Вы находитесь на странице: 1из 9

Electrical Activity of Extrinsic Laryngeal Muscles

in Subjects With and Without Dysphonia


 Oliveira Amorim,
 jo Pernambuco, Geova
*Patricia Maria Mendes Balata, *Hilton Justino Silva, Leandro Arau
*Renata Souto Maior Braga, *Elthon Gomes Fernandes da Silva, *Leilane Maria de Lima,
and *Slvia Regina Arruda Moraes, *Recife, Pernambuco, yNatal, Rio Grande do Norte, and zMaceio, Alagoas, Brazil
Summary: Objective. To compare dysphonic individuals to nondysphonic with regards to electrical activity of
extrinsic laryngeal muscles related to perceptual and acoustic vocal parameters.
Hypothesis. Dysphonic individuals have higher electrical activity in the supra and infrahyoid muscles than those
nondysphonic.
Study Design. Prospective, cross-sectional, case series study.
Methods. Forty-one subjects, divided into two groups according to the presence of dysphonia, underwent evaluation
of surface electromyography, auditory-perceptual, and acoustic evaluations of voice during the vocal rest and sustained
emissions of the vowel // and count of 20 to 30 at usual and strong intensities.
Results. The dysphonic group differed significantly from the nondysphonic by (1) lower electrical activity normalized
by the maximum sustained voluntary activity evaluated in all tasks of phonation in the suprahyoid group; (2) lower
recruitment of electrical activity in emissions of strong intensity compared with those of usual intensity in the suprahyoid muscles to emit the vowel // (13.66 5.17 in dysphonic group and 35.20 7.60 in the nondysphonic group,
P 0.029) and in the infrahyoid muscles in the count of 20 to 30 (14.90 4.69 vs 42.01 6.15; P < 0.001) and to
emit the vowel // (11.47 6.52 vs 22.66 9.05, P < 0.001); (3) lower vocal intensity to produce the vowel // in usual
and strong intensities and count in strong intensity. The electrical activities of the maximum sustained voluntary activity
were reduced with increasing degree of dysphonia.
Conclusions. There was lower electrical activity of the extrinsic laryngeal muscles in dysphonic individuals
compared with nondysphonic, and related to the degree of dysphonia.
Key Words: ElectromyographyAcoustic analysisPhonationDysphonia.
INTRODUCTION
Dysphonia is assessed by perceptual, acoustic, and visual parameters. Recently, the number of research seeking to establish
the relation between electrical activity of laryngeal extrinsic
muscles and dysphonia has grown.13
These research use the surface electromyography (sEMG).
The sEMG is a different method from the electromyography using needles; which despite allowing the assessment of smaller
and deeper muscles, has limitations, as the discomfort to the patient and difficulty to obtain data of dynamic tasks, that is, during speech, suffering more probability of noise or spurious
interference.
The sEMG did not establish itself as a tool for diagnosis of
dysphonia. The research about this topic are few and it is
hard to compare them, because they vary in analyzed muscle
groups, examination technique of phonation muscles, sample
size, and sign standardization method.
Last decade studies point out that dysphonic patients electrical activity is higher than that of normal subjects, but because of
methodological differences these data were not corroborated by
recent research.46
Accepted for publication March 18, 2014.
From the *Speech and Language Department, Universidade Federal de Pernambuco,
Recife, Pernambuco, Brazil; ySpeech and Language Department, Universidade Federal do
Rio Grande do Norte, Natal, Rio Grande do Norte, Brazil; and the zInstitute of Human Sciences, Communication and Arts, Universidade Federal de Alagoas, Maceio, Alagoas, Brazil.
Address correspondence and reprint requests to Patricia Maria Mendes Balata, Av.
Domingos Ferreira 636 sala 208, Pina, Recife, PE, Brazil, CEP 51011-010. E-mail:
pbalata@uol.com.br
Journal of Voice, Vol. 29, No. 1, pp. 129.e9-129.e17
0892-1997/$36.00
2015 The Voice Foundation
http://dx.doi.org/10.1016/j.jvoice.2014.03.012

Vocal tasks requested in sEMG assessment are also diverse


and there is no consensus about which one is more precise or
adequate. This procedure may suffer the action of diverse variables, for instance the adipose tissue conditions, muscle
strength, mechanic artifact by electrodes movements during dynamic tasks as fundamental frequency variations, singing,
reading, and others. Moreover, there is an important gap in
knowledge in laryngeal muscles contraction physiology, which
would allow understanding its behavior in laryngeal functions
such as protection, breathing, swallowing, and phonation.
Considering the previous research results, the hypothesis to
this study is that dysphonic subjects have higher electrical activity than normal subjects at suprahyoid (SH) and infrahyoid
(IH) muscles, because they recruit more muscles fibers during
vowel emissions and passage of speech in usual and strong
intensities.
Observing the proper technique to sEMG, defining a standard
of normalization that keeps the direct relation to voice and
emissions that do not require excessive muscle movements,
the present study has as purpose to compare dysphonic and
normal subjects regarding laryngeal extrinsic muscles electric
activity related to voice perceptual and acoustic parameters.
METHODS
Forty-one volunteer subjects, aged between 28 and 57 years,
mean 37.92 1.46 years, median 41 years were included. In
sex distribution the subjects were predominantly female (36
[87.8%]). These participants did not have the following selfreported conditions or evident to physical examination: cervical
joint dysfunction, hearing impairment of any degree

129.e10

Journal of Voice, Vol. 29, No. 1, 2015

compromising the examination, and the use of metallic orthoses


or prostheses. All participants attended to Divis~ao de Reabilitac~ao do Hospital dos Servidores do Estado de Pernambuco
between February and November, 2012.
The sample was divided in two groups: dysphonic (D group),
with 19 subjects (46.34%); and nondysphonic (ND Group),
with 22 subjects (53.66%). The division considered the possibility of conflict between voice and speech proposed tasks, sustained vowel //, and counting numbers, respectively, due to
idiosyncratic supraglottic adjusts during counting numbers,
and the vocal complaint presence itself does not characterize
dysphonia, although it is one indicative.
The criteria to dysphonia group inclusion considered the occurrence of divergence between vocal emissions and took the necessary rigor that this kind of examination requires, considering that
counting numbers is speech representative and, therefore, where
the main complaint lies. It was agreed once a vocal deviation
was detected, regardless presence of voice complaint, the subject
was classified as dysphonic. But, if there was a deviation detected
in sustained vowel, and there was vocal complaint reported, this
subject was also placed in the D group. To the ND group, it was
agreed that the participant should have absence of vocal
complaint and voice deviation; but the subject was also placed
in this group if there was a complaint and the voice kept its normal
variability in voice quality during sustained vowel and counting
numbers tasks according to visual analog scale evaluation (VAS).7
VAS was used because it is widely applied to vocal
screening.7 This scale is a 100-mm line, with cutoff, which
the numeric correspondence allows the characterization of
vocal quality variability in degrees: 1normal, from zero to
35.5 mm; 2slight to moderate deviation, from 35.5 to
50.5 mm; 3moderate deviation, from 50.5 to 90.5 mm; and
finally, 4severe deviation, values higher than 90.5 mm. The
criteria are summed up at Table 1.
The intention is not to affirm that vocal complaint should not
be a value, yet many of these complaints may not be related to
voice deviation but to other disorders, as sore throat, for
instance, different from pain during speaking.8 Similarly, vocal
disorders may be present without vocal complaint also is, as
pointed out in a study by Corazza et al.9
Perceptual evaluation has high sensitivity to detect voice disorders, although it has no condition to establish the presence of
laryngeal diseases.8,9 However, once this procedure showed
high alpha Cronbach coefficient among the speech-language
pathologists that were judges in this research (a .810;
P < 0.001, to VAS, and a .686; P < 0.001 to numeric scale),

it was possible to consider the procedure proper to subjects classification in groups. The sEMG results corroborated the group
classification because it showed differences between the
groups, which will be presented and discussed as follow.
It was adopted convenience simple random sample, using table of random numbers, by the premise of homogeneous sample
exhaustion identified for the absence of discrepancy in sEMG
evaluation (outliers), as well for the Altman nomogram, considered the significance level adjustments, of test power and effect
to be identified by the research.
The MIOTOOL 200 (Miotec, Brazil) electromyography
was used to collect the electrical potentials of SH and IH muscles
groups in microvolt (mV), with the possibility to select four independent channels, with 32 windowing and 2000 gain for each
channel. Because there are no references about the gain range
used to capture the electrical activity of these muscles, the gain
range previously mentioned was adopted, which allowed the
adjustment of signal to muscles reaching 574.93 mV, according
to the manufacturers instruction. Three channels were used,
each one connected to an active sensor SDS500 by claws; reference cable; calibrator; universal serial bus communication cable;
all from Miotec, Brazil, and disposable child surface electrodes from MEDITRACE (Kendall, Canada). The sign analysis was performed using the Miograph 2.0 software.
The sEMG equipment was connected to an LG notebook (LG
Electronics, S~ao Paulo, Brazil), with main configurations
160 GB HD, Intel Dual-Core Inside 1.7 GHz processor (Intel Corporation, USA), Windows Vista Premium operational system.
Voice assessment was made by digitally recording voice in a
portable computer (Sony Vaio, Intel 2.3 GHz processor; Sony,
Brazil). Voice was captured by a unidirectional head set microphone, Sennheiser PC-20 (Sennheiser, Germany), placed
3 cm at the side of the mouth, to avoid exhale noise interference.
To catalog the voice recordings, the register were processed in
VoxMetria software, 4.7 h version (CTS Informatica, Brazil).
Considering the interest of this research in acoustic data was to
assess mean fundamental frequency and loudness in usual and
strong intensities, the voice analysis mode was chosen to register
sustained vowel // and counting numbers from 20 to 30.
All participants were guided about their rights and duties in
this research, purpose of the study and then signed the informed
consent. Later, all of them passed through a structured interview
to identify the complaints and the evident signs indicating voice,
hearing, and cervical disturbances, to reduce selection bias.
The electromyography and voice recording were made in the
speech-language room, where the study took place, with the

TABLE 1.
Criteria for Classifying the Individuals According to Groups of Analysis
Criteria
Count of 20 to 30 in usual intensity
Vocal complaint
Emission of vowel // in usual intensity

Group D
(Dysphonic)

Group D
(Dysphonic)

Group ND
(Nondysphonic)

Group ND
(Nondysphonic)

VAS > 1
+ or 

VAS 1
+
VAS > 1

VAS 1


VAS 1
+
VAS 1

Abbreviations: VAS, degree of vocal change assessed by visual analog scale; +, present; , absent; + or , present or absent.

Patricia Maria Mendes Balata, et al

Electrical Activity of Extrinsic Laryngeal Muscles by SEMG

environment temperature of 18 C maintained by airconditioner. The subject was requested to remain seated
straight, with eyes opened, feet on the ground, arms resting
on lower members, with the back to the equipment, to avoid
any attempt to visual feedback. It was not necessary to shave
the participants. After cleaning the place with gauze and 70%
ethyl alcohol, for which abrasion is higher than cotton, the electrodes were placed: one channel with two electrodes in submandibular region, to obtain signs mainly from milohyoid and
digastrics muscles considered as mouth floor, toward anterior
venter of digastrics muscle; and two channels placed bilateral
to larynx, 1 cm from the thyroid notch,10 measured using a digital caliper rule. The distance between the electrodes was 1 cm,
center from center.
Assessment started following the order presented in Figure 1.
To capture maximum voluntary sustained activity (MVSA) it
was required of the subject to perform incomplete swallowing
maneuver with effort and retracted tongue with opened mouth
maneuver, as recommended by Balata et al.10 Each procedure
was performed taking 10 seconds rest between tasks, and the
researcher always provided an explanation about the task to
be performed, afterward the voice and electromyography signs
were recorded.
All tasks were timed, assuming the command emission by the
researcher as zero for the subject to start. Total time to collect
data varied from 45 to 60 minutes, depending on subjects
comprehension to perform the tasks and the electric sign quality
captured by the electrodes.
Data analysis had three moments. The first was the electromyography data preparing, that is, selection, filtering, and
analyzing the sign. The second was perceptual and acoustic
vocal analysis; and the third was the statistic data analysis.
At the first moment, electromyography normalization sign of
SH and IH was started by the raw sign register in microvolt (mV).

129.e11

For the signs of MVSA, sustained vowel //, and rest, the first and
last second of each recording were eliminated. In muscle maneuvers, the register of the three middle seconds were considered the
most stable ones, and, therefore selected to be analyzed. At rest,
the selected register was at 58 seconds interval.
At the analysis of electromyography signs during counting
numbers in habitual and loud volumes, the whole register was
selected and its duration varied according to speed rate and/or
voice disorders which could cause pauses along the speech. After
the interval selection, each one of them was doubled to posterior
mean square root analysis (Root Mean SquareRMS).
Online Butterworth digital filter was used to avoid noises and
mechanic artifacts of electromyography (EMG) sign. This is a
band pass filter in 20500 Hz frequency, configured as superior
and inferior filtering cut point. Frequency spectrum through
Fourier Fast Transform was used which allows the manual
filtering using Notch, at 60 Hz frequency the same of energy
supply interference.
Once the sign was filtered and the interferences taken off, the
software module that allows RMS values extraction, was used
to obtain maximum and minimum statistic parameters, and the
mean electric activity in microvolt (mV). Muscle electric activities
were captured during sustained vowel and speech, in usual and
strong intensities, converted to microvolt in percentage of normalization parameter, previously defined,11 according to Table 2.
At second analysis moment, GRBASI international scale, in
Likert, was used to measure general dysphonia grade, which allowed quantifying and qualifying the impact of general voice
deviation (G parameter) as follows: 0 , absence of dysphonia;
1 , slight deviation; 2 , moderate deviations; and 3 , severe
deviations.
Assuming MVSA as standard for electric activity sign
normalization, associations between electromyography and
acoustic analysis were searched, comparing dysphonic and

FIGURE 1. Flowchart of research data collection.

129.e12

Journal of Voice, Vol. 29, No. 1, 2015

TABLE 2.
Statement of Criteria and Formulas for Converting the Electrical Activity of the Extrinsic Laryngeal Muscles from Microvolts
to Percentage

Task
MVSA
Emission of
vowel //
Count emission

Normalizing
Parameter
Mean
Peak

Emission
Formula for Calculation
EA of emission %

EA of emission mV3100
EA of parameter admitted as 100%mV

Peak

nondysphonic subjects. Therefore, in acoustic analysis, mean


fundamental frequency and intensity in sustained vowel and
speech to both intensities (habitual and strong), and the
dysphonia degree were used.
After data consistence analysis, the sample description and
those used to place the subjects in the studied groups were
summed up in statistic descriptive parameters, according to
measurement level.
At D and ND group comparison of physical examination,
electromyography, acoustic analysis, the summed up parameters, and the statistic descriptive variation parameters were
used ANOVA test (Analysis of Variance) to mean difference
with F test of variance homogeneity, Student t test to mean of
independent samples; Kendall correlation coefficient to identify the association between vocal deviation and extrinsic laryngeal muscles electric activity. The significance level of 0.05 was
adopted for all tests as the cut to reject null hypothesis.
This research was approved by the Ethic in Research Committee of Universidade Federal de Pernambuco under the
number 0469.0.000.17209.
RESULTS
The sample was characterized by higher frequency of subjects
aged between 38 and 47 years in D group, which derived the
mean 43.53 1.89 years, similarly at ND group, with predominance of subjects up to 47 years, mean 39.91 1.57 years. This
result was not statistically significant.
Comparing the groups regarding vocal complaint duration, D
group (mean 246.39 28.03 days, corresponding to 8.2 months,
median 10 months) had vocal complaint for more time than ND
group (mean 125.80 21.71 days, corresponding to 4.3 months
and median 2.4 months) and this difference was statistically significant (P 0.003) according to Figure 2.
At Table 3, mean distribution and standard error of MVSA
normalized electric activity of both groups are presented, according to laryngeal extrinsic muscles and normalization tasks.
For most of the normalization tasks, D group had lower normalized mean than ND group, at the two muscle groups, with
exception to counting numbers from 20 to 30 in habitual volume at IH group. It was evident that, to SH muscle group,
this difference was statistically significant, except to vocal rest.
It is relevant to note that the lowest means at MVSA normalized electric activity in D group were also the ones with lowest
variability values, expressed by the mean standard error
(Table 3).

Vowel //

Count

Usual and strong


Usual and strong

Usual and strong

Usual and strong

Besides, the vocal rest in IH muscles, in D group, had lower


normalized electric activity mean than ND group, reaching intermediate P values between 0.05 and 0.10, therefore suggestive of a possible difference between groups at this task, but
not evident (Table 3).
At Table 3, it also observed that loud emissions were the ones
with major electric activity MVSA normalized when compared
with the same emissions in usual intensity. except the sustained
vowel // in IH muscle.
Through the mean difference of MVSA normalized electric
activity identification between emissions in habitual and loud
volumes, D and ND group were compared identifying the
lowest variability in SH muscle to sustained vowel, and, in IH
muscles to counting number from 20 to 30 (Table 4).
Observe that both P values at the sustained vowel, the difference in mean IH electric activity and in SH muscle during
counting numbers, are equal 0.054, indicating the possibility
of difference between groups, which should be confirmed in
other research with large samples.
At Table 5 are presented the means, standard error of mean of
acoustic parameters assessed in sustained vowel, and counting
numbers in usual and strong intensities, which allows identifying the mean fundamental frequency values in usual intensity.
In both, emissions were higher in D group than ND group, but in
strong intensity emissions biggest values remained in ND
group. These differences of fundamental frequency values
were not significant.

FIGURE 2. Box-plot of the overall median time of vocal complaints, according to study groups.

Patricia Maria Mendes Balata, et al

Electrical Activity of Extrinsic Laryngeal Muscles by SEMG

129.e13

TABLE 3.
Distribution of Mean and Standard-Error of the Mean Normalized Electrical Activity by MVSA of Muscle Groups, According
to the Task and Comparison Groups
Means and Standard Errors of the Mean
Normalized Electrical Activity by MVSA
According to Comparison Group
Tasks

D (n 19) (M Sem)

ND (n 22) (M Sem)

P Value

Vocal rest
Vowel // in usual intensity
Vowel // in strong intensity
Count of 20 to 30 in usual intensity
Count of 20 to 30 in strong intensity
Vocal rest
Vowel // in usual intensity
Vowel // in strong intensity
Count of 20 to 30 in usual intensity
Count of 2030 in strong intensity

17.30 2.32
63.33 11.72
69.62 11.83
87.63 13.28
95.41 15.44
14.48 3.08
27.92 2.77
41.57 6.37
45.32 5.02
63.69 7.94

30.54 5.83
71.69 12.61
100.29 16.00
91.34 14.50
128.06 17.36
17.51 4.07
44.96 6.09
80.16 10.49
73.24 11.14
107.97 14.54

0.053
0.622
0.142
0.866
0.174
0.567
0.016
0.003
0.030
0.012

Muscle Evaluated
IH

SH

Abbreviations: M, mean; sem, standard error of mean.


Notes: P-value for the analysis of variance (ANOVA) by F test.
Significance level < 0.05 shown by bold type; significance level between 0.05 and 0.10 shown with underline.

Regarding intensity, the groups behavior was different.


Indistinct to all emissions, regardless the vocal intensity, the
acoustic means were lower in Group D than ND, with statistic
significance at sustained vowel // to both volumes, as well at
counting numbers from 20 to 30 exclusively to loud volume
(Table 5).
Table 6 shows the vocal deviation distribution of dysphonic
subjects assessed by G of GRBASI which deviation grade 2
to emissions in usual intensity predominated.
Reduction in electric activity is observed with the increase of
dysphonia degree to all emissions and evaluated muscles. Even
distinct, none of these differences reached statistical significance,
also there was no significant relation between electric activities
and dysphonia degree assessed by the Kendall correlation coefficient, which may be verified with larger sample size.

DISCUSSION
The results of this article had broken a paradigm: dysphonics
would have major electric activity (EA) because they recruit
more muscles fibers during sustained vowels emissions and

speech, due to a possible increase in muscle tension generated


by dysphonia.11,12
The strength of data presented in Table 3 point out dysphonics recruited less motor units that generated lower electric
activity of SH muscles than normal subjects for all tasks.
To IH muscle group, even without significance relation,
values close to 0.05 at rest indicate that this data are clinically
relevant and may be considered.
These results were surprising because other authors proved
dysphonics to have major electric activity than normal because
of its compensatory muscle condition to phonate, but it is
important to consider that sEMG technique from these studies
are different.4,11
When intrinsic laryngeal muscles reach fatigue level, it is
common that the person activates extrinsic muscles compensatorily, which may generate a habitual standard to phonate, characterizing hyperfunctional dysphonia or the muscle tension
syndrome.1214
Current references, as Van Houtte et al,3 Stepp et al,1517
evaluated muscle tension dysphonia relating them with sEMG
data concluding that this tool (1) do not have diagnostic

TABLE 4.
Distribution of the Differences of Means Normalized Electrical Activities by MVSA of Extrinsic Laryngeal Muscles Between
Emissions in Usual and Strong Intensities
Differences of Means Normalized Electrical Activities Between
Emissions in Usual and Strong Intensities
Muscle

Emission

D (n 19) (Mean Sem)

ND (n 22) (Mean Sem)

P Value

IH

Vowel //
Count
Vowel //
Count

11.47 6.52
14.90 4.69
13.66 5.17
18.37 4.17

22.66 9.05
42.01 6.15
35.20 7.60
34.73 6.73

<0.001
<0.001
0.029
0.054

SH

Abbreviation: sem, standard error of mean percentage normalized by MVSA.


Notes: P value by the Students t test for paired samples.
Significance level < 0.05 shown by bold type; significance level between 0.05 and 0.10 shown with underline.

129.e14

Journal of Voice, Vol. 29, No. 1, 2015

TABLE 5.
Distribution of Means and Standard Errors of Acoustic Parameters Measured in Emissions in Usual and Strong Intensities
Means and Standard-Errors
Acoustic Parameters
Fundamental frequency (Hz)

Intensity (dB)

Emission

D (n 19) Mean Sem

ND (n 22) Mean Sem

P Values

195.08 14.10
244.77 10.88

194.66 7.59
262.52 12.48

0.979
0.297

184.86 10.67
255.14 11.13

180.97 7.64
264.99 8.91

0.764
0.489

64.20 1.05
79.54 1.36

68.74 1.23
83.38 0.94

0.009
0.023

50.22 1.15
61.70 1.38

53.32 1.40
66.24 1.15

0.101
0.015

Vowel //
Usual intensity
Strong intensity
Count of 20 to 30
Usual intensity
Strong intensity
Vowel //
Usual intensity
Strong intensity
Count of 20 to 30
Usual intensity
Strong intensity

Notes: Significance level < 0.05 shown by bold type.

power to hyperfunctional dysphonia; (2) do not have specificity


to identify the presence of vocal nodules, and, (3) do not have
sensitivity to evaluate vocal hyperfunction. On the other
hand, last decades studies, as Redenbaugh and Reich6;
Hocevar-Boltezar, Janko and Zargi4 identify that EA in studied
muscles was higher in dysphonics.
The main problem, these authors discuss, is the fact that the
studies did not converge to assessed muscles, suprahyoid,6,15
infrahyoid,2,16,17 sternocleidomastoid,2,16 not even to
normalization use or method.18
Because the present study seeks to evaluate exclusively the
extrinsic laryngeal muscles, assuming normalization by
maximum sustained voluntary electric activity and it was
observed differences in the method used at the other studies,
it is not possible to compare the results to discuss them.
To support the presented results, it is worth talking about the
physiology of skeletal muscles, although few research about types

of SH and IH muscle fibers were found, different from the intrinsic


laryngeal muscles which authors are dedicating time to understand the effects of voice techniques in dysphonia treatment.1921
Laryngeal biomechanics require extrinsic and intrinsic control. At phonation, this relation is defined regarding elevation
or reduction in fundamental frequency, which may depend on
cricothyroid intrinsic muscle, the extrinsic muscle (IH) and
sternocleidomastoid (SCM) interaction. The first, when activated increases the fundamental frequency and the second enhances lowering voice because it pulls larynx down, reducing
the angle between cricothyroid and thyroid cartilages.22
Although the studies are few, Vilman et al23 revision points
out some needle electromyography studies in which sternohyoid and SH had major EA during speech and singing in
raising and falling fundamental frequency.
Skeletal muscles are composed of fiber matrix types, which
keep relation with their function and are distributed to all

TABLE 6.
Distribution of 19 Dysphonic Subjects According to the Vocal Deviation
Electric Activity (mV) by Dysphonic Grade
Muscles
Dysphonic in vowel // in usual intensity
Dysphonic in count of 20 to 30
Measurements in MSVA
IH
SH
Measurements in vocal rest
IH
SH

Emissions

5 (26.32)
6 (31.58)

9 (47.36)
9 (47.36)

5 (26.32)
4 (21.06)

Vowel//
Count of 20 to 30
Vowel//
Count of 20 to 30

43.50 14.11
40.78 11.79
56.25 13.12
50.62 12.10

35.59 5.54
35.34 5.60
43.38 9.60
43.27 9.63

20.27 3.13
19.56 3.99
39.37 7.42
46.87 7.63

Vowel//
Count of 20 to 30
Vowel//
Count of 20 to 30

5.28 1.24
4.84 1.07
4.22 1.60
3.98 1.33

4.30 0.64
4.30 0.64
1.45 0.48
3.97 0.49

3.61 0.51
3.85 0.60
3.58 0.69
3.22 0.77

Notes: Correspondence was maintained between the emission in usual intensity of vowel // and the emission of count of 20 to 30 by respective G consensus.

Patricia Maria Mendes Balata, et al

Electrical Activity of Extrinsic Laryngeal Muscles by SEMG

skeletal muscle tissue in variable portions. Slow contraction fibers are entitled type I, and fast contraction fibers, type II. These
last ones are subdivided into three subgroups: IIa, IIb, and
IId/x.24,25
Type I fibers are involved in aerobic and resistance activities
once they have major quantity of mitochondria, higher blood
flow, and more myoglobin, characterizing them as red fibers
and are, therefore, more resistant to fatigue. Type II fibers
have lower expressed myoglobin, therefore characterized as
white fibers. Type IIa has fast oxidative contraction and higher
contraction speed, but low resistance to fatigue. Type IIb has
fast glycolytic contraction and has lower mitochondria, activated when short and intense activity shots are necessary.
Type IId/x has intermediate properties between IIa and IIb.24,25
Hoh25 revised the reason of intrinsic laryngeal muscles to
have hybrid fiber type. It was identified that these muscles
composed by dynamic tissues, which characteristics may be
altered to attend the complexity of laryngeal functions (protection, breathing, swallowing and phonation, and last one phylogenetic) and therefore present different standards that may
suffer adaptive change with hormone and neural regulation.
However, the study does not research the extrinsic muscles.
Extrinsic laryngeal muscle has participation in voice modulation, contributing to size of vocal folds variations and resonant
characteristics.23,26 They act indirectly in phonation because
of its anchor in neck, pulling larynx and hyoid bone in four
directions: anterior, posterior, cranial, and caudal. These
muscles fixated on hyoid and other structures as jaw, mastoid,
and thorax bones are divided into SH and IH muscles.23
These muscles are theoretically antagonists because the first
has mainly the function to raise and the second the function to
lower the larynx. Muscle dynamics may favor change in function allowing that the antagonist muscle, for example, act as
synergist in other movement, but there is no evidence if this
biomechanics happens during phonation.
In cases of hyperfunction of extrinsic laryngeal muscles,
because of intensive or inappropriate use of voice, this tension
may raise the larynx, increase muscle mass in the neck, cause
pain during palpation of the regions, and therefore collaborating
to functional dysphonia or its worsening.27
Studying vocal tremor in six subjects using laryngeal EMG
by needle in two intrinsic muscles and two extrinsic: thyroid
and sternothyroid, during sustained vowel /i/ task, Finnegan
et al28 verified that even showing unsystematic responses
among the subjects, there was electric activity in these muscles
during emission.
Although the function of extrinsic muscles during phonation
is recognized, some aspects still require research. They act in
fixation of hyoid bone. However, looking to topographic dynamics of hyoid bone in laryngeal functions, because of its
importance in cranium-cervical stability, German et al29 identified the existence of an information gap about aspects as resistance to fatigue, contraction time, and hysteresis of muscles
fibers related to passive or active balance. Furthermore, the
lack of knowledge about the fiber composition disfavor inferences about muscle condition related to strength, resistance,
and fatigue.

129.e15

Physiologic explanation most likely to the results of this


research is that dysphonic subjects had lower muscle resistance
than normal subjects, hypothesis that is supported by the duration of 8 months of complaints, dysphonia of moderate degree,
characterizing chronic voice deterioration. This possible low
resistance may have generated lower recruitment of fibers and
caused muscle fatigue, phenomenon that was not the purpose
of this research.
Besides, the fact that the tasks were performed by the subjects continuously may have contributed to muscle fatigue, after
all, the order of two MVSA maneuvers, repeated three times,
even with rest breaks, followed by assessment of rest and
then to three emissions of sustained vowels and counting
numbers in different volumes, required major muscle recruitment and may indicate that dysphonics may not hold long
and higher effort tasks.
Because the composition of fiber types of the assessed muscle groups is unknown, some speculations about the functional
behavior were made.
The two types of fibers differ in their metabolic properties.
Type I is proper to sustained contraction, due to fatigue resistance; type II proper to fast contraction. It is possible to suppose
that sustained vowel emissions were tasks that recruit mainly fibers type I, because of the need of major laryngeal stability.
Muscle fatigue, however, may be characterized for the reduction in the capacity of neuromuscular system to generate force
to execute this action.
Counting numbers from 20 to 30 is a more dynamic and complex task when compared with sustained vowel //. It requires
more breathing support and orofacial muscles activation to
articulate the sounds of speech, and may also require extrinsic
laryngeal muscles adjustments, especially in strong intensity
which produced higher EA.
It is assumed that this correlation is correct once the speechlanguage treatment in voice has as purpose to improve vocal
resistance in functional and organic-functional dysphonia.
Other data that stand out is the MVSA mean to be higher in
dysphonics, especially in SH muscles, which seems to be a
paradox comparing to the lowest EA detected in the emissions.
This aspect may have foundation at absolute force concept that
differs from resistance force, because it implies at the capability
to overcome fatigue in longer tasks.30 Absolute force is the
maximum tension that a muscle or a muscle group may
generate, which at this study was the type isotonic concentric,
performed during MVSA.
At this context it is possible to affirm that dysphonics may
have major absolute force potential because they can generate
more muscle tension at a maximum contraction action; however, they are not efficient to hold the required sustenance at performing phonatory tasks. According to Blanc and Dimanico,31
SEMG do not allow extracting force measures, therefore this
argumentative proposition is based only on clinical thinking,
which is fundamental to validate electromyography data.
Taking as base the dysphonia general deviation degree (G) it
was observed reduction in muscle recruitment and as consequence lower electrical activity, as the dysphonia degree
increased, which corroborates the presented argument.

129.e16
Dysphonia severity seems to produce the worst muscle performance. It is interesting to remember that extrinsic muscles have
a direct relation to phonation, especially to fundamental
frequency.
Acoustic analysis of the mean loudness values showed lower
variability in the dysphonic group, which agree to the argument
that it is a necessary major energetic and metabolic support to
increase loudness and these subjects could not do it as normal
ones can.
Analyzing acoustic parameters of fundamental frequency
and loudness, finding lower value between dysphonics when
compared with normal seems to reinforce the hypothesis that
lower electric activity of extrinsic laryngeal muscles may
have contributed to its reduction.
To detect significant difference between dysphonic and
normal exclusively in loudness, P-values between 0.002 and
0.02, it is a strong indicative of association between dysphonia
and the difficulty to control loudness.
The results of rest test were also surprising. It was expected
that basal activity was increased due to muscle tension maintenance which is verified at dysphonic phonation. However, physiologically, it would be odd if this happened in this research
because of the lowest value of electric activity detected during
phonation. It is myoelectric principle that once a lower functional recruitment happens, the same must happen at basal activity, pointing out that evaluated muscle has low action
potential, limiting the contraction process.
It is necessary to suggest possible explanations to the presented results. The most likely as the methodology adopted,
which was based on previous studies that did not exclusively
evaluate SH and IH groups. Task repetitions, common in
myoelectric evaluations, were not arbitrary; the electromyography assessment method respected proper technical
directions.
MVSA normalization of SH and IH muscles also constitute a
differential, because it promoted the maximum contraction of
muscles directly involved in vocal function, differing from
other studies where normalization was performed using
maximum or submaximum contraction of muscles with indirect
relation to phonation.

CONCLUSION
Dysphonic subjects showed lower electric activity of extrinsic
laryngeal muscles when compared with nondysphonic. The
fact that the studied muscles groups had lower electric activity
in this population may be indicative of a reduction in muscle
resistance.
Another factor to be considered is the possibility of muscle
fatigue in dysphonics, perhaps by the excessive use of these
muscles, thus this finding may indicate the need to change the
voice therapy for these subjects. It seems reasonable to admit
that this therapy should invigorate these muscles instead of promote its relaxation.
Biochemical and histological analysis of these muscles
would better explain these results, because of the relation of
electric activity with the kind of muscle fiber and its conditions

Journal of Voice, Vol. 29, No. 1, 2015

in dysfunctions. Such relation is yet unknown in suprahyoid and


infrahyoid muscles.

Acknowledgments
The authors thanks the National Council of Technological and
Scientific Development (CNPq), which had a financial support
with Universal Edictal MCT/CNPQ 14/2009, area B, process
476412/2009. There are no conflicts of interest.

REFERENCES
1. Stepp CE, Heaton JT, Braden MN, Jette ME, Stadelman-Cohen TK,
Hillman RE. Comparison of neck tension palpation rating systems with surface electromyographic and acoustic measures in vocal hyperfunction. J
Voice. 2011;25:6775.
2. Stepp CE, Heaton JT, Jette ME, Burns JA, Hillman RE. Neck surface electromyography as a measure of vocal hyperfunction before and after injection laryngoplasty. Ann Otol Rhinol Laryngol. 2010;119:594601.
3. Van Houtte E, Van Lierde K, Claeys S. Pathophysiology and treatment of
muscle tension dysphonia: a review of the current knowledge. J Voice.
2011;25:202207.
4. Hocevar-Boltezar I, Janko M, Zargi M. Role of surface EMG in diagnostics
and treatment of muscle tension dysphonia. Acta Otolaryngol. 1998;118:
739743.
5. Milutinovic Z, Lastovka M, Vohradnk M, Janosevic S. EMG study of hyperkinetic phonation using surface electrodes. Folia Phoniatr (Basel).
1988;40:2130.
6. Redenbaugh MA, Reich AR. Surface EMG and related measures in normal
and vocally hyperfunctional speakers. J Speech Hear Disord. 1989;54:
6873.
7. Yamasaki R, Le~ao SHS, Madazio G, Padovani M, Azevedo R, Behlau M.
Correspond^encia entre escala analogico-visual e a escala numerica na
avaliac~ao perceptivo-auditiva de vozes. In: XVI Congresso Brasileiro de
Fonoaudiologia. 2008; Campos de Jord~ao, SP.
8. Eckley CA, Anelli W, Duprat AC. Sensitivity and specificity of perceptual
hearing analysis of voice in screening laryngeal disorders. Rev Bras
Otorrinolaringol. 2008;74:168171.
9. Corazza VR, Silva VFC, Queija DS, Dedivitis RA, Barros APB. Correlation
between stroboscopic perceptual hearing and acoustic findings in adults
without vocal complaints. Rev Bras Otorrinolaringol. 2004;70:3034.
10. Balata PMM, Silva HJ, Nascimento GKBO, et al. Incomplete swallowing
and retracted tongue maneuvers for electromyographic signal normalization of the extrinsic muscles of the larynx. J Voice. 2012;26:813.e1813.e7.
11. Guirro RRJ, Bigaton DR, Silverio KCA, Berni KCS, Distefano G,
Santos FL, Forti F. Transcutaneous electric nerve stimulation in dysphonic
women. Pro Fono. 2008;20:189194.
12. Altman KW, Atkinson C, Lazarus C. Current and emerging concepts in
muscle tension dysphonia: a 30-month review. J Voice. 2005;19:261267.
13. Belafsky PC, Postma GN, Reulbach TR, Holland BW, Koufman JA. Muscle
tension dysphonia as a sign of underlying glottal insufficiency. Otolaryngol
Head Neck Surg. 2002;127:448451.
14. Morrison M, Rammage L, Nichol H. The management of voice disorders.
4th ed. San Diego, CA: Singular Publishing; 1994.
15. Stepp CE, Heaton JT, Rolland RG, Hillman RE. Neck and face surface electromyography for prosthetic voice control after total laryngectomy. IEEE
Trans Neural Syst Rehabil Eng. 2009;17:146155.
16. Stepp CE, Heaton JT, Stadelman-Cohen TK, Braden MN, Jette ME,
Hillman RE. Characteristics of phonatory function in singers and nonsingers with vocal fold nodules. J Voice. 2011;25:714724.
17. Stepp CE, Hillman RE, Heaton JT. Modulation of neck intermuscular Beta
coherence during voice and speech production. J Speech Lang Hear Res.
2011;54:836844.
18. Balata PMM, Silva HJ, Moraes KJR, Pernambuco LA, Moraes SRA. Use of
surface electromyography in phonation studies: an integrative review. Int
Arch Otorhinolaryngol. 2013;17:329339.

Patricia Maria Mendes Balata, et al

Electrical Activity of Extrinsic Laryngeal Muscles by SEMG

19. Cordeiro GF, Montagnoli AN, Nemr NK, Menezes MH, Tsuji DH.
Comparative analysis of the closed quotient for lip and tongue trills in relation to the sustained vowel //. J Voice. 2012;26:e17e22.
20. Cielo CA, Elias VS, Brum DM, Ferreira FV. Thyroarytenoid muscle and
basal sound: a literature review. Rev Soc Bras Fonoaudiol. 2011;16:
362369.
21. Azevedo LL, Passaglio KT, Rosseti MB, Silva CB, Oliveira BFV, Costa RC.
Vocal performance evaluation before and after the voiced tongue vibration
technique. Rev Soc Bras Fonoaudiol. 2010;15:343348.
22. Ludlow CL. Central nervous system control of the laryngeal muscles in humans. Respir Physiol Neurobiol. 2005;147:205222.
23. Vilkman E, Sonninen A, Hurme P, Korkko P. External laryngeal frame
function in voice production revisited: a review. J Voice. 1996;10:7892.
24. Liu G, Gabhann FM, Popel AS. Effects of fiber type and size on the heterogeneity of oxygen distribution in exercising skeletal muscle. PLoS One.
2012;7:e44375.
25. Hoh JFY. Laryngeal muscle fibre types. Acta Physiol Scand. 2005;183:
133149.

129.e17

26. Simonyan K, Horwitz B. Laryngeal motor cortex and control of speech in


humans. Neuroscientist. 2011;17:197208.
27. Behlau M, Madazio G, Feijo D, Azevedo R, Gielow I, Rehder MI. Vocal
improvement and speech and langugae treatment of dysphonia. In:
Behlau M, ed. (org): Voice: The specialist Book. S~ao Paulo, Brazil: Revinter; 2001:410564. (2).
28. Finnegan EM, Luschei ES, Barkmeier JM, Hoffman HT. Synchrony of
laryngeal muscle activity in persons with vocal tremor. Arch Otolaryngol
Head Neck Surg. 2003;129:313318.
29. German RZ, Campbell-Malone R, Crompton AW, Ding P, Holman S,
Konow N, Thexton AJ. The concept of hyoid posture. Dysphagia. 2011;
26:9798.
30. Mitchell WK, Williams J, Athernon P, Larvin M, Lund J, Narici M. Sarcopenia, dynapenia, and the impact of advancing age on human skeletal muscle size and strength; a quantitative review. Front Physiol. 2012;3:260.
31. Blanc Y, Dimanico U. History of the study of skeletal muscle function with
emphasis on kinesiological electromyography. Open Rehab J. 2010;3:
8493.

Вам также может понравиться