. I.
.5515‘
i 1.219..

warn. I...
13‘: fl
‘ ..

flu!
WWW!

 

all. 5%
A; lilrvffiilV 1'! 5A. -Iuh“
livn‘rfitp 7-, 001‘

.3}. [in-3:; {at
a .. w
.\.£t:l....‘t.11}
3!. {fiftiil
si/l. «. 1:50 .!
1|!!!W

 

 

_

 

—-i
f
c2

’35

9
1C,
(“J

?>

LIBRARY
Michigan State
University

This is to certify that the

dissertation entitled

ANALYSIS OF COMMUNICATION AND SWALLOWING
IN THE SLEEP APNEA POPULATION
presented by

Vicki S. Parker

has been accepted towards fulfillment
of the requirements for

Ph.D. degree in Audiology

and
Speech Sciences

he Ollie

Major professor

Date 12111/01

k

MS U is an Affirmative Action/Equal Opportunity Institution

0-12771

 

PLACE IN RETURN Box to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 cJCIRCJDateDuepSS-sz

I————- .._ — 7—.

ANALYSIS OF COMMUNICATION AND SWALLOWING IN THE SLEEP APNEA
POPULATION

By

Vicki S. Parker

A DISSERTATION
Submitted to

Michigan State University
In partial fulfillment of the requirements for the degree of

DOCTOR OF PHILOSOPHY

Department of Audiology and Speech Sciences

2001

ABSTRACT
ANALYSIS OF COMMUNICATION AND SWALLOWING IN THE SLEEP APNEA
POPULATION
By

Vicki S. Parker

This is a descriptive study of the communication and swallowing characteristic of
35 subjects diagnosed with sleep apnea. An additional component of this study is
comparison of a reference group on non-treated sleep apnea subjects with a group of
sleep apnea subjects receiving continuous positive air-pressure (CPAP) treatment for 1 to
3 months. The five areas of study were respiration during speech, speech production,
resonance, phonation/voice and swallowing. Information was gathered by the use of a
patient questionnaire, clinician perceptual judgment and objective clinical measures. The
subjects had high Body Mass Index (BMI) scores that may have had a negative effect on
communication. Communication and swallowing ratings based on the perceptual
judgment of a clinician revealed mild difficulties in (32) 94% of the subjects. Non-
normative ratings occurred for 65% of the subjects on speech production, 50% of the
subjects on voice, 47% of the subjects on resonance, and 9% had abnormally judged
swallowing. The objective data revealed that sleep apnea subjects had frequent abnormal
scores in the respiration during speech area that were sufficient inhalation, duration of
exhalation and maximum phonation time. There were also a high number of sleep apnea
subjects with low fundamental frequencies. Individuals with high BMI ratings and low

Apnea Hypopnea Index (AHI) ratings were all rated lower in the area of speech

production and had below normal maximum phonation times. In addition, self-reported
smokers all had below normal abilities in the respiratory tasks.

The reference and CPAP treatment groups showed the same trend between the
initial and follow—up data. There were no statistical significant results that differentiated
between the two groups. However, there was a tendency for improvement with the
CPAP group in the areas of resonance, voice and swallowing abilities perhaps related to
less edema in the upper airway after CPAP use.

In summary, the 35 sleep apnea subjects used in this study were obese and had
severe sleep apnea scores. Almost all of them were evaluated as outside of normal limits
in one or more components of communication. There was no significant change for those
subjects who went on a CPAP treatment protocol for 1 to 3 months compared to those

subjects who choose no regime of treatment.

To Mom and in memory of Dad, for always having the greatest faith in me.
May I Show that same faith in Sally.

And for David who has assisted me in realizing one of my dreams.

iv

ACKNOWLEDGMENTS

I would like to express very special words of gratitude to Dr. Cooke for his guidance,
honest feedback, patience, kindness, and sense of humor.
Thank you, Drs. Deal, LaPine and Marmion for assistance with this project.
Deep appreciation to Donna Gas for her assistance as a listener.
A special thanks to my sister, Laurie for emotional support.
Thank you Rae Zimmerman, my friend and mentor.
Thank you Ulrike Berzau for German translation.
Thank you Eileen Pierce and Rae Zimmerman for your reviews and feedback.
Thank you all for providing above and beyond childcare: Laurie and Terry Phillips;
Brock and Erika, Rene Vernick, Amy and Greg Peterson,

Melody Potter and Rae Zimmerman.

TABLE OF CONTENTS

LIST OF TABLES

CHAPTER 1

INTRODUCTION ................................................................................................... 1
Prevalence .................................................................................................... 4
Sleep Apnea and Weight .............................................................................. 4
Anatomical Relationships ............................................................................ 7
Normal Anatomy and Function ................................................................... 7
Management of Obstruction Sleep Apnea ................................................. 10
Communication and Swallowing Characteristics ...................................... 14
Systems Approach ..................................................................................... 19
Purpose ....................................................................................................... 23

CHAPTER 2

METHODS ............................................................................................................ 24
Subjects ...................................................................................................... 24
Equipment .................................................................................................. 26
Procedures .................................................................................................. 27
Data Analysis ............................................................................................. 32

CHAPTER 3

RESULTS .............................................................................................................. 34
Self-Perceptions ......................................................................................... 38
Clinician Perceptual Judgments ................................................................. 39
Oral Abilities .............................................................................................. 41
Respiratory Measures ................................................................................. 42
Resonance .................................................................................................. 44
Voice Parameters ....................................................................................... 45
Swallowing ................................................................................................ 46
Summary of Baseline Characteristics ........................................................ 46
Follow-up Procedure Results ..................................................................... 47
Self-Perception Follow-up ......................................................................... 48
Clinician Perceptual Judgment Follow-up ................................................. 49
Objective Data Initial versus Follow-up .................................................... 51
Oral Motor Follow-up ................................................................................ 51
Respiratory Follow-up ............................................................................... 54
Resonance Follow-up ................................................................................. 55
Voice Parameters Follow-up ...................................................................... 56
Swallowing Follow-up ............................................................................... 57
Summary of Follow-up Data ..................................................................... 58

vi

CHAPTER 4

DISCUSSION ........................................................................................................ 60
Limitations of Study .................................................................................. 67
Areas to be Explored in the Future ............................................................ 68

APPENDICIES
A Baseline Questionnaire .......................................................................... 72
B Follow-up Questionnaire ........................................................................ 73
C Emotional Communication Measure - Speech Production Disorder ..... 74
D Functional Communication Measure — Voice Disorder ........................ 75
E Resonance Scale ..................................................................................... 76
F Emotional Communication Measure — Ability to Swallow Function ....77
G Initial Raw Data ..................................................................................... 78
H Follow-up Raw Data .............................................................................. 79

REFERENCES ...................................................................................................... 81

vii

LIST OF TABLES

Table 1 — Physical Characteristics of Subjects ...................................................... 36

Table 2 — Percentage of Subjects Self-Reported Difficulties Across
Various Categories ....................................................................................... 39

Table 3 — Percentage of Subjects Rated Outside of the Normal Range
for Various Perceptual Categories ............................................................... 41

Table 4 — Percentage of Subjects Judged as Within or Outside of the
Normal Range by Combined Perceptual Categories (Speech Production,
Voice, Resonance, and Swallowing) .......................................................... 41

Table 5 — Distribution of Subjects on the Oral Speech Mechanism

Screening Examination — R (OSMSE — R) Data ......................................... 42
Table 6 - Descriptive Characteristics for Several Respiratory Parameters ........... 43
Table 7 — Nasalance Data Distribution .................................................................. 44
Table 8 — Fundamental Frequency Data ................................................................ 45

Table 9 - Percentage of Subjects Self—Reporting Difficulties Across
Various Categories at F ollow-up, ............................................................... 49

Table 10 - Percentage of Subjects as Within or Outside of the Normal Range
by Combined Perceptual Categories (Speech Production,
Voice, Resonance, and Swallowing) at Follow-up ...................................... 50

Table 11 — Percentage of Subjects Scoring a 6 or 7 Rating on the
Emotional Communication Measure (FCM-SP)—— Speech Production,
Initial and Follow-up Comparison ............................................................... 51

Table 12 - Objective Data Comparison of Initial and Follow-up Results ............. 52

Table 13 — Percentage of Subjects Scoring Below or at Norm or Better
on the Oral Speech Mechanism Screening Examination — R
(OSMSE - R), Initial and Follow-up Comparison Summary
of the Reference and CPAP Groups ............................................................. 53

Table 14 — Percentage of Subjects Scoring Within or Outside of Norm for
Maximum Phonation Time (MPT), Initial and
Follow-up Comparison ................................................................................ 54

viii

Table 15 — Nasalance Data Distribution, Initial and Follow-up Comparison ........ 55

Table 16 — Percentage of Subjects Within or Outside of Norm for
Fmdamental Frequency, Initial and F ollow-up ........................................... 57

Table 17 - Percentage of Subjects Within or Outside of Norm for
Jitter Perturbation, Initial and Follow-up ..................................................... 57

Table 18 — Percentage of Subjects Within or Outside of Norm for 3/2 Ratio,
Initial and F ollow-up .................................................................................... 57

Table 19 — Percentage of Subjects With Timely Swallows of Solids,
Initial and F ollow-up Comparison ............................................................... 58

ix

INTRODUCTION

Obstructive sleep apnea (OSA) is a respiratory disorder related to the collapse of
the upper airway during sleep. Snoring may be one of the early signs of sleep apnea.
Sleep apnea is defined as the cessation of breathing lasting 10 seconds or longer during
sleep. Hypopnea, a related disorder, is defined as a reduction of airflow causing arterial
oxygen desaturation greater than 4% (Reeves-Hoche, Meek, and Zwillich 1994; Sforza,
Addati, Cirignotta, and Lugaresi 1994). Sleep apnea is diagnosed by a study called
polysomnogram that provides information on brain wave activity, eye movements,
muscle tone, heart rate, oxygen saturation and respiration during a minimal six-hour
period. A confirmed diagnosis of sleep apnea is often given after the apnea-hypopnea
index (AHI) is computed, which is defined as the average number of apneas plus
hypopneas per hour of sleep. Hypopneas are a reduction of _>_ 30% reduction in airflow
or thoracoabdominal excursion accompanied by a 2 4% drop in oxyhemoglobin
saturation. The AHI provides information on the severity of the apnea. AHI scores of
0 — 4 are normal, 5 — 9 asymptomatic, 10 — 19 represents a mild apnea, 20 — 39 a
moderate apnea and any score above 40 is considered to be a severe apnea (L. Marmion,
personal communication, April 2001). However, Crofi and Golding-Wood (1990, p. 871)
point out "There is no miversally accepted grading system for snoring or apnoea."

Snoring and sleep apneas are both caused by a combination of anatomical and
functional problems. Snorers have anatomical and physiologic abnormalities of the
pharynx that distinguish them from non-snorers and snorers with obstructive sleep apnea

(Bradley et al., 1986). Drugs, alcohol, fatigue, and smoking all contribute to snoring and

sleep apnea by increasing muscle hypotonia and decreasing respiratory drive (Krespi,
Pearlman, and Keidar, 1994; Lugaresi, Cirignotta, and Montagna, 1988; Robinson,
White, and Zwillich, 1985).

A large number of individuals with obstructive sleep apnea present with structural
deviations that differentiate them from the normal population. Obstruction in the airway
may be caused by a multitude of anatomical or physiological reasons including
hypertrophic tonsils and adenoids, rhinitis, deviated nasal septum, upper airway edema,
vocal fold paralysis, airway stenosis, neoplastic disease, vocal cord dysfunction, goiter,
relapsing polychondrites and tracheomalacia. Anatomical characteristics that are
frequently noted in the OSA population include micrognathia — small jaw, retrognathia -
mder developed jaw, and macroglossia — large tongue (Shelton, Woodson, Gay, and
Suratt 1993). Ono, Lowe, Ferguson and Fleetham (1996) noted that craniofacial
structures have a major influence on jaw and tongue muscle function. Men may have a
greater prevalence of nasal obstruction secondary to contact sports and physical fighting.
On CAT Scan cross-sectional analysis from the hard palate to the tip of the epiglottis, a
group of 15 obstructive sleep apnea subjects demonstrated a reduction in the size of the
pharynx when compared to 10 normal controls (Stauffer, White, and Zwillich, 1990).

Emotional differences in size and shape caused by relaxation of the muscles of
the soft palate, uvula, and pharynx may result in more complex obstructive sleep apneas.
Contributing to the airway obstruction may be the difference between the closure shape
of the pharynx between normal subjects and subjects with sleep apnea. Rodenstein et al.
(1990) reported that the healthy subjects in their study had a coronal closure shape while

the apnea subjects had a sagittal or circular closure shape. Skolniok and Cohen (1989)

have described four velopharyngeal closure patterns: coronal, circular with Passavant's
ridge, sagittal, and circular. The coronal pattern is when there is opposition between the
soft palate and the posterior pharyngeal wall. Circular closure with Passavant's ridge is a
sphincter type closure with sofl palate opposition to a localized bulge of the posterior
pharyngeal wall. The sagittal closure results in greater medial movement of the lateral
pharyngeal walls. Equal movement between the velum, lateral and posterior pharyngeal
wall is observed with a circular closure pattern. The type of closure pattern an individual
exhibits can result in a need to modify the type of surgical reconstruction planned with
children who exhibit velopharyngeal insufficiency and may have implications for other
medical procedures that involve this mechanism.

Contributing to the upper airway collapse are over-generous amomts of tissue
and weakened muscle tone. A study by Yoshida, Thumm, and Siebert (1995) (cited in
Wiltfang et al., 1999) reported that the muscle tone of the geniohyoid muscles are
decreased in OSA patients. It is the genioglossus and geniohyoid muscles that are
involved in providing upper airway patency. In an effort to increase muscle tone,
Wiltfang et a1. (1991) used electrical stimulation on the suprahyoid muscle and fomd that
this stimulation resulted in less tongue collapse and apnea.

Individuals with airway obstruction have increased pharyngeal airflow resistance
when compared with control subjects (Stauffer et al., 1990; White, Lombard, Cadieux,
and Zwillich, 1985). Pharyngeal resistance is measured by assessing pressure between
the area of the choanae to the epiglottis. Some hypothesize that there is a delayed and
insufficient contraction of the pharyngeal dilator muscles during inspiration (Lugaresi et

al., 1988). Moreover, repeated heavy inspirations over time may change the individual's

anatomy by lengthening and narrowing the laryngo-tracheo-bronchial tree downwards
and shifting rib cage movement as well as contributing to changes in the tongue and jaw
anatomy (Lugaresi et al., 1988).
Prevalence

Phillips, Cook, Schmitt, and Berry (1989) estimated that the incidence of sleep
apnea for the general population in the United States is 1%. However, other studies
report the incidence for middle aged males in the United States to be as high as 4%,
whereas women have an OSA incidence of 2% (Yomg, Palta, Dempsey, Skatrud, Weber,
and Badr 1993). The likelihood of sleep apnea increases above age 50 years for both
men and women. Zamarron, Gude, Otero, Alvarez, Golpe, and Rodriguez (1999)
recently published some information on an older population that is thought to have a
greater incidence of sleep apnea. Zamarron and his colleagues fomd the prevalence of
Sleep apnea syndrome in 50 —- 70 year olds in Spain to be almost 7%. He and his
colleagues also cited a difference in the prevalence between men and women.
Sleep Apnea and Weight

Obese individuals have many chronic health problems. Sturm and Wells (2001)
conducted a telephone survey of 9585 adults and fomd that obese people have an average
of twice as many chronic health problems as those with normal weights. Sloan and
Shapiro (1995) define obesity as body weight of 20% greater than ideal on the Body
Mass Index (BMI). Body Mass Index (BMI) is a method of adjusting body weight of
individuals related to their height. The formula for BMI is: BMI kg/m2 = [(weight
(lb.)/height any] x 705 (Mahan and Escott-Stump, (Eds), 2000). In addition, the BMI

ranges are established ranging from malnutrition to obese. The ranges are severe

malnutrition (<16 kg/m2), moderate malnutrition (16.0 — 16.9 kg/m2), mild malnutrition
(17.0 — 18.4 kg/m2), normal (18.5 — 24.9 kg/m2), overweight (25 — 29.9 kg/m2), obese
(30 -— 34.9 kg/m2) and very obese (35+ kg/m2) (Nieto et al., 2000; Shils, Olson, and
Shike, 1994). Bray, Jordan, and Sims (1976) report there is increased risk of health
problems if the BMI is greater than 27.

The frequency of OSA has been documented to be 25 times higher in patients
who are obese compared to non-obese patients (Peiser, Lavie, Ovnat, and Charuzi 1984).
About 80% of individuals with obstructive sleep apnea are more than 20% or more
overweight; the fatty infiltration of the oropharyngeal musculature most likely reduces
efficiency (Lugaresi et al., 1988). Sloan and Shapiro (1995) reported on a group of 40
women who were consecutive referrals from an eating disorders clinic and were not
receiving assessments for sleep disturbances mtil participation in the study. They fomd
10% of these women had OSA. The 10% of the women subjects that had OSA also
tended to have a higher percentage of body fat and higher Body Mass Index scores than
did the non-apnea group.

Body fat distribution typically differs in men and women. Men typically carry
their body fat in the upper body while women typically carry their body fat in the lower
body (Sloan and Shapiro, 1995; Stradling and Crosby, 1991). Sloan and Shapiro (1995)
feel that this distribution of body fat may be more important than the amomt of obesity in
OSA patients. Rollheim, Osnos, and Miljeterg (1997) fomd that there was a correlation
between increasing BMI scores and anatomically lower pharyngeal obstructions.
Stradling and Crosby (1991) reported that the most significant predictors of OSA are

nocturnal hypoxemia and neck size. Nieto et a1. (2000) in a descriptive study of 6132

subjects reported that neck circumference and waist-to-hip ratio were higher in
participants with high AHI values.

Horner et a1. (1989) studied the sites and sizes of fat deposits using MRI aromd
the pharynx in weight-matched references and OSA patients. Their study reported large
fat deposits located posterior-lateral to the oropharyngeal space at the level of the soft
palate. The study revealed that there was a greater amomt of fat present with the OSA
subjects than with the references. Shelton et a1. (1993) measured volume of fatty tissue
in the upper airway and compared this to AHI severities. There was a positive correlation
between the volume of fatty tissue and number of apneas and hypopneas per hour. The
location of the fatty tissue was located laterally possibly explaining the frequent lateral
pharyngeal collapse. Two of the patients from this study who lost weight showed a
decrease in fatty tissue in the pharyngeal area and also had fewer apneas and hypopneas.

These studies provide evidence that when there is increased weight there may be
an increased amount of fat in the tongue and sofi tissue contributing to increased
collapse/obstruction of the airway during sleep. Increased weight could also decrease the
cross-sectional area or increase the muscle load needed to keep the airway area patent.
The changes in the airway due to weight may also affect the airway during waking hours.
Therefore, increased weight may affect many body systems including respiration, speech
production, voice, resonance, and swallowing. Increased weight is a common risk factor
with many chronic diseases such as pulmonary disease, cardiovascular disease, diabetes,

and cancer (Centers for Disease Control, 1999)

Anatomical Relationships

Respiration, communication and swallowing all involve interdependent anatomy
and complex motor planning that is dependent on the central nervous system. The
anatomy that serves these overlapping systems is the oral and nasal cavities and the
oropharynx (often referred to as the upper airway). It is important to mderstand normal
anatomy and function of the airway area during respiration, speech production,
phonation, and swallowing to get an appreciation for the integration of these systems. An
mderstanding of a normal model will also assist in observations of an abnormal system
and the need for corrective measures when appropriate. When there is an abnormality in
one system, the performance of another system may be affected. For example, an
appropriate valving of the oral air stream is needed to produce pressure consonants; when
there is insufficient valving, intelligibility to a listener maybe diminished. Individuals
with respiratory disorders may be at higher risk for speech deficits, voicing disorders
and/or swallowing difficulties because respiration supports these activities or is
coordinated with these tasks (Logemann, 1983; Sataloff, 1991). Changes by disease or
surgical revisions directed at improvement of function in one system might affect the
performance in another related system such as removal of tonsils to aid breathing or
swallowing may at times improve upon or hinder resonance.
Normal Anatomy and Function

The normal route of breathing during wakefulness is predominantly through the
nose (Sher, 1990). This route allows the air to be warmed and humidified. Inhalation
occurs when alveolar pressure within the lmgs is sufficiently decreased from outside

atmospheric pressure (Hixon, 1973). Exhalation occurs when the opposite conditions are

met. Airflow during both inhalation and exhalation proceeds from an area of high
pressure to an area of low pressure. Both passive non-muscular forces and active
muscular forces are involved in this function. The muscle activity assists in stiffening or
dilating the pharynx. In addition to sustaining life, breathing during wakefulness
supports speech and phonation as well as other activities.

During sleep many breathing modifications occur. (Ventilation is decreased
during sleep (Gleeson, Zwillich, Braier, and White 1986). Men more frequently change
the route of their breathing during sleep to increase the amomt of mouth breathing versus
nasal breathing (Gleeson et al., 1986). During sleep the skeletal muscles are generally
more flaccid, especially during rapid eye movement sleep, and may contribute to
pharyngeal collapse (Krespi et al., 1994; Wiegand and Zwillich, 1994). This pharyngeal
soft tissue collapse in part explains the greater pharyngeal airflow resistance and
decreased patency of the nasal-oral-pharyngeal area relative to other more fixed portions
of the respiratory system such as the nasal passage or trachea. Unfortunately, information
on the patenoy of the pharynx in normal individuals is limited. Patency is typically
determined by size of the pharynx from the hard palate to the tip of the epiglottis and by
pharyngeal and supraglottio resistance. Phillipson (1993) feels that patenoy in the upper
airway depends critically on the action of the dilator muscles.

The upper airway is supported by neurological signals that are involved in
maintaining patency and coordinating movement for breathing, speech and swallowing.
The pharyngeal plexus innervates the soft palate and upper pharyngeal muscles, which
are part of the velopharyngeal port. The pharyngeal plexus comprises the branches of the

glossopharyngeus and vagus cranial nerves. There are two exceptions to this

velopharyngeal mechanism innervation. The tensor is innervated by the mandibular
branch of the trigeminal nerve, and the musculus uvula is innervated by the spinal
accessory nerve and may receive some innervation from the palatine nerves of the facial
nerve (McWilliams, Morris, and Shelton 1990; Nishio, Matsuya, Machida, and Miyazaki
1976a; Nishio, Matsuya, Ibuki, and Miyazaki et al., 1976b).

During respiration the velopharyngeal port is open, forming a continuous passage
between the nasal and oral pharynx. The velopharyngeal musculature begins at the
posterior edge of the hard palate and extends to the posterior pharyngeal wall. The
mechanism consists of the soft palate, uvula, posterior pharyngeal wall and lateral
pharyngeal walls. The anterior faucial pillars are on the lateral borders and the tongue is
the inferior border. The palatoglossus muscles form the anterior faucial pillars. These
pillars assist in lowering the velum and retracting the tongue. The branches of the
palatopharyngeus muscle form the posterior faucial pillars extending from the posterior
pharyngeal wall to the soft palate. The palatopharyngeus acts to narrow the
velopharyngeal orifice and raise the larynx. The superior constrictor muscle forms the
lateral and posterior pharyngeal walls, which is responsible for medial movement of the
lateral walls and anterior movement of the posterior pharyngeal wall.

During speech the velopharyngeal mechanism works as one of the components of
articulation by changing the direction of airflow and providing shaping (Moon and
Kuehn, 1996), as it is open for nasal somds and closed for oral somds. This modulation
affects the shape of the vocal tract and, hence, the resonance of one's voice. The musculus

uvula is an intrinsic muscle. Cassell and Elkadi (1995) stated:

It has been suggested that the musculus uvulae assists velopharyngeal closure by
thickening during phonation, a result of its contraction, which shortens the
muscle, adding bulk to the point at which it overlies the levatores. The actual
contribution of the musculi uvulae to speech remains mcertain, however. (p 55)
During swallowing the velum lowers and bulges forward to control the transit of

the bolus (Logemann, 1983) while the faucial pillars (palatoglossus muscle) are thought
to be involved in the initiation of the swallow response (Logemann, 1983; Moon and
Kuehn, 1996). Finkelstein, Talmi, and Zohar (1988) reported that in addition to its role
in speech, the uvula’s function is related to closing off the nasal cavity from the
oropharynx while bending over and drinking, such as from a drinking fomtain. Later,
Finkelstein, Meshorer, Talmi, Zohar, Brenner, and Gal (1992) reported on an expanded
role of the uvula to produce saliva to lubricate the oropharynx and prevent dryness.

The nasal-oral-pharyngeal area is a highly vulnerable and, in part, collapsible
system. Maintenance of a patent nasal-oral-pharyngeal area depends on neurologic
control, air pressure and peripheral feedback. Intervention directed at the patenoy
problem results in the various management techniques used for OSA patients, including
lifestyle changes, medical non-surgical procedures, and surgery.

Management of Obstructive Sleep Apnea

Many treatments and strategies are used to help improve sleep apnea. The first
recommendations made are usually lifestyle changes such as weight loss, sleep position
modifications, meal scheduling, and smoking cessation (Lombard and Zwillich, 1985;
Wiegand and Zwillich, 1994). Compliance with these first-line recommendations has

been problematic. For some milder forms of sleep apnea, dental appliances that assist in

10

forward positioning of the tongue and jaw are options (Lombard and Zwillich, 1985).
However, such dental appliances are often discontinued because of discomfort.

The two most common treatment regimes for moderate and severe sleep apnea are
continuous positive air pressure (CPAP) or surgical intervention. CPAP therapy is the
most widely used treatment for OSA (Wiegand and Zwillich, 1994). The CPAP device is
a blower that delivers room air of a set pressure via a tube and nasal mask to the nasal
airway when worn by a patient. CPAP keeps the upper airway open by generating high
airflow and pharyngeal resistance. It is thought that nasal stimulation occurring during
CPAP therapy may result in reflex activation of the upper airway muscles and contribute
to increased patenoy (Lombard and Zwillich, 1985). A contrasting view is reported by
Strohl and Redline (1986) - they report that upper airway muscle activity is mohanged
during CPAP therapy. Anecdotal complaints by patients using this intervention usually
have to do with the noise of the CPAP machine and difficulty traveling with this type of
intervention. In addition, there are some complaints of dryness in the mouth area.
Lombard and Zwillich (1985) stated that dry nasal mucosa, sinusitis, middle ear
infections, pneumothorax and pneumomediastinum are potential problems arising from
CPAP use; however, there are no documented studies on these possible side-effects.
Lombard and Zwillich (1985) stated that “Medical therapy for the OSA (obstructive sleep
apnea) patient is unlikely to be beneficial if surgically correctable anatomic abnormalities
of the upper airway are present.” In other words, CPAP therapy may not be as effective
if anatomical problems that obstruct airway access such as a deviated nasal septum or

enlarged tonsils are not taken care of first prior to the initiation of CPAP.

11

Surgical remediation typically is a combination of tonsillectomy and/or
resectioning of the nasal turbinates or septoplasty and the uvulopalatopharyngoplasty
(U PPP) procedure. The UPPP procedure trims tissue from the uvula and posterior soft
palate. The goal of these procedures is to remove the obstruction in the nasal and/or
pharyngeal area and improve the free flow of air in the upper respiratory system. The
reviews on the effectiveness of surgery for treating OSA have been disappointing (Janson
et al., 1997; Troell, Riley, Powell, and Li, 1998). There are a number of studies that
report on the commmication or swallowing characteristics post-surgery of the UPPP
patients. Interestingly, most studies determine whether afier UPPP surgery there are
changes in communication or swallowing. Yet the post-surgery condition of the
pharyngeal area may be similar to the normal airway of the general population. In fact,
the pre-surgical or mtreated OSA population to begin with was not similar to a normal
population because of the large amomt of airway obstruction they experience.

There are a number of reported changes in communication and swallowing after
UPPP surgery. Most of the studies reporting these changes (Croft and Golding-Wood,
1990; Haavisto and Suopaa, 1994; Prichard, Marshall, Ahmed, Thomas, and Hanning,
1994) are based on questionnaires that reported on occurrences after surgery or that ask
questions about communication and swallowing only afier surgery. A study by Murry
and Bone (1989) did compare the OSA subjects’ pre-surgery results from specific speech
tasks to post-surgery results. They also compared the results obtained at both the initial
and final sessions to published norms in the literature from the normal population. They
fomd that the length of time it took for their sleep apnea subjects to read a passage prior

to surgery was below norm for all the subjects, whereas after the UPPP surgery all of

12

their subjects had improved reading rates. The UPPP procedure and its effects on
communication appears to have been more strongly researched than the abilities of the
OSA population without intervention or with the more widely used intervention of
CPAP. Results from the combined studies reveal relationships to communication and
swallowing. Complications from the UPPP surgery include velopharyngeal
incompetence (Croft and Golding-Wood, 1990; Sher, Thorpy, Shprintzen, Spielman,
Burack, and McGregor, 1985), pharyngeal dryness (Croft and Golding-Wood, 1990;
Haavisto and Suonpaa, 1994; Salas-Provance and Kuehn, 1990), loss of taste (Haavisto
and Suonpaa, 1994; Croft and Golding-Wood, 1990), nasal regurgitation of fluids
(Haavisto and Suonpaa, 1994; Prichard et al., 1994; Sher et al., 1985; Zohar, Finkelstein,
Talmi, and Bar-Ilan, 1991), swallowing dysfimction (Salas-Provance and Kuehn, 1990),
tongue numbness (Fairbanks, 1990), middle ear dysfunction (Zohar et al., 1991),
sneezing (Zohar et al., 1991), speech/voice changes (Haavisto and Suonpaa, 1994; Murry
and Bone, 1989, Prichard et al., 1994; Zohar et al., 1991), difficulty singing (Haavisto
and Suonpaa, 1994), and difficulty playing a wind instrument (Haavisto and Suonpaa,
1994)

One reported type of speech change resulting from UPPP surgery was change in
pronmciation. Zohar et a1. (1991) commented that 10 individuals in their study of 71
subjects afier UPPP had difficulty with pharyngeal productions especially guttural /ch/
and /r/. These authors noted that the /r/ and /ch/ were changed to alveolar or dental
productions. Nakai, Sakakura, Takahashi, Sadaoka, and Kakitsuba, (1996) examined

monosyllabic speech productions of speakers pre- and post-UPPP surgery and fomd no

13

significant differences in any of their productions. As noted above, reading rates
improved for subjects following UPPP surgery (Murry and Bone, 1989).

Vocal quality was also reported as changed by some UPPP patients. On
perception tasks listener-judges were able to tell the difference between the reference
subjects versus the UPPP subjects particularly with judgments related to phonation and
articulation (Salas-Provance and Kuehn, 1990). Instrumental analysis of resonance has
described little change by actual aerodynamic measurement (Salas-Provance and Kuehn,
1990)

Swallowing has also been reported as changed following the UPPP procedure.
Levring-Jaghagen, Nilsson, and Isberg, (1999) studied individuals post-UPPP surgery.
They fomd that after the UPPP procedure 29% of the subjects had complaints of
persistent dysphagia, and upon follow-up with videofluoroscopy the researchers fomd
pharyngeal difficulties in 71% of these individuals with complaints.

There have been a handful of studies on communication and swallowing abilities
after surgical remediation of OSA. However, most of these have explored UPPP surgery;
the more frequent remediation technique, CPAP, has fewer studies on the effects of this
therapy on commmication and swallowing abilities. Brander, Soirinsuo, and Lohela,
(1999) reported that OSA patients tend to describe nasopharyngeal problems such as
dryness, sneezing, mucus in the throat, blocked nose and rhinorrhea. These complaints
are noted prior to treatment and increase afier CPAP use.

Communication and Swallowing Characteristics
Many of the communication and swallowing characteristics that could provide

some additional information on this patient population have yet to be studied. One of the

14

important characteristics to evaluate with this population is respiration. The link between

speech production and respiration is described well by Hixon (1973):
To be more specific, the respiratory pump participates in speech by displacing
structures, creating pressures behind valves, and generating flows through
constrictions within the larynx and upper airways. These activities, in association
with intricate and rapid maneuvers of other parts of the speech apparatus, create
the disturbances of air that constitute speech at the acoustical level. Since
respiratory forces provide the basic energy source for all speech and voice
production in the English language, the events of speech respiration are of
fundamental importance in any accomt of oral communication. Within the broad
spectrum of physiological function in speech, the respiratory pump is involved in
the regulation of such important parameters as speech and voice intensity

(loudness), vocal fundamental frequency (pitch), linguistic stress (emphasis), and
the division of speech into various mits (syllables, words, phrases, etc.). (pp. 98 -

99)

Individuals with respiratory disorders are often fomd to have traumatized and
swollen pharyngeal tissue that may also correspond to voice and swallowing changes.
Respiratory measures can help determine the severity of a speech and/or voice disorder,
monitor change and assist in treatment planning. Measures of maximum phonation time,
breath groups per minute during speech production, sufficient inhalation gauged by chest
wall movement, and average duration of exhalation, are important measures to provide
descriptive objective information regarding respiratory support for speech. Maximum
phonation time assists in determining effectiveness of breath support for speech. Colton
and Casper (1990) also feel that maximum phonation time has a role in estimating glottal
efficiency and laryngeal control. Breath groups per minute during reading and picture
description tasks provide information on glottic efficiency and respiratory use.
Sufficiency of inhalation provides information on gross adequacy of respiratory support
for speech production. In addition, the individual's average length of exhalation while

Speaking assists in observing trends of efficient use and the reserves that an individual

15

may be able to call upon. The obstructive sleep apnea population may have smaller
respiratory capacities to support dynamic functions such as speech, voice and swallowing
coordination.

Investigations of oral motor movements and swallowing in the obstructive sleep
apnea population can improve our mderstanding of subtle changes in the anatomy and
function of the respiratory and speech mechanism. Ruscello, Tekieli, and Van Sickels,
(1985) studied speech production before and after orthognathic surgery and fomd that
when collapsing data from multiple studies, 72% of the individuals had articulation errors
before surgery. However, after surgery 88% of the subjects with pro-operative errors had
either reduced or eliminated articulation errors.

Speech production is a combination of articulatory positioning, neurological
motor programming and coordination with the airstream. Perceptual evaluation of
articulation can provide inferences about anatomical structure and function. Dental
occlusions, poor tongue position, or weakened muscles can result in distorted speech
somds. Poor respiratory support can result in consonants with weak pressure,
diminishing intelligibility.

Speech production is usually assessed by having the individual produce a series of
words with a variety of consonants targeted or by taking a sample of speech produced
during reading, talking about a picture, or conversation. Speech production is judged by
listener perception. Standardized oral motor examinations and swallowing trials provide
additional insight into the patient self-reports. Oral agility measures may help provide

some additional information related to articulation and swallowing changes.

16

Phonation quality and characteristics of phonation are products of the laryngeal
mechanism. These contribute to our knowledge about health, flexibility and the
efficiency of the respiratory, communication, and swallowing systems. The size, shape,
mucosa] quality of oral, nasal, and pharyngeal cavities, the function of the glottis, and
respiratory support for phonation determine the individual's vocal quality. Vocal abuse
or misuse, such as using an abnormally low pitch, can result in laryngeal hyperfmction
and decreased respiratory efficiency. After a more global perceptual assessment of
phonation is made, specific tasks that assist in describing the voice and inferring structure
and function are typically performed. For example, to better mderstand normal or
abnormal laryngeal vocal quality, the acoustic measures of fundamental frequency and
jitter perturbation can be analyzed. Jitter perturbation is the cycle-to-cycle variability of
firndamental frequency (Horii, 1979). Perturbation measures provide information about
the stability of the vocal fold vibratory pattern. Gilbert (1975) reported that jitter
perturbation is sufficiently sensitive to pathological changes in the phonatory process and
perhaps even to severe respiratory insufficiency. In addition, variations of the vocal tract
configuration such as the production of different vowels result in changes of jitter
perturbation. Excessive jitter perturbation is an acoustical physical correlate for
hoarseness. Another task that probes for laryngeal function is the s/z ratio, a gross
measure of laryngeal pathology. The cognates voiceless /s/ and voiced /z/ are produced
at the longest duration possible and the ratio reflects the influence of laryngeal valving.

Another characteristic that needs to be studied more in the OSA population is
resonance. Resonance is the quality that is produced by sound waves caused by air flow

vibrating in the nasal, oral, pharyngeal and laryngeal cavities. Here, a unique quality is

17

produced based on the size and shape of the cavity. Movable parts of the pharynx,
pyriform sinus, chest, trachea, tongue, jaw, soft palate, and lips are involved in changing
resonance. There is a range of this quality that is perceptually pleasing to our car.
Resonance is judged perceptually by listening and can be quantified by physical measures
such as nasalance, a ratio of oral versus oral-and-nasal acoustic energy. If resonance
occurs outside a normal range, it is considered to be either denasal/hyponasal, or
conversely, hypemasal. Deviant resonance is often perceived to be distracting to the
listener and may result in decreased intelligibility of the speech signal. For example, an
individual with a retracted tongue and little mouth opening may somd hypemasal with a
tense, pinched somding resonant quality. Enlarged tonsils may result in snoring,
obstructive sleep apnea, and denasal vocal quality. Denasal children often mdergo
removal of their tonsils for remediation. After tonsillectomy and/or adenoidectomy some
individuals experience transient hypemasality, whereas others experience a more
permanent increase in hypemasality as a complication of the surgery which enlarges the
oropharyngeal area.

Changes to the structure or function of the respiratory mechanism may lead to
changes in speech production, voice, resonance and swallowing. Monoson and Fox
(1987) used 39 age-matched subjects divided into three groups: OSA, chronic obstructive
pulmonary disease (COPD) and subjects’ without OSA or COPD. They assessed speech
production in these three groups by having the subjects read sentences and then having
trained listeners make binary judgments (disorder present or absent) of the subject’s
speech productions. They fomd abnormal speech in 62% of their sleep apnea group, in

23% of their COPD subjects, and in just 8% of their reference group. The clinicians that

18

rated the speech samples commented on the dysarthric-like speech with abnormal
resonance of the sleep apnea subjects.

There are studies present in the literature that reflect that changes in the oral
cavity through disease or intervention do change communication and swallowing.
Ichikawai, Komado, Horiucki, and Matsumoto, (1995) fomd that both voice onset time
(time between the release of the stop consonant and the voicing of the vowel) and
consonant duration increased with palatal augmentation, possibly due to changes in oro-
sensory feedback, laryngeal compression or indirect aerodynamic feedback. Siefert,
Ronte, Riebandt, Lamprecht-Dinnessen, and Bollmann, (1999) fomd that alterations in
the oral cavity from a dental prosthesis caused perceptual changes in voice; however,
fundamental frequency was not changed at a statistically significant level. More directly,
a study by Zohar, Grusko, Sulkes, and Melloul, (1998) fomd that during swallows, as
documented by scintigraphy, the OSA subjects had longer oral transit times than both
healthy subjects or OSA subjects following UPPP intervention.

Systems Approach

It appears that the sleep apnea group may already be at higher risk for various
communication and swallowing disorders. The increased risk may relate to the abmdant
pharyngeal tissue, crowded oral cavities, decreased neurologic function, or a combination
of any of these deviancies. Likewise, the interventions used to help sleep apnea by
decreasing the amomt of pharyngeal tissue or having air pressure forced into the airway
may also create changes in communication and swallowing abilities.

Noninvasive measurement may be helpful for early detection of neurologic or

muscular changes. chtions of the upper airway are dependent on a sequence of

19

volitional and reflexive neural control. Although coordinated functions exist between the
shared structures for respiration, speech, resonance, phonation, and swallowing, it should
be appreciated that differences in the motor control and peripheral feedback occur for
these functions. In both speech and swallowing, pharyngeal muscle contraction is present
for vocal tract posturing to produce various somds and resonance patterns or to propel
the bolus. Tongue strength, range and coordination can affect articulatory precision,
nasality of voice (by directing the airflow), vocal quality (by adding tension to the
muscles of the pharynx and larynx), and swallowing (by initiating the swallow in
response to tongue movement). There is a need to interface clinical and instrumental
findings of respiration, resonance, phonation, oral mechanism movements and
swallowing.

One example of how a functional change in one system can result in an
unintended change in a second area can be fomd in examining the relationship between
dysphagia and dysarthria. Consideration of long-held insights regarding the shared
symptomology for dysarthria and dysphagia relates to issues that may gain further
enlightenment when examined from the perspective of obstructive sleep apnea research.
Robbins (1985) discusses a significant link between individuals who present with
dysarthric speech and have symptoms of a swallowing disorder. She identified this link
as the coordination of respiration. Respiration stops momentarily to allow for a swallow
to occur and at the same time protect the nasopharynx and airway. Respiration
coordination is required for speech production, enough pressure to produce somds and

put together a string of words. When breath support is reduced additional stress may be

20

put on these systems which are dependent on fine timing and coordination of muscular
movements.

On the other hand, an example of how a structural change can result in a change
of function in one system and an mexpected change of function in a second can be fomd
in the relationship between airflow and voice. Children with borderline velopharyngeal
inadequacy are more likely to have vocal nodules (McWilliams, Morris, and Shelton,
1969) because they have a routinely increased demand on the respiratory system for
pressure. Hyperfunction of the vocal folds occurs to create adequate power (i.e., voice
intensity) to overcome the open chamber and decreased resistance in the upper pharynx.
There are many vocally abusive behaviors that result in malfirnctioning or organic
changes in the laryngeal mechanism. Years of heavy snoring could result in repetitive
vocal abuse to the laryngeal mechanism as the individual has sudden choking or gasping
at the end of the snore cycle. Similarly throat clearing is considered vocally abusive to
the system as the individual abruptly closes his or her vocal folds together. Further vocal
hyperfunction may be more likely to occur with the tongue collapsed during sleep by
increasing mass to the pharynx and possibly adding tension to the laryngeal mechanism.

The relationship among multiple systems seems obvious when noting that
changes in loudness, pitch, and syllable emphasis effects subglottic pressure. Voice
therapy utilizes breathing exercises and efficient use of respiration to improve laryngeal
vocal quality and resonance. Articulation also affects subglottic pressure by changing
airflow and resistance as well as having an effect on resonance. Brancewicz and Reich
(1989) fomd perceptions of nasality increased in normal speakers when the rate of

speech decreased, possibly related to changes in nasal airflow. Therefore, if respiration is

21

significantly compromised, it seems intuitive that laryngeal quality, resonance and speech
production may also be adversely effected.

During an initial clinical evaluation for snoring and obstructive sleep apnea, an
oral examination of structure and function is conducted to learn how the characteristics of
the mouth may have an effect on snoring. Evaluation may reveal a small jaw or large
tongue for the oral cavity, redmdant faucial arches and tonsillar tissue, and a reduced
distance between the velum and posterior pharyngeal walls. The above traits are risk
factors for snoring and obstructive sleep apnea. However, a similar evaluation of the oral
mechanism assists in providing information about articulation and intelligibility. The
tongue carriage may be too far back so that there are distortions in some phoneme
productions or a throaty tone may be present. The oral cavity may be small or
disproportionally sized and result in a muffled tone. Enlarged tonsillar tissue can inhibit
velar and/or tongue movement, contribute to making the individual somd denasal and/or
may be responsible for containing food in the oral cavity during swallowing. The mouth
may be so crowded that airflow is directed out of the nose or there is decreased tongue
propulsion during the initiation of the swallow. Obstruction in the oropharyngeal area
could also cause anterior displacement of the tongue and interfere with chewing and

swallowing.

22

Purpose

A detailed study of the communication and swallowing abilities in the obstructive
sleep apnea population is missing from the clinical literature. Moreover, there has been
no clinically well-documented outcome data profiling whether the most recommended
therapy technique for sleep apnea, CPAP, may effect changes in respiration during
speech, speech production, phonation, resonance, and swallowing. The first part of this
study provides descriptive information regarding sleep apnea subjects’ characteristics in
the areas of contextual speech, oral motor abilities, respiration for speech,
phonation/voice, resonance and swallowing. It is important to investigate patterns and
characteristics across multiple dimensions that may vary within a diagnostic group that
specifically deviates from the normal population. A wide net of clinical tasks was
developed to gain information on the areas that might be most affected with changes in a
normal upper airway system.

The second part of this study compares the commmication and swallowing
results of two different sleep apnea groups; those that did not receive treatment (reference
group) and those who received CPAP treatment. Multiple task analysis can help us learn
about how connected or independent the systems may be. The null hypothesis of the
experimental protocol was as follows: there will not be a significant difference between

the non-treatment (reference group) and CPAP treatment groups afier CPAP therapy.

23

Chapter 2
METHODS

Typically the investigations into sleep or respiratory disorders, dysarthria, voice,
or dysphagia are narrowly focused with the investigators looking at variables within a
single area of clinical focus or contrasting and comparing one area to a second (e. g.
dysarthria/dysphagia, apnea/respiratory function). Broader conclusions to appreciate the
larger linkages among these areas (i.e. respiration during speech, speech production,
resonance, phonation, swallowing) may be better grasped by examining the
interdependence between structure and their changes affecting functional equilibrium.
Consequently, permutations of change may be set up to influence systems that are beyond
the scope of a narrow study. Models which allow recognition and accomting for this
linkage may be valuable in making suitable predictions when intervention is
contemplated in any single dimension.

The characteristics to be studied--respiration during speech, speech production,
voice, resonance and swallowinguare outlined below. Self-reported questionnaires,
clinical and instrumental measures of assessment were used to measure these
communication and swallowing characteristics in an attempt to evaluate structural and
functional abilities and deficits. The clinical measures chosen were selected because they
are used in Speech pathology departments in academic and medical facilities.

Subjects

The medical director at a mid-western Sleep Lab referred potential sleep apnea

patients to the Speech - Language Pathology department for the purpose of this study.

The primary investigator would schedule a week per month at the sleep lab to collect data

24

from subjects prior to their scheduled overnight sleep study. The potential subjects were
then called and asked if they would be willing to participate in a research study prior to
their overnight polysomnographic test at the Sleep Lab. All subjects gave written consent
to participate in this study. The subjects participated in the initial part of the study prior
to their sleep apnea test and the follow-up component after the diagnosis was confirmed.
The subjects in this study had a confirmed diagnosis of obstructive sleep apnea with an
AfH score of 10 or greater from their overnight sleep study. The subjects self-reported
no neurological history of CVA, brain injury or progressive degenerative disease, and no
histories of cleft lip and/or palate or vocal disease or surgery (e.g. vocal nodules, polyps,
cancer, uvulopalatopharyngoplasty). Subjects that were not excluded from the study
included those having had tonsillectomy, adenoidectomy, dental procedures and/or nasal
surgeries. Additional information collected included whether the subject had any
allergies and what medications the subject was taking.

Thirty-five adult volmteer subjects with a confirmed diagnosis of sleep apnea
were studied. The subjects were not assigned to a group mtil after their initial visit and
the sleep lab had contacted them with the results of their polysomnographic testing.
Subjects were then put into either the reference group or CPAP group depending on their
future plans for treatment. The 16 subjects in the reference group returned from 10 to 62
days after their initial assessment. Individuals in the reference group decided on a variety
of treatment plans. Some were not going to receive any treatment, others were planning
on exploring a dental prosthesis, whereas still others were going on to CPAP in the
future. The 19 subjects in the CPAP group had been on CPAP for 30 days or more at the

time of the follow up.

25

There was a wide representation of participants by gender and age in the
obstructive sleep apnea population sampled. There were (11) 31% females and (24) 69%
males. This is consistent with the general sleep apnea population where the prevalence is
two to three times more in men than in women (Yomg et al., 1993). The subjects ages
ranged from 23 years and 1 month to 73 years 6 months, with a mean age of 50.2, SD.
11.9, and a median age of 48.4. This again is consistent with the general sleep apnea
population as this condition peaks during age 40 to 50 years (Wiegand and Zwillich,
1994)

Equipment

Several standard clinical tasks and instruments were used to provide a reference
of the normal population. Cafet-R for Voice, a microcomputer based instrument, uses a
strain gauge to monitor chest wall movement and provide information on relative airflow
while the subject is speaking (Goebel, 1992). It uses a pressure transducer to convert air
pressure to an electrical signal. The Signal estimates movement of the lmgs during
volume displacement. There is a microcomputer attachment and audio input to

coordinate voicing with respiratory measures. Baseline calibration occurred with each

subject during quiet breathing. The Cafet—R for Voice provides visual feedback on

respiratory movement and vocal fold vibration.
Kay Elemetrics Nasometer R model 6200 R with PC attachment was used. The

Nasometer R is a microcomputer-based instrument that provides numerical values of
nasal energy from a ratio of nasal energy over total oral and nasal energy. This device
records oral and nasal somd pressure using two condenser microphones momted on each

side of a somd separator plate that rests between the upper lip and nose. The Nasometer

26

R was calibrated according to the manufacturer's instructions before sample collection

was taken each day.

The Visi- Pitch R model 6097 is a microcomputer instrument that is used to

capture acoustic information providing a waveform that represents fundamental

frequency (f0). The Visi-Pitch R was used to gain information on fundamental frequency
and jitter perturbation. The data were obtained by the production of the /a/ phoneme

produced across the Visi-Pitch screen. Selection of the steadiest visual tracing was done,

and then the cursors were set for a 2-second interval. The Visi-Pitch R microphone was
hand held by each subject, positioned one inch to the right of the center and one inch in
front of the subject’s mouth during the data recording.
Procedures

Each subject filled out an initial and follow-up questionnaire (Appendices A and
B). The subjects performed various well-established clinical tasks that allowed for a
baseline and follow-up comparison of clinician-perceptual ratings and objective
measurements of respiration, speech, voice, resonance, and swallowing characteristics. A
hearing screening was given in a quiet room at the sleep lab during the initial session
only. The test was given in both cars at 25db SPL at 500, 1000, 2000, and 4000 Hz. The
initial data collection session took approximately 45 minutes and the follow-up session
approximately 30 minutes.

Information about respiration for speech consisted of four measures. Maximum
phonation time was obtained by having the individual prolong the /a/ vowel for as long as
possible after a maximum inhalation. Maximum phonation time was recorded on two

consecutive trials. The longer of the two trials was used in the database. The maximum

27

phonation time normative data used as a benchmark was from Hirano, Koike, and von
Leden, (1968). Chest wall movement and coordinated acoustic measures were gained

through use of instrumental analysis on a Computer Assisted Fluency Evaluation Trainer

(Cafet-R for Voice). The data were gathered while the subject read the California
Merge (Hoit and Hixon, 1975) and talked spontaneously about the Cookie Thief
(Goodglass and Kaplan, 1983) picture for sixty seconds with each stimulus. The two
samples were than averaged together. The Cafet-R instrumentation calculated measures
of number of breaths per minute, inhalation sufficiency (60% of the maximum voltage
level of expansion reached during calibration when the individual is demonstrating quiet
breathing) and average length of exhalation while speaking. The information calculated
by the Cafet — R was collected while the subjects were standing. If the strain gauge did
not fit aromd the subject at the recommended level (approximately l-inch below the
bottom rib), it was typically moved to a superior position for improved fit and calibration.
Information gathered for this study from the CAFET- R instrumentation was compared to
normative data provided in the CAFET-R training manual (Goebel, 1992).

Oral mechanism evaluations were done by the investigating speech - language
pathologist to gain information on structure, function and diaodochokinetic rates through
administration of the @l Speech Mechanism Screening Examination - Revised
(OSMSE - R) (St. Louis and Ruscello, 1987). Two judges computed the
diaodochokinetic rates (rapid repetitive movements) by agreeing on the frequency of
productions from audio and visual cues. One of the listeners was the primary investigator

and the other a non—clinical volmteer.

28

Speech production was assessed by having the individual read the California
Passgge and Spontaneously describe the Cookie Thief picture for 60 seconds each. One
subject that had difficulty reading repeated a stimulus sentence. If the individual had
difficulty describing the picture, they were told to talk about an interest of theirs or about
their day. There may be a difference between speech production when reading and
during spontaneous speech; therefore, using only a spontaneous speech sample does not
insure a variety of phonemes will be used. The speech samples were recorded onto
audiocassettes.

The passages were judged using a 7-point scale for speech production developed
by the American Speech- Language- Hearing Association - ASHA (1995) for clinical
outcome (Appendix C). The ASHA-functional communication scales are based on a
severity hierarchy with the low number representing profomd impairment ascending to
severe, moderately severe, moderate, mild-moderate, mild and within normal limits. One
of the key concepts used in the speech production scale is the concept of intelligibility,
which is the degree of mderstandability of the utterances Spoken. Specific criteria for
each severity level have been outlined. A speech-language pathologist who had no
contact with the subjects listened to all of the audiotapes and applied the perceptual
scales.

The individual who listened to the audiotapes and made all the perceptual
judgments on the subjects knew the subjects had a diagnosis of sleep apnea. However,
she had no knowledge of whether the individual was in the reference group or the CPAP
group. The perceptual rater/listener is a certified speech-language pathologist. Her

primary areas of expertise include articulation, cleft palate and voice. She is active on

29

two cleft palate teams and therefore an experienced listener to articulation, voice and
resonance.
There were four measures used to contribute to the assessment of resonance.

Information related to resonance was evaluatedby audiorecording and by use of a

Nasometer R. An audio recording of each individual's reading was subjected to
perceptual ratings of nasality using a 7-point scale ranging from denasal to hypemasal.
The scale used was modified from a resonance scale reported by Wilson in 1971

(Appendix E). The subjects read or repeated the Zoo Passage (Kay Elemetrics) for the

perceptual rating. The Nasometer R provides a quantitative assessment of nasalance.

Two condenser microphones placed near the nose and mouth detect acoustic energy.

Data collected while using the Nasometer R were derived from the following stimuli
selections: a high-pressure oral passage, the Zoo Passage, oral Low Pressure (Karnell,
1995) sentences and Nasal Sentences (Kay Elemetrics). One subject who had difficulty
with reading repeated the passages after the investigator. The Zoo Passage is a standard
passage that has no nasal consonants and is frequently used during evaluations of
velopharyngeal insufficiency. The Low-Pressure Sentences were designed to minimize
or eliminate the effects of turbulent nasal airflow (Karnell, 1995). The Nasal Sentences
passage has many nasal consonants; decreased scores during production of this stimulus
indicate nasal or upper pharyngeal obstruction. A facial mask was fitted on the subj ect’s
head and remained in position for all recordings. With the facial mask in place, the
subjects were asked to read each passage silently before the recording to assist with
gaining an accurate recording. Nasalance data were compared to norms of Seaver,

Dalston, Leeper, and Adams, (1991).

30

There were four measures used to evaluate voice. Phonation/voice was judged
perceptually by a speech - language pathologist using a 7-point scale of firnctional
communication measures - voice disorders from the American Speech-Language-Hearing
Association, 1995 (Appendix D). The perceptual rating was made based on the
California Passage and Cookie Thief picture audiocassette recordings. Voice was
assessed by production of a s/z ratio over two trials and taking the longest trial during
each session. The investigator used a stopwatch to measure the /s/ and /z/ productions.
The subjects were instructed to take a diaphragmic breath and produce the target

phoneme. Voicing was further evaluated by having the individual prolong the vowel /a/

at a comfortable pitch. This was viewed on the ViSi-Pitch R screen set at 15 seconds.
The measures of fundamental frequency and jitter perturbation were fomd from
positioning the cursor on the most consistent 2-second sample (from visual inspection) of
the total sample taken. Jitter perturbation is the cycle-to-cycle change in fundamental
frequency. The Visi-PitchR calculated relative average perturbation using Koike’s
formula (Kay Elementrics). This was measured concurrently from the same /a/ sample.
Information on fmdamental frequency was compared with norms from Fitch (1990).

To gain information on swallowing abilities, clinical perceptual observation was
used. The individual took a sip of water while the clinician noted by a yes/no judgment if
there was a change in vocal quality such as a wet and gurgly voice. The same procedure
was followed when the individual took a small bite of a cracker. Linden and Siebens
(1983) fomd that the most suggestive clinical signs of laryngeal penetration were the
combination of wet-hoarse voice quality and impaired pharyngeal gag reflex. Daniels,

Brailey, Priestly, Herrington, Weisberg, and Fomdas, (1998) fomd the presence of 2 of

31

the following six indicators were predictors of dysphagia in stroke patients: abnormal
volitional cough, abnormal gag reflex, dysphonia, dysarthria, cough after swallow or
voice change after swallow. The investigating clinician estimated the timeliness of the
swallow response from observation of when oral formulation of the bolus has stopped to
when the larynx was felt to rise (as felt by the clinician’s fingers positioned on the
submandibular region and thyroid, and thumb near the temporomandibular joint). The
clinician gathering the data used the American Speech-Language-Hearing Association —
functional communication measure on swallowing to rate swallowing abilities (Appendix
F).
Data Analysis

Data collected from the previously described variables were ratings or numerical
scores. Descriptive data were analyzed by computing percentages, mean, standard
deviation and ranges. Visual inspection of the raw data reduced the number of variables
for further analysis. For example, the clinician’s perceptual rating for swallowing was
collapsed into one group for both the tasks of liquid and solid intake since visual
inspection of the data was so similar. Since the perceptual judgments for reading or
spontaneous speech were nearly identical, further analysis was only made on the
spontaneous speech passage that more closely represents daily speech activities. Each
subject’s data was compared to published critera to look for the number of occurrences
that fell within and outside the normal range. These data were then reported in a
frequency distribution, indicating the number of individuals falling into a specific
category (e.g., normal, above norm, below norm). The various norms used were those

from previously published studies.

32

Further analysis was accomplished by comparing the initial and follow-up results
and the reference and CPAP groups. The areas chosen for statistical analysis were based
on visual inspection of the data, noting change in the frequency data from the initial to
follow-up session or differences between the two groups. Parametric testing was done
when the distributions of the data were appropriate. In most cases, non-parametric
statistics using the number of occurances were employed. The non-parametric analysis
was chosen because the subject pool was different from a normal distribution and the data
were not continuous. The score data on a number of variables were skewed toward the
normal range. This was an expected finding, as clinically the sleep apnea population is

not identified on many speech clinician’s caseloads.

33

Chapter 3

RESULTS

There were 66 individuals who participated in the initial gathering of data. There
were 36 individuals who participated in the initial and follow-up phase of this research
project. The thirty individuals that did not complete the follow-up were those without a
confirmed diagnosis of sleep apnea (i.e. low AHI score) or who, for personal reasons,
were mable to return for the follow-up study. One individual was disqualified after data
collection because of a low AHI score. Thus, there are 35 subjects with initial and
follow-up data that form the following results.

Hearing screenings were done at 500, 1000, 2000 and 4000 Hz with passing
criteria at 25 decibels in both ears. The screening revealed that (11) 31% of the subjects
passed and (24) 69% of the subjects did not pass. The majority (19) 79% of the subjects
that did not pass the screen had difficulty in both ears. Difficulties were distributed fairly
evenly between low, high and mixed frequencies respectively (9) 38%, (7) 29% and (8)
33%. Recall that the mean age of the subjects in this study was 50.2 with a range from
23 years 1 month to 73 years 6 months. It is not musual for the age of the subjects in this
study to have difficulty on a hearing screening.

The subjects’ physical characteristics are displayed in Table 1. This was a clinical
population and subjects in the reference and CPAP groups were not matched according to
any criteria (e.g. gender, BMI, AHI). Body Mass Index (BMI) is used to objectively
compare one subject’s size to another subject’s size, taking stature into accomt. BMI
calculations were based on self-reported height and weight. The subjects had their

weight and height measures taken at their physician’s appointment at the Sleep Lab just

34

prior to the initial data collection. Approximately 10% of the sample was checked with
medical records and fomd to be accurate between the self-report and recorded chart
information. The BMI of the subjects (N = 35) revealed a heavy group of subjects.
Subcategories revealed (1) 3% of our subjects had BMI scores in the normal range, (8)
23% were in the overweight range, (9) 26% in the obese range and the majority of our
subjects (17) 49% were in the very obese range. Body mass index mean was 36.71 , SD.
8.9 with a range from 23.74 - 57.66, compared to approximately 38% of Michigan
residents that have a healthy body weight (BMI =18.5 to 25) (Behavioral Risk Factor
Surveillance System).

The apnea-hypopnea index (AHI) scores revealed that over half of the subjects in
this study (21) 60% had scores of 40 or greater. These scores were in the severe range.
The next largest group of subjects (1 1) 31% had moderate AHI scores of 20 — 39
followed by (3) 9% of the subjects having mild scores of 10 — 19. The mean AHI score
was 55.10, 8D. 29.83 with a range of 11.2 - 100.8. Recall that to participate in this
study, a person had to have an apnea-hypopnea index score of at least 10 (mild AHI

range.

35

 

 

 

 

\2

Table l - Physical Characteristics of Subjects

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Subject No. Gender Age Wt/Ht. Body Mass Apnea
lbs. — ff, in Index Hypopnea
Kg/m2 Index

1 -— Reference F 48.5 170 - 5’2 31.18 13.4
2 — Reference F 38.3 250 — 5’4 V; 42.37 92.5
3 — Reference F 41.8 180 —5’5 30.04 71.9
4 — Reference M 49.2 204 - 5’ 10 29.35 11.2
5 -— Reference M 46.6 250 - 6’2 32.19 29.5
6 - Reference F 37.3 280 - 5’4 48.19 18.7
7 -— Reference M 55.5 230 - 6’5 27.35 37.8
8 — Reference M 40.9 340 - 5’ 10 48.92 48.5
9 — Reference M 44.8 260 — 5’ WV: 36.88 48.1
10 - Reference M 50.6 240 - 5’11 33.56 91.7
11 -— Reference F 59.8 152 — 5’3 27.0 34.3
12 — Reference M 48.3 165 — 5’10 23.74 96

13 — Reference M 65.10 340 — 6’3 42.61 32.6
14 — Reference M 61.6 254 - 6’0 34.54 45.1
15 — Reference M 45.7 290 — 5’ 10 44.39 72.4
16 - Reference M 60.10 185 - 6’0 26.77 18.7
17 —CPAP M 48.10 385 —6’1 50.93 31.1
18 — CPAP M 30.9 189 — 5’9 J/4 27.39 23.0
19 -CPAP F 73.6 163 —5’0 31.92 44.3
20 - CPAP M 54.9 300 -— 6’0 40.80 88.7
21 — CPAP M 44.4 371 — 5’11 51.89 32.8
22 — CPAP M 31.11 280 — 6’1 37.04 92.4
23 — CPAP F 64.9 235 — 5’1 '/2 43.8 80.6

 

 

 

 

 

 

36

 

Continue Table l

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Subject No. Gender Age Wt/Ht. Body Mass Apnea
lbs. — ft., in Index Hypopnea
Kg/m2 Index
24 — CPAP M 45.3 227 — 6’1 30.03 82.3
25 — CPAP M 61.4 335 — 5’4 57.66 100.1
26 — CPAP F 40.5 200 —5’4 34.42 88.9
27 — CPAP M 60.5 240 —5’ll V2 33.10 100.8
28—CPAP M 49.8 349—5’11 48.81 94.7
29 —CPAP F 62.3 245 ~5’3 43.52 63.7
7) — CPAP M 71.4 245 — 6’2 43.52 21.4
31 —CPAP F 43.10 158-5’6 25.57 21.
32 - CPAP M 23.1 210 — 6’2 27.04 25.1
33 - CPAP M 48.9 263 — 5’10 37.84 41.5
34 —- CPAP M 41.2 333 —— 6'] 44.05 75.9
35 — CPAP F 69.5 176 — 5'6 28.48 57.9
Mean F = 11 50.2 F_wt = 200.82 36.71 55.10
M = 24 M_wt = 270.21 [135.14
F_ht =5'3
M), = 5.9 v. M_37.93
8D. 11.90 F_wt = 44.11 8.9 29.83
M_wt = 62.03
F_ht = 1.88
M_ht = 2.49
Range 23.08 — 73.5 F_wt = 152 — 23.74 — 57.66 11.2 — 100.8
280 F_25.57-48.l9
M_wt = 165 —
385 M_23.74-57.66
F_ht = 5’ — 5’5
M_ht = 6’ —
6’6

 

 

37

The initial data were gathered prior to the investigator or the subject’s knowing
into which group they would be placed. After the subjects received a confirmation of
their diagnosis and had a planning session occur with their physician regarding treatment
they were placed in the reference or CPAP group. Subjects in the reference group were
either going to try medication, weight reduction, a dental appliance or CPAP at a later
time. The reference subjects had their follow-up at an average of 36.69 days after their
initial data collection. The range for the reference group was from 10 - 62 days; a few of
these extremes were secondary to scheduling purposes. The subjects receiving CPAP
treatment were the independent experimental variable. They were asked to come to the
follow-up appointments after 30 days on CPAP. Their mean number of days from the
initial data gathering to the follow-up averaged 56.74 days, with a range of 32 — 105
days. Delays were caused by scheduling difficulties, initiating CPAP use, or with
returning for the follow-up study appointment.

Self-Perceptions

There were three main areas of data collection including self-perception, clinical
judgment and objective measures. Information on self-perception was gathered using
initial and follow-up questionnaires. A baseline questionnaire (Appendix A) was given to
all subjects to obtain a self-report on their perception of their respiration support for
intelligibility and capacity for sustained speech, speech production, voice quality,
resonance (i.e., tone of voice) and swallowing. The total group had a self-perception of
adequate (89% or better) breath support for speech, clear speech productions, appropriate
resonance and vocal quality. There were fewer self-reports of adequate swallowing (77%)

(see Table 2).

38

Table 2 - Percentage of Subjects Self-Reported Difficulties Across Various

 

 

 

Categories
Subjects Respiration Speech Voice Resonance Swallowing
N = 35 for speech Production

Total 6% 9% 1 1% 9% 23%

 

 

 

 

 

 

Clinician Perceptual Judgments

Perceptual judgments were made by a speech-language pathologist to gain a

better understanding regarding the functional ability of each subject’s communication

process. There were four areas in which perceptual judgments were made, three using

the chtional Communication Measures as established by ASHA, the national

professional organization. These areas were speech production, voice and swallowing.

Scales for speech, voice and swallowing are based on a seven point scale going from 0 —

7. Zero is mable to test and one is the most profomd moving toward seven as normal.

The judgments for speech production, voice and resonance were made by a speech

pathologist listening to the subject’s audio-recordings. One subject had a noisy

recording; therefore, perceptual judgments of her data were not made by the therapist and

the swallowing data were not used in the aggregate data. Resonance was judged using a

modified scale from Wilson (1971)(Appendix E). The resonance scale is a seven-point

scale from —2 denasal moving to 0 normal and +4 severe hypemasal. Swallowing was

judged at the time of the data gathering by the primary investigator. The primary

investigator works in a large metropolitan acute care hospital with experience in clinical

and instrumental dysphagia evaluations. Ratings on respiration were not made

perceptually.

39

 

Data describing spontaneous speech production indicated that (12) 35% of the
subjects received a normal perceptual rating and (22) 65% of the subjects received a
rating of six, revealing speech is intelligible in and out of context, but the production is
sometimes distorted (see Table 3). The deviations noted by the clinician included
distortions of the /s/ and /r/ phonemes, which are both lingual alveolar somds.

Perceptual judgment of voice indicated a wider distribution of ratings, from
normal to mild-moderately difficulty. There were (17) 50% subjects who received a
normal rating; (13) 38% received a rating of six indicating voice production is
appropriate in most situations although minimal difficulty may occur. Another (4) 12%
of subjects had ratings of 5 described as a mild-moderate disorder where voice is
consistent in most contexts and voice quality is distracting to some listeners (see Table
3). The clinician judge commented repeatedly that many samples were “low-pitched at
the bottom of their range.”

Resonance/vocal tone judgments indicated that the majority of subjects (18) 53%
were judged as having normal resonance. This was followed by (15) 44% being
perceived as having moderate hypemasal a score of 2 and (1) 3% of the subjects as
having a mild denasal pattern a score of —1 (see Table 3).

Swallowing abilities were primarily judged as normal with (32) 91% of the
subjects receiving this rating. Mild difficulties, a rating of six, were observed with (3)
9% of the subjects, indicating functional eating activity with periodic difficulties or the

need for additional time when eating (see Table 3).

40

Table 3 — Percentage of Subjects Rated Outside of the Normal Range for Various
Perceptual Categories

 

 

 

Type of rating Speech Production Voice Resonance Swallowing
N=34 N=34 N=34 N=35
Outside of Normal Range 65% 50% 47% 9%

 

 

 

 

 

Of the 34 participants used for analysis, (32) 94% had one or more abnormal
perceptual ratings (see Table 4). Only (2) 6% had normal scores on all four of the
perceptual measures. None of the subjects were abnormal in all four ratings. Specific

information on clinician ratings in each of the perceptual areas can be fomd in Table 3.

Table 4 - Percentage of Subjects Judged as Within or Outside of the Normal Range
by Combined Perceptual Categories (Speech Production, Voice, Resonance, and
Swallowing)

 

 

 

Subjects Normal All One Abnormal Two Abnormal Three
N = 34 Parameters Parameter Parameters Abnormal
Parameters

TOTAL 6% 41% 38% 15%

 

 

 

 

 

Objective Information
Oral Abilities

On the oral motor screen of structure and fmction using the OSMSE-R (N = 35),
(11) 32% of the subjects scored below norm, (18) 51% in the appropriate norm range for
their age and (6) 17% above the norm for their age. Reference norms were taken fi'om
the OSMSE-R manual according to the age of the subject (St. Louis and Ruscello, 1987).
Mean score for our total group was 51.82, SD. 2.74, range 44 — 55, (see Table 5).

Review of the completed forms indicated the most frequent reason that subjects were

41

 

 

rated down in structure was due to significant ridges on their tongue. This is often caused
by a forward resting position and may be secondary to trying to enlarge the airway. On
the fmction portion of the OSMSE-R the area most frequently marked down was also
tongue range of motion. Diadochokinetic rates of “puh”, “tuh”, “kuh”, “puh ” and
“puhtuhkuh” showed a wide range of abilities. Four 11% of the subjects producing all 5
repetitions within or better than norm (St. Louis and Ruscello, 1987). The remaining (31)
88% produced the targeted repetitions below norm: one repetition below norm (7) 20%,
two of the repetitions below norm (6) 17%, three targets (8) 23%, four targets (3) 9% and

below norm on all targets (7) 20%.

Table 5 - Distribution of Subjects on the Oral Speech Mechanism Screening
Examination -R

 

 

Subjects Below -2 SD. -2 SD. -1 SD. Normal +1 S.D. + 2 SD.
Total N = 35 7 l 3 l8 6 0
% 20% 3% 9% 51% 17% 0%

 

 

 

 

 

 

 

 

 

 

Respiratory Measures

The majority of subjects (24) 70% had appropriate number of breath groups per
minute (norm, 12 — 26) while talking for one minute. For the three other respiratory
measures the majority of scores revealed abnormal findings. Duration of exhalation
during speech was mostly too short or too long. A longer than normal duration of
exhalation (norm, 2.6 — 3.2 seconds) while speaking was noted for (19) 55% and too
short of a duration was noted for (10) 30% of all subjects. The data for sufficient
inhalation (norm, 0%) revealed the majority of subjects (32) 97% were not reaching 60%

of tidal volume when speaking and had insufficient inhalation. The data were segmented

42

into groups of l — 25, 26 - 50, 51 - 75, 76 — 100 and 101 - 125 percent of occurrences of

insufficient inhalations and indicated peaks at 26 - 50 and 76-100. The further away

from zero, the more frequently the individual demonstrated a shallow type of breathing.

Insufficient inhalation scores can occur as a percentage expressed over 100% because the

measures are calculated from voiced occurrences of speech and at times the individual

will have unvoiced cycles (M. D. Goebel, personal communication, September 2001).

The maximum phonation times were below the normal range for subjects (33) 94% of the

time and normal performance was noted (2) 6% of the time for our subjects. Norms for

maximum phonation time used were 30.2 — 39.4 seconds for males and 22.9 — 28.7

seconds for females (Hirano, Koiki, and von Leden, 1968). See Table 6 for Respiratory

data as well as comparative norms from the CAFET-R training manual (Goebel, 1992).

Table 6 — Descriptive Characteristics for Several Respiratory Parameters

 

Area

Number
of
Subjects

Below
Norm

Norm

Above
Norm

Mean

Standard
Deviation

Range

 

Breath
Groups per
Minute Norm

<11

12—26

>27

 

No. of
Subjects

33

18%

70%

12%

16.83

6.88

 

Sufficient
Inhalation
Norm

 

No. of
Subjects

33

97%

3%

0%

61.18

33.84

0—105

 

Duration of
Exhalation
Norm

< 2.4
seconds

2.5 — 3.2
seconds

> 3.3
seconds

 

No. of
Subjects

33

30%

15%

55%

3.30

1.3

1—6.4

 

Maximum
Phonation
Time Norm

<

M = 30.2—

39.4

F =22.9—

28.7

>

 

 

No. of
Subjects

 

35

 

94%

 

6%

 

0%

 

19.57

 

7.22

 

 

43

 

Th4

the l

maj

DE.)

11‘}:

5U

1111

\ L33

 

Resonance

The subjects’ scores were compared to norms noted in a study by Seaver et a1.
(1991); the Zoo Passage mean is 15 % with a standard deviation of 6 and the m
Sentences mean is 62% with a standard deviation of 6. The Zoo Passage, a high-pressure
oral passage, indicated the majority of subjects’ nasalance scores were in the normal
range with a tendency of some subjects to have slightly lower nasalance (see Table 7).
The group’s mean was 13.86, SD. 5.27 and the range was 7.19 - 28.99. The results of

the Nasal Sentences a group of sentences with abmdant nasal consonants, indicated the

 

majority of scores were in the normal range followed by low nasalance scores for the
next largest group of subjects. The total group’s mean was 55.37, 8D. 10.68 and range
of 11.17 — 68.17. Each subject’s Low Pressure Sentence, nasalance score, was compared
with his or her scores on the Zoo Passage; the comparisons revealed that the passage
scores were within 10 points of each other on (34) 97% of the subjects’ recordings,
indicating there were no indications of nasal emission.

All the subjects had a mean of 15.17, SD. 7.27 and a range of 7.28 - 38.83 on the
Low Pressure Sentences. The Low Pressure Sentences data were not present in the tables
as there are no norms for this passage. The Low Pressure Sentences passage was used as

a screening to help identify nasal turbulence from an overall high nasalance outcome.

Table 7 - Nasalance Data Distribution

 

 

 

 

Stimulus - 3 SD. - 2 SD. - 1 SD. Normal +1 S. D. + 2 S. D.
Zoo N = 35 0% 0% 17% 69% 11% 3%
Nasal N = 35 6% 11% 23% 57% 3% 0%

 

 

 

 

 

 

 

44

 

Voice Parameters

Norms for fundamental frequency of vowels (males 115.6 —164.4 Hz and females
229 — 288 Hz.) were made using data from Fitch (1990). The fundamental frequency (f0)
was in the normal range for (10) 29% of the subjects and below norm for (25) 71% of the
subjects. A lower than expected f0 was predominant with the female subjects (N = 11):
(9) 82% were below norm and (2) 18% were in the normal range. There were (16) 67%
of males (N = 24) with lower than expected f0 and (8) 33% had scores in the normal
range. No one in the study had a f0 higher than the normal range. The males had a mean
fo of 107.86 Hz, 8D. 17.45 and a range of 79.1 to 146.5. The females had fo averaging

203.02 Hz, 8D. 23.22 and a range of 170.8 to 239.70 (see Table 8).

Table 8 — Fundamental Frequency Data

S ects N
T N = 35 29%

F es N=Il 18%
CSN=24 3%

 

For these subjects the s/z ratio indicated (8) 25% of the subjects had ratios of less
than .79, the physical correlate of breathy voice, the majority (18) 51% had s/z ratios
falling in the norm area .8 — 1.29 (Hirano et al., 1968). Finally, (9) 26% had ratios above
1.3, raising suspicion regarding deviant voice with the physical correlate of harshness.
The mean for the total group, s/z ratio was 1.11, SD. .41 and range of .4 — 2.4. Jitter
perturbation indicated the majority of subjects, (31) 89% had normal scores of 1.25 % or
less (Kay Elemetrics). The group mean for jitter was .79 % relative average perturbation;

the S.D.was .71% and the range was .21 — 3.8%.

45

Swallowing

There was no audible change in vocal quality after eating or drinking for any of
the 35 subjects. The majority of swallows were timely. The swallows were prompt for
(33) 94% of the subjects with liquid swallows and for (22) 65% of the subjects with solid
consistencies. Delays with liquids were 1-second in duration for (2) 6% of the subjects.
Delays with solids were l-second in duration for (10) 29% of the subjects and 2 -Seconds
in duration for (2) 6% of the subjects.

Summary of Baseline Characteristics

There was a broad distribution by age, and approximately a third of our subjects
were female. Increased body weights are common in the OSA population (Wiegand and
Zwillich, 1994). It is common for 80% of the OAS population to have overweight ratings
(Lugaresi, 1988). This sample had (34) 97% of the subjects with BMI scores above the
normal range. The majority of our subjects had moderate or severe sleep apnea, with
(10) 29% and (21) 60% respectively based on their AHI scores (see Table 1).

The subjects most typically rated their own communication abilities as adequate,
with a small number of subjects self-reporting some swallowing difficulties. The
perceptual baseline as judged by a clinician revealed that a majority of subjects had less
than completely normal communication. Only (2) 6% of the subjects had completely
normal communication ratings across all areas of speech, voice, resonance and
swallowing. Most judgments put subjects in the category of mild difficulties. In the area

of resonance, subjects were frequently judged as having moderate difficulties.

46

 

511i

511’

the

i To

abi

 

 

sut

lllZ

F.

The OSA subjects revealed differences from the normal population in the area of
respiration (Goebel, 1992). Three respiratory tasks revealed inadequate performance:
suflicient inhalation, duration of exhalation and maximum phonation time.

Nasalance was normal for the majority of subjects with both the Zoo Passage
stimuli for (24) 69% of the subjects, and with the Nasal Sentence stimuli for (20) 57% of
the subjects. This was followed by low nasalance for the next largest group of subjects.
Vocal quality did not appear to be a large problem for those subjects showing normal
abilities with s/z ratio (Hirano etal., 1968) and jitter (Kay Elemetrics). A majority of
subjects, (25) 71% had lower fundamental frequencies than published norms (Fitch,
1990)

Oral motor abilities as a group were fomd to be in the normal range (1 8) 51% the
majority of the time with the next largest percentage of subjects receiving scores below
norm (11) 32%. However, combining normal and above normal abilities resulted in (24)
68% of the subjects falling in positive performance areas. The only notable swallowing
finding was a delayed swallow for (12) 35% of the subjects when eating solids.

F allow-up Procedure Results

At the time of follow-up it was known what group, reference or CPAP, the
subjects were placed in. There were 16 subjects in the reference group and 19 subjects in
the experimental CPAP group. The reference group’s mean age was 49.19, SD. 8.65
with range of 37.03 — 65.10, whereas the CPAP groups mean age was 50.38, 8D. 14.33
with a range of 23.01 - 73.06. The BMI and AHI severity were similar for the reference
group and the CPAP group. The BMI mean for the reference group was 34.94, SD. 8.04,

with a range of 23.74 — 57.66, whereas the BMI for the CPAP group was 38.20, SD.

47

8.04, with a range of 23.74 — 48.92. A parametric t-test (or = .05, df = 33) did not show a
significant difference between the two groups. The AHI scores indicating the severity of
the sleep apnea appeared to show higher scores for the CPAP group as a whole. AHI
scores over 40 are considered to be severe. The AHI mean scores for the reference group
were 47.65, 8D. 28.88, with a range of 11.2 —- 96.0. The CPAP group had more severe
AHI scores with the mean of 61.38, 8D. 29.91 and a range of 21 .0 —- 100.8. However, a
t-test analysis (or = .05, df = 33) showed no significant differences between the two
groups. Both the reference and CPAP groups had the same basic characteristics.
Self-Perception Follow-up

All subjects participated in a second trial of the various communications and
swallowing tasks. Subjects completed a questionnaire (Appendix B) which asked
whether their abilities had stayed the same, were better, or were worse since the subject’s
initial participation. Further examination of the initial data revealed that there were
differences between the group of subjects who would be in the reference and CPAP
groups. The CPAP group had more frequent complaints than did those who would be in
the reference group. Compared to the reference subjects, the CPAP subjects had more
descriptions of problematic vocal tone (0% vs. 16%), impaired vocal quality (0% vs.
21%) and difficulty swallowing (19% vs. 26%). At follow-up there were also five
questions asked; the questions included one on respiratory support for speech, speech
production, voice quality, resonance and swallowing. Subjects rated their abilities
compared to the initial session as mohanged 94% of the time or greater. The reference
group rated themselves as having the same skill in all five areas 100% of the time. A

number of subjects from the CPAP group (4) 21% reported improvements in three areas:

48

(2) 11% laryngeal vocal quality, (1) 5% resonance/tone of voice, and (1) 5% swallowing.

The area of respiration for speech became worse for one person, 5% of the CPAP group.

Table 9 - Percentage of Subjects Self-Reporting Difficulties Across Various
Categories, at Follow-up

 

 

 

Change Respiratory Speech Voice Resonance Swallowing
N=35 N=35 N=35 N=35 N=35
Same 97% 100% 94% 97% 97%
Better 0% 0% 6% 3% 3%
Worse 3% 0% 0% 0% 0%

 

 

 

 

 

 

 

 

Clinician Perceptual Judgment Follow-up

On follow-up the four areas involving clinician perception continued to show that

the majority of subjects had at least one or more areas that were judged to fall outside of

the normal range (see Table 10). However, the areas judged perceptually had the most

consistent and greatest change in the positive direction. The number of subjects that had

all 4 perceptual areas (speech, voice, resonance, and swallowing) rated as normal

improved nearly 5 times, moving from (2/34) 6% initially to (10/3 5) 29% at the time of

follow-up. Notably the largest positive change occurred for the CPAP group moving

from (0) 0% of the subjects receiving normal ratings in all four areas initially to (7) 20%

receiving normal ratings at follow-up. Similar to the initial data collection, no subjects

had abnormalities in all four perceptually rated areas. A chi square (x2) analysis (or = .05,

(if = 1) revealed no significant difference between the two groups when comparing the

perceptual judgments.

49

 

Table 10 - Percentage of Subjects as Within or Outside of the Normal Range by

Combined Perceptual Categories (Speech Production, Voice, Resonance, and

Swallowing) at Follow-up

 

 

 

 

 

 

 

 

Groups Initial N = 34 Follow-up N=35
All 1 2 3 All Normal 1 2 3
Normal Abnormal Abnormal Abnormal Parameters Abnormal Abnormal Abnormal
Parameter Parameter Parameters Parameters Parameter Parameters Parameters
Total 6% 41% 38% 15% 29% 20% 29% 23%
Reference 13% 31% 44% 13% 19% 19% 38% 25%
CPAP 0% 50% 33% 17% 37% 21% 21% 21%

 

 

 

 

 

 

 

When looking at the perceptual ratings area by area, the largest movement in

judgments corresponded to changes in speech production (see Table 11). During the

initial data collection the majority of subjects (22) 65% scored a 6, mild difficulty on the

chtional Communication Measure — Speech Production (ASHA, 1995). At the time of

the follow-up the majority of subjects (19) 54% scored a 7, normal. Both the reference

and CPAP groups improved 19%. A chi square (x2) analysis (or = .05, df = 1) revealed

no significant difference between the reference and CPAP group with regard to the

perceptual ratings of speech production.

For the perceptual judgments of resonance (18) 53% of the subjects initially

received a normal rating compared to (17) 49% receiving a normal rating at the time of

follow-up. Shifts occurred with more individuals being rated in the hypemasal area

scores of +1 and +2. There were no group changes observed in the areas of voice

perception or swallowing as judged perceptually. For voice perception the majority of

subjects scored a chtional Communication Measure of 7 - normal, during the initial

(17) 50% and follow-up session (19) 54%. Swallowing results were the same for the

initial and follow-up rating with (33) 94% of the subjects receiving a normal rating of 7

on the Functional Communication Measure — Swallowing.

50

 

Table 11 — Percentage of Subjects Scoring a 6 or 7 Rating on the Functional
Communication Measure- Speech Production (FCM-SP), Initial and Follow-up
Comparison

-p:

 

. .-_.. I
.,
n

Objective Data — Initial versus Follow-up
When comparing the total group’s initial to final scores on the variables within

the 5 main areas, a consistent pattern was observed. There was not one area where the

 

trend changed significantly (see Table 12). Observation of the major areas (i.e., oral F
motor abilities, respiration, voice, resonance and swallowing) revealed respiration was
consistently the measure falling most frequently outside of the normal range, followed by
a split on the responses to the nasalance tasks. On the nasalance task the stimuli used
affected the results with the subjects falling into the normal range or showing low
nasalance, with infrequent scores of high nasalance.
Oral Motor Follow-up

Oral motor abilities, which are often associated with speech production, did not
change for the total group. During the initial and follow-up session, (24) 68% of the total
group of subjects were ranked as within or better than norm, and there were (11) 32%
below norm. It was noted that there was an inversion of performance for the two groups,
with the CPAP group initially placing more frequently at norm or better (14) 74% and
reference group (10) 63%. However, at follow-up it was the reference group (12) 76%

more frequently at norm and not the CPAP group (12) 64% (see Table 13). Chi —squared

51

Table 12 - Objective Data Comparison of Initial and Follow-up Results

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Rating Session Abnormally Normal Abnormal Comments
Low High and Findings
OSMSE- R Initial 32% 51% 17%
N = 35
F ollow-up 32% 57% 1 1%
N = 35
BGPM Initial 18% 70% 12%
N = 33
F ollow-up 29% 65% 6%
N = 34
Sufficient Initial 97% 3% NA 170W = % 0f
Inhalation N = 33 "me {‘0‘
Follow-up 88% 13% NA mashing 60%
N = 32 of;t1da1 volume
X NS
Duration of Initial 30% 15% 55% LOW = short
Exhalation N = 33 High = “mg
F ollow-up 26% 12% 62%
N = 34
Maximum Initial 94% 6% 0%
Phonation N = 35 -
Time F ollow-up 86% 11% 0% X‘ NS
N = 35
Fmdamental Initial 71% 29% 0%
Frequency N = 35 ,
Follow-up 69% 31% 0% X1 NS
N = 35
Jitter Initial NA 89% l 1% Higher scores
Perturbation N = 35 1991“” .
Follow-up NA 94% 6% d'fficult'es
N = 35 X NS
S/Z Ratio Initial 23% 51% 26% Low scores
N = 35 indicate
0 o o breathiness
F 011113;?!) 34 /0 54 /o 1 1 /o High scores
indicate
harshness
Nasalance — Initial 17% 69% 14%
Z00 N = 35
F ollow-up 26% 63% 11%
N = 35
Nasalance — Initial 40% 57% 3%
Nasal N = 35
Follow-up 37% 51% 12%
N = 35
Vocal Initial 0% 100% NA LOW = change
Quality — N = 35 in quality
Liquid Follow-up 3% 97% NA
N = 35

 

 

 

52

 

 

 

 

Continue Table 12

Quality —
Solid

- Liquid

 

Analysis (or = .05, df = 1) revealed no significant difference between the reference
and CPAP group. Diadochokinetic rates at follow-up mirrored the initial results with a
wide range of abilities and a similar number of subjects producing all repetitions of
“puh”, “tuh”, “kuh”, “puhtuh” and “puhtuhkah” at or above expected norms (St. Louis
and Ruscello, 1987). Initially, (4) 11% of the subjects produced all 5 targets at or above

norm on follow-up (5) 14% obtained this status.

Table 13 — Percentage of Subjects Scoring Below or at Norm or Better on the Oral
Speech Mechanism Screening Examination (OSMSE-R) - Initial and Follow-up
Comparison, Summary of the Reference and CPAP Groups

N=35

 

53

 

Respiratory Follow-up

The same trend was noted on the initial and follow-up data for the total group for
all the respiratory tasks (breath groups per minute, sufficient inhalation, duration of
exhalation and maximum phonation time). Comparison of the data by reference and
CPAP group was mremarkable for sufficient inhalation and duration of exhalation.

Similar to the initial trials, on three of the four respiratory tasks the subjects showed

. «If-y

abnormal performance, indicating reduced respiratory support for speech. For breath
groups per minute, the both groups, reference and CPAP repeatedly showed a majority of

subjects had abilities in the normal range. The CPAP group had more subjects in the

 

normal category than the reference group at the time of both the initial and follow—up
trials. In addition, there were changes made during the maximum phonation time task.
There were a small number of subjects from the reference group (2) 13% that initially
scored in the normal range at follow-up however, this doubled to (4) 25% in the normal
range. By contrast, at follow-up; (19) 100% of the CPAP group remained below normal
range for maximum phonation time (MPT)(see Table 14). At both sessions no

individuals had MPT above norm.

Table 14 — Percentage of Subjects Scoring Within or Outside of Norm for Maximum
Phonation Time (MPT), Initial and Follow-up Comparison

 

 

 

 

 

 

Subjects Initial N = 35 F ollow-up N = 35
<Norm Norm > Norm <Norm Norm >Norm
Total 94% 6% 0% 89% 1 1% 0%
Reference 88% 13% 0% 75% 25% 0%
CPAP 1 00% 0% 0% 1 00% 0% 0%

 

 

 

 

 

 

 

 

54

Resonance Follow-up

The objective nasalance data Showed the same trend for the large group during the
initial and follow-up sessions as well as for the reference and CPAP groups (see Table
15). The majority of subjects had normal nasalance for both stimulus passages, followed
by low nasalance scores. With the Zoo Passage, (24) 69% initially had scores in the
normal range, and at follow-up there were (22) 63% of the subjects in the normal range.
With the Nasal Sentences measure the majority of subjects performed within the normal
range of nasalance, (20) 57% initially and (18) 51% at follow-up. The next largest group
of subjects had low nasalance scores for both sets of stimuli. There did not appear to be a
considerable difference between the reference and CPAP group for nasalance on the high
pressure Zoo Passage. There was a trend of low nasalance for the reference group on the
Zoo Passage at follow-up. Chi-squared analysis (or = .05, df = 1) on nasalance data on
the Zoo Passage and the Nasal Sentences both revealed no difference between the
reference and CPAP groups. Thirty-two of the 35 subjects (91%) had similar data
between the Low Pressure and Zoo Passage. This did not differ from the comparison
initially of (34) 97%, indicating agreement between the scores received for two oral

passages.

Table 15 - Nasalance Data Distribution, Initial and Follow-up Comparison

2 o =

- - - orm + -
8D 8D 8D 8D.
11 26%

65
16%
1% % 51%

%

 

55

Voice Parameters Follow-up

In two areas where quantitative voice measures were taken (i.e., jitter perturbation
and s/z ratio), scores fell in the normal range at both the initial and follow-up sessions for
the total group. The fundamental frequency data remained below normal for the majority
of subjects. There was a slight movement to the normal range for subjects in the CPAP
group at follow-up. This positive movement occurred for one female in the CPAP group
(see Table 16). The jitter perturbation data did not differentiate between the reference
and CPAP groups initially. However, upon follow-up the reference group demonstrated
slight improvement toward norm while the CPAP group was mohanged (see Table 17).
The s/z follow-up data had the same distribution as the initial data with the majority of
the total group of subjects in the normal range. There were some differences between the
reference and CPAP groups; however, the movement trend was the same. Initially and at
follow-up the reference group was more pronomcedly placed in the normal range. At
follow-up some movement to below the .8 level occurred, a finding that indicated the
correlate of a more breathy voice for the reference subjects. The largest portion of the
CPAP group (7) 37% initially were in the above normal range 1.3+; and upon follow-up
after they had participated in the CPAP treatment, a greater number of CPAP subjects
(1 l) 58% then placed in the normal range. Thus, both the reference and CPAP groups at
the time of follow-up moved to lower ratio levels, possibly indicating more breathiness

with their phonation (see Table 18).

56

Table 16. Percentage of Subjects Within or Outside of Norm for Fundamental

Frequency, Initial and Follow-up
F 0

01111

ects N= 5

T
32

 

Table 17. Percentage of Subjects Within or Outside of Norm for Jitter

Perturbation, Initial and Follow-up
Fo ow-up Itter

rtter
N

N=35 N
%

1

 

Table 18. Percentage of Subjects Within or Outside of Norm for s/z Ratio, Initial

 

 

 

 

 

 

 

 

 

 

 

 

and Follow-up
Initial s/z F ollow-up s/z
Subjects Below Norm Above Below Norm Above
N = 35 Norm norm norm norm
Total 23% 51% 26% 34% 54% 1 1%
Reference 1 3% 75 % 1 3% 44% 50% 6%
CPAP 32% 32% 3 7% 26% 5 8% 10%

 

 

Swallowing Follow-up
There was no substantial change in the data for the total group from the initial to

follow-up data gathering with regard to vocal quality or promptness of swallowing
liquids. The same trend also emerged for the promptness of swallowing a solid, with the

majority of subjects in both the reference and CPAP group showing timely swallows (22)

57

65% initially and (28) 80% at the time of follow-up. There were greater improvements
within the CPAP group versus the reference group while both moved in a positive
direction (see Table 19), chi-squared analysis (or = .05, (if = 1) revealed no significant
difference between the reference and CPAP groups with the promptness of their

swallowing solids.

Table 19 - Percentage of Subjects With Timely Swallowing of Solids, Initial and
Follow-up Comparison

 

 

 

 

 

Subjects Initial Timeliness of Swallow F ollow-up Timeliness of Swallow
N = 34 N = 35
Timely 1 sec 2 sec Timely 1 sec 2 sec
Total 65% 29% 6% 80% 14% 6%
Reference 56% 38% 5% 69% 25% 6%
CPAP 72% 22% 6% 89% 5% 5%

 

 

 

 

 

 

 

 

Summary of Follow-up Data

Both groups started with similar baseline characteristics including body mass
index and apnea-hypopnea index scores. The follow-up data revealed that self-perception
of abilities and the objective data taken showed little change from the initial trends
established for the total group. There was, however, a movement toward normal ratings
for the stimuli judged perceptually by the speech clinicians. A closer look was taken of
the differences between subjects in the reference group and those participating in the
CPAP group when reviewing the follow-up data. A positive trend was noted during
speech production judged perceptually, maximum phonation time, and the timeliness of
the subject’s swallow. The improvements were more frequently observed with the CPAP

subjects. There were no significant differences between the reference and CPAP groups

58

 

on areas where additional statistical analysis was done. A review of the data by each
individual subject revealed that most subjects stayed the same on tasks. However, there
was individual variation in positive and negative directions with some individuals, but

these were not sufficient to change the group means.

59

 

Chapter 4
DISCUSSION

This sample of obstructive sleep apnea patients is different from the nationally
normal population in a number of communication areas. The difference is primarily
noticeable in two areas of this study. The first is with the perceptual judgments made by
the speech-language pathologist. The second is with the subjects’ abilities, as recorded
during the objective collection of data, in the major area of respiratory support for speech
production.

A large percentage of the study population was judged as having mild
commmication deficits in speech, voice or resonance. Only (2) 6% of the 34 subjects
had perceptual ratings that were normal in all four perceptual areas. However, there was
a noticeable difference between patients’ self-report of communication and swallowing
abilities and the clinician’s judgment. The majority of patients judged their abilities as
normal 89% of the time or better in the communication areas and normal 77% of the time
with swallowing. This may be due to the subtle communication deviations of the
subjects at a time when the patient is focused on improvement in the somdness of their
sleep and daytime alertness. The clinician providing the perceptual ratings for speech,
voice and resonance did not have visual contact with the subjects. It is possible that the
subjects’ physical characteristics would have made the speech sample more acceptable to
the clinician, thus creating the difference between the self-rating and clinician perception.

There are other reports citing that perceptual judgment of resonance only
moderately correlate with nasalance measures (Dalston, R.M., Warren, D.W. and

Dalston, ET, 1991). This was true in our study as well. However, the deviant

60

perceptual scores did parallel the deviant nasalance scores, but not always in the
corresponding direction. For example, there were a higher percentage of low nasalance
scores and a high number of hypemasal perceptual ratings, both ratings showing
deviance. The mid-western sample used in this study should not present dialectical
differences that would affect comparing this sample to published norms. Previously, an
mpublished study on local nasalance norms showed that the published norms fit our
normal subject population of adults (Parker, 1995).

Speech production was the area most often rated by the clinician judge as mildly
impaired, this may relate to the more frequent oral narrowing and less flexible airways
fomd in the sleep apnea population (Lowe, 1990; Sher, 1990). Oliver and Evans (1986)
fomd speakers with a noticeable articulation deficit have a tendency toward having
smaller oral dimensions than those individuals who have normal articulation. The
subjects in this current study also may be similar to the subjects from the Ruscello,
Tekieli, and Van Sickels, (1985) study, which showed 72% articulation errors before jaw
surgery with improvement after surgery highlighting that abnormal oral structure can
have an effect on articulation abilities. The current study supports the previous study by
Monoson and Fox (1987) which fomd that 62%, the majority of sleep apnea subjects in
their study, were judged as having abnormal speech. The perceptual findings in this
study were also similar to the findings of Salas-Provance and Kuehn (1990) who fomd
listener-judges were able to tell the difference between the reference (non-sleep apnea)
subjects versus the UPPP subjects primarily in the areas of phonation and articulation.

The body mass of this population showed 98% of their BMI scores were above

the normal range. There is an abundance of information in the literature demonstrating a

61

connection between being overweight and many types of chronic disease including
hypertension, pulmonary disease, cardiovascular disease and digestive deficits. There are
also reports of a relationship between sleep apnea and high blood pressure, heart failure,
heart attack, pulmonary difficulties, and stroke (Koskenvuo, Kapno, Partinen,
Langinuainio, Sarna, and Heikkila, 1985, Waller and Bhopal, 1989). The question then
emerges: are there typical communication and swallowing features that relate to
increased weight? The decreased respiratory support for speech may indeed be related to
the increased body mass of our subjects. The increased body mass may result in an
overall decrease in their respiratory systems’ capacity secondary to obstruction and bulk.
The additional body mass may result in an overly stressed system with correspondingly
less flexibility to control rapidly changing airway dynamics, especially with the overlaid
fmction of voice production. Also, the instrumental equipment may have reduced
capacity to work effectively with individuals of larger size secondary to the fit of the
strain gauge, or decreased detection of movement for abdominal/thoracic excursion, often
a sign of hypopneas.

Respiration is the fomdation for speech, voice and resonance. Decreased
respiratory support for speech may impact the intelligibility of the overall message by
producing weaker endings for utterances or diminishing the crispness of certain sounds.
None of our subjects’ had above normal maximum phonation times this reveals an
opportunity for further investigation. Maximum phonation time provides information on
expiratory airflow during phonation requiring the upper airway to remain relatively open

and mobstructed. Were the reduced maximum phonation times related to the large size

62

of the subjects, and would a sample of OSA subjects with reduced size obtain similar
results?

The sample studied mostly showed characteristics of normal voice. The
perceptual difficulties noted with voice and the objective data indicating lower
fundamental frequency and reduced length of non-vibratory/vibratory productions for this
population may also be related to the reduced respiratory support and subject’s increased
size. The high number of subjects with low fundamental frequencies, especially in
women, may be related to each subject’s size and concurrent increased mass in the area
of the vocal folds, causing slower vibrations and lower pitch. Because none of the
subjects in this current study had above normal fundamental frequencies, additional
investigation in this area would be beneficial. For example, the use of reading as a
stimuli has demonstrated the most consistent measure of fundamental frequency and may
provide additional information (Fitch, 1990). A study looking at fundamental frequency
from a read sample, which is a more consistent measure (Fitch, 1990), may be helpful.
Also, investigation of appropriate subgroups within OSA may provide additional insight.
For example, subjects were not excluded from the study if they were smokers; this also
may have had an impact on respiratory and voice measures. Smoking is an irritant to
mucosal tissue and may cause redness, swelling and inflammation (Sataloff, 1991). The
edema is thought to increase vocal fold mass and change the vibratory movement of the
vocal folds. Information on smoking was gathered informally during the medical history
interview; however, it was not specifically questioned. There were 5 self-reported
smokers in our total group, 3 from the reference group and 2 from the CPAP group.

There were 4 male smokers and 1 female smoker. Sorensen and Horii (1984) fomd a

63

trend of lower fundamental frequencies, without statistical significance, with smokers
versus non-smokers. This pocket of subjects showed some interesting trends. When
reviewing data of these five subjects, it was fomd that they all had abnormal sufficient
inhalation, duration of exhalation and maximum phonation times. Four out of five of
these subjects had abnormal breath groups per minute. There was a strong tendency for
abnormal respiratory findings from this subset. Another subset of OSA subjects would
be those with gastroesophageal reflux disease (GERD). GERD is known to cause edema
in the pharynx/larynx and to contribute to voice and swallowing disorders (Koufman,
1991).

Nasalance could be affected by increased body weight causing additional airway
obstruction, especially in the oropharynx and nose area. This obstruction would typically
result in decreased nasalance scores. The subjects in the study showed a mix of normal
and low nasalance scores, an indication of obstruction. However, there was not a
significant difference between the reference and CPAP groups. This current study
supports the study by Salas-Provance and Kuehn (1990) who fomd aerodynamic
measurements pre- and post-UPPP surgery used to estimate resonance did not change.

With regard to swallowing, there were two areas that revealed difficulties. The
subjects (8) 23% self-rated themselves as having difficulty with swallowing. The second
area was that (12) 35% of the subjects initially had delayed swallows with solids and (7)
20% at the time of follow-up. This was similar to findings by Zohar et a1. (1998) which
fomd longer oral transit times for OSA subjects compared to healthy subjects or those
who have had UPPP intervention. The fmctional commmication measure regarding

swallowing and vocal quality judgment following eating and drinking were in the normal

64

range. However, the swallowing assessments were not done during a contextual meal
when a greater number of distractions occur and when there is less regulation of bite size
and speed. Based on the overall clinical swallowing data, there may be a subset of the
OSA population that may have mild swallowing difficulties. Further instrumental
investigation would be warranted.

Of interest was a subgroup of 5 subjects with very high BMI scores all above 42
(falling in the very obese range 35+) and relatively low AHI scores falling below 32
(mild 10 — 19 and moderate 20 — 39). In addition to decreased maximum phonation times
these subjects all showed abnormal speech production ratings as judged by a speech
pathologist. Four out of the five also had decreased fundamental frequency and
insufficient inhalation. These parameters may be fertile areas for further investigation of
relationships between communication and increased weight.

An additional part of this study was whether the treatment of CPAP appeared to
effect the communication and swallowing characteristics of the subjects. In most
instances there did not appear to be a change after the CPAP treatment. This confirms
the concept of a dynamic upper airway that is able to compensate for minor adjustments.
There was a tendency for improvement in the area of voice and swallowing with CPAP
use for a number of individuals. On follow-up there were (11) 58% of the individuals
from the CPAP group were placed in the normal range with the s/z task, previously 32%.
Improvement was noted in the promptness of their swallowing with solid foods for (4)
22% of the CPAP group. One possibility could be that with the shmting effect of CPAP,
there was less repetitive vocal abuse occurring in the upper airway from snoring and

obstructive sleep apnea. With CPAP forcing air through the nasal passage, there may be

65

more warming and humidification of air than when the air passes through the oral cavity
for obstructed non-CPAP breathing. A previous study has reported a decrease in
gastroesophageal reflux with CPAP use, possibly due to a reduction in symptoms that
may predispose individuals to nocturnal gastroesophageal reflux, e.g., lowered
intrathoracic pressure, increased arousal and movement (Kerr, Shoenut, Millar, Buckle,
and Kryger, 1992). Another study using MRI has shown that after CPAP use there is an
increase in airway volume (Ryan, Lowe, Li, and Fleetham, 1991). Both of these studies
may reflect a reduction in inflammation of the airway tissue. A reduction in
inflammation may relate positively to better voice production and swallowing comfort
and speed. A frequent anecdote from the CPAP subjects seen at the follow-up session
was that they reported drinking a great deal of water to combat the dryness caused by the
CPAP. This increase in hydration is a well-known and positive strategy for optimizing
voice and keeping the upper airway healthy. More and more CPAP machines also
feature humidification options. On the other hand, CPAP might increase evaporation so
increased hydration might mean no net gain of hydration; there may even be a loss in
hydration.

The speech-language pathologist’s current caseload may have some individuals
who have not been diagnosed as having sleep apnea, although signs and symptoms of
sleep apnea may be present (e.g., self-report of snoring, daytime sleepiness, choking and
awakening episodes at night). Knowledge of the signs and symptoms of sleep apnea can
alert the speech-language pathologist to initiate comseling with their patient and make an
appropriate referral to a specialist. Other health practioners would benefit from increased

awareness of subtle communication and swallowing deficits with the sleep apnea

66

population. However, with multiple medical co-morbitites their communication skills
and swallowing may be a lower priority issue to individuals with obstructive Sleep apnea.
Limitations of the Study

The subjects in this study were volmteers, thereby having a self-selection bias.
Either those with more concerns about their commmication and swallowing may have
volmteered, or conversely those with high confidence in their communication and
swallowing abilities may have participated.

The respiratory instrumental measures may be suspect because of the inconsistent
placement of the strain gauge resulting from the girth of several of the subjects. The
gauge was stretched to its maximum capacities at times and for one subject data were not
collected because the strain gauge did not fit aromd the subject. Conversely, with
increased body girth there could have been internal absorption of muscle movement by
the surromding body fat; therefore, the strain gauge did not sense the movement that may
have occurred. However, non-instrumental respiratory measures such as maximum
phonation time and sustained productions for [s] and [2] also demonstrated a reduction
from normal for a majority of subjects. The clinical tasks supported the instrumental
outcomes obtained with the strain gauge.

There were no controls made for the time of day during the experiment, as every
attempt was made to accommodate the subject’s schedules in order to gain their
participation. However, there may be an effect related to the time of day the data were
obtained. For example, many subjects had their initial testing done in the mid-evening

and follow-up typically late afternoon or early evening. The amomt of typical voice use

67

may have been different for these subjects and possibly greater amomts of vocal fatigue
at the initial testing.

Perceptual ratings of live voice instead of tape-recorded voice may also have
resulted in different ratings. The visual observation of the subjects may have influenced
the rating, as the size of the subject would have been taken into consideration. Additional
raters would have strengthened the reliability of the ratings. Also, there may be a
difference in the ratings if lay raters were used opposed to speech-language pathologists.

Data were not analyzed by age, and there may have been some contribution of
aging to some of the measures taken such as fundamental frequency, maximum
phonation time and other respiratory measures. Lynne-Davies (1977) reported
respiratory changes with aging, including less stability, less elasticity of the system, less
capacity and decreased muscle mass. Fmdamental frequency lowers in females and
increases in males with advanced age (Boone, Bayles, and Koopmann, (1982).

There may not have been a significant difference between the reference and
CPAP groups on the communication or swallowing tasks because the length of time that
the individuals were on the CPAP approximately 36 days was not a long enough time
period to create the changes in these parameters. There may also have been moontrolled
variables (e.g., compliance with CPAP, trial of nasal strips, change in medications) that
were maccomted for with the subjects.

Areas to be Explored in the Future
A set of 5 individuals with high BMI scores and low AHI scores highlight the

need for additional studies of speech and respiratory measures as they relate to size of an

68

individual. Recall that this subset of 5 individuals was all rated as having reduced
perceptual ratings in speech and decreased maximum phonation times.

The large number of subjects (24) 69% that did not pass the hearing screening
provokes the question of whether individuals that are prone to snoring have an increased
frequency of acquired loss of hearing. This population is known to have a high incidence
of mouth breathers possibly resulting in Eustachian tube dysfunction. Complete
audiological testing would be an area of follow-up that is warranted.

It would be interesting to explore the results of respiratory support exercises such
as working on diaphragmatic breathing or adjusting volume to see whether there would
be improved outcomes with this population. A protocol for muscle strengthening with
OSA patients who are mtreated or prior to the CPAP treatment option maybe beneficial
for this population to further investigate the effects on resonance, voice and swallowing.
Oral motor exercises that are tailored to strengthen the posterior tongue and palate are
supported by the literature as interventions with positive outcomes. Kuehn (1991) uses
nasal CPAP with contrasting oral and nasal somds to strengthen the velopharyngeal
muscles and assist with palatal closure when speaking. Wiltfang et a1. (1999) fomd
benefit to stimulation of the suprahyoid muscle on decreasing tongue collapse.

The relationship between CPAP use and hydration would also be important to
test. This would enable health care practitioners to be in a better position to educate
CPAP users on their need for fluids. Another interesting study would be the effect of
introducing a hydration protocol with this population to determine whether there would

be a positive impact on structure, physiological parameters, and voice.

69

More information on the relationship between communication (an activity that
occurs when the individuals are awake), and changes in physiology with sleep is
warranted. For example, does the sleep deprivation that occurs with OSA affect complex
neural timing that may affect physiological function used during communication?
Harrison and Home (1997) fomd a reduction in word fluency and articulation/intonation
changes following a night without sleep.

The group of sleep apnea individuals in this study revealed that there are
differences with the commmication abilities of those individuals with sleep apnea
compared to normal references cited in the literature. In conclusion, this study accepted
the null hypothesis that there would not be significant differences between the non-
treatrnent (reference group) and CPAP treatment groups. Use of the CPAP does not
significantly affect the commmication and swallowing abilities of the subjects. There is
a trend in the areas of resonance, voice and swallowing for improvement with CPAP use

possibly because of the effects of decreased edema.

70

 

APPENDICES

71

APPENDIX A

Baseline Questionnaire

Subject No.:
Date:
Is your breath strong enough to support speech production (voice does not somd weak,
there is enough air to produce a string of syllables)?

Yes No If no, explain:
Are all of your speech somds pronomced clearly (without lisp, slurring, slowness,
misarticulations)?

Yes No In no, explain:
Is the quality of your voice clear (without breathiness, hoarseness, strained quality)?

Yes No If no, explain:
Does the tone of you voice seem appropriate (without too much air coming out of your
nose or somding like you have a cold)?

Yes No If no, explain:
Do you swallow without difficulty (without coughing, choking, or frequent throat
clearing, wet voice after swallowing, the feeling of something getting stuck in your
throat)?

Yes No If no, explain:

Please add any descriptions you have about your speech, voice or swallowing?

72

 

APPENDIX B
Follow-up Questionnaire
Subject No.:
Date:
Have there been any changes in your respiration for speech?
Worse Same Better Please explain changes:
Have there been any changes in your pronmciation of somds?
Worse Same Better Please explain changes:
Have there been any changes in the quality of your voice?
Worse Same Better Please explain changes:
Has there been any change in the tone of your voice?
Worse Same Better Please explain changes:
Have you had any changes in your swallowing abilities?

Worse Same Better Please explain changes:

 

If you have noticed any other changes in your speech, vice or swallowing please

describe:

73

Level 0

Level 1

Level 2

Level 3

Level 4

Level 5

Level 6

Level 7

APPENDIX C
Functional Communication Measure:
Speech Production Disorder
Unable to test
Production of speech is unintelligible
Spontaneous production of speech is limited in intelligibility. Some
automatic speech and imitative words or consonant/vowel combinations

may be intelligible.

Spontaneous production of speech consists primarily of automatic words
or phrases with inconsistent intelligibility.

Spontaneous production of speech is intelligible at the phrase level in
familiar contexts; out of context, speech generally is unintelligible unless
self-cueing and self-monitoring strategies are applied.

Spontaneous production of speech is intelligible for meeting daily living
needs; out of context, periodic repetition, rephrasing, or provision of a one

is required.

Spontaneous production of speech is intelligible in and out of context, but
the production is sometimes distorted.

Production of speech is normal in all situations.

74

 

Level 0

Level 1

Level 2

Level 3

Level 4

Level 5

Level 6

Level 7

APPENDIX D
Functional Communication Measure
Voice Disorder
Unable to test

Voice production (pitch, quality, loudness) is non-functional for
communicating.

Voice production is functional for brief episodes; most communication
must be accomplished by non-vocal means.

Voice production is unreliable, although some vocal communication may
occur in limited contexts.

Appropriate voice production is limited. Self-monitoring and self- .
correcting skills are inconsistent. Voice quality is distracting to most A
listeners.

 

Appropriate voice production (pitch, quality, loudness) is consistent in
most contexts. Self-correcting skills are used appropriately. Voice quality
is distracting to some listeners.

Voice production is appropriate in most Situations, although minimal
difficulty may occur.

Voice production is normal for all speaking situations.

75

APPENDIX E

Resonance Scale

Denasal

Mild denasal

Normal Resonance

Mild hypemasal (somd specific)
Moderate hypemasal (intermittent)
Marked hypemasal

Severe hypemasal

76

 

Level 0

Level 1

Level 2

Level 3

Level 4

Level 5

Level 6

Level 7

APPENDIX F
Functional Communication Measure

Ability to Swallow Function

Unable to test
Swallowing is not functional.

Swallowing disorder prevents eating for all nutritional needs, but some
swallowing is possible.

Swallowing disorder prevents eating for a portion of nutritional needs and
one-to-one supervision is required for eating.

Swallowing disorder does not prevent eating to meet nutritional needs,
although general supervision is required to ensure use of compensatory

techniques.

Swallowing is functional to meet nutritional needs, although self-
monitoring and compensatory techniques are used.

Swallowing is functional for most eating activity, although mild difficulty
may occur periodically; additional time may be necessary for eating.

Swallowing is normal in all situations.

77

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

APPENDIX G Initial Raw Data
3
g c 5 .e a 2 e
g, s 2 8 _ s e 3; LI; é Ce; 3 ‘9 E L _ E
_e a 382‘: .3 egg d3 .2 A 2 § E‘s-Escher,
g g a L: a g E 5 a 8‘ 0.. 13 v E g 131-). d) S 9:). 6 g 2" g F; 0 g 0
C14) 9‘ Dd 9‘ Di ‘0- : C c 8 § 8 g 1:: o ’5‘ 8 an ‘101) E 0611 3 81) m L“ A “" m g A g
=0 m U U) 0 C B ‘1’ Cl) 4.) <1) .... "“ “’3 :3 <1.) cs ca ‘38 O 1 o 1 '
c. _: _: LLI LL] LIJ a, 2.; CHILEVC : .3 .2 E g... mu 2.. n o>4v>.535
8 0 0 (I) m m 5 .3 z: .3 -; 2 a, Q, ._ ._ =3 > c n. c H r: n. c ._ DD): 60,: mo 9.2
-—. 432 452 2 2 2 a: 8?, game _O‘DS :3 9‘ v ‘— 02 8 n8 3 :5 ::~ :— 5: 5—5
a ac aommm BE 6: Bessie a N o g 6285 as Ee-gfigfigo“0m
m mu. mu. 0 o o an: 6.5 Qm> >I—m m a u. a on?) NIL Zn. 4:13 4 OOOOEEEI
1 7 7 31 24 55 12.0 79.2 4.4 7 12 11 11 1 0 179.00 .43 0 15.31 63.05 11.14 6 0 0 0 0
2 7 7 29 24 53 12. 4.3 4.3 6 11 12 14 86 229.8 .593 2 12.14 43.54 12. 7 0 0 0 0
3 6 6 29 21 50 8. 93.3 6.1 6 22 20 21 95 239 7 .294 0 8.54 57.53 9.64 7 0 0 0 1
4 7 7 27 21 48 20.5 21.1 2.3 7 24 20 30 67 94.2 .568 0 11.54 59.78 16.98 7 0 0 0 0
5 7 7 29 24 53 14.5 86.2 3.8 7 28 16 40 4 120.7 .326 0 8.17 56.33 9.45 7 0 0 0 1
6 6 6 31 24 55 37 101. 1. 6 14 28 20 14 192.8 .428 0 11.66 61.98 18.14 6 0 0 0 1
7 6 6 27 21 48 20. 65.8 2.5 7 29 28 30 93 95.4 3.82 2 11.73 61.14 11.75 7 0 0 0 1
8 6 6 29 24 53 31.5 70. 1.2 5 25 27 25 108 103.2 .271 0 9.10 51.16 11.02 7 O 0 O 0
9 6 6 28 20 48 11.5 100. 4.6 6 24 22 18 1 22 104.3 .525 0 11.17 52.15 9.64 7 0 0 0 0
10 6 6 28 22 50 16. 25.8 3.3 7 13 26 22 1 18 90.1 .232 0 15.49 56.62 12.22 7 0 0 0 1
11 6 6 30 24 54 11.5 13. 4.5 6 23 25 20 1 25 170.8 .365 0 12.82 65.36 10.75 7 0 0 0 1
12 7 7 31 24 55 13. 50. 3.9 6 22 20 19 1 05 97.9 .607 0 13.63 48.33 18.92 7 0 0 0 2
13 6 6 29 21 5O 15. 26.7 3.3 5 10 24 10 2 4 84.8 .533 2 16.34 55.85 15.64 7 0 0 0 0
14 6 6 30 22 52 14. 7.1 3.1 5 39 33 28 1 18 116.6 .827 2 24.11 65.19 22.8 7 0 0 0 0
15 6 6 30 24 54 27.5 50.9 1.8 7 21 22 22 1 0 1 146.5 .964 0 23.76 63.87 29.12 7 0 0 0 0
16 6 6 31 23 54 9. 100. 6.4 7 2O 25 20 1 25 ' 100. 1.45 0 12.54 61.69 15.86 7 0 0 0 0
17 6 6 25 22 47 * * * 6 16 17 25 68 | 134.2 .340 2 15.65 49.72 17.11 7 0 0 0 0
18 7 6 31 24 55 13. 80.8 4.1 7 15 17 22 77 . 114.7 1.2 0 9.14 53.87 9.29 7 0 0 0 0
19 * * 3O 23 53 21.5 72.1 1.7 * 15 18 9 2. 5' 208 2.89 * 20.28 67.62 17.82 7 0 0 1 1
20 6 7 28 24 52 12. 69.6 4. 7 16 15 19 .79 3 122.8 .381 2 8.59 45.39 7.28 7 0 0 0 0
21 7 6 29 23 52 19.5 105.6 2.5 7 28 27 30 .9 1 88.7 .867 2 14.54 55.86 14.26 7 0 0 0 0
22 6 6 30 24 54 9. 29.4 5.1 7 25 32 24 1.33 1 111.7 .338 2 13.57 46.25 11.76 6 0 0 0 0
23 7 7 28 23 51 14. 82.1 3.7 6 9 14 10 1.4 3 222.6 .371 2 10.72 50.64 7.3 7 0 0 0 0
24 6 6 26 22 48 15. * 3.7 7 6 13 10 1.3 1 125.5 .798 2 28.99 68.17 26.36 7 0 0 0 0
25 7 7 28 23 51 18. 77.8 2.8 7 14 27 17 1.59 1 102.9 .219 2 14.43 33.29 13.74 7 O 0 0 0
26 7 7 31 24 55 23. 0.0 2.1 6 17 29 23 1.26 5 215.4 1.04 0 22.0 59. 29.83 7 0 0 0 0
27 7 7 28 24 52 12.5 48. 4.1 6 28 24 29 .83 99. .651 0 8.19 56.54 7.48 7 0 0 0 0
28 6 6 31 21 52 33. 103.4 1.2 6 18 10 15 .67 1 140.1 .578 0 18.66 53.64 22.09 7 0 0 0 *
29 7 7 30 24 54 13.5 30.8 3.7 7 24 10 19 .53 171.8 .833 2 8.78 60.63 10.59 7 0 0 0 1
30 6 6 27 22 49 16.5 43.8 3.1 7 13 18 13 1.38 79.1 .588 0 7.19 53.05 7.33 7 0 0 0 0
31 6 6 30 23 53 17.5 17.6 2.1 7 13 32 18 1.78 195.8 .998 2 10.3 62.17 9.2 7 0 0 0 0
32 7 7 31 22 53 11. 95.5 3.8 6 29 41 32 1.28 103.1 .478 2 13.13 58.11 16.45 7 0 0 0 0
33 6 6 24 20 44 20.5 97.4 2.2 7 15 12 18 .67 117.3 .774 0 21.63 58.75 38.83 7 0 0 0 1
34 6 6 31 24 55 15.5 70.1 2.3 6 27 35 29 1.21 95.8 .953 2 9.96 58.34 10.79 7 0 0 0 1
35 6 6 30 21 51 14.5 100. 3.6 5 20 15 18 .83 207.5 1.26 -1 11.46 64.41 18.43 7 0 0 1 2

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

* Missing data point

78

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

APPENDIX H Follow~up Raw Data
(1) Q)
60 § :5) E _ 8 E 73 E E 8 ’39: CE .2 _
8 g .3 8 8 OS :3 8 g, 5 8 23: E 8 8 6 E 8%: 82 “5 ‘3 “5%
c: O U) LL P :3 L4:: S V <1) ‘3 U) N .3 U) a) 5 S O > "T > O m g m m
52’ a 62.2.. 9 a: 53 = 'o 13 o a 8Eg%E% aamgrfie‘fi’aefisl
53.: E m we 9.. ~§ Sea. 2 a a -.: .2 a2 we are eseogagsase
8 o o m m ”3 5“ Fe! :3 <1) O-fi-— -— N > c953: —‘3 9‘5 72 00.: 00.: Geo 21.3
*2 92 02 E 2 E “5 873 ““92x853 «‘3 m v ‘- 08 8 “8 8 : :_. C~ E: Em
8 8080mwm 9.3-5 “I: ‘53? 59“08 ‘8 N o 3 8r8‘r 8‘- Srvi—“SESONC’g
m COLT-IUJLLO O O (:02 313.5 QLLI >LLEE’3U1 m E LL. ;; odd: N15: 233 Ad": :1 OO’UO’EEEVI
1 7 7 31 23 54 13.5 7.4 3.1 6 16 9 10 .9 234.2 .395 2 11.2 59.13 11.75 6 O 0 0 0
2 7 7 29 24 53 8. 93.8 6.7 6 16 17 17 1.0 233.1 .233 2 11.13 49.13 11.46 7 0 0 0 1
3 7 7 29 23 52 7. 100 7.7 6 24 25 2.. 1.14 213.8 .275 0 6.48 50.93 6.26 7 0 0 0 0
4 7 7 27 23 50 14.5 65.5 3.4 7 21 19 28 .68 101.3 .214 O 15.82 64.32 18.75 7 0 0 0 0
5 7 7 28 24 52 12.5 100. 3.9 7 36 14 38 .37 113.3 .324 0 7.31 56.31 5.61 7 0 0 0 1
6 6 6 31 24 55 27. 1.9 1.5 7 12 29 24 1.21 204.9 1.12 0 19.72 71.36 24.44 7 O 0 0 0
7 6 6 28 23 51 11. 86.4 4.7 6 29 24 23 1.04 95.1 .398 O 7.6 57.91 8.83 7 0 0 0 0
8 6 7 29 23 52 20 * 2.5 7 19 18 23 .78 143.5 .357 O 8.64 51.53 6.14 7 0 0 0 0
9 6 6 28 23 51 16.5 0.0 3.2 6 23 18 24 .75 122.8 .346 2 8.33 45.76 7.34 7 0 0 0 0
10 6 6 28 23 51 12.5 41.7 4.5 7 14 29 25 1.16 82.8 .179 0 12.9 58.49 12.77 7 0 0 0 2
11 6 7 30 23 53 11. 10. 4.6 6 26 19 26 .73 161.5 .168 2 5.09 58.49 5.14 7 0 0 0 0
12 6 6 31 24 55 15. 20. 3.5 7 17 17 17 1.0 103.7 .189 2 15.56 56.32 17.69 7 0 0 0 1
13 6 7 30 24 54 10. 100. 5.3 6 10 21 10 2.1 98. .173 2 10.05 53.67 9.71 6 0 0 0 0
14 6 6 30 23 53 11. 13.6 4.7 5 35 37 39 .95 100.2 .581 2 19.43 64.92 19.65 7 0 0 0 1
15 6 6 30 23 53 21.5 46.5 1.8 6 2O 13 22 .59 144.1 .221 O 25.61 65.44 36.21 7 0 0 O 0
l6 6 6 31 22 53 11.5 82.6 4.5 6 29 20 29 .69 102.7 .472 1 18.42 65.41 21.02 7 0 0 0 0
17 6 6 27 22 49 * * * 6 18 13 26 .62 118.5 .539 2 12.13 54.12 16.63 7 0 0 0 0
18 6 6 31 24 55 18. 55.6 2.9 7 10 38 38 1. 138.4 .854 0 10.49 65.17 9.87 7 0 0 0 0
19 7 7 29 23 52 11.5 0.0 4.1 4 12 16 12 1.33 235. 2.2 2 26.34 70.03 23.59 7 0 0 0 0
20 6 7 28 23 51 12.5 0.0 1.1 7 19 22 19 1.16 108.5 .627 O 12.2 50.17 10.78 7 0 0 0 0
21 6 6 30 24 54 13. 75. 4.1 7 27 29 26 1.12 87.7 .317 0 9.98 61.14 11.7 7 0 0 0 0
22 7 7 30 24 54 7. 78.6 8.6 6 30 32 33 .97 113.1 .382 2 14.41 44.13 10.74 7 0 0 0 0
23 6 7 28 22 50 26. * 1.6 7 8 10 12 .83 227.4 .368 2 12.94 67.98 10.53 7 0 0 0 0
24 6 6 26 23 49 14.5 13.8 3.5 7 11 10 10 1. 122.2 .714 2 22.68 68.82 19.14 7 0 0 0 0
25 7 7 28 24 52 20.50 80.5 2.3 7 12 23 18 1.28 105.0 .401 0 13.59 51.82 13.52 7 0 0 0 1
26 7 7 31 23 54 13.5 29.5 3.7 7 14 26 22 1.18 223. 1.35 0 11.44 49.66 10.0 7 0 0 0 0
2L 7 7 28 24 52 13.5 61.5 3.7 7 21 15 22 .68 86.6 .864 0 4.94 48.07 4.36 7 0 0 0 0
2i 6 6 31 I7 48 43.5 147.4 .8 6 21 9 15 .6 128. .889 2 11.12 51.12 26.77 7 0 0 0 0
29 7 7 30 24 54 12. 62.5 4.2 7 22 9 19 .47 183.4 .359 0 10.01 57.86 14.09 7 0 0 O 0
30 6 7 27 24 51 13.5 0.0 3.5 7 11 13 15 .87 92. .191 2 11.37 54.77 12.38 7 0 0 0 O
31 7 7 30 24 54 17.5 34. 2.1 7 14 21 14 1.5 195.2 .917 0 8.72 67.16 6.71 7 0 0 0 0
i 7 7 30 24 54 13.5 84. 3.5 7 24 34 29 1.17 94.1 .557 0 20.7 66.9 20.19 7 0 0 0 0
i 6 6 24 22 46 21.5 102.4 1.1 7 9 19 16 1.19 124.6 .24 2 19.88 59.72 23.55 7 7 O 0 0
34 6 6 31 24 55 9.5 31.6 4.1 6 29 39 29 1.34 87.5 .501 1 7.78 58.98 10.14 7 O 0 0 0
35 6 6 30 19 49 17. 35.3 2.9 5 20 9 15 .6 221.7 1.19 1 21.78 75.72 32.47 7 1 0 l 2

 

* Missing data point

79

 

 

REFERENCES

80

.K‘I

 

REFERENCES

American Speech-Language-Hearing Association (1995) Functional Communication
Measures (FCMs). Meeting the marlaged care challenge strzflgies for professionals and
the profession. Rockville, MD.

 

American Thoracic Society (1994) Indications and standards for use of Nasal Continuous
Positive Airway Pressure (CPAP) in Sleep Apnea Syndromes. American Journal
Respiratory Critical Care Medicine. 150, 1738 —- 1745.

Behavioral Risk Factor Surveillance System (BRF SS) (2001) Prevalence of healthy
lifestyle characteristics — Michigan 1998 and 2000. Morbidity and Mortality Weefly
Report, 50, 35, 758 — 761.

Boone, D.R., Bayles, K.A. and Koopmann, C.F. Jr. ( 1982) Communicative aspects of
aging. Otolarvngologic Clinics of North America, 15, 2, 313 - 327

 

Bradley, T.D., Brown, 1.6., Grossman, R.F., Zamel, N., Martinez, D., Phillipson, EA,
and Hoffstein, V. (1986) Pharyngeal size in snorers, nonsnorers, and patients with
obstructive sleep apnea. New England Journal of Medicine. 315, 21, 1327 - 1331.

Brancewicz, T.M. and Reich, AR. (1989) Speech rate reduction and “nasality” in normal
speakers. JOILHQ of Speech and Hearing Research. 32, 837 -— 848.

Brander, P.E., Soirinsuo, M., and Lohela, P. (1999) Nasopharyngeal symptoms in
patients with obstructive sleep apnea syndrome. Respiration, 66, 128 — 135.

Bray,G.A., Jordan, H.A., and Sims, EA. (1976) Evaluation of obese patients I, an
algorithm Journal of the American Medical Association. 14, 235, 1487 - 1491.

Brosch, S., Matthes, C., Pirsig, W., and Verse, T. (2000) Uvulopalatopharyngoplasty
changes fundamental frequency of the voice — a prospective study. The Journ_al of

Lgngology and Otology, 114, 113 - 118.

Cassell, MD. and Elkadi, H. (1995) Anatomy and physiology of the palate and
velopharyngeal structures. In R]. Shprintzen and J. Bardach (Eds), Cleft Palate Speech

Management A Multidisciplinag Approach, (pp. 45 - 62) St. Louis, MO: Mosby-Year
Book, Inc.

Centers for Disease Control (1999) Chronic diseases and their rig factors: the n_ation’s

leading cause of death: a report with expanded state-bv-sgite information Atlanta,
Georgia: US Department of Health and Human Services.

81

Coleman, RP. and Sly, DE. (1991) Preoperative and postoperative voice analysis of
uvulopalatopharyngoplasty patients. Archives of Otolmgologv — Head and Neck
Surgegg, 117, 1345 — 1349.

Colton, RH. and Casper, J. K. ( 1990) Understanding Voice Problems. A Physiological
Perspective for Diagnosis and Treatment. Baltimore, MD: Williams & Wilkins.

Croft, CB. and Golding-Wood, D.G. (1990) Uses and complications of
uvulopalatopharyngoplasty. The Journal of Larypgology and Otology, 104, 871 — 875.

Dalston, R.M., Warren, D.W., and Dalston, ET. (1991) Use of Nasometry as a diagnostic r
tool for identifying patients with velopharyngeal impairment. Cleft Palate-Craniofacial
Journal 28, 2, 184 - 189.

 

Daniels, S.K., Brailey, K., Priestly, D.H., Herrington, L.R., Weisberg, LA, and Foundas,
A. L. (1998) Aspiration in patients with acute stroke. Archives of thsicfi Medicine and
Rehabilitation 79, 14 — l9.

 

 

Fairbanks, D.N.F. (1990) Uvulopalatopharyngoplasty complications and avoidance 6
strategies. Otolmngology, Head and Neck Surgeg, 102, 3, 239-24.

Finkelstein, Y., Meshorer, A., Talmi, Y.P., Zohar, Y., Brenner, J ., and Gal, R., (1992)
The riddle of the uvula. Otolarvngologv Head Neck Surgeg, 107, 444 - 450.

Finkelstein, Y., Talmi, Y., and Zohar, Y. (1988) Readaptation of the velopharyngeal
valve following the uvulopalatopharyngoplasty operation. flastic and Recon_structive
Surgeg, 82, 1, 20 - 27.

Fitch, J .L. (1990) Consistency of fundamental frequency and perturbation in repeated
phonations of sustained vowels, reading, and connected speech. Journ_al of Speech and
Hearing Disorders, 55, 360 — 363.

Gilbert, HR. (1975) Speech characteristics of miners with black lung disease
(pneumoconiosis). Journal of Communication Disorders 8, 129 - 140.

 

Gleeson, K., Zwillich, C.W., Braier, K., and White, DP. (1986) Breathing route during
sleep. American Review of Respiratory Disease, 134, 115 -120.

Goebel, MD. (1992) Cafet R-for-Voice user’s Manual for Installation and Therapy.
Annandale Fluency Clinic. Annandale, VA.

Goodglass, H. and Kaplan, E. (1983) Boston Diagnostic Aphasia Examination Stimulus
Materials: The Assessment of Aphasia and RelaLted Disorders (2nd Ed.) Baltimore, MD:
Williams and Wilkins/Lea and Febiger.

82

Haavisto, L. and Suonpaa, J. (1994) Complications of uvulopalatopharyngoplasty.
Clinical Otolagyngology, 19, 243-247.

Harrison, Y. and Home, J .A. (1997) Sleep deprivation affects speech. Sleep, 20, 10, 871
- 877.

Hirano, M., Koike, Y., and von Leden, H. (1968) Maximum phonation time and air usage
during phonation. Folia Phoniatrica. 20, 185-201

Hixon, T.J. (1973) Respiratory function in speech. Norm_al Aspects of Speech. Hearing.
and Language. (pp 73 -125) Minifie, F .D., Hixon, T.J. and Williams, F. (Eds).
Englewood Cliffs, NJ: Prentice-Hall, Inc.

Hoit, J. and Hixon, T. (1975) Age and speech breathing. Journal of Speech and Hearing
Research, 30, 351-366.

Horii, Y. (1979) Fundamental frequency perturbation observed in sustained phonation.
Journal of Speech and Hearing Research, 22, 5 - 19.

Homer, R.L. , Mohiaddin, R.H., Lowell, D.G., Shea, S.A., Burman, E.D., Longmore,
DB, and Guz, A. (1989) Sites and sizes of fat deposits around the pharynx in obese

patients with obstructive sleep apnea and weight matched controls. European Respiratory
Journal 2, 613 — 622.

 

Ichikawai, T., Komado, J ., Horiucki, M., and Matsumoto, N. (1995) Influence of
alterations in the oral environment on speech production. Journal of Oral Rehabilitation
22, 295 — 299.

 

Janson, C., Gislason, T., Bengtsson, H., Eriksson, G., Lindberg, E., Lindholm, C.E.
Hultcrantz, E., Hetta, J ., and Boman, G. (1997) Long-term follow-up of patients with
obstructive sleep apnea treated with uvulopalatopharyngoplasty. Archives of

Otolmngology Head and Neck Surgeg, 123, 257 — 262.

Karnell, MP. (1995) Nasometric discrimination of hypemasality and turbulent nasal
airflow. Cleft Palate —Craniofacial Journal 32, 2, 145 — 148.

 

Kay Elemetrics Corp. Msometer TM Applications Notes. Pine Brook, NJ
Kay Elemetrics Corp. Visi-Pitch R Applications Manual. Pine Brook, NJ
Kerr, P., Shoenut, J .P., Millar, T., Buckle, P., and Kryger, M.H. (1992) Nasal CPAP

reduces gastroesophageal reflux in obstructive sleep apnea syndrome. Chest, 101, 1539 —
1544.

83

. I- 5 £41.11“!

 

5.1— _—

Koskenvuo, M., Kapno, J ., Partinen, M., Langinuainio, H., Sarna, S., and Heikkila, K
(1985) Snoring as a risk factor for hypertension and angina pectoris. Lancet 1, 893 —-
896.

 

Koufman, J .A. (1991) The otolaryngologic manifestations of gastroesophageal reflux
disease (GERD): a clinical investigation of 225 patients using ambulatory 24-hour pH
monitoring and an experimental investigation of the role of acid and pepsin in the
development of laryngeal injury. Lmngoscopy, 101, (supplement 53).

Krespi, Y.P., Pearlman, S. J ., and Keidar, A. (1994) Laser - assisted uvula-palatoplasty
for snoring. The Joum_alfiof Otolaryngology, 23, 5, 328 - 334.

Kuehn, DP. (1991) New therapy for treating hypemasal speech using continuous positive
airway pressure (CPAP). Plastic and Reconstructive Surgery. 88, 6, 959 —966.

Lai, V.W.S. and Corey, J .P. (1993) The objective assessment of nasal patency. ENT
Journal 72, 6, 395 — 400.

 

 

Levring-Jaghagen E., Nilsson, ME, and Isberg, A. (1999) Persisting dysphagia after
uvulopalatoplasty performed with steel scalpel. La_ryngoscope, 109, 1, 86 — 90.

Linden, P. and Siebens, AA. (1983) Dysphagia: Predicting laryngeal penetration.
Archivea of ths_i_cal Medicine apd Rehabilitation, 64, 281 - 284.

Logemann, J .A. (1983) Evaluation aid trement of swallowing disorder., San Diego,
CA: College-Hill Press, Inc.

Lombard, R.M. Jr. and Zwillich, CW. (1985) Medical therapy of obstructive sleep apnea.
Medical Clinics of North America,_69, 6, 1317 — 1335.

 

Lowe, AA. (1990) The tongue and airway. Otolarvngologic Clinics of North America.
23, 4, 677 — 698.

Lugaresi, E., Cirignotta, F., and Montagna, P. (1988) Pathogenic aspects of snoring and
obstructive apnea syndrome. Schweiz. Med Wschr, 118, 1333 - 1337.

Lynn-Davies, P. (1977) Influence of age on the respiratory system. Geriatrics 32, 57 —
60.

 

Mahan, KL and Escott-Strump, S. (Eds). (2000) Krause’s: Food. Nutrition. a_n_d Diet
Therapy (9th ed.). Philadelphia, PA: W.B. Saunders Company.

McWilliams, B.J., Bluestone, CD, and Musgrave, RH. (1969) Diagnostic implications
of vocal cord nodules in children with cleft palate. Lmngoscope, 79, 2072 - 2080.

84

McWilliams, B.J., Morris, H.L., and Shelton, R.L. (1990) The nature of the
velopharyngeal mechanism. Cleft PaLate Speech. 2nd Ed. B.C. Becker Inc. Philadelphia,
PA.

Monoson, PK. and Fox, AW. (1987) Preliminary observation of speech disorder in
obstructive and mixed sleep apnea. Chest, 92, 4, 670 - 675.

Moon, J. and Kuehn, D. (1996) Anatomy and physiology of normal and disordered
velopharyngeal function for speech. National Center for Voice and Speech. 9, 143 - 158.

Murry, T. and Bone, RC. (1989) Acoustic characteristics of speech following
uvulopalatopharyngoplasty. Lgngoscope, 99, 1217 — 1219.

Nakai, K., Sakakura, A., Takahashi, H., Sadaoka, T., and Kakitsuba, N. (1996)
Articulation after uvulopalatopharyngoplasty. European Archives of

Otorhinolagngology, 253, 417 — 420.

Neito, J .F., Young, T.B., Lind, B.K., Shahar, E., Samet, J .M., Redline, S., D’Agostino,
R.B., Newman, A.B., Lebowitz, M.D., and Pickering, T.G. (2000) Association of sleep-

disordered breathing, sleep apnea, and hypertension in a large community-based study.
Journal of the American Medical Association 283, 1829 —l836.

 

Nishio, J ., Matsuya, T., Ibuki, K., and Miyazaki, T. (1976) Roles of the facial,
glossopharyngeal and vagus nerves in velopharyngeal movement. Cleft Palate Joumal,
13, 201 - 214.

Nishio, J ., Matsuya, T., Machida, J ., and Miyazaki, T. (1976) The motor nerve supply of
the velopharyngeal muscles. Cleft Pagte Journal, 13, 2O - 30.

Oliver, R.G and Evans, SP. (1986) Tongue size, oral cavity size and speech. The Angle
Orthodontist. 7, 234 — 243.

Ono, T., Lowe, A.A., Ferguson, K.A., and Fleetham, J .A. (1996) Associations among
upper airway structure, body position and obesity in skeletal class 1 male patients with

obstructive sleep apnea. American Journal of Orthodonics and Dentofacial Orthopedics.
109, 6, 625 — 634.

 

Parker, VS. (1995) Local West Michigan Nasalance Norms. Unpublished manuscript.
Peiser, J ., Lavie, P., Ovnat, A., and Charuzi, I. (1984) Sleep apnea syndrome in the
morbidly obese as an indication for weight reduction surgery. Annuals of Surgery. 199,
1 12 - l 15.

Phillips, D., Cook, Y., Schmitt, F ., and Berry, D. (1989) Sleep apnea: Prevalence of risk
factors in a general population. Southern Medical Journ,a_lL 82, 9, 1090 —— 1273.

85

 

Phillipson, EA. (1993) Sleep apnea - A major public health problem. The New
Egland Journal of Medicine. 328, 17, 1271 -1273.

Prichard, A.J.N., Marshall, J ., Ahmed, A., Thomas, R.S.A., and Hanning, CD, (1994)

Uvulopalatopharyngoplasty; The Leicester experience. The Journal of Lagngology and
Otology, 108, 649 — 652.

Reeves-Hoche, M.K., Meek, R., and Zwillich, CL. (1994) Nasal CPAP: An objective
evaluation of patient compliance. American Journal of Respiratorv Criticg Care
Medicine 149, 149 - 154.

 

Robbins, J. (1985) Swallowing and speech production in the neurologically impaired T—
adult. Seminars in Speech and Language, 6, 4, 337 - 349.

Robinson, R.W., White, DP, and Zwillich, CW. (1985) Moderate alcohol ingestion
increases upper airway resistance in normal subjects. American Review of Respiratog
Disease 132, 1238 - 1241.

 

 

Rodenstein, D.O., Dooms, G., Thomas, Y., Liistro, G., Stanescu, D.C., Culee. C., and ;—
Aubert-Tulkens, G. (1990) Pharyngeal shape and dimensions in healthy subjects,
snorers, and patients with obstructive sleep apnoea. Thorax 45, 722 — 727.

 

Rollheim, T., Osnos, T., and Miljeterg, H. (1997) The relationship between obstructive
sleep apnea and body mass index. Clinical Otolgyngology, 22, 419 -— 422.

Ruscello, D.M., Tekieli, ME, and Van Sickels, J .E. (1985) Speech production before
and after orthognathic surgery: a review. Oral Surgeg, 59, 10 - 14.

Ryan, C.F, Lowe, A., Li, D., and F leetham, J .A. (1991) Magnetic resonance imaging of
the upper airway in obstructive sleep apnea before and after chronic nasal continuous
positive airway pressure therapy. American Review of Respiratory Diseasa, 144, 939 —
944.

Salas-Provance, MB. and Kuehn, D. (1990) Speech status following
uvulopalatopharyngoplasty. Chest, 97, l, 111 — 117.

Sataloff, RT. (1991) Professional voice: The science and art of clinical care, New York,
NY: Raven Press

Seaver, E.J., Dalston, R.M., Leeper, H.A., and Adams, L.E. ( 1991) A study of
nasometric values for normal nasal resonance. Journal of Speech and Hearing Research,
34, 715 — 721.

Sforza, E., Addati, G., Cirignotta, F., and Lugaresi, E. (1994) Natural evolution of sleep

apnoea syndrome: A five year longitudinal study. European Respiratory Journya_l, 7, 1765
- 1770.

86

Shelton, K.E., Woodson, H., Gay, S., and Suratt, PM. (1993) Pharyngeal fat in
obstructive sleep apnea. American Review of Respiratory Disease; 148, 462 — 466.

Sher, AB. (1990) Obstructive sleep apnea syndrome: A complex disorder of the upper
airway. Otolmngologic Clinics of North America, 23, 4, 593 - 608.

Sher, A.E., Thorpy, J .J ., Shprintzen, R.J., Spielman, A.J., Burack, B., and McGregor,
PA. (1985) Predictive value of Muller maneuver in selection of patients for
uvulopalatopharyngoplasty. Lmngoscope, 95, 1483 - 1487.

Shils, M.E., Olson, J .A., and Shike, M. (1994) Modern Nutrition in Health and Disw
Philadelphia, PA: Lea and Febiger.

Siefert, E., Ronte, C., Riebandt, M., Lamprecht-Dinnessen, A., and Bollmann, F. (1994)
Can dental prosthesis influence vocal parameters? Journal of Prosthetic Dentistry. 81,
579 - 555.

Skolniok, ML. and Cohn, ER. (1989) VideofluorsoscopLStudies of Speech in Patients
With Cleft Palate New York, NY: Springer-Verlag.

 

Sloan, ER and Shapiro, GM. (1995) Obstructive sleep apnea in a consecutive series of
obese women. International Joumg of Eating Disorders. 17, 2, 167 — 173.

Sorensen, D. and Horii, Y. (1984) Cigarette smoking and voice fundamental frequency.
Journal of Communication Di§orders. 15, 2, 135 - 144.

St. Louis, KC. and Ruscello, D.M. (1987) Oral Speech Mechanism Screening
Examination —Revised. Pro-Ed, Austin TX.

Stauffer, J .L., White, D.P. , and Zwillich, CW. (1990) Pulmonary function in obstructive
sleep apnea. Chest, 97, 302 - 307.

Stradling, J .R. and Crosby, J .H. (1991) Predictors and prevalence of obstructive sleep
apnoea and snoring in 1001 middle age men. Thorax 46, 85 — 90.

 

Strohl, K.D. and Redline, S. (1986) Nasal CPAP therapy, upper airway muscle activation,
and obstructive sleep apnea. American Review Respiratory Disease. 134, 555 — 558.

Sturm, R. and Wells, KB. (2001) Does obesity contribute as much to morbidity as
poverty or smoking? Public Health. 115, 3, 339 — 235.

Troell, R.J., Riley, R.W., Powell, N.B., and Li, K. (1998) Long-term results of surgical

management of sleep disordered breathing. Otolgyngologic Clinics of North America.
31, 6, 1031 — 1035.

87

Waller, RC. and Bhapal, RS. (1989) Is snoring a cause of vascular disease? Lancet, 1,
143 — 146.

White, D.P., Lombard, R.M., Cadieux, R.J., and Zwillich, CW. (1985) Pharyngeal
resistance in normal humans: influence of gender, age, and obesity. Journal of Applied
Physiology, 58, 2, 365-371.

Wiegand, L. and Zwillich, CW. (1994) Obstructive sleep apnea. Diseasefi-a—Month. XL,
4, 197 — 252.

Wiltfang, J ., Klotz, S., Wiltfang, J., Jordon, W., Cohrs, S., Engelbe, W., and Hajak, G. 5!
(1999) First results on daytime submandibular electrostimulation of suprahyoidal muscles
to prevent night-time hypopharyngeal collapse in obstructive sleep apnea syndrome.
Interaatiorall Journal of Oral Maxillofacial Surgeg, 28, 21 — 15.

Wilson, D. (1971) The voice-disordered child. L_anguage. Speech and Hearing Services
in the Schools. 2, 14 - 22.

 

Yoshida, K., Thumm, J ., and Siebert, GK. (1995) Kauund zungenmuskelaktivitat bei
schlafapnoikem und schnarchern. Dtsch Zahnarztlz 50, 387 — 389.

 

Young, T., Palta, M., Dempsey, J ., Skatrud, J ., Weber, S., and Badr, S. (1993) The
occurrence of sleep-disordered breathing among middle-aged adults. New England
Journal of Medicine. 328, 1230 — 1235.

Zamarron, C., Gude, F ., Otero, Y., Alvarez, J .M., Golpe, A., and Rodriguez, J .R. (1999)
Prevalence of sleep disordered breathing and sleep apnea in 50- to 70-year—old
individuals. A survey. Respiration, 66, 4, 317 - 322.

Zohar, Y., Finkelstein, Y., Talmi, Y.P., and Bar-Ilan, Y. (1991)
Uvulopalatopharyngoplasty: evaluation of postoperative complications, sequelae, and
results. La_ryngoscope, 101, 775 — 779.

Zohar, Y., Grusko, I., Sulkes, J ., and Melloul, M. (1998) Oropharyngeal scintigraphy: A
computerized analysis of swallowing in patients with obstructive sleep apnea.

La_ryngoscope, 108, 37 — 41.

88

IIIIIIIIIIIIIIIIIIIII

1m131attttttttttt11111111II