE testing was performed using the Echoport ILO 288 USB system with V6 software (Otodynamics, Hatfield, UK) installed on a laptop computer. An Otodynamics UGD TEOAE+DPOAE probe (Otodynamics, Hatfield, UK) was used and the calibration was tested daily prior to use, by placing the probe in a 1 cc test cavity and running the calibration test. If the results were out with the manufacturer's recommendations, the probe was replaced. An appropriately sized rubber probe tip was chosen and the probe was placed in the ear. An ear muff EP-101 (Parkson Safety Industrial Corporation, Taipei, Taiwan) was gently held against the test ear throughout the test to reduce environmental noise. Environmental noise was further minimised by closing the doors and ensuring only the tester was in the room during the test procedure.

TEOAE testing

Prior to testing, the amplitude, polarity, duration and frequency of the stimulus was assessed using the machine's Checkfit function. Optimal probe placement resulted in a short positive and negative deflection followed by minimal oscillation of the waveform and a smooth rounded frequency spectrum spanning the 1–4 kHz range. The probe position was altered until optimum placement was achieved. The gain setting was altered until it approximated 90 decibels sound pressure level (dB SPL). The non-linear Quick screen software mode (Otodynamics, Hatfield, UK) was used for data collection. The device and software automatically acquired and analysed responses, producing a spectral record of the emissions (sound) and the mean background noise in half octave frequency bands (centred at 1 kHz, 1.5 kHz, 2 kHz, 3 kHz and 4 kHz). The responses to alternate clicks were stored in two averaging bins and the overall reproducibility was calculated by the Echoport software by comparing the waveform of the two bins and was reported as a percentage. The higher the value, the more similar the waveforms were. A rejection level of 10 mPa was used and results with a noise level greater than this threshold were discarded. Stimulus stability was measured by the software throughout data collection by comparing the stimulus waveform at the start of the test to that of the current stimulus. The stimulus stability is given a value between 0 and 100. If the initial stimulus waveform and current stimulus waveform are identical, the value would be 100. If the value dropped below 75 the test was aborted. The stimulus stability values given for each run are those generated by the Echoport software which is the stability at the end of the run. Data collection continued until either: (a) passes were obtained in all frequency bands and a minimum of 30 low noise samples were collected, or (b) 260 low noise responses were collected, whichever occurred first.

For each run, the response and noise in each of the five frequency bands, whole response reproducibility, stimulus stability, achieved stimulus intensity and run duration were recorded by the software. Runs were considered valid if:

1. Achieved stimulus intensity was between 87.5 and 92.5 dB SPL, and

2. Stimulus stability was greater than 75%.

The sound-to-noise ratio was calculated for each frequency band by subtracting the noise from the response in that frequency band. A pass for a frequency band was defined as a sound-to-noise ratio ≥6 dB, except for the 1 kHz band in which a pass was defined as a sound-to-noise ratio ≥3 dB. An overall pass for the run was achieved when three or more frequency bands achieved a pass. One run was performed in each ear. If the run was invalid, the results were discarded and the run repeated. If the run was valid but failed, the probe was removed, the coupling tubes and probe tips replaced, the ear cleaned if necessary and the run repeated. If the run was valid and passed, it was not repeated. A maximum of three runs were performed in each ear.

DPOAE testing

Twenty-seven pairs of primary tones (f₁ and f₂₎ were used to generate DPOAEs. The ratio of the frequency of these tones (f₂/f₁) was 1.21. The primary tone f₂ varied from 0.84 to 8 kHz with eight pairs per octave. The tones were presented sequentially from highest to lowest frequency. The whole sequence of tones was repeated three times (three sweeps) before recording was manually stopped. The DPOAE response at 2f₁ — f₂ and the noise around the 2f₁ — f₂ frequency region were recorded for 1.54 s for each pair of frequencies for each sweep. If at any time, the noise exceeded the rejection level of 10 mPa, then no response was recorded during that time. If the noise exceeded the rejection level throughout the 1.54 s for all three sweeps, no result was recorded for that frequency pair.

A pass for a pair of frequencies was defined as a response exceeding the mean noise plus two standard deviations by more than 3 dB (ie, a sound-to-noise ratio >3 dB) and an overall pass for the run was achieved if five or more of eight specific, pre-selected frequency pairs achieved a pass. The f₂ of the eight frequency pairs included were 1.54, 1.83, 2.19, 2.60, 3.09, 3.66, 4.36 and 5.19 kHz. If, due to excessive noise, no data was collected for three or more frequency pairs, the run was excluded. The overall outcome for a run was defined as a pass if the above criteria were met or a fail if they were not. Due to time constraints, only one DPOAE run was performed in each ear unless no response at any frequency was seen after the first sweep in which case the run was aborted, the probe removed, the probe tip and coupling tubes changed, the ear canal cleaned (if necessary) and the run restarted. Data collected for each run included the sound-to-noise ratio for each frequency pair, the frequency pairs with no data due to excessive noise and the overall outcome.

Data analysis

For TEOAE testing, the number of ears passing after one, two and three runs was calculated. The final outcome of the TEOAE test for an ear was defined as the result of the last run performed. The number of runs achieving a pass in each frequency band and the median whole response reproducibility were calculated. The median and range of test times for all TEOAE runs were also calculated. For the DPOAE test, the number of ears passing and failing was established. To simplify the results of the runs that passed, only data from 13 pre-selected pairs of frequencies (four pairs per octave) out of the 27 pairs collected were presented in the descriptive results. The number of ears reaching the sound-to-noise threshold of more than 3 dB for the 13 frequencies, and the numbers of ears with insufficient data collected due to excessive noise at each of the 13 frequencies, were determined. An estimate of the test time required to achieve the data necessary for the pass/fail protocol (ie, to collect data for eight frequency pairs) was calculated by multiplying the number of frequencies (eight) by the number of sweeps (three) and the time for each sweep for each frequency (1.54 s). The final outcome for each ear for the TEOAE test (after a maximum of three runs) and the DPOAE test (after one run) were compared.

Results

Eighteen cats were included in the study, comprising 17 domestic shorthair cats and one domestic longhair cat. None had blue eyes nor completely white coats, although nine had white patches. Twelve cats were estimated to be under 1-year-old, four between 1 and 5 years old and two more than 5 years old. Eleven were male (10 entire, one neutered) and seven were female (five entire, two neutered). All had been chemically restrained with medetomidine, ketamine and buprenorphine for a variety of procedures including castration (n = 9), ovariohysterectomy (n = 4), leg amputation (n = 1), umbilical hernia repair (n = 1) and dental treatment (n = 2). In nine cats, an endotracheal tube had been placed during the procedure. All cats had been given a non-steroidal anti-inflammatory, either carprofen (n = 16) or meloxicam (n = 2). Behavioural testing of hearing was normal in all cats following recovery from anaesthesia.

Click-evoked TEOAEs can be elicited and recorded in cats within a clinical setting

At least one valid TEOAE run was obtained from both ears of all cats, giving 36 ears in the analysis (Fig 1). Eighty-nine percent of the ears (32/36) passed after a maximum of three runs. Of the four ears that failed, only one had had all three runs performed and only one ear showed no evidence of cochlear function (ie, failed all five frequencies). This ear had two runs performed. The maximum, minimum and median sound-to-noise ratios at each frequency for the 32 runs that passed are shown in Fig 2. Of the 32 runs, 13 (41%), 19 (59%), 30 (94%), 31 (97%) and 25 (78%) passed in the 1kHz, 1.5kHz, 2kHz, 3 kHz and 4kHz frequency bands, respectively. Within the 1 kHz band there is a wide range of sound-to-noise ratios and the median sound-to-noise ratio for this frequency band is almost 0 dB (Fig 2). The median whole response reproducibility for all 32 runs that passed was 90.5% (range: 70–99%). The median test time for all TEOAE runs (regardless of pass or fail) was 69 s (range: 9–137 s).

Fig 1.

The outcome of TEOAE testing (three runs) and DPOAE testing (one run) in 36 ears from 18 cats. The number of ears included is shown above each bar. TEOAE testing was repeated if the run was failed, up to a maximum of three runs.

Fig 2.

Box and whisker plot of the sound-to-noise ratios in all five frequency bands (centre of bands shown) for the 32 TEOAE runs that achieved an overall pass (ie, a pass in more than three frequency bands). The boxes represent the interquartile range, the line within the box represents the median and the whiskers represent the range of values (dB = decibels).

DPOAEs can be elicited and recorded in cats within a clinical setting

Thirty-five ears were included in this analysis (in one ear the stimulus settings were incorrect and this ear was excluded). Noise exceeded the rejection level throughout recording for three or more of the eight frequencies for three ears and these were excluded due to excessive noise. Twenty-nine ears out of the remaining 32 passed DPOAE testing, giving a pass rate of 94% after one run (Fig 1). Of the three ears that failed, all showed some evidence of cochlear function, passing 1, 3 and 4 frequency pairs. The maximum, minimum and median sound-to noise ratios for the 13 frequency pairs of the 29 ears that achieved an overall pass for the DPOAE testing are shown in Fig 3. Frequency pairs for those ears in which no data was collected due to noise exceeding the preset rejection threshold were excluded from Fig 3. The number of ears included at each frequency, the median sound-to-noise ratio for the included ears and the number and percentage of ears passing at each pair of frequencies is shown in Table 1. The overall test time for DPOAEs was 123 s, however, 27 frequency pairs were presented.

Fig 3.

Box and whisker plot of the sound-to-noise ratios for 13 frequency pairs from the 29 ears achieving an overall pass for the DPOAE test (ie, a pass in more than five of eight frequency pairs). The boxes represent the interquartile range, the line within the box represents the median and the whiskers represent the range of values (dB = decibels). The shaded boxes are those frequencies included in the pass/fail analysis protocol. The number of runs included at each frequency is shown in Table 1.

Table 1.

DPOAE testing in 29 apparently normal cat ears in which an overall pass was achieved. The median sound-to-noise ratio for the included ears, the number of ears included for each pair of test frequencies and the number and percentage of ears passing at each pair of frequencies is shown. The shaded rows represent the eight frequency pairs included in the pass/fail analysis

f2 frequency (kHz)	Median sound-to-noise ratio (dB)	Number of ears included	Number of ears passing	Percentage of ears passing (n = 29)
0.92	—6.30	25	4	13.8
1.09	—5.00	25	5	17.2
1.30	0.10	27	12	41.4
1.54	8.60	28	19	65.5
1.83	14.80	29	25	86.2
2.18	16.70	29	28	96.6
2.59	20.90	29	29	100
3.08	17.40	29	27	93.1
3.67	15.20	27	26	89.7
4.36	14.80	29	25	86.2
5.19	12.20	29	24	82.8
6.17	19.80	29	29	100
7.34	20.20	29	29	100

TEOAE and DPOAE assessment of cochlear function correlates well with behavioural assessment of hearing in a population of apparently normal cats

Out of the 36 ears, 35 ears passed either the TEOAE testing, the DPOAE testing or both. The outcome for the final TEOAE and DPOAE test for each ear are shown in Fig 4. Of the 36 ears assessed, 26 passed both tests, one failed both tests, three failed the TEOAE but passed the DPOAE, two passed the TEOAE but failed the DPOAE and four passed the TEOAE but were excluded from the DPOAE test.

Fig 4.

The number of frequency bands/pairs passed for each ear in the TEOAE and DPOAE tests to show a comparison of the outcome of the two tests. Ears represented by the symbol ‘•’ had results from both tests, those represented by the symbol × did not have results for DPOAE testing and are shown to the right of the graph.

Discussion

This is the first study examining the feasibility of evoked OAE testing for the assessment of feline hearing within a clinical environment. Valid TEOAE tests were achieved in all ears tested and robust click-evoked TEOAEs were recorded in 32/36 ears after a maximum of three runs and in 29/32 ears with one run of DPOAE testing. All cats in this study had normal behavioural responses to sound and comparison of the results of the TEOAE and DPOAE tests suggested that all but one ear had adequate cochlear function. This study provides test and analysis protocols for TEOAE and DPOAE testing in cats.

When using TEOAE testing a high failure rate (about 30%) was evident if only one run was used, however, if ears that failed were immediately retested up to a maximum total of three tests, 89% of ears achieved a pass in this study. Both TEOAE and DPOAE tests are widely used in human neonates for hearing screening.^29–34 Children that do not achieve a pass after the first test are often retested before being referred for further hearing assessment.^29,30 In children, a pass on a TEOAE test using less stringent pass/fail criteria to those used in this study is considered evidence of adequate auditory function and neither repeat TEOAE testing, nor BAER testing is usually performed. ³⁰ Whole response reproducibility may be used as additional evidence of cochlear function, with a value of over 50% suggesting that a cochlear emission is present. ¹⁶ All TEOAE runs that passed in this study had a whole response reproducibility of 70% or more. In contrast to the other frequency bands, less than 50% of ears passed at 1 kHz despite using a pass criterion of only 3 dB for this frequency band. Possible explanations include louder noise at low frequencies, an imbalance between the pressure in the outer and middle ear, a weaker stimulus intensity at lower frequencies or a poor seal of the ear canal by the probe. ¹⁵ In addition, a wide range of sound-to-noise ratios were found amongst the runs that achieved an overall pass and many ears did not achieve a response greater than the mean noise (ie, sound-to-noise ratio was ≤3 dB). These findings suggest that removal of the 1 kHz frequency band from the stopping criteria in the future would help to reduce the test time.

As with TEOAE testing, a pass for one run of DPOAE testing in children is considered evidence of adequate auditory function. In some protocols an overall pass is based on demonstrating that the amplitude of emissions for at least three to five frequency pairs fall within a normative range. ¹⁸ As normative data was not available for cats, a pass for a frequency pair was determined as a response exceeding the mean noise plus two standard deviations by a minimum of 3 dB. ³¹ This was required for a minimum of five of the eight pre-determined frequency pairs. Even though only one run of DPOAE tests was performed in each ear, the DPOAE test had a high pass rate, suggesting that DPOAE testing is feasible in sedated cats in a clinical environment. Repeated DPOAE testing would likely further improve the pass rate in hearing ears. Three ears were excluded from the DPOAE analysis due to excessive noise during the collection period; however, passes were achieved with TEOAE testing in these ears. Repeating the test after ensuring a good probe seal and after moving to a quieter environment might have enabled testing in these ears. DPOAE testing was performed after TEOAE testing and the animals may also have been less sedate at this stage. Similar to TEOAE, less than 50% of ears achieved a pass during DPOAE testing for low frequency pairs (with f₂ frequencies less than 1.5 kHz). Incorporating low frequency pairs into the analysis protocol appears to be of little benefit for hearing screening in cats. Similarly, it is recommended that frequency pairs below 1.5 kHz are not included in human neonatal screening protocols. ³² Results for two high frequency pairs (above the range used to define a pass) were also described in this study. Both demonstrated high sound-to-noise ratios and a high percentage of ears passed at these frequencies. However, including these in the test protocol would have increased test duration and may not have improved the detection of normal cochlear function. The use of high frequency stimuli for DPOAE generation (above 5 kHz) may also result in artifactual responses, either through generation of standing waves within the ear, ¹⁸ or through non-active processes.^33,34

Test duration is important if OAE testing is going to be adopted in clinical practice. The time taken to perform one TEOAE run was a median of 69 s. The total estimated TEOAE test time, if the maximum of three tests were performed in each ear would still be less than 10 min per cat. However, many cases would pass on the first TEOAE run and refinement of the TEOAE test protocol (such as removing the 1 kHz frequency band from the stopping criteria) would further reduce this. The overall test time for a DPOAE run was 123 s, making TEOAE runs shorter than DPOAE runs in this study; however, 27 frequency pairs were presented for the DPOAE test. The estimated DPOAE test duration if only the eight frequency pairs included in the pass/fail analysis were presented was 37 s and if three runs are required for TEOAE testing and only one run of eight frequency pairs are presented for DPOAE testing, the overall test time would be considerably shorter for DPOAE testing.

A comparison of the outcomes for the TEOAE tests and the DPOAE test revealed discordant results in five ears (Fig 4). As false passes are unlikely with either test, it is likely that the failed test runs were false fails. The pass criteria were deliberately stringent in order to correctly identify any ears with reduced cochlear function, but the consequence was that some ears may incorrectly be defined as having reduced cochlear function. Of the three ears that failed the TEOAE test but passed the DPOAE test only one had no evidence of a TEOAE response; however, it is possible that excessive environmental noise or patient movement prevented the detection of a response despite adequate cochlear function. Two ears failed the DPOAE test (but passed the TEOAE test) and both had evidence of cochlear function in at least one frequency, again it is possible that high noise levels may have masked the DPOAE response at other frequencies in these ears. One ear in this study failed to achieve a pass in either the DPOAE or TEOAE tests. It achieved a pass for only the highest of the eight frequency pairs for the DPOAE and on both TEOAE runs performed, passed only the two highest frequency bands. The other ear of this cat passed both tests. This cat was one of the oldest cats in the study and it is possible that this cat had acquired, unilateral hearing deficits in this ear (particularly affecting low frequencies) which would have been difficult to detect with behavioural testing.

This study had various limitations, including the limited available history (these were rescued cats) and lack of concurrent BAER. BAER testing was considered an invasive test and could not be justified without any clinical suspicion of deafness. Inherited congenital deafness was unlikely in these cats as none had a completely white phenotype or blue eyes. Acquired hearing deficits were possible, although given the young population age (12/18 cats were less than 1-year-old) and the absence of clinical evidence of aural or neurological disease, the prevalence was likely to be low.

This is the first study to investigate the use of evoked OAEs for hearing testing of cats in a clinical environment. This study demonstrates that both TEOAE and DPOAE testing are feasible within the clinical setting in sedated cats. Analysis of the test parameters suggests that DPOAE testing may be quicker and more efficient than TEOAE testing for hearing screening in cats, although further investigation in cats with hearing impairment and which are undergoing concurrent hearing assessment with BAER is necessary.