Annals of Otology. Rhinology & Laryngology 119(7):460-467.

2010 Annals Publishing Company. All rights reserved.

Vocal Function Exercises for Presbylaryngis:

A Multidimensional Assessment of Treatment Outcomes
Cara Sauder, MA; Nelson Roy, PhD; Kristine Tanner, PhD; Daniel R. Houtz, MA;
Marshall E. Smith, MD
Objectives: Presbylaryngis, or aging of the larynx, can adversely affect vocal function and quality of life in the elderly.
This preliminary investigation examined the effects of vocal function exercises, a physiologic voice therapy approach, as
a primary treatment for presbylaryngis.
Methods: Nine consecutive elderly patients with presbylaryngis (2 female, 7 male) underwent a 6-week course of voice
therapy employing vocal function exercises. Pretherapy-versus-posttherapy comparisons were made of self-ratings of
voice handicap and phonatory effort level, as well as auditory-perceptual voice assessments, acoustic analyses, and visual-perceptual evaluations of laryngeal images.
Results: After treatment, patients reported significant reductions on Voice Handicap Index scores, phonatory effort levels, and voice disorder severity. Blinded listeners rated the posttreatment voices as significantly less breathy and strained.
However, comparison of pretreatment and posttreatment maximum phonation times, acoustic measures, and laryngeal
images did not reveal significant changes.
Conclusions: These preliminary data suggest that vocal function exercises produce significant functional and perceptual
improvements in voice, and deserve further attention as a treatment for elderly patients with presbylaryngis.
Key Words: dysphonia, presbylaryngis, vocal function exercise, voice analysis.

INTRODUCTION
At presetit, the US Census Bureau estimates that
there are 36 million elderly persons in the United
States. It is projected that the number of older adults
will double by 2030 and will represent 20% of the
population. Voice disorders are relatively common
among the elderly, with 29% of noninstitutionalized
(ie, community-dwelling) seniors (over 65 years of
age) reporting a current voice disorder.' Whereas
age-related changes in pitch, pitch range, loudness,
and voice quality can alter quality of life and limit social interaction, older adults may also be at increased risk for voice disorders due to the possible
alteration of voice use patterns, the presence of vocal fold disorders (eg, carcinoma, Reinke's edema,
and paralysis), the development of systemic diseases
known to be associated with alterations in laryngeal
function and voice production (eg, stroke, respiratory disease, and arthritis), or degenerative changes
in the structure and function of the vocal fold mucosa, musculature, or peripheral nerve supply. When
specific underlying disease processes or vocal fold
disorders are excluded as possible explanations for

the presence of a voice disorder, the diagnosis of

presbylaryngis or "aging of the larynx" is frequently offered. Presbylaryngis implies that age-related
changes are primarily responsible for the observed
dysphonia. In treatment-seeking populations, however, the prevalence of presbylaryngis varies widely,
ranging from 4% to 30%,^'^ likely reflecting differences in diagnostic criteria, referral patterns, and patient populations. For instance. Woo et aP reviewed
the medical charts of 151 elderly patients with voice
complaints (age at least 60 years) and found that
only 6 patients had features of presbylaryngis as
defined by vocal fold bowing with breathy voice
characteristics. In contrast, Hagen et aP sampled 47
consecutive patients over the age of 60 years who
were referred to a specialty voice center, and found
that presbylaryngis was the most common cause of
voice complaints, accounting for 30% of cases.
Although prevalence estimates vary, there is near
consensus that bowing of the vocal folds is considered to be a hallmark of presbylaryngis. Additional
age-related changes in laryngeal appearance have
been reported, including concavity of vocal fold

From the University of Utah Voice Disorders Center, University Hospitals and Clinics (Sauder, Tanner, Houtz), and the Department
of Communication Sciences and Disorders (Roy) and the Division of Otolaryngology-Head and Neck Surgery (Smith), University of
Utah, Salt Lake City, Utah. This work was supported by a faculty creative grant from the College of Health, University of Utah.
Correspondence: Cara Sauder, MA, University of New Mexico Hospitals, Dept of Speech-Language Pathology, 2211 Lomas Blvd
NE, Albuquerque, NM 87106.
460

Sauder et al, Vocal Exercises for Presbylaryngis

medial edges during abduction or adduction with or

without a spindle-shaped glottal gap, prominent vocal processes, predominant open phase, and/or abnormal vibratory amplitudes,'*'^ Some of the changes in visual appearance and vibratory characteristics
associated with advanced age can be explained by
structural changes of the vocal folds observed in the
elderly. For example, Sato et al^ reported many agerelated changes in the appearance, density, and ratio of collagenous and reticular fibers. Butler et af
also reported age-related reductions in the distribution of hyaluronic acid in the extracellular matrix of
the lamina propria. These changes potentially affect
vocal fold vibratory characteristics and reduce vocal
efficiency. Immunocytochemical and histochemical
stains also identified relative and absolute age-related differences in type I and type II fibers of the
thyroarytenoid muscle. Type I fibers have slow contraction speeds but increased endurance, whereas
type II fibers have faster contraction rates but less
endurance. Age-related reductions in type II fibers
are consistent with the presumed reduction in muscle tone that might explain the appearance of vocal
fold bowing,^ Furthermore, electromyographic studies have confirmed age-related neurogenic changes,
as well.^ln addition, decreased lung elasticity and
respiratory strength might further degrade phonatory efficiency.'^ Therefore, it is likely that age-related systemic, neurogenic, and structural changes
all contribute to a decline in voice quality and vocal
function over time in the elderly.
Despite the high prevalence of voice disorders
among the elderly, and current knowledge regarding age-related changes that can adversely affect
the subsystems underlying phonatory function, relatively little is known regarding outcomes related to
treatments aimed at improving age-related vocal
dysfunction. Few well-controlled studies have specifically examined the effects of surgical or behavioral therapy in this patient population. Lu et al"
reported only modest functional improvements following surgical intervention (type I thyroplasty) in
patients with presbylaryngis. One alternative to surgical management would be to exercise the phonatory system to putatively strengthen muscles and
improve glottal closure and phonatory efficiency.
In this regard, vocal function exercises (VFEs) have
received attention as promising techniques to improve phonatory function across a variety of disordered and vocally normal populations. Vocal function exercises represent a series of systematic exercises designed ostensibly to strengthen and rebalance the laryngeal musculature, increase or improve
vocal fold adduction, and coordinate the subsystems
of voice production. In a previous study, an elderly

461

male cohort with presbylaryngis (n = 19) underwent

a 12-week course of VFEs,'*^ The investigators reported improvement in maximum phonation times
during VFE performance, as well as improvements
in laryngeal airflow characteristics, suggesting improved glottal sufficiency,'" Although the results of
this investigation are encouraging, further study is
warranted, using a multidimensional approach to
treatment outcomes assessment. From the above
study, it is unknown whether improvements in airflow characteristics translate into clinically relevant
improvements in voice-related handicap, vocal effort, voice quality, and/or changes in laryngeal appearance and function. From recent epidemiological data, it is clear that many older adults will experience age-related voice problems that will require
treatment. Therefore, this investigation used a multidimensional assessment approach to examine the
effects of VFEs as a primary treatment for presbylaryngis,
MATERIALS AND METHODS
All procedures were approved by the University of
Utah Institutional Review Board (IRB# 00018230).
Participants. Nine consecutive elderly (over 65
years of age) treatment-seeking patients (mean age,
76 years; SD, 6,9 years; range, 67 to 90 years; 2 femaie, 7 male) evaluated at The University of Utah
Voice Disorders Center who received a diagnosis of
presbylaryngis and completed a 6-week course of
VFEs were included in the study. Two additional
subjects consented to participate, but were unable
to complete a full course of therapy. One of these
subjects was lost to follow-up after a single treatment session, and the second withdrew before initiating treatment because of personal obligations. In
all cases, the diagnosis of presbylaryngis was made
by a multidisciplinary team that included an otolaryngologist and a speech-language pathologist (both
specializing in voice disorders). They performed a
comprehensive clinical evaluation including medical and voice history, auditory-perceptual evaluation, and rigid laryngostroboscopy as part of routine clinical care. The diagnosis of presbylaryngis
was based upon the presence of dysphonia within
the context of 1) no visible mucosal disease, evidence of paralysis or paresis, or other disorder sufficient to explain the dysphonia, and 2) concave medial vocal fold edges (ie, bowing) with incomplete
glottal closure or a dominant open phase observed
during phonation. Exclusion criteria were treatment
for presbylaryngis, a history of previous vocal fold
paralysis or paresis, or neurologic disease with the
potential to affect voice or speech (ie, cerebrovascular accident, tremor, Parkinson's disease). Subjects

462

Sauder et al. Vocal Exercises for Presbylaryngis

TABLE 1. VOCAL FUNCTION EXERCISE PROGRAM

Four specific exercises are performed 2 times each, twice daily (morning and night) for 6 weeks.
Exercise 1; "Warm up" - Sustain "ee" softly but with extreme forward focus.
Women on musical note F above middle C.
Men on musical note F below middle C.
Extreme forward focus "almost but not quite nasal."
Goal: Sustain "ee" for as long as you can.
Exercise 2: "Stretching" - Glide smoothly up from your lowest to your highest note on the word "knoll."
Goal: No voice breaks.
Exercise 3: "Contracting" - Glide downward from your highest to your lowest note on the word "knoll."
Goal: No voice breaks.
Exercise 4: "Power exercise" - Sustain the musical notes (C-D-E-F-G) for as long as possible on the word "knoll" without the "kn.'
Women start on middle C.
Men start 1 octave below middle C.
'
Focus on open pharynx and constricted lips, focus on the vibration.
Goal: Sustained "knoll" without the "kn" for as long as possible.
Important: Produce all sounds as softly as possible, but keep the voice "engaged."
No glottal attacks. Start the sound in the lips.
Make sure the tone is forward (think inverted megaphone shape).
Use good breath support throughout.
Track your progress each day, and use the CD recording to assist you.

with other significant medical or surgical histories

were not excluded from the study. None of the participants underwent surgery or reported changes in
pulmonary or medical status during the treatment
period.
Procedures. Two speech-language pathologists
specializing in voice disorders provided 6 weekly
60-minute sessions of behavioral voice therapy involving VFEs. Vocal function exercises, considered
a physiologic approach, were selected because they
ostensibly address many aspects of voice production, including laryngeal tension, breath support,
voice onset, and resonance attributes, that can be potentially affected in the elderly voice. These exercises, as described by Stemple et al'^ and Sabol et al,'^
are reputed to strengthen and rebalance the subsystems involved in voice production (ie, respiration,
phonation, and resonance) through a program of
systematic exercise. Those authors speculated that
by improving the strength, endurance, and coordination of the systems involved in voice production, the
exercises help rehabilitate the voice.'^,13 Although
the assumptions underlying the physiological bases
of the exercises have not been empirically validated,
the exercises have proven useful in improving and
enhancing selected aspects of vocal performance of
speakers with healthy voices,'^ singers,'-^ voice-disordered teachers,''* and elderly individuals.'o In the
VFE approach used in this investigation, 4 specific
exercises were practiced at home 2 times each, twice
daily for a period of 6 weeks. The exercises included
maximum vowel prolongations and pitch glides using specific pitch and phonetic contexts. All exer-

cises were produced as softly as possible, combined

with a forward placement of" the tone (ie, maximizing midfacial vibratory sensations). Compact discs
containing the exercises were provided to guide the
home practice sessions, and the patients recorded
their progress and compliance with the twice-daily practice sessions. The patients were instructed to
complete the VFEs, but were not prescribed specific
time requirements or limitations, as the duration of
practice was expected to vary with maximum phonation times. However, the total time required to
complete the twice-daily practice regimen was estimated to be less than 40 minutes daily. The 4 exercises and their salient procedural characteristics are
described in detail in Table 1. During therapy sessions, the exercises were reviewed, and facilitating
techniques used to elicit extreme forward focus and
instruction in breath support as it related to performance of VFEs were included to assist with accurate performance of the exercise regimen.
Data Acquisition and Analysis. To assess the effects of the VFE regimen in a multidimensional manner, we acquired multiple treatment outcomes measures as part of the clinical evaluation before and after the 6-week treatment period. In the following sections, the acquisition and analysis of each treatment
outcome measure are described.
Acoustic Analysis. Digital audio recordings were
acquired immediately before treatment and again 6
weeks after the initiation of treatment by means of
a research-quality recording system. The recordings
included the "Rainbow Passage,"'^ three 5-second
tokens of the sustained vowel lal, and maximum

Sauder et al, Vocal Exercises for Presbylaryngis

sustained phonation of the vowel /a/. To assess

changes in phonatory instability, we computed the
acoustic measures jitter, shimmer, and harmonicsto-noise ratio from the central 3 seconds of the second 5-second sustained /a/ token using the Multidimensional Voice Profile (MDVP version 3.1.1, KayPENTAX, Lincoln Park, New Jersey). These measures of phonatory instability have shown potential to discriminate between disordered and normal
voices, and are sensitive to improvements associated with treatment. Speaking fundamental frequency
was also calculated from the middle 2 sentences of
the Rainbow Passage with the MDVP. Age-related changes in fundamental frequency for both men
and women have been observed, with an increased
speaking fundamental frequency in elderly men and
a decreased speaking fundamental frequency in elderly women, although there is some variability in
these findings between studies.'^ Finally, maximum
phonation times were calculated from the acoustic
waveform. Increases in maximum phonation times
are thought to reflect improved glottal sufficiency
and vocal efficiency.
Auditory-Perceptual Ratings. Four speech-language pathology students who had completed graduate course work in voice disorder assessment and
management (with no reported hearing loss) rated
each patient's pretreatment and posttreatment voice
samples. The identical middle 3-second segment
used in the acoustic analysis, and the middle 2 sentences of the first paragraph of the Rainbow Passage,
were selected for listener judgments. All samples underwent amplitude normalization using Adobe Audition 2.0. Separate listening experiments were developed for the sustained vowel and Rainbow Passage voice samples. Pretreatment and posttreatment
voice samples for each patient were presented to the
listeners as a set, with the order randomized within
each pair. The listeners rated voice samples in a quiet laboratory at a self-selected comfortable loudness
level. The listening tasks were based on the method
described by Kreiman et al.'^-'^ Four sample pairs
were presented at the beginning of the session to orient the listeners to the range of severity and to the
listening task. Ten percent of the voice samples were
repeated at the end of each listening experiment for
later analysis of intrajudge reliability. The listeners
were asked to indicate the overall severity of each
sample by placing a mark on a 10-cm visual analog
scale, ranging from "normal voice" on the extreme
left of the scale to "profoundly abnormal voice"
on the extreme right. Similarly, the listeners were
asked to indicate breathiness on a second 10-cm visual analog scale ranging from "no breathiness" on
the extreme left and "profound breathiness" on the

463

extreme right, and to indicate strain on a third 10-cm

visual analog scale ranging from "no strain" on the
extreme left and "profound strain" on the extreme
right. The listeners were asked to label the first sample in the pair as "sample A" and the second sample
in the pair as "sample B."
Self-Ratings. To assess the effects of therapy, each
patient also completed the Voice Handicap Index
(VHI) before and after the 6-week treatment period.
The VHI is a statistically robust instrument designed
to assess the self-perceived psychosocial consequences of voice disorders.'^ This self-report inventory consists of 30 statements that evaluate a patient's judgment regarding the relative impact of his
or her voice disorder upon daily activities. The subjects rate each statement indicating how frequently
he or she has the experience on a 5-point equal-appearing interval scale whereby 0 = never, 1 = almost
never, 2 = sometimes, 3 = almost always, and 4 = always. The VHI generates a total score (ranging from
0 to 120) and 3 subscale scores: functional, physical,
and emotional. It has been psychometrically validated and shows strong internal consistency, reliability,
and test-retest stability. According to its authors, the
VHI can be used as a measure of the effectiveness
of specific treatment techniques and as a component
of functional outcomes measurement.'^
In addition to the VHI, the patients were asked
to self-assess their level of phonatory effort while
reading the Rainbow Passage. Before and after treatment, the patients rated their phonatory effort on a
scale from 0 to 3 with 0 = no effort, 1 = minimal effort, 2 = moderate effort, and 3 = maximal effort.
Finally, the patients were also asked to rate the severity of their present voice disorder wherein 0 = no
problem, 1 = mild problem, 2 = moderate problem,
and 3 = severe problem.
Visual-Perceptual Ratings. To assess laryngeal
changes possibly associated with treatment, we reviewed video samples obtained from rigid videolaryngostroboscopy performed before and after the
6-week treatment. Laryngostroboscopic examinations were accomplished with the KayPENTAX
70 rigid endoscope model 9106, Rhino-Laryngeal
Stroboscope (RLS) light source 9100, and a 3CCD
Toshiba Camera. Images were captured at a video
rate of 30 frames per second so that the time between 2 consecutive frames accounts for 33 ms.
Two speech-language pathologists with more
than 5 years of experience in voice disorders independently reviewed pretreatment and posttreatment stroboscopic samples of the same 60 consecutive frames (2.2 seconds) captured during sustained
phonation at comfortable pitch levels for each sub-

Sauder et al, Vocal Exercises for Presbylaryngis

464

TABLE 2. PRETREATMENT AND POSTTREATMENT

OBJECTIVE MEASURES OF PHONATORY FUNCTION
Pretreatment
Mean SD

Posttreatment
SD
Mean

Z*

P
0.14 2.35 -0.42 0.67
0.16 3.15
Harmonics-tonoise ratio
1.22 0.862 -0.30 0.77
1.28 1.23
Jitter (%)
3.84 2.41 -1.36 0.17
Shimmer (%)
4.98 2.27
17.78 5.35 18.62 5.81 -0.42 0.68
Maximum
phonation
time (s)
146.42 39.48 141.59 26.18 -0.53 0.60
Speaking
fundamental
frequency (Hz)
*From Wilcoxon matched-pairs test.

ject. The raters were asked to track, frame by frame,

the degree of glottal closure so as to calculate open
phase versus closed phase. The raters were permitted to review 60 consecutive frames up to 3 times for
each subject during both tasks. Change scores were
calculated for each rater (pretreatment and posttreatment) to determine whether there was an increase
in the duration of complete glottal closure during
sustained phonation following treatment. Second,
the raters were asked to track, frame by frame, the
degree of glottal closure so as to calculate opening
versus closing phases during these same 60 frames,
and the ratio of opening phase to closing phase. This
rating of opening versus closing phase was included
to increase sensitivity to possible vibratory changes,
particularly when complete glottal closure was not
observed during phonation.
To further assess changes in laryngeal appearance, a second rating task was accomplished by 2
blinded laryngologists who examined still images of
subjects' vocal folds obtained during maximal adduction. Pretreatment and posttreatment still images
were paired and presented as a set, but the order of
the pair was randomized. Written instructions were
provided to assist in the rating task. Ratings of the
still images along parameters of glottal gap size,
glottal closure pattern, and supraglottic hyperfunction (mediolateral and anterior-posterior) were accomplished with the Stroboscopic Evaluation Rating Form (SERF) for these parameters .^o The size of
the glottal gap was rated as none or no gap (0), mild
gap (1), moderate gap (2), or severe gap (3).
RESULTS
Statistical Analysis. To assess changes associated with treatment, both parametric and nonparametric statistical analysis procedures were undertaken,
given the nature of the data and the small sample
size. Wilcoxon signed ranks tests and paired samples /-tests were used as appropriate. If the results of

the parametric and nonparametric tests rendered opposite conclusions, both results were reported. Because of the preliminary and exploratory nature of
this clinical research study and the relatively small
number of participants, a more liberal alpha level of
0.10 was selected to minimize the likelihood of failing to identify clinically significant differences because of power limitations (ie, to minimize the type
II error rate) .2'
Acoustic Analysis. Inspection of Table 2 reveals
no statistically significant differences between pretreatment and posttreatment acoustic measures, including harmonics-to-noise ratio, jitter, and shimmer. In addition, there was no statistically significant change in fundamental frequency or maximum
phonation time after treatment (Table 2).
Auditory-Perceptual Ratings.B&foreexaminingthQ
results of listener ratings, we calculated and assessed
interrater and intrarater reliability estimates. The interrater reliabilities for overall severity, breathiness,
and strain during both sustained /a/ and reading of
the Rainbow Passage were evaluated with intraclass
correlation coefficients. For the overall severity parameter, an intraclass correlation coefficient of 0.97,
with a 95% confidence interval (CI) of 0.91 to 0.99
was observed. For the breathiness and strain parameters, intraclass correlation coefficients of 0.77 (95%
CI, 0.31 to 0.95) and 0.88 (95% CI, 0.88 to 0.99)
were observed, respectively. These estimates indicate
adequate interrater reliability. The intrarater reliability was assessed for the 10% repeated measures with
Pearson's product moment correlations (r). The intrarater correlations were 0.86 for overall severity, 0.92
for breathiness, and 0.91 for strain. All correlations
were statistically significant with a p value of 0.10 or
less and were considered evidence of acceptable intrarater reliability.
The results of the listener ratings of pretreatment
and posttreatment voice samples are displayed in
Table 3. For the Rainbow Passage, statistically significant reductions in perceived breathiness and
strain were observed after treatment, and ratings of
overall severity approached significance at the adjusted alpha level (p - 0.11). No statistically significant differences were observed for the pretreatment
and posttreatment perceptual ratings of sustained
vowels, although the perceptual ratings of strain approached significance (p = 0.11).
Self-Ratings. Significant reductions in the mean
VHI scores were observed across participants (p =
0.01; Table 4). The patient-based assessments of the
severity of their voice disorders also improved significantly after treatment, indicating milder levels of

Sauder et al, Vocal Exercises for Presbylaryngis

465

TABLE 3. PRETREATMENT AND POSTTREATMENT LISTENER RATINGS FOR RAINBOW PASSAGE

(MIDDLE TWO SENTENCES) AND SUSTAINED VOWEL PHONATION
Perceptual Parameter

Task

Rainbow Passage
Rainbow Passage
Rainbow Passage
Sustained lal
Sustained lal
Sustained lal

Overall quality
Breathiness
Strain
Overall quality
Breathiness
Strain
*Significant difference at p s 0.10 level.

Pretreatment
Mean
SD

Posttreatment
Mean
SD

32.17
27.00
29.39
39.00
23.10
39.15

22.56
15.45
20.97
36
24
33

dysfunction (p = 0.10). Similarly, the self-rated phonatory effort levels were also significantly reduced
after treatment, indicating reduced physical effort
associated with voice production after treatment (p
= 0.04).
Visual-Perceptual Ratings. Before examining the
results of visual-perceptual ratings of frame-by-frame
stroboscopic images, we calculated and assessed the
interrater reliability estimates. Interrater reliability
was evaluated with Pearson's correlation coefficient
(r). The results of these analyses show that there was
a statistically significant correlation between raters 1
and 2 for ratings of open, closed, opening, and closing for both pretreatment and posttreatment visualperceptual ratings. Correlations (r) between raters
for number of open, closed, opening, and closing
frames ranged from 0.784 to 0.935 (p < 0.05). Table
5 reveals no significant changes in duration of closed
phase, duration of closing phase, or ratio of opening
to closing phases.
Before examining the results of visual-perceptual
ratings of still images of the larynx obtained during
maximal adduction, we calculated and assessed the
interrater reliability estimates. Interrater reliability
was evaluated with Pearson's correlation coefficient
(r). The results of these analyses show that there was
a strong correlation between raters 1 and 2 for ratTABLE 4. PRETREATMENT AND POSTTREATMENT
SELF-RATINGS OF VOICE HANDICAP INDEX SCORE,
VOICE DISORDER SEVERITY, AND PHONATORY
FUNCTION
Measure

Pretreatment Posttreatment
Test
Mean SD Mean SD Statistic

Voice Handicap 39.11 21.35 23.44 19.54 = 3.27 0.01*

Index score
Voice disorder
1.7 0.71
1.1 0.93 z= 1.89 0.10*
severity
Phonatory effort 1.78 1.09 0.89 0.60 z = 2.53 0.04*
For voice disorder severity scale, 0 = no problem; 1 = mild problem; 2 = moderate problem; 3 = severe problem. For phonatory effort scale, 0 = no effort; 1 = mild effort; 2 = moderate effort; 3 =
maximal effort.
Significant difference at p s 0.10 level.

25.26
22.04
25.43
14.78
17.68
16.14

17.83
7.60
19.08
13.36
14.00
14.29

Z
-1.61
-1.72
-1.66
-1.24
-0.889
-1.60

P
0.11
0.09*
0.10*
0.21
0.37
0.11

ings for size of glottal gap (r = 0.98), glottal closure

pattern (r = 1.0), and medial compression of the supraglottis (r = 0.80; all correlations, p < 0.01), indicating acceptable interrater reliability. However, ratings of anterior-posterior supraglottic compression
were associated with weak interrater reliability (r =
0.458; p = 0.25). Comparisons of pretreatment-toposttreatment laryngeal changes are shown in Table
6, which reveals no significant changes in glottal
closure, glottal gap size, or supraglottic compression following treatment.
Age as Potential Covariate. To examine the possible influence of age on any of the observed treatment effects, we entered age as a covariate in a repeated-measures analysis of variance. There were
no significant interactions between age and VHI,
duration of open versus closed phase, duration of
opening versus closing phase, or perceptual improvement between pretreatment and posttreatment
measures when tested with the multivariate criterion
of Wilks's lambda.
DISCUSSION
This investigation evaluated the effects of a
6-week vocal exercise program as a primary treatment for presbylaryngis. The preliminary results appear to support the role of behavioral intervention
as a potentially effective treatment for age-related
dysphonia. All participants in the present study reported posttreatment improvement in voice-related
dysfunction or handicap, as evidenced by reduced
VHI scores. Furthermore, significant improvements
in self-assessed voice disorder severity and phonaTABLE 5. PRETREATMENT AND POSTTREATMENT
VIBRATORY DYNAMICS: CLOSED, CLOSING, AND
RATIO OF OPENING PHASE TO CLOSING PHASE
Pretreatment Posttreatment
Test
Mean SD Mean SD Statistic

% Closed
15.63 0.21 21.34 0.25 z = 0.26 0.20
% Closing phase 46.45 13.23 57.29 11.64 ? = 0.26 0.23
Opening-to1.01 0.24 0.81 0.38 i = l . l
0.30
closing ratio

466

Sauder et al, Vocal Exercises for Presbylaryngis

TABLE 6, PRETREATMENT AND POSTTREATMENT

VISUAL PERCEPTUAL CHANGES IN GLOTTAL GAP
SIZE, COMPLETE GLOTTAL CLOSURE, AND
SUPRAGLOTTIC COMPRESSION
Pretreatment Posttreatment
jg^f
Mean SD Mean SD Statistic p
Glottal gap size* 0,625 0,74 0,688 0,59 i = 0,26 0,80
% Complete
50
38
z = -0,58 0,56
glottal closure
Medial
1,3
1,5
1,2
1,1 / = -0,48 0,65
compressiont
Anterior-posterior
1,6 0,79
1,6
1,2 i = 0,00
1,0
compressiont
*Rated on scale of 0 = none, 1 = mild, 2 = moderate, 3 = severe,
tRated on scale of 0 = no compression to 5 = complete compression
using concentric circles, as indicated on Stroboscopy Evaluation
Rating Form,

tory effort levels were also observed after treatment.

Taken together, these findings provide particularly
compelling evidence for functional improvements
following treatment. In addition, these patientbased improvements were corroborated by blinded listeners who observed reduced breathiness and
strain during the reading task after treatment. The
acoustic analysis and perceptual ratings of the middle 3-second portion of the 5-second sustained /a/,
however, did not display any significant posttreatment improvement. The lack of a strong association between acoustic measures and patient-based
measures of voice disorder severity and functional
disability is well documented, especially in mildly
dysphonic patients ,^2 This lack of correlation between acoustic and patient-based measures typically reflects differences in the nature of the measures:
acoustic measures serve as an impairment-level assessment, whereas handicap-level measures such as
the VHI are sensitive to voice-related restrictions in
participation and activity, reflecting the experiences of the person who has the disorder. Although no
measurable improvements in perceptual or acoustic
measures were observed during sustained-phonation tasks, auditory-perceptual improvement during
more complex speech tasks (ie, reading), combined
with reductions in self-perceived handicap, effort,
and voice disorder severity, provides compelling
evidence of successful treatment outcomes.
The results of our investigation generally corroborate the positive findings reported by Gorman et al,'^
who also employed VFEs as a treatment for presbylaryngis. However, a lack of significant change
in visual-perceptual ratings of laryngeal dynamics
and maximum phonation time, presumed to reflect
phonatory sufficiency in our study, was somewhat
unexpected, considering the results of Gorman et
al,"^ who did observe increases in maximum phona-

tion time and improved aerodynamic measures after

VFEs, However, we did not use high-speed video
imaging or collect aerodynamic measures, and it is
likely that the ratings of laryngeal dynamics used in
this clinical study are less sensitive to improvement
in glottal sufficiency. Furthermore, there exist some
distinct methodological differences between the two
studies. Gorman et al'" measured maximum phonation time during VFEs, using the extreme forward
focus and the semi-occluded vocal tract posture required to perform VFEs, rather than during a sustained /a/, as in our study. Additionally, the amount
of therapy also varied between studies; the subjects
in the study of Gorman et al"^ underwent 12 sessions of therapy, as compared to the 6 in our study.
Finally, another possible explanation is that the reduction in breathiness and strain observed in our
study, in conjunction with reduced physical effort,
might reflect the combined effects of modifying respiratory and resonance characteristics, in addition
to attempts to improve glottal closure and phonatory
efficiency. These differences aside, our results of a
multidimensional assessment of treatment outcomes
appear to extend the positive phonatory effects of
VFEs in presbylaryngis beyond improvements in
aerodynamic function alone,
CONCLUSIONS
These preliminary findings suggest improvement
in functional disability, voice problem severity, effort level, breathiness, and strain following VFEs,
and provide both objective and patient-centered
outcomes that support voice therapy as a primary
treatment approach for patients with presbylaryngis.
In this small group of subjects, it appeared that improvements in perceptual and functional outcomes
were not necessarily related to discernible changes
in glottal closure, but were perhaps related to improved vocal efficiency resulting from simultaneous
altering of all phonatory subsystems (ie, phonation,
resonance, respiration).
Further investigation of VFEs using a larger number of elderly treatment-seeking subjects is warranted, along with direct comparisons of surgical procedures aimed solely at improving glottal closure,
such as collagen injection laryngoplasty, which has
been suggested as the treatment of choice for presbylaryngis if speech therapy is insufficient.^^ Additionally, future studies should include an alternative
or nontreatment control group to confirm that observed treatment effects are not merely related to
the reactive effects of clinical interaction. Finally,
although immediate treatment effects were encouraging, long-term follow-up data are needed to determine whether improvements can be sustained in

Sauder et al. Vocal Exercises for Presbylaryngis

the absence of treatment, and/or whether tapering

or long-term maintenance is required. These caveats notwithstanding, the preliminary results reported

467

here suggest that voice therapy should be considered

as a treatment option for patients with presbylaryngis.

REFERENCES
1. Roy N, Stemple J.Merrill RM, Thomas L. Epidemiology
of voice disorders in the elderly: preliminary findings. Laryngoscope 2007; 117:628-33.
2. Woo P, Casper J, Colton R, Brewer D. Dysphonia in the
aging: physiology versus disease. Laryngoscope 1992; 102:13944.
3. Hagen P, Lyons GD, Nuss DW. Dysphonia in the elderly:
diagnosis and management of age-related voice changes. South
MedJ 1996;89:204-7.
4. Honjo I, Isshiki N. Laryngoscopic and voice characteristics of aged persons. Arch Otolaryngol 1980; 106:149-50.
5. Linville SE. Vocal aging. San Diego, Calif: Singular Publishing Group, 2000.
6. Sato K, Hirano M, Nakashima T. Age-related changes of
collagenous fibers in the human vocal fold mucosa. Ann Otol
Rhinol Laryngol 2002; 111:15-20.
7. Butler JE, Hammond TH, Gray SD. Gender-related differences of hyaluronic acid distribution in the human vocal fold.
Laryngoscope 2001 ; 111:907-11.
8. Malmgren LT, Lovice DB, Kaufman MR. Age-related
changes in muscle fiber regeneration in the human thyroarytenoid
muscle. Arch Otolaryngol Head Neck Surg 2000; 126:851-6.
9. Luschei ES, Ramig LO, Baker KL, Smith ME. Discharge
characteristics of laryngeal single motor units during phonation
in young and older adults and in persons with Parkinson disease. J Neurophysiol l999;8l:2l3l-9.
10. Gorman S, Weinrich B, Lee L, Stemple JG. Aerodynamic
changes as a result of vocal function exercises in elderly men.
Laryngoscope 2008; 118:1900-3.
11. Lu FL, Gasiano RR, Lundy DS, Xue JW. Vocal evaluation of thyroplasty type I in the treatment of nonparalytic glottic
incompetence. Ann Otol Rhinol Laryngol 1998;107:l 13-9.
12. Stemple JG, Lee L, D'Amico B, Pickup B. Efficacy of
vocal function exercises as a method of improving voice pro-

duction. J Voice 1994;8:27I-89.

13. Sabol JW, Lee L, Stemple JG. The value of vocal function
exercises in the practice regimen of singers. J Voice I995;9:2736.
14. Roy N, Gray SD, Simon M, Dove H, Gorbin-Lewis K,
Stemple JG. An evaluation of the effects of two treatment approaches for teachers with voice disorders: a prospective randomized clinical trial. J Speech Lang Hear Res 2001;44:28696.
15. Fairbanks G. Voice and articulation drillbook. New York,
NY: Harper & Bros, 1960.
16. Ramig LA, Ringel RL. Effects of physiological aging on
selected acoustic characteristics of voice. J Speech Hear Res
1983;26:22-30.
17. Kreiman J, Gerratt BR, Kempster GB, Erman A, Berke
GS. Perceptual evaluation of voice quality: review, tutorial, and
a framework for future research. J Speech Hear Res 1993;36:21 40.
18. Kreiman J, Gerratt BR, Ito M. When and why listeners
disagree in voice quality assessment tasks. J Acoust Soc Am
2007; 122:2354-64.
19. Jacobson B, Johnson A, Grywalski G, et al. Voice Handicap Index: development and validation. Am J Speech Lang
Pathol I997;6:66-7O.
20. Poburka BJ. A new stroboscopy rating form. J Voice
1999; 13:403-13.
21. Pett MA. Nonparametric statistics for health care research: statistics for small samples and unusual distributions.
Thousand Oaks, Galif: Sage Publications, 1997.
22. Wheeler KM, Gollins SP, Sapienza GM. The relationship
between VHI scores and specific acoustic measures of mildly
disordered voice production. J Voice 2006;20:308-l7.
23. Remade M, Lawson G. Results with collagen injection
into the vocal folds for medialization. Gurr Opin Otolaryngol
Head Neck Surg 2007;15:148-52.

Copyright of Annals of Otology, Rhinology & Laryngology is the property of Annals Publishing Company and
its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's
express written permission. However, users may print, download, or email articles for individual use.