THE EFFECT OF AIR FLOW AND MEDIAL ADDUCTORY COMPRESSION ON VOCAL EFFICIENCY AND GLOTTAL VIBRATION G E R A L D S. BERKE, MD. D A V I D G . H A N S O N , MD, BRUCE R. G E R R A T T , MD. T E R R E N C E K. T R A P P , MD, C A R O L Y N M A C A G B A , MD, and M A N U E L N A T I V I D A D , AB, Los Angeles, California
This s t u d y u s e d a n In v i v o c a n i n e m o d e l t o Investigate t h e effects of v a r y i n g v o c a l f o l d resistance b y e l e c t r i c a l l y stimulating t h e recurrent l a r y n g e a l n e r v e w h i l e monitoring m e d i a l a d d u c t o r y c o m p r e s s i o n of t h e v o c a l folds, glottal airflow, a n d v o c a l intensity. T h e effects o f I n c r e a s i n g a i r f l o w o n glottal v i b r a t i o n w e r e a l s o e x a m i n e d s t r o b o s c o p i c a l l y a n d b y m e a s u r e m e n t of o p e n quotient. T h e results I n d i c a t e d that Increasing intensity b y m e d i a l a d d u c t o r y c o m p r e s s i o n w a s m o r e efficient t h a n b y Increasing airflow. Increasing airflow p r o d u c e d a significantly g r e a t e r o p e n quotient a n d v o c a l fold v i b r a t o r y e x c u r s i o n . (OTOLARYNGOL HEAD NECK S U R G 1 9 9 0 ; 1 0 2 : 2 1 2 . )
The study of the factors affecting vocal intensity is complicated by difficulty in measuring all the variables of interest. Most often, investigators have measured subglottal pressure (Psub), glottal airflow (U), and vo cal intensity (I). Vocal fold resistance has not usually been measured directly, but rather has been calculated from these other measures. This study, using an in vivo canine model, investigated the effects of varying vocal fold resistance by electrical stimulation of the recurrent laryngeal nerve (RLNS) while monitoring medial ad ductory compression (MAC) of the vocal folds, glottal airflow, and vocal intensity.
BACKGROUND 1
Van den Berg was one of the first investigators to emphasize the importance of subglottic power in vocal intensity, which he estimated as the product of mean subglottic pressure and mean flow rate. Rubin et a l . used a tracheal puncture technique to study the effect of Psub and U on intensity. They concluded that sub glottic pressure had a much greater effect on intensity 2
From the UCLA School of Medicine, Division of Head and Neck Surgery and VA Medical Center. Supported by Veterans Administration Medical Research Funds and NIH Grant NS 20707-05. Presented at the Annual Meeting of the American Academy of Otolaryngology-Head and Neck Surgery, Washington, D.C., Sept. 25-29, 1988. Submitted for publication Dec. 5, 1988; accepted May 15, 1989. Reprint requests: Gerald S. Berke, MD, UCLA School of Medicine, Division of Head and Neck Surgery, CHS-62- ! 39, 10833 Le Conte Ave.. Los Angeles, CA 90024-1624. 23/1/13905
production than airflow, which had little or no effect. Isshiki reported that resistance or subglottic pressure had the greatest effect on intensity at low fundamental frequencies (F s), whereas at high frequencies airflow had the greater effect. Koyama et al. , using an in vivo canine model, concluded that airflow was more im portant than RLNS in controlling the intensity of voice production. They assumed that RLNS was directly pro portional to laryngeal stiffness. Other authors have not concurred with their conclusions. Timcke et al. reviewed the relationship of open quo tient (OQ) to intensity of voice production. Using high speed cinematography, they found that as open quotient (i.e., the proportion of time the glottis was open within each glottal cycle) decreased, vocal intensity usually increased. The above studies and others appear to have as sumed that resistance of the vocal folds is proportional to vocal fold stiffness, implying a linear relationship of resistance to airflow. One aim of this study is to reex amine the validity of this assumption. Previous in vivo animal studies have demonstrated that, under conditions of steady subglottic flow, in crease in electrical stimulation to the recurrent laryngeal nerves was associated with induced phonation of a higher pitch and greater loudness. This study exam ined effects of variation in airflow and electrical stim ulation of the laryngeal nerves on intensity and open quotient. 3
0
4
2 3 5
6
711
59
METHODS In v i v o p r e p a r a t i o n . Mongrel dogs were anesthe tized with an intramuscular injection of 2 ml ketamine, followed by intravenous pentobarbital titrated to loss
212
Downloaded from oto.sagepub.com at SAGE Publications on June 21, 2016
02 Number3
tev
9 9 0
V o c a l efficacy a n d glottal vibration
CAMERA CONTINUOUS XENON -*»•' LIGHT SOURCE
Video or 35 m m S L R
STROBOSCOPE
213
Subsequent Data Analysis
(strobe trigger)
4 CHANNEL FM TAPE RECORDER
Sup. l a r y n g e a l π
- (dEGG) NERVE STIMULATOR' (bilateral)
-(EGG)—I
NERVE STIMULATOR (bilateral)
-(PGG)-
PRESSURE TRANSDUCER SPONTANEOUS RESPIRATIONS
MULTICHANNEL STORAGE OSCILLOSCOPE
- (press.)-)
R e c u r r e n t n. X-Y
PLOTTER
Fig. 1. A n in vivo c a n i n e p r e p a r a t i o n a l l o w e d simultaneous monitoring of v i d e o stroboscopy, p h o t o g l o t t o g r a p h y , e l e c t r o g l o t t o g r a p h y , a n d subglottic pressure a n d airflow.
°f the corneal reflex. The animals were placed in supine Position on an operating table (Fig. 1) and direct lar yngoscopy was performed to confirm normal laryngeal anatomy. A 7-mm oral endotracheal tube was inserted, through which the animal breathed spontaneously. Through a vertical midline incision, the strap muscles and sternocleidomastoid muscles were retracted later ally to expose the larynx and trachea. The external branches of the superior laryngeal nerves were isolated at their entrance into the cricothyroid muscle. Harvard bipolar electrodes were applied to the nerves. The re current laryngeal nerves were isolated 5 cm inferior to the larynx, and bipolar electrodes were applied. A sue through the thyrohyoid membrane was used to sus pend the epiglottis anteriorly to improve visualization °f the vocal folds. A distal tracheotomy was made for Placement of an endotracheal tube to permit the animal to breathe spontaneously. An additional proximal tra cheotomy was performed through which a cuffed tra cheotomy tube was placed, with its tip resting 10 cm below the glottis. The cuff of the superiorly directed tube was inflated to just seal the trachea. Room air was bubbled through 5 cm Η,Ο at 37° C for warming and t U r
humidification and passed through the cephalad trache otomy tube. The temperature in the animal's trachea was measured at 15-minute intervals to assure a con stant air temperature of 37° C. A Grass mode) 54H stimulator (Grass Instruments, Quincy, Mass.) provided variable voltage stimulation to both recurrent laryngeal nerves. A second (WPI 301-T) stimulator was used to provide a low level of constant current stimulus for the superior laryngeal nerves. Voltages ranged from 0.5 to 0.9 volts for the Grass stimulator. Currents ranged from 0.1 to 0.15 mA for the WPI stimulator. Frequency of stimulus was 80 Hz, with a pulse duration of 1.5 msec for both the Grass and WPI units. M e d i a l a d d u c t o r y c o m p r e s s i o n . Medial adductory compression (MAC) of the vocal folds was mea sured by a device consisting of a small soft polyethylene (1 x 3 x 3 mm) H 0-filled sac connected to a 2-mm diameter catheter. The device was calibrated against a mercury manometer with a pediatric blood pressure transducer for pressure levels from 10 to 90 mm Hg. The device was held between the arytenoids and vo cal processes, with the catheter arising from the sub-
Downloaded from oto.sagepub.com at SAGE Publications on June 21, 2016
2
214
OTOLARYNGOLOGYHEAD AND NECK SURGERY
BERKE et al.
I N T E R A C T I V E C O M P U T E R I Z E D G L O T T A L E V E N T IDENTIFICATION MAXIMAL OPENING O P E t >J I N G
PGG
dEGG
/
C L O SURE
\
\
Y
I
Fig. 2. O p e n quotient w a s c a l c u l a t e d f r o m simultaneously o b t a i n e d p h o t o g l o t t o g r a p h y a n d t h e first derivative of e l e c t r o g l o t t o g r a p h y . T h e points of o p e n i n g a n d closing w e r e m a r k e d a t r e s p e c t i v e p e a k s in increasing a n d d e c r e a s i n g velocity of t h e E G G i m p e d a n c e signal. Peak glottic a r e a w a s estimated f r o m t h e p e a k v o l t a g e of t h e photosensor signal. O p e n quotient w a s t h e n c a l c u l a t e d as t h e p e r c e n t a g e of t h e glottal c y c l e during w h i c h t h e v o c a l folds w e r e o p e n ,
glottic tracheostomy. A measurement was taken before and after each trial to ascertain a relative gauge of MAC. Photoelectric, Intensity, a n d pressure m e a s u r e ments. A photosensor (Centronics OSD 50-2, Moun tainside, N.J.) was placed on the animal's trachea ap proximately 3 cm below the larynx. A halogen flashlight provided supraglottic illumination for photoglottogra phy (PGG). A microphone (Sennheiser, Old Lyme, Conn.) was placed 15 cm from the vocal folds and connected to a Storz model 8000 laryngostroboscope (Storz, Culver City, Calif.) for frequency analysis of the induced phonation. The stroboscope source was connected with a fiberoptic cable to a 0-degree Storz telescope for observation of vocal fold vibratory ex cursion. Electroglottography (EGG) electrodes (Synchrovoice, Briarcliff Manor, N.Y.) were placed in direct contact on either side of the thyroid cartilage while the reference electrode was sutured to the skin. A catheter-tipped pressure transducer (Millar model # S P C 330, Houston, Texas), inserted through the upper tracheotomy, rested 2 cm below the glottis. The trans ducer was calibrated at the temperature of the animal's trachea by submerging the transducer in a water bath at 37° C to a depth just covering the sensor (0.5 cm) and then calibrating it against a manometer from 0 to 120 cm H 0 pressure. Intensity was measured with a linear scale sound level meter (Quest Electronics Model # 2 0 8 L , Oconomowoc, 2
Wis.) positioned 1 m from the anterior canine. Rotation of the sound level meter at constant radius, 1 m from the animal's mouth, showed less than a 2-dB fluctuation in intensity, indicating isotropic sound radiation. PGG, EGG, Psub, and MAC signals were digitized at 20 kHz using a 12-bit A / D board and a 16-bit per sonal computer microprocessor. The signals were mon itored on oscilloscopes (Tektronix 5116, Beavetton, Ore.) and (Hitachi V1050-F, Carson, Calif.). Files were stored on disk, and a 0.5-second sample of stable pho nation was used for analysis. A multipurpose computer software program was used to choose points of opening and closing by using the differentiated EGG. " Peak opening was chosen using the peak of the PGG (Fig. 2). Twenty-five consecutive cycles were used to cal culate a mean open quotient for each trial. 1
E x p e r i m e n t a l d e s i g n . The first part of this study compared the effects of airflow vs. laryngeal nerve stimulation on intensity. Four related experiments were performed on five animals. 1. Subglottal air flow was provided at a constant 318 cc/second. Superior laryngeal nerve stimu lation was set to sustain half maximal contraction of the cricothyroid muscle at 0.1 mA. Voltage was increased to the recurrent laryngeal nerve until phonation ensued and then voltage was in creased in steps to result in change in F of 20 Hz per increment. Intensity was measured at each F„ increment (Fig. 3).
Downloaded from oto.sagepub.com at SAGE Publications on June 21, 2016
0
V o l u m e 102 M a r c h 1990
Number 3
V o c a l efficacy a n d glottal vibration
215
100
500
ι 240
1
1
1
Γ
300
360
420
480
540
AIR F L O W (cc/sec)
AIR F L O W ( m l / s e c ) F,
8 - 3. Intensity m e a s u r e d in dB SPL Is plotted against rate of subglottal airflow as f u n d a m e n t a l f r e q u e n c y w a s v a r i e d b y c h a n g e In v o l t a g e of stimulation of t h e recurrent l a r y n g e a l nerves ( n = 2).
Fig. 5. At a constant m e d i a l a d d u c t o r y pressure, intensity in dB SPL Is plotted against three levels of subglottal air flow (n = 1).
H
105
100-
Ο = How 150cc/sec
ω 90
t
ao-i
Ο
95 /
EN
ζ LU
Ο
Δ 90'
•
85 §
/
>-
ω T3
• = flow 3 0 0 c c / s e c Δ = How 4 5 0 c c / s e c
Η
6080
AIR F L O W = 3 1 8 c c / s e c
5 0 H
4080
100
I 120
-Γ"
—r—
140
160
ι— 180
Ν β - 4. Intensity In dB SPL Is plotted against F at a constant airflow of 318 c c / s e c o n d ; m = slope, r = Pearson's correlation, " = 1. D
2. Stimulus voltage to the recurrent laryngeal nerves was held constant at 0.5 volts, and stimulation to the superior laryngeal nerves was held at a con stant current of 0.1 mA. Subglottal airflow was increased by increments of 60 cc/second from 120 to 480 cc/second, and the F and intensity of the induced phonation were measured (Fig. 4). 3. Stimulus voltage to the recurrent laryngeal nerves was varied in 0.1-volt increments, and MAC and intensity of phonation were measured. This ex periment was performed at three different rates of subglottal flow (150, 300, and 450 cc/second). It was observed that equal increments of increas ing voltage did not correspond to proportional increases in MAC. This was the result of imped ance fluctuation from electrode-nerve connection or nerve-muscular fatigue. For this experiment,
55
60
65
70
75
80
85
90
95
100
MEDIAL ADDUCTORY COMPRESSION (cm/Η,Ο)
I 200
FREQUENCY (Hz)
0
50
F i g . 6. For t h r e e different flow rates, Intensity in dB SPL Is plotted against m e a n a d d u c t i o n pressure m e a s u r e d b e t w e e n the v o c a l folds [n = 1).
f>=tlow
300cc'sec
O=flow
150cc/sec
20001800· ^ LU
A = 180cc/sec
1600-1
(J
1400-J
LU LU
1200 H
_!
IOOOH
8 >
Ο
Δ =250cc/sec 80
°1
Δ = 31β00/3«0
6
600-1
388cc/sec
τ—\—ι—ι—i^—ι——ι—\—i^—Γ
400 40
45
50
55
60
65
70
75
80
85
90
95
100
MEDIAL ADDUCTORY COMPRESSION (cm/Η,Ο) Fig. 7. T h e open triangles d e m o n s t r a t e v o c a l efficiency for four rates of subglottic flow (180, 250, 318, a n d 388) at a 59.5 c m H 0 constant m e d i a l a d d u c t o r y pressure, T h e o p e n circles a n d closed circles represent v o c a l efficiency, with vari ation in m e d i a l a d d u c t o r y pressure at subglottic flow rates of 150 a n d 300 c c / s e c o n d , respectively ( n = 2). 2
Downloaded from oto.sagepub.com at SAGE Publications on June 21, 2016
216
OtolaryngologyH e a d a n d N e c k Surgery
BERKE et al
dEGG J " * - 4 *
AIR F L O W (ml/sec)
308
175
442
577
Flg. β. Representative samples of signals, representing I m p e d a n c e ( E G G , with increasing i m p e d a n c e up), light transillumination ( P G G , Increasing light up), subglottic pressure (PRS), a n d t h e first derivative of t h e E G G signal a r e d e m o n s t r a t e d for four subglottal flow rates. T h e duration of t h e o p e n p e r i o d i n c r e a s e d with rate of flow,
phonation at 450 cc/second was elicited solely at an MAC of 59.5 cm H 0 (Figs. 5, 6, and 7). 4. MAC was maintained constant at 59.5 cm H 0 by stimulation of the recurrent laryngeal nerves at 0.7 volts. Subglottal airflow was varied (180, 250, 318, and450 cc/second) while intensity was measured (Fig. 7). Vocal efficiency, the ratio of the radiated intensity of production to the subglottic power, was calculated from experiments 3 and 4 using the following equation: 2
2
2
(I x R x 4 pi)/(U x Psub)
in which I = intensity (dB), R = radius (cm), U = mean flow (cc/sec), and Psub = mean subglottic pressure (cm H 0 ) . The second part of this study examined the effect of airflow on open quotient. MAC was held constant at 59.5 cm H 0 . This was done at a constant superior laryngeal nerve stimulation of 0.1 mA and recurrent laryngeal nerve stimulation of 0.4 to 0.8 volts. Airflow was varied in four levels (175, 308, 442, and 577 cc/second). Seven animals were studied. Not all prep arations produced phonation at 175 cc/second, so anal 2
2
ysis was limited to the three flow rates of 308, 442, and 577 cc/second.
RESULTS Fig. 3 shows the experimental effect of varying air flow at a constant level of RLNS. There was little change in I or F for increasing levels of flow. In con trast, Fig. 4 demonstrates that at a constant level of U (318 cc/second), increasing F„ by greater RLNS pro duced approximately a 25-dB increase in I. A similar comparison is seen in Figs. 5 and 6. Fig. 5 demonstrates the effect of increasing flow while monitoring constant medial adductory compression at 59.5 cm H 0 . As U was increased from 150 to 450 cc/second, I increased by only 8 dB. Fig. 6 shows the effect of increasing MAC at three different rates of air flow (150, 300, and 450 cc/second). MAC showed a profound influence on the intensity of production during constant flow. Fig. 7 shows that at a constant level of MAC (59.5 cm H 0 ) , increasing levels of U (180, 250, 318, and 388 cc/second) were associated with a decrease in the efficiency of vocal production (triangles). When flow was held constant at 150 cc/second (open circles) u
2
2
Downloaded from oto.sagepub.com at SAGE Publications on June 21, 2016
V o l u m e 102 M a r c h 1990
Number 3
V o c a l efficacy a n d glottal vibration
or 300 cc/second (closedcircles), increasing MAC had relatively less effect on vocal efficiency. These data indicated that producing intensity by increasing medial adductory compression was more efficient than by in creasing airflow. Fig. 8 shows signals from simultaneously obtained EGG, PGG, subglottic pressure, and differentiated EGG in a typical experiment for four levels of airflow at constant MAC. It was observed that as airflow in creased, there was a longer glottal open period. Sub glottic pressure changed relatively little as flow was increased. Fig. 9 shows mean data for seven experiments. Open quotient significantly increased with greater airflow (ANOVA for repeated measures within subjects, P < 0 . 0 0 l ; F [ l , l 2 ] = 30.84). Post hoc NeumanKeuls testing indicated that all levels were significantly different. Stroboscopic examination of the glottis at constant MAC demonstrated an increase in the lateral excursion of the vocal folds with increasing flow.
DISCUSSION Little if any increase in subglottic pressure was ob served when flow was increased during constant MAC. In contrast, greater RLNS significantly increases sub glottic pressure." Open quotient increased with greater airflow in this study, whereas OQ decreased with greater R L N S . " Results of this study indicate that the phonating lar ynx may be analogous to a small balloon filling with air each cycle until it bursts, leading to resealing and refilling again. Once the stiffness of the walls of the balloon are overcome and it begins to fill, the pressure within the balloon stays constant and is independent of the rate of airflow used to inflate it. Furthering this analogy, it appears that greater airflow ( U ) can inflate the balloon to a larger circumference (OQ) without changing the intraballoon pressure (Psub). Fant' has pointed out that the primary factor involved in determining voice intensity is the lung or subglottic Pressure. This occurs because of the increase in the mean particle velocity produced by the transference of Potential energy from the elevated subglottic pressure to the kinetic energy of the molecules in the supraglottic airflow jet. Two additional covarying factors include increase in the velocity of vocal fold closing caused by increased medial adductory compression and an in crease in the fundamental frequency or pulse repetition. These intensity-related factors are a function of laryn geal nerve stimulation, but not of airflow. The living innervated animal model of phonation used for this study appears to be closer to the physiology of normal human phonation than would be expected of 2
217
O P E N Q U O T I E N T A S A F U N C T I O N O F AIR F L O W 0.5η
1
•
0.4-
Z
0.2H 0.H
η
308
1
1
448
577
AIR F L O W (ml/sec) F i g . 9. T h e open a n d line bars represent m e a n d a t a a n d s t a n d a r d deviation, respectively, for variation in o p e n quotient for t h r e e m a g n i t u d e s of subglottal airflow for s e v e n subjects.
excised laryngés or other models with fixed mechanical properties of the aperture walls. During phonation in duced in these experiments, the resistance of the vocal folds varied, even though the measured medial adduc tion compression was kept constant. The pressure mea sured as MAC probably resulted from a number of forces, including adduction of the vocal processes, in trinsic muscular contraction of the thyroarytenoid, and linear tensile stretch of the vocal fold tissues. It seems, however, that MAC represents a reasonable indirect measure of muscular effort that results in resistance t o flow of air through the glottis." As flow increased at an otherwise steady state of medial adductory compres sion, subglottic pressure was observed to remain con stant. This necessarily implies a decrease in the dy namic resistance of the vocal fold walls in response to greater airflow. Stroboscopy of the vocal fold move ments indicated that the relative stability of subglottic pressure as subglottic flow increased was associated with greater lateral movement of the vocal folds in each cycle and a greater cross-sectional area for air escape in each cycle (resulting in greater measured open quo tient). Thus, the resistance of the vocal folds is not linear across various levels of airflow, and the assump tion of glottal resistance as the quotient of Psub and U may consequently be incorrect. It appears that the me chanical properties of the folds changed in response to the effect of greater flow, without comparable change in the muscular effort indicated by medial adductory compression. The concept of a flow-controlled nonlin ear resistance has been proposed for a number of fluid
Downloaded from oto.sagepub.com at SAGE Publications on June 21, 2016
21β
OtolaryngologyH e a d a n d N e c k Surgery
BERKE et al.
mechanical systems, such as collapsible t u b e s .
14
It has
been suggested that vocal fold oscillation may dem onstrate a specific subset of collapsible tube behavior.
15
Data from these experiments would appear to support that theory. Further exploration of the nature of flowcontrolled
resistance
in the larynx
should be
en
couraged.
CONCLUSIONS 1. During
constant
medial
adductory
compression
of the vocal folds, increasing airflow from 100 to 500 cc/second increased intensity by only 5 to 10 d B . 2. During constant airflow, increasing medial adduc tory compression from 60 to 90 cm H 0 increased 2
intensity by 30 d B . 3. Vocal efficiency markedly decreased for increasing airflow and remained essentially unchanged for in creasing levels of medial adductory compression. 4. Increasing airflow during constant medial adductory compression produced a significant increase in the open quotient and a larger vocal fold vibratory ex cursion. 5. The dynamic resistance of the larynx is controlled by the flow during phonation and is not constant, even for constant medial adductory compression. REFERENCES 1. Van den Berg JW. Direct and indirect determination of the mean subglottic pressure. Folia Phoniatr (Basel) 1956;8:1-24. 2. Rubin HJ, LeCover M, Vennard W. Vocal intensity, subglottic pressure and air flow relationships in singers. Folia Phoniatr (Basel) 1967;19:393-413. 3. Isshiki N. Regulatory mechanism of voice intensity variation. J Speech Hear Res 1964;7:17-29.
4. Koyama T, Kawasaki M, Ogura J. Mechanics of voice produc tion: regulation of vocal intensity. Laryngoscope 1969;79:33754. 5. Rubin HJ. Experimental studies on vocal pitch and intensity on phonation. Laryngoscope 1963;73:973-1015. 6. Timcke R, von Leden H, Moore P. Laryngeal vibrations: mea surements of the glottic wave. Part I the normal vibratory cycle. Arch Otolaryngol 1958;68:1-19. 7. Netsell R, Shaughnessy AL, Lotz WK. Laryngeal aerodynamics for selected vocal pathologies. Presented to the Association for Research in Otolaryngology. St. Petersburg Beach, Fla., January 1983. 8. Smitheran JR, Hixon TJ. A clinical method for estimating la ryngeal airway resistance during vowel production. J Speech Hear Disord 1981;46:138-46. 9. Berke GS, Moore DM, Hanson DG, Hantke DR, Gerratt BR, Burstein F. Laryngeal modeling: theoretical, in vitro, in vivo. Laryngoscope 1987;97(7):871-81. 10. Childers D, Naik J, Krishnamurthy A, et al. Electroglottography, speech, and ultra-high-speed cinematography. In: Titze IR, Scherer RC, eds. Vocal Fold Physiology. Denver, Colo: Denver Center for the Performing Arts, 1983:202-21. 11. Moore DM, Berke GS. The effect of laryngeal nerve stimulation on phonation: a glottographic study using an in vivo canine model. J Acoust Soc Am 1988;83(2):705-15. 12. Fant G. Speech production: preliminaries to analysis of the hu man voice source. Quarterly Progress and Status Report Stock holm, Sweden: Speech Transmission Laboratory, Royal Institute of Technology (KTH), STL-QPSR 1982;4:1-27. 13. Scherer RC, Cooper DS, Alipour-Haghighi F, Titze IR. Vocal process contact pressures. Van Lawrence, ed. Transactions of the twelfth symposium: care of the professional voice. Presented .at New York: Lincoln Center, The Juilliard School, June 6-10, 1983. 14. Conrad WA. Pressure-flow relationships in collapsible tubes. IEEE Trans Biomed Eng 1969;16(4):284-95. 15. Shapiro AH. Steady flow in collapsible tubes. J Biomech Eng 1977;99:126-47.
Downloaded from oto.sagepub.com at SAGE Publications on June 21, 2016
This s t u d y u s e d a n In v i v o c a n i n e m o d e l t o Investigate t h e effects of v a r y i n g v o c a l f o l d resistance b y e l e c t r i c a l l y stimulating t h e recurrent l a r y n g e a l n e r v e w h i l e monitoring m e d i a l a d d u c t o r y c o m p r e s s i o n of t h e v o c a l folds, glottal airflow, a n d v o c a l intensity. T h e effects o f I n c r e a s i n g a i r f l o w o n glottal v i b r a t i o n w e r e a l s o e x a m i n e d s t r o b o s c o p i c a l l y a n d b y m e a s u r e m e n t of o p e n quotient. T h e results I n d i c a t e d that Increasing intensity b y m e d i a l a d d u c t o r y c o m p r e s s i o n w a s m o r e efficient t h a n b y Increasing airflow. Increasing airflow p r o d u c e d a significantly g r e a t e r o p e n quotient a n d v o c a l fold v i b r a t o r y e x c u r s i o n . (OTOLARYNGOL HEAD NECK S U R G 1 9 9 0 ; 1 0 2 : 2 1 2 . )
The study of the factors affecting vocal intensity is complicated by difficulty in measuring all the variables of interest. Most often, investigators have measured subglottal pressure (Psub), glottal airflow (U), and vo cal intensity (I). Vocal fold resistance has not usually been measured directly, but rather has been calculated from these other measures. This study, using an in vivo canine model, investigated the effects of varying vocal fold resistance by electrical stimulation of the recurrent laryngeal nerve (RLNS) while monitoring medial ad ductory compression (MAC) of the vocal folds, glottal airflow, and vocal intensity.
BACKGROUND 1
Van den Berg was one of the first investigators to emphasize the importance of subglottic power in vocal intensity, which he estimated as the product of mean subglottic pressure and mean flow rate. Rubin et a l . used a tracheal puncture technique to study the effect of Psub and U on intensity. They concluded that sub glottic pressure had a much greater effect on intensity 2
From the UCLA School of Medicine, Division of Head and Neck Surgery and VA Medical Center. Supported by Veterans Administration Medical Research Funds and NIH Grant NS 20707-05. Presented at the Annual Meeting of the American Academy of Otolaryngology-Head and Neck Surgery, Washington, D.C., Sept. 25-29, 1988. Submitted for publication Dec. 5, 1988; accepted May 15, 1989. Reprint requests: Gerald S. Berke, MD, UCLA School of Medicine, Division of Head and Neck Surgery, CHS-62- ! 39, 10833 Le Conte Ave.. Los Angeles, CA 90024-1624. 23/1/13905
production than airflow, which had little or no effect. Isshiki reported that resistance or subglottic pressure had the greatest effect on intensity at low fundamental frequencies (F s), whereas at high frequencies airflow had the greater effect. Koyama et al. , using an in vivo canine model, concluded that airflow was more im portant than RLNS in controlling the intensity of voice production. They assumed that RLNS was directly pro portional to laryngeal stiffness. Other authors have not concurred with their conclusions. Timcke et al. reviewed the relationship of open quo tient (OQ) to intensity of voice production. Using high speed cinematography, they found that as open quotient (i.e., the proportion of time the glottis was open within each glottal cycle) decreased, vocal intensity usually increased. The above studies and others appear to have as sumed that resistance of the vocal folds is proportional to vocal fold stiffness, implying a linear relationship of resistance to airflow. One aim of this study is to reex amine the validity of this assumption. Previous in vivo animal studies have demonstrated that, under conditions of steady subglottic flow, in crease in electrical stimulation to the recurrent laryngeal nerves was associated with induced phonation of a higher pitch and greater loudness. This study exam ined effects of variation in airflow and electrical stim ulation of the laryngeal nerves on intensity and open quotient. 3
0
4
2 3 5
6
711
59
METHODS In v i v o p r e p a r a t i o n . Mongrel dogs were anesthe tized with an intramuscular injection of 2 ml ketamine, followed by intravenous pentobarbital titrated to loss
212
Downloaded from oto.sagepub.com at SAGE Publications on June 21, 2016
02 Number3
tev
9 9 0
V o c a l efficacy a n d glottal vibration
CAMERA CONTINUOUS XENON -*»•' LIGHT SOURCE
Video or 35 m m S L R
STROBOSCOPE
213
Subsequent Data Analysis
(strobe trigger)
4 CHANNEL FM TAPE RECORDER
Sup. l a r y n g e a l π
- (dEGG) NERVE STIMULATOR' (bilateral)
-(EGG)—I
NERVE STIMULATOR (bilateral)
-(PGG)-
PRESSURE TRANSDUCER SPONTANEOUS RESPIRATIONS
MULTICHANNEL STORAGE OSCILLOSCOPE
- (press.)-)
R e c u r r e n t n. X-Y
PLOTTER
Fig. 1. A n in vivo c a n i n e p r e p a r a t i o n a l l o w e d simultaneous monitoring of v i d e o stroboscopy, p h o t o g l o t t o g r a p h y , e l e c t r o g l o t t o g r a p h y , a n d subglottic pressure a n d airflow.
°f the corneal reflex. The animals were placed in supine Position on an operating table (Fig. 1) and direct lar yngoscopy was performed to confirm normal laryngeal anatomy. A 7-mm oral endotracheal tube was inserted, through which the animal breathed spontaneously. Through a vertical midline incision, the strap muscles and sternocleidomastoid muscles were retracted later ally to expose the larynx and trachea. The external branches of the superior laryngeal nerves were isolated at their entrance into the cricothyroid muscle. Harvard bipolar electrodes were applied to the nerves. The re current laryngeal nerves were isolated 5 cm inferior to the larynx, and bipolar electrodes were applied. A sue through the thyrohyoid membrane was used to sus pend the epiglottis anteriorly to improve visualization °f the vocal folds. A distal tracheotomy was made for Placement of an endotracheal tube to permit the animal to breathe spontaneously. An additional proximal tra cheotomy was performed through which a cuffed tra cheotomy tube was placed, with its tip resting 10 cm below the glottis. The cuff of the superiorly directed tube was inflated to just seal the trachea. Room air was bubbled through 5 cm Η,Ο at 37° C for warming and t U r
humidification and passed through the cephalad trache otomy tube. The temperature in the animal's trachea was measured at 15-minute intervals to assure a con stant air temperature of 37° C. A Grass mode) 54H stimulator (Grass Instruments, Quincy, Mass.) provided variable voltage stimulation to both recurrent laryngeal nerves. A second (WPI 301-T) stimulator was used to provide a low level of constant current stimulus for the superior laryngeal nerves. Voltages ranged from 0.5 to 0.9 volts for the Grass stimulator. Currents ranged from 0.1 to 0.15 mA for the WPI stimulator. Frequency of stimulus was 80 Hz, with a pulse duration of 1.5 msec for both the Grass and WPI units. M e d i a l a d d u c t o r y c o m p r e s s i o n . Medial adductory compression (MAC) of the vocal folds was mea sured by a device consisting of a small soft polyethylene (1 x 3 x 3 mm) H 0-filled sac connected to a 2-mm diameter catheter. The device was calibrated against a mercury manometer with a pediatric blood pressure transducer for pressure levels from 10 to 90 mm Hg. The device was held between the arytenoids and vo cal processes, with the catheter arising from the sub-
Downloaded from oto.sagepub.com at SAGE Publications on June 21, 2016
2
214
OTOLARYNGOLOGYHEAD AND NECK SURGERY
BERKE et al.
I N T E R A C T I V E C O M P U T E R I Z E D G L O T T A L E V E N T IDENTIFICATION MAXIMAL OPENING O P E t >J I N G
PGG
dEGG
/
C L O SURE
\
\
Y
I
Fig. 2. O p e n quotient w a s c a l c u l a t e d f r o m simultaneously o b t a i n e d p h o t o g l o t t o g r a p h y a n d t h e first derivative of e l e c t r o g l o t t o g r a p h y . T h e points of o p e n i n g a n d closing w e r e m a r k e d a t r e s p e c t i v e p e a k s in increasing a n d d e c r e a s i n g velocity of t h e E G G i m p e d a n c e signal. Peak glottic a r e a w a s estimated f r o m t h e p e a k v o l t a g e of t h e photosensor signal. O p e n quotient w a s t h e n c a l c u l a t e d as t h e p e r c e n t a g e of t h e glottal c y c l e during w h i c h t h e v o c a l folds w e r e o p e n ,
glottic tracheostomy. A measurement was taken before and after each trial to ascertain a relative gauge of MAC. Photoelectric, Intensity, a n d pressure m e a s u r e ments. A photosensor (Centronics OSD 50-2, Moun tainside, N.J.) was placed on the animal's trachea ap proximately 3 cm below the larynx. A halogen flashlight provided supraglottic illumination for photoglottogra phy (PGG). A microphone (Sennheiser, Old Lyme, Conn.) was placed 15 cm from the vocal folds and connected to a Storz model 8000 laryngostroboscope (Storz, Culver City, Calif.) for frequency analysis of the induced phonation. The stroboscope source was connected with a fiberoptic cable to a 0-degree Storz telescope for observation of vocal fold vibratory ex cursion. Electroglottography (EGG) electrodes (Synchrovoice, Briarcliff Manor, N.Y.) were placed in direct contact on either side of the thyroid cartilage while the reference electrode was sutured to the skin. A catheter-tipped pressure transducer (Millar model # S P C 330, Houston, Texas), inserted through the upper tracheotomy, rested 2 cm below the glottis. The trans ducer was calibrated at the temperature of the animal's trachea by submerging the transducer in a water bath at 37° C to a depth just covering the sensor (0.5 cm) and then calibrating it against a manometer from 0 to 120 cm H 0 pressure. Intensity was measured with a linear scale sound level meter (Quest Electronics Model # 2 0 8 L , Oconomowoc, 2
Wis.) positioned 1 m from the anterior canine. Rotation of the sound level meter at constant radius, 1 m from the animal's mouth, showed less than a 2-dB fluctuation in intensity, indicating isotropic sound radiation. PGG, EGG, Psub, and MAC signals were digitized at 20 kHz using a 12-bit A / D board and a 16-bit per sonal computer microprocessor. The signals were mon itored on oscilloscopes (Tektronix 5116, Beavetton, Ore.) and (Hitachi V1050-F, Carson, Calif.). Files were stored on disk, and a 0.5-second sample of stable pho nation was used for analysis. A multipurpose computer software program was used to choose points of opening and closing by using the differentiated EGG. " Peak opening was chosen using the peak of the PGG (Fig. 2). Twenty-five consecutive cycles were used to cal culate a mean open quotient for each trial. 1
E x p e r i m e n t a l d e s i g n . The first part of this study compared the effects of airflow vs. laryngeal nerve stimulation on intensity. Four related experiments were performed on five animals. 1. Subglottal air flow was provided at a constant 318 cc/second. Superior laryngeal nerve stimu lation was set to sustain half maximal contraction of the cricothyroid muscle at 0.1 mA. Voltage was increased to the recurrent laryngeal nerve until phonation ensued and then voltage was in creased in steps to result in change in F of 20 Hz per increment. Intensity was measured at each F„ increment (Fig. 3).
Downloaded from oto.sagepub.com at SAGE Publications on June 21, 2016
0
V o l u m e 102 M a r c h 1990
Number 3
V o c a l efficacy a n d glottal vibration
215
100
500
ι 240
1
1
1
Γ
300
360
420
480
540
AIR F L O W (cc/sec)
AIR F L O W ( m l / s e c ) F,
8 - 3. Intensity m e a s u r e d in dB SPL Is plotted against rate of subglottal airflow as f u n d a m e n t a l f r e q u e n c y w a s v a r i e d b y c h a n g e In v o l t a g e of stimulation of t h e recurrent l a r y n g e a l nerves ( n = 2).
Fig. 5. At a constant m e d i a l a d d u c t o r y pressure, intensity in dB SPL Is plotted against three levels of subglottal air flow (n = 1).
H
105
100-
Ο = How 150cc/sec
ω 90
t
ao-i
Ο
95 /
EN
ζ LU
Ο
Δ 90'
•
85 §
/
>-
ω T3
• = flow 3 0 0 c c / s e c Δ = How 4 5 0 c c / s e c
Η
6080
AIR F L O W = 3 1 8 c c / s e c
5 0 H
4080
100
I 120
-Γ"
—r—
140
160
ι— 180
Ν β - 4. Intensity In dB SPL Is plotted against F at a constant airflow of 318 c c / s e c o n d ; m = slope, r = Pearson's correlation, " = 1. D
2. Stimulus voltage to the recurrent laryngeal nerves was held constant at 0.5 volts, and stimulation to the superior laryngeal nerves was held at a con stant current of 0.1 mA. Subglottal airflow was increased by increments of 60 cc/second from 120 to 480 cc/second, and the F and intensity of the induced phonation were measured (Fig. 4). 3. Stimulus voltage to the recurrent laryngeal nerves was varied in 0.1-volt increments, and MAC and intensity of phonation were measured. This ex periment was performed at three different rates of subglottal flow (150, 300, and 450 cc/second). It was observed that equal increments of increas ing voltage did not correspond to proportional increases in MAC. This was the result of imped ance fluctuation from electrode-nerve connection or nerve-muscular fatigue. For this experiment,
55
60
65
70
75
80
85
90
95
100
MEDIAL ADDUCTORY COMPRESSION (cm/Η,Ο)
I 200
FREQUENCY (Hz)
0
50
F i g . 6. For t h r e e different flow rates, Intensity in dB SPL Is plotted against m e a n a d d u c t i o n pressure m e a s u r e d b e t w e e n the v o c a l folds [n = 1).
f>=tlow
300cc'sec
O=flow
150cc/sec
20001800· ^ LU
A = 180cc/sec
1600-1
(J
1400-J
LU LU
1200 H
_!
IOOOH
8 >
Ο
Δ =250cc/sec 80
°1
Δ = 31β00/3«0
6
600-1
388cc/sec
τ—\—ι—ι—i^—ι——ι—\—i^—Γ
400 40
45
50
55
60
65
70
75
80
85
90
95
100
MEDIAL ADDUCTORY COMPRESSION (cm/Η,Ο) Fig. 7. T h e open triangles d e m o n s t r a t e v o c a l efficiency for four rates of subglottic flow (180, 250, 318, a n d 388) at a 59.5 c m H 0 constant m e d i a l a d d u c t o r y pressure, T h e o p e n circles a n d closed circles represent v o c a l efficiency, with vari ation in m e d i a l a d d u c t o r y pressure at subglottic flow rates of 150 a n d 300 c c / s e c o n d , respectively ( n = 2). 2
Downloaded from oto.sagepub.com at SAGE Publications on June 21, 2016
216
OtolaryngologyH e a d a n d N e c k Surgery
BERKE et al
dEGG J " * - 4 *
AIR F L O W (ml/sec)
308
175
442
577
Flg. β. Representative samples of signals, representing I m p e d a n c e ( E G G , with increasing i m p e d a n c e up), light transillumination ( P G G , Increasing light up), subglottic pressure (PRS), a n d t h e first derivative of t h e E G G signal a r e d e m o n s t r a t e d for four subglottal flow rates. T h e duration of t h e o p e n p e r i o d i n c r e a s e d with rate of flow,
phonation at 450 cc/second was elicited solely at an MAC of 59.5 cm H 0 (Figs. 5, 6, and 7). 4. MAC was maintained constant at 59.5 cm H 0 by stimulation of the recurrent laryngeal nerves at 0.7 volts. Subglottal airflow was varied (180, 250, 318, and450 cc/second) while intensity was measured (Fig. 7). Vocal efficiency, the ratio of the radiated intensity of production to the subglottic power, was calculated from experiments 3 and 4 using the following equation: 2
2
2
(I x R x 4 pi)/(U x Psub)
in which I = intensity (dB), R = radius (cm), U = mean flow (cc/sec), and Psub = mean subglottic pressure (cm H 0 ) . The second part of this study examined the effect of airflow on open quotient. MAC was held constant at 59.5 cm H 0 . This was done at a constant superior laryngeal nerve stimulation of 0.1 mA and recurrent laryngeal nerve stimulation of 0.4 to 0.8 volts. Airflow was varied in four levels (175, 308, 442, and 577 cc/second). Seven animals were studied. Not all prep arations produced phonation at 175 cc/second, so anal 2
2
ysis was limited to the three flow rates of 308, 442, and 577 cc/second.
RESULTS Fig. 3 shows the experimental effect of varying air flow at a constant level of RLNS. There was little change in I or F for increasing levels of flow. In con trast, Fig. 4 demonstrates that at a constant level of U (318 cc/second), increasing F„ by greater RLNS pro duced approximately a 25-dB increase in I. A similar comparison is seen in Figs. 5 and 6. Fig. 5 demonstrates the effect of increasing flow while monitoring constant medial adductory compression at 59.5 cm H 0 . As U was increased from 150 to 450 cc/second, I increased by only 8 dB. Fig. 6 shows the effect of increasing MAC at three different rates of air flow (150, 300, and 450 cc/second). MAC showed a profound influence on the intensity of production during constant flow. Fig. 7 shows that at a constant level of MAC (59.5 cm H 0 ) , increasing levels of U (180, 250, 318, and 388 cc/second) were associated with a decrease in the efficiency of vocal production (triangles). When flow was held constant at 150 cc/second (open circles) u
2
2
Downloaded from oto.sagepub.com at SAGE Publications on June 21, 2016
V o l u m e 102 M a r c h 1990
Number 3
V o c a l efficacy a n d glottal vibration
or 300 cc/second (closedcircles), increasing MAC had relatively less effect on vocal efficiency. These data indicated that producing intensity by increasing medial adductory compression was more efficient than by in creasing airflow. Fig. 8 shows signals from simultaneously obtained EGG, PGG, subglottic pressure, and differentiated EGG in a typical experiment for four levels of airflow at constant MAC. It was observed that as airflow in creased, there was a longer glottal open period. Sub glottic pressure changed relatively little as flow was increased. Fig. 9 shows mean data for seven experiments. Open quotient significantly increased with greater airflow (ANOVA for repeated measures within subjects, P < 0 . 0 0 l ; F [ l , l 2 ] = 30.84). Post hoc NeumanKeuls testing indicated that all levels were significantly different. Stroboscopic examination of the glottis at constant MAC demonstrated an increase in the lateral excursion of the vocal folds with increasing flow.
DISCUSSION Little if any increase in subglottic pressure was ob served when flow was increased during constant MAC. In contrast, greater RLNS significantly increases sub glottic pressure." Open quotient increased with greater airflow in this study, whereas OQ decreased with greater R L N S . " Results of this study indicate that the phonating lar ynx may be analogous to a small balloon filling with air each cycle until it bursts, leading to resealing and refilling again. Once the stiffness of the walls of the balloon are overcome and it begins to fill, the pressure within the balloon stays constant and is independent of the rate of airflow used to inflate it. Furthering this analogy, it appears that greater airflow ( U ) can inflate the balloon to a larger circumference (OQ) without changing the intraballoon pressure (Psub). Fant' has pointed out that the primary factor involved in determining voice intensity is the lung or subglottic Pressure. This occurs because of the increase in the mean particle velocity produced by the transference of Potential energy from the elevated subglottic pressure to the kinetic energy of the molecules in the supraglottic airflow jet. Two additional covarying factors include increase in the velocity of vocal fold closing caused by increased medial adductory compression and an in crease in the fundamental frequency or pulse repetition. These intensity-related factors are a function of laryn geal nerve stimulation, but not of airflow. The living innervated animal model of phonation used for this study appears to be closer to the physiology of normal human phonation than would be expected of 2
217
O P E N Q U O T I E N T A S A F U N C T I O N O F AIR F L O W 0.5η
1
•
0.4-
Z
0.2H 0.H
η
308
1
1
448
577
AIR F L O W (ml/sec) F i g . 9. T h e open a n d line bars represent m e a n d a t a a n d s t a n d a r d deviation, respectively, for variation in o p e n quotient for t h r e e m a g n i t u d e s of subglottal airflow for s e v e n subjects.
excised laryngés or other models with fixed mechanical properties of the aperture walls. During phonation in duced in these experiments, the resistance of the vocal folds varied, even though the measured medial adduc tion compression was kept constant. The pressure mea sured as MAC probably resulted from a number of forces, including adduction of the vocal processes, in trinsic muscular contraction of the thyroarytenoid, and linear tensile stretch of the vocal fold tissues. It seems, however, that MAC represents a reasonable indirect measure of muscular effort that results in resistance t o flow of air through the glottis." As flow increased at an otherwise steady state of medial adductory compres sion, subglottic pressure was observed to remain con stant. This necessarily implies a decrease in the dy namic resistance of the vocal fold walls in response to greater airflow. Stroboscopy of the vocal fold move ments indicated that the relative stability of subglottic pressure as subglottic flow increased was associated with greater lateral movement of the vocal folds in each cycle and a greater cross-sectional area for air escape in each cycle (resulting in greater measured open quo tient). Thus, the resistance of the vocal folds is not linear across various levels of airflow, and the assump tion of glottal resistance as the quotient of Psub and U may consequently be incorrect. It appears that the me chanical properties of the folds changed in response to the effect of greater flow, without comparable change in the muscular effort indicated by medial adductory compression. The concept of a flow-controlled nonlin ear resistance has been proposed for a number of fluid
Downloaded from oto.sagepub.com at SAGE Publications on June 21, 2016
21β
OtolaryngologyH e a d a n d N e c k Surgery
BERKE et al.
mechanical systems, such as collapsible t u b e s .
14
It has
been suggested that vocal fold oscillation may dem onstrate a specific subset of collapsible tube behavior.
15
Data from these experiments would appear to support that theory. Further exploration of the nature of flowcontrolled
resistance
in the larynx
should be
en
couraged.
CONCLUSIONS 1. During
constant
medial
adductory
compression
of the vocal folds, increasing airflow from 100 to 500 cc/second increased intensity by only 5 to 10 d B . 2. During constant airflow, increasing medial adduc tory compression from 60 to 90 cm H 0 increased 2
intensity by 30 d B . 3. Vocal efficiency markedly decreased for increasing airflow and remained essentially unchanged for in creasing levels of medial adductory compression. 4. Increasing airflow during constant medial adductory compression produced a significant increase in the open quotient and a larger vocal fold vibratory ex cursion. 5. The dynamic resistance of the larynx is controlled by the flow during phonation and is not constant, even for constant medial adductory compression. REFERENCES 1. Van den Berg JW. Direct and indirect determination of the mean subglottic pressure. Folia Phoniatr (Basel) 1956;8:1-24. 2. Rubin HJ, LeCover M, Vennard W. Vocal intensity, subglottic pressure and air flow relationships in singers. Folia Phoniatr (Basel) 1967;19:393-413. 3. Isshiki N. Regulatory mechanism of voice intensity variation. J Speech Hear Res 1964;7:17-29.
4. Koyama T, Kawasaki M, Ogura J. Mechanics of voice produc tion: regulation of vocal intensity. Laryngoscope 1969;79:33754. 5. Rubin HJ. Experimental studies on vocal pitch and intensity on phonation. Laryngoscope 1963;73:973-1015. 6. Timcke R, von Leden H, Moore P. Laryngeal vibrations: mea surements of the glottic wave. Part I the normal vibratory cycle. Arch Otolaryngol 1958;68:1-19. 7. Netsell R, Shaughnessy AL, Lotz WK. Laryngeal aerodynamics for selected vocal pathologies. Presented to the Association for Research in Otolaryngology. St. Petersburg Beach, Fla., January 1983. 8. Smitheran JR, Hixon TJ. A clinical method for estimating la ryngeal airway resistance during vowel production. J Speech Hear Disord 1981;46:138-46. 9. Berke GS, Moore DM, Hanson DG, Hantke DR, Gerratt BR, Burstein F. Laryngeal modeling: theoretical, in vitro, in vivo. Laryngoscope 1987;97(7):871-81. 10. Childers D, Naik J, Krishnamurthy A, et al. Electroglottography, speech, and ultra-high-speed cinematography. In: Titze IR, Scherer RC, eds. Vocal Fold Physiology. Denver, Colo: Denver Center for the Performing Arts, 1983:202-21. 11. Moore DM, Berke GS. The effect of laryngeal nerve stimulation on phonation: a glottographic study using an in vivo canine model. J Acoust Soc Am 1988;83(2):705-15. 12. Fant G. Speech production: preliminaries to analysis of the hu man voice source. Quarterly Progress and Status Report Stock holm, Sweden: Speech Transmission Laboratory, Royal Institute of Technology (KTH), STL-QPSR 1982;4:1-27. 13. Scherer RC, Cooper DS, Alipour-Haghighi F, Titze IR. Vocal process contact pressures. Van Lawrence, ed. Transactions of the twelfth symposium: care of the professional voice. Presented .at New York: Lincoln Center, The Juilliard School, June 6-10, 1983. 14. Conrad WA. Pressure-flow relationships in collapsible tubes. IEEE Trans Biomed Eng 1969;16(4):284-95. 15. Shapiro AH. Steady flow in collapsible tubes. J Biomech Eng 1977;99:126-47.
Downloaded from oto.sagepub.com at SAGE Publications on June 21, 2016
