Content uploaded by Pedro De Lemos Menezes
Author content
All content in this area was uploaded by Pedro De Lemos Menezes on Nov 19, 2017
Content may be subject to copyright.
Acta Oto-Laryngologica. 2014; Early Online, 1–6
ORIGINAL ARTICLE
Amplitude modulated vestibular evoked myogenic responses: a study
of carrier and modulating frequencies
ALINE CABRAL DE OLIVEIRA
1
, LILIANE DESGUALDO PEREIRA
2
,
JOSÉ FERNANDO COLAFÊMINA
2
& PEDRO DE LEMOS MENEZES
3
1
Federal University of Sergipe, Brazil,
2
Federal University of São Paulo, Brazil and
3
Estadual university of health sciences
of Alagoas, Brazil
Abstract
Conclusion: Responses with greater amplitude were recorded when carrier frequencies were modulated at 37, 40, and 43 Hz.
These responses can be recorded even in patients with significant sensorineural hearing loss, from the sternocleidomastoid
(SCM) muscle for a 500 Hz tone, 100% modulated at 40 Hz. Objective: To determine the best carrier and modulating
frequencies to evoke steady-state myogenic responses. Methods: The present study investigated 156 ears of 78 normal-hearing
young adults, with carrier frequencies of 250, 500, and 1000 Hz, modulated at 20, 37, 40, 43, 70, 77, and 80 Hz, with an
intensity of 95 dBA. Furthermore, we observed responses evoked by stimulus carrier frequency of 500 Hz, modulated at 40 Hz,
with an intensity of 95 dBA in a group of five subjects with severe sensorineural loss. Results: Responses were found for all
stimuli studied (p<0.01). Modulated stimuli at frequencies of 37, 40, and 43 Hz evoked better steady-state vestibular evoked
myogenic potential (S-VEMP) (p<0.05). No statistically significant differences were found between the group of normal
hearers and the group of subjects with hearing loss (p=0.431), for the stimulus used.
Keywords: Evoked potentials, vestibule, labyrinth, vestibular nerve, amplitude modulation, evoked responses, electromyography
Introduction
The vestibular evoked myogenic potential (VEMP) is
used in a number of specialized centers for vestibular
evaluation, specifically analyzing saccule and inferior
vestibular nerve function, regions not detected by
traditional vestibular examinations [1–6].
Low-frequency tone-burst stimuli are the best
option for obtaining this potential, given that they
are frequency-specific brief envelopes of pure tones
[5–7]. However, VEMPs are present over a wide
frequency range (between 100 and 3200 Hz), with
greater response amplitude for the low frequencies
(between 250 and 500 Hz) [5,6].
A study by Bell et al. [8] demonstrated that VEMP
can be obtained at a frequency of 500 Hz amplitude,
modulated at frequencies of 5, 39, 59, 78, 98, and
122 Hz and that the properties of these responses are
consistent with a saccular origin, although more
clinical testing is required to confirm this finding.
The author denominated this potential ‘steady-state’
VEMP (S-VEMP). In this study better responses were
found at 500 Hz, modulated at 39 Hz, but the
next lowest modulating frequency was 5 Hz and
the next highest was 59 Hz. Thus, the best modulating
frequencies, with less variation, and at frequencies
between 20 and 80 Hz, remain unknown. Moreover,
it is important to observe the response behavior for
different carrier frequencies. Accordingly, the present
study investigated which frequencies close to 39 Hz
would provide the best responses for carrier frequen-
cies of 250, 500, and 1000 Hz.
Sedimentation of the S-VEMP and its application
in clinical practice is important, since it would
decrease exam time, when compared with VEMP
in the time domain, and would allow simultaneous
observation of the two labyrinths, given that sounds
can be presented to the two ears at the same time.
Correspondence: Aline Cabral de Oliveira, PhD, Street G, 38 –Porto Sul, Aruana, Aracaju –SE, CEP 49039-282, Brazil. Tel: +55 79 99211700.
Fax: +55 82 33152200. E-mail: alinecabralbarreto@gmail.com
(Received 4 February 2014; accepted 18 March 2014)
ISSN 0001-6489 print/ISSN 1651-2251 online 2014 Informa Healthcare
DOI: 10.3109/00016489.2014.909605
Acta Otolaryngol Downloaded from informahealthcare.com by University of Utah on 06/25/14
For personal use only.
This simultaneous ear stimulation, associated with
contraction of the two sternocleidomastoid (SCM)
muscles (frontal protrusion of the neck) ensures the
contraction of these muscles with the same tension
and allows more adequate comparison of response
amplitudes between the two sides, without the use of
electromyography.
Thus, preforming simultaneous S-VEMP will allow
more precise direct observation of asymmetries in
response amplitudes between the two sides, without
the need for formulas, as is currently the case.
Material and methods
Three experiments were carried out in the present
study. The first investigated S-VEMP responses
evoked by different modulating and carrier frequen-
cies in normal hearers. The second investigated
S-VEMP responses evoked by a frequency of 500
Hz, and amplitude modulated at 40 Hz, in a small
group of subjects with hearing loss (sensorineural
deafness). The third experiment was conducted to
observe the existence of electromyographic interfer-
ences in the capture of electrophysiological responses.
Subjects
All subjects were recruited from a student population
and screened for normal hearing using pure-tone
audiometry, immittance testing, and determination
of acoustic reflexes. Subjects with hearing loss (sen-
sorineural deafness) were recruited among patients
undergoing audiologic rehabilitation therapy. All
measurements were approved by the Research Ethics
Committee of the Alagoas State University of Health
Sciences (UNCISAL), process no. 1010/2009.
For the first experiment, the sample was composed
of 78 individuals (156 ears), 40 females (80 ears) and
38 males (76 ears), aged between 18 and 35 years,
with no hearing alterations. The second experiment
was conducted with 5 deaf subjects (10 ears), with
pure-tone threshold greater than or equal to
90 dBHL, in addition to a normal immittance test,
carried out to ensure that the responses obtained did
not originate in the auditory system. The third group
was composed of five subjects randomly drawn from
the first experiment.
Apparatus
VEMP examinations were conducted using a specific
recording device. The apparatus was conceived at
the Instrumentation and Acoustics Laboratory
(LIA) of the Health Sciences University of Alagoas
State (UNCISAL, Brazil) and at the Dosimetry
Instrumentation and Radioprotection Center of the
Faculty of Philosophy, Sciences and Arts of Ribeirão
Preto (University of São Paulo-USP, Brazil). It
consists of biological amplifiers, filters, electrical pro-
tection system, and a logic system that enables a
thorough investigation of VEMP [9].
The equipment used in S-VEMP exhibited the
following basic traits for processing this type of signal:
digital analog signal conversion performed at a sam-
pling rate of 44.1 kHz, with resolution of 16 bits;
analog digital conversion conducted at an acquisition
frequency of 7.35 kHz, exactly 1/6 of the signal
generation rate, to ensure a fast fourier transform
(FFT) of up to 3.6 kHz. After rejection of electrical
artifacts, the FFT computational algorithm [10] was
applied to each window and the resulting strengths of
the frequency spectrum were promediated.
Stimuli were delivered to the participant via ER-3A
insert headphones, initiated in the right ear and later
repeated in the left ear. The responses were recorded
twice on both sides. Responses were recorded with the
active electrode on the upper half of the SCM muscle
of the ipsilateral SCM, the reference electrode on the
upper edge of the ipsilateral sternum, and the ground
electrode on the mid-frontal line. For each recording,
subjects were instructed to turn their heads as far as
comfortably possible and look over the shoulder of the
non-test ear, maintaining tonic muscle contraction
between 60 and 80 mV, monitored by electromyog-
raphy. Impedances and inter-electrode impedances
were kept below 5 kW.
Stimuli
In experiment 1, responses from the SCM were
recorded for 250, 500, and 1000 Hz carrier frequen-
cies, 100% modulated at 20, 37, 40, 43, 70, 77, and
80 Hz for a fixed stimulation level of 95 dBA,
presented in ER-3A insert earphones. The order of
presentation was randomized between subjects. Thus,
21 different stimuli were applied, 7 for each carrier
frequency. In other words, each carrier frequency was
modulated by each of the modulating frequencies,
configuring a specific stimulus; for example, a
frequency of 250 Hz modulated at 20 Hz, 250 Hz
modulated at 37 Hz, etc.
To avoid fatigue of the SCM muscle, the S-VEMP
was only applied at a carrier frequency (250, 500 or
1000 Hz) for all the modulators. Thus, each individ-
ual was submitted to seven examinations in the
frequency domain. These were conducted in random
order in terms of modulating frequency. The duration
of the examination (for each ear) at each carrier
frequency was approximately 2 min. It should be
pointed out that the carrier frequencies were analyzed
2A. C. de Oliveira et al.
Acta Otolaryngol Downloaded from informahealthcare.com by University of Utah on 06/25/14
For personal use only.
separately only to determine test methodology, as was
done in the study by Bell et al. (2010) [8]. However,
in clinical practice, these components should be
mixed in only one stimulus, maintaining an interval
between the modulators that is much greater than
3 Hz, to be determined in a future study.
The maximum sound exposure used for any of the
measurements was 95 dBA. The total sound dose for
any subject was kept within the limit for ‘usual’
exposures, defined by the Institute of Sound and
Vibration Research Ethics Committee as an 8 h
equivalent exposure of 76 dBA.
For S-VEMP analysis, the strengths of the frequen-
cies calculated by FFT were promediated 200 times
and analyzed using the magnitude-squared coherence
(MSC) method, calculated by equation 1 below to
confirm the responses. However, all the peaks could
be discerned with the naked eye.
MSC f
Xf
MXf
i
i
M
i
i
M
()
()
()
()==
=
∑
∑
1
2
2
1
1
The MSC method is based on coherence between the
electroencephalographic (EEG) signal and the stim-
ulus. When there is no synchronism, the value of
MSC is close to zero, and when it exists, it is close
to 1. Equation 2 demonstrates that M identical parts
of a signal are related to statistical distribution F and a
critical value of MSC is obtained for significance level
aat 2 and 2M-2 degrees of freedom.
MSC critical FMz
MFMz
zz
zz
() ,
,()
,
,
−
−+ −
a
a
12
Since the value of MSC(f) is higher than that of
MSC
critical
, it can be concluded that positive
identification of the response occurred for frequency
component f[11]. Values were considered significant
for p<0.05.
Experiment 2 used the same parameters as exper-
iment 1; however, a 500 Hz amplitude modulated at
40 Hz was the only stimulus used.
Finally, the third experiment was performed to
confirm that the responses were not a result of
electromagnetic interference in the earphone used.
Thus, five examinations were carried out with normal
hearers using the same parameters as experiment 1;
however, a 500 Hz amplitude modulated at 40 Hz was
the only stimulus used. Furthermore, the duct of the
insert earphone was clipped to prevent the passage of
sound.
Data were tabulated and processed using PASW
Statistics data editor 20.0. The means and standard
errors were presented in graph form.
Results
Pilot study
Initially, piloting was conducted to demonstrate the
existence of the VEMP response to amplitude mod-
ulated sounds, and to demonstrate that it exhibited
key characteristics of the VEMP response [12]. A 250,
500, and 1000 Hz tone modulated at 20, 37, 40, 43,
70, 77, and 80 Hz was presented at levels up to
95 dBA. In eight ears, responses to ipsilateral and
contralateral stimulation were compared at various
test levels. This demonstrated ipsilateral dominance
(no contralateral responses were observed) and an
increase in response amplitude with stimulation level
consistent with the VEMP response. An increase in
response amplitude with SCM tonicity was observed,
again consistent with the VEMP response.
Experiment 1: tuning of steady-state vestibular responses
to modulation frequency
The subjects were aged between 18 and 31 years,
mean age of 21.28 years, and standard deviation (SD)
of 2.90 years.
Figure 1 shows the amplitude of responses the
SCM recorded for a 250 Hz carrier frequency,
0
5
10
15
20
20 37 40 43 70 77 80
Amplitude (mV)
Modulation frequencies (Hz)
250 Hz carrier
Figure 1. Amplitude of responses of the sternocleidomastoid (SCM) muscle recorded to a 250 Hz carrier frequency 100% modulated at
different frequencies. Error bars represent standard error of the mean.
Steady-state vestibular evoked myogenic responses 3
Acta Otolaryngol Downloaded from informahealthcare.com by University of Utah on 06/25/14
For personal use only.
100% modulated at different frequencies. The
ANOVA test revealed statistically significant differ-
ences between the modulating frequencies used
(p<0.01). Tukey’spost hoc test estimated that mod-
ulating frequencies between 20 and 43 Hz exhibit
statistically higher responses than those between
70 and 80 Hz (p<0.01).
Figure 2 shows the amplitude of responses the
SCM recorded for a 500 Hz carrier frequency,
100% modulated at different frequencies. The
ANOVA test showed statistically significant differ-
ences between the modulating frequencies used
(p<0.01). Tukey’spost hoc tests estimated that
modulating frequencies between 37 and 43 Hz
exhibited statistically higher responses than those of
20, 70, 77, and 80 Hz (p<0.01).
Figure 3 shows the amplitude of responses the
SCM recorded for a 1000 Hz carrier frequency,
100% modulated at different frequencies. Analyses
of this carrier frequency produced similar results to
those obtained at a frequency of 500 Hz, when the
same tests were applied. Thus, statistically significant
differences were found between the modulation
frequencies used (p<0.01). Modulating frequencies
between 37 and 43 Hz showed statistically higher
responses than those of 20, 70, 77, and 80 Hz
(p<0.05).
Finally, the paired ttest showed that a carrier
frequency of 500 Hz exhibited statistically higher
responses than the others investigated, for modulating
frequencies between 37 and 43 Hz (p<0.05).
Experiment 2: responses of a small group of subjects with
hearing loss recorded in SCM, evoked by a 500 Hz
carrier frequency stimulus, 100% modulated at 40 Hz
The ages of the group of five subjects with severe
sensorineural hearing loss varied between 19 and
23 years, with a mean of 19.4 years and SD of
2.2 years. The mean amplitude of responses recorded
by the SCM recorded for a 500 Hz carrier frequency,
100% modulated at 40 Hz, was 17.4 mV, with a
standard deviation of 4.5 mV. The Mann–Whitney
U test did not reveal statistically significant differences
between these results and those obtained with the
same stimuli in normal hearers (p=0.431).
Experiment 3
The five subjects randomly selected from group one,
submitted to the experiment with blocked earphones
to prevent the passage of sound, did not exhibit any
electrophysiological response whatsoever.
Discussion
We chose tone-burst stimuli at frequencies of 250,
500, and 1000 Hz given that earlier studies [5]
showed that the VEMP may be present between
100 and 3200 Hz. We presented the stimuli, consist-
ing of a single modulated carrier frequency, and not
the three frequencies simultaneously, given that we
wanted to analyze how the frequencies behaved
0
5
10
15
20
25
20 37 40 43 70 77 80
Amplitude (mV)
Modulation frequencies (Hz)
500 Hz carrier
Figure 2. Amplitude of responses of the sternocleidomastoid (SCM) muscle recorded to a 500 Hz carrier frequency 100% modulated at
different frequencies. Error bars represent standard error of the mean.
0
5
10
15
20
20 37 40 43 70 77 80
Amplitude (mV)
Modulation frequencies (Hz)
1000 Hz carrier
Figure 3. Amplitude of responses of the sternocleidomastoid (SCM) muscle recorded to a 1000 Hz carrier frequency 100% modulated at
different frequencies. Error bars represent standard error of the mean.
4A. C. de Oliveira et al.
Acta Otolaryngol Downloaded from informahealthcare.com by University of Utah on 06/25/14
For personal use only.
individually so as to be able to determine the best
modulators for each carrier frequency. Thus, analyses
of the recordings demonstrated responses for all the
carrier frequencies tested, with a modulating
frequency between 20 and 80 Hz. This result was
expected, since earlier studies obtained more effective
responses for tone-burst stimuli at low frequencies
(£1000 Hz) [1,7].
The choice of modulating frequencies for the
examination in the frequency domain should also
be discussed. Following the auditory steady-state
evoked potential model, we opted initially for stimuli
with modulation rates of around 40 Hz [8], since
these are used to assess mid-latency auditory poten-
tials. Moreover, stimuli in the frequency range
between 70 and 110 Hz (used to record short-latency
potential, such as auditory steady-state responses –
ASSRs) did not allow visualization of the initial
components of VEMP, since this mid-latency exam-
ination requires greater temporal observation [13].
Response assessment for carrier frequencies up to
1000 Hz followed the same logic and was most
efficient at 37, 40, and 43 Hz, similar to that observed
in an earlier study and another conducted with a
500 Hz pure tone, which showed higher responses
to amplitudes modulated at 39 Hz [8]. Furthermore,
this modulating range shows better response for
awake individuals, as well as in auditory brainstem
response (ABR) [14].
The present study showed that a 500 Hz carrier
frequency evoked better results for frequencies
between 37 and 43 Hz than the other frequencies
tested. We found no literature studies, other than that
of Bell et al. (2010) [8], that tested only the 500 Hz
carrier frequency.
These results can be used to select carrier and
modulating frequencies for the S-VEMP exam, which
may further decrease execution time, when compared
with the VEMP of the time domain, and allow obser-
vation of more than one frequency simultaneously,
thereby avoiding possible false negatives, resulting in
stimulations with slightly weaker intensities.
Finally, S-VEMP responses were observed in deaf
subjects. These findings corroborate earlier studies
that also found VEMP responses in individuals with
hearing loss [4,15]. These results indicate that the
responses found in S-VEMP are not generated in the
auditory system, since they are present when subjects
exhibit severe hearing impairment.
Conclusions
This study demonstrated that S-VEMP can be
recorded from sound stimuli with carrier frequencies
of 250, 500, and 1000 Hz. Moreover, responses
exhibited higher amplitude when carrier frequencies
were modulated at 37, 40, and 43 Hz. These
responses can be recorded even in patients with severe
sensorineural hearing loss, from the SCM for a
500 Hz tone, 100% modulated at 40 Hz. In addition,
the examinations performed with blocked earphones
to prevent the passage of sound exhibited no electro-
physiological response.
Acknowledgment
This work received financial support from CNPq
(National Council for Scientific and Technological
Development). ClinicalTrials.gov Identifier:
NCT01260077.
Declaration of interest: The authors report no
conflicts of interest. The authors alone are responsible
for the content and writing of the paper.
References
[1] Akin FW, Murnane OD, Panus PC, Caruthers SK,
Wilkinson AE, Proffitt TM. The influence of voluntary tonic
EMG level on the vestibular-evoked myogenic potential.
J Rehabil Res Dev 2004;41:473–80.
[2] Basta D, Todt I, Ernest A. Normative data for P1/N1 –
latencies of vestibular evoked myogenic potentials induced by
air- or bone-conducted tone bursts. Clin Neurophysiol 2005;
116:2216–19.
[3] Cheng PW, Huang TW, Young YH. The influence of clicks
versus short tone bursts on the vestibular evoked myogenic
potentials. Ear Hear 2003;24:195–7.
[4] Sazgar A, Dortjai V, Akrami K, Akrami S, Yazdi AK.
Saccular damage in patients with high-frequency sensori-
neural hearing loss. Eur Arch Otorhinolaryngol 2006;263:
608–13.
[5] Sheykholeslami K, Kermany MH, Kaga K. Frequency sen-
sitivity range of the saccule to bone-conducted stimuli mea-
sured by vestibular evoked myogenic potentials. Hear Res
2001;160:58–62.
[6] Todd NPM, Rosengren SM, Colebatch JG. Tuning and
sensitivity of the human vestibular system to low-frequency
vibration. Neurosci Lett 2008;444:36–41.
[7] de Oliveira Barreto AC, Colafêmina JF, Menezes PL.
Saccular sensitivity function measured by vestibular
evoked myogenic potential. Acta Otolaryngol 2011;131:
618–23.
[8] Bell SL, Fox L, Id Bihi R. Vestibular evoked myogenic
responses to amplitude modulated sounds. J Acoust Soc
Am 2010;128:559–62.
[9] Oliveira AC, Colafêmina JF, Menezes PL. Vestibular evoked
myogenic potential: we propose a new instrument. Int Arch
Otorhinolaryngol 2010;14:410–16.
[10] Press WH, Teukolsky SA, Vetterling WT, Flannery BP.
Numerical recipes in C: art of scientific computing. New
York: Cambridge University Press. 1997. p 994.
[11] Infantose AFC, Melges DB, Tierra-Criollo CJ. Use of
magnitude-squared the maximum driving response band of
Steady-state vestibular evoked myogenic responses 5
Acta Otolaryngol Downloaded from informahealthcare.com by University of Utah on 06/25/14
For personal use only.
the somatosensory evoked potential. Braz J Med Bol Res
2006;39:1593–603.
[12] Colebatch JG, Halmagyi GM, Skuse NF. Myogenic poten-
tials generated by a click-evoked vestibulocollic reflex.
J Neurol Neurosurg Psychiatry 1994;57:190–7.
[13] Hall JW. New handbook of auditory evoked responses.
Boston: Pearson Education Inc. 2007. p 736.
[14] Joseph B, Roberson M, Carolann O, Stidham K. Auditory
steady-state response testing in children –evaluation of a new
technology. Arch Otolaryngol Head Neck Surg 2003;129:
107–13.
[15] Rosengren SM, Colebatch JG. Vestibular evoked potentials
(VsEPs) in patients with severe to profound bilateral hearing
loss. Clin Neurophysiol 2006;117:1145–53.
6A. C. de Oliveira et al.
Acta Otolaryngol Downloaded from informahealthcare.com by University of Utah on 06/25/14
For personal use only.
