Download Diagnostic Auditory Brainstem Response Training Manual

Diagnostic Auditory
Brainstem Response
Training Manual
Compiled by Renée Janssen
June 2008
-1-
Acknowledgements
The underlying protocols for this ABR training manual has been compiled from
various sources, including the excellent work done by the Ontario Infant Hearing
Program.
Renée Janssen, compiler and editor of this manual, thanks the members of the
BCEHP Diagnostic Protocol Advisory Group for their contributions to this document.
BCEHP Diagnostic Protocol Advisory Group
(Past and Present Members)
Dr. David Stapells
Terence Miranda
Jess Rainey
Dr. Anna Van Maanen
Dr. Ieda Ishida
Karin Rennert, past chair
.
-2-
Table of Contents
SECTION 1: BCEHP PROGRAM INFORMATION
-6-
1.1
Background
-7-
1.2
Program Principles and Goals
-8-
1.3
Program Description
-9-
1.4
Family Centred Care
- 10 -
1.5
Roles and Responsibilities
1.5.1
Regional Coordinator
1.5.2
Audiologists
SECTION 3: PATIENT MANAGEMENT
- 12 - 12 - 12 -
- 13 -
3.1
Suggestions for Working with Patients and Families
- 14 -
3.2
Ensuring adequate infant preparation for ABR
- 16 -
3.3
Instructions to Parents for Appointment Preparation
- 16 -
3.4
Sedation
- 18 -
SECTION 4: AUDITORY BRAINSTEM RESPONSE – PREPARATION
4.1
Definition of Auditory Brainstem Response
- 20 - 21 -
4. 2
Room and Equipment Requirements
4.2.1. Facility Requirements:
4.2.2. Supply Requirements
4.2.3
Equipment Components
4.2.4
Hardware Setup
- 21 - 21 - 23 - 24 - 25 -
4.3
Room and Equipment Preparation
- 26 -
4.4
Infection Control
- 26 -
4.5
Preparing Baby for ABR testing
4.4.1
Infant handling and preparation
4.4.2
Skin preparation
4.4.3
Electrode placement and impedance
4.4.4
Transducer placement
SECTION 5: AUDITORY BRAINSTEM RESPONSE - TESTING
5.1
ABR Protocol and Strategy
5.1.1
General Strategy
5.1.2
Strategy for Stimulus levels
- 27 - 27 - 27 - 28 - 29 -
- 32 - 33 - 33 - 34 -
-3-
5.1.3
5.1.4
5.1.5
5.2
Stimulus Frequency Protocol
BCEHP minimum ABR requirements
Example Test Sequences
Air Conduction
- 34 - 35 - 36 - 38 -
5.3
Bone Conduction
5.3.1
Bone conduction – general information
5.3.2
Bone conduction stimulus artifact
- 38 - 38 - 39 -
5.4
Waveform interpretation
5.4.1
Definition and Identification:
5.4.2
Replication:
5.4.3
Morphology:
5.4.4
Interpretation:
5.4.5
Air conduction – interpretation
5.4.6
Bone conduction – interpretation
- 41 - 41 - 41 - 42 - 43 - 46 - 48 -
5.5
Objective aids to waveform interpretation
5.5.1
Residual Noise (RN):
5.5.2
SNR
5.5.3
Cross correlation (CCR)
- 51 - 51 - 52 - 55 -
5.6
Non-physiologic artifact
- 56 -
5.7
Troubleshooting a noisy EEG
- 59 -
5.8
Logsheet
- 61 -
SECTION 6: TECHNICAL PARAMETERS
- 62 -
6.1
ABR Tone Pip Parameters
- 63 -
6.2
ABR Amplifier Gain Settings
- 64 -
6.3
ABR Artifact Rejection Parameters
- 65 -
SECTION 7: DIAGNOSING AUDITORY NEUROPATHY/DYS-SYNCHRONY
- 66 -
7.1
Background
- 67 -
7.2
Differential diagnosis – techniques
- 68 -
SECTION 8: USING THE INTELLIGENT HEARING SYSTEM (IHS) SMARTEP - 72 8.1
IHS manual - sections
- 73 -
8.2
Using the IHS – detailed instructions
- 73 -
SECTION 9: REPORTING
9.1
ABR waveform display
- 87 88
-4-
9.2
Estimated Hearing Levels (EHL)
88
9.3
Sample reports
91
APPENDIX A - RISK INDICATORS
95
APPENDIX B: BCEHP DIAGNOSTIC CLINICAL PATHWAYS
96
APPENDIX C: BCEHP ABR PROTOCOL FLOWCHART
103
APPENDIX D: ABR RECORDING SHEETS
104
APPENDIX E: ABR PROTOCOLS FOR CHALLENGING CASES
106
-5-
Diagnostic ABR Training Manual
Section 1: BCEHP Program Information
1.1 Background
1.2 Program Principles and Goals
1.3 Program Description
1.4 Family Centred Care
1.5 Roles and Responsibilities
-6-
1.1
Background
The first few years of life are critical for the development of language, speech
and social skills. The importance of acquiring communication skills early in life is
well understood. Research has shown that delay in language development can
have a significant impact on cognitive, emotional and psychosocial development
and that language development is the prime indicator of future academic
success. Children born with permanent hearing loss or who acquire permanent
hearing loss during the early years of life are at risk for delay in language
development. The age at which a hearing loss occurs and the age at which it is
identified are both critical factors in determining the child’s development of both
receptive and expressive language. Generally, the earlier and more severe the
hearing loss, the greater the potential influence on the child. However, even a
minimal or unilateral hearing loss in infants and young children can impact the
development of language and communication skills (American Speech Language
Hearing Association, 2002; Jamieson, 1994).
Studies conducted in the United States have determined the incidence of
permanent hearing loss at birth to be 3 to 6 out of every 1000 births. The range
depends upon whether both unilateral and bilateral losses were included
(American Speech Language Hearing Association, 2002). In addition, there are
risk factors for progressive (delayed onset) hearing loss in young children.
Results of universal newborn hearing screening programs have indicated that
appropriate intervention, including amplification, before six months of age
improves outcomes in speech and language development (American Speech
Language Hearing Association, 2000).
Early attempts to identify hearing loss in infants were limited by the technology
and the dependence upon a required behavioural response. Screening
techniques using sounds coupled with observation of resulting infant motion can
only identify infants with severe to profound hearing loss (American Speech
Language Hearing Association, 2000). Objective techniques such as auditory
brainstem response (ABR) provide additional information, but are timeconsuming and expensive. The infant must reach a developmental age of six
months before techniques such as Visual Reinforcement Audiometry (VRA) can
be administered reliably.
Starting in the 1980’s the Public Health Audiology Clinics in British Columbia used
a High Risk Hearing Registry to screen all births for factors associated with
hearing loss. Children initially identified at risk were then referred for Visual
Reinforcement Audiometry at age six to six and a half months. This achieved an
average age of identification at eighteen months. In addition to the delay in
-7-
identification, risk registries only capture approximately 50 percent of congenital
hearing loss.
The discovery of otoacoustic emissions (OAE) and a growing acceptance of a
need for earlier identification and intervention resulted in the international push
in the 1990’s and start of the twenty first century to develop and implement
Universal Newborn Hearing Screening—Early Intervention (UNHS-EI) programs.
The Vancouver Island Health Authority-South Hearing Program piloted the first
such program in British Columbia in 2001.
1.2
Program Principles and Goals
BCEHP is built as a set of “foundational” principles. These principles guide the
ongoing development and review of the program and how providers work with
families.
The principles are:
• Accountability and Sustainability
• Acceptable and Timely
• Accessible and Equitable
• Contribute and Develop BC and Canadian Research
• Continuum of Integrated and Coordinated Service
• Consultation
• Cultural and Geographical Diversity
• Evidence-Based, consistent standards across the Province
• Family and Child-Centred Care
Goal of the Diagnostic ABR Program: to test all babies who have “referred”
on a second stage screen.
Purpose of ABR: to confirm hearing status (if hearing loss, type, degree and
configuration), to identify infants who have target disorder, and to provide
diagnostic ABR results useful for decisions regarding intervention and
habilitation.
Target Disorder: Permanent unilateral or bilateral hearing loss (conductive or
sensorineural) over 25 dBnHL between 500 and 4000 Hz.
Targets: 100 percent of babies who “referred” on second stage screen by 3
months of age.
-8-
The goals of the BC Early Hearing Program are:
•
•
•
A child with hearing loss will have age-appropriate language skills by school
entry.
The child and his/her family will have meaningful and accessible
communication.
Parents will have knowledge, skills and confidence to parent/advocate for
their child with hearing loss.
The goal of the diagnostic ABR component of the BCEHP is to provide an earand frequency-specific estimation of behavioural thresholds for infants with
hearing loss (i.e. type, degree and configuration of hearing loss), in a timely and
family sensitive manner.
The program is provided to families based on fully informed parent/guardian
choice and consent, and will comply with confidentiality requirements.
The program was developed using the principles of evidence-based practice.
The program is monitored and evaluated on an ongoing basis. The quality of the
program is to be continuously improved, based on evaluation and new evidence.
1.3
Program Description
Each year in BC, about 120 babies are born with a congenital (present at birth)
hearing loss and about 5-6 percent of babies are born with risk factors for late
onset hearing loss. The average age of identification for children with hearing
loss in BC is estimated to be age 2-2 ½ years. The BCEHP will facilitate early
screening (by one month of age), early confirmation of hearing status (by age 3
months) and early intervention for those babies with confirmed hearing loss (by
age 6 months), leading to improved outcomes for families and children.
BCEHP is a continuum of integrated services and support including:
• Public Education
• Hearing Screening – both at birth and surveillance for later onset hearing loss
• Diagnostic Assessment – Audiological and medical
• Hearing Equipment Fitting and ongoing Care – both hearing aids and FM
Equipment
• Communication and Family Supports
• Evaluation and Quality Improvement
The BCEHP screening was implemented in stages across the province, starting
with screening infants in the NICU in June-Dec 2006. The remaining screening
will be phased in beginning in 2007, with full roll out by 2009.
-9-
1.4
Family Centred Care
What is family centred care?
•
Family Centred Care builds parent/professional partnerships and promotes
parents as the decision makers for their infants. Family choice and decision
making occurs at all levels.
•
Family Centred Care recognizes the strengths and expertise that all parties
bring to the relationship but it also recognizes that it is the parents who will
be the constant in the child’s life—not the service providers.
•
Families are treated with dignity and respect. Service providers are sensitive
to diversity. The expertise, preferences, language, and culture of the family
are valued.
•
Service Providers listen to and honour the family’s knowledge, perspective
and choices; they avoid making judgments about the family’s decisions.
•
Complete and unbiased information is shared with families in ways that are
useful. Families receive timely and accurate information in order to
understand and effectively participate in the care and decision making
regarding their infant. Service Providers avoid the use of any jargon that they
don’t first explain.
•
Families have opportunities to discuss their options and understand
implications of all decisions.
Family Centred Care results in practices in which the pivotal role of the family is
recognized and respected. Families are supported in their caregiving roles by building
on their unique strengths as individuals and families. Opportunities are created for
families to make informed choices for their children and more importantly, these choices
are respected.
Sheldon &Stepanek 1994
- 10 -
Family-Centred Care Checklist
Here is a checklist for you to use to evaluate your own work. The questions
below can be used to help you consider what changes you might want to make
in your interactions with families.
‰ Do I communicate, to parents that they are “in charge” of what will happen
with their infant, and I will respect their choices?
‰ Do I communicate to the parents that I value their infant?
‰ Do I tell them what they should do or instead - do my actions show that I
recognize their important role as the decision maker and I provide them with
information including the benefits and limitations of the different choices they
might make?
‰ When parents aren’t doing what I “want them to do”, do I view them as
uncooperative?
‰ Do I allow enough time to actively listen to family’s concerns, questions, and
priorities?
‰ Do I provide information without using any “jargon” that I do not at first
explain?
‰ Do I communicate to the parents that I care about their well being, too?
(Paying attention to how the parents are receiving the information, ensuring
their questions and anxieties about their child or the testing are addressed)
‰ Do I consider the family’s unique culture, language preferences, and values?
‰ Do I assist the family in obtaining other supports that might help them make
decisions?
- 11 -
1.5
Roles and Responsibilities
Successful diagnostic ABR requires a team approach. The team consists of
family members, audiologists, and Regional Coordinators, audiometric
technicians, and public health nurses. The general roles are outlined below.
Individual screening sites and Health Authorities may adjust these slightly to fit
local needs. The Regional Coordinator is responsible to ensure all roles are
fulfilled and personnel are clear on their responsibilities.
1.5.1 Regional Coordinator
The Regional Coordinator is responsible for:
• Local Program development, management and evaluation
• Ongoing support for Screening Personnel
• Ensuring audiological follow-up of all referred infants
1.5.2 Audiologists
The audiologists are responsible for:
• Identifying infants who have the target disorder
• Communicating results to families and other professionals involved with the
child
• Collecting data, recording on the BCEHP ABR Recording Sheet and entering
ABR results into data base
• Monitoring, tracking and reporting equipment and supply inventory and
maintenance
- 12 -
Diagnostic ABR Training Manual
Section 2: Patient Management
2.1 Suggestions for Working with Patients and Families
2.2 Ensuring adequate preparation of infant for ABR
2.3 Instructions to Parents for Appointment Preparation
2.4 Sedation
- 13 -
2.1
Suggestions for Working with Patients and Families
Primary factors are sleep promotion and the best use of clinical time to obtain
the priority data to inform the next steps of care for this infant. It is important to
instruct the family to have the infant arrive tired (not sleeping) and hungry, so
that once the infant is fed, they are more likely to be in a deep sleep. The goal
is to conduct as much testing as quickly as possible.
In general it is recommended that testing be completed by one audiologist.
Circumstances where additional staffing would be needed include when the
audiologist is less experienced, or when assistance in placement of transducer
(i.e. hold a bone conductor) is required. Where feasible, it is recommended that
the tester and instrumentation be inside the sound booth, together with the
infant, who may be in a bassinet, crib or pram or in the mother’s arms in a
reclining chair. Attendance of family members/caregivers during ABR testing is
determined by the family requirements, testing requirements and the goal for
obtaining maximum results in minimal clinic time. The presence of a family
member or third party may be desirable for reasons unrelated to test quality,
such as to secure compliance or to manage perceived medico-legal risk.
However, family members differ widely in their knowledge and skills related to
infants' sleeping habits. They may distract or excite the infant, may use
inappropriate strategies to promote sleep, or may otherwise compromise efficient
and accurate testing. However, allowing parent(s) into the test room, at least
initially, can alleviate their anxiety. For ABRs conducted without sedation
(typically, under 6 months of age in BC), babies often will not sleep soundly for
the duration of the test, especially if they are over 3 months of age. In these
cases, it is important to have the caregiver present to aid in getting the child
back to sleep, as the caregiver is familiar to the child, and will often know the
best techniques for sleep induction for their child.
When the baby arrives:
1. Quick introduction and explanation of ABR (see below for suggestions). No
history at this time to ensure infant does not fall asleep prior to testing.
2. Make sleep arrangements at this time (find out the place/position the baby
will sleep the best).
3. Start prepping forehead and mastoids as soon as mother and infant are
comfortable.
4. Once electrodes are in place and impedance is acceptable, detach the
electrodes from the box and place the electrodes somewhere they will be
accessible to you later.
5. If possible, put inserts in at this point. In some cases, these can be put in
without being attached to the EAR box (mind the tubes).
- 14 -
6. Have caregiver try to get the baby to sleep by feeding or comforting. Make
the booth dark. If the baby is not yet asleep, it may be appropriate to leave
at this point; ensure to check sleep status frequently.
7. Have caregiver move baby into the best sleep position once they are asleep
(i.e., to bed, stroller or in arms).
8. When you re-enter the booth, plug the electrodes in box, check impedance
and status of inserts while disturbing the baby as little as possible.
Miscellaneous:
1. With several people in the booth, heat can be an issue. Make sure that
everyone has any heavy clothes (sweaters, jackets etc.) removed prior to
testing.
2. If caregiver is holding the infant, try to make them as comfortable as possible
(provide pillows), as the more they move around the more likely the infant
might wake up.
3. Have a puppet or toy in the room to distract for OAEs or tympanograms
(assuming that the ABR is complete and the baby is potentially awake for
these procedures).
4. Have a basket, cupboard or area where supplies are sorted and easy to get
to. Having supplies organized and accessible will be helpful in a dark booth.
5. Having counselling tools in the booth can also be helpful. If the baby needs
to be fed after waking up and the booth recliner is the best place to do that,
you can counsel while the caregiver is feeding the baby.
6. Have earmold impression material close by.
Explanation of ABR and why deep sleep is needed:
“ABR works by recording electrical energy from your baby’s hearing nerve in
response to sound. Muscles also generate electrical energy, which is much larger
than the hearing nerve activity. This is why the baby has to be asleep, so that
the muscles are quiet and we can see the responses from the hearing nerve.”
Explanation of test length:
“If the hearing is normal the test will take less time than if it is not. With normal
hearing I test at a low intensity at several pitches and stop if I see a response. A
response at each pitch tells me the hearing is within normal range. If there is a
hearing loss, I have to do much more testing to determine where the hearing
loss is in the ear as well as the lowest intensity where I will still see a response.”
Explanation of skin preparation (often the baby will cry; it helps to warn
parents):
- 15 -
“We need a good connection between our sensors (electrodes) and your baby’s
skin to pick up the responses. The skin preparation doesn’t hurt your baby, but
if your baby is tired and hungry, it might make your baby cry” (using the scrub
tape or gel on the caregiver’s hand can be useful to demonstrate the scrubbing
feeling).
Explanation of what the audiologist will be doing during the test (this will allow
you to explain test results only when the testing is over instead of providing
commentary as the testing progresses):
“Because we don’t know how long your baby will sleep, I’m going to be focusing
on the testing, and making sure we get as many results as possible. For this
reason, I won’t be interacting with you, or telling you about the results until the
test is over. I will share all my results with you at the end of the testing.”
2.2
Ensuring adequate infant preparation for ABR
For assessments in natural sleep, every reasonable effort will be made to ensure
that the infant arrives for testing in an appropriate state. It is recommended that
the infant be tired (but not overtired) and hungry on arrival. From a risk
management standpoint, families who drive to assessments will be STRONGLY
encouraged to be accompanied by a third party who will manage the infant. The
potential futility of attempting assessment in an inappropriately prepared infant
will be stressed.
The infant’s behavioural state upon arrival for assessment is important for
successful testing in natural sleep. The family should be made fully aware of the
importance of appropriate preparation for testing, and should be given detailed
instructions on what and what not to do. Written instructions and telephone
confirmation are recommended. The importance of preparation increases with
age, due to a decrease in the amount of the infant’s daytime sleep. Wherever
possible, the infant should arrive at the test tired and hungry. It is normally
appropriate to deny sleep and food for at least an hour before testing, where not
medically contraindicated. If the child is being brought to the test by car, it is
important that every reasonable effort be made (consistent with safety) to keep
the child awake on the journey. Because of the soporific effect of car journeys on
infants, another person in addition to the driver is usually necessary.
2.3
Instructions to Parents for Appointment Preparation
Written Information:
- 16 -
This is sent to parents with their appointment letter.
Dear Parents,
Your child has been scheduled for a diagnostic Auditory Brainstem Response
(ABR) assessment.
Diagnostic ABR assessments are done while your baby sleeps. We place four
recording sensors on your baby’s head and play soft sounds through earphones
that fit in the ears. The ABR is painless and will not harm your baby in any way.
We schedule 3 hours for the ABR appointment, although the testing time is
usually not that long. We allow extra time as it can take a while for your baby to
fall asleep.
Your baby must be asleep for testing. Please do what you can to bring your baby
to the appointment awake but ready to sleep once the testing has started. We
realize this is not easy, but please do your best. Here are some suggestions to
help your baby sleep during testing:
How parents can prepare their baby for diagnostic testing
• Keep your baby awake for at least an hour before the test.
• If driving to the appointment, have an extra adult in the car to play with your
baby and keep the baby awake.
• Bring any special blankets or items that may help your baby to sleep.
• Please try to delay feeding your baby for about one and a half hours before
the appointment. It often works well if parents feed their baby as testing
begins, helping the baby to fall asleep.
You will be in a comfortable chair with your baby for the duration of the test.
There are no supervised play areas in the hearing clinic. If you must bring other
children with you, please bring another adult to look after them.
In most cases, the results of the test will be explained to you immediately after
the test.
If you have any questions, please call us at 604-123-4567. Please phone us
immediately if you are unable to keep this appointment.
Verbal Information:
This information is given to parents over the phone by the BCCH Audiology clerk.
•
"This is a sleeping hearing test so we need your baby to sleep for it. We need
your baby to arrive here awake, yet tired and ready to go to sleep".
- 17 -
•
•
•
•
2.4
“Hold off on feeding your child until you arrive and wake them up a bit earlier
that day, or no nap in the morning if appt. is in the afternoon”.
"You know your baby best so whatever you can do to get baby to arrive here
tired, but awake, and ready to go to sleep."
Some parents mention that their baby will fall asleep in the car on the way in,
especially if they come from a long way away. I suggest to them that
someone else come to help keep baby awake, but this isn't always a
possibility ... so I just tell them to do their best.
I suggest that they book the time when the baby is more likely to nap.
Sedation
Assessments under sedation will comply with generally accepted standards of
care and all local risk management protocols. The BCEHP strongly recommends
written, informed consent, medical referral and specification of sedative and
dosage, administration by medical/nursing staff, appropriate supervision of the
child post-medication, and adequate access to emergency services.
Extensive experience from Ontario and from British Columbia indicates that
under about 5 months of age, tone pip ABR testing can almost always be done
satisfactorily with the baby in natural sleep. For example, in over 20 years of
experience with tone pip ABR in infants at Toronto’s Mount Sinai Hospital, with
over 10,000 infant diagnostic assessments, the rate of sedation requirement is
less than 2%. Appropriate training, test protocols and infant management
methods are necessary and sufficient. Appropriate and effective instruction to
families about pre-test preparation is crucial. Family members routinely
underestimate their babies’ inclinations to sleep and adopt inappropriate
strategies if involved in the test. Routinely resorting to sedation or general
anaesthesia in infants under six months’ corrected age is not recommended, and
is largely unnecessary given adequate skills at sleep induction.
Testing under sedation may be necessary in infants for whom acceptable
behaviour and EEG conditions cannot otherwise be obtained. Usually, at least
one attempt to test in natural sleep would have failed before resorting to
sedation. It is reasonable to consider fairly routine use of sedation in children
older than 5-6 months or for children who have to travel long distances for
assessment, because it is especially important to have a reasonable assurance of
success.
The audiologist determines that sedation is indicated on audiometric grounds;
whereas, the family, in consultation with the audiologist and appropriate
physicians, determines whether sedation will actually occur. The infant’s
paediatrician or family physician would normally be involved, as he or she may
- 18 -
have unique knowledge of contraindications or risk indicators in the client’s
history. Where specific centres have established high-quality protocols, they
should take precedence. Documented, informed consent would normally be
required.
If sedation is indicated, and the parent consents, a physician should prescribe
the sedative agent (usually oral chloral hydrate). Appropriate risk management
procedures to guard against rare, adverse events such as respiratory depression
should be in place. While there is wide variation in practices for sedation, the
BCEHP strongly recommends a conservative standard of care. Testing under
sedation should normally be done under medical order and, preferably, with
medical or nursing supervision of the infant from the time of administration
through to the end of the indicated recovery period. Immediate access to
respiratory support and emergency services is appropriate, but local safety
protocols are the determining factor of what is required in a given test setting.
In a few infants, especially those with neurological and/or behavioural disorders,
the response to sedation may be paradoxical activation. This has been addressed
in various ways: by increasing the dosage of the sedative, by using alternative
medications, or by resorting to light, general anaesthesia. The indications for
these procedures are a matter for local risk management protocols and
standards of care.
While testing under sedation is generally easier than under natural sleep, not all
BCEHP audiologists will have access to the required medical coverage. The
decision whether or not to accommodate testing under sedation rests with the
individual audiologist. Where necessary, cross-referral to another BCEHP
audiologist who has a sedation practice may occur. The BCEHP has facilitated the
best possible access to testing under sedation, across the province, within
resource constraints.
- 19 -
Diagnostic ABR Training Manual
Section 3: Auditory Brainstem Response – Preparation
3.1 Auditory Brainstem Response Definition
3.2 Room and Equipment Requirements
3.3 Room and Equipment preparation
3.4 Infection Control
3.5 Preparation of Baby for testing
- 20 -
3.1
Definition of Auditory Brainstem Response
By the time “sound” leaves the cochlea or inner ear, it has become a series of
neuroelectric events traveling in sequence to the acoustic nerve, the brainstem,
and finally to the cortical areas of the brain. These electrical responses have
been referred to by several names but they are now commonly known as
auditory evoked potentials. One type of auditory evoked potential is the
Auditory Brainstem Response (ABR), which occurs within approximately 10-20
ms of the onset of the stimulus (depending on frequency and intensity of the
stimulus). Voltage (potentials) can be measured at the skin with surface
electrodes; the electrode montage consists of a minimum of three electrodes
(but more typically four electrodes). As the amplitude of the ABR is quite small
compared to the “noise” component of the EEG, the signal to noise ratio is
enhanced by averaging. The amplitude of the ABR is also quite small compared
to voltages generated by myogenic activity; therefore, children must be tested
when sleeping. The ABR consists of a series of positivities (at the vertex) that are
named by their relative order (waves I through VI). ABR is typically elicited by
click or brief tone stimuli. For the purposes of ear- and frequency-specific
threshold determination, brief tone-bursts are the stimuli of choice.
Wave Number Hypothesized generator Site
I VIIIth cranial nerve
II VIIIth cranial nerve
III Pons
IV Pons
V Upper Pons and lower Midbrain (inferior colliculus)
Since wave V has the largest amplitude as well as being the most reliable of the
responses, its presence is considered to be the most significant. At or near
threshold, it is the waveform most likely to be reliably present.
3. 2 Room and Equipment Requirements
3.2.1. Facility Requirements:
In many test situations, it is feasible for a single audiologist with appropriate
training to conduct tone pip ABR testing. Where feasible, it is recommended
that the tester and instrumentation be inside the soundroom, together with the
infant, who may be in a bassinet, crib or pram. This may be reassuring to the
parents, who may be reluctant to leave the child alone in what they may see as
- 21 -
an intimidating test environment. It also facilitates the option of single-handed,
unassisted testing.
The current BCEHP diagnostic ABR/OAE equipment (IHS Smart-EP) is laptopbased and noise levels from that unit are not a significant concern. Printing of
records (laser or inkjet) should be done off-line, so it is not absolutely necessary
to power up the printer during testing. Alternatively, the printer may be located
outside the soundroom, given adequate cable routing though the trap or
connector panels.
In special circumstances, exceptions may be made, subject to BCEHP review and
acceptance of the proposed environment. In general, Assessments in any test
environment other than an audiometric soundroom will not qualify for BCEHP
funding unless the environment was approved by BCEHP management.
When not possible to test in the soundbooth, threshold ABR should be done in a
quiet room with measured sound levels (octave band measurements) not
exceeding: (NB: exact levels may be updated pending further information.)
o
o
o
o
•
•
•
•
•
22
30
35
43
dB
dB
dB
dB
SPL
SPL
SPL
SPL
500 Hz
1000 Hz
2000 Hz
4000 Hz
Electrical isolation is required – i.e., more than 25 ft away from elevator
shaft; more than 25 ft away from XRay equipment & power doors.
Other sources of electrical interference may exist that cannot be predicted.
Facility management & clinicians must be aware of possible sources of
electrical interference and be prepared to relocate the ABR room if other
measures to isolate are not successful.
Dedicated circuit is not a requirement.
Access to a sink within close proximity to testing room.
Furnishings:
•
•
•
•
•
Upholstered recliner for parent to hold sleeping infant for one to two hours
Rubberized pillow with disposable case for comfort while holding child
Alternate arrangement for baby to be sleeping (e.g. crib)
ABR equipment on a moveable cart/computer desk
Comfortable chair with wheels for examining audiologist
Other Equipment Required:
•
Diagnostic otoacoustic emissions equipment (within same room).
- 22 -
•
•
Diagnostic immittance equipment with high frequency tympanometry and
acoustic reflexes in close proximity.
Soundsuite for VRA and BOA assessment as per typical set up, i.e., a
minimum of two reinforcers, two soundfield speakers, insert phones, and
bone conductor, diagnostic audiometer with capability of testing pure tones
250 Hz through 8000 Hz and testing of speech (testing to profound levels,
meeting ANSI calibration specifications, with microphone and CD input).
3.2.2. Supply Requirements
Skin Preparation:
Suggested products:
1. 3M Red Dot Trace Prep 2236 (available from 3M Canada, London Ontario)
2. NuPrep ECG & EEG Abrasive Skin Prepping Gel (available from D.O. Weaver &
Co., 565-B Nucla Way, Aurora CO, 80011 – http://www.doweaver.com )
3. Cotton tipped applicators (15mm) (available through Stevens Company
Limited at 8188 Swenson WayDelta, B.C. V4G 1J6 604-634-3088 http://www.stevens.ca )
Electrodes and attachment:
1. Ten20 conductive EEG paste (4 or 8 oz) (available from D.O. Weaver & Co.,
565-B Nucla Way, Aurora CO, 80011 – www.doweaver.com )
2. Disposable electrodes
a. Ambu Neurology
b. Kendall Puppydog 1011pts Small (Neonatal Radiolucent Monitoring
Electrodes) (available through Ludlow 909-605-6572)
c. Viasys disposable 4-disk electrodes with 2.0m leads (1-800-356-0007 in
U.S.)
3. Non-disposable electrodes
a. Silver EEG Surface Electrodes – 6mm cup, 1M cable (or shorter)
4. 3M Micropore surgical tape
IHS backup supplies:
1. 2 channel fibre optic cable (ref #132200)
2. “Y” adapter (ref #013901)
3. ER-10D OAE Probe tip replacement
Other extra supplies:
1. EAR 3A insert foam tips (small)
- 23 -
2. EAR 3A insert tubes and tube nipples
3. AA batteries (for use with battery operated opti-amp)
4. Headband for bone-conduction testing (available from Design Veronique
“Universal Facial Band- 2" Width” from http://designveronique.com/cgibin/ic/dv2/210-2.html)
Recommended miscellaneous supplies:
1.
2.
3.
4.
5.
6.
7.
Flashlight
Ziplocks/Tupperware for organization of supplies
Scissors to cut foam tips
Hemostats, with teeth taped if metal, or green plastic clamp scissors
Counselling tools such as a diagram of the ear, speech audiogram, pamphlets
Linens (towels, blankets)
Adhesive remover (avoid use of adhesive removers containing alcohol; also,
there have been reports of some babies exhibiting allergic reaction to the
chemicals in these products)
8. Alcohol swabs (for disinfecting equipment – not recommended for use on
baby’s skin)
9. Antiseptic instant hand sanitizer
3.2.3 Equipment Components
o laptop
o power transformer for routing of power to IHS hardware and laptop
o IHS hardware (image below) connected to computer by USB cable
o opti-amp transmitter box (image below) connected to IHS hardware by
infrared cables
- 24 -
o
o
o
o
o
insert headphones, connected to IHS hardware
supra-aural headphones, connected to IHS hardware (when needed)
bone conduction oscillator, connected to IHS hardware
OAE probe, connected to IHS hardware
printer connected to laptop
3.2.4 Hardware Setup
Calibration
Instrumentation will be calibrated and maintained according to BCEHP
specification. The factory calibrations supplied with the BCEHP ABR
instrumentation are not acceptable for BCEHP Assessments and will be modified
according to BCEHP specifications.
Many factors affect the proper calibration of stimuli for tone pip ABR testing in
infants. Important variables include subject age, stimulus route and transducer,
frequency, envelope and repetition rate, the amount of averaging, EEG criteria
for residual averaged noise, and definition of threshold. The amount of highquality published data is limited but sufficient to establish normative calibrations.
Provided that the key features of BCEHP test protocols match those of the
protocols used in deriving the published data, the use of published, large-sample
norms is generally superior to the development of local, small-sample norms.
The BCEHP is responsible for the determination and dissemination of reference
setting for the calibration files that govern stimulus levels for ABR testing.
Normally, ABR calibration files or lists of specific values will be distributed by
email or CD. The calibration data will be updated from time to time, as further
information becomes available from published research or BCEHP clinical
practice.
The main reason for routine, full acoustical calibration is mechanical change in
the stimulus transducer or defects in leads and connections. Annual acoustical
calibration is a formal BCEHP requirement. Calibration services will be arranged
by the individual Assessment centres, in conjunction with the BCEHP. Because
transducer malfunction can occur at any time, additional, frequent listening
checks are required. The individual BCEHP audiologist is responsible for routine
listening checks.
- 25 -
3.3
Room and Equipment Preparation
Routine daily checks
Visually check that the earphone and sensor cables are properly connected and
in good condition. Check that the cable connectors are in good condition and
that the metal connecting pins are not bent, broken or damaged.
Listening checks should be performed routinely to ensure that your equipment is
functioning properly.
Daily/Weekly Listening Check:
a. Increase the intensity to 50 dB
b. Change Mode: Both
c. Listen through the insert phones
3.4
Infection Control
Infection control is an important part of the BCEHP. It is essential that
audiologists wash their hands between each baby and after handling the cart
and supplies. This is to protect from baby-to-baby infection, baby-to-equipment
infection, and equipment-to-baby infection, as well as to protect the audiologist
from infection. The following techniques should prove helpful in providing a
healthy environment for babies and clinic personnel. Consult with your facility’s
Risk Manager (Infection Control person) to obtain specific procedures required
for your hospital/clinic.
To be testing babies, the audiologist must be:
•
•
•
Free of transmissible infectious disease, for example, conjunctivitis, “active”
cold/cough, and dermatitis.
Any Herpes lesions (cold sores) must be dry and scabbed and covered
whenever possible.
Vaccinated against rubella if the staff member is susceptible.
Hand Washing
Proper Hand washing techniques:
• Remove all rings, bracelets, etc. and put in a safe place.
• Adjust the temperature of the water.
• Apply soap from dispenser.
- 26 -
•
•
•
•
•
•
•
Rub hands together for 15 seconds (time it takes for you to sing “Happy
Birthday”).
Scrub between fingers, knuckles, backs of hands and fingernails.
Rinse hands under running water.
Wring hands of excess water.
Use paper towel to thoroughly dry hands.
Turn off water with paper towel.
Discard paper towel into open, plastic lined receptacle or one with a footoperated cover.
Clean equipment and work surfaces
Establish a daily cleaning routine including anything that will come in direct
contact with the baby, or the assessment materials, such as cart or tables.
•
•
•
3.5
Use cleaning materials recommended by your facility
Discard individual baby disposable supplies in an approved receptacle after
the assessment of each baby.
Wipe cables and electrode leads with alcohol or disinfectant wipes between
babies.
Preparing Baby for ABR testing
3.4.1 Infant handling and preparation
1. Follow BCEHP protocol re: infection control procedures (disposables)
2. A sleeping, comfortable baby is optimal for testing
3. A baby that is too alert or uncomfortable will create high test artifact and
compromise the result or prolong test time
3.4.2
Skin preparation
The skin at the electrode sites must be prepared to ensure that the impedance is
low enough for a good recording (see 4.4.3).
1. Scrub skin with abrasive tape or gel at the place of electrode placement
2. Use of alcohol wipes is not recommended, as it dries out the skin and can
cause irritation.
- 27 -
3.4.3 Electrode placement and impedance
Placement:
1. Use four electrodes (two on the forehead and two for the mastoids).
2. Place the non-inverting electrode on the high forehead as close as possible to
the hairline, in the midline. The electrode lead should be run away from the
face.
3. Place an inverting electrode on each mastoid area (as low as possible on the
mastoid bone to avoid the post-auricular muscle and to be farther away from
the bone-conduction transducer). The electrode lead should be pointing
down.
4. The four electrodes properly placed provide two differential recording
channels: forehead to left mastoid and forehead to right mastoid.
5. Place the common electrode on the forehead, ≥ 3 cm over from the
noninverting electrode (centre to centre). The electrode lead should be run
away from the face.
6. Keep the electrode wires close together and if possible braid to decrease 60Hz artifact. Electrodes should be led away from where the transducers (air or
bone, but especially bone) are to be placed.
Impedances:
1. Impedances should be less than 3 kOhms (if possible). The impedance does
not affect the ABR itself, but the larger the impedance, the larger the amount
of pickup of external electromagnetic interference and of artifact from
movement of the electrode leads.
2. More important than the overall impedance is the symmetry of the two
electrodes that form each differential pair. These should be as similar as is
possible with reasonable effort (not more than 1 kOhm is a desirable target).
3. Given reasonable efforts to achieve satisfactorily low and symmetrical
impedances, testing may proceed despite less than ideal conditions. If this is
- 28 -
the case, document the impedance values and be alert to the possible need
for larger averages and more frequent replication of records.
3.4.4
Transducer placement
Cursory otoscopy will be conducted at the start of any BCEHP ABR assessment.
Its main purpose is to detect foreign bodies, canal occlusion and any physical
condition of the ear that indicates referral to a physician under standard red
flags.
Detailed otoscopy and TM visualization can be difficult in the young infant and
are, therefore, the domain of the experienced physician, but it is recommended
that the audiologist conduct at least a cursory otoscopic examination at the
outset of the assessment, primarily for the visualization of any significant
debris/cerumen in the ear canal.
The ear canals of young infants frequently contain varying amounts of debris
and/or cerumen. Hearing testing remains viable unless the canal is completely
occluded acoustically, but total acoustical occlusion is difficult to determine
visually. If the canal appears totally occluded, which is infrequent, or if there is a
foreign body or evidence of acute infection, then referral for management by an
experienced physician is mandatory.
In the absence of a red flag condition, the decision to undertake testing with
insert phones, when there is partial occlusion by debris or cerumen, is at the
discretion of the audiologist. If the results of such testing are not normal,
removal and replacement of the ear tip often gives improved results and may
remove significant debris or cerumen. Supra-aural earphones are an option, with
the caveats noted below. Bone-conduction testing is an option, but a return visit
for AC testing would be required after ear cleaning in both cases.
Earphones:
1. Use foam insert ear tips. If not small enough, cut down to size for a young
infant. These are preferred over impedance probe tips (and required for any
later amplification fitting procedures). Use of impedance probe-tip adaptors
often does not allow for both insert phones to be inserted in both ears, and
are very prone to falling out. Thus, under typical situations, impedance
probe-tip adaptors should not be used.
2. Insert foam tips into both ear canals prior to beginning the test. This allows
efficient switching between ears.
3. Supra-aural earphones (TDH/MX41) may be used for special circumstances
(e.g., atresia), or when infants do not tolerate insert earphones.
- 29 -
Insert phones have the following advantages over supra-aural earphones:
o
o
o
o
o
o
reduced stimulus artifact
decreased background noise
less acoustic cross-over
decreased likelihood of collapsed canals
increased comfort
easier for audiologist to perform test, as supras require more skill and
attention to maintain proper placement
Bone conductor:
1. Place the transducer supero-posterior to the pinna of the individual test ear
(on the temporal bone or very high mastoid).
2. Accurate BC ABR tests require proper placement and stable retention of the
transducer, with adequate contact force. To achieve proper force and stability
of the bone oscillator, the BCEHP audiologist may either use a band of elastic
fabric with Velcro attachments, or, with training, use hand-holding.
3. Velcro bands are simple to construct. The width of material should be
sufficient to envelop the transducer and hold it securely in place. A
disadvantage of the Velcro band is that placing it may waken the infant.
4. Alternatively, the bone-conduction transducer may be hand-held firmly in
place by an individual specifically trained in this procedure. Recent research
at UBC (Small, Hatton & Stapells, Ear & Hearing, 2007) as well as clinical
experience has demonstrated that handholding, under controlled conditions,
allows quick and effective screening of BC-ABR at minimum test levels.
Moreover, experience has shown that this method is less likely to awaken the
infant compared to the elastic strap. Provided they are seated comfortably
next to the infant being tested, the BCEHP audiologist performing the
assessment can often hand-hold the transducer while testing. However, when
BC-ABR results are complicated and/or the setup does not allow this, either
another individual trained in hand-holding or the use of the elastic headband
will be required. The transducer should not be hand-held by an individual not
trained in hand-holding the transducer, such as the infant’s parent.
5. The metal bone-conductor band used in behavioural testing should not be
used for BC-ABR testing as it is uncomfortable, easily slips off during testing,
and does not provide sufficient or calibrated application force for young
infants.
6. Insert earphones need not be removed for BC-ABR testing. The occlusion
effect is not present at 2000 Hz in adults; more importantly, recent research
at UBC (Small, Hatton & Stapells, Ear & Hearing, 2007) indicates young
infants do not show an occlusion effect at 500, 1000, 2000 or 4000 Hz.
- 30 -
7. In contrast to adult testing, in the infant, intracranial transmission losses are
sufficiently large that each ear MUST be tested individually; that is, it cannot
be assumed that a given mastoid placement stimulates both cochleae equally.
- 31 -
Diagnostic ABR Training Manual
Section 4: Auditory Brainstem Response - Testing
4.1 ABR Protocol and Strategy
4.2 Air conduction
4.3 Bone conduction
4.4 Waveform interpretation
4.5 Objective aids to waveform interpretation
4.6 Non-physiologic artifact
4.7 Troubleshooting a noisy EEG
4.8 Logsheet
- 32 -
4.1
ABR Protocol and Strategy
4.1.1 General Strategy
A general theme underlying the clinical strategies identified here is to constantly
review the specific clinical information that is most important at any point
throughout the course of a clinical assessment, and to implement the precise
procedural step that will yield that information in a valid, accurate and efficient
manner. This principle applies to the strategic selection and sequencing of
stimulus frequencies and routes, as well as to the detailed tactics of level
selection within individual frequencies and routes of stimulation.
In general, testing for the most infants seen for BCEHP ABR Assessment aims
to answer the following three questions, in order of priority:
1. Is an ear’s AC threshold normal or elevated? Is the other ear’s AC
threshold normal or elevated?
2. If elevated, is the elevation conductive in nature or is there a
sensorineural component?
3. If elevated, what are the specific thresholds (AC and/or BC)?
The first question is answered by testing each ear at the minimum level required
for normal by BCEHP. It would not normally involve a threshold search. If the
baby wakes up at the end of this, the BCEHP clinician is still able to state
whether one or both ears’ thresholds are normal/elevated.
The second question is answered by BC testing the ear(s) with AC elevation(s) at
the minimum BC level. If the infant wakes up at the end of this stage, the BCEHP
clinician is able to state whether the elevation in AC threshold is conductive in
nature or has a sensorineural, and thus PCHI, component. As the majority of
BCEHP infants with elevated AC thresholds will turn out to have conductive
losses, this procedure will most often quickly identify an infant’s elevation as
conductive in nature, providing important information for subsequent
management and for the parents.
The third question is answered by detailed determination of AC (and BC)
thresholds. AC thresholds for each required frequency are required for
subsequent interventions, including amplification (when chosen by the family)
when sensorineural hearing loss is present.
- 33 -
In general, greatest priority is given to 2000 Hz, and results for this frequency
are normally obtained first. 500 Hz is usually next in priority, with 4000 Hz and, if
required, 1000 Hz following. Prior information, history (e.g., ototoxic
medications) and actual results obtained during the Assessment may alter the
relative priority of frequencies, but the above sequence should be appropriate for
most infants requiring BCEHP ABR Assessment.
4.1.2 Strategy for Stimulus levels
The default strategy for threshold bracketing includes starting at the BCEHP
minimum required level, followed by ascent in steps of at least 20 -30 dB and
descent in 10-dB steps. This procedure is efficient, given that many initial
assessments will reveal normal hearing. Ascent by 10 dB should be avoided
except in the single situation of a questionable positive (replicated) response at a
given frequency. The BCEHP protocol does not involved routine use of an
intensity-latency input-output function approach to threshold estimation, whether
replicated or unreplicated.
The most common cause of inefficiency is to use a step size that is too small, or
to fail to make use of prior information in setting levels that will be close to
threshold. The optimal result for any particular frequency and route of
stimulation is to bracket threshold with only two levels that are no more than 10
dB apart, in which case only two pairs of averages are needed to define the
threshold. The more intensity levels required, the less efficient the strategy. An
approach for selecting intensity levels that is generally efficient for the early
stages of threshold estimation in the absence of prior knowledge is to
approximately bisect the current range of intensities in which the threshold is
believed to lie. For example, if there is clearly no response at 30 dBnHL, then an
efficient next level is 60 dBnHL. This approximately bisects the current range of
uncertainty about the location of the true threshold (30 – 90+ dBnHL), and is
unlikely to disturb the infant.
However, for thresholds that are likely to be greater than 70 dBnHL, a final
bracketing with 5-dB steps is recommended, because 5 dB may be important
given a very limited residual dynamic range of hearing.
4.1.3 Stimulus Frequency Protocol
The frequencies at which tone pip ABR testing by AC will be done include at a
minimum 500 Hz and 2000 Hz. Both are important generally and also particularly
with regard to prescription of amplification. Even when 500 Hz and 2000 Hz are
- 34 -
normal, time permitting, it is preferable to also attempt to obtain results for 4000
Hz in order to rule out any high-frequency impairment.
If there is a severe or profound impairment at 2000 Hz, 500 Hz is a key measure
of residual low-frequency hearing. If reliable thresholds have been obtained at
500 and 2000 Hz, the clinical utility of measurements at 1000 Hz depends on the
500/2000 Hz threshold difference. If the threshold difference in dBnHL values is
less than or equal to 20 dB, subsequent clinical interpolation of the 1000 Hz
threshold as the average value will rarely be seriously in error. If the difference
in dBnHL thresholds is greater than 20 dB, then threshold measurement at
1000 Hz is mandatory, with the exception of a finding of a substantial conductive
component at 500 Hz. In that case, AC thresholds may change with time or after
medical management, so a detailed audiogram is not justified.
ABR threshold measurement at 4000 Hz is mandatory if the ABR threshold at
2000 Hz is more than 10 dB above the minimum mandatory level. Testing at
4000 Hz is recommended also if the 2000 Hz threshold is within normal limits but
OAE results indicate a clear abnormality isolated at 4000 Hz.
•
•
•
•
•
2000 Hz: In the absence of prior BCEHP Assessment results indicating
otherwise, testing will begin by AC at 30 dBnHL at 2000 Hz. Response or noresponse at 30 dB will be followed by testing the opposite ear at 2000 Hz/30
dB. If no response to AC 2000 Hz at 30 dBnHL in either ear, then the
mastoid(s) of the side where elevated by AC should be tested using BC 2000
Hz at a minimum of 30 dBnHL
If no response to BC 2000 Hz at 30 dB, increase to BC 2000 Hz intensity of 60
dBnHL.
500 Hz: AC at a minimum of 35 dBnHL.
BC 500 Hz at a minimum of 20 dBnHL is recommended (when 500 AC is
elevated), where time permits, but is not mandatory if BC at 2000 Hz has
been obtained. If the only AC abnormality is at 500 Hz, BC 500 Hz is
mandatory
BC testing will not be done at any frequency other than 500 and 2000 Hz.
Currently, BCEHP does not provide calibration values for other frequencies,
and the current literature is not sufficient to support BC ABR testing at
frequencies other than 500 and 2000 Hz.
4.1.4 BCEHP minimum ABR requirements
•
Tonepip ABR thresholds
o 2000 Hz AC (minimum of 30 dBnHL) both ears
o 500 Hz AC (minimum of 35 dBnHL) both ears
- 35 -
o 4000 Hz AC (minimum of 25 dBnHL) time permitting; mandatory
if 2000 AC threshold is greater than 40 dBnHL
o 1000 Hz AC (minimum of 35 dBnHL) time permitting; mandatory
if there is more than a 20 dBnHL difference between thresholds
at 500 and 2000 Hz.
o 2000 BC if 2000 Hz AC is elevated (minimum of 30 dBnHL)
o 500 BC is not mandatory given elevated 500 Hz AC if you have
2000 BC
o 500 BC is mandatory if 500 Hz AC is only elevation (BC 500 Hz
minimum of 20 dBnHL) and AC threshold is greater than 40
dBnHL
•
Given absent or severely abnormal wave V at all frequencies in one
or both ears, click-ABR for AN/AD, including cochlear microphonic
potentials and stimulus artifact analysis.
o Rarefaction and condensation (at least 2 replications) at 95
dBnHL; clamp tube run if early non-neural waveform seen
o Clicks are recommended for more severe SNHL losses even if you
do have wave V
4.1.5 Example Test Sequences
A. Normal:
1. 2000 Hz AC at 30 dBnHL each ear
if 2kHz present at 30 dB both ears then go to step 2.
2. 500 Hz AC at 35 dBnHL each ear
if 500 Hz present at 35 dB both ears then go to step 3.
3. 4000 Hz AC at 25 dBnHL each ear
if 4000 Hz present at 25 dB both ears then go to step 4.
4. 1000 Hz AC at 35 dBnHL each ear
ABR complete
Rationale: If the baby wakes up after 2000 and 500 Hz, you have lots of
information about the status of their hearing (4000 and 1000 Hz are not
"mandatory" if 500 Hz & 2000 Hz are normal).
B. Conductive:
- 36 -
1. 2000 Hz AC at 30 dBnHL both ears
if either or both ears show no response
2. 2000 Hz BC at 30 dBnHL for ear(s) without a response
if response(s) then go to step 3.
3. 500 Hz BC at 20 dBnHL
if response(s), then go to step 4.
4. 2000 Hz AC thresholds both ears
5. 500 Hz AC thresholds both ears
Rationale: If the bone conduction responses are present and the baby then
wakes up, you have information about the status of their cochlea. If middle ear
fluid is suspected (e.g. based on immittance results), AC thresholds can be
deferred as they are temporary and likely fluctuating.
C. Sensorineural:
1. 2000 Hz AC at 30 dBnHL both ears
if either or both ears show no response
2. 2000 Hz BC at 30 dBnHL for ear(s) without a response
if no response(s)
3. 2000 Hz BC at 60nHL dB and threshold for ear(s) with no response
4. Either:
a) 500 Hz BC
b) 2000 Hz AC threshold
5. 500 Hz AC threshold both ears
6. 4000 Hz AC threshold both ears
7. 1000 Hz AC threshold both ears, if greater than 20 dBnHL difference
between thresholds at 500 and 2000 Hz.
Rationale: If the bone conduction responses are absent and the baby then wakes
up, you have information about the type of hearing loss and limited information
about the degree of hearing loss.
D. Auditory neuropathy/auditory dys-synchrony (AN/AD) (follow protocol
sequence for SNHL): see below for more details re: AN/AD
- 37 -
1. 2000 Hz AC at 30 dBnHL first ear
2. 2000 Hz AC at 30 dBnHL second ear
if either or both ears show no response
3. 2000 Hz BC at 30 dBnHL for ear(s) without a response
if no response(s)
4. 2000 Hz BC at 60 dBnHL and threshold for ear(s) with no response
5. Either:
a) 500 Hz BC (no response at maximum level) or
b) 2000 Hz AC threshold (no response at maximum level)
6. Clicks:
a) Rarefaction first ear at 95 dBnHL
b) Condensation first ear at 95 dBnHL
c) Tube Clamp (use your haemostat) first ear
(Repeat a) through c) second ear if no wave V present at any
frequency)
7. 500 Hz AC (no response at maximum level)
8. 4000 Hz AC (no response at maximum level)
4.2
Air Conduction
Recording channels:
The contralateral channel has limited value and need not be retained in the
audiologic records. However, it is sometimes helpful on-line in identifying errors
of stimulated ear selection. It also may be useful to view ipsilateral/contralateral
response asymmetries when a significant interaural threshold asymmetry is
suspected (see Section 5.4.5).
4.3
Bone Conduction
4.3.1 Bone conduction – general information
- 38 -
1. Conduct BC-ABR testing at 2000 Hz, and if possible, 500 Hz if air conduction
thresholds are elevated at those frequencies. There is no research to support
the use of 1000 or 4000 Hz at this time.
2. It is important to obtain replications from both the ipsilateral and
contralateral channels. Information from both channels can be used to
determine which cochlea is responding in young infants; wave V is earlier and
generally larger in the channel ipsilateral to the responding cochlea. At nearthreshold levels, V-V’ complex is usually larger and wave V latency is usually
earlier in the channel ipsilateral to the cochlea responding most effectively.
The latency measures are the more reliable indicator: if amplitude and
latency indicators are in conflict, more weight should be given to the latency
information. The asymmetries are most evident at near-threshold levels, and
at 2000 Hz. They decrease with maturation; the asymmetries are not reliably
present in children older than 3-4 years of age. See below for more details.
3. Identify the ear tested (mastoid placement) and whether the EEG channel is
ipsilateral or contralateral to the BC transducer side (it will be important to
track this closely).
4. In the absence of prior BC-ABR information, all BC-ABR measures should
begin at the specified minimum levels (30 dB at 2000 Hz and 20 dB at 500
Hz), as ipsilateral/contralateral asymmetries are easiest to assess at these
levels. BC-ABR at the minimum stimulus levels provides an excellent
indication of whether the observed AC elevation (e.g., no response at
30dBnHL for AC 2 kHz) is conductive in nature (BC-ABR present with normal
asymmetry) or has a sensorineural component of undetermined size (BC-ABR
absent or opposite asymmetry).
5. Note that BC limits for 2000 Hz are 60 dBnHL and limits for 500 Hz are 40-45
dBnHL.
6. Because of limitations of the transducer, BC-ABR is not used to provide a
specific estimate of the air-bone gap; rather, it provides an excellent
estimation of the presence/absence of a sensorineural component. It also
may indicate whether the sensorineural component is in the mild/moderate
range or is greater. An absent BC-ABR at maximum BC level (45 dB at 500 Hz
and 60 dB at 2 kHz) indicates the presence of a significant sensorineural
component.
7. Elevated BC-ABR threshold with a present response at intensity above
minimum level (30 dB at 2000 Hz and 20 dB at 500 Hz) suggests a
sensorineural component in the mild/moderate range; however, the specific
amount of this elevation cannot be determined from current knowledge.
4.3.2 Bone conduction stimulus artifact
At the higher BC stimulus levels (40 dBnHL for 500 Hz, 60 dBnHL for 2000 Hz),
stimulus artifact can be very large. The IHS RN and SNR measures may be
- 39 -
contaminated and therefore unreliable when significant stimulus artifact is
present, especially for 500 Hz. Especially for this reason, alternating stimulus
polarity must be used for BC stimuli. Even with alternating polarity, however,
artifact is not entirely removed due to asymmetry of the BC transducer response
with stimulus polarity inversion.
Image of stimulus artifact with 500 Hz bone conduction at 45 dBnHL:
For BC 500 Hz, the problem is such that artifact can extend over a substantial
region of the analysis epoch and can increase the difficulty of reliable ABR wave
V identification, because there is no useful EEG display prior to the putative
response. Nevertheless, in the majority of cases, wave V remains identifiable.
The amount of artifact and the difficulty caused by it vary dramatically from
subject to subject, and the underlying factors are not fully understood.
For BC 2000 Hz, uncancelled stimulus artifact may appear similar to a wave I.
Evaluation of the rarefaction and condensation sub-averages, made possible
using the IHS “split buffer” feature, will usually indicate the “wave I” is due to
stimulus artifact. BCEHP clinicians must be careful to differentiate this artifactual
result from a real wave I, which might indicate a neurologic or AN/AD problem.
- 40 -
Image of stimulus artifact with 2000 Hz bone conduction at 60 dBnHL:
4.4
Waveform interpretation
4.4.1 Definition and Identification:
ABR threshold is defined as the lowest level giving replicated response-positive
records, either at the minimum required level or with replicated responsenegative records at a level not more than 10 dB lower. If such a procedure
yields an ABR threshold of 80 dBnHL or greater, use of a 5-dB final step is
recommended, because increased precision may be clinically useful with a limited
dynamic range of hearing.
4.4.2 Replication:
1. Typical past practice has required a minimum of two averages of about 2000
accepted sweeps each for any given stimulus condition; with liberal use of a
third average in cases of uncertainty about response presence or absence
strongly recommended (Ontario IHP Protocols, 2005; Stapells, 2000). The
use of less than 1000 trials is discouraged, and the routine use of larger
numbers of averages (per replication) of more than about 4,000 sweeps is
discouraged because it is inefficient, due to the law of diminishing returns
within averaging.
- 41 -
2. Typically, each replication will have at least 2000 trials in each average,
although averaging may either be stopped
a. early if waveform residual noise levels (see below) reach criterion levels
b. after more than 2000 trials if waveform noise levels are high
3. The routine use of more than four replications for a given stimulus condition
is not recommended. Rather than doing more than 3-4 replications
(assuming these are relatively noise free), it is more effective to increase the
stimulus level, where feasible, given persistent uncertainty about the
response.
4. Replication of waveforms is not required under the following special
circumstances:
a. waveform well below threshold that is below noise criterion, is flat, and
shows no suggestion of expected waveform: this may be determined as
“no response” provided it is not at the level bracketing threshold (i.e., not
at level immediately below threshold); and
b. waveform above threshold showing clear response provided the response
is typical/expected for condition and has a peak-to-peak amplitude that is
at least four times (4x) the residual noise level amplitude and/or the
Smart-EP “SNR” value is greater than 1.0. The non-replicated “response”
must not be at threshold.
4.4.3 Morphology:
Tone Pip/Threshold: The morphology of the ABR to tone pips is very different
from that seen in otoneurologic ABR testing with click stimuli. The waveform
changes are most marked near threshold.
1. Typically, the earlier waves of the ABR are absent, and the response is a slow
and late V-V’ negative-going transition.
2. There may be no wave V at all, but only a negative V’ peak.
3. There may also be a positive-going deflection following V’, at the end of the
analysis interval.
4. The tonepip response at 2 kHz usually shows a wave V that is more clearly
defined and sharper than at 0.5 kHz, and the 4 kHz response can be quite
similar to conventional click responses.
5. At moderate sensation levels there may even be earlier ABR waves (especially
at 2000 and 4000 Hz).
Click/Neurological: The morphology of the ABR to click stimuli is a sharp
waveform where waves, I, III and V are identifiable (waves II and IV might also
be identifiable).
- 42 -
4.4.4 Interpretation:
The interpretation of replicated waveforms as a “response” itself relies on four
main cues:
1. the occurrence of a response-like waveform at a grossly reasonable latency
(there will be a range depending on the tonepip frequency and intensity)
2. the prominence of that feature relative to the fluctuations in the remainder of
the average (a sort of visual signal-to-noise ratio),
3. the reproducibility of the feature across averages (a better measure of signalto-noise ratio)
4. comparison of the peak-to-peak amplitude of the supposed response to the
residual noise level (this should be 3-4 times the size of the residual noise)
Response present:
Visual criteria:
o the response amplitude is at least 3 times the amplitude of the average
difference (noise) of the waveform within the SNR region (see Section
5.5)
Statistical criteria:
o the SNR of the overall average of the waveforms is greater than 1.0 (see
Section 5.5)
o the CCR of the waveforms within the SNR region must be greater than or
equal to 0.46 (see Section 5.5)
Example of response present:
difference (noise) between waves)
- 43 -
Response absent:
Visual criteria:
o waveform should appear visually quiet, with no apparent replicable
waveform
Statistical criteria:
o waveform should be below RN criterion: 0.11 μV for single waveform, and
0.08 μV for combined (added) waveform (see Section 5.5)
Example of response absent:
Added wave:
SNR = 0.20 RN = 0.08μV
Cannot evaluate
If the putative response is not 3 times the amplitude of the average difference of
the waves, and either the SNR is <1.0 or the CCR is ≤ 0.46, then you cannot call
the waveform a response present. In addition, if the waveform is visually noisy,
or does not meet the criterion value for noise (ie, RN is not below criterion), you
cannot call the waveform no response. In both of these cases, the waveforms
cannot be evaluated for response presence or absence. Usually, a third (or more)
waveforms are indicated, until criteria for response presence or absence are met.
Errors of clinical judgment most often arise from over-interpretation of
uninterpretable waveforms. A conservative approach to waveform interpretation
will aid in ensuring the reliability of those judgments.
Examples of cannot evaluate:
- 44 -
In the image above, the waveforms do not correlate significantly over the SNR
region selected. You therefore cannot say there is a response present. While the
waveforms have RN values below criterion, the waveforms are not visually quiet
enough to say no response. At least one more repetition is required.
- 45 -
In the image above, there appears to be a small-amplitude peak that replicates
to some degree. However, visually, the amplitude of the waveform is not 3 times
the average “noise” of the recordings. The CCR of the two waveforms is below
criterion. You cannot say there is a response present here. The waveforms are
not visually quiet enough to say no response (in fact, it appears that there might
be a small possible response). At least one more repetition is required.
4.4.5 Air conduction – interpretation
1. Focus on the presence or absence of a wave V or negative V’ peak (there
might also be a positive-going deflection following V’).
2. The higher the frequency of the tonepip, the earlier the latency of wave V
3. A 2 kHz response usually shows a wave V that is more clearly defined and
sharper than lower frequencies.
4. A 0.5 kHz response often shows only a negative V peak.
5. A 4 kHz response can be quite similar to click responses.
6. Responses that are suprathreshold can show much shorter latencies than
those at threshold.
Example of normal ABR at 500, 2000, and 4000 Hz as per BCEHP criteria:
- 46 -
Unilateral hearing loss
In the case of unilateral hearing loss, care should be taken at high stimulus
levels (i.e., above approximately 60 - 70 dBnHL) to ensure that responses seen
to stimulation of the impaired ear are not due to cross-over.
As long as you are getting no response from the impaired ear, then cross-over is
not a concern. However, as soon as you see a response from the impaired ear
with intensities of 60 - 70 dBnHL and greater, you should ascertain which
cochlea the response is coming from.
There are two strategies for ensuring that responses are from the ipsilateral ear:
- masking of the contralateral (good) ear
- 2-channel recording.
- 47 -
Contralateral masking should be enabled prior to commencing acquisition of a
new waveform. To add contralateral masking, open the Stimulus parameters
dialogue box, accessed from the control panel on the bottom of the acquire
page.
A 2-channel recording must be chosen prior to acquiring a new waveform. To
change to a 2-channel recording, change “Mode:ipsi” to “Mode:both” on the
control panel at the bottom of the acquire page. Waveform interpretation is the
same as with bone conduction (see Section 5.4.6).
4.4.6 Bone conduction – interpretation
Best arrangement of waveforms for bone conduction:
ipsilateral waveforms on top of contralateral waveforms (for a given set of
stimulus parameters) for easiest viewing of relative latencies
in the case of multiple repetitions for a given intensity, it can be difficult to
select a definitive latency for wave V. Use the “add” feature of the IHS
(see Section 8) to combine all waveforms into one averaged waveform; it
will now be easier to mark the appropriate latency and amplitude of wave
V.
Example of best layout for bone conduction waveforms:
- 48 -
Determining which cochlea is responding
The latency measures are the more reliable indicator: if amplitude and latency
indicators are in conflict, more weight should be given to the latency information.
1. Start at minimum levels (30 dBnHL at 2000 Hz and 20 dBnHL at 500 Hz).
a. If the response is present/larger/earlier ipsilaterally vs.
absent/smaller/later contralaterally, you can assume the ipsilateral cochlea
is responding, and there is normal cochlear function at 2000 and 500 Hz.
b. If the response is present/larger/earlier contralaterally vs. the
absent/smaller/later ipsilaterally, you can assume that the ipsilateral
cochlea has a sensorineural component (although degree is not
determined). You can also assume that there is normal cochlear function
in the contralateral cochlea.
- 49 -
c. If there is no response in either the ipsilateral or contralateral channels,
you can assume that the ipsilateral cochlea has a sensorineural
component (although degree is not determined). You can make no
assumptions about status of the contralateral cochlea.
d. If the ipsilateral and contralateral responses are equal, you can assume
that one cochlea is responding, but you do not know which one.
Decreasing intensity and/or switching mastoids will likely clarify.
2. If there is no response at (30 dB at 2000 Hz and 20 dB at 500 Hz), go to the
limits of the transducer (60 dB at 2000 Hz and 40-45 dB at 500 Hz).
a. If the response is present/larger/earlier ipsilaterally vs.
absent/smaller/later contralaterally, you can assume the ipsilateral cochlea
is responding. Find threshold.
b. If the response present/larger/earlier contralaterally vs.
absent/smaller/later ipsilaterally, you can assume that the contralateral
cochlea is responding and the ipsilateral cochlea has a significant
sensorineural component.
c. If there is no response in either the ipsilateral or contralateral channels,
you can assume that the ipsilateral cochlea has a significant sensorineural
component. You can make no assumptions about status of the
contralateral cochlea.
d. If the ipsilateral and contralateral responses are equal, you can assume
that one cochlea is responding, but you do not know which one.
Decreasing intensity and switching ears might clarify.
- 50 -
Assuming that these waveforms are in
response to 2000 Hz bone conduction at 30
dBnHL (minimum level; i.e. if response
present at this level, cochlea is normal);
mastoid placement is on ipsilateral cochlea:
A. ipsi cochlea is responding and therefore
normal; status of contralateral cochlea is
unknown
B. contralateral cochlea is responding, and
therefore normal; the ipsilateral cochlea is
abnormal (i.e. sensorineural component)
C. the ipsilateral cochlea is abnormal (i.e.
sensorineural component); the status of the
contralateral cochlea is unknown
D. one of the cochlea is responding, and
therefore normal; it cannot be determined
which cochlea is responding
In the case of D in the figure above, it may be possible to resolve which cochlea
is responding by decreasing the intensity by 5 to 10 dBnHL.
In making the above interpretation, you must be assessing ABR wave V-V’ (i.e.,
wave V to the following negativity). Be careful not to confuse the Post-Auricular
Muscle Response (PAMR) with ABR wave V-V’
4.5
Objective aids to waveform interpretation
4.5.1 Residual Noise (RN):
The IHS has an on-line measure of the noisiness of the waveform being collected
called residual noise. RN can help one determine the overall “noisiness” of a
waveform.
1. No response/small response: RN is useful for those replications that appear to
have “no response” or where there is a possible small response as is seen at
or near threshold. If there is too much noise, as indicated by the RN, this
noise might mask a response, and thus the clinician can not reliably state that
there is no response present. RN should be used in determining when
enough trials/replications have been obtained, especially if no clear response
- 51 -
is present. Importantly, any determination of “no response” requires that the
average of all waveforms/ replications for a given condition not be greater
than a minimum noise criterion (see below) before “no response” might be
determined. If the overall residual noise level is higher than the criterion,
these waveforms apparently without a response must not be interpreted for
threshold purposes.
2. RN “no response” criterion: The final residual noise (after averaging all
replications together) should be no more than 0.08 µV. If two replications are
obtained, then recording each to an RN value of 0.11 µV or lower is
reasonably quiet. If the RN is not below the noise criterion, one cannot say
“no response”.
3. Response: Often, a large and clear wave V (for example) may be seen well
before noise criterion is reached. Thus, the requirement to reach noise
criterion applies to determination of no response, and not to an obvious
response. However, the residual noise level can aid in determining when a
response is likely significantly greater (3.2 – 4x) than the noise and thus
when averaging may be stopped and a second (or third) replication initiated.
4. Note: if significant stimulus artifact is present, the RN values are likely not
valid. This most often occurs for high-level 500-Hz air conduction (i.e. above
90 dBnHL) or 500 Hz bone conduction (30 to 45 dBnHL).
4.5.2 SNR
The SNR measure is also obtained online, and is a measure of the response
signal-to-noise ratio for the replication (or, if you have “added” all replications,
for the overall average). The larger the SNR, the more likely a response is
present. Recent research at UBC has indicated that the IHS SNR criterion may be
useful as an objective tool to waveform interpretation. SNR values for waveforms
collected at BCCH were compared to response presence/absence determinations
by 3 expert ABR clinicians. From these data, it was determined that an SNR value
of greater than 1.0 is indicative that a response is highly likely to be present.
Clinicians should use the SNR criterion in conjunction with visual appraisal of
replications to determine response presence.
Previously, CCR (see 5.5.3 below) has been the approved BCEHP objective
means of determining a high degree of waveform replicability, to be used in
conjunction with visual waveform judgment. As CCR and SNR are statistically
related measures on the IHS, it is redundant to use both. Now that a criterion
value for SNR has been established, it is recommended that this measure be
used instead of CCR. The main advantage is that the SNR is much quicker to
obtain for a given set of waveforms as it does not require setting the cursors and
splitting buffers.
- 52 -
On the IHS SmartEP, the SNR region will be set to whatever setup file is loaded
on the Acquire page (each stimulus setup file has its own SNR region set),
regardless of what page you are on. So if you just completed one frequency (e.g.
1000 Hz), and still have that stimulus loaded, then you go to another page
where you have responses to another frequency (e.g. 4000 Hz), the SNR for the
waveforms at 4000 Hz will be wrong as they will be based on the SNR region of
1000 Hz. So you must ensure that the correct SNR region is loaded before
obtaining SNR values for a given waveform.
When using the SNR criterion, you must still obtain replicate waveforms in
determining threshold – that is, at the lowest intensity at which there is a
response, and the intensity below (either 5 or 10 dB below) where there is no
response. The exception to this is at intensities that are not defining threshold,
i.e. when the level is at least 10 dB above or 20 dB below threshold.
To obtain the SNR of the average of all the waveforms that you have collected at
a given intensity and frequency, you will have to add all those waveforms and
then obtain the SNR region from the added wave. Make sure the SNR region is
correct first before adding a waveform, as the SNR value for a given waveform
will be based on the SNR region at the time of processing.
How to use the SNR criterion:
1. Collect waveforms. If you think you have a response, but want an objective
measure, you need to determine the SNR value for those waveforms. The
waveforms below are 4 runs of 500 Hz at 45 dBnHL.
- 53 -
2. Ensure the correct SNR region is set. Either you can load the stimulus of the
correct frequency (you might still be in that setup file), or you can manually
enter the SNR region from the top menu. Go to System, and check the SNR
region (circled in green). Here it is set correctly for 500 Hz AC.
3. If it isn’t correct, it must be changed. To do so, click on “Set SNR Region”. A
dialogue box will open; enter the desired SNR region (see 5.5.3 below).
4. Now you need to add the waveforms together to obtain your added wave.
Make sure your added waveform is activated, and you should be able to read
the SNR value from the display at the top of the window. In the waveforms
below, the added wave has an SNR value of 1.04. The response is therefore
likely present.
- 54 -
With noisy waveforms, the RN will be high, and this will of course affect the SNR,
making it lower. Also, at high intensities of air conduction and bone conduction,
a waveform that appears very quiet may have an abnormally high RN value due
to stimulus artifact, which will drive the SNR down. In this case, you might have
2 waveforms that replicate very well visually, and have a poor SNR. Therefore, at
high intensities of air conduction and bone conduction (especially at 500 Hz), RN
and SNR values may not be valid, and you may have to rely on visual
replicability.
4.5.3 Cross correlation (CCR)
The CCR is a measure of replicability between 2 waveforms (or between the split
buffers of a single waveform), and is determined after the recording is completed
(“under the Process menu”). To perform a cross-correlation on a pair of
waveforms, use the cursors (under the “Show” menu) to select the appropriate
CCR window (see below). Once the window is set, select both waveforms (using
Ctrl+click), and select “Cross-correlated selected” under the “Process” menu. In
general, a correlation of 0.46 or higher is strongly suggestive of a response
present. CCR and SNR are statistically related, therefore one does not need to
calculate both measures. As with SNR, however, visual observation of replicated
waveforms by an experienced/skilled clinician is the standard/accepted method.
Windows (or “SNR Regions”) must be set-up prior to any recordings to use RN
and/or SNR. (Windows for CCR can be set and modified by the clinician, and are
not pre-set). SNR region is set under “System”, and can be saved as part of the
settings. BCEHP stimulus setup files on BCEHP-issued IHS equipment have been
pre-set with the appropriate SNR regions. Windows are 10 ms in length, as
follows:
Stimulus
AC/BC Clicks
AC 500-Hz tones
BC 500-Hz tones
AC 1000 Hz tones
AC/BC 2000 Hz
tones
AC 4000 Hz tones
RN, SNR & Cross Correlation (CCR) Window
1.8 – 11.8ms
10.50 – 20.5
(Caveat: RN and SNR with 90-100 dBnHL AC stimuli)
20 dBnHL: 10.5 – 20.5 ms
30-45 dBnHL: 14 – 24 ms (only RN valid)
Note: higher BC is later because of stimulus artifact
7 – 17 ms
6.5 – 16.5 ms
5 – 15 ms
Note: Current IHS Smart-EP software (post August 8, 2005) allows one to recalculate (offline) RN and SNR values using different SNR regions (for example if
- 55 -
you used the wrong SNR regions). Simply set the new/correct region, then “split
buffer” then “add”.
CCR and Multiple Replications: Normally, one must have an even number of
waveforms to average to then obtain a single CCR. With current Smart-EP
software (post August 8, 2005), it is possible to “add” all the replications and
then obtain CCR the “split buffer”.
Otherwise one can use the following:
2 replications: between replication 1&2:
3 replications: between replication 1&2,
1&3 and 2&3:
4 replications: Two choices:
a. Preferred: average offline 1&3 and
2&4 to double the number of trial in
each replication then calculate CCR
between new waveforms
b. CCR between replication 1&2, 1&3,
1&4, 2&3 and 3&4:
4.6
Response present if r≥0.46
Response present if r≥0.46
Response present if r≥0.46
Response present if r≥0.46 for 3 or
more comparisons
Non-physiologic artifact
Any ABR recording may be contaminated by non-physiologic artifact, notably 60
Hz interference that is partially phase-locked. This is more likely to occur with the
lower high-pass cut-off frequencies used in tone pip ABR threshold
measurement. Given no adverse changes in the test environment (such as power
cable routing or proximity), by far the most common cause of child-specific 60 Hz
artifact is asymmetry of electrode impedances.
Some methods to confirm non-physiologic artifact:
1. clamped-tube recording
2. online EEG
3. comparison of recordings with and without the “60-Hz notch filter”
If non-physiologic artifact is present, efforts must be made to reduce electrode
impedance asymmetries as well as reducing the source of the artifact (e.g. turn
off dimmer switches and lights, unplug other electrical devices in the room).
Other strategies include braiding electrodes, and ensuring that the transducer
cables are as far away as possible from recording electrode cables. Using the 60Hz notch filter is not recommended however, as energy of the response
waveform at this frequency will be lost. Consultation with BCEHP is
recommended if artifact problems are persistent.
- 56 -
Image below: a nice quiet EEG, optimal for threshold ABR testing
Image below: an example of high frequency interference, most pronounced in
the bottom tracing. A slight change in rate (e.g. 38.5/s instead of 39.1/s) seems
to get rid of this. `
Image below: this interference superimposes a slow wave which inverts with
channel (A vs. B) (25 ms window shown). This greatly interferes with waveform
interpretation and is difficult to get rid of.
- 57 -
Image below: a slow-wave “interference”, which a clamp tube run indicates is
physiologic in origin (i.e. disappears with a clamp tube run). This slow-wave is
therefore actually a component of the response EEG.
clamp
tube run
Image below: 2 runs with significant 60 Hz interference (25 ms window shown).
Note that the waveform marked LF (line filter) has a slightly reduced amplitude
- 58 -
as compared to the one marked “no LF”. The line filter has therefore helped to
reduce the amplitude of the 60 Hz interference somewhat, but not enough to
allow for waveform interpretation.
4.7
Troubleshooting a noisy EEG
Things to try to reduce/investigate noise while testing (i.e. patient
present)
•
Reduce potential sources of artifact: turn off all unnecessary dimmer lights,
lights, equipment.
•
Ensure that if you are using a cart which has a powerbar feature (i.e. you can
plug the cart itself into an A/C source and use it to power other devices, such
as the laptop), do NOT plug in the cart as it will cause major electrical
interference.
•
Ensure electrodes are braided/taped together.
•
Observe the EEG – does the noise look rhythmic (more likely electrical) or
random (more likely physiologic)? If physiologic, perhaps baby needs to fall
into a deeper sleep. If electrical, focus efforts on reducing potential sources
of electrical artifact.
•
Stretch out all cables/wires, as coiled wires act more like an antenna.
•
Ensure wires from electrodes do not cross wires from transducers.
•
Have patient as far away from computer, and other sources of electrical
artifact as possible.
- 59 -
•
Ensure electrode impedance is as low as possible and the same on all
channels.
•
Try changing rate from 39.1 to 38.5 or 39.8 (only small changes are
necessary).
•
Use line filter (tick line filter box in EEG window to activate when not
acquiring) as a diagnostic: does the amplitude decrease significantly when
line filter is on? If so, you have 60 Hz noise, and you may need to reposition.
Only as a last resort should you actually collect data with the line filter on, as
low-frequency EEG information contributes significantly to the response.
•
Wait until baby is in a deeper sleep.
•
Reposition baby for more neck relaxation.
•
Move amplifier box to see if another position might be better (sometimes
small moves can make a big difference).
•
If nothing works to improve situation, start recording anyway: if the baby is
normal, you may still be able to see waveforms. If baby is not normal, you
may not be able to get a low enough RN to establish NR.
Things to try when patient not present
1. Turn on amplifier (USB box) without electrodes or Y-cord, and begin
collecting. You should get RN of around 0.04 μV. If the RN is large, this is an
indication of internal (IHS) noise.
2. Short the + and – together for one channel (using the y-cord), and collect
data from that channel. You should get a low RN value (0.08 μV or less).
3. Plug in all cords as you would for a 2-channel recording, place the recording
end of the electrodes in the location of interest (e.g. where baby’s head will
likely be in a real testing situation) and begin collecting. This will give you an
indication of noise for that given location (assuming no internal IHS noise
issues). You should find that the EEG is noisier when closer to sources of
interference such as lights and computers. You should also be able to find
areas of less EEG noise as well, which should be better for testing.
4. Attach electrodes in normal array for 2-channel recording to a volunteer.
Attach headphones; collect with your subject as relaxed as possible (neck
- 60 -
supported, as close to asleep as possible) using a stimulus at 40 to 50 dBnHL.
You should be able to see a response (ensure you are using a normal-hearing
subject) as well as obtain low RN values.
4.8
Logsheet
BCEHP audiologists will record information of acquired ABR waveforms on a
recording form during the assessment. Please see Appendix D for two versions of
the BCEHP ABR recording form. Information to be recorded includes the
filename, stimulus parameters, IHS online noise measures (RN, SNR), waveform
determination, and any observations of relevant patient/test environment details.
The recording sheet should be filled in concomitantly with waveform collection.
The purpose is to provide a record of the order of acquisition of ABR waveforms,
and the conditions in which they were recorded. If, for example, it is discovered
that an insert earphone is out while acquiring data, this should be noted on the
recording form. If the waveforms are reviewed at a later date, consultation of
the log sheet will indicate why certain waveforms were to be disregarded; this
information would not be available unless recorded at the time of data
acquisition. It will also provide information on the online decision analysis for a
given diagnostic ABR assessment; this will be important for later secondopinion/file review processes.
- 61 -
Diagnostic ABR Training Manual
Section 5: Technical Parameters
5.1 ABR Tone Pip Parameters
5.2 ABR Amplifier Gain Settings
5.3 ABR Artifact Rejection Parameters
- 62 -
5.1
ABR Tone Pip Parameters
Mandatory recording parameters for the various components of ABR testing are
given below. The bandwidth must be appropriate relative to the frequency
spectrum of the ABR waveforms of interest. Optimal high-pass cut-off
frequencies for ABR threshold estimation are lower than that typically used for
otoneurologic ABR testing. The analysis window must be long enough to
encompass the target response, so it must be greater for near-threshold
recording, especially for 500 Hz stimuli. Also, for threshold estimation the
stimulus repetition rate should be as fast as possible, given a required analysis
window length, because wave V is relatively unaffected by high repetition rates,
which yields more average per unit test time.
Air conduction:
Stimuli: "2-1-2" (cycles) linearly-gated tones OR 5-cycle Blackman without
plateau
Polarity: Alternating onset polarity
500 Hz: 4 ms rise/fall, 2 ms plateau (i.e., 10 ms total duration)
TDH-49: 0 dBnHL = 25 dB peak-to-peak equivalent (ppe) SPL
ER-3A: 0 dBnHL = 22 dB ppe SPL
1000 Hz: 2 ms rise/fall, 1 ms plateau
TDH-49: 0 dBnHL = 23 dB ppe SPL
ER-3A: 0 dBnHL = 25 dB ppe SPL
2000 Hz: 1 ms rise/fall, 0.5 ms plateau
TDH-49: 0 dBnHL = 26 dB ppe SPL
ER-3A: 0 dBnHL = 20 dB ppe SPL
4000 Hz: 0.5 ms rise/fall, 0.25 ms plateau
TDH-49: 0 dBnHL = 29 dB ppe SPL
ER-3A: 0 dBnHL = 26 dB ppe SPL
Bone conduction
BC (B71 Transducer):
Stimuli: "2-1-2" (cycles) linearly-gated tones OR 5-cycle Blackman tones (no
plateau):
Polarity: Alternating onset polarity
- 63 -
500 Hz: 4 ms rise/fall, 2 ms plateau (i.e., 10 ms total duration)
0 dBnHL = 67.0 dB peak-to-peak equivalent (ppe) re: 1 µN RMS
(dB ppe SPL = dB peak (peak hold) minus 3 dB (for tones)
2000 Hz: 1 ms rise/fall, 0.5 ms plateau
0 dBnHL = 48.7 dB ppe re: 1 µN RMS
Epoch length: 23 to 25 ms
Filters:
High-pass
Low-pass
Tone pip thresholds
All click recordings
30 Hz
30 Hz
Tone pip thresholds
All click recordings
1500 Hz
1500 Hz
Notch (“line”) filter: off
Ipsilateral noise: none (at present)
Contralateral noise: discretional – white noise 30 dB below tone behavioural
masking level
Clicks:
Repetition rate 21.1/s
condensation, rarefaction polarity as specified
5.2
ABR Amplifier Gain Settings
Amplifier gain must be set to a minimum of 100,000. A general recommendation
is that the gain should be such that about 5-10 percent of sweeps are routinely
rejected when the EEG is quiet. Operation such that the quiet EEG occupies only
a small proportion of the amplitude range within the rejection limits is highly
undesirable, because it results in negligible protection against substantial
artifacts.
Any infant may manifest high-amplitude myogenic bursts during a period of
otherwise quiet EEG. Artifact reject systems, even if set as just indicated, do not
provide complete protection against such bursts, which may rapidly distort an
otherwise clean average. Such bursts are preceded and followed by a few
sweeps of high-amplitude activity that may not reach artifact rejection levels.
Careful and continuous monitoring of the ongoing EEG is essential through
- 64 -
averaging: EEG amplitude increase should trigger immediate interruption of the
averaging, which should then be resumed after a quiet EEG is re-established.
Should the infants’ EEG deteriorate substantially during the test, the amplifier
gain should not be reduced because that will simply admit more noise into the
average. The reason for increase in myogenic noise levels must be dealt with at
the source. Actions include quieting the child, checking electrode impedances, or
simply waiting for the child to settle. It is strongly emphasized that if none of
these actions are successful, the ABR test must be terminated, because useful
ABR threshold estimates simply cannot be obtained in the presence of active
EEG. No result at all is preferable to an incorrect interpretation based on noisy
and unreplicable averages.
Infants may manifest EEGs with high myogenic levels even if they appear to be
resting quietly or sleeping. It is the EEG activity that determines whether it is
worthwhile to commence or continue averaging, not the child’s overt behaviour.
5.3
ABR Artifact Rejection Parameters
Stimulus artifact: BC 30 dBnHL or higher produces large artifact, which often
triggers EEG artifact reject. To correct, either set "artifact reject region" to after
end of stimulus (IHS SmartEP) or turn off EEG artifact reject.
- 65 -
Diagnostic ABR Training Manual
Section 6: Diagnosing Auditory Neuropathy/Dyssynchrony
6.1 AN/AD background
6.2 Differential diagnosis - techniques
- 66 -
6.1
Background
Auditory neuropathy/dys-synchrony (AN/AD) is defined conventionally by a
cluster of findings that includes an indicator of normal cochlear amplifier function
(normal or near-normal OAEs and/or present cochlear microphonic), absent or
severely abnormal ABR, absent middle-ear muscle reflexes, and actual hearing
impairment of any degree from mild to profound. The disorder is usually but not
always bilateral. OAEs may be absent for trivial reasons related to middle ear
conditions, so absence of OAEs does not rule out AN/AD. In a small proportion of
cases, in the absence of confounding middle-ear conditions the OAEs are absent
or degrade over time. The proposed mechanisms of AN/AD include inner hair
cell, synaptic and primary neuronal dysfunction, and any or all of these may be
operate in the individual case. AN/AD may or may not include a genuine
neuropathy, and there may or may not be other, concurrent peripheral
neuropathies present.
AN/AD may represent as much as 5-10 percent of all sensorineural hearing
impairment in infancy, but its actual prevalence is not yet well-understood. There
are several underlying variations, mechanisms and causes, at least some of
which appear to be genetic. Perinatal hyperbilirubinemia and anoxia are also risk
factors. It is also possible that auditory brainstem immaturity or damage
recovery such as from perinatal hypoxia may masquerade as AN/AD. In the
presence of AN/AD, ABR threshold estimates are not reliable. Slow cortical
potential recordings may be contributory.
ABR is often the test within the audiologic battery that provides the indicators
leading to a diagnosis of AN/AD (in conjunction with OAEs). A high intensity click
in a child with AN/AD will usually reveal cochlear microphonics, and absent or
highly abnormal ABR. It is possible to see a replicable low amplitude wave V at
high intensities with AN/AD (see below for an example); therefore, it is advisable
to record responses to high intensity clicks in all patients manifesting ABR
thresholds in the severe-profound range. In interpreting the high intensity click
response, care must be taken to differentiate CM, stimulus artifact and
neurogenic response, using the method noted below. The CM is readily apparent
as a series of large deflections in the 0-5 ms latency region that should invert
with change in click polarity. It is currently believed that given an absent or
highly abnormal click ABR, the presence of a large CM suggests a provisional
differential diagnosis of AN/AD, especially when OAEs are present.
The appropriate course of action, given a probable diagnosis of AN/AD, is at the
discretion of the individual BCEHP audiologist, in conjunction with discussion with
the family.
- 67 -
6.2
Differential diagnosis – techniques
When a clear and reproducible wave V or V’ is seen at any stimulus level in the
course of tone pip ABR measurements, significant audiometric inaccuracy due to
a retrocochlear disorder is very unlikely. The presence of prolonged interwave
intervals, by themselves, does not typically indicate ABR audiometric inaccuracy.
Therefore, it is not considered necessary to measure click ABRs routinely.
However, if there is no clear wave V-V’ at the highest tone pip levels available, or
only a small replicable waveform is seen to very high intensity tone pips, an ABR
measurement using a click at or near the highest level available (typically about
95 dBnHL) is required.
The waveforms on the left above indicate clear robust replicable waveforms,
indicative of good neural synchrony once thresholds are reached. AN/AD is not
indicated, and a high-intensity click is not required. The waveforms on the right
are replicable, but are quite low amplitude. A high-intensity click should be
performed to investigate the possibility that AN/AD is a contributor to this
hearing loss.
If there is no detectable ABR with identifiable wave V to tonepips at the highest
available intensity levels at all frequencies measured in any ear, by AC or BC (if
indicated), then an AC click ABR test will be done at 95 dBnHL in that ear.
Condensation and rarefaction recordings must be replicated, have at least 2000
sweeps per average with an overall noise criterion at or below criterion (as
above).
- 68 -
If a high intensity click is run, and an early multi-peaked waveform that inverts
with a change in polarity is seen, then care must be taken to ensure that the
waveform is indeed physiologic before determining that it is a cochlear
microphonic, and not due to stimulus artifact, which may also appear as an early
waveform that inverts with stimulus polarity.
The waveform on the left above is due to stimulus artifact, and the waveform on
the right is a cochlear microphonic. They are indistinguishable without further
investigation. A diagnosis of AN/AD cannot be made until a clamp tube run has
been performed.
In the case of early waveforms that invert with stimulus polarity, you must rule
out stimulus artifact. To rule out stimulus artifact, the stimulus delivery tube is
clamped (using hemostats). The insert and transducer must not be moved from
their position for the previous click recordings.
a) If the early waves remain, then they are due to stimulus artifact and there
is no reason to question the reliability of the tone pip ABR thresholds.
b) If the early waves are abolished, then they are likely to be CM, in which
case confidence about tonepip thresholds is decreased, just as it would be
if the early waves were of neural origin.
- 69 -
Image above: Example of a cochlear microphonic that disappears with a clamp
tube run (bottom set of waveforms). Time window shown is 0 to 13 ms. Stimulus
used was a click at 95 dBnHL, rarefaction and condensation. Note the reversal of
waveforms with the change in polarity (rarefaction vs. condensation).
The finding of a clear CM and absent neurogenic waves suggests AN/AD, and the
validity of ABR thresholds as estimators of perceptual thresholds is highly
questionable in that case. In the presence of AN/AD, ABR thresholds will
overestimate perceptual thresholds by an amount that may be very large.
If either record individually shows a clear wave V, the likelihood of substantial
inaccuracy in the tonepip threshold caused by AN/AD is low. If no waveforms are
seen (either CM-like, or ABR), then this is consistent with tonal findings of a
severe-profound sensorineural hearing loss, and AN/AD is unlikely. However,
OAEs should still be obtained.
- 70 -
Image below: Flowchart for differential diagnosis of AN/AD:
- 71 -
Diagnostic ABR Training Manual
Section 7: Using the Intelligent Hearing System (IHS)
SmartEP
7.1 IHS Manual – sections
7.2 Using the IHS – detailed instructions
- 72 -
7.1
IHS manual - sections
The Manual contains some useful sections, listed for you.
Launch Pad (following chapter on USB box installation)
1. Tutorial (Chapter 2)
2. Overview (Chapter 3)
3. Patient Utilities (Chapter 4)
SmartEP (following Glossary)
1. Tutorial (Chapter 2)
2. Main Menu (Chapter 3)
3. Dialog Boxes (Chapter 4)
4. Display Area (Chapter 5)
5. Control Panel (Chapter 6)
6. Troubleshooting (Chapter 10)
7.2
Using the IHS – detailed instructions
1. Starting the IHS Smart EP
a. Once Windows is loaded, double-click on the Intelligent Hearing Systems
icon and the “Launch Pad” should open
b. You can enter a patient ID at this point or open a module – if you do not
enter the patient ID at this point you will have to enter it each time you
open a new module
c. Open SmartEP
Launch Pad:
- 73 -
2. Loading Patient information
a. Enter the patient information from the “Launch Pad”. By doing so, this
patient information will be recalled for every program (i.e., Smart DPOAE,
Smart TrOAE, ASSR).
b. If you forget, and you go immediately to SmartEP, you can easily retrieve
the patient information when you go from SmartEP to another program.
c. Click on “Patient” in the main menu
d. Click on “New”
e. Enter the patient’s demographic data on the displayed screen
f. You’ll see 5 tabs including “Personal”, “Age”, “Contact”, “Medical” and
“Summary”.
g. One helpful feature is the “Age” feature.
h. IHS assigns an identifier. You can change this to another identifier
relevant to your patients (BCCH uses a hospital number). Note that once
you enter an identifier it cannot be changed.
- 74 -
3. SmartEP screen layout
a. Main Menu at the very top
b. Indicator bar underneath the Main Menu
c. Page Parameter (PP) Menu on the right hand side
d. One Acquisition page
e. Nine Display pages
f. Plot size adjustment
g. Page scroll
h. Control Panel on the bottom: allows you to manipulate parameters for
new recordings (i.e., doesn't affect previously recorded waveforms)
- 75 -
4. Loading SmartEP "setup" files for correct stimulus & recording parameters
a. Default parameters (i.e., setup automatically when SmartEP turned on)
should be for acquisition of AC 2000 Hz at 30dBnHL
b. To setup for different frequencies (or air vs bone)
c. Click “Load Settings”
d. Click appropriate setup file (i.e., Insert tone 2k)
e. Note that all parameters (such as SNR region, EEG window) have been
preset for each setup file.
- 76 -
5. Beginning recording of waveforms – click acquire
6. Keeping track of acquired waveforms
a. IHS automatically saves all completed recordings
i. The file name includes the initials of the patient, the ear of
stimulation, the EP type, intensity, and channel of the recording.
ii. If you were to record a waveform at 2000 Hz at 30 dB for Bart
Simpson’s left ear, the IHS file would be BSLA30B.1. The second
recording would be BSLA30B.2. If you change frequency and
record at 500 Hz at 30 dB for Bart Simpson’s left ear, the IHS file
would be BSLA30B.3
- 77 -
7. Loading SmartEP files to review acquired waveforms
a. Click on “Patient” in the main menu.
b. Click on “Open”.
c. Move your mouse or arrows to highlight your chosen patient.
d. Two options:
i. Under “Report” (Main Menu)
1. Highlight and Scroll down to “Load Report” (this is only
applicable if waveforms have been previously saved in report
form)
ii. Under “Data” (Main Menu)
1. Highlight and Scroll down to “Load File”. A listing of all data
files will be displayed
2. Highlight those files you wish to display
8. Activating and moving waveforms
a. To make a waveform active click on the “recording handle” – the small
circle to the left of the recording.
b. The waveform should display green and the recording handle black
i. You can drag waveforms using the recording handle
c. Inactive recordings are displayed red for the right ear and blue for the left
ear
d. To move all the waveforms to another page.
i. Go to the Page Menu at the side
ii. Click on PP (Page Parameters)
iii. Click on “Send all waveforms to Page __
9. Scaling waveforms
a. Scaling is found in the Display Area under PP
b. A submenu appears and the following options are given
i. 1, 0.9, 0.8, 0.7, and Normalized
- 78 -
ii. Do not use “Normalized” as this scales all responses to appear to
be the same size (unfair comparison)
iii. 0.7, 0.5, 0.3 is typically a good scale for all tones (you might need
to increase the scale to see small waveforms)
iv. 0.8 to 1.0 is typically a good scale for clicks
10. Viewing recording parameters of a given waveform
a. Click on the “recording handle” to activate waveform of interest
b. Right click on “recording handle” for parameter display
c. Select “File Name: … Info”
d. All the recording parameters will be displayed
- 79 -
11. Clearing waveforms
a. Click on the “recording handle” to activate
b. Right click on “recording handle”
c. Click the “clear” option in the submenu (This won’t clear this waveform
from the hard drive!)
12. Averaging (“adding”) waveforms
a. Click on the “recording handle” to activate first waveform
b. Hit CTRL and click on “recording handle” of second waveform
c. Go to “Process” (Main Menu)
d. Select “Add Selected”
e. Note that added waveforms are not saved unless you go to “Data” and
“Save file”
- 80 -
f. Experience indicates labelling new waveforms will also save them.
13. Splitting buffers
a. All tones are recorded in the alternate polarity; alternating data is
acquired the IHS stores rarefaction and condensation in alternate buffers
(they are displayed together). (If single-polarity clicks recorded, then split
buffers are each same polarity.)
b. What are some advantages to split buffers?
i. Can see rarefaction and condensation responses after the fact
ii. If the baby wakes up and there appears to be a response, but
there is only one run of 2000 sweeps, one can split the buffers . . .
If there is a replicable response in the two buffers, you can be
more sure that there truly is a response there
iii. Note – if you split buffers, these “new” responses will not be saved
unless you save them
14. Labelling Waveforms
a. Click on the “recording handle” to activate waveform
b. Right click on “recording handle”
c. Select “Mark Peak”
- 81 -
15. Measuring amplitude
a. Click on the “recording handle”
b. Drag the lower peak marker (upward pointing triangle) to the most
negative (largest) following valley within 8 ms (tones) or 6 ms (clicks)
following wave V. (Largest valley within 2 ms following wave I.) Note: this
may not be the “next” or “closest” valley after the peak.
c. The peak-to-peak amplitude will be printed if the negativity (valley) is
marked as per above
d. You can also view the amplitude by
i. Right clicking on “recording handle”
ii. Select “peak latency”
iii. Or Select “show text side” or “show text below” (this will be
displayed on the screen)
- 82 -
16. Removing labels (Page 11 of the SmartEP manual)
a. Click on the “recording handle” to activate waveform
b. Right click on “recording handle”
c. Select “Remove Peak”
17. SmartEP RN and SNR measures
- 83 -
a. The IHS has an online real time analysis of the “noise” in the recording
called the Residual Noise.
b. The residual noise (RN, in μV) calculation is there to assist you, but does
not take the place of your judgment.
i. RN values should be less than 0.11 μV each or less than 0.08 μV
when waveforms added together to conclude “no response” (see
5.5.1)
c. There is also a signal-to-noise ratio (SNR) calculation
i. If the SNR is greater than 1.0, it is likely that there a response is
present (see 5.5.2).
18. Using your cursors
a. Click on “Show” (Main Menu)
b. Click on “Cursors” – The rightmost cursor displays time position and
latency
c. Drag to the start and end of the desired 10 ms window (see 5.5.3) using
the handles
- 84 -
19. Using the Cross-correlation tool (CCR)
a. The SmartEP has a statistical tool that allows you to compare two
waveforms for degree of correlation. This can be used as an aid to your
visual appraisal of waveform replicability.
b. To use, set the cursors to the desired window as above
c. Select the 2 waveforms of interest using ctrl+click
d. Click on “Process” (Main Menu)
e. Select “cross-correlate selected”
f. A dialogue box will open with the CCR value of the 2 waveforms
g. the CCR must be 0.46 or greater to call the waveforms “response present”
- 85 -
20 Printing
c. Click on “Print” (Main Menu)
d. Click on “Page” or “All pages”
19. Backing up your data
a. Insert disk into alternative appropriate drive
b. Click on “Patient” (Main Menu)
c. Click on “Open” or “Edit”
d. After selecting appropriate patient
e. Click on “Backup” (Demographics Dialog Box)
- 86 -
Diagnostic ABR Training Manual
Section 8: Reporting
8.1 ABR waveform display
8.2 Estimated Hearing Levels (EHL)
8.3 Sample reports
- 87 -
8.1
ABR waveform display
Display:
1. AC and BC waveforms
a. Superimpose all waveforms
b. Split Page
c. Display AC and BC waveforms on separate pages
d. Waves should be “blown up” reasonably large (so that a small response
may be seen). Scaling should be between 0.3 and 0.7 for tones, and
approximately 0.8 to 1.2 for clicks.
e. Consider not having pages too crowded
2. Order of AC waveforms:
a. Separate from BC waveforms (not the same page)
b. Consider not having pages too crowded with waveforms
c. Low frequencies on top, high frequencies on the bottom
3. Order of BC waveforms:
a. Place ipsilateral and contralateral waveforms on top of one another (not
side by side) for comparison of latencies
4. Order of pages attached to report:
a. Log sheets
b. AC tones
c. BC tones
d. Clicks
8.2
Estimated Hearing Levels (EHL)
Tone pip ABR thresholds in dBnHL are not directly equivalent to perceptual
thresholds in dBHL, and both dBnHL and dBHL are defined with reference to
adult norms. ABR dBnHL thresholds are converted to bias-free estimates of true
perceptual threshold in dB HL by applying adjustment factors based on
longitudinal validation studies.
ABR thresholds will be converted to estimates of the true perceptual threshold in
dBHL by the application of the threshold adjustment factors listed below. The
resulting thresholds will be referred to in the BCEHP context as Estimated
Hearing Level or EHL thresholds, with units in dB EHL. All reports and outcomes
data will indicate both the dBnHL threshold and the EHL value.
ABR thresholds are conventionally expressed in dBnHL and are not generally
equal to perceptual thresholds in dBHL. Therefore, an offset adjustment for bias
of ABR thresholds is required. The adjustments are derived from normative data
relating ABR threshold in early infancy to subsequent behavioural thresholds.
- 88 -
The BCEHP adjustments for EHL are between -15 and 0.
The thresholds derived by ABR measurements have an indirect and statistical
relationship to true hearing levels. There are many factors affecting the
relationship. First and foremost, the level at which an ABR is detectable depends
upon a host of variables that affect ABR detectability, including EEG noise levels,
filter bandwidths, averaging parameters, response detection criteria and
threshold bracketing procedure. Second, the units of ABR thresholds are dBnHL,
which is itself subject to many variables such as stimulus envelope and repetition
rate, which affect psychophysical energy integration. Such integration itself is
affected by the type and degree of hearing impairment, so threshold
relationships may also depend on the type and degree of impairment. Also,
dBnHL are defined with reference to adults, not infants. Furthermore, dBHL
itself is defined with reference to adults, and it is known that in the maturing
infant ear, canal SPLs and middle-ear transfer functions are different from those
of adults, and also may change significantly over time, especially at frequencies
above 1 kHz and in the first six months of life.
The adjustment factors will be applied to each ABR threshold in order to derive
an “Estimated Hearing Level" or “EHL”, which is a relatively bias-free estimate of
the actual hearing level in dBHL that would be obtained if the child developed to
adult anatomical and psycho acoustical status with no change in actual level of
hearing impairment. Given this approach, the target disorder is equivalent to
ABR threshold estimates that are adjusted to greater than 25 dBEHL, and the
ABR thresholds in dBEHL may be used directly in any subsequent prescription for
amplification.
The EHL conversion adjustments are derived from longitudinal follow-up studies,
primarily comparing early ABR with subsequent VRA thresholds (see Stapells,
2000). It should be noted that VRA thresholds are themselves generally greater
than the true psycho acoustical thresholds, which tends to reduce the observed
differences between ABR and behavioural thresholds. Conversely, because of
the effects of ear canal maturation, the observed relationships between ABR and
behavioural thresholds will incorporate the effects of maturational SPL changes in
the developing ear. The actual threshold SPLs in early infancy will be greater
than those for the same stimulus at the point of subsequently behavioural
threshold measurement, especially at higher frequencies, so the results may give
an impression of progressive impairment.
There is a clinical impression that ABR thresholds are closer to behavioural
thresholds when hearing impairment is severe, and this is often explained by
appeal to a recruitment-like phenomenon. Another factor that affects threshold
relationships is spectral spread of ABR stimuli, which tends to lower ABR
thresholds relative to true perceptual thresholds, especially at high frequencies
and with severe, sloping high-frequency impairments.
- 89 -
Many factors influence the key elements in normative threshold relationships that
determine the adjustments to derive EHL estimates. Such adjustments are
dependent on stimulus frequency, but currently the best evidence is that they do
NOT depend substantially on stimulus level or, concomitantly, the true hearing
level, over the range of interest in BCEHP threshold measurements. Accordingly,
the current BCEHP conversion adjustments are specific to stimulus frequency,
but not to observed ABR threshold level.
All BCEHP ABR reports must provide the ABR thresholds in dBnHL and the
estimated behavioural thresholds in dB EHL.
BCEHP Minimum Required Levels in dBnHL and ABR threshold adjustment factors
for Estimated Hearing Level (EHL) derivation.
Air Conduction
Bone Conduction
Frequency (Hz)
500
1k
2k
4k
500
2k
Minimum Level (dBnHL)
35
35
30
25
20
30
Adjustment (dB)*
-15
-10
-15
-0
NA
NA
* For AC ABR threshold estimates greater than 70 dBnHL, if 5-dB final step size
is used the absolute value of the Adjustment should be reduced by 5 dB at all
frequencies.
The rationale is that with a 10-dB step size, the possibility of response presence
at a level 5 dB lower (untested) is included in the statistical adjustment for bias,
whereas with a 5-dB step there is no such possibility, since the lower level is now
demonstrated to be response-negative.
Examples:
2k 80dBnHL (+), 70dBnHL (-): EHL = 80 -10 = 70 dBEHL
2k 80dBnHL (+), 75dBnHL (-): EHL = 80 -10+5 = 75 dBEHL
where (+) and (-) represent definite response detection outcomes.
** For any AC ABR threshold, it is discretional to reduce the absolute value of
the Adjustment by 5 dB, if the response at the lowest level considered positive is
minimal AND the EEG noise level is very low.
The rationale is that with exceptionally quiet EEG, the ability to identify small,
near-threshold responses is increased, and if such a response is seen, the
negative offsets normally used are likely to be on average excessive.
Examples:
- 90 -
500 60dBnHL (+), 50dBnHL (-):
EHL = 60 -15 = 45 dBEHL
500 70dBnHL (+), 60dBnHL (small V; lowRN), 50dBnHL (-): EHL = 60 -15+5 = 50
dBEHL
Note that application of these adjustment factors may occasionally yield small,
negative air-bone gaps. Such a finding is to be expected, given that the
adjustments are based on group mean normative data.
8.3
Sample reports
Sample Report # 1:
AUDIOLOGY REPORT
Patient name: DW
DOB:
Address:
Assessment: May 15, 2006
Unit #:
Telephone:
cc:
Foster Mother
Dr. G. B.
Public Health Unit.
File: BCCH Health Records; Aud Dept
History:
• DW was seen for an auditory brainstem response (ABR) assessment upon
referral from Dr. G. B, because she referred on an otoacoustic emissions
screening at St. Paul’s Hospital.
• DW’s foster mother indicated that DW was discharged from St. Paul’s Hospital
with neurological concerns including decreased muscle tone and floppy head,
tremors, and vacant spells.
• DW’s foster mother indicated that DW’s father had hearing loss as a child,
and wears bilateral hearing aids.
• DW’s foster mother indicated that she has observed DW to startle to loud
sounds.
Results:
• Distortion Product Otoacoustic Emissions (DPOAEs): DPOAEs were present
from 1500-6000 Hz; indicative of normal outer hair cell function at those
frequencies.
• Acoustic Immittance: Tympanograms indicated normal tympanic membrane
movement bilaterally.
- 91 -
•
•
Auditory Brainstem Response (ABR): ABR testing for estimation of pure-tone
thresholds was conducted using insert phones. ABR thresholds were
considered to have “good” reliability and were consistent with peripheral
hearing within the normal range from 500-4000 Hz bilaterally.
Below is a summary of the ABR:
ABR results
Stimulus
Right
ear
(dB
nHL)
Left ear
(dB
nHL)
dBnHL
to EHL
conversion
Right
ear (dB
EHL)
Left ear
(dB
EHL)
500 Hz air conduction (normal = ≤ 35 dBnHL)
≤ 35
≤ 35
-15
≤ 20
≤ 20
1000 Hz air conduction (normal = ≤ 35 dBnHL)
≤ 35
≤ 35
-10
≤ 25
≤ 25
2000 Hz air conduction (normal = ≤ 30 dBnHL)
≤ 30
≤ 30
-5
≤ 25
≤ 25
4000 Hz air conduction (normal = ≤ 25 dBnHL)
≤ 25
≤ 25
0
≤ 25
≤ 25
Impressions & Recommendations:
• ABR results are consistent with peripheral hearing thresholds within the
normal range from 500-4000 Hz bilaterally.
• Because of DW’s family history of hearing loss, DW should have follow-up
behavioural testing at the Public Audiology Clinic. The clinic will contact the
family for an appointment. For the information of DW’s parents, the Public
Audiology Clinic is at 00 Highway and can be reached at (604) 000-000.
• Please feel free to call (604) 123-4567 should there be any questions.
_____________________________
Jane Doe, M.Sc., Aud (C)
Registered Audiologist
- 92 -
Sample report #2
Cc:
otolaryngologist
family physician
pediatrician
Public Health Audiology Clinic
IDP
Parents
Aud Dept.
BCCH Health Records
Patient name: EL
Unit number:
DOB:
Date seen:
Address:
Phone:
Background information:
EL, aged 10 months, was seen for an auditory brainstem response (ABR) assessment at
BC’s Children’s Hospital on October 4, 2006. EL has Down Syndrome; she had surgery
to repair an AV septal defect. Her parents report that they are not sure that she hears
well, as she responds inconsistently to sounds. She has not had any ear infections;
there is no family history of hearing loss. She was seen at the Public Health Audiology
clinic for otoacoustic emissions testing on three occasions, but results were
inconclusive. She was therefore referred for a sedated ABR.
Audiological Results:
•
•
•
•
•
Auditory Brainstem Response (ABR): ABR testing for estimation of pure-tone
thresholds was conducted using insert headphones under conscious sedation. ABR
thresholds were considered to have “good” reliability.
Dr. ENT reported that EL’s ears appeared clear bilaterally.
Distortion Product otoacoustic emissions were present at 2500, 3000, 7000 and
8000 Hz in the right ear, and from 3500 to 7000 Hz in the left ear. Otoacoustic
emissions are indicative of normal outer hair cell function at those frequencies
present.
ABR thresholds were consistent with a normal hearing bilaterally from 500 to 4000
Hz.
Below is a summary of the ABR:
- 93 -
ABR results – dBnHL
Right ear
Left ear
dBnHL to
eHL
(estimated
hearing level)
conversion
500 Hz air conduction (normal = ≤ 35 dBnHL)
≤35
≤35
-15
1000 Hz air conduction (normal = ≤ 35 dBnHL)
DNT
DNT
-10
2000 Hz air conduction (normal = ≤ 30 dBnHL)
≤30
≤30
-5
4000 Hz air conduction (normal = ≤ 25 dBnHL)
≤25
≤25
0
Stimulus
DNT = did not test
Impressions:
1. The ABR results are consistent with normal peripheral hearing sensitivity from
500 to 4000 Hz bilaterally.
Recommendations:
1. EL should have follow-up behavioural testing given that she is at risk for middle
ear pathology. This is best accomplished at the Public Health Audiology clinic. For
the information of the family, the clinic is located at ****, and can be reached at
*****.
Please contact me at 604-123-4556, local xxx with any questions or concerns.
______________________
John Doe, M.Sc., Aud (C)
Registered Audiologist
- 94 -
Appendix A - Risk Indicators
BCEHP Risk indicators for progressive/late onset hearing loss
A. Craniofacial – an obvious craniofacial anomaly (not pits or tags), e.g. cleft palate
(NOT cleft lip in isolation), microtia/atresia
B. Family history of permanent hearing loss in early childhood before age 12 years,
first cousin or closer to baby (parents, siblings, uncle/aunt, cousin, grandparent),
irrespective of degree of hearing loss.
C. Syndrome associated with progressive/late onset hearing loss (e.g. Pendred,
Branchio-Oto-Renal, Alport, Usher, LVA, neurofibromatosis, osteopetrosis, Down
Syndrome)
D. Birthweight less than 1200 grams
E. Breathing problems:
i. Five-minute APGAR score less than or equal to 3
ii. Hypoxic-Ischemic Encephalopathy (HIE) moderate/severe, (Sarnat II or
III)
iii. Congenital Diaphragmatic Hernia (CDH)
iv. Extra-Corporeal Membrane Oxygenation (ECMO) or inhaled Nitrous Oxide
(iNO) or High-Frequency Oscillatory (HFO) or Jet (HFJ) ventilation
F. Brain dysfunction:
i. Intra-ventricular Hemorrhage (IVH), Grade III or IV (IV are seen by
Neonatal Follow-up)
ii. Peri-ventricular Leukomalacia (PVL)
G. Hyperbilirubinemia ≥ 400 μM OR meeting any standard criteria for exchange
H. Lab proven infection:
i. perinatal (in the baby) TORCHES infection (toxoplasmosis, rubella,
Cytomegalovirus (CMV), herpes, syphilis)
ii. Meningitis, irrespective of the pathogen.
I. Accidental overdose of Gentamycin or other aminoglycosides, five-fold or greater
- 95 -
Appendix B: BCEHP Diagnostic Clinical Pathways
Diagnostic Audiology Clinical Path
Stage I or II Screen Pass
With Late Onset Risk Factor
Determine Risk
Factor
Low Birthweight, Low
Apgar, TORCHS,
Respiratory Distress,
Head Trauma or Brain
Disorder,
Suspected syndrome
Cleft Palate and
Craniofacial
1
2
ECMO
CDH
3
Neonatal
Meningitis
4
Family
History
5
- 96 -
Low Birthweight, Low Apgar, TORCHS, etc
Behavioural
Audiology appt at
6 to 9 months C.A.
Pass/Refer
Refer
Report to RC
NORMAL HEARING
Report to RC
Behavioural
Audiology appt at
36 months
Go To Diagnostic
Clinical Path
Refer
Report to RC
Pass/Refer
NORMAL HEARING
Report to RC
1 – Low Birthweight,
Low Apgar, TORCHES,
Respiratory Distress,
Head Trauma or Brain
Disorder,
Suspected syndrome
(not including Cleft
Palate – see 2)
Discharge from
BCEHP
At 36 months
- 97 -
Cleft Palate and Craniofacial
Behavioural
audiology at 6 to 9
months C.A.
2 – Cleft Palate and
Craniofacial
Pass/Refer
Go To Diagnostic
Clinical Path
REFER
Report to RC
NORMAL HEARING
Report to RC
Behavioural
audiology workup
every 6 months
- 98 -
ECMO & CDH
3
Diagnostic ABR
workup
At 4 months
C.C.A. at BCCH
Pass/Refer
REFER
NORMAL HEARING
Report to RC
Report to RC
Behavioural
audiology workup at 8
months C.C.A. at
BCCH
REFER
Pass/Refer
Report to RC
Go to Diagnostic
Clinical Path
NORMAL HEARING
Report to RC
Behavioural audiology
workup at 12 months
C.C.A. in community
REFER
Report to RC
Pass/Refer
NORMAL HEARING
Report to RC
Behavioural audiology
workup at 18 months
C.C.A. at BCCH
REFER
Report to RC
Pass/Refer
Refer
Behavioural audiology
workup at 24 months
C.C.A. in community
3 – ECMO & CDH
Pass/Refer
Behavioural audiology
workup at 36 months
C.C.A. at BCCH
NORMAL HEARING
Pass/Refer
Report to RC
NORMAL HEARING
Report to RC
Discharge from
BCEHP
At 36 months
- 99 -
- 100 -
- 101 -
Level One
By 1
month
Stage II
screen
Refer
or refer
from DORM
Initial Diagnostic Appointment
made within 4 -8 weeks C.A.
Or at refer from DORM
Discharge from
BCEHP
Pass
report to RC
Threshold ABR
OAE/Immittance
when possible
Pass/Refer
Refer
Report to RC
Pass/Refer?
Pass
report to RC
Discuss w. R.C. or
BCEHP
Diagnostic
Consultant
Refer
Report to RC
Medical Referral/
ENT consult
OAE/Immittance
Conductive HL
Medical Diagnosis
Permanent CHL OR
Conductive HL
Yes
By 3rd
month
Audiology
monitoring
(Immittance/OAE)
Permanent Hearing Loss is
Suspected and/or Confirmed
Report to RC
Permanent
HL?
Referral to BCEHP Early
Intervention Coordinator
as soon as PCHL is
suspected – may be prior
to completion of
Diagnostic workup
By 6th
month
Medical Referral
Team Conference: Identify
case manager, plan
coordination of early
intervention, medical,
audiological management
Amplification
May in some cases be
initiated prior to
completion of
diagnostic work up
No
Report to RC
Discharge from
BCEHP; monitor
per Public Health
mandate
- 102 -
Appendix C: BCEHP ABR Protocol Flowchart
- 103 -
Appendix D: ABR Recording Sheets
- 104 -
- 105 -
Appendix E: ABR PROTOCOLS FOR CHALLENGING CASES
ATRESIA
Why test AC (supra-aural) before BC?
There have been cases missed by ENT, confirmed by ABR, where a pin-hole opening
was present in canal allowing some normal or near-normal thresholds.
Physical examination does not differentiate bone vs. cartilage in place of the meatus
(ABR thresholds might provide a clue).
UNILATERAL ATRESIA (Typical Results):
Tonepip ABR by air conduction (AC):
2 kHz at 30 dBnHL non-atretic ear
if 2kHz present at 30 dB non-atretic. . .
500 Hz at 35 dBnHL non-atretic ear
500 Hz present at 35 dB non-atretic ear
2 kHz AC at 30 dB with supra-aural phones in the atretic ear
if 2kHz AC absent at 30 dB in atretic ear. . .
2 kHz BC at 30 dB for the atretic ear
if 2kHz BC present at 30 dB in atretic ear. . .
500 Hz BC at 20 dB for the atretic ear
if 500 Hz BC present at 20 dB in atretic ear. . .
2 kHz AC with supra-aural phones (find threshold) atretic ear
500 Hz AC with supra-aural phones (find threshold) atretic ear
4000 Hz AC non-atretic ear
4000 Hz AC with supra-aural phones (find threshold)
ATYPICAL RESULTS
1.
2.
3.
4.
5.
Conductive hearing loss in the non-atretic ear
Sensorineural hearing loss in the atretic ear
Sensorineural hearing loss in the non-atretic ear
Combinations of 1, 2, and/or 3 (ABR Hell)
Normal hearing in both ears (ABR Heaven)
- 106 -
UNILATERAL ATRESIA (Atypical Results – Conductive HL in the non-atretic
ear):
2 kHz AC at 30 dBnHL non-atretic ear
if 2kHz AC absent at 30 dB non-atretic ear. . .
2 kHz BC at 30 dBnHL non-atretic ear
if 2kHz BC present at 30 dB non-atretic ear. . .
500 Hz BC at 20 dBnHL non-atretic ear
if 500 Hz BC present at 20 dBnHL non-atretic ear. . .
AND/OR:
2 kHz AC threshold non-atretic ear
2 kHz AC at 30 dBnHL with supra-aural phones
if 2kHz AC absent at 30 dBnHL in atretic ear
2 kHz BC at 30 dBnHL for ear with atresia
if 2kHz BC present at 30 dBnHL in atretic ear. . .
500 Hz BC at 20 dBnHL for ear with atresia. . .
if 500 Hz BC present at 20nHL dB in atretic ear. . .
Go back to the non-atretic ear:
2 kHz AC threshold if not already obtained
500 Hz BC at 20 dBnHL if not already obtained
500 Hz AC threshold
4000 Hz AC threshold
BILATERAL ATRESIA (Typical Results):
2 kHz AC at 30 dBnHL first ear (supra-aural phones)
2 kHz AC at 30 dBnHL second ear (supra-aural phones)
if either or both ears show no response…
2 kHz BC at 30 dBnHL for ear(s) without a response
if BC response(s). . .
500 Hz BC at 20nHL dB
Use ipsi/contra asymmetries to determine primary responding cochlea
Once BC has been determined:
2 kHz AC with supra-aural phones (find threshold) for each ear
500 Hz AC with supra-aural phones (find threshold) for each ear
4000 Hz AC with supra-aural phones (find threshold) for each ear
OTOTOXIC MONITORING:
Infant should have already had AABR or OAE hearing screening once BCEHP
screening is fully rolled-out
- 107 -
Tonepip ABR by air conduction (AC):
4 kHz –25 dBnHL in each ear
if no response at 25 dBnHL in one or both ears
2 kHz – 30 dBnHL in each ear
if no response at 30 dBnHL in one or both ears
2 kHz – find threshold for one or both ears
4 kHz – find threshold for each ear
6 kHz – find threshold for each ear
8 kHz – find threshold for each ear
AN/AD or Other Neurological Issues:
Tonepip ABR:
2 kHz AC at 30 dBnHL first ear
2 kHz AC at 30 dBnHL second ear
if either or both ears show no response
2 kHz BC at 30 dBnHL for ear(s) without a response
if no response(s)
2 kHz BC at 60 dBnHL and threshold for ear(s) with no response
Then, either:
500 Hz BC
or
AC 2kHz threshold
If you
•
•
•
get:
No responses for BC and AC for 2000 Hz at the limits of the Smart-EP, and/or
Clear early waves without Wave V and/or
Unidentifiable responses (i.e., Wave V not clear)
You must ensure you’re not dealing with AN/AD or neurological issue instead of a
cochlear SNHL. How?
ABR using rarefaction clicks both ears
at least 2 replications of 2000 sweeps at maximum intensity (around 95 dBnHL)
ABR using condensation clicks both ears
(Important: RN must be ≤0.11 for each replication)
If there is a clear no response for the rarefaction and condensation buffers you
don’t need the following “clamp tube” run; rather go back to finding threshold
with tones
- 108 -
If there appears to be a response:
Click ABR using rarefaction or condensation clicks for ear with a response while
CLAMPING tube with hemostat (2 replications)
(note: keep everything else identical to your previous runs – don’t move the
transducers or the box)
Repeat sequence for the second ear (if there was a response to initial clicks)
What might you see?
1. No response to clicks (unclamped) (previously mentioned) – no indicators for
retrocochlear pathology
2. A clear wave V response (the response might be originating from frequencies not
tested if until now you have not obtained any clear wave Vs) – AN/AD is ruled
out
3. CM present, which disappears in clamp-tube run. The direction of the large
cochlear microphonic will reverse with the change in polarity of the stimulus
(click-CM occurs approximately at 0.3 ms and lasts for about 2 ms with major
peaks at 0.5 and 1 ms) – AN/AD
4. Early waves present (e.g. I, III) (which do not invert with rare vs. cond) with
absent, degraded or delayed wave V: brainstem dysfunction
- 109 -