Non-linear Frequency Compared to Conventional Processing in Patients With and Without Cochlear Dead Regions.

Non-linear Frequency Compared to Conventional Processing in Patients With and Without Cochlear Dead Regions.

A Comparison of Conventional Hearing Aid Processing (CP) and Non-linear Frequency Compression (NLFC) Processing in Adult Patients With and Without Cochlear Dead Regions Who Have a Moderate to Severe Sensorineural Hearing Loss.

The investigators are interested in an additional hearing aid feature called non-linear frequency compression (NLFC). This aims to improve audibility of high frequency sounds by converting them into lower frequencies and has been shown to benefit those with moderate-severe sensorineural hearing loss (SNHL). Cochlear dead regions (DRs) are areas of the inner hearing organ (the cochlea) where there is little or no function and are commonly found in regions responsible for detecting high pitched (frequency) sounds. Not all people with hearing loss have DRs. The investigators would like to determine whether based on the presence or absence of DRs, patients with moderate-severe SNHL perform better or prefer their hearing aids programmed conventionally, or with NLFC activated. To do this, two participant groups will be created based on findings from the threshold equalising noise (TEN) test which identifies cochlear DRs. Two participant groups will be created; one group with DRs and one group without DRs. All participants will receive two hearing aids and will wear these programmed conventionally for ~six weeks and with NLFC activated for ~six weeks in a counterbalanced manner. Following each condition, participants will complete a questionnaire and various speech tests will be performed. This involves participants repeating sentences, words or speech sounds they hear from a speaker in quiet and in the presence of background noise. Individuals' scores will be calculated for each test and their performance when NLFC was activated and deactivated will be compared. This will be analysed alongside the questionnaire data to compare the 'DR' and 'no DR' group in both conditions. Findings may help to determine whether NLFC should be activated for all moderate-severe SNHL patients, or just those with DRs, helping clinicians to optimise hearing aid settings for patients.

Potential participants will be recruited from the audiology database at Manchester Foundation Trust. Those who meet the inclusion criteria will be sent a letter in the post accompanied by a participation information sheet to raise their awareness of the project. This criterion includes those aged 45 and over with a bilateral moderate to severe sensorineural hearing loss who wear two hearing aids. They must report to wear these for > 5 hours each day which, will then be confirmed by their hearing aid data logging. English must be their first language due to the test material that will be used. Patients will be provided with a participant information sheet, outlining what will happen in each of the four sessions and a consent form. This participant information pack will inform potential participants about how they may indicate their interest or contact the researchers for more information about what the study entails. The investigators will try and recruit 30 participants in total to allow for attrition in the event of withdrawal or exclusion from the study. The researchers hope to find 24 (12 in each group) eligible participants who will be able to complete the study. The researchers anticipate recruitment will be done over one month and data collection over four. It is possible that participants may be excluded from the study once testing begins if asymmetrical DRs are found (present at frequencies < 3kHz in one ear only) or any otological symptoms arise that prevent participants from wearing their aids for >5 hours a day. The sample size was taken based on pragmatic ground and time constraints as this is a MSc project. Participants who provide written, informed consent will attend a reassessment appointment (session one) a fitting appointment (session two) and a follow up appointment (session three). These appointments are routinely offered to all patients who are beginning a new hearing aid pathway. Additionally, the researchers are asking participants to attend one further follow up appointment (session four). Each session will last between one and one and a half hours. Session One- A reassessment is routinely offered to patients every three years and involves; A brief medical history to identify potential ototoxic medication, any otological symptoms and factors which may impact patient's ability to manage hearing aids. Establishing the patient's hearing aid history to determine use and understand the patient's lifestyle. Otoscopy (where the clinician will look in the ears) and tympanometry (where the clinician will assess the ear drums). A pure tone audiogram (PTA) to determine the patient's hearing thresholds. If required, impressions of the ears will be taken for new earmoulds to be made. Additionally, a TEN test will be performed in session one to assess for DRs. The TEN test procedure is similar to that of a standard PTA but involves ipsilateral masking. This prevents signal tones from being detected at other regions of the cochlea that surround the DR. By preventing off-frequency/place listening, a DR can be detected. Based on findings from the TEN test, participants will be split into two groups; those with significant bilateral cochlear DRs < 3kHz (the DR group) and those without cochlear DRs or a DR > 4kHz (the no DR group). Those with unilateral DRs < 3kHz will be excluded. Session two- During session two, all participants will be familiarised with the test material that will be used in sessions three and four. The adaptive Bamford Kowal Bench (BKB) English-language sentences, the Arthur Boothroyd (AB) word list and auditory speech sounds evaluation (ASSE) are all routine speech tests readily available within our audiology department. These will be the speech tests used to measure performance with NLFC and with CP. All testing will take place in a soundproof booth. None of these intensities should be too loud for participants as they aim to replicate normal conversational levels. Following any complains of loudness discomfort, the dB SPL will be reduced, and this will be recorded. The questionnaire used will be The SSQ12, a validated shortened version of the speech, spatial and qualities of hearing scale (SSQ) questionnaire that asks participants about their listening experiences in various situations. The SSQ12 will be shown to participants in session two, so they are aware of the situations they should be monitoring whilst wearing their hearing aids. BKB sentences will be played in the presence of noise presented at 55 A- weighted decibels (dBA) and the speech level adapts depending on whether the participant repeats the sentence correctly, or not. If two or more key words are repeatedly correctly, the response is considered correct. Following two correct responses, the speech intensity will be decreased. Following one incorrect response, the intensity of the signal will be increased. Participants will be given a score in dB which corresponds to the intensity level at which 71% of the sentences are correctly identified. The speech signal will always be presented from the front (0° azimuth). The noise signal will be presented from the front, from the left (-90° azimuth), and from the right (90° azimuth). One sentence list will be administered for the noise presented at 0°, +90° and -90° and an average overall score in dB will be calculated. The AB word list will be performed in the absence of background noise. Words will be played from a loudspeaker at 65 decibels sound pressure level (dB SPL). The test material consists of ten words per list and each word is made of three phonemes. A point is awarded for each correct phoneme (0,1,2,3) and the average score from three lists will be calculated. Additionally, participants will complete a phoneme discrimination task, ASSE at a level of 70dB SPL. They will be asked to discriminate between various phoneme pairs by a 'same/ different' procedure. Following three to eight presentations of a single speech sound, another speech sound will be played e.g. V-Z. If participants respond correctly to two out of three presentations, it can be concluded that the contrast between the background phoneme and the odd speech sound is well discriminated and this will be marked as correct. Seven pairs will be used, and participants will be given a percentage score based on how many they are successfully able to discriminate between. Also, during session two, participants will be fitted with bilateral Phonak Auto hearing aids. These will be programmed to NAL-NL2 prescription targets following departmental protocol and British Society of Audiology (BSA) recommended procedures, which includes making real ear measurements (REMs). NLFC will be programmed into all participants hearing aids but will be deactivated in half of the participants from each group in a randomised manner. This will be generated for each group separately as there will be 12 participants in the 'DR group' and 12 in the 'no DR group'. Using an Excel formula, participants in each group will be placed in a random order in a list. The first six participants listed in each group will experience NLFC first. The remaining six participants listed in each group will experience CP first. Participants will be unaware of how their aids are programmed. NLFC in the 'no DR' group will be validated according to BSA recommended procedures. NLFC in the 'DR group' will be set based on where their dead region begins. This will be set at 0.75 times their individual edge frequency (0.75Fe). Any fine tuning requested by participants of the hearing aids will be permitted up to 7 days following fitting. Session three- Around six weeks after the first fitting, session three will take place. This will consist of testing participants with their hearing aids set to their first condition using the adaptive BKB sentence test, AB word list and ASSE, as outlined above. Participants will also complete the SSQ12 in an interview style format. Following this, the participant's hearing aid settings will be swapped to the opposite condition i.e. if NLFC was activated in session two, this will be disabled in session three and vice versa. Any fine tuning requested by participants of the hearing aids will be permitted up to 7 days following fitting. Session four- Around six weeks later, session four will take place. This will involve testing participants with their hearing aids set to their second condition using the same test material and completing the same questionnaire as in session three. Participants will then be given a choice of which setting they would like permanently or can decide to go away with both settings programmed into the hearing aids and decide at a later date.

Participants will be split into two groups depending on whether or not they have 'cochlear dead regions'. All participants will experience both conventional processing and frequency compression for six weeks. The order of which will be randomised and counterbalanced between groups.

Participants will not be aware of the type of hearing aid setting they are experiencing. All participants will experience both .

Participants who have cochlear dead regions, identified by the thresholds equalising noise test will be placed in this group.

Participants who have no cochlear dead regions, identified by the thresholds equalising noise test will be placed in this group.

BKB is a standard sentence list comprised of three or four key words per sentence. The adaptive BKB sentence test requires the noise to be presented at 55 A- weighted decibels (dBA) and the speech level adapts depending on whether the participant repeats the sentence correctly or not. Participants will be given a score in dB which corresponds to the intensity level at which 71% of the sentences are correctly identified. One sentence list will each be administered when the noise is presented at 0°, +90° and -90° and an average overall score in dB will be calculated. The lower the score in dB, the better the participant has performed.

The AB word list aims to assess patient's speech discrimination ability, awarding one point for each phoneme repeated correctly in each word. Words will be presented from a loudspeaker at 65 decibels sound pressure level (dB SPL) in the absence of background noise. All words are monosyllabic and comprised of three phonemes, one mark is awarded for each correct phoneme (0,1,2,3). A single list contains ten words, giving a total of 30 phonemes per list. For each test session, three different word lists will be used and the number of phonemes correctly repeated will be recorded, allowing participant's scores out of 90 to be calculated. The greater the score, the better the performance.

ASSE is a psychoacoustic test that directly assesses participant's ability to discriminate between different phonemes. ASSE discrimination is an oddity test aiming to assess supra threshold performance. At a level of 70dB SPL, following three to eight presentations of a single speech sound, a different speech sound will be played, and participants should respond to the 'odd one out'. If participants respond correctly to two out of three presentations, it can be concluded that the contrast between the background phoneme and the odd speech sound is well discriminated and this will be marked as correct. Seven pairs will be used that include both high and low frequency containing phonemes. Participants will be given a percentage score based on how many of the seven speech pairs they are successfully able to discriminate between. The higher the percentage, the better able the participant is to discriminate.

A validated shortened version of the speech, spatial and qualities of hearing scale (SSQ) questionnaire. The questionnaire will be administered during sessions three and four to compare patient's experiences with NLFC and CP. This will be carried out in an interview style format to ensure participants understand the questions. Participant's scores out of 120 will be calculated and compared in the two conditions. The higher the score, the better the patient finds the hearing aids.

Inclusion Criteria: Aged 45 years and above. Bilateral moderate to severe symmetrical sensorineural hearing loss, defined as PTA thresholds from 0.5- 4kHz falling between 40dB and 85dB. Normal otoscopic examination and normal middle ear function. Currently wears bilateral hearing aids. Good hearing aid user >5 hours a day bilaterally, assessed by data logging. English as a first language, due to the speech testing material Exclusion Criteria: Under 45 years old Asymmetrical, mixed or conductive hearing loss, Recent history (<1 year) of otalgia, otitis externa, otitis media with effusion. Current or recent (<12 months ago) exposure to ototoxic medication e.g. cisplatin (or other platinum-containing chemotherapy drugs) or amikacin (or other aminoglycoside antibiotics), as stable hearing thresholds are required for the duration of the study. Poor hearing aid user with data logging <5 hours a day. English as a foreign language, illiterate or unable to read due to poor eyesight. Diagnosis of conditions such as muscular dystrophies or chronic fatigue syndrome that may cause tiring during testing. Diagnosis of dementia or a neurological condition that would suggest poor cognitive function. Identification of a significant cochlear dead region (< 3kHz) in one ear only.

This is a Masters Dissertation project. All data will be anonymised at the first available opportunity.

Salorio-Corbetto M, Baer T, Moore BCJ. Comparison of Frequency Transposition and Frequency Compression for People With Extensive Dead Regions in the Cochlea. Trends Hear. 2019 Jan-Dec;23:2331216518822206. doi: 10.1177/2331216518822206.

Vinay, Moore BC. Prevalence of dead regions in subjects with sensorineural hearing loss. Ear Hear. 2007 Apr;28(2):231-41. doi: 10.1097/AUD.0b013e31803126e2.

Salorio-Corbetto M, Baer T, Moore BC. Quality ratings of frequency-compressed speech by participants with extensive high-frequency dead regions in the cochlea. Int J Audiol. 2017 Feb;56(2):106-120. doi: 10.1080/14992027.2016.1234071. Epub 2016 Oct 11. Erratum In: Int J Audiol. 2017 Feb;56(2):144.

Salorio-Corbetto M, Baer T, Moore BCJ. Evaluation of a Frequency-Lowering Algorithm for Adults With High-Frequency Hearing Loss. Trends Hear. 2017 Jan-Dec;21:2331216517734455. doi: 10.1177/2331216517734455.

Pepler A, Munro KJ, Lewis K, Kluk K. Prevalence of cochlear dead regions in new referrals and existing adult hearing aid users. Ear Hear. 2014 May-Jun;35(3):e99-e109. doi: 10.1097/AUD.0000000000000011.

Glista D, Scollie S, Bagatto M, Seewald R, Parsa V, Johnson A. Evaluation of nonlinear frequency compression: clinical outcomes. Int J Audiol. 2009;48(9):632-44. doi: 10.1080/14992020902971349.

Ellis RJ, Munro KJ. Benefit from, and acclimatization to, frequency compression hearing aids in experienced adult hearing-aid users. Int J Audiol. 2015 Jan;54(1):37-47. doi: 10.3109/14992027.2014.948217. Epub 2014 Dec 3.