Lecture 5 Flashcards
When was the first ABR recorded in infants?
1974
Why are ABRs most useful in infants?
Because behavioural audiometry is not possible so we need an objective measure
What is the estimate of prevalence of infant hearing loss? At what dB HL?
1/1000 at 40 dB HL or greater
What are the main principles from the joint committee on infant hearing (2007)?
-All infants should be screened within 1 month
-Full audiologic/medical evaluation by 3 months
-All infants with HL should have intervention by 6 months
What 2 kinds of OAEs are used to screen healthy babies?
TEOAE & DPOAE
What is the automated auditory brainstem response used for (AABR)?
-AABRs are used for screening in NICU
-Assesses more of the auditory system compared to OAEs
What are the 3 possible categories of results for ABR?
- Response (hearing)
- No response (not hearing)
- Noisy result (inconclusive)
Which wave is most important in an ABR?
Wave V because it’s big enough to give you thresholds
Electrode placement locations: active, reference, and ground
-Active (non-inverting) at vertex (Cz) or forehead (FPz)
-Reference (inverting) at mastoid (TP7 or TP8)
-Ground on lower forehead / contralateral mastoid / clavicle
Target impedance for babies
< 10 kOhms and within 1 kOhm between electrodes, but higher absolute impedances may be warranted
For threshold ABR, why is the low-pass filter generally 1500 Hz?
Since we’re doing threshold ABR instead of neurodiagnostic ABR, the exact latencies aren’t important, so we care more about having a smoother result
For threshold ABR, why is high-pass generally 20-30 Hz?
To get a stronger wave V
Weighted Averaging
Weighted averaging ensures that very noisy sweeps are not included by setting a rejection level (this is preferred over non-weighted averaging which averages all responses)
What correlation value is strongly suggestive of a response?
.46 or greater
T/F: we can measure evoked potential (EPrms) on its own
FALSE: always measuring evoked potential + background noise
How do you estimate SNR?
-Estimate amplitude of background noise
-Divide response amplitude (which includes noise) by this
-Subtract 1
What are the 3 basic methods of estimating background noise?
-Signal absent recording (e.g., a pre-stimulus baseline)
-The plus/minus average
-Trial-to-trial variance
Estimating BN: Signal Absent Methods
-Compare amplitude of trial with stimulus present (EP + BN) with amplitude of trial with no stimulus present (BN)
-This requires two runs, one with stimulus and one without
-Problem: the brain changes after stimulation (even the noise isn’t the same) —> induced activity
Estimating BN: the plus/minus average
-Create an estimate of the noise, by inverting every other response before averaging
-Anything consistent from trial to trial should get subtracted away (e.g., you average a positive wave V and a negative wave V)
-Advantage to this approach is that the background noise is obtained from the same time period as the response, so the normal equation holds
Estimating BN: Trial-Trial Variance
-Basic approach:
·Pick a single time point (e.g., wave V peak)
·Record response amplitude at this time point over individual trials
·Find the difference between each trial amplitude and the across-trial amplitude
·Since these differences will be positive or negative, the average difference will be zero, so we square them to make them positive
·The sum of the squared values (SS) divided by the number of trials is the mean squared value (this is variance or power of the noise)
·The square root of the mean squared value is the standard deviation (this is the background noise amplitude)
What is residual noise also known as?
Background noise estimate
What is the best approach for estimating background noise?
The trial-trial variance because although it is similar to the plus/minus average, it uses actual trial-trial differences instead of just average differences (where some of the noise gets averaged away)
T/F: signal-to-noise ratio is a ratio of amplitudes and if you square it, you get a ratio of powers
TRUE: SNR^2 is a ratio of powers, and therefore is an F statistic
What is Fsp?
When the noise is estimated using trial-to-trial variance, measured at a single time point. Fsp = F single point
What Fsp value gives you a 99% confidence interval?
3.1
What is Fmp?
F multiple points (most clinical systems include this)
The RN should be ~ ___ nV before deciding that no response is present
20
Subjective vs. Objective methods to determine when you have enough averages
-Subjective: when response can be seen,
·Replicate to visualize variability (better: calculate cross-correlation)
·Run at multiple levels to visualize latency-intensity function
-Objective: when response is sufficiently large relative to background noise (e.g., SNR > 1.2, Fsp > 3.1): BN can be estimated by baseline, +/- averaging or trial-trial variance or amplitude (but baseline is a poor choice)
·OR when background noise is sufficiently low (20 nV), even if no response
__ dB nHL is the level most commonly used in ABR screening
35
Automated vs. Non-Automated ABR Screening
-Automated: must use an objective criterion
-Non-Automated: usually uses an objective criterion (SNR, Fsp, residual noise)
ABR is the most accurate test of hearing for infants < ____ months
6
What is the ABR stopping criteria?
-70 dB nHL stimulus —> stop when response observed
-30 & 50 dB nH stimuli —> stop if one of the following:
·Fsp criterion reached AND at least 768 sweeps AND observed
·Low residual noise (<15 nV) was reached without response
·Maximum number of sweeps (6144) without response
What peaks are visible in infant ABR and when do they mature to adult-like?
-Wave I: adult-like by 3 months
-Wave III: adult-like by 8-16 months
-Wave V: adult-like by 18-36 months
Latency change with age
-Wave I settles first while wave V settles last
Threshold is not affected by rates up to ___ Hz. Usually we use a rate of ___ /sec
50; 40
T/F: stimulus phase (condensation or rarefaction) affects thresholds
FALSE: stimulus phase does NOT affect thresholds. We use alternating polarity to eliminate cochlear microphonic OR combine rarefaction and condensation after recording
Procedural Considerations for Babies
-Newborn babies naps are short (roughly an hour at a time)
·Sedation with chloral hydrate when necessary (usually not required though with newborns)
-Prepare baby before feeding
·Make sure the baby is not fed the hour before the appointment (want a hungry and tired baby!)
-Begin testing at lower level (30 dB)
·Less chance of waking baby
·Less wasted time
Procedural Considerations: Frequency & dB nHL
-Start with 2000 Hz at 30 dB nHL, then 60 dB nHL
-If no response, try bone conduction at 30 dB nHL
-Then find threshold in 10 dB steps
-Get 2000 Hz in both ears first (and maybe 4000 Hz) before going on to 500 Hz, which is more difficult
T/F: masking is generally not required for ABR
TRUE: if the other ear is responding, the response will be larger contralaterally. True for bone conduction in infants (10-15 dB IA for bone)
What are the normal infant physiologic thresholds? (At 500 Hz, 1000 Hz, 2000 Hz, 4000 Hz)
-500 Hz: 42-53 dB
-1000 Hz: 37-38 dB
-2000 Hz: 31-38 dB
-4000 Hz: 31-43 dB
How is estimated behavioural threshold expressed?
eHL
How are ABR thresholds typically expressed?
dB nHL
ABR threshold estimates have standard deviations that range from about __ dB to __ dB across various studies
5, 15
