Physiologic and Acoustic Effects of Opera Performance

*Paul E. Kwak, †C. Richard Stasney, †Jeremy R. Hathway, ‡Danielle Guffey, ‡Charles G. Minard, and
§Julina Ongkasuwan, *Boston, Massachusetts, and †‡§Houston, Texas

Summary: Introduction/Hypothesis. Opera performance is physiological and emotional, and singing perform-
ers utilize their larynges in often strenuous ways. Historically, the training of a classical voice has been considered the
paragon of healthy singing. However, the natural history of a performing larynx has not been studied systematically.
There is paucity of scientific studies to guide practice patterns, particularly with regard to the course and extent of post-
performance physiologic and acoustic changes.
Study Design. A prospective case series was carried out.
Methods. Principal singers in the Houston Grand Opera’s 2012–2013 repertory were enlisted, for a total of seven
singers. Stroboscopy was performed prior to the start of rehearsals, and at the completion of the opera’s run. Data points
included erythema, edema, masses or lesions, mucosal waveform, supraglottic posture; acoustic measurements were
also performed.
Results. There were statistically significant differences (P < 0.05) in the mucosal wave on pre- and postperformance
stroboscopic examinations. Acoustical measures did not achieve statistical significance, but there was a trend toward
increased harmonic-to-noise ratio in postperformance measures, as well as decreased frequency range and reading F0.
Measures of intra- and inter-rater reliability indicated varying levels of intra-rater reliability, and generally poor inter-
rater reliability.
Conclusions. This pilot study describes physiologic and acoustic changes that may occur over the course of a series
of rehearsals and performances in the operatic larynx. In so doing, it highlights a need for larger studies with in-
creased frequency of serial examinations to study in a systematized way what may be natural reactive changes that
occur during performance.
Key Words: Professional voice–Stroboscopy–Acoustic measures of voice–Opera–Vocal performance.

INTRODUCTION
Opera fans are among the most impassioned of any devotees of
any musical genre, their fervor fueled largely by the sonic wonder
of a solitary human larynx trained to reach, unamplified, above
an orchestra and chorus across space to the back of a cavern-
ous hall. This most inspiring of artistic endeavors remains,
simultaneously, the most athletic of challenges to the anatomy
and physiology of the human larynx. In the evolution of oper-
atic performance, an arguable pinnacle of which was described
famously by Richard Wagner in 1849 in his vision of
Gesamtkunstwerk (approximately, “total artwork”), in which the
composer conceived opera, specifically, his expansive Ring cycle,
as the summation of all arts,1 more has been demanded of opera
singers through each century. Vocal and theatrical exigencies trans-
late into taxing demands on the microanatomy and gross
physiology of the larynx. This notwithstanding, opera perfor-
mance and the training of a classical voice has historically been
Accepted for publication March 7, 2016.
This study was presented at the American Laryngological Association’s 2014 Spring
Meeting at COSM in Las Vegas, Nevada, May 15, 2014.
Financial disclosures: None.
Conflicts of interest: None.
From the *Division of Laryngeal Surgery, Massachusetts General Hospital, Harvard Medical
School, Boston, Massachusetts; †Texas Voice Center, Houston Methodist Hospital, Houston,
Texas; ‡Dan L. Duncan Institute for Clinical and Translational Research, Baylor College
of Medicine, Houston, Texas; and the §Bobby R. Alford Department of Otolaryngology-
Head and Neck Surgery, Baylor College of Medicine, Houston, Texas.
Address correspondence and reprint requests to Paul E. Kwak, Division of Laryngeal
Surgery, Massachusetts General Hospital, One Bowdoin Square, 11th Floor, Boston,
Massachusetts 02114. E-mail: pkwak@partners.org
Journal of Voice, Vol. 31, No. 1, pp. 117.e11–117.e16
0892-1997
© 2017 The Voice Foundation. Published by Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.jvoice.2016.03.004
considered the paragon of healthy singing, with its focus on breath
support, dimension of tone, and minimizing of strain.
Medical understanding of the natural history of a perform-
ing larynx is based on decades of anecdotal observation and
experience. However, a review of the literature indicates that the
scientific studies to guide practice patterns are sparse. Both recent
and more historic studies have worked to characterize the dynamic
function of the larynx during singing, by analyzing supraglot-
tic activity and posture, and by challenging ideas about what
constitutes normal function versus hyperfunction during speak-
ing and singing.2–4 These studies have contributed to the idea
that the singing voice comprises much more than the approxi-
mation and vibration of the vocal folds.
What remains missing, however, is knowledge about how these
highly coordinated functions of the larynx and glottis, in fact,
change the larynx and glottis over time, if at all. There is not
yet reproducible evidence about the changes that occur either
in the acute setting following a performance, or over the course
of an operatic career. For example, it is often cited in informal
conversation that vocal fold edema lasts 36–48 hours after an
arduous performance, but there are no rigorous studies to support
that time course.
This pilot study set out to begin to investigate what, if any,
physiologic and acoustic changes happen over the course of op-
eratic performance. Recognizing the enormous multiplicity of
variables, ranging from an individual singer’s anatomy and phys-
iology, to variations in training and technique, to performance
conditions, to influences of lifestyle and environment, the goals
of this study were focused, and tailored to a small pilot cohort
of elite singers over a specific period of time.
117.e12
MATERIALS AND METHODS
Principal singers in two of the productions from the Houston
Grand Opera’s 2012–2013 repertory productions were en-
listed, for a total of seven principal singers. Stroboscopy was
performed 1 week prior to the start of rehearsals, and 1 week
following the completion of the performance period. All togeth-
er, this represented 4 weeks of rehearsals and 2 weeks that
included 7 performances.
At each examination, acoustic measurements were taken using
the Multi-Dimensional Voice Program (MDVP). Singers were
asked to read the standard “Marvin Williams” passage; the average
F0 was calculated from the first 10 seconds of this reading sample
(rF0). They were then asked to phonate /α/. They were then asked
to sing /α/ at a comfortable pitch (sF 0 ); during their
postperformance testing, they were also asked to sing /α/ at the
same sF0. Finally, they were asked to glissando on /α/ from their
lowest to highest pitch and then again from their highest to lowest
pitch. They then were asked to sing chromatically ascending and
descending scales to the extremes of their ranges. The highest
and lowest frequencies with stable phonation were recorded. All
sustained tasks were sustained for 4–5 seconds and were re-
peated for a total of three tokens per task. The first 0.25 second
of each token was then discarded and the following 3 seconds
were analyzed; the analysis of the three tokens within each task
was averaged using the Multiple Tokens Protocol within the
MDVP program.
Stroboscopic examinations were performed by a single
laryngologist (C.R.S.); singers were asked to phonate a sus-
tained /i/ at the rF0. These examinations were reviewed by two
laryngologists not involved in the performance of the examina-
tion. No identifying information was shown in the examinations,
nor was sound heard. The two raters were blinded to the date
of the examination as well, so that they were not aware whether
the examination was recorded before or after the series of re-
hearsals and performances. All examinations were shown in a
randomized order determined by computer-randomization soft-
ware. Data points evaluated in stroboscopy included: (1) extent
of erythema, (2) extent of edema, (3) presence of masses or
lesions, (4) evaluation of mucosal waveform, (5) closure pattern,
and (6) supraglottic posture.
The acoustic measures were summarized by means and stan-
dard deviations at initial and follow-up time points. Acoustic
measures were compared pre- and postperformance for differ-
ences using Wilcoxon signed rank tests. Correlations between
acoustic measures were evaluated both pre- and postperformance
using Spearman rank correlation.
Stroboscopy measures were described for each rater pre- and
postperformance. Changes in stroboscopy measures were de-
scribed as improved, no change, or worsened using frequencies
for both raters separately. Changes in stroboscopy measures were
compared pre- and postperformance using McNemar’s test or
Bower’s symmetry test for each rater. Agreement between the
change in rater’s ratings (worsen or not) was measured using a
kappa statistic. Agreement within a rater’s rating for an exam-
ination was measured using a weighted kappa statistic for
erythema and edema (they have more than two levels) and a kappa
statistic for the other stroboscopy measures. Statistical signifi-
Journal of Voice, Vol. 31, No. 1, 2017
cance was assessed at the 0.05 level, and no adjustments were
made for multiple hypothesis testing.
RESULTS
Stroboscopy data
When comparing pre-and postperformance stroboscopy mea-
sures, there is a significant worsening of mucosal waveform when
measured by the first rater (P = 0.046) but not by the second rater.
All kappa statistics comparing the two raters’ ratings were less
than 0.2, indicating poor agreement. Complete stroboscopy rating
data from rater 1 is given in Table 1.
When comparing the readings of the examinations within a
rater (assessing intra-rater reliability), rater 1 had higher kappa
statistics than rater 2. Rater 1 had substantial agreement for ery-
thema and lesions, moderate agreement for edema, supraglottic
posture and mucosal waveform, and fair agreement for closure
pattern. Rater 2 had moderate agreement for erythema and slight
or poor agreement for all others (Figure 1).
Acoustic data
There was a statistically significant change to lower frequency
in the rF0 (P = 0.047) pre- and postperformance. Although there
were changes in the glissando and chromatic ranges in the low
and high ends, these did not reach statistical significance. There
was also a trend toward increased jitter in the rF0 (P = 0.078)
(Table 2).
DISCUSSION
One is constantly reminded in a study such as this one of the
gulf between the summative art of a performance and the re-
ductionist lens of laryngoscopic examination and acoustical
measurement. It may never become possible to quantify or
measure the effects of emotion or interpretation on the micro-
anatomy of the singing voice, nor would such measures
necessarily be desirable or even useful. The dearth of literature
on the physiologic and acoustic effects of performance, however
defined, on the singing voice, regardless of genre, is surpris-
ing, given the frequency with which singers seek out care and
the window of insight that a highly trained, athletic singing voice
would seem to provide into general principles of managing
phonotrauma and high vocal demand.
The crux of the challenge may well be found in the gap
between isolated, deconstructed larynxes that can be easily studied
in the lab, and living, breathing, singing voices in humans on
stage, which are much more difficult study. The extensive and
innovative work of Titze and Jiang, through computer model-
ing and study of canine anatomy, has afforded insight into the
biomechanics of phonotrauma. An oft-cited study by Gray and
Titze in 1988 found in a canine model that hyperphonated vocal
folds caused sloughing of the superficial epithelial layers of the
vocal folds, as well as breakdown of the anchoring filaments in
the basement membrane zone.5 Indeed, histopathological inves-
tigations have generally established a model of injury and
remodeling that underlies the development of pathologies like
vocal fold nodules and polyps over time.6,7 But the temporal re-
lationships of these developments to a performance—what some
Paul E. Kwak et al Physiologic and Acoustic Effects of Opera Performance 117.e13

between the first and second viewing of each examination, the rating given in the second viewing is listed underneath in parentheses. At the time of viewing, these ratings were performed in random order with each rater blinded to the
Notes: When there was consistency, ie no difference in the rating given between the first and second viewing of each examination, only one rating is listed in the corresponding box. When there was a difference in the rating given
(normal)
(lateral)
SP Post
Normal

Normal

Normal

Normal
Lateral

Lateral
AP

AP (normal)
(lateral)
SP Pre
Normal

Normal

Normal

Normal
Lateral
(AP)
AP
(complete)

(complete)

(complete)
CP Post
Hourglass

Hourglass

Hourglass

Hourglass
Complete

Complete

Complete
(complete)

(complete)

FIGURE 1. Radar plot of the kappas for each measure for each rater.
Hourglass

Hourglass
CP Pre

Complete

Complete

Complete
Posterior

Irregular
chink

might be willing to classify as an acute episode of sustained

phonotrauma—remains unclear.
Diminished

Diminished

Diminished
(present)

This pilot study, then, endeavored to close the gap of knowl-

MW Post

Present

Present

Present

Present

edge about the physiologic and acoustic changes which may occur
in situations of high vocal demand. The specific scenario of vocal
demand chosen for this report was the performance of opera,
(diminished)

(diminished)

(diminished)

largely for the emphases in classical singing on breath support,

Diminished

minimizing of laryngeal tension, and production of multidimen-

MW Pre

sional sound. The pilot design of the study, with a small number
Present

Present

Present

Present

Present

Present

of subjects in a highly specific vocal endeavor, makes general-

izations premature and the achievement of statistical significance
elusive.
Yes (No)
Lesions

Indeed, the primary limitation of this study was sample size

Post

(N = 7), which provides <50% power to detect an effect size 1.0.

Yes

Yes

Yes

Yes
No

No

No adjustments were made for multiple hypothesis testing due

to the small sample size. This study was not randomized, and
Yes (No)
Lesions

confounding variables, such as singer demographics—including

Pre

Yes

Yes

Yes

Yes

gender, background, experience, and medical histories—were not

No

No

collected. Nonetheless, several findings are worth highlighting.

Abbreviations: CP, closure pattern; MW, mucosal wave; SP, supraglottic posture.
Moderate

Moderate

Moderate

When examining the acoustical data, it is equally interesting

Edema

(mild)
Post

to note the measures which did not achieve statistical signifi-

None

None

None

None

cance as those that did. For example, the reading F0 was 7.9 units
smaller postperformance and was statistically significant at the
(moderate)

0.05 level. However, the mean change in sung F0 was only 0.08
(severe)

unit smaller postperformance (P = 0.59). Similarly, the jitter in

Edema
Complete Ratings from the First Rater.

(mild)
Pre

the reading passage approached statistically significant change

None

None

None

Mild

Mild

Mild

Mild

but jitter in the sung tones did not. In fact, none of the acous-
tical measures in the sung tones approached statistically significant
change from the first to the second examination (P ≥ 0.17). It
(severe)
Erythema

Moderate

might be reasonable to hypothesize that these singers are better

Post

Severe
None

None
Mild

Mild

Mild

able to control their voice in singing, or even that they simply

are more vigilant about the quality of sound production when
identity of the examination.

singing.
(moderate)

(moderate)
Erythema

In similar fashion, the stroboscopy data are as compelling for

(severe)
Moderate

Moderate
Pre

what was seen as for what was not seen. Specifically, it is worth
None
TABLE 1.

noting that both raters observed lesions in the first examination

Mild

Mild

Mild

Mild

of the same three subjects. This is a point that bears repeating

in the literature; that is, elite singers who are healthy, sought after,
Pt
1

and successful, are not immune to vocal fold pathology. A recent

117.e14 Journal of Voice, Vol. 31, No. 1, 2017

TABLE 2.
Summary Statistics for Acoustic Data
Pre Post Delta P Value (Signed
Variable N Mean SD N Mean SD N Mean SD Rank Test)
readingF0 7 171.4 51.4 7 163.5 53.6 7 −7.9 7.6 0.047
glissandolow 7 124.5 39.1 7 113.9 33.2 7 −10.6 11.7 0.063
chromaticlow 7 119.8 37.2 7 109.3 26.1 7 −10.6 15.8 0.063
glissandohigh 7 1091.1 306.9 7 1078.8 369.4 7 −12.4 214.0 0.612
chromatichigh 7 1083.8 299.3 7 1024.9 252.3 7 −58.9 60.4 0.031
rF0 7 169.9 53.3 7 165.6 48.8 7 −4.3 8.8 0.156
rF0std 7 1.5 0.84 7 1.5 0.69 7 0.04 0.22 0.469
rF0jitt 7 0.51 0.28 7 0.68 0.35 7 0.17 0.18 0.078
rF0shdb 7 0.22 0.09 7 0.27 0.11 7 0.04 0.07 0.156
rF0nhr 7 0.13 0.01 7 0.13 0.02 7 0.004 0.01 0.469
rF0vti 7 0.06 0.02 7 0.06 0.02 7 0.01 0.01 0.234
rF0spi 7 5.5 3.7 7 5.7 3.4 7 0.26 2.1 1
sF0 7 319.1 176.0 7 319.0 177.6 7 −0.08 8.1 0.578
sF0std 7 9.1 5.4 7 9.6 6.6 7 0.47 1.8 0.688
sF0jitt 7 0.48 0.23 7 0.66 0.33 7 0.18 0.30 0.172
sF0shdb 7 0.21 0.09 7 0.21 0.07 7 −0.005 0.08 0.938
sF0nhr 7 0.23 0.16 7 0.21 0.15 7 −0.02 0.06 1
sF0vti 7 0.13 0.11 7 0.09 0.06 7 −0.04 0.07 0.109
sF0spi 7 3.8 1.6 7 5.8 4.6 7 1.9 3.4 0.219
sF0ftri 7 2.7 0.65 7 2.8 1.0 7 0.11 0.55 0.469
sF0atri 7 8.4 3.3 7 8.4 2.6 7 −0.07 1.7 0.938
Glissando range 7 966.7 297.8 7 964.9 353.8 7 −1.8 213.0 0.735
Chromatic range 7 963.9 279.7 7 915.6 235.5 7 −48.3 55.0 0.018
CPP 10s reading 7 7.8 0.62 7 7.8 0.85 7 0.02 0.66 0.866
CPP reading 2nd 7 8.6 0.58 7 8.9 1.0 7 0.26 0.95 0.237
CPP spoken rFo 7 14.9 1.9 7 14.4 2.3 7 −0.53 1.30 0.311
CPP sung rFo 7 11.4 2.6 7 10.9 2.4 7 −0.46 1.66 0.311
CSID spoken rFo 7 −2.64 6.0 7 −2.1 6.9 7 0.52 6.5 0.866
CSID sung rFo 7 18.5 9.1 7 19.4 8.2 7 0.94 8.99 0.237

Females
Pre Post Delta P Value (Signed
Variable N Mean SD N Mean SD N Mean SD Rank Test)
readingF0 4 209.7 12.6 4 204.5 16.2 4 −5.2 5.3 0.144
glissandolow 4 150.3 28.9 4 135.0 24.1 4 −15.3 13.1 0.144
chromaticlow 4 145.5 24.1 4 127.2 13.7 4 −18.3 17.5 0.068
glissandohigh 4 1234.5 324.6 4 1291.3 265.9 4 56.8 212.2 0.715
chromatichigh 4 1265.9 219.4 4 1168.7 177.3 4 −97.1 51.2 0.068
Glissando range 4 1084.2 343.8 4 1156.2 283.2 4 72.0 202.3 1
Chromatic range 4 1120.3 236.8 4 1041.5 185.4 4 −78.8 55.3 0.068

Males
Pre Post Delta P Value (Signed
Variable N Mean SD N Mean SD N Mean SD Rank Test)
readingF0 3 120.3 29.2 3 108.8 19.3 3 −11.5 9.9 0.109
glissandolow 3 90.0 15.1 3 85.7 18.6 3 −4.3 7.2 0.285
chromaticlow 3 85.5 13.8 3 85.3 15.7 3 −0.22 2.2 0.999
glissandohigh 3 900.0 169.3 3 795.5 304.4 3 −104.5 218.0 0.285
chromatichigh 3 843.6 205.4 3 833.1 217.5 1 −7.9 13.7 0.317
Glissando range 3 810.0 155.9 3 709.8 290.3 3 −100.3 222.0 0.285
Chromatic range 3 755.4 190.9 3 747.7 202.1 3 −7.7 11.7 0.109
Paul E. Kwak et al Physiologic and Acoustic Effects of Opera Performance
study observed a 30% incidence of vocal fold pathology in young
classical singers-in-training.8 In addition, no statistically signif-
icant differences were seen in pre- and postperformance ratings
of erythema, edema, supraglottic posture, or closure pattern. It
may be simplest to ascribe this to the small number of subjects
and the pilot study’s low power. One other hypothesis is worth
entertaining: that the time points of measurement being 6 weeks
apart may have facilitated recovery and resolution of any lesions
or abnormalities that may have been perceived on an immedi-
ate (within 24 hours) postperformance examination. It is
reasonable to suspect that while the vocal folds may sustain acute
“injury” in the course of an evening’s performance, over time,
given proper time between performances for rest and recovery,
normal forces of wound healing are recruited to achieve long-
term homeostasis. A study to examine immediate postperformance
changes and their effects is underway currently at this center.
These effects are most easily observed in the presence or
absence of such measures as erythema, edema, and lesions. But
this study and others recently provoke questions about assess-
ing what is “normal” on a stroboscopic examination. The raters
of this study assessed lateral or anterior-posterior compression
of the supraglottis during phonation as abnormal. But the notion
that beautiful sound can emerge from vocal folds that are not
pristine or which do demonstrate some supraglottic posturing
enlarges to the idea that there is likely an important involve-
ment of the supraglottis in the production of sound at the level
of the larynx. A recent multi-institutional study examined the
function of the supraglottic musculature in healthy singers and
found indeed that supraglottic involvement is common among
trained singers.9 This study bears out these findings, noting
anterior-posterior or lateral compressive supraglottic posture in
a majority of the subjects. Therefore, this pilot study high-
lights the challenges in assessing and understanding the effects
of strenuous vocal performance on the larynx when normative
values for typical data points have yet to be definitively
established.
It must also be noted that there was generally poor agree-
ment between the two raters of the stroboscopic examinations.
This is not an unprecedented finding; at least two notable studies
in recent years have demonstrated poor inter-rater and intra-
rater reliability when reviewing laryngeal examinations for benign
pathology. A study from the University of Chicago in 2011 dem-
onstrated less than 50% agreement of two fellowship-trained
laryngologists asked to identify benign vocal fold lesions.10 In
the current study, the one stroboscopic measure to achieve sta-
tistical significance was a change in the mucosal waveform as
evaluated by the first rater. Otherwise, none of the stroboscop-
ic measures achieved statistical significance. This finding, along
with the general disagreement between raters, is a vivid re-
minder of the challenges of accurate diagnosis, appropriate
counseling, and treatment recommendations.
If the study inadvertently confirmed the difficulty of agree-
ing on findings on stroboscopy, one might have hoped for more
definitive findings in acoustic data, but only one of the acous-
tic measures achieved statistically significant change between
examinations, and only a few more approached statistical sig-
nificance. A meta-analysis from 2009 found generally poor
117.e15
reliability and validity of acoustic data in assessing vocal
pathology.11 This study’s findings with regard to the acoustic data
add to these questions about the role of acoustic measurements
in evaluation of voice.
The results of this study thus crystallize the challenges at the
academic center of the care of the professional voice. How does
one study in a systematic way a process that is so inherently in-
dividual? A larger sample would, by statistical standards, increase
the power and reliability of the findings, but the central issue
of style, genre, and fach confounds the issue significantly. One
would suspect that the vocal coordination and technique of a lyric
coloratura soprano, versus a Wagnerian heldentenor, versus a Verdi
baritone—versus a country-western singer, a Broadway belter,
a lead singer in a heavy-metal band, and the list goes on—
would be different in ways that are at the very least likely to
contribute to the extent and characteristics of trauma to the vocal
folds, and to the extent of supraglottic involvement and kind of
breath control that contribute to the singing.
CONCLUSIONS
This pilot study highlights opportunities and challenges in the
ongoing study of the effects of demanding vocal performance
on physiologic and acoustic measures of voice. Highly trained
singers may have more control over their singing voice and be
better able to mask inefficiencies than in their speaking voices,
and may, as a result, demonstrate more evidence of trauma in
singing rather than spoken acoustic parameters. Statistically sig-
nificant differences in pre- and postperformance stroboscopic
measures were not identified, but the time points of each ex-
amination were distant, making it difficult to assess whether the
lack of change represents recovery and healing over time, or
whether it is simply a function of the study’s small cohort as
well as low inter- and intra-rater reliability. Further study is needed
of this relatively homogeneous cohort in larger numbers, with
data collected in narrower timeframes, to add subtlety and detail
to these preliminary findings.
Acknowledgments
The authors are grateful to Dr. Stephen King, Lynette S. Autrey
Professor of Voice and Chair of Vocal Studies at the Shepherd
School of Music at Rice University, and to Diane Zola, Direc-
tor of Artistic Administration, Houston Grand Opera; and Laura
Canning, Director of the Houston Grand Opera Studio.
REFERENCES
1. Wagner R, Ellis WA, ed. Richard Wagner’s Prose Works, Vol. 1. The
Art-Work of the Future, and Other Works. London: Kegan Paul, Trench,
Truber & Co, Ltd; 1982:69–213.
2. Mayerhoff RM, Guzman M, Jackson-Menaldi C, et al. Analysis of
supraglottic activity during vocalization in healthy singers. Laryngoscope.
2013;124:504–509.
3. Yanagisawa E, Estill J, Kmucha S, et al. The contribution of aryepiglottic
constriction to “ringing” voice quality: a videolaryngoscopic study with
acoustic analysis. J Voice. 1989;3:342–350.
4. Pershall K, Boone S. Supraglottic contribution to voice quality. J Voice.
1987;1:186–190.
5. Gray S, Titze I. Histologic investigation of hyperphonated canine vocal cords.
Ann Otol Rhinol Laryngol. 1988;97:381–388.
117.e16
6. Johns MM. Update on the etiology, diagnosis, and treatment of vocal fold
nodules, polyps, and cysts. Curr Opin Otolaryngol Head Neck Surg.
2003;11:456–461.
7. Altman KW. Vocal fold masses. Otolaryngol Clin N Am 2007;40:
1091–1108.
8. Kwak PE, Stasney CR, Hathway J, et al. Knowledge, experience,
and anxieties of young classical singers-in-training. J Voice. 2014;28:
191–195.
Journal of Voice, Vol. 31, No. 1, 2017
9. Mayerhoff RM, Guzman M, Jackson-Menaldi C, et al. Analysis of
supraglottic activity during vocalization in healthy signers. Laryngoscope.
2014;124:504–509.
10. Cipriani NA, Martin DE, Corey JP, et al. The clinicopathologic spectrum
of benign mass lesions of the vocal fold due to vocal abuse. Int J Surg Pathol.
2011;19:583–587.
11. Carding PN, Wilson JA, Mackenzie K, et al. Measuring voice outcomes:
state of the science review. J Laryngol Otol. 2009;123:823–829.