USF Libraries
USF Digital Collections

Are Respiratory Behaviors Affected in Individuals With Adductor Spasmodic Dysphonia?

MISSING IMAGE

Material Information

Title:
Are Respiratory Behaviors Affected in Individuals With Adductor Spasmodic Dysphonia?
Physical Description:
Book
Language:
English
Creator:
Biedess, Katie
Publisher:
University of South Florida
Place of Publication:
Tampa, Fla
Publication Date:

Subjects

Subjects / Keywords:
Inductive plethysmography
Respiration
Laryngeal disorders
Voice disorders
Respitrace
Dissertations, Academic -- Audiology -- Masters -- USF   ( lcsh )
Genre:
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Abstract:
ABSTRACT: Adductor spasmodic dysphonia (ADSD) is a focal dystonia that is characterized by voice breaks due to involuntary contractions of the adductor muscles of the vocal folds. These spasms can interfere with the coordination and balance of the respiratory and phonatory systems interfering with normal voice production. Disruptions in normal respiratory behaviors are well documented in inviduals with laryngeal disorders, including ADSD. Previous research regarding respiratory processes in ADSD has focused on airflow and pressure; however, there are many other parameters that have not been considered and may shed new light on the respiratory behaviors of individuals with ADSD. Therefore, the current pilot study attempted to determine if individuals with ADSD differed from controls in various breathing parameters while engaged in conversational and reading tasks.Thirty individuals were tested; fifteen in the ADSD group and fifteen in the age- and gender-matched control group.^ ^Respitrace, an inductive plethysmography device, calculated 14 different respiratory measures related to volume, timing, thoracic displacement and respiratory efficiency. The results of the study indicated that various significant differences existed between groups. Those with ADSD were found to have statistically higher ventilation rates, a greater frequency of breaths per minute, a higher degree of muscular inefficiency/breathlessness and labored breathing. These results indicated that individuals with ADSD suffered from disordered breathing due to the neurologically related obstruction at the level of the larynx. Differences according to task were also found. Specifically, the rib cage contributed to a lesser extent in voice production and the participants utilized longer inspiratory times, exhaled a larger volume of air and took longer to reach peak expiratory flow during conversational tasks when compared to reading tasks.^ ^These differences were attributed to a higher cognitive-linguistic demand required during conversational speech. Overall, the results of this study have many clinical implications. Most importantly, these findings support the idea that individuals with ADSD may experience difficulties with respiration as the effects of their Botox injection begin to wear off. Further research is needed with regards to the effects laryngeal spasms have on other respiratory behaviors.
Thesis:
Thesis (M.S.)--University of South Florida, 2006.
Bibliography:
Includes bibliographical references.
System Details:
System requirements: World Wide Web browser and PDF reader.
System Details:
Mode of access: World Wide Web.
Statement of Responsibility:
by Katie Biedess.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains 76 pages.

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 001920442
oclc - 189810195
usfldc doi - E14-SFE0001846
usfldc handle - e14.1846
System ID:
SFS0026164:00001


This item is only available as the following downloads:


Full Text
xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam Ka
controlfield tag 001 001920442
003 fts
005 20080109130802.0
006 m||||e|||d||||||||
007 cr mnu|||uuuuu
008 080109s2006 flu sbm 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0001846
035
(OCoLC)189810195
040
FHM
c FHM
049
FHMM
090
RF290 (ONLINE)
1 100
Biedess, Katie.
0 245
Are Respiratory Behaviors Affected in Individuals With Adductor Spasmodic Dysphonia?
h [electronic resource] /
by Katie Biedess.
260
[Tampa, Fla] :
b University of South Florida,
2006.
3 520
ABSTRACT: Adductor spasmodic dysphonia (ADSD) is a focal dystonia that is characterized by voice breaks due to involuntary contractions of the adductor muscles of the vocal folds. These spasms can interfere with the coordination and balance of the respiratory and phonatory systems interfering with normal voice production. Disruptions in normal respiratory behaviors are well documented in inviduals with laryngeal disorders, including ADSD. Previous research regarding respiratory processes in ADSD has focused on airflow and pressure; however, there are many other parameters that have not been considered and may shed new light on the respiratory behaviors of individuals with ADSD. Therefore, the current pilot study attempted to determine if individuals with ADSD differed from controls in various breathing parameters while engaged in conversational and reading tasks.Thirty individuals were tested; fifteen in the ADSD group and fifteen in the age- and gender-matched control group.^ ^Respitrace, an inductive plethysmography device, calculated 14 different respiratory measures related to volume, timing, thoracic displacement and respiratory efficiency. The results of the study indicated that various significant differences existed between groups. Those with ADSD were found to have statistically higher ventilation rates, a greater frequency of breaths per minute, a higher degree of muscular inefficiency/breathlessness and labored breathing. These results indicated that individuals with ADSD suffered from disordered breathing due to the neurologically related obstruction at the level of the larynx. Differences according to task were also found. Specifically, the rib cage contributed to a lesser extent in voice production and the participants utilized longer inspiratory times, exhaled a larger volume of air and took longer to reach peak expiratory flow during conversational tasks when compared to reading tasks.^ ^These differences were attributed to a higher cognitive-linguistic demand required during conversational speech. Overall, the results of this study have many clinical implications. Most importantly, these findings support the idea that individuals with ADSD may experience difficulties with respiration as the effects of their Botox injection begin to wear off. Further research is needed with regards to the effects laryngeal spasms have on other respiratory behaviors.
502
Thesis (M.S.)--University of South Florida, 2006.
504
Includes bibliographical references.
516
Text (Electronic thesis) in PDF format.
538
System requirements: World Wide Web browser and PDF reader.
Mode of access: World Wide Web.
500
Title from PDF of title page.
Document formatted into pages; contains 76 pages.
590
Adviser: Ruth Huntley Bahr, Ph.D.
653
Inductive plethysmography.
Respiration.
Laryngeal disorders.
Voice disorders.
Respitrace.
690
Dissertations, Academic
z USF
x Audiology
Masters.
773
t USF Electronic Theses and Dissertations.
4 856
u http://digital.lib.usf.edu/?e14.1846



PAGE 1

Are Respiratory Behaviors Affected In Individuals With Adductor Spasmodic Dysphonia? by Katie Biedess A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science Department of Communicati on sciences and disorders College of Arts and Sciences University of South Florida Major Professor: Ruth Huntley Bahr, Ph.D. Marion B. Ridley, M.D. Darla Freeman-LeVay, M.A. Date of Approval: November 3, 2006 Keywords: inductive plethysmography, respirati on, laryngeal disorders, voice disorders, respitrace Copyright 2006 Katie Biedess

PAGE 2

Acknowledgements I thank the following individuals for their assistance in the completion of this study. Dr. Ruth Bahr, my thesis advisor, fo r all your generous support. You provided me a great deal of academic and emotional encour agement, that without, I would have never completed this study. Thank you for keeping me motivated and pushing me towards the goal. Mrs. Freeman-LeVay, I am appreciativ e of your support and my experiences with you in voice treatment. Thank you Dr. Ridley for providing me the opportunity to study these individuals. I am gracious to both of you for taking the time to help me with the thesis. Our discussion proved invaluable to my understanding of the im plications of this research on spasmodic dysphonia.

PAGE 3

Dedication I would like to dedicate th is thesis to my family. Mom and Dad, your unyielding encouragement and unconditional love provided me the support to pursue this degree and complete this study. Scott and Julie, tha nk you for believing in me and reassuring me whenever I needed it. Millicent, without your constant companionship, this journey would have been a lonely one. Katie Biedess

PAGE 4

i Table of Contents List of Tables iii List of Figures iv Abstract vi Introduction 1 Instrumentation for Measuring Breathing 4 Breathing with a Disordered Larynx 10 Airflow 11 Lung Volumes 12 Subglottal Pressure 14 Physical Respiratory Movements 15 Spasmodic Dysphonia 16 Respiration in Adductor Spasmodic Dysphonia 18 Statement of the Problem 21 Methods 23 Participants 23 Materials 24 Procedures 27 Measured Breathing Parameters 29 Volume Measures 29 Inspiratory volume (ViVol) 30 Expiratory volume (VeVol) 30 Ventilation (Vent) 30 Timing 30 Breaths per minute (Br/M) 31 Inspiratory time (Ti) 31 Expiratory time (Te) 31 Total breath time (Tt) 31 Fractional inspiratory time (Ti/Tt) 31 Time to reach peak expiratory flow (PefTTe) 31 Thoracic Displacement 32 Percentage of rib cage contribution (%RC) 32 Labored breathing index (LBI) 32 Respiratory Efficiency 33 Rapid shallow breathing index (F/Vt) 33

PAGE 5

ii Peak inspiratory flow (PifVt) 33 Respiratory efficiency and breathlessness (VePif) 33 Data Reduction 34 Reliability 35 Statistical Analysis 36 Results 37 Volume 37 Timing 40 Thoracic Displacement 45 Respiratory Efficiency 46 Summary of Findings 49 Discussion 51 Volume 51 Timing 55 Thoracic Displacement 57 Respiratory Efficiency 61 Conclusions 64 Clinical Implications 65 Strengths of the Present Study 67 Limitations of the Present Study 68 Directions for Future Research 70 A Final Word 72 References 73

PAGE 6

iii List of Tables Table 1. A table re presenting the randomization of sp eaking tasks presented to participants. 28

PAGE 7

iv List of Figures Figure 1. An example of a waveform produced by a control participant as viewed via the RespiEvents software. 26 Figure 2. Chart of the mean inspiratory volumes compared between partic ipant group and speaking task. 38 Figure 3. Chart of the mean expiratory volume averaged across speaking task. 39 Figure 4. A chart of the mean ventilation (Vent) rate compared across participant groups. 39 Figure 5. A chart of average breaths per minute compared across participant group. 41 Figure 6. A chart of the av erage inspiratory time compared across speaking tasks. 41 Figure 7. A chart of the mean expiratory times compared between participant group and across speaking tasks. 42 Figure 8. A chart of the aver age total breath duration compared acr oss participant group and speaking task. 42 Figure 9. A chart of the averag e percentage of inspiratory time compared across gender. 43 Figure 10. A chart of the average percentage of expiratory time to reach peak expiratory flow compared across speaking tasks. 44 Figure 11. A chart showing the pe rcentage of rib cage contribution across speaking tasks. 45 Figure 12. A chart of the index of labored breathing compared across participant group. 46 Figure 13. A chart demonstrating the values for peak inspiratory flow compared across participant group and speaking task. 47

PAGE 8

v Figure 14. A chart depicting the differences in respiratory muscular efficiency and breathlessness compared across speaking tasks and participant group. 48 Figure 15. A chart showing the di fference in respiratory muscular efficiency and breathlessness compared across participant group. 48 Figure 16. An example of a RespiEvents waveform produced by a participant with ADSD. 60

PAGE 9

vi Are Respiratory Behaviors A ffected In Individuals With Adductor Spasmodic Dysphonia? Katie Biedess ABSTRACT Adductor spasmodic dysphonia (ADSD) is a focal dystonia that is characterized by voice breaks due to involunt ary contractions of the adductor muscles of the vocal folds. These spasms can interfere with th e coordination and balance of the respiratory and phonatory systems interfering with norma l voice production. Disruptions in normal respiratory behaviors are well documented in inviduals with laryngeal disorders, including ADSD. Previous research rega rding respiratory processes in ADSD has focused on airflow and pressure; however, there are many other parameters that have not been considered and may shed new light on th e respiratory behaviors of individuals with ADSD. Therefore, the current pilot study at tempted to determine if individuals with ADSD differed from controls in various breathing parameters while engaged in conversational and reading tasks. Thirty individuals were te sted; fifteen in the ADSD group and fifteen in the ageand gender-matched control group. Respitrace, an inductive plethysmography device, calculated 14 different respiratory meas ures related to volume, timing, thoracic displacement and respiratory efficiency. The results of the study i ndicated that various significant differences existed between groups Those with ADSD were found to have

PAGE 10

vii statistically higher ventilation ra tes, a greater frequency of breaths per minute, a higher degree of muscular ineffici ency/breathlessness and labor ed breathing. These results indicated that individuals with ADSD suffe red from disordered breathing due to the neurologically related obstruction at the level of the larynx. Differences according to task were al so found. Specifically, the rib cage contributed to a lesser extent in voice production and the pa rticipants utilized longer inspiratory times, exhaled a larger volume of air and took longer to reach peak expiratory flow during conversational tasks when compar ed to reading tasks. These differences were attributed to a higher cognitive-li nguistic demand required during conversational speech. Overall, the results of this st udy have many clinical implications. Most importantly, these findings support the idea th at individuals with ADSD may experience difficulties with respiration as the effects of their Botox injection begin to wear off. Further research is needed w ith regards to the effects la ryngeal spasms have on other respiratory behaviors.

PAGE 11

1 Introduction The process of respiration is multifaceted and intricate. It requires the coordination of various struct ures and muscles to function appropriately. The rib cage and the abdomen are the two basic structures that assist in resp iration (Hixon, Goldman, & Mead, 1973). These two areas of the body work together (and typically move as a unit) to facilitate lung ex pansion and contraction, wh ich results in a physical displacement of the thoracic cavity. Total ri b cage and abdominal displacement are equal to the total lung volume change; i.e., the amount of air that is inhaled or exhaled (Hixon et al., 1973). In other words, the volume of air that is inhaled and exhaled on a single breath is directly related to the physi cal movement of these two structures. The rib cage is a highly deve loped region. It encases the lungs, which are held to the interior of the rib cage by pleural forces. The rib cage expands and contracts through the influence of muscle contra ction and external/internal pre ssures (Hixon et al., 1973). Inspiratory muscles, such as the external intercostals, contract and lift the rib cage superiorly, anteriorly and horizontally. Becau se the lungs are attached to the rib cage, they also expand. This expansion creates a negative internal lung pressure when compared to the external environmental positiv e pressure. This resu lts in a rush of air into the lungs (Hixon et al., 1973). This process can be li kened to the effects of a vacuum, because air will flow from an area of positive pressure into an area of negative pressure to stabilize the environment. E xhalation works in a similar manner. Through

PAGE 12

2 elastic recoil of the rib cage and some muscle contraction, a positive pressure is exerted on the lungs. This pressure pushes the air out of the lungs and through the respiratory tract. The abdomen, the second mechanism assisting in the breathing process, is comprised of the diaphragm and the muscles of the abdominal wall. The diaphragm is a muscle located at the inferior boundary of the lungs. During inhalation, the diaphragm moves downward, which in turn, pulls the lungs down and increases their size (Hixon et al., 1973). The abdominal wall is, therefore, pushed forward to allow for the downward movement of the diaphragm. This process al so creates a negative pressure within the rib cage. As described above, this negative pre ssure facilitates air fl ow into the lungs. Once the lungs fill with air, the positive pressure within the lungs triggers the process of expiration. In this process, the diaphragm relaxes, pushing the inferior edge of the lungs upward and the air outward. The abdominal muscle s may also contract, pulling inward and exerting further force to move the diaphragm upward and air outward. The air that exits the lungs will proceed through the bronchi into the trachea and pass through the larynx. It is at the level of the larynx that vo ice production occurs. As the air reaches the level of the larynx, the vocal folds are drawn toward the midline by means of the adductor muscles. Air pressure builds up subgl ottally and blows the vocal folds apart. The folds are drawn back together through elas tic recoil and the Bernoulli effect (Titze, 1994). This cycle of opening and closing crea tes a “buzz” or the complex acoustic signal known as the voice. This buzz is then car ried through the vocal tract (a resonance system), going through various strictures and cl osures to eventually pass through the lips and/or nose. The strictures and closures al ong the length of the vocal tract allow for the

PAGE 13

3 production of consonants and vowels (or phoneme s). The phonemes, when produced in sequence, create spoken words and sentences. The production of speech, therefore, reli es on the coordination and balance of three subsystems: respiration, phonation and resonance. These systems must work together as their functions often overlap. For example, one responsibility of the respiratory system is to maintain proper subglottal pressure (Haynes & Netsell, 2001), which must be regulated, as it is the foundation to voice production. The maintenance of subglottal pressure occurs through the coordi nation of the muscles of the respiratory system and those within the larynx (Haynes & Netsell, 2001; Redstone, 2004; Sapienza, Stathopoulos & Brown, 1997). Un fortunately, the coordinati on of these three systems does not always function properly. One system may be forced to act in a different fashion as a result of the other system’s failure. The respiratory system is constantly exchanging depleted air for newer, oxygenrich air. The amount of air that is inhaled and exhaled dur ing a single breath has been termed tidal volume. Tidal volume will chan ge depending on the activities the individual is performing. For example, while exercising, individuals will have larger tidal volumes because more oxygen is required (Hixon et al., 1973). Similarly, tidal volume is increased for the purposes of producing speech (Hixon & Hoit, 2005). In fact, evidence has shown that people will typically inhale to about twice the resting tidal volume and expire to the resting tidal volume’s end e xpiratory volume (Hixon & Hoit, 2005). This means, for the purposes of speech, twice the amount of air is needed than that during resting respiration.

PAGE 14

4 Researchers have labeled other respirator y volumes and capacities in addition to tidal volume. For example, the volume of air that an individual can inhale above the tidal volume is called inspiratory reserve volume, wh ile the air that an individual can expel after the end expiratory volume of tidal vol ume is the expiratory reserve volume (Hixon & Hoit, 2005). When two or more lung volumes are combined, a lung capacity is formed. Vital capacity refers to the total amount of air a person can inhale if they exhale all of the expiratory reserve volume and then fu lly inhale. Therefore, vital capacity is the sum of the expiratory reserve volume, tidal volume and the inspiratory reserve volume. For a more thorough description and explana tion of lung volumes and capacities, please refer to Hixon and Hoit (2005). Instrumentation for Measuring Breathing Various instruments have been used to obtain data regarding volumes, respiratory patterns, etc. Wet spirometers were among the first instruments used to measure air volumes (Hixon & Hoit, 2005). These instrume nts contained a cylindrical chamber that housed water and a floating bell. Individuals exhaled into tubing that allowed for the expired air to become trapped under the bell resulting in a rising of the bell. The measurement of the air volume expelled wa s calculated from the distance the bell traveled. Although innovative at the time, th is instrumentation ha s obvious limitations. For example, it was not easily transportable to other test sites and it did not allow for measurements of expelled air during con tinuous speaking tasks (Hixon & Hoit, 2005). Researchers later realized that there was a direct relationship between the physical displacement of the thorax a nd the resulting volume of ai r within the lungs (Hixon & Hoit, 2005). Therefore, a new method to de termine lung volume change was developed

PAGE 15

5 by measuring the changes in body surface. These instruments were developed recognizing that the volume of air is equal to the amount of chest wall and abdomen displacement (Hixon & Hoit, 2005). The magne tometer is one such instrument that establishes lung volumes through the calculation of the surface displ acement. With this instrumentation, electromagnetic coiled wires are attached to the surface of the body at the sternum and above the umbilicus and at adjacent areas on the back. Magnetometers calculate the changing distance between the adjacent anterior a nd dorsal wires (i.e. between the sternum and the upper back and be tween the umbilicus and the lower back). The increasing and decreasing distances that occur during inhalation and exhalation are then converted to lung volumes (Hixon & Hoit, 2005). Respiratory inductive plethysmography opera tes in a similar way. It determines lung volumes by calculating the surface displace ment using elastic bands with embedded coiled wires (Hixon & Hoit, 2005). The bands are wrapped around a person’s chest and abdomen and the wires calculate the expansi on and contraction of the body’s surface. As the body surface is displaced, the elastic trans duction bands send an equivalent voltage to a calibration device, which then translates th is information into volume measures (Hixon & Hoit, 2005). Because these two instruments (magnetometers and inductance plethysmography) calculate expansion and contraction of the thorax, they allow measurements to be made dur ing inspiration and expiration. These surface measuring devices are not without strengths an d weaknesses. Due to the fact that the instruments are noninvasive, they do not restrict the respiratory or articulatory systems and allow the measurem ent of total lung volume change (Hixon & Hoit, 2005). Further, these instruments produce immediate measurements, which allow

PAGE 16

6 the study of rapid breathing processes, such as speech breathing. In comparison, instruments such as the spirometer, require th e individual to exhale into a tube. This process is cumbersome due to the size of th e instrument and the t ubing is obstructive to normal respiratory behaviors. Further, it can only measure the volume of expired air. Spirometry does not allow measurements of inhalation because the device works by measuring the rising movements of the bell as expired air forces the bell up. On the other hand, surface-measuring devices may errone ously measure body m ovements that are unrelated to breathing, such as, posture change s. Also, band slippage is a possibility when using inductance plethysmography (Hixon & Hoit, 2005). Band slippage occurs when the band is too large for the person’s circum ference, or the Velcro fastener is loose. In these cases, distorted data ma y be obtained during measurements. Respitrace (produced by Non-Invasive Monitoring Systems; Nims, 2002) is one version of inductance plethysmography. It is a commonly used instrument to determine various aspects of the breathing process. Acco rding to the manual, Respitrace is accurate to within 10% of respiratory changes when compared to other instruments that have proven to be valid measures of respirati on (Nims, 2002). Like any instrument, the Respitrace is not without setbacks. A shif ting baseline is an example of problems associated with this instrument (Neuma nn, Zinserling, Haase, Sydow, & Burchardi, 1998). When used over a period of time, the calibrated baseline of end expiratory level has been known to drift upward, possibly skewing data. A few researchers have examined this shift. For example, Neumann et al. (1998) examined the shift in individuals who were ventil ator-dependent due to lung injury, Chronic Obstructive Pulmonary Disease (COPD) or sedation. The researchers found the drift did not increase

PAGE 17

7 or decrease at a steady rate within or between subjects. Specificall y, during the first five minutes after calibration, the baselines drifted between 28.2 and 48.9 mL/min. The researchers reported that this device is therefore, not accurate enough to make quantitative measures in lung volumes (Neu mann et al., 1998). However, this finding was not supported by the data obtained in a study by Lei no, Nunes, Valta, and Takala (2001), who specifically compared the newest model of Respitrace to an older version. Better measurement accuracy and maintenance of the calibration was found. These researchers made two pertin ent recommendations: the instru mentation should be turned on for several hours before using on patient s in order to redu ce the baseline shift (maximum baseline shift during this study wa s under 8 mL/min) and repeated measures should be completed to assure accuracy. Over all, the resear chers reported Respitrace to be “accurate enough for clinical and research purposes” (Leino et al., 2001, p. 111) even when their recommendations were not followed. Since Respitrace has been accepted as a va lid measure of the various breathing parameters (Leino et al., 2001; Nims 2002), it has been used in a wide variety of research studies. Although commonly used for sleep apnea studies, inductance plethysmography has been used in studies related to speech production as well. Schaeffer, Cavallo, Wall and Diakow (2002) used induc tive plethysmography to examine the physical respiratory movements in individuals with dysphonia. Participants were asked to read two paragraphs containing either 10or 60-syllables per sentence. With the use of this device, the researchers were able to determ ine that those with dysphonia terminate speech below the resting expiratory level and us ed paradoxical movements in the abdomen during exhalation. They hypothesized that th e dysphonic group utili zed the paradoxical

PAGE 18

8 movements because they coul d not efficiently exhale due to difficulty using their abdominal muscles. The researchers assume d that the paradoxical movements would not have been so prevalent in this group if th ey had used more appropriate lung volumes during speaking tasks. Hence, inductive plet hysmography allows for analysis of specific aspects of the breathing process, especially in the rapidly changing patterns associated with speech breathing. Respitrace has also been used to determin e the level of synchronicity between the abdomen and rib cage during respiration. Br aun, Abd, Baer, Blitzer, Stewart and Brin (1995) evaluated respiratory behaviors in indi viduals with dystonia. Many measures of respiratory performance were used in this study including flow volume loops, inspiratory and expiratory muscle pressure, and measures of arterial blood gas values in relation to respiratory tasks (speaking and quiet breathing) and positioning (supine and sitting). Specific to the purposes of the present paper, chest wall and abdominal movements were also examined in order to determine if aberrant movements could be found during pulmonary testing while exercising. The re searchers found some degree of asynchrony; however, the majority of the participants pe rformed at near normal levels during the exercise task. On the other hand, a greater number of dystonic beha viors interrupted the breathing process while performing the other ta sks (quiet breathing, sp eech breathing) or while in other positions (supine). Therefor e, the authors concluded that resting, speech and exercise respiration result from different neural pathways and the dystonia can affect one pathway more than the others. Interestin gly, these authors also briefly reported on previous research that found this same populat ion to have irregular measures of rapid shallow breathing (Braun et al., 1995).

PAGE 19

9 Inductance plethysmography has also been used to evaluate the differences in breathing patterns in healthy children while in the sitting, standing and supine positions (Mayer, Clayton, Jawad, McDonough & A llen, 2003). The thoracic/abdominal coordination of children was examined in each of the three positions. The researchers found significant differences in the coordina tion between the positions. For example, nearly synchronous movements between the subsystems were found in the sitting position, while asynchronous movements were us ed in the supine position. In addition, the researchers examined the feasibility and, what the aut hors termed, success of using this instrument with this young population. F easibility was defined as a willingness to participate in the study, while success was defi ned as the ability for the participants to perform the study’s procedure. Out of the 50 pa rticipants, 49 were w illing to pa rticipate (feasibility) and 42 had success in performi ng the desired tasks. Therefore, the researchers felt the instrument was both f easible and successful for this population (Mayer et al., 2003). These results can be considered universal to the general public because, if the instrument is non-threatening and easy to use for children, adults should have no problem with it. In another study, Iwarsson (2001) used i nductive plethysmography to examine the positioning of the larynx during different abdominal postures while breathing. She postulated that abdominal wall expansi on would be accompanied by tracheal pull resulting in a lowered laryng eal position. Subjects were asked to alter their abdominal expansion by either keeping it pushed out or pulled in during speech tasks. Plethysmography was used as a visual feedb ack method for the subjects, as they were asked to begin speaking when they reached about 70% vital capacity. Results of the

PAGE 20

10 study indicated that clear differences in the vertical laryngeal pos itioning occurred during the two abdominal speaking positions. A hi gher laryngeal position resulted when the abdomen was pushed out. These results indica ted that the author’s original hypothesis was not supported. These study descriptions have provide d examples of the research use of inductance plethysmography. It has been used to measure various breathing parameters and evidence validating its use in research settings has been provided. Therefore, researchers can use this instrument to determ ine the patterns of br eathing in individuals with a disordered larynx, th e results of which could posse ss significant clinical value Breathing with a Disordered Larynx Variations in speech breathing have been reported in individuals with disordered larynges (Bunton, 2005; Plant & Hillel, 1998 ; Makiyama, Kida & Sawashima, 1998; Saarinen, Rihkanen, Malmberg, Pekkanen & Sovijari, 2001; Sapienza et al., 1997; Schaeffer et al., 2002; Vertigan, Gibson, The odoros, Winkworth, Borgas & Reid, 2006). The phonatory and respiratory systems must work in concert to yield the precise framework from which to produce normal sp eech. If one system is not functioning appropriately, the other system may begin to function differe ntly. Some authors believe an individual alters his/her respiratory behavior in order to compensate for the disordered larynx (Sapienza et al., 1997; Vertigan et al ., 2006). These investig ators have described that individuals with voice disorders tend to produce deep inhalations and initiate speech at different lung volumes compared to controls These behaviors were attributed to an attempt to “overcome respiratory difficulty, co mpensate for air loss at the glottal level and regulate subglottal pressure during phona tion” (Vertigan et al., 2006, p. 648). In

PAGE 21

11 addition, increased or decreased air flow rate s, differing end inspiratory or expiratory lung volumes, changes in subglottal pressu re and paradoxical movements have been reported in persons with disordered larynge s (Bunton, 2005; Makiyama et al., 1998; Plant & Hillel, 1998; Saarinen et al., 2001; Sapienza et al. 1997; Schaeffer et al., 2002; Vertigan et al., 2006). Airflow The most apparent change observed in the speech production of those with disordered larynges is from the effects of a ltered airway resistance at the level of the glottis. Depending on the type of laryngeal disorder, glottal resistance can be increased or decreased. For example, vocal nodules will prevent the vocal folds from fully closing. There will be small spaces around the nodul e area and air will escape through these spaces during phonation (Sapienza et al., 1997). Therefore, a subsequent decrease in laryngeal resistance with an increase in airflow would be expected. Increased airflow rates were found when researchers examined the respiratory behavi ors in women with vocal nodules. Further, the production of each spoken syllable resulted in a larger expulsion of air when compared to controls. The conclusions of this study highlight the tendency for airflow to be altered when pat hology is located within the larynx. Makiyama et al. (1998) examined expira tory lung pressure and airflow rates during a sustained vowel task in individuals with Reinke ’s edema and patients with recurrent nerve paralysis. Individuals were required to sustain a vowel at a comfortable level and then increase and d ecrease the intensity without a ltering the pitch. Results of the study indicated that all part icipants increased subglottal pr essure to increase intensity, resulting in elevated airflow rates. When compared across groups, however, differences

PAGE 22

12 in the degree of increased pressure and ai rflow were found. For example, the group with paralysis was found to have substantially in creased airflow. Th e authors hypothesized that this group increased airflow to incr ease vocal intensity because they could not increase laryngeal tension. Saarinen et al. (2001) studied airflow in individuals with vocal fold paralysis. Flow-volume spirometry and body-plethysmogr aphy were used to examine respiratory patterns in quiet and forced breathing. Sp ecifically, forced inspiratory and expiratory flow was lower than that of controls. Th ese findings would suggest that because the paralyzed vocal fold created an obstruction at the level of the larynx, airflow is hindered. In this case, obstruction at the level of the la rynx can inhibit force respiration. Lung Volumes In addition to altered airflow, there is evidence showing there are differences in lung volumes during speech when comparing individuals with la ryngeal pathology to controls. Sapienza et al. (1997) hypothesi zed that individuals compensate for the increased airflow related to the laryngeal pathology by initiating speech production at high lung volumes. These researchers f ound the women with vocal nodules expended larger volumes of air per syllable and utte rance during connected speech when compared to controls. Because airflow was increased with each syllab le and a larger volume of air was expended during the speaking tasks, the res earchers postulated that larger volumes of air are needed to maintain the nece ssary subglottal pressure for phonation. Bunton (2005) examined lung volume use in individuals with Parkinson’s disease (PD). Lung volumes at the initiation and te rmination of a conversational speech task were measured. Findings from this study indicated many differences in breathing

PAGE 23

13 parameters between the PD and control gr oups. The participants in the PD group produced fewer syllables per breath group and evidenced a shorter du ration of expiratory time during speech. In addition, this group ini tiated and terminated speech at lower lung volumes. This was attributed to an increased effort during speaking. These results are in contrast to Sapienza et al. (1997), who hypothe sized increased initia ting lung volumes to maintain subglottal pressure. The differen ces may lay in the different populations examined. While Bunton (2005) examined indi viduals with PD (a disorder interfering with neural innervations), Sapienza et al. (1997) examined women with vocal nodules (a laryngeal lesion). Hence, the nature of th e laryngeal pathology may affect breathing patterns. In another study of individuals with nonneurological relate d laryngeal pathology, Schaeffer et al. (2002) examined the venit ilatory behaviors of ten women with abuserelated dysphonia. Specifically, participants were asked to re ad two sets of paragraphs: a ten syllable-per-sentence paragraph and a 60 syllable-per-sentence paragraph. Those with dysphonia had lower end-expiratory l ung volumes while speaking than controls. This finding, exacerbated in the longer speaki ng task, was attributed possibly to the use of the grammatical periods as linguistic ma rkers within the text indicating acceptable places to replenish the air supply. Theref ore, the dysphonic group increased the demands on the respiratory system by producing speech we ll below the resting e nd expiratory level in order to maintain continuity within th e reading text. The control group, however, utilized more efficient breathing patterns by term inating speech at or above this level. Data has confirmed that individuals will change lung volumes at the initiation of speech (Bunton, 2005). In addition, reliable da ta confirms lung volumes at the end of

PAGE 24

14 speech are also altered (Bunton, 2005; Schaeffer et al., 2002). Interestingly, all of the above mentioned studies reported that their subjects terminated speech at lower lung volumes compared to controls (Bunton, 2005; S aarinen et al., 2001; Sapienza et al., 1997; Schaeffer et al., 2002). Speaking at lower l ung volumes may create an unnecessary need to increase muscle effort in order to maintain proper subglottal pressure. This increased muscular effort may, in turn, create tension within the respiratory mechanism (Saarinen et al., 2001). The tension may further exacerbate changes in respiratory behaviors. Subglottal Pressure As mentioned previously, prope r subglottal pressure must be maintained in order to phonate. Not surprisingly, researchers have found variations in subglottal pressure in populations who exhibit voice di sorders (Schaeffer et al., 2002). Jiang, O’Mara, Chen, Stern, Vlagos & Hanson (1999) examined the aerodynamics of 24 individuals with PD. They found significantly greater subglottal pressures in this group when compared to the control group. They attributed this findi ng to increased laryngeal resistance during phonation. It was concluded that the participants with PD used an increased subglottic pressure (possibly with incr eased expiratory effort) to compensate for the laryngeal resistance because of the presence of a neurological laryngeal pathology. Makiyama et al. (1998) examined the expi ratory lung pressures before and after increasing vocal intensity in three groups of participants: individuals with Reinke’s edema, individuals with recurrent laryngeal nerve paralysis and a control group. Results revealed increased subglottal pressure and ai rflow in all groups when increasing vocal intensity. Those with Reinke’s edema exhi bited extremely elevated expiratory lung pressures that it differentiate d this group from the other two. This finding was attributed

PAGE 25

15 to increasing expiratory effort in order to overcome the increased laryngeal resistance because of increased vocal fold mass. Those with paralysis exhibited elevated expiratory lung pressures in both the comf ortable and increased vocal in tensity task, however to a lesser degree than the controls or those with edema. In paralysis, there would be less physical resistance because the laryngeal mu scles are paralyzed, a nd therefore, cannot resist airflow or maintain increased pressure Not surprisingly, this group experienced extremely elevated airflow rates that distinguished them as a group. Physical Respiratory Movements The research described above has demons trated that larynge al disorders may affect respiration in a variety of ways, such as altered airflow, lung volumes and subglottal pressure. In addition, evidence has been provided in the literature suggesting that laryngeal disorders can a ffect the physical framework of respiration in regards to rib cage and abdominal movements. For example, Vertigan et al. ( 2006) briefly commented on previous research that i ndividuals with voice disorders exhibit paradoxical chest wall movements while speaking in their disc ussion of the pulmonary functioning of individuals with a voice disord er. Research conducted by Schaeffer et al. (2002) support this claim. In this latter study, the resear chers used inductive plethysmography in order to evaluate the movements of the respiratory system. They found that the subjects with dysphonia exhibited a greater frequency of abnormal expiratory abdominal movements during speech, suggesting that th e physiological process of resp iration is altered in these participants. In her study of lung volume use in indivi duals with Parkinson’s disease, Bunton (2005) observed greater effort dur ing the speaking tasks. This effort was attributed to the

PAGE 26

16 increased abdominal activity needed to overc ome a rigid rib cage that may occur in Parkinson’s disease. This increased abdo minal contribution may have been used to maintain the subglottal pressure needed wh ile speaking at low lung volumes. In other words, because the rib cage was less flexib le, the abdomen took on a larger role in the respiratory process, thereby allowing the indi viduals to maintain and control the expiring air for speech production purposes. The information provided above has demons trated that laryngeal disorders have various influences on measurements of respir ation related to airflo w, volumes, pressures and the physical movements. For the purposes of the present study, one specific disorder of the larynx and its influen ce on respiration is examined. Spasmodic Dysphonia The spasmodic dysphonias (SD) are one set of phonatory disorders that are characterized by voice breaks due to involuntary contractions of the adductor (closing) or abductor (opening) muscles of the vocal fold s. According to the National Spasmodic Dysphonia Association (NSDA, 2006), ther e are approximately 50,000 people diagnosed with SD in North America. This number is thought to be an underestimate because the disorder can often be misdiagnosed as Mu scle Tension Dysphonia (MTD). MTD is a form of dysphonia due to excess muscle tens ion within the larynx. MTD and SD share similar perceptual characteristics, such as hoarseness and limited pitch and loudness range, however SD has more of a strained/s trangled vocal quality (Sapienza, Walton, & Murry, 2000). The spasmodic dysphonias can be cla ssified into three groups: adductor spasmodic dysphonia (ADSD), abductor spasmodic dysphonia (ABSD), and mixed

PAGE 27

17 spasmodic dysphonia. For the purposes of th e present study, this paper will focus on the adductor type. As the name implies, ADSD involves spasms of the adductor laryngeal muscles. These muscles are responsible for closing the vocal folds. Adduction of the folds occurs at each muscular contraction, re sulting in excessive medial compression of the folds (Cannito & Woodson, 2000). These sp asms severely disrupt the normal closing and opening of the vocal folds. The result is a strained/strangled vocal quality that sounds characteristically similar to gl ottal fry (Cannito & Woodson, 2000). The perceptual characteristics of ADSD are often very similar to other voice disorders. However, differentiating the di sorders is important in terms of underlying origin, disease progress and treatment op tions (Lundy, Roy, Xue, Casinao, & Jassir, 2004). Sapienza et al. (2000) performed an acoustical analysis on individuals with ADSD and MTD with hopes of developing a fr amework to differentially diagnose ADSD from MTD. The researchers specifically examined aperiodicit y, phonatory breaks, and frequency shifts. They found that thos e with ADSD produced significantly more phonatory breaks, aperiodic segments and fre quency shifts during su stained vowel tasks (Sapienza et al., 2000). Most notable was the presence of phonatory br eaks in those with ADSD, while there were no instances of breaks in voicing in the MTD group. The cessation of phonation occurs when vocal fo lds are tightly compressed. Continuous phonation is not possible because the invol untary muscular contractions prevent phonation due to the tight medialization. The researchers ther efore claimed, that determining the presence of phonatory break s could provide useful information when diagnosing ADSD. Overall, acoustical analysis, although us eful, cannot alone provide a basis for differential diagnosis.

PAGE 28

18 Researchers have attempted to find othe r non-invasive means for differentially diagnosing ADSD. For example, Lundy et al. (2004) examined speech production parameters by means of a motor speech prof ile analysis in individuals with ADSD, amyotrophic lateral sclerosis (ALS), and tremor These three disorders were chosen for study because they shared similar perceptual features (strained, strangled and tremulous vocal quality) but had different etiological origins. The researchers concluded that although the voice symptoms of these disord ers perceptually sounded similar (i.e. strangled), differences did exist. First, th ey found the individuals with ADSD to have a speaking fundamental frequency closer to that of normative data when compared to the other two groups. Also, those with ADSD pe rformed diadochokinetic tasks faster than those with ALS or Tremor; nevertheless, al l three groups had a sl ow connected speech rate (Lundy et al., 2004). On the other ha nd, pitch variability ta sks did not show a statistically different range between the groups and should not be used in differentially diagnosing these disorders. Respiration in Adductor Spasmodic Dysphonia Researchers have described the necessity fo r the respiratory system to compensate for laryngeal tension caused by the spasms in SD (Cannito & Woodson, 2000; Woodson, Zwirner, Murry & Swenson, 1992). Woods on et al. (1992) reported a decreased phonatory airflow rate with increased subglotta l pressure in this population. Presumably, the increased subglottal pressure is needed to overpower the medial compression while flow is impeded due to the spasms. In another study, Plant & Hillel (1998) exam ined subglottal pressure and intraoral air pressure during continuous syllable produc tion in seven indivi duals with ADSD.

PAGE 29

19 Pressure detecting devices were inserted in the trachea usi ng needle electrodes and into the oral cavity via a facemask. This allowe d for the comparison of subglottic and oral pressures. They found a decrease in airflow with an increase in s ubglottal air pressure that, at times, surpassed the pressure measured intraorally. Therefore, these participants, with an increased expiratory lung pressure, produced decreased airf low. These findings seemed to indicate that laryngeal spasms affected both airflow and air pressure. Specifically, the individuals with ADSD were compensating for the spasms by increasing subglottal pressure, while the spasms were simultaneously hindering airflow. Airflow was reduced because the spasms are constantly creating an obstruction within the larynx. Another aerodynamic study of ADSD was co mpleted by Higgins, Chait & Schulte (1999). These researchers examined phonatory airflow in this population. Specifically, the authors measured mean phonato ry airflow, the variation of airflow (breath to breath changes in airflow) and airflo w perturbations (an increase a nd then decrease of airflow of 75ml/s) via an intraoral air pressure sensing device and a pneumotactograph. Those with ADSD exhibited greater airflow variability with overall lower phonatory airflow. Further, this group consistently demonstrated more airflow perturbations when compared to a control group and a group diagnosed with MTD. The authors contributed the large airflow variability to both glottal deficien cy and to changes in respiratory driving pressures. In other words, when individua ls with ADSD experienced the spasms, they increased respiratory pressure to overcome the spasm in order to continue to phonate. The previous studies have suggested th at individuals with ADSD are likely to experienced increased subglottal pr essure with decreased airflow rates. It is important to know if these aerodynamic parameters are return to normal in patients with ADSD after

PAGE 30

20 Botox injections. Therefore, various studies have focused on the phonatory aerodynamics of individuals post-botox injectio n. Adams, Durkin, Irish, Wong and Hunt (1996) completed a study in which the aer odynamics of individuals with ADSD were compared to a control group. Individuals in the ADSD group were measured three times: prior to injection, two to four weeks post-inje ction, and eight to ten weeks post-injection; while data was obtained from the control gr oup two times with two weeks separating the measurements. The speaking task consisted of repeated /pa/ syllables and the participants used a face mask with an in traoral pressure sensing devi ce to measure airflow and pressure. The results of the study indicated si gnificant decreases in laryngeal resistance and variability of airflow with an increase in average airflow post-in jection. It seems, therefore, the injections inhibited the spas ms, which decreased the glottal resistance. With lowered resistance, airflow naturally in creased and airflow variability decreased. Therefore, the participants with ADSD experienced near normal values for these measures after treatment in the form of Botox injection. Other studies of the aerodynamics of resp iration post-Botox in jection have also found increased airflow rates. For example, an increase in airflow during sustained vowels was found in a study by Cantarella, Be rusconi, Maraschi, Ghio, and Barbieri, (2006). The post-injection airflow values were significantly different from pre-injection values, but were not significantly different when compared to controls. In addition, airflow variability was measured and post-in jection airflow values were significantly more stable compared to pr e-injection levels. Intere stingly, although these values indicated a more stable airflow after Botox than pre-injection, patients with ADSD still evidenced a higher level of variabi lity in airflow than controls.

PAGE 31

21 Similarly, increased airflo w rates post-injection were found in a study completed by Woo, Colton, Casper and Brew er (1992). This study specif ically examined airflow preand post-Botox injection a nd in unilateral nerve block (a nother form of treatment for ADSD). The researchers found that those who were injected with Botox improved their airflow rates to values within the normal range, while those who underwent nerve block had post-treatment airflow rates well above the normal range (Woo et al., 1992). The authors concluded that Botox injections allo wed for the participants to phonate with a more normal airflow rate than those w ho chose nerve block (Woo et al., 1992). Statement of the Problem Evidence has been presented that suggests laryngeal movements not only relate to respiratory events, but also that respirator y behaviors may be altered in those with disordered larynges. This interdependent relationship demonstrates how the two (respiratory and phonatory) system s rely on each other and e xplains how deviations of normal respiratory functioning may occur when the phonatory system is defective. The studies related to ADSD that are presented a bove basically inspected respiration related to measurements of airflow and neglected so me of the other respiratory parameters. What is lacking in these studies is info rmation regarding the respiratory behaviors involving lung volumes, respiratory times and the physical movements related to respiration. Given the evidence base, it seems intuitive that these respiratory parameters would also be altered. Research needs to be conducted on individuals with ADSD that investigates these specific measures of resp iration. Therefore, th e present study focuses on respiration and ADSD. Specifically, are re spiratory patterns al tered in individuals with ADSD?

PAGE 32

22 As discussed above, the Respitrace instru mentation has been proven as a valid and sensitive instrument to measure vari ous breathing parameters. The Respitrace generates several different measures of breathi ng. For the purposes of this investigation, measures of volume, timing, thoracic disp lacement and respiratory efficiency were considered. The following questions were asked: 1. Do individuals with ADSD differ from ageand gender-matched controls on measures of respiratory volume, timing, thoracic displacement and efficiency? 2. Do differences in measures of respiratory volume, timing, thoracic displacement and efficiency vary by speaking task or gender?

PAGE 33

23 Methods In order to produce voice, a balance must be maintained between the respiratory and phonatory systems. When one of these systems becomes dysfunctional, the other system will no longer perform normally. Many researchers believe the changes at the respiratory level are the direct result of a n eed to compensate for the disruption at the laryngeal level (Sapienza et al., 1997; Vertigan et al., 2006 ). Individuals with ADSD may experience reduced phonatory airflow ra tes and increased subglottal pressure (Cannito & Woodson, 2000). Previous rese arch, however, has only focused on these parameters. Yet to be studied are other pa rameters of breathing that may broaden our understanding of the level of compensation by the respiratory system. Therefore, this study was designed to examine parameters i nvolving respiratory volume, timing, thoracic displacement and efficiency in the sp eech of individuals with ADSD. Participants Twenty-five adults were recruited from the Ear, Nose and Throat (ENT) Clinic at a medical university in southwes t Florida. These individuals were at the clinic for their regularly scheduled Botox injection. They were asked to participate in this project prior to receiving their injections. Individuals were consider ed eligible for the addu ctor spasmodic dysphonia (ADSD) group if they had a medical diagnosis of ADSD. Individuals with a co-existing neurological or neuromotor di sorder or an inadequate read ing ability were excluded from

PAGE 34

24 this study. Participants self-re ported a co-existing neurological disorder Reading ability was judged during the oral r eading tasks and one person wa s excluded from the study due to poor oral reading skills. This participan t’s reading ability was judged to be nonfluent because that participant demonstrated many false starts and long pauses while reading aloud. Of the 25 participants, fifteen were u ltimately included in the ADSD group; nine women and six men. The mean age for the females was 60.77 years ( s.d. = 12.96 years) and the mean age for the males was 50.33 years ( s.d. = 11.39 years). The data collected from the remaining ten participants were not included because of computer corruption of data files, researcher error when conducting the study, and poor quality speech samples. Sixteen ageand gender-matched individuals agreed to participate as the matched control group. These participants were me mbers of the community, faculty from the Communication Sciences and Diso rders department or acquainta nces of the researchers. Individuals were asked to part icipate in the study if they reported being a non-smoker and were free of a neurological/neuro motor disorder or respiratory disease. Data from fifteen of the individuals were in cluded: nine women and six men. The mean age for the females was 56.33 years ( s.d. = 7.62 years) and the mean age for the males was 53.5 years ( s.d = 11.39 years). Data collected from the remaining individual was excluded because of poor quality of the speech sa mple and Respitrace band slippage, which resulted in questionable Respitrace waveforms. Materials Respitrace and RespiEvents (Nims, 2002) were used to quantitatively measure the breathing process. Specifical ly, the Respitrace bands act as an inductive plethysmograph.

PAGE 35

25 Plethysmography is a method of determin ing air pressure, volume and flow by calculating changes in air volumes while breat hing. The Respitrace, when compared to spirometry or other pneumotachography, has be en shown to be accurate within 10% of the estimated changes while breathing (Nims, 2002). Respitrace requires the use of a vest that is wrapped around an individual’s rib cage and abdomen. This vest consists of tw o elastic bands with embedded coiled wires. These wires calculate and quan tify any thoracic movement by forming a stretchable loop around the body so that the wires change size with the rib cage and abdominal excursions. These changes are quantified and a sinusoidal waveform that represents inhalation and exhalation is produced within Re spiEvents. Figure 1 is an example of a waveform obtained during the study. The t op display represents the tidal volume, the middle display represents the rib cage trace a nd the bottom display is the abdomen trace. These tracings were extracted from a control participant during a speaking task.

PAGE 36

26 Figure 1. An example of a waveform produced by a control participan t as viewed via the RespiEvents software. An initial calibration period of five minutes was completed for each participant. This calibration period allowed for Respitrace to track typical breat hing patterns for each individual and to calibrate th e equipment for experimental use. After this calibration period, the data obtained from Respitrace were converted to arbitrary volume units (Aml) by RespiEvents. Participants were asked to complete four tasks in a random order: two readings and two conversational tasks. The reading tasks involved two readings of the first

PAGE 37

27 paragraph of the Rainbow Passage (Fairbanks, 1960), while the conversational tasks consisted of two separate tasks. The firs t task involved a description of the “Cookie Theft” picture from the Boston Diagnostic Aphasia Examination 3rd Edition (Goodglass, Kaplan, & Barresi, 2000). The second convers ational task varied by group. The ADSD group answered the question, “Tell me about your experiences being diagnosed with spasmodic dysphonia”. Individuals in the cont rol group were instructed to speak about a topic of interest for approximately one minut e. These tasks were designed to elicit conversational speaking patterns. Procedures All participants were briefed on the purpose and procedures of the study and asked if they were willing to participate. If they agreed, an informed consent was signed and the study began. Each individual was carefully fitted with the Respitrace bands. One band was wrapped around the participant’ s rib cage and the other around his/her abdomen. The bands were securely attached to prevent band slippage. To ensure a good fit, Velcro was used to attach the two ends of the bands together and any excess band was tightly pinched together, folded over, and secured with duct tape Wires were then connected to the bands and held in place vi a a snap. These wires then lead to the Resptrace device, which was also connected to a computer. RespiEvents converted the traces into waveforms to be further proce ssed into measures of various aspects of breathing. Table 1 illustrates the presentation orde r of speaking tasks utilized during this experiment. For example, participant 1 was asked to read the paragraph, describe the “Cookie Theft” picture, reread the paragr aph and finally engage in 1.5 minutes of

PAGE 38

28 dialogue. The next participant was asked to complete the same tasks, however in a different order: respond to the question, read the paragraph, descri be the picture and finally reread the paragraph. Table 1. A table representing the randomi zation of speaking tasks presented to participants. Participant 1 Participant 2 Participant 3 Participant 4 Order 1 Paragraph Rdg Response to ? Paragraph Rdg Picture Descrip Order 2 Picture Descrip Paragraph Rdg Response to ? Paragraph Rdg Order 3 Paragraph Rdg Picture Desc rip Paragraph Rdg Response to ? Order 4 Response to ? Paragraph R dg Picture Descrip Paragraph Rdg At the start and finish of each task, the researcher pulled either the rib cage or abdominal band, which produced a notable spik e within the computer waveform. This motion allowed for the easy identification of the beginning and end of each speaking task within the waveform. To verify that the correct portion of data was being analyzed, the duration between spikes was compared to the speaking duration obtained from audio recordings. The participants were audio recorded using an Optimus voice activated full autostop cassette recorder (model number CTR-117 14 -1123). Recordings were played back and the duration of the speech sample during each task, in order to match recording time with the waveforms recorded within RespiEvents. All data was transferred from RespiEvents to an Excel file via an ASCII cut. This process converted the data into a readable form, by representing a ll data as numerical

PAGE 39

29 values on a breath-by-breath basi s. These values represen ted breathing parameters in terms of volumes, times, derivatives, and rib cage and abdominal movements. A spreadsheet was created, to or ganize and store the data. Measured Breathing Parameters RespiEvents (Nims, 2002) allows for the acquisition and analysis of many breathing parameters, such as measures of tid al flow-volume loops, breath and heart rate parameters and Electrocardiogram (ECG) wave forms. For the purposes of the present study, fourteen breathing parameters were analyzed and compared across groups and tasks. Specifically, three volume measures were analyzed: inspir atory and expiratory amplitude (ViVol and VeVol, respectively) and the minute ve ntilation (Vent). Six timing measures were recorded: breaths per minute (Br/M), inspiratory tim e (Ti), expiratory time (Te), total breath time (Tt), a fractional inspiratory time (Ti/Tt) and the time to reach expiratory flow (PefTTe). Two measures that related specifically to thoracic displacement were analyzed: the percenta ge of rib cage contribution (%RC) and a Labored Breathing Index (LBI). And finally, three measures of respiratory efficiency were recorded: a rapid shallow breathing inde x (F/Vt), peak inspiratory flow (PifVt) and a measure of respiratory muscular efficiency and breathlessness (VePif ). More detailed definitions follow. Volume measures A total of three measurements relating to volume were chosen for this study. Previous evidence has shown that individuals with disordered larynges utilize different volumes of air during speaking tasks (Bunton, 2005; Saarinen et al., 2001; Sapienza et al., 1997; Schaeffer et al., 2002). Therefore, the present investigator s wanted to examine

PAGE 40

30 measures of respiratory volume in order to de termine if the obstruction at the level of the larynx impacted that amount of air inhaled or exhaled. It was thought that increased lung volumes would be needed in order to ove rcome this obstruction. The measurements related to volume are as follows: Inspiratory volume (ViVol). This measurement represents the inspiratory amplitude of the tidal volume. The tidal volum e is the volume of air that is inhaled and exhaled on a single breath. This parameter, therefore, represents the volume of air inhaled during each breath. Expiratory volume (VeVol). VeVol represents expiratory amplitude of the tidal volume, or the volume of air that is exha led on a single breath. During quiet breathing, this measurement should equal the inspiratory amplitude. RespiEvents begins to measure this when the inspiratory amplitude peak for each breath is reached and the thoracic volume begins to decrease. Ventilation (Vent). Ventilation is calculated by multiplying the ViVol by breaths per minute. This provides an assumed minute ve ntilation rate that represents respiratory muscle efficiency. Timing measures A total of six measurements relating to timing were chosen for this study. Previous research has not specifically targeted the respiratory parameters related to timing. Therefore, measures of respiratory tim e were analyzed in order to determine if the obstruction at the level of the larynx impacted the duration of breathing patterns. It was thought that because this population expe riences insufficient gl ottal pressure, the timing of respiratory patterns may be cha nged or adjusted to create an increased

PAGE 41

31 subglottal pressure needed to overcome this obstruction. The measurements related to volume are as follows: Breaths per minute (Br/M). Breaths per minute is the respiratory rate calculated breath-by breath. In other words, Br/M repr esents how many breaths would be taken if the individual maintained the respiratory rate seen at each breath. For example, if one breath lasted 10 seconds then the Br/M woul d be approximately 6, for that particular breath. However, if the next breath last ed 6 seconds, then the Br/M would be 10. Intuitively, quiet breathing results in a more stable and regular respiratory rate and ultimately a lower Br/M value, whereas, during activity (or for our purposes, speech breathing) the rate at which a breath is taken will be more likely to vary from breath to breath. Inspiratory time (Ti). This is the time in seconds from the initiation of a breath to its peak inspiratory volume. Expiratory time (Te). Te measures the time in seconds from the peak inspiratory volume to the end expiratory volume. Total breath time (Tt). Tt is the duration in second s from the initiation of a breath to the completion of that breath. Fractional inspiratory time (Ti/Tt). In this instance, inspiratory time (Ti) is divided by the total breath time (Tt) and given on a breath-by-breath basis. Time to reach peak expiratory flow (PefTTe). This value is presented as a percentage of expiratory time and represents the time to reach peak expiratory flow.

PAGE 42

32 Thoracic displacement A total of two measurements relating to the displacem ent of the rib cage and abdomen were chosen for this study. A st udy by Schaeffer et al. (2002) examined the physical displacement of the respiratory struct ures during speech breathing in individuals with dysphonia. They found a greater fre quency of abnormal expiratory abdominal movements in this population. Therefore, measures of thoracic displacement were analyzed in order to determine if the obstruc tion at the level of the larynx impacted the coordination or contribution of the ri b cage and abdominal movements. The measurements related to thorac ic displacement are as follows: Percentage of rib ca ge contribution (%RC). This measurement calculates the contribution of rib cage movement to the gene ration of tidal volume. It is obtained by dividing the rib cage amplitude by the tidal volum e at the peak of inspiratory volume. Its value is given as a percentage on a breath-by-br eath basis. In normal adults during quiet breathing, the rib cage involvement tends to exceed abdominal involvement and a typical percentage of rib cage invol vement is 60% (Nims, 2002). Labored breathing index (LBI). The labored breathing index measures thoracicabdominal coordination by comparing the power generated at the rib cage and abdominal level to the actual amount of power delivered (in the tidal vo lume). Perfect coordination would produce a ratio of 1.0. A mildly in creased LBI would produce a ratio of 1.3-2.0 and extremely high LBI values are above 3.0 (Nims, 2002). A high LBI value could possibly indicate muscular dysf unction or pulmonary obstruction.

PAGE 43

33 Respiratory Efficiency A total of three measurements relating to the respiratory efficiency were chosen for this study. These parameters are obtai ned through a comparison of the tidal volume to one or more measures of time, volume or movement. Little to no research has specifically examines these parameters in relation to speech breathing in patients with ADSD Therefore, these measures of respiratory efficiency were analyzed in order to determine if the obstruction at the level of the larynx impacted re spiratory functioning. The measurements related to respiratory efficiency are as follows: Rapid shallow breathing index (F/Vt). F/Vt is computed by dividing Br/M by the tidal volume, giving an indication of the respiratory frequency. Peak inspiratory flow (PifVt). This derivative reflects resp iratory drive. It is the peak inspiratory flow derived from the tidal vol ume. The higher the value, the greater the respiratory drive. Respiratory efficiency and breathlessness (VePif). This parameter reflects respiratory muscular efficiency and breathle ssness. It is calc ulated by dividing the minute ventilation by the peak inspiratory flow of the tidal volume. Minute ventilation is the volume of air inhaled and exhaled in one minute. VePif compares the respiratory drive to the respiratory output (ventilation). Respiratory driv e has shown to be greater in those with respiratory disease th an in controls (Nims, 2002). After much deliberation, these parameters were chosen because of the potential to show the changes in respiratory patterns. As mentioned in the litera ture review, measures of lung volume and physical displacement are altered in individuals with disordered larynges (Bunton, 2005; Sapienza et al., 1997; Schaeffer et al ., 2002; Vertigan et al.,

PAGE 44

34 2006). Specifically, evidence has shown that in dividuals may initiate speech at lower or higher lung volumes and terminate speech at lo wer lung volumes. In addition, aberrant respiratory movements may have been eviden t in this population. Therefore, these two categories, in addition to resp iratory timing and efficiency, were included in the present study to determine if individuals with ADSD utilize different respiratory behaviors when compared to a control group. Data reduction Once RespiEvents processed the informati on coming from the Respitrace vest, the data was converted to an Excel file via an ASCII cut. This process converts the waveform data into a readable form, by repres enting all data as numer ical values. More specifically, the data was placed into an Ex cel spreadsheet (Microsoft, 1998), in which the data was organized and stored. The cuts were performed on a br eath-by-breath basis as the guide for the rows of the spreadsheet and each breathing pa rameter served as a column name. The breathing parameters described above were extracted from the RespiEvents spreadsheet. The values obtained during each task (given in a breath by breath fashion) were averaged together and the standard devi ation was computed for each participant. This data was then combined to get an average for the reading tasks and an average for the conversation tasks for each participant. The averages for each breathing parameter within each task were used to determine if the differences seen between groups, gender and task were statistically significant.

PAGE 45

35 Reliability The data obtained from two participants from each group were randomly selected to determine the intra-judge reliability. The goal was to perform the same analysis used during the study to determine the beginning and end breaths for each ta sk resulting in the determination of which breath groups were used during the speaking tasks. This process was completed through review of the audiot apes, analysis of th e duration of each speaking task and the waveforms. The vertic al spike in the waveform, which correlates to the quick pull of the bands during the task s, gave a general marking of the placement of the tasks. The actual determining of the speaking portion was completed by comparing the waveform to the duration of th e speaking tasks. The information from the audiotape helped to establish if the sample contained initial paus es, large inhalations (without a speaking component), c oughing, throat clearing, etc. If this were true, then the initial breaths were excluded from the anal ysis. For example, one participant began a reading passage by inhaling a large volume a nd quickly exhaling while saying “alright.” The actual reading began with the next inha lation. This was excluded from the study because the respiratory pattern did not refl ect the rest of the sample. Because each participant engaged in four sp eaking tasks, and four particip ant’s data were analyzed, a total of 16 speaking tasks were re-determine d. Out of the 16 tasks, 14 beginning and end breaths were matched correctly. The remaining two anal yses were off +/1 breath. These had the beginning breath off by +/1 breath with correctly matching the end breath. In total, these four participants contribu ted a total of 148 breaths for the study’s analysis. During reliability te sting, 146 breaths matched correct ly. This means that when using the methods described above, 98.6% of the breaths would match correctly.

PAGE 46

36 One factor seemed to have influenced why the reliability testing did not match exactly. The discrepancy was attributed to small errors in judgment of which wave within the waveform indicated the initial sp eaking breath. This was the case because, often, the waveform was a shaky line with br eaths squeezed closely together. When this was the case, the next breath or set of breat hs that were obviously part of the speaking task was used as the initial breath. During reliability testing, the same judgment calls were attempted and judged differently. Ne vertheless, this occurred infrequently and reliability was considered to be very good. Statistical Analysis Each breathing parameter was analyzed using a multivariate analysis of variance (MANOVA). All dependent variables (i.e. th e number measures of breathing) were compared across the three independent va riables: group (ADSD vs. control), gender (male vs. female) and task (reading vs. c onversation). When analyses yielded an interaction between the three independent variables, post hoc an alysis testing was completed using Tukey A procedures. Effect sizes were calculat ed as appropriate.

PAGE 47

37 Results The purpose of this study was to collect data on various breathing parameters while participants were engaged in two sp eaking tasks. In order to achieve this, participants wore a vest connected to co mputer software that computed breathing parameters via inductance plethysmography. The resulting breathing parameters were analyzed across groups (ADSD vs. control), gender and speaking tasks. The various breathing parameters were chosen due to suspic ion that the obstruction at the glottal level would alter the selected parame ters. All parameters were cl assified and grouped into one of four categories: volume, timing, thoracic displacement and respiratory efficiency. Results are presented below according to each category. Volume The first set of MANOVAs were perform ed for the volume measures, inspiratory volume (ViVol), expiratory volume (VeVol) and minute ventilation (Vent). The first analysis considered inspiratory volume (V iVol). Results revealed a significant interaction between task and group, F (1,26) = 4.212, p = .05 p 2 = .139. Post hoc testing with the Tukey A proce dures indicated that two out of three pairwise comparisons of interest were significant ( p <.05) As illustrated in Figure 2, the amount of air inhaled during reading for the control group was signifi cantly less than the amount of air inhaled during reading for the ADSD group and in c onversation for the control group. To be specific, the ADSD group had average Vi Vol measures of 564.11 Aml (arbitrary

PAGE 48

38 milliliters) for reading and 586.43 Aml for c onversation, while the control group had ViVol measures of 393.42 Aml for reading and 530.46 Aml for conversation. This would suggest that the speaker s with ADSD were using equi valent inhaled air volumes while the control speakers i nhaled less air for reading. Figure 2. Chart of the mean inspiratory volum es compared between participant group and speaking task 0 200 400 600 ADSDControlGroup Aml Reading Convo The second analysis considered expiratory volume (VeVol). The results of this MANOVA indicated that only the main effect of speech sample was significant, F (1,26) = 5.499, p = .027, p 2 = .175. Therefore, differences in volume expired depended on the task. Both groups expired a greater volu me of air during conversation tasks when compared to reading tasks (see Figure 3).

PAGE 49

39 Figure 3. Chart of the mean expira tory volume averaged across speaking task 0 200 400 600 ReadingConversationTaskVolume in Aml The last MANOVA measuring volume was ventilation (Vent) In this case, there was a significant main effect for group, F (1,26) = 5.519, p = .027, p 2 = .175. As illustrated in Figure 4, the ADSD group had an average ventilation of 11.24 Al/min while the control group had an average ventilation of 7.645 Al/min. Theref ore, those in the ADSD group have, on average, a higher ventila tion rate no matter th eir speaking task. Figure 4. A chart of the mean ventilation (Vent) rate compared across participant groups. All MANOVAs performed for the volum e parameters yielded a statistical difference. The control group inhaled a significantly lower volume of air while reading when compared to conversation and when compared to the group with ADSD in both tasks. Both groups exhaled a significantly lo wer volume of air during reading tasks than 0 5 10 15 ADSDControl GroupVolume in Al/minute

PAGE 50

40 during conversation. There was a significan t difference across groups when comparing the minute ventilation. Specifically, the ADS D group had significantly higher ventilation rates than the controls when the two tasks were averaged together. Th e effect sizes for all parameters were small, but suggested practical significance and that the inclusion of more participants may increase these effect sizes. Timing The next set of analyses considered m easurements related to time. Specifically, breaths per minute (Br/M), inspiratory time (T i), expiratory time (Te), total breath time (Tt), a fractional inspiratory time (Ti/Tt) and the time to reach peak expiratory flow (PefTTe) were the timing parameters th at were analyzed. The first MANOVA considered breaths per minute (Br/M). The re sults revealed a significant main effect for group, F (1,26) = 4.769, p = .038, p 2 = .155 The ADSD group had an average of 22.66 breaths/min while the control group’s averag e was 17.58 breaths/min. As illustrated in Figure 5, the ADSD group, on average, inhaled more frequently during speaking tasks than did those in the control group.

PAGE 51

41 Figure 5. A chart of average breaths per minute compared across participant group 0 5 10 15 20 25 ADSDControl GroupRate in Al/min The next MANOVA considered inspiratory time (Ti) differences across groups, gender and tasks. Results revealed a significant main effect for task, F (1,26) = 5.809, p = .023, p 2 = .183. The average inspiratory time during the reading task was 1.19 seconds and 1.34 seconds during conversation (see Figure 6). The Ti value was greater in the conversational tasks than in the reading tasks. Figure 6. A chart of the average inspirator y time compared across speaking tasks. 0 0.5 1 1.5 ReadingConversation TaskTime in seconds The next analysis considered expirato ry time (Te). The results revealed no significant main effects or in teractions. Therefore, there were no differences across groups, gender or tasks for the durat ion of expiration (see Figure 7).

PAGE 52

42 Figure 7. A chart of the mean expiratory times compared between participant group and across speaking tasks. Te0 1 2 3 4 ReadingConversation TaskTime in seconds ADSD Control The next analysis of the timing measures considered total breath time (Tt). Once again, there were no significan t main effects or interactions. Both groups had similar respiratory cycle durations regardless of speaking task or gender (see Figure 8). Figure 8. A chart of the aver age total breath duration comp ared across participant group and speaking task. 0 1 2 3 4 5 ReadingConversationTaskTime in seconds ADSD Control

PAGE 53

43 The fifth MANOVA consider ed the fractional inspiratory time (Ti/Tt). The results indicated a significant inte raction between task and gender, F (1,26) = 7.442, p = .011, p 2 = .223. In other words, differences am ong the Ti/Tt values for a specific task differed by gender. Post hoc testing with Tuke y A procedures revealed that four out of six pairwise comparisons were significant ( p < .05). As illustrated in Figure 9, the men’s values for Ti/Tt in conversation were greater than the men’s values for this parameter in reading and greater than the women’s values in conversation. The men’s values were equivalent to the women’s in reading and the women showed no differences in Ti/Tt values when comparing conversation and read ing. Hence, Ti/Tt values seem to be greatest for men in conversation. Figure 9. A chart of the average percentage of inspiratory time compared across gender. 0% 10% 20% 30% 40% 50% ReadingConversation TaskPercentage of total breath Men Women The last MANOVA for the timing paramete rs considered the time to reach peak expiratory flow (PefTTe). In this case, only the main effe ct for task was significant, F (1,26) = 5.599, p = .026, p 2 = .177. The average time to re ach peak expiratory flow (as a percentage of expiration time) was 45.72 % for reading and 50.09 % for conversation.

PAGE 54

44 This indicates that, in general, speakers take l onger to reach peak expiratory flow while in conversation than during read ing tasks (see Figure 10). Figure 10. A chart of the averag e percentage of expiratory ti me to reach peak expiratory flow compared across speaking tasks 0% 20% 40% 60%ReadingConversation TaskPercentage of expiratory time A total of four out of the six parameters relating to measures of time that were analyzed were found to be statistically si gnificant when comparing across groups, gender and task. A main effect for group was found in one parameter, breaths per minute, with the ADSD group having a higher frequency of br eaths per minute. Two parameters were found to be statistically signifi cant with reference to task: inspiratory time and the time to reach peak expiratory flow. Specifically, a l onger period of inhalation with a shorter time to reach peak expiratory flow was found in reading tasks. One parameter was found to have an interaction between task and gender. Men had a higher fractional inspiratory time in conversation than in reading and grea ter than women’s values in both speaking tasks. No significant main effects or inte ractions were found in two timing parameters: expiratory time and the averag e duration of the total breath. Again, the effect sizes for

PAGE 55

45 the parameters were small, but suggest practi cal significance. If mo re participants were included in this study, larger e ffect sizes may have been found. Thoracic Displacement Two parameters were included in this st udy that related directly to rib cage and abdominal movements: the percent of ri b cage contribution (%RC) and the labored breathing index (LBI). The first MANO VA considered the amount of rib cage contribution to the tidal volume. The results indicated that the main effect of task was significant, F (1,26) = 18.329, p <.001, p 2 = .413. The rib cage contributed to 57.7 % of the tidal volume during read ing tasks and 51.7 % during conve rsational tasks. This indicates that breathing pattern s during reading tasks utilize more rib cage movement (see Figure 11). Figure 11. A chart showing the percentage of rib cage contribution across speaking tasks. 0% 25% 50% 75%ReadingConversation TaskPercentage of rib cage contribution The second analysis considered the labor ed breathing index (LBI). The results revealed a group difference, F (1,26) = 7.716, p = .010, p 2 = .229. The group with ADSD had a higher index of labored breathing than the controls. Speci fically, those in the ADSD group used more movements to produce le ss ventilatory output than the controls (see Figure 12).

PAGE 56

46 Figure 12. A chart of the index of labored br eathing compared acro ss participant group. 0.00 0.25 0.50 0.75 1.00 1.25ADSDControl GroupLBI Analysis of parameters related to rib cage and abdominal movement yielded significance in both parameters. The main effect of task was significant for the percentage of rib cage contri bution with the rib cage cont ributing more during reading tasks. The effect size ( p 2 = .413) supporting this finding was large. Also, a group difference was found when comparing th e LBI. Specifically, the ADSD group demonstrated a higher degree of labored breath ing than the controls. In this case, the effect size ( p 2 = .229) was moderate, supporting the si gnificance of this finding as well. Respiratory Efficiency The next category of measures involved measures of respiratory efficiency. These parameters were derived through the co mparison of tidal volume with one or more parameters related to volume, timing or t horacic displacement. The values obtained represented a rapid shallow br eathing index (F/Vt), peak in spiratory flow (PifVt) and respiratory muscular effici ency (VePif). The first MANOVA considered the rapid shallow breathing index (F/Vt). There were no significant main effects or interactions. Therefore, there were no differences across groups, tasks or gender for the rapid shallow breathing index.

PAGE 57

47 The results of the MANOVA for peak insp iratory flow (PifVt) also revealed no significant main effects or interactions. However, th e task by group interaction approached significance, F (1,26) = 3.746, p = .001, p 2 = .126. While the effect size is small, it would suggest that there may be differences by task and group if more participants were run. As illustrated in Figur e 13, there are comparable values in regards to conversation; however, ther e are obvious differences in the reading tasks between the two groups. This finding merits continued investigation. Figure 13. A chart demonstrating the values for peak inspiratory flow compared across participant group and speaking task. 0 500 1000 1500 ReadingConversation TaskAml/second ADSD Control The last analysis for the derivational measures consider ed muscular efficiency and breathlessness (VePif). This analysis reveal ed a significant interac tion between task and gender, F (1,26) = 6.681, p = .016, p 2 = .204. Post hoc testing revealed that males evidenced higher values for PefVt in the conve rsational tasks than males in the reading tasks or the females in both tasks (see Figure 14). In addition, the analysis revealed that the main effect for group was significant, F (1,26) = 11.880, p = .002, p 2 = .314. Those with ADSD experienced a greate r degree of muscular inefficiency and breathlessness in both speaking tasks when compared to the control group (see Figure 15).

PAGE 58

48 Figure 14. A chart depicting the differences in respiratory muscul ar efficiency and breathlessness compared across sp eaking tasks and participant group. 0 0.08 0.16 ReadingConversation TaskIndex of muscular efficiency Men Women Figure 15. A chart showing the difference in respiratory muscular efficiency and breathlessness compared across participant group. 0 0.08 0.16 ADSDControl GroupIndex of muscular efficiency Results of the statistical analysis regarding the derivative parameters revealed only one parameter with significant findings fo r the independent vari ables. Specifically, an interaction between task and gender was found for VePif with men having a higher degree of muscular inefficiency during the conversation when compared to reading and for both measures in women. In addition, a significant main effect for group was found for this parameter. Specifically, the ADSD group was found to have a higher degree of muscular inefficiency and breathlessness than th e controls in both task s. The effect sizes for this parameter were moderate ( p 2 = .314), supporting the si gnificance of these findings. No significant diffe rences were found in the ra pid shallow breathing index,

PAGE 59

49 while the results of the analysis of peak inspiratory flow approached significance, indicating that there may be an interac tion between task and group. The ADSD group, during reading tasks, appeared to have had higher values of peak inspiratory flow than the controls during reading or both groups dur ing conversational task s. The effect sizes for these parameters were relatively small, po ssibly indicating that a larger sample size would be needed in order to establish significance. Summary of Findings MANOVAs were performed for each of th e 14 breathing parameters included in this study. Many significant differences were found when comparing across group, gender and task. A main effect according to group was found in four parameters: VENT, Br/M, VePif and LBI. The ADSD group was found to ventilate more liters of air per minute and take more breaths per minute than the control group, re gardless of speaking task. This group performed the speaking task s with a higher index of labored breathing and respiratory inefficiency and breathlessne ss. In addition, main effects according to task were found in VeVol, Ti, PefTTe and %RC. Both groups had longer inspiratory times, larger expiratory volumes and took l onger to reach peak expiratory flow during conversational tasks. Also during conversation, the rib cage contribute d less to the tidal volume than during reading tasks. Additiona lly, several interactio ns were found to be significant. An interaction between task and group was found to be statistically significant in ViVol and approaching significan ce in PifVt. Those in the control group inhaled significantly less air in convers ation than the ADSD or both groups in conversation. In addition, the ADSD group in creased their respiratory drive in reading tasks when compared to the control group in reading or both groups in conversational

PAGE 60

50 tasks. Also, an interaction was found between task and gender in Ti/Tt and VePif. Men demonstrated a greater fractional inspiratory ti me with a greater muscular efficiency in conversation versus in reading or when compar ed to women in both tasks. Overall, these results indicate that individuals with ADS D evidenced more difficulties with breathing efficiency than the control group. These fi ndings are consistent with the variable presence of obstruction at the level of the larynx.

PAGE 61

51 Discussion The primary objective of the current pilo t study was to determine if individuals with adductor spasmodic dysphonia differed from ageand gender-matched controls in various breathing parameters while engaged in speaking tasks. It was hypothesized that individuals with ADSD might suffer from diso rdered breathing due to obstruction at the level of the larynx. To determine if the part icipants utilized alte red breathing patterns while speaking, volume, timing, thoracic disp lacement and measures of respiratory efficiency were computed and compared across group, speaking task and gender. The results indicated that, indeed, various signifi cant differences existed between these two groups. Main effects according to group were found in four out of the 14 parameters analyzed. Those with ADSD were found to have statistically higher ventilation rates and frequency of breaths per minute. In additi on, this group experienced a higher degree of muscular inefficiency/breathlessness and labored breathing. Differences according to task were found as well. Specifically, the pa rticipants utilized l onger inspiratory times, exhaled a larger volume of air and took longe r to reach peak expi ratory flow during conversational tasks compared to reading tasks. There was also less rib cage contribution to the tidal volume in the c onversational tasks. There were no main effects related to gender, however, various inte ractions between task, gender and group were found. These findings will be discussed further in light of the research questions.

PAGE 62

52 Volume First, the research questions focused on potential differences in volume across groups, task and gender. Thr ee different parameters were measured: inspiratory volume (ViVol), expiratory volume (VeVol) and th e minute ventilation (V ent). Statistical analyses of these parameters revealed main effects for task and group, as well as an interaction between task a nd group. No differences according to gender were found related to volume measures. These findings suggested that, in relation to volume measures, differences do exist between groups, as well as in speaking task. Most critical to the current study ar e the differences found between the two groups. Individuals with ADSD utilized highe r ventilation rates in both speaking tasks when compared to the controls. The ventilat ion values obtained by RespiEvents refer to the total volume of air i nhaled during a minute of speaking. Since the ADSD group experienced higher Vent values, they utilized more air while speaking. Interestingly, the two groups did not differ in their values of av erage inspiratory volume (ViVol). In other words, the ADSD group utilized a greater vol ume of air throughout the speaking tasks, without actually increasing th e volume of air inspired on each breath. To accomplish this, therefore, the ADSD group increased the number of breaths taken per minute (Br/M). For that reason, the ADSD group was not able to use their breaths as efficiently as the control group because they required more air to complete the tasks. Increased ventilation rates coupled with similar inspiratory volumes reveals that the ADSD group used their inspired air at a quick er rate. This finding was verified by an analysis of the audiotapes to compute averag e number of syllables produced per minute. On average, across speaki ng tasks, the group with ADS D produced 189.47 syllables per

PAGE 63

53 minute. On the other hand, the control group produced 230.51 syllables per minute. Therefore, this assumption was upheld as the ADSD group did produce fewer syllables per minute during the speaking tasks. In gene ral, these individuals increased ventilation rates to overcome the glottal resistance, which in turn lead to a decrease in the number of syllables per minute. And, because of the hi gher rate of ventilation, they replenished their air supply more frequently in order to initiate speech. The individuals with ADSD attempted to overcome the increased laryngeal resistance associated with the spasms by increasing the overall minute inspiratory volume. In addition to the group effect for ventil ation, a task effect was found related to expiratory volume (VeVol). Participants e xhaled a smaller volume of air during reading tasks compared to conversational tasks. Thes e findings support eviden ce that respiratory behaviors related to volume are altered in di fferent speaking tasks. Schaeffer et al. (2002) examined expiratory volumes in oral r eading tasks. On average, smaller values were found in the less linguistically comple x speaking task. Although this difference was not statistical, this tendency can be applied to the current study’s findings. The reading task presented the participants with a lower cognitive-linguistic demand. Linguistic markers, such as commas and periods, in a ddition to the grammatical markers, such as phrase boundaries, provided the participants with acceptable places to replenish their breath supply. Conversationa l speech, on the other hand, demanded a higher degree of cognitive planning, because it did not provide these markers. In conversational tasks, therefore, the participants experienced a higher demand on discourse planning, along with the absence of given ma rkers indicating when to perf orm the respiratory events. This suggestion is further supported by the fact that the average syllables per minute for

PAGE 64

54 the reading task across all participants wa s 255.58, while for the conversation tasks, the average was 194.15. Task differences were also found by Hixon and Hoit (2005) who reported an overall slower speech rate for speaking ta sks that have a high cognitive-linguistic demand. Hence, both groups in this study e xpired more air during conversation because the cognitive-linguistic demand was higher. Re ading of the Rainbow Passage (Fairbanks, 1960), on the other hand, was a more familiar (i.e., they had read it before in other speech evaluations) and linguistically determined ta sk that structured breathing patterns. Therefore, it seems as though a more effici ent utterance plan was developed for the reading tasks, as all participants produced mo re syllables on less air during these tasks. Lastly, a task by group interaction was found. The control group inhaled a lower volume (ViVol) during reading task s. This implies that the control group utilized an even more efficient utterance plan than the ADSD group to complete the reading tasks because they completed the tasks with a lower volume of air. It appears the ADSD group did not demonstrate this tendency to inhale less air during reading be cause the increased ventilation was required to overcome laryngeal obstruction regardless of speaking task. The results of the statistical analyses revealed that di fferences between the groups and tasks exist in relation to volume meas ures. Specifically, those with ADSD were found to have a higher rate of ventilation than the controls, signifyi ng the need for more air to complete the speaking tasks. This was the case, presumably because this group increased airflow through the glottis in or der to prevent the sp asms from halting phonation. Also, the measures related to expi ratory volume yielded a difference between tasks, as both groups expired a smaller volu me of air during the reading tasks. In

PAGE 65

55 addition, the control group inhale d a significantly lower amount of air during these tasks. These latter differences demonstrate how the cognitive-linguistic demand influenced the respiratory behaviors related to volume. The speaking demands on reading were lower, and, therefore, the participants used less ai r. The control grou p developed an overall more efficient utterance plan as evidenced by their overall lower inspiratory lung volume during the reading tasks. Timing The research questions also focused on differences in the timing of breathing patterns. Six measurements related to timi ng of respiratory patterns were considered: breaths per minute (Br/M), inspiratory time (T i), expiratory time (Te), total breath time (Tt), a fractional inspiratory time (Ti/Tt) and the time to reach peak expiratory flow (PefTTe). The results of statistical anal yses revealed that differences were found depending on the type of timing measure. Mo st striking was the finding that those with ADSD tended to have a higher number of breat hs per minute (Br/M). In other words, these individuals replenished their breath supp ly more frequently th an controls. This seems obvious, considering increased ventilat ions rates were also found. This finding further suggests that the ADSD group required a greater vol ume of air during speaking tasks, presumably to overcome the laryngeal resistance. In addition to the main effect regardi ng group, main effects regarding task were found in inspiratory time (Ti) and the time to reach peak expiratory flow (PefTTe). Both groups utilized longer periods of inhalation a nd took longer to reach pe ak expiratory flow during conversational tasks. This may be rela ted to a reduced need for discourse planning during reading tasks because linguistic materi al was provided. Perhaps, because more

PAGE 66

56 cognitive and linguistic pl anning was required during c onversation, the individuals increased inspiratory times as a met hod to increase planning time. According to Hixon and Hoit (2005) there are manifestations in speech breathing during “activities that require on-line formula tion” (p. 90). For example, the effects of demanding cognitive-linguistic processing can yield brief s ilent pauses (200-500 ms), breath holds and non-speech expirations. Alt hough these symptoms were not specifically analyzed, it is likely that the increased cognitive load gene rated these behaviors in the conversational samples. These behaviors will influence the time to reach peak expiratory flow because they are directly interfering with expiration. In addition, as mentioned above and evidenced by the syllables per minute data, the participants had a lower speaking rate in conversational tasks. Ag ain, Hixon and Hoit (2005) reported that this data should be expected in more demanding speaking tasks. Therefore, the increased cognitive load in the conversational tasks yielded changes in the timing of the speech breathing behaviors. Increased in spiratory times were used as extra time to formulate the upcoming speech. In addition, the demand on fo rmulating the discourse manifested itself in the expiratory flow because it took longe r to reach peak flow rates in conversation. Reading, on the other hand, did not require utterance planning to the same extent; hence, the participants were able to initiate speech qui cker and more efficiently during this task. Lastly, a task by gender interaction was found, in that, men in both groups tended to have a larger fractional inspiratory tim e (Ti/Tt) than women during the conversation tasks or both genders during r eading. In other words, in c onversational tasks, men spent a larger percentage of the total breath duration inhaling than the women did. This can be explained through the basic anatomical diffe rences between men and women. Men, in

PAGE 67

57 general, have a larger lung capacity because they have la rger body sizes (Hixon and Hoit, 2005). This fact, coupled with the a bove-mentioned finding that both genders experienced longer inspiratory time during c onversation tasks, explains why men would experience a longer percentage of inspirat ory time during conversational tasks. Given these findings related to the timing of respiratory events, it is interesting that no differences according to group, task or gender were found in relation to the expiratory (Te) and total breath (Tt) durations. Phra sing during speaking tasks is controlled via a linguistic plan. The lack of a statistical difference he re would suggest that individuals varied greatly in their expiratory patterns. The results of statistical an alyses of measures related to the timing of respiratory events revealed differences between groups, tasks and an interact ion between gender and speaking tasks. Specifically, those with ADS D were found to replen ish their air supply more often than the control group. This is not surprising, given the finding related to increased ventilation rates with in this group. Also, the measures related to inspiratory time and the time to reach peak expiratory fl ow yielded a difference between tasks, as both groups took longer to inhale and to reach peak expiratory flow during conversation. Again, these differences were attributed to the differences in the demand of the cognitive and discourse planning between the two task s. In addition, the men used a greater portion of the total breath inhaling duri ng conversational tasks than the women, presumably because of the larger lung capacity in men. Thoracic Displacement Next, the research questions focused on measurements related to the thoracic displacement during respiration. The results of statistical analysis revealed a group

PAGE 68

58 difference when considering the labored br eathing index (LBI) with the ADSD group experiencing a higher degree of labored breath ing. In addition, a difference regarding task in the percentage of rib cage contribu tion (%RC) was found with the rib cage contributing to a greater extent in the reading tasks. A significant difference between gr oups was found with the ADSD group obtaining higher labored breathing (LBI) values. This was one of the most interesting findings because it supported the presence of disordered breathing patterns in this population. Because this measure consider s the degree of rib cage and abdominal coordination during respiration, it can be said that those with ADSD are experiencing altered and disordered breathing patterns. Pe rfect coordination between the two systems would yield an LBI value of 1.0. On averag e, the ADSD group obtaine d an LBI value of 1.13, while the controls had a value of 1.06. These findings can be compared to th ose found by Schaeffer et al. (2002), who examined thoracic displacement in individuals with dysphonia. Results from this study indicated that the individuals with dysphoni a utilized paradoxical abdominal movements by using physical breathing pa tterns that are not typical ly observed. The purpose and contribution of the abdominal patterns to the respiratory cycle was not clear and considered aberrant. Theref ore, the presence of lar yngeal pathology affected the physiological patterns of respiration in this population by influencing physical displacement of the abdomen. This can be c onnected to the results of the present study, in that those with ADSD are experienci ng discoordination between rib cage and abdomen. In a similar manner as Sch aeffer’s participants, the physiological underpinnings of respiration during speech are modified in individuals with ADSD. The

PAGE 69

59 laryngeal tension in the ADSD group resulted in the two sub-systems not functioning as a single unit, but, at times, wo rking in dyssynchrony. Figure 16 is an example of a RespiE vents waveform tracing in an ADSD participant. The quick vertical spikes in th e waveform represent the band pulls indicating the beginning or end points of the speaking tasks. It is interesting to see the very different waveforms the rib cage (middle tracing) and abdomen (bottom tracing) are producing. When comparing these traces to the ones produced by a control participant (see Figure 1), the differences in the respir atory patterns are obvious. For example, on the control tracing, the actual contribution of each tracing to the tidal volume (top tracing) is obvious. On Figure 16, howeve r, it is not clear how the abdomen is contributing to the tidal volume.

PAGE 70

60 Figure 16. An example of a RespiEvents wa veform produced by a participant with ADSD. According to Hixon and Hoit (2005), “whe ther reading aloud or speaking [in conversation], speech breathing tends to s how similar mechanical patterning … and engagement of similar muscular strategies” ( p. 89). With this in mind, it is interesting that the rib cage exhibited diffe rent levels of contribution between the speaking tasks. In this study, the rib cage (%RC) contributed to a greater extent in the reading tasks. This finding can be explained through the differences in discourse planning. It would seem that the increased recruitm ent of abdominal muscle occurred when the demand on the cognitive and linguistic pla nning systems were higher. The abdomen may have been utilized as a method to support expiration, in vi ew of the fact that it took longer to reach

PAGE 71

61 peak expiratory flow during the more c ognitively demanding tasks. Increasing the abdominal workload allowed the participants to control expiration while formulating the discourse and linguistic pl an needed to perform th e conversational tasks. Statistical analysis of measures relate d to thoracic displacement revealed that differences existed between group and task. A higher degree of labored breathing was found in the ADSD group. This was attri buted to physiological changes in the respiratory behaviors due to the obstruction at the level of the glottis. Measurements related to the contribut ion of the rib cage to the tidal volume showed a greater rib cage contribution in the reading tasks, presumably related to differences in discourse planning. It appeared the abdomen was recruited to a gr eater extent in conversational tasks to allow for the respiratory changes that occurred when the individual was formulating the linguistic plan. No differences related to gender were found regarding measures of thoracic displacement. Respiratory Efficiency The last set of research questions addre ssed measures of respiratory efficiency. These measures involved timing or flow rates being derived from tidal volume and provided a rapid shallow breathing index (F/Vt) peak inspiratory flow (PifVt) and an index of respiratory muscular efficiency and breathlessness (VePif). With regards to these parameters, there was a high degree of variability in both groups that may have masked group differences. Results of the statistical analysis indi cated that the ADSD group demonstrated an overall higher level of musc ular inefficiency and breathlessness when compared to controls. Again, these results are related to the disruption in normal respiratory patterns

PAGE 72

62 in those with obstructed larynges. This fi nding, coupled with the si gnificant group effect for labored breathing, exemplified the redu ced level of respir atory functioning in individual with ADSD. It appeared that those with ADSD were bot h inefficient in the muscular contributions to re spiration and in regards to the ease at which respiration occurs. In addition, an inte raction between task and gend er revealed that men during conversational tasks exhibited a higher degr ee of inefficiency a nd breathlessness than men in reading or women in both tasks. It is speculated that, again, the level of cognitive-linguistic utterance pl anning influenced this parameter. Because conversation required a higher degree of discour se planning, males in this ta sk became less efficient in their respiratory behaviors. It is not clear however, why women did not also demonstrate this trend. An interaction between task and group in values reflecting peak inspiratory flow (PifVt) approached significan ce. Here, the ADSD group had larger PifVt values during reading when compared to the control group in reading and both groups in conversation. According to the RespiEvent manual, the higher the PifVt value, the higher the respiratory drive (Nims, 2002). With this in mind, the ADSD group experienced greater respiratory drive during reading tasks. These individuals used a higher degree of effort in order to keep up with the flow of readi ng. The reading tasks were structured and respiratory events related to oral reading app eared to be linguistically controlled. On the other hand, conversational tasks allowed greater resp iratory freedom. These tasks did not have linguistic and grammatical markers in dicating when to perform the respiratory events. Therefore, the effort needed to produce conversational speech was lower.

PAGE 73

63 Because of the obstruction at th e larynx, this group needed to increase respiratory drive in the reading tasks to maintain the proper flow as it was determined by the linguistic utterance plan. Lastly, no differences were found in the rapid shallow breathing index (F/Vt). While the ADSD group was found to inhale more frequently during speaking tasks, they did not inhale so frequently as to indi cate rapid breathing. In addition, the ADSD group, when compared to the control group, was found to inhale a comparable volume of air (ViVol), further supporting that shallow br eathing was not evident in this population. The results of the statistical analyses revealed that di fferences between the groups, tasks and gender existed in rela tion to measures of respirator y efficiency. Specifically, those with ADSD were found to have a hi gher degree of muscular inefficiency and breathlessness. This finding reflects disorder ed breathing within thes e participants. Also, males were found to be less efficient in conversation than women. This finding was attributed to an increased need for discourse planning during conversat ional tasks that led to less efficient respiratory pa tterns. An interaction betw een group and task approached significance for the parameter of peak inspirat ory flow. Those with ADSD were found to have higher PifVt values, or a higher respiratory drive in the reading tasks. It was hypothesized that there was an in crease in the level of contro l required to complete the reading tasks. The ADSD group, therefore, need ed to increase effort in order to perform the respiratory events at the designated boundaries. Although tw o out of the three measures related to respiratory efficiency were not significant, the results still are valuable to the understanding of altered resp iratory patterns in indi viduals with ADSD.

PAGE 74

64 Most significant to the presen t study is the finding that t hose with ADSD have decreased muscular efficiency and breathlessness. Conclusions The results of the present study indicat e those with ADSD exhibit disordered breathing when compared to an ageand gender-matched control group. The objective data obtained in this study can be linked to these patient’s subjectiv e reports of a higher degree of effort needed during speaking. This effort is perceived as the effects of Botox treatment dissipates. Because of this, it a ppeared that the respir atory manifestations found in this current study during speech brea thing are influenced by the neurologicallyrelated laryngeal obstruction. Statistical differences be tween the two groups were found in measurements of volume, timing, thoracic displacement and resp iratory efficiency. In general, the ADSD group ventilated more air per minute (Vent), thereby requiring more breaths per minute (Br/M). In addition, their re spiratory behaviors were perf ormed with a lower degree of muscular efficiency and breathlessness (V ePif) and with a higher index of labored breathing (LBI). These results illustrated a high level of discoordination between the respiratory and phonatory systems, indicating th at the participants with ADSD exhibited some speech breathing parameters that were deviant from normal respiratory behaviors. Further, it added to the evidence base that those with disordered larynges experience alterations in their respiratory pa tterns and behaviors during speech. In addition to the group differences, significant findings were found when comparing tasks. A higher volume of exha led air (VeVol) and l onger inspiratory times (Ti) were found in conversational tasks. Also during these tasks, the rib cage contributed

PAGE 75

65 less to the tidal volume (%RC) and participants took longer to reach peak expiratory flow (PefTTe). These differences supported the previous research that suggested that respiratory patterns were altered when e ngaged in different speaking tasks (Hixon & Hoit, 2005). These task differences were attributable to the demands of cognitivelinguistic planning. This hypothesis was supported by Schaeffer et al. (2002) who confirmed changes in respiratory behaviors wh en the demand on planning was altered. In that study, longer speaking tasks, without gra mmatical indicators for when to replenish air supply, resulted in deviant respiratory pa tterns. This type of speaking condition can be likened to the conversational task used in this study because the demand for linguistic planning was high and linguistic indicators mark ing appropriate places for breath renewal were not present. On the other hand, gramma tical markers were in cluded in the reading tasks, which reduced cognitive load by providing a linguistically appropriate indicator to replenish air supply or alter the respirator y patterns for speech producing purposes. Clinical Implications The results of this study are clinically important to those who experience ADSD as well as for those who will work with th is population. Those who will work with ADSD in voice therapy will want to be cognizan t of the difference in respiration between reading and conversational tasks when choosing therapeutic tasks. Respiration appears to be more difficult in conversation for indivi duals with ADSD. Because the demands on cognitive and linguistic planning are higher dur ing conversational task s, reading seems to produce respiratory behaviors th at more closely resemble th e patterns in non-disordered individuals. In addition, read ing provides individuals with linguistic cues that seem to affect the respiratory behaviors. For example, periods and commas will indicate

PAGE 76

66 appropriate pauses or places to renew the br eath supply. These factors seem to positively influence breathing patterns during speaking tasks. Clinicians working with this population should understand the possible negativ e implications of spontaneous speech on the respiratory system. In order to cr eate the most supportiv e therapeutic tasks, reading could be used to allow for more cont rol of speech breathing. Once the client is ready to move into less structured therapy tasks, spontaneous speech can be attempted. Treatment of ADSD has shown to be most effective when Botox injections are combined with behavioral voice therapy (Cannito & Woodson, 2000; Murry & Woodson, 1995; Sapienza et al., 2000; Woo et al., 1992). Results of st udies that have examined respiratory behaviors prea nd post-Botox injection indicate a greater degree of airflow stability (Cantarella et al., 2006) as well as overall increased airflow rates with decreased laryngeal resistance (Adams et al., 1996). This establishes that, post-injection, individuals are experiencing cl oser to normal respiratory f unctioning. As reported above, increased respiratory effort is perceived in th ese patients as the Botox wears off. Thus, it is reasonable to suppose that physical respir atory symptoms may be a precursor for the emergence of the voice symptoms and signals a need for reinjection of Botox. In the researchers’ experience in sp eaking with individua ls with ADSD, a common complaint is a tightness in their ch est while speaking. Although this phenomena of sensing breathing difficulty was not spec ifically examined in this study, the implications are obvious. Results of this study confirmed disordered breathing in individuals with ADSD, especi ally in regards to respiratory efficiency and labored breathing. In addition, at the time of te sting, many subjects’ vocal quality was subjectively determined to be mildly strained /strangled. This indi cates that the voice

PAGE 77

67 symptoms need not be severe for the signifi cant respiratory behaviors to be evident. Therefore, the physical manifestations of th e disorder (i.e. tightness) may be a strong indicator of the physio logical manifestations (i.e. diso rdered breathing patterns) that occur as the effects of Bot ox wears off. A longitudina l study investigating this relationship would provide insigh t into this hypothesis. Overall, the results of the analysis regard ing measures of volume, timing, thoracic displacement and respiratory efficiency indicate that the Respitrace is a sensitive measure to determine differences in speech breathing. Those who will work with this population may want to invest in this instrumentati on and use it to monitor breathing patterns over the course of therapy and possibly train to breathe more effici ently during laryngeal spasms. Strengths of the Present Study There are a few strengths of this stud y. First, this pilot study included many parameters related to respiration that have not previously been examined in research related to ADSD. Previous studies have examined respiratory behaviors related to airflow and volume; however they have not examined timing, the physical displacement of the respiratory structures, or indices related to respir atory efficiency. This study, therefore, provides evidence regarding disorder ed speech breathing in this population that has not previously been reported. The results of this study showed th e participants with ADSD exhibited a higher degree of muscular inefficiency and breathlessness and labored breathing. Also, they were found to ventilate more air per minute and subsequently renew their breath supply more often.

PAGE 78

68 A second strength is that a control group was carefully established for a more reliable comparison of the data between gr oups. This adds further support to these findings because many studies compare their findings to the existing normative data. Normative data is not necessarily determined using the same methods and tasks as an experimental study. Therefore, since bot h groups completed similar tasks, the comparison between groups may be more va lid. Further, both men and women were included in this study, which allowed for a more thorough evaluation of the disorder. According to the National Spasmodic Dysphonia website, more women than men are diagnosed with this disorder (NSDA, 2006). Through the inclusion of six men in this study, gender comparisons were possible. Frequently only men or only women are selected for studies; however, th is severely limits the degree to which the results can be generalized. The study’s design is a third strength. Two different speaking tasks (reading and conversation) were included due to the evid ence that respiratory behaviors are altered depending of the speaking task (Hixon & Hoit, 2005). Use of different speaking tasks allowed for comparison between the task s to determine which one would better approximate non-disordered respirat ion for therapeutic tasks. Limitations of the Present Study Six limitations may have affected the results of this study. First, the study included a total of 30 participants. This is a small number of indi viduals and a larger number of individuals would ha ve increased the reliability of the results. Nevertheless, 15 individuals is a fairly large number of part icipants with ADSD for a research study. Fifteen people served the purposes of this study, although more would have been better.

PAGE 79

69 Second, limited health, medical and previ ous treatment history related to the disorder was obtained from the participants Although a few basic questions were asked, more information regarding th eir health, injection histor y, smoking history, time postonset, etc. would have allowed for a more comprehensive investigation of this population. If this data were obtained, differences accordin g to lifestyle, health, and other factors may have been found to influe nce speech breathing. In addition, if these individuals underwent voice therapy, they most likely experienced some degree of respiratory behavior modification. The effect s of this training may obviously influence respiratory parameters and this possibility was not controlled. The individuals in the ADSD group particip ated in this study prior to receiving their regularly scheduled Botox injection. ENT physicians frequently recommend that individuals receive Botox re-injections every three months in order to maintain a more stable voice over time. Therefore, in this experiment, the full effects of the Botox may not have fully worn off in the participants at the time of testing. Interestingly, however, patterns of disordered breathing were still evident, providing suppor t to this study’s conclusions. The Respitrace was determined to be a se nsitive measure for the purposes of the current study; however, problems with this instrument have been well documented. Band slippage is one problem that may have influenced the study. Although careful consideration was taken to pr event slippage, it may still have occurred unbeknownst to the researchers. Band slippage could sk ew the data obtained in the Respitrace waveforms. In addition to the slippage, base lines have been noted to drift upwards with this instrumentation (Leino et al., 2001; Ne umann et al., 1998). This possibility was

PAGE 80

70 taken into consideration and the waveforms were analyzed to make sure the drift did not influence the data. Even so, there could have been a small degree of baseline shift that influenced the data. Lastly, the tasks chosen for this study were fairly short in duration. On average, the reading passage took approximately 30 s econds to complete and most people were engaged in conversation for one minute. This fact has two possible im plications. First, baseline drift is thought to even itself out wh en the instrument is used for a longer period of time (Leino et al., 2001). Since the en tire procedure, including calibration and completion of the speaking tasks, would have taken less than 15 minutes, stabilization of the drift (if the drift took pl ace) probably did not occur. On the other hand, however, because the entire procedure was only 15 minutes or less, the baseline possibly would not have had the opportunity to drift upwards so mu ch as to significantly skew the data. On a different note, more in depth info rmation may have been obtained had the speaking tasks been longer in duration. If longer speaking tasks were used, a more in depth examination of respiratory behaviors in extended speaking tasks would have been obtained. It is possible that breathing patterns would change as the length of the speech sample increased. Directions for Future Research This study attempted to establish if respiratory behaviors were altered in individuals with ADSD. A lthough, it was found that there were critical important differences in breathing patterns between the controls and patients with ADSD, future research should consider other factors. Firs t, longer speaking tasks should be included. This will help to establish more data on th e breathing parameters utilized during normal

PAGE 81

71 speech. Stressing the system, so to say, will gi ve a more realistic l ook at the respiratory behaviors in this population. It must be said, however, that in creasing the required duration of conversational ta sks will also increase the de mand on discourse planning. Therefore, the differences regarding task may be even more apparent in the longer utterances. In the current study, participants were asked to partake in the study immediately preceding their Botox injections. Comparisons of the respiratory patterns preand postinjection will provide valuable information in regards to the effec tiveness of Botox. In addition, this will provide more detailed information regarding respiration when the vocal symptoms are not present. Clinically, if individuals with ADSD resume normal respiration post-injection, be havioral therapy may focus on establishing and maintaining these normal respiratory behaviors. It would also be benefici al to examine the respiratory behaviors in individuals who do not have any amount of Bo tox within their laryngeal musculature. Individuals who have never been injected or have not been injected for at least six months will allow for examination of the respiratory behaviors associated with ADSD that have not been influe nced by the effects of Botox. If this study were to be replicated, a fe w differences should be considered. The Respitrace allows for the calculation of severa l parameters not specifically examined in this study. For example, this study did not examine the time to reach peak inspiratory flow (PifTTi), the peak expira tory flow (PefVt) or the per centage of agreement between the direction of rib cage and abdominal movements during inspiration (PhRIB), expiration (PhREB) and the total breath (PhRTB). A lthough these parameters were not deemed critical to include in this study, it would be interesting to see if differences

PAGE 82

72 according to group are found in regard to these parameters. Examining the agreement between the direction of the surface displacement in the insp iratory, expiratory and total breath phases can further indicate disordered breathing as it relates to the physiological properties of respiration. Lastly, many participants felt awkwar d or uneasy about the conversational speaking tasks. They expressed anxiety about what to say and how long to talk. Picture description seemed to work well for many pa rticipants because it gave a referent to discuss. Others, however, were unsure or la cked confidence to perform this task. Therefore, the conversational ta sks should be reconsidered to determine a more effective way of obtaining data related to spontaneous speech. In addition, more group differences may have been seen if an unfamiliar reading task was used. The Rainbow Passage (Fairbanks, 1960) is commonly us ed in voice assessments. It would seem, therefore, that the majority of the ADSD participants would be, at least, somewhat familiar with the passage. Future research should include a r eading passage that would be unfamiliar to all participants. A Final Word Adductor spasmodic dysphonia is a focal dystonia that manifests itself during voice production. It has se rious detrimental influences in th e lives of those it affects. If additional and more comprehensive information is obtained regarding this disorder, better treatment and management methods can be establ ished. It is unfortuna te that little is known regarding this disorder. Further res earch should be conducted to describe the respiratory behaviors used by individuals with ADSD and the impact these behaviors have on their lives.

PAGE 83

73 References Adams, S. G., Durkin, L. C., Irish, J. C., W ong, D. L. H., & Hunt, E. (1996). Effects of Botulinum toxin type A injections on aerodynamic measures of spasmodic dysphonia. Laryngoscope, 106, 296-300. Braun, N., Abd, A., Baer, J., Blitzer, A., Stew art, C., & Brin, M. (1995). Dyspnea in dystonia. Chest, 107. 1309-1318. Bunton, K. (2005). Patterns of lung volume use during an extemporaneous speech task in persons with Parkinson disease. Journal of Communication Disorders, 38, 331348. Cannito, M. P., & Woodson, G. E. (2000). The spasmodic dysphonias. In Kent, R., & Ball, M., J. (Ed.) Voice quality measurement. San Diego: Singular Publishing Group. Cantarella, G., Berlusconi, A ., Maraschi, B., Ghio, A., & Ba rbieri, S. (2006). Botulinum toxin injection and airflow st ability in spasmodic dysphonia. Otolaryngology – Head and Neck Surgery. 134, 419-423. Fairbanks, G. (1960). Voice and articulation drillbook (2nd ed.). New York: Harper & Row. Goodglass, H., Kaplan, E., & Barresi, B. (2000). Boston Diagnostic Aphasia Battery – 3rd Edition. Boston, MA. The Psychological Corporation

PAGE 84

74 Haynes, J. R., & Netsell, R. (2001) The mechanics of speech breathing: A tutorial. Retrieved April16, 2006. Available from Southwest Missouri State University. www.missouristate.edu Higgins, M. B., Chait, D. H., & Schulte, L. (1999). Phonatory air flow characteristics of adductor spasmodic dysphonia. Journal of Speech, Language and Hearing Research, 42, 101-111. Hixon, T. J., Goldman, M. D., & Mead, J. ( 1973). Kinematics of the chest wall during speech production: Volume displacement of the rib cage, abdomen, and lung. Journal of Speech and Hearing Research, 16, 78-115. Hixon, T. J., & Hoit, J. D. (2005). Evaluation and management of speech breathing disorders: Principles and methods. Tucson, AZ: Redington Brown. Iwarsson, J. (2001). Effects of inhalatory abdominal wall movement on vertical laryngeal position during speech. Journal of Voice 15. 384-394. Jiang, J., O’Mara, T., Chen, H., Stern, J. I., Vlagos, D., & Hanson, D. (1999). Aerodynamic measurements of pati ents with Parkinson’s disease. Journal of Voice, 13, 583-591. Leino, K., Nunes, S., Valta, P., & Takala, J. (2001). Validation of a new respiratory inductive plethysmograph. Acta Anaesthesiologic a Scandinavia Avica, 45 104111. Lundy, D. S., Roy, S., Xue, J. W., Casiano, R. R., & Jassir, D. (2004). Spastic/spasmodic vs. tremulous vocal qua lity: Motor speech profile analysis. Journal of Voice, 18, 146-152.

PAGE 85

75 Makiyama, K., Kida. A., & Sawashima, M. ( 1998). Evaluation of expiratory effort on dysphonic patients on increasing vocal intensity. Otolaryngology Head and Neck Surgery, 118, 723-727. Mayer, O. H., Clayton, R. G., Jawad, A. F., McDonough, J. M., & Allen, J. L. (2003). Respiratory inductance plethysmography in healthy 3to 5year-old children. Chest, 124. 1812-1819. Murry, T., & Woodson, G. E. (1995). Co mbined-modality treatment of adductor spasmodic dysphonia with botulin um toxin and voice therapy. Journal of Voice, 9, 460-465. NSDA. (2006). FAQ: Who gets spasmodic dysphonia?. Retrieved November 30, 2005. Available from National Spasmodi c Dysphonia Association Web site, http://www.dysphonia.org Neumann, P., Zinserling, J., Haase, C., S ydow, M., & Burchardi, H. (1998). Evaluation of respiratory inductive plethysm ography in controlled ventilation. Chest, 113, 443-451. Nims. (2002). RespiEvents operation manual fo r heath care practitioners. Non-Invasive Monitoring System. RespiEvents, v5.2. North Bay Village, FL: NIMS. Plant, R. L., & Hillel, A. D. (1998). Dire ct measurement of sunglottic pressure and laryngeal resistance in normal s ubjects and in spasmodic dysphonia. Journal of Voice, 12, 300-314. Redstone, F. (2004). The effects of seati ng position on the respir atory patterns of preschoolers with cerebral palsy. International Journal of Rehabilitation Research, 27, 283-288.

PAGE 86

76 Saarinen, A., Rihkanen, H., Malmberg, L. P ., Pekkanen, L., & Sovijarvi, A. R. (2001). Disturbances in airflow dynamics and tracheal sounds during forced and quiet breathing in subjects with unila teral vocal fold paralysis. Clinical Physiology, 21 712-717. Sapienza, C. M., Stathopoulos, E. T., & Br own, W. S. (1997). Speech breathing during reading in women with vocal nodules. Journal of Voice, 11, 195-201. Sapienza, C. M., Walton, S., & Murry, T. (2000) Adductor spasmodic dysphonia and muscular tension dysphonia: Acoustic analysis of sustained phonation and reading. Journal of Voice, 14, 502-520. Schaeffer, N., Cavallo, S., Wall, M., & Dia kow, C. (2002). Speech breathing behavior in normal and moderately to severely dy sphonic subjects during connected speech. Journal of Medical Speech – Language Pathology, 10, 1-18. Titze, I. R. (1994). Principles of voice production. Englewood Cliffs, NJ: Prentice Hall Inc. Vertigan, A. E., Gibson, P. G., Theodoros, D. G., Winkworth, A. L., Borgas, T. B., & Reid, C. (2006). Involuntary glottal closur e during inspiration in muscle tension dysphonia. Laryngoscope, 116, 643-649. Woo, P., Colton, R., Casper, J., & Brewer, D. (1992). Analysis of spasmodic dyphonia by aerodynamic and laryngostroboscopic measurements. Journal of Voice, 6, 344-351. Woodson, G. E., Zwirner, P., Murry, T., & Swenson, M. R. (1992). Functional assessment of patients with spasmodic dysphonia. Journal of Voice, 6, 338-343.