Introduction
Auditory-evoked potentials (AEPs) are neuroelectric changes that occur in the peripheral
and central nervous systems when they are exposed to certain sound stimulation. Such
potentials can be classified according to latency as short-, medium- and long-latency
AEPs.[1]
The brainstem auditory-evoked potential (BAEP) is a short-latency, objective and non-invasive
electrophysiological test that records the electrical activity of the auditory system
from the auditory nerve to the brainstem in a window of up to 10 milliseconds (ms).[2]
[3]
The BAEP responses are obtained through seven waves, and each wave comprises a structure
of the auditory system; therefore, the distal portion of the auditory nerve is represented
by wave I, the proximal portion is represented by wave II, wave III starts in the
cochlear nuclei, waves IV and V can be assigned to the ipsi and contralateral lemniscus,
and waves VI and VII can be characterized as the mesencephalic potential activities.[4]
To evoke BAEP responses, the most commonly-used stimuli are click and tone burst,
with click as the most frequent as it is a transient and short-lived stimulus (100
µsec), which produces synchronous firing of auditory neurons, whose primary frequency
is determined through a transducer-resonant frequency. Such a stimulus also has a
broad spectrum, with maximum peak energy in regions ranging from 1,000 Hz to 4,000 Hz.
The best relationship for pure-tone thresholds is between 2,000 Hz and 4,000 Hz.[5]
Some non-pathological variables should be considered in order to achieve a better
understanding of the possibilities of BAEP responses. These include variables related
to the individual, such as age, gender, drugs taken, body temperature, muscle artifacts,
among others.[6]
[7]
[8]
Several stimulus-related variables influence AEPs, including polarity, filter, stimulus
intensity and stimulation rate. The following is an example of a protocol used in
the clinical practice to obtain BAEP recordings: click stimulus, condensation polarity,
stimulus intensity of 80 dB HL, a rate of 27.7 stimuli/s and a bandpass filter of
100 Hz to 3,000 Hz.[9] The stimulation rates can be applied in different ways, and the rate of 27.7 stimuli/s
is one of the most used in audiology services.[1]
[10] The presentation of higher rates of stimuli per second enables the collection of
a greater number of responses in a given period of time, promoting a shorter testing
time; however, the collected recordings are subject to changes related to wave morphology.[11] These changes are important and must be considered to standardize data and define
protocols for the clinical practice.[12]
An important fact that must be considered when proposing studies on this topic is
the transmission frequency of the power grid in Brazil. It runs at a 60 Hz sine wave;
however, this value is not always found in the sockets. Thus, in an electrophysiological
assessment, the use of a presentation rate adjusted based on a measurement of the
local transmission frequency of the power grid can provide important information.
As modifications in this parameter may cause deformations in wave morphology, including
the possibility of changes in latency and amplitude, the present study aimed to compare
the BAEP responses with click stimulus at different rates in adults, including a presentation
rate adjusted based on the measurement of the local frequency transmission of the
electricity grid, and determine the best response possible with minimum wave deformation.
Methods
The present cross-sectional analytical study was performed at the Laboratory of Hearing
and Technology after obtaining approval from the institutional Ethics in Research
Committee Research under number 4.118.743. All participants signed the free and informed
consent form.
The study included 15 individuals (30 ears), according to the central limit theorem,
which states that the distribution of sample means approximates a normal distribution
as the sample size gets larger.[13] The sample was aged between 19 and 35 years, with normal hearing thresholds (≤ 25 dB
at frequencies from 250 Hz to 8000 Hz), presence of acoustic reflexes, and type-A
tympanograms. The adopted exclusion criteria were: individuals with hearing impairment;
any complaints relating to a central-auditory-processing disorder (CAPD); history
of ear surgery; having had more than three ear infections in the current year; use
of ototoxic medication; and presence of tinnitus, vertigo, dizziness or other cochleo-vestibular
changes. A detailed anamnesis was applied to verify all these preestablished criteria.
After selecting the participants, the following procedures were performed:
-
Inspection of the external auditory canal to check its integrity and that of the tympanic
membrane using a model mini-3000 (Heine Optotechnik GmbH & Co. KG, Gilching, Germany)
otoscope;
-
Immittance audiometry, performed using the Interacoustics (Middelfart, Denmark) AT235
clinical tympanometer, to select participants with type-A tympanograms and acoustic
reflexes;
-
Tonal and vocal audiometry, using the Interacoustics model Ad 629 audiometer and supra-aural
model DD45 headphones, in an acoustic booth by Vibrasom (São Bernardo do Campo, SP,
Brazil), according to S3.1 recommendations of the American National Standards Institute
(ANSI).[14] The objective was to identify normal auditory tonal and vocal thresholds. Tonal
audiometry is a psychoacoustic method to research hearing thresholds, and it was performed
using the descending technique, with 10-dB intervals, and the ascending technique
with 5-dB intervals, to confirm the responses. Frequencies with octave ratios between
250 Hz and 8,000 Hz were evaluated, including the interoctave frequencies of 3,000 Hz
and 6,000 Hz;
-
The BAEP was performed using the Bio-logic Navigator PRO AEP equipment(Natus Medical,
Inc., Pleasanton, CA, United States). The participant was instructed to lie down in
an armchair, with the body and muscles relaxed to reduce the artifacts from muscle
action. The testing site remained poorly lit, silent, and with few visual distractions.
The skin was cleaned with abrasive paste and then the electrodes were fixed with microporous
tape, using an electrolytic paste to improve electrical conductivity. Surface electrodes
were placed at predetermined positions: the inverter electrode at M1 and M2 (related
to the right ear and left ear respectively, fixed on the mastoids); the non-inverter
electrode at Fz; and the reference (ground) at the Fpz position, according to the
10–20 International Electrode System (IES). The impedance between the electrodes was
maintained at a level below 5 kΩ. The protocol used to capture the recordings was
the click stimulus, with a duration of 0.1 μV, rarefaction polarity, presented monaurally
at 80 dB HL, at different stimulation speeds per second, employing a total of 2,000
stimuli, with 150 Hz high-pass filters and 3,000 Hz low-pass filters, 10-ms window,
and artifact rejection with peaks over 23.73 μV or valleys below -23.73 μV.
The stimuli were randomly presented at rates of 21.1, 26.7, and 27.7 stimuli/s. In
addition, a fourth stimulation rate was presented, based on a mathematical calculation
performed on the Excel (Microsoft Corp., Redmond, WA, US) software, version 14.4.2
for Mac OS. The transmission frequency of the power grid at the time of the examination
was measured using an oscilloscope.
The mathematical calculation is as follows:
in which, % SRE is the percentage of stimulation rate efficacy in relation to the
transmission frequency of the power grid, TFPG is the transmission frequency of the
power grid, and ASR is the adjusted stimulation rate.
The TFPG was obtained with an oscilloscope. Thus, after recording the electrical voltage
(mV) in relation to time (ms), a Fast Fourier Transform was applied, by the oscilloscope
itself, to obtain the frequency spectrum. Finally, after obtaining the TFPG, the data
were inserted into the formula programmed in the spreadsheet, to obtain the ASR. The
spreadsheet automatically searched for the ASR close to 27.7 stimuli/s with the highest
%SRE. This was the fourth rate used in the investigation, and it varied according
to the TFPG at the time of the examination.
The power grid in Brazil runs at a 60 Hz sine wave; however, this value is not always
exactly the one that reaches the sockets. In addition, it is not recommended that
the AEP stimulation rate be applied at multiples or submultiples of the TFPG, for
a mechanism called “phase-locked” (phase synchronism) may occur. In this condition,
the acquisitions of the biopotentials would always occur in the same phase of the
power grid, which would cause a distortion in the wave's amplitude, especially if
this phase synchronism tends to the extremities of the wave's peak or valley.
Thus, the adjusted stimulation rate, in other words, is the frequency closest to 27.7
stimuli/s, that has the lowest resonance with the power grid, which, in theory, would
enable the obtainment of signals with less interference from that grid.
Two recordings were obtained for each stimulation rate and for each of the 30 ears,
thus verifying the tracing reproduction and confirming the existence of responses.
The BAEP was recorded by a trained examiner, and wave marking was performed by two
examiners experienced in electrophysiology. When the tracing was difficult to analyze
and there was no agreement regarding the marking, discussions were held among all
professionals involved in the study, and a consensus was reached. The marking and
identification of each peak was performed manually, thus evaluating its morphological
characteristics and relevant temporal aspects. The absolute latencies and amplitudes
of waves I, III, V were analyzed.
Statistical Analysis
The statistical analysis was performed using the Statistical Package for the Social
Sciences (IBM SPSS Statistics, IBM Corp., Armonk, NY, United States), version 23.0
for Macbook. For the description of the data, tables and graphics with means and standard
deviations (SDs) were used. Initially, an evaluation of the sample was performed to
observe its adherence to the normal distribution using the Shapiro-Wilk test.
The analysis of latencies and amplitudes according to the stimulation rates was performed
using the Friedman test, and the Wilcoxon signed-rank test was used to analyze the
differences between stimulation rates by pairs,. The evaluation of artifacts according
ot the stimulation rates was performed using the Friedman test. The differences were
considered statistically significant with values of p < 0.05.
Results
The sample consisted of 15 participants of both genders, 6 (40%) male and 9 (60%)
female subjects. The average age of the men was 20.5 years (SD: 1.3 years), and that
of the women was 23.4 years (SD: 4.6 years). The age of the total sample ranged from
19 to 35 years.
The ASRs from the mathematical calculation and those obtained with the aid of the
oscilloscope at the time of the examination varied between two values: 26.6 stimuli/s
for 3 participants, and 27.0 stimuli/s for 12 participants.
[Table 1] presents the data obtained from the descriptive analysis (mean, SD, and confidence
interval) of the latencies and amplitudes of waves I, III, and V in the studied stimulation
rates.
Table 1
Descriptive analysis of the latencies (ms) and amplitudes (μV) of waves I, III and
V in the right and left ears of the 15 participants (30 ears) according to the stimulation
rate
|
Mean ± standard deviation
|
95% confidence interval
|
|
Stimulation rate
(stimuli/s)
|
Wave I
|
Wave III
|
Wave V
|
(lower limit–upper limit)
|
|
Wave I
|
Wave III
|
Wave V
|
|
Latency (ms)
|
|
21.1
|
1.39 ± 0.25
|
3.51 ± 0.30
|
5.44 ± 0.24
|
1.29–1.48
|
3.40–3.62
|
5.35–5.53
|
|
26.7
|
1.42 ± 0.28
|
3.52 ± 0.28
|
5.43 ± 0.33
|
1.32–1.53
|
3.42–3.43
|
5.31–5.56
|
|
27.7
|
1.38 ± 0.29
|
3.47 ± 0.32
|
5.50 ± 0.26
|
1.27–1.50
|
3.35–3.59
|
5.41–5.60
|
|
ASR
|
1.46 ± 0.24
|
3.50 ± 0.30
|
5.47 ± 0.25
|
1.37–1.56
|
3.39–3.61
|
5.38–5.57
|
|
Amplitude (μV)
|
|
21.1
|
0.36 ± 0.18
|
0.41 ± 0.19
|
0.56 ± 0.16
|
0.29–0.43
|
0.34–0.48
|
0.50–0.62
|
|
26.7
|
-0.02 ± 0.21
|
0.13 ± 0.26
|
0.46 ± 0.22
|
-0.10–0.05
|
0.03–0.22
|
0.38–0.55
|
|
27.7
|
-0.08 ± 0.24
|
0.09 ± 0.21
|
0.36 ± 0.18
|
-0.17–0.01
|
0.01–0.17
|
0.29–0.43
|
|
Adjusted stimulation rate
|
-0.91 ± 4.93
|
0.21 ± 0.26
|
0.44 ± 0.17
|
-2.75–0.93
|
0.10–0.31
|
0.37–0.50
|
The general analysis of latencies and amplitudes according to the stimulation rates,
performed using the Friedman test, showed a statistically significant difference (p < 0.01).
For the analysis of the latencies by pairs of used stimulation rates, with the Wilcoxon
test, there was a statistically significant difference only in the latency of wave
V between the rates of 21.1 and 27.7 stimuli/s (mean: 5.38 ms and 5.45 ms respectively;
p = 0.048) and between the rates of 21.1 stimuli/s and the ASR (mean: 5.38 ms and 5.46 ms
respectively; p = 0.049).
[Table 2] shows the amplitude results with a statistically significant difference by pairs
of used stimulation rates. We observed that the rate of 21.1 stimuli/s had the highest
amplitudes whenever compared with other rates.
Table 2
Results with statistically significant differences in amplitude according to the pairs
of stimulation rates evaluated
|
Pairs of stimulation rates (stimuli/s)
|
Wave
|
Mean rate 1
(μV)
|
Mean rate 2
(μV)
|
p-value
|
|
21.1–26.7
|
I
|
0.40
|
0.09
|
0.000
|
|
III
|
0.48
|
0.25
|
0.000
|
|
V
|
0.58
|
0.49
|
0.009
|
|
21.1–27.7
|
I
|
0.40
|
0.05
|
0.000
|
|
III
|
0.48
|
0.22
|
0.000
|
|
V
|
0.55
|
0.39
|
0.000
|
|
21.1–adjusted stimulation rate
|
|
|
|
|
|
I
|
0.40
|
0.09
|
0.000
|
|
III
|
0.48
|
0.32
|
0.004
|
|
V
|
0.55
|
0.46
|
0.001
|
|
26.7–adjusted stimulation rate
|
III
|
0.25
|
0.32
|
0.043
|
|
27.7– adjusted stimulation rate
|
III
|
0.22
|
0.32
|
0.013
|
|
V
|
0.39
|
0.46
|
0.022
|
|
26.7–27.7
|
V
|
0.49
|
0.39
|
0.013
|
Note: The Wilcoxon signed-rank test was used to analyze the differences between stimulation
rates by pairs.
[Fig. 1] shows the grand average of the BAEP wave recordings obtained according to the stimulation
rate used.
Fig. 1 Grand average of the BAEP waves according to the stimulation rate used in the study
(21.1, 26.7, 27.7 stimuli/s, and the ASR).
[Table 3] presents the data obtained from the descriptive analysis (mean, SD, and confidence
interval) of the artifacts according to the stimulation rate used.
Table 3
Descriptive analysis of the artifacts according to the stimulation rate
|
Stimulation rate (stimuli/s)
|
Mean
|
Standard deviation
|
95% confidence interval
|
|
Lower
|
Upper
|
|
21.1
|
150.13
|
129.15
|
116.77
|
183.50
|
|
26.7
|
170.77
|
139.33
|
134.77
|
206.76
|
|
27.7
|
161.48
|
177.29
|
115.68
|
207.28
|
|
Adjusted stimulation rate
|
172.47
|
174.33
|
127.43
|
217.50
|
The general evaluation of the artifacts, using the Friedman test, did not result in
a statistically significant difference (p > 0.05).
Discussion
The rate of presentation of stimuli is a parameter related to the speed the acoustic
stimulus travels through the auditory pathways to the brainstem, as it corresponds
to the refractory period of the neural structure. Higher rates can be used to sensitize
the BAEP recording, to assess neural synchronism, and to identify neurological disorders.[11]
Due to the possibility of using different rates of presentation of stimuli, it is
important that all possible findings and their implications for the clinical practice
are clarified and discussed, as changes in the BAEP wave morphology can lead to mistakes
in markings and, consequently, mistakes in assessments and incorrect diagnoses.
The present study evaluated four different click stimulus presentation rates in the
BAEP, which were determined based on the following criteria: 1) frequent use in the
clinical practice (27.7 stimuli/s);[11] 2) proximity to the rate used in the clinical practice, but with less resonance
with the power grid, corresponding to 60 Hz (26.7 stimuli/s);[15] 3) mentions in the literature as usual (21.1 stimuli/s);[16] amd 4) ASR according to the power grid at the time of the examination. This adjustment
would provide better knowledge about latency and amplitude recordings, because, theoretically,
it would control the artifact inputs from the power grid, which could contribute to
better wave morphology.
Among the used rates, there was a better morphology of waves I, III, and V with the
rate of 21.1 stimuli/s ([Fig. 1]), which comprises the slowest investigated stimulation rate. The use of slower rates
is indicated when there is a need to identify the components and amplitude of the
first BAEP waves.[17]
In 1991, the Committee on Infant Hearing American Speech-Language-Hearing Association
suggested the use of click stimulus with a 21.1 stimuli/s repetition rate in the rarefaction
and condensed polarities, with the justification that, at this rate, wave morphology
showed better visualizations, which would assist with the verification of the presence
of cochlear microphonism, to identify auditory neuropathy spectrum disorders.[18]
In a study[19] with 59 patients diagnosed with severe or profound hearing loss, the authors modified
the characteristics of the click stimulus to sensitize the BAEP performance. According
to the authors, changes in the characteristics of the acoustic stimulus in the BAEP
with the use of faster stimulation rates may cause less reproduction and clarity of
the waves, in contrast to slower stimulation rates, which can increase amplitude and
significantly improve the wave pattern. Thus, the study[19] suggested the use of the slowest recording technique in all patients with and without
hearing loss. Likewise, the present study showed, when comparing the rates of 27.7
and 21.1 stimuli/s, statistically significant differences in the values referring
to the amplitudes of waves I, III, and V, with better recordings for the rate of 21.1
stimuli/s.
Another important issue refers to the differences that can be found in the BAEP recording
depending on the assessed population. The absolute latency time and wave amplitude
can be influenced, for instance, by factors related to age, with increased latency
and decreased amplitude with increasing age.[6] For young children, due to the incomplete myelinization of neuron axons, prolonged
neural transmission and increased latency of responses are found.[20]
A study[21] evaluated the BAEP of children with suspected CAPD and showed a significantly-prolonged
latency of wave V at faster stimulation rates when compared with age-matched typically
developing children. The BAEPs were recorded at slow (13.3 stimuli/s) and faster (57.7
stimuli/s) rates.
The choice of rate of presentation of stimuli can be determined by age.[11] In infants, for example, the rate of 27.7 stimuli/s can result in a quicker test,
as the behavioral pattern of infants or younger children does not enable the use of
slower rates, and this factor increases the examination time, making the procedure
more difficult. However, in older children (or younger ones, but using sedation) and
adults who respond to commands and, in turn, can contribute by remaining at rest during
the test, the stimulation rate could be slower.
In addition, another point of discussion is quite relevant when evaluating the clinical
applicability of the technique of presenting stimuli based on the concept of refractory
period. It is known that a nerve fiber needs a critical amount of time to recover
from a previous stimulus and, during this period, it cannot trigger another recording
response. Therefore, when nerve fibers are changed, they require a longer refractory
period. Therefore, the stimulus presented at a high repetition rate causes a shorter
recovery period in the nerve fibers, which can trigger poor synchrony, with changed
waves in the BAEP.[22]
All of these aspects are important and should be considered to standardize data and
define protocols in the clinical practice. The increase in this specific parameter
can cause changes in wave latency and amplitude, with a reduction in amplitude and
a wave prolongation, which alter the BAEP wave latencies. For individuals with normal
hearing, when the stimulus presentation rate is increased from 10 to 100 stimuli/s,
the lagtency of wave V increases ∼ 0.5 ms, and the amplitude decreases.[12] Similar results were found in the present study, in which an increase in the latencies
of wave V at higher rates of stimulation (21.1 versus 27.7 stimuli/s and 21.1 versus
the ASR) was observed, and higher amplitudes were always observed at lower rates (21.1
stimuli/s).
A study[23] on the effect of the click rate on the latency of BAEPs was performed with six individuals
with normal hearing at rates of 10, 30, 50 and 100 stimuli/s, and at intensities of
30, 40, 50 and 60 dBSPL. The authors[23] found differences in the absolute latency of wave five of around 0.5 ms, when comparing
the repetition rates of 10 and 100 stimuli/s. In addition, the four tested intensities
triggered the same change in the latency of wave V, even with the variation in stimulation
rates. The authors also observed a reduction in the amplitude of wave V with an increase
in the stimulation rate.
In a study with adults with normal hearing, the author[24] observed that changes in stimulus speed from 10 to 80 stimuli/s resulted in an increase
of 0.14, 0.23 and 0.39 ms for waves I, III, and V respectively. In neonates, the same
increase in rate triggers a latency increase of 0.8 ms for wave V. In relation to
amplitude, this increase promotes a 50% reduction in waves I and III, and only 10%
in wave V. The amplitude of wave I decreases only 10% with stimulation rates of ∼
20 stimuli/s. In addition, the author[24] noted that no significant changes are observed in amplitude with an increase in
the rate from 11 to 61 stimuli/s in infants. Finally, repetition rates of around 50
stimuli/s are more efficient to visualize wave V; however, to observe wave I, rates
between 20 and 40 stimuli/s are more appropriate.
In the present study, when comparing the rate of 27.7 stimuli/s with the ASRs, significant
differences were found for the amplitudes of waves III and V, attesting that the ASRs
are responsible for the amplitudes with better visualization. It should be noted that
the ASRs were of 26.6 stimuli/s for 3 participants and 27.0 stimuli/s for 12 participants.
The comparison of the rate of 26.7 stimuli/s with the ASRs, in turn, only showed a
difference for the amplitude of wave III, with higher amplitudes for the ASRs. This
important finding strengthens the thesis that automated adjustments in stimulation
rates that take into account the local TFPG can improve the acquisition of the tested
signals.
It is a consensus in the literature that the increase in the speed of acoustic stimuli
affects the BAEP wave behaviors differently. Waves I and III, for example, are more
influenced than wave V, which remains more robust at high stimulation rates.[11]
To delve deeper into the morphological differences presented in [Fig. 1], explanations in the field of Biophysics should be made. It is impossible to have
negative neuron responses as the neuron potential is linked to positive ions. Therefore,
the positive/negative nomenclature is based only on the habit of using this term,
as the electrodes can only generate positive responses.[25]
According to the findings of the present study, the responses generated with the rate
of 21.1 stimuli/s did not show negative amplitudes; however, the faster rates, such
as 27.7 stimuli/s, showed negative amplitude peaks ([Table 1]). Apparently, according to the presented results, the increase in speed decreases
the amplitude, and lower amplitudes are more susceptible to signal changes due to
fluctuations in the power grid.
The amplitude analysis, following the adjustment criteria according to the power grid,
showed only positive peaks; however, all of them were below the rate of 21.1 stimuli/s.
This data indicates that the rate of 21.1stimuli/s is less influenced by the power
grid, a fact evidenced in [Fig. 1] of the present study, which presents the rate of 21.1 stimuli/s with a prominent
morphology and impeccable visualization when compared with the other evaluated rates.
Another analysis is necessary when considering the use of different rates of presentation
of stimuli rates due to the possibility of reducing the test time. Currently, given
the short time available to carry out all activities, adjusting a certain protocol
parameter to reduce testing time is often more valued than the implications of this
choice to the evaluation findings.
In addition, the time to perform 2,000 recordings is of 1.60 minute, 1.24 minute and
1.20 minute respectively for the rates of 21.1, 26.7, and 27.7 stimuli/s. A gain of
∼ 40 seconds per test alone would not justify the use of a higher rate. It would also
be necessary to maintain the wave's morphology.
So, this increase in the stimulation rate can cause changes in wave morphology, especially
regarding amplitude, which may result in difficulties in marking the peaks and interfering
with the reliability of the diagnosis. The decrease in amplitude could make it difficult
to visualize wave V in the threshold search, for example.
Regarding the artifacts presented according to the stimulation rate, we found no statistically
significant difference, a fact possibly justified by two factors: 1) the high variability,
with SDs that approximate or exceed the mean values. Such variability does not enable
one to safely define of differences between groups; 2) the low sensitivity of artifact
rejection, which is configured to eliminate records with peaks above 23.73 μV or valleys
below -23.73 μV. Thus, small improvements or worsening in noise levels may have gone
unnoticed by this measure. It is likely that a detailed noise assessment, with an
analysis of the frequency domain, and also with the possibility of individual analysis
of the records, will reveal important findings to be considered in the clinical practice.
Despite the limited number of participants, the present study yielded important and
reliable findings regarding the discussion on the use of different rates of presentation
of click stimuli in the BAEP. It is noteworthy that the BAEP was performed at 80 dB
HL, focusing on the peaks of waves I, III, and V, analysis condition restricted to
the assessment of the auditory pathway integrity. However, further studies to delve
deeper into the topic are essential, with the possibility of assessments with different
age groups, populations with hearing loss, as well as a study aimed at researching
the electrophysiological threshold from the change in this parameter.
Thus, the importance of the examiner's knowledge about the choice of parameters is
reinforced, in particular, the rate of presentation of stimuli rate for each protocol
defined during the performance of electrophysiological assessments.
