Experimental setup used to obtain high-speed photographs (A) and record stroboscopic images (B) of the lower surface of the vocal fold during phonation. VCR indicates videocassette recorder.
Schematic drawing of the hemilarynx showing the method used to adjust the length of the vocal fold by pulling the anterior cut end of the thyroid cartilage.
Schematic representation of the method used to measure pliability of the vocal fold mucosa. The box indicates a close-up view of the vocal fold mucosa and needle tip.
Graphic representation of pliability (circles) of the mucosa and the estimated range (squares) where the mucosal upheaval occurred in the absence (open symbols) and presence (black symbols) of vocal fold lengthening. Parts A through E represent dogs 1 through 5, respectively.
Frontal section of the larynx of dog 1 (hematoxylin-eosin, original magnification ×4). Arrows indicate the points 2 mm and 4 mm below the free edge.
Schematic illustration of the positional relationships among the lower border of the sparse deep layer of the lamina propria, the point of minimal pliability, and the mucosal upheaval in the absence of vocal fold lengthening. Arrows indicate pliability; arrowhead, location of the mucosal upheaval. Length of each arrow corresponds to magnitude of pliability of the mucosa.
Customize your JAMA Network experience by selecting one or more topics from the list below.
Yumoto E, Kadota Y. Pliability of the Vocal Fold Mucosa in Relation to the Mucosal Upheaval During Phonation. Arch Otolaryngol Head Neck Surg. 1998;124(8):897–902. doi:10.1001/archotol.124.8.897
Copyright 1998 American Medical Association. All Rights Reserved. Applicable FARS/DFARS Restrictions Apply to Government Use.1998
To quantitatively evaluate the effect of vocal fold lengthening on pliability of the mucosa measured along the superior-inferior axis and to examine the relation of the location of mucosal upheaval (MU) during phonation to the changes in pliability pattern of the mucosa when the vocal fold was lengthened.
Investigation of mechanical characteristics of the vocal fold in relation to the MU during phonation.
Five excised canine larynges.
Vibrations with and without vocal fold lengthening were recorded from the tracheal side via high-speed photography or video recording combined with stroboscopic illumination. Tattooed marks on the lower surface of the vocal fold were used to locate the MU. Pliability was defined as the maximal distance elevated in response to a constant focal negative pressure.
Pliability decreased significantly (P=.05) when the vocal fold was lengthened. The point of minimal pliability and MU without vocal fold lengthening were located slightly above the area where the muscular layer approached the epithelial layer. They were located closer to the free edge of the vocal fold when it was lengthened than when it was not. Discrepancy of their locations when the vocal fold was lengthened was suggested.
The MU occurs around the point of minimal pliability when the vocal fold is not lengthened, whereas the MU occurs slightly more laterally than the point of minimal pliability when the vocal fold is lengthened. Although further study is necessary to explain this discrepancy, the presence of the sparse deep layer of the lamina propria seems to be essential in the generation of the mucosal wave.
THERE IS a small mucosal ridge on the lower surface of the vocal fold in canine and human excised larynges during phonation, which is known as the mucosal upheaval (MU).1,2 Since the mucosal traveling wave has been reported to be elicited above the MU,3,4 it is thought to be the lower border of the vibrating structure of the vocal fold or the starting line of the mucosal wave. Further, based on histological examination of the canine larynx used for observation of the MU during phonation, its location in the absence of thyroarytenoid muscle contraction was shown to be slightly closer to the free edge of the vocal fold than the area where the muscular layer approached the epithelial layer. It was reported previously3that the MU shifts medially toward the free edge of the vocal fold when it was lengthened.
According to Hirano,5 the vocal fold consists of 2 functionally different layers: the cover and the body. In the canine vocal fold, the whole thickness of the mucous membrane is regarded as a cover, while the body is made up of the medial part of the thyroarytenoid muscle. During the closed phase of a vibratory cycle, the cover slides up from the underlying body because of the increased subglottic pressure. Subsequently, as air in the subglottic space escapes through the glottis as a result of the increased subglottic pressure, the mucosal wave propagates to the upper lateral surface of the vocal fold. Thus, deformation during vibration occurs mainly in the cover. The lower limit of this sliding of the cover is thought to be the portion where the MU appears.
Previous attempts have been made to measure the Young modulus of different layers of the excised canine vocal fold.6-8On the other hand, using excised human larynges, Haji et al9measured the maximal distance over which the vocal fold mucosa was lifted up by suction at a constant negative pressure. Their method was designed to quantitatively estimate pliability of the mucosa while preserving the influence of layer-by-layer connections on stiffness of the vocal fold. We have modified their method to measure pliability of the vocal fold mucosa as an index of its elasticity.10
The purposes of this study were to quantitatively evaluate the effect of lengthening of the vocal fold on pliability of the mucosa measured along the superior-inferior (vertical) axis of the vocal fold and to examine the relation of the location of the MU during phonation to changes in the pliability pattern of the mucosa when the vocal fold was lengthened. We also histologically examined the structure of the vocal fold in relation to the point of minimal pliability (Pomin).
Five adult mongrel dogs were anesthetized with an intraperitoneal injection of pentobarbital sodium (25 mg/kg). Their larynges looked normal under direct laryngoscopic observation. Local subcutaneous infiltration of 1% lidocaine hydrochloride with 1:100000 epinephrine was applied before the larynx was excised. The larynx was severed between the cricoid cartilage and the first tracheal ring inferiorly. The supraglottic structures, including the false vocal folds, were removed to obtain the conventional supraglottic view of the vocal fold during phonation.
The bilateral vocal processes were sutured together to attain the glottic closure necessary for phonation. The midpoint of the membranous vocal fold on the free edge and 2 points 1.5 to 2.0 mm and 3.0 to 5.0 mm below the free edge, respectively, were marked with India ink using a 27-gauge needle. These marks were aligned along the vertical axis of the vocal fold. The free edge was defined as the most medial portion of the vocal fold without lengthening of the vocal fold when viewed from the oral side. These points served as landmarks on the vocal fold during phonation to evaluate whether the MU, which occurs only when the vocal fold vibrates, is located medially or laterally compared with each reference point. The larynx was fixed in the hole of a glass box with a clay compound so that air could pass only through the glottal opening during phonation. Moisture-saturated warm air was supplied to the glass box. The mean airflow rate was regulated with a valve and measured with a flowmeter (FS-300; Kawashige Safety Service Industries Ltd, Tokyo, Japan).
The lower surfaces of the vibrating vocal fold of the larynges from dogs 1 and 2 were filmed with a high-speed motion camera (Hymac; Hitachi Ltd, Tokyo) at about 5000 frames per second (Figure 1, A). Generated sound was recorded from the beginning of phonation until several seconds after the end of filming. Because of background noise generated by high-speed filming, the sound recorded just after filming was used to measure the fundamental frequency (F0). Stroboscopy (Type 4914; Bruel & Kaer, Copenhagen, Denmark) via a rigid endoscope (SFT-II; Nagashima, Tokyo) was used in the larynges of dogs 3 through 5. The charge coupled device camera (WV-KS152; Matsushita, Yokohama, Japan) image was recorded onto a videotape (VO5850; Sony, Tokyo) (Figure 1, B). The generated sound was recorded to an audio channel of the videotape simultaneously and then fed into the stroboscope via an external trigger junction to synchronize flashing.
Cricothyroid approximation11 was performed on each larynx. This procedure lengthened the vocal fold compared with its original length and thereby increased longitudinal tension. The length of the vocal fold, defined as the length of its membranous portion viewed from the oral side, was measured with calipers before and after cricothyroid approximation. Then, the infraglottic aspect of vocal fold vibration was recorded as described above. The mean airflow rates in each larynx were kept constant under these different conditions if stable sound was produced. The vocal fold was misted with physiological isotonic sodium chloride before each blowing to prevent drying of the vocal fold mucosa. The F0 of each phonation was measured by the method described previously.3
Each film was repeatedly projected (Filmosound 1658; Bell & Howell, Tokyo) at normal speed (24 frames per second) to evaluate the location of the MU in relation to the reference points. Videotapes were also viewed repeatedly for the same purpose. Thus, the locations of the MU under the 2 conditions were evaluated in relation to the reference points in each larynx.
After recording vibrations under the 2 conditions, the larynx was vertically incised at the midline of the thyroid laminae anteriorly and at the interarytenoid portion posteriorly. The hemilarynx was held firmly on a metal plate. A nylon thread was secured through the arytenoid mound and was fixed posteriorly. Another thread was secured through the thyroid cartilage at the anterior commissure, and this thread was used to adjust vocal fold length to its original length by loading a certain amount of weight via a pulley (Figure 2). Because some of the reference points faded after recording of vocal fold vibration, they were marked again at the free edge, and 2 mm and 4 mm below the free edge to accurately identify the points to be measured. The point 6 mm below the free edge was also marked in dogs 1 through 3. These marks were aligned along the vertical axis of the vocal fold. They also served as landmarks for histological examination. The hemilarynx ipsilateral to the reference points during phonation was used for measurement of pliability of the vocal fold mucosa.
The tip of a 22-gauge injection needle (0.7 mm in inner diameter) was filed under an operating microscope (OpMi-1; Carl Zeiss Ltd, Oberkochen, Germany) until it became flat, and was applied at a right angle to points to be measured on the surface of the vocal fold mucosa. Figure 3 shows the method used to measure pliability of the mucosa. After confirming contact of the needle tip with the mucosal surface under the operating microscope, a constant suction pressure of −300 mm Hg was applied to the needle by connection to a vacuum source at the laboratory wall inlet. A pressure gauge (Data Module AA4632; Toyoda Machine Works, Kariya, Japan) was used to monitor the negative pressure, which was kept constant at −300 mm Hg by adjustment at the vacuum source. The needle tip applied to the mucosa was then slowly moved away with a micromanipulator (MM-3; Narishige, Tokyo). Thus, the mucosa was pulled off its original position by the needle tip until it was detached from the mucosa. These 2 critical moments when the mucosa and the needle tip came into contact and separated were carefully observed under the operating microscope. The distance between the 2 points of contact and detachment was read on the scale of the micromanipulator to the order of 10 µm. The mobility of the vocal fold mucosa was measured at several (5-7) points at a distance of 1 mm starting from the reference point on the free edge. We repeated the measurement 3 times for each point.
Subsequently, more weight was loaded so that vocal fold length was increased to that after cricothyroid approximation. We repeated the procedure in the same way as described above. The free edge of the vocal fold when it was lengthened coincided with the point marked as the free edge without lengthening of the vocal fold. Although distances between 2 neighboring marks were less than 2 mm after lengthening of the vocal fold, points to be measured were accurately set as those marked with India ink and those at the midpoint between 2 neighboring marks. Thus, pliability was measured at fixed points on the mucosa under the 2 conditions.
After measuring pliability of the vocal fold mucosa under the 2 conditions, a small linear cut was made with a No. 11 scalpel oriented along the longitudinal axis of the vocal fold at the 2 reference points 2 mm and 4 mm below the free edge. Thus, these wounds were aligned along the vertical axis of the vocal fold and served as indicators of the 2 reference points, making it possible to histologically identify the position of the points where pliability was measured. The hemilarynx was then fixed with 10% formaldehyde solution, decalcified, embedded in paraffin, sectioned serially in the frontal plane, and stained with hematoxylin-eosin and elastica van Gieson. Histological examination was performed on 3 larynges (dogs 1, 3, and 4).
This study was conducted in compliance with the Guide for Animal Experimentation at Ehime University School of Medicine, Ehime, Japan.
Table 1 summarizes mean airflow rates, F0, and length of the vocal fold under the 2 conditions in 5 dogs. The membranous vocal fold was lengthened 1.3 to 2 mm, which always resulted in an increase of F0. Figure 4, A through E, illustrates pliability measurements (averages of 3 measurements for each measured point) and the estimated range of the vocal fold mucosa where the MU occurred during phonation in the absence and presence of vocal fold lengthening.
Pliability in the presence of vocal fold lengthening was smaller than that in the absence of vocal fold lengthening at all measured points except at the point 3 mm below the free edge in dog 4. These differences in each larynx were statistically significant (P=.05, Mann-Whitney test).
In the absence of vocal fold lengthening, the free edge of the vocal fold showed a maximal pliability that gradually diminished toward the tracheal side and reached a minimum pliability. Then, pliability of the mucosa showed a slight increase further toward the tracheal side in all larynges and was followed by a gradual decrease in dogs 1 and 5 as the mucosa of the vocal fold showed a visible transition to that of the trachea. The Pominwas located 2 mm below the free edge in dog 5, 3 mm in dog 4, 3 to 4 mm in dog 3, 4 mm in dog 1, and 5 mm in dog 2. When the vocal fold was lengthened, the free edge of 3 specimens (dogs 2, 3, and 5) showed a maximal pliability, which diminished gradually toward the tracheal side, and reached a minimum pliability. Then, pliability showed a slight increase, which was followed by a gradual decrease. In dog 1, although the free edge did not show maximal pliability, minimum pliability was reached at 1 mm below the free edge. Then, pliability showed an increase to a point 5 mm below the free edge except at the point 3 mm below the free edge, and showed a gradual decrease at the point 6 mm below the free edge. In dog 4, the free edge showed minimal pliability, which increased and was followed by a gradual decrease. The Pominwith vocal fold lengthening was located at the free edge in dog 4, 1 mm below the free edge in dogs 1 and 5, 1 to 2 mm below the free edge in dog 3, and 3 mm below the free edge in dog 2. Thus, the Pomin shifted toward the oral side in all larynges.
Table 2 summarizes distances of 2 reference points from the free edge during phonation, estimated location of the MU, and location of the Pomin under the 2 conditions in each larynx. In the absence of vocal fold lengthening, the location of the MU was estimated to be around the Pomin in all larynges. In contrast, when the vocal fold was lengthened, the MU observed in 4 larynges (dogs 1, 3, 4 and 5) was located laterally to the Pomin. In dog 2, the location of the MU was estimated to be around the Pomin.
Figure 5 shows a frontal section of the larynx of dog 1 stained with hematoxylin-eosin. The arrows indicate the points 2 mm and 4 mm below the free edge. The Pomin in the absence of vocal fold lengthening, 4 mm below the free edge, was located slightly above the area where the muscular layer approached the epithelial layer. In 2 other larynges, the Pomin in the absence of vocal fold lengthening was also located slightly above the area where the muscular layer approached the epithelial layer.
Although attempts have been made to measure stiffness of the Young modulus of the excised canine vocal fold,6-8these studies did not take into account the effect of layer-by-layer connection between the cover and body on stiffness of the vocal fold. Other authors reported stiffness of the human vocal fold in vivo.12-15Tanaka et al12 measured the maximal distance at which the vocal fold was drawn to the forceps channel of the fiberscope connected to a constant negative pressure. Their results suggested that the suction method was useful in estimating the stiffness of the vocal fold in vivo. Several investigators13-15measured displacement caused by pressing the vocal fold while the subject was anesthetized with orotracheal intubation. As they stated, the modulus of elasticity of the vocal fold is determined by both the stiffening of the body and the tension of the cover. However, they were not able to examine subtle changes in stiffness along the vertical axis as the vocal fold mucosa showed a visible transition to the tracheal mucosa. We previously reported that contraction of the thyroarytenoid muscle significantly increased pliability of the vocal fold mucosa.10 The present study quantitatively demonstrated that lengthening of the vocal fold, which simulated contraction of the cricothyroid muscle, significantly decreased pliability of the vocal fold mucosa. Thus, lengthening of the vocal fold invariably resulted in elevation of F0 in all larynges.
The vocal fold mucosa showed a visible transition to the tracheal mucosa, but without any anatomical landmark that can be identified during phonation. As reported previously,1-4the MU appears on the lower surface of the vocal fold only when it vibrates. Therefore, we used tattooed marks as reference points to estimate the location of the MU during phonation. In the absence of vocal fold lengthening, the location of the MU was estimated to be around the Pomin. Histologically, the lamina propria in dog 1 became thinner as it shifted downward and the muscle layer came into close proximity with the epithelial layer 4 mm below the free edge (Figure 5). Sparse connective tissue, the deep layer of the lamina propria of the canine vocal fold, disappeared just above the area where the muscle layer neared the epithelial layer.16 Thus, the mucosa was relatively tightly connected to the muscle layer around this area. This seems to be the most likely origin of the Pomin. Deformation of the canine vocal fold during vibration occurs mainly in the cover.17It slides up from the underlying body because of the increased subglottic pressure during the closed phase. This movement of the cover generates repeated mucosal waves. The tight junction between the muscle layer and the epithelial layer may interfere with supple mucosal mobility. In this respect, the presence of the sparse deep layer between these 2 layers allows upward motion of the mucosa during vibration. Therefore, the Pomin might coincide with the MU, which is the lower margin where the mucosal wave occurs. Figure 6 illustrates the positional relationships among the lower border of the sparse deep layer of the lamina propria, the Pomin, and the MU in the absence of vocal fold lengthening.
Distance between neighboring marks became shorter after lengthening of the vocal fold than that before lengthening, but the location of the MU could be evaluated in relation to the reference points under the 2 conditions in each larynx. Thus, the results of dogs 1 through 3 in this study indicated that the actual shift of the MU toward the oral side occurred in the presence of vocal fold lengthening. Although the results of dogs 4 and 5 failed to prove the actual shift of the MU in the presence of vocal fold lengthening, it was assumed that the shift of the MU toward the oral side occurred because a previous study3 described that the position of the MU always showed an actual shift to the oral side in the presence of vocal fold lengthening. When the vocal fold was lengthened, the Pomin also shifted upward toward the free edge of the vocal fold. As reported previously,16 the distance between the free edge and the portion where the deep layer of the lamina propria is not visible becomes smaller than that in the absence of vocal fold lengthening. This shift of the lower border of the sparse connective tissue seems to be the main cause of the upward shift of the Pomin and the MU. In 4 larynges, however, the MU occurred more laterally than the Pomin, although the MU when the vocal fold was lengthened invariably occurred more medially than that in the absence of vocal fold lengthening. One possible explanation for this discrepancy in location of the MU and the Pomin would be differences in the procedures used to lengthen the vocal fold. Cricothyroid approximation was performed when driving vibration, while direct anterior traction lengthened the vocal fold when its pliability was measured. Kadota et al18lengthened the vocal folds of 2 canine larynges by cricothyroid approximation and those of 2 other larynges by direct anterior traction, and, using the frozen section of these larynges, measured the distance between the free edge and the portion where the sparse deep layer of the lamina propria was not visible. They normalized the measurements of the larynges lengthened by cricothyroid approximation with those obtained from 2 other larynges, each of which had the same vocal fold length as that of one of the larynges used for cricothyroid approximation. The measurements of the larynges lengthened by direct anterior traction were normalized with those obtained from the other side of the larynx used. According to their results, the distance between the free edge and the portion where the deep layer of the lamina propria was not visible was slightly longer when the vocal fold was lengthened by cricothyroid approximation than when it was lengthened by direct anterior traction. However, this result was not conclusive because the number of larynges used in the studies was too small. Further studies are required to investigate the cause for the discrepancy in the positional relationship between the MU and the Pomin in the presence of vocal fold lengthening.
Hirano et al19 used a fiberstroboscope inserted through a tracheostoma to view the vocal fold from below, and they reported the occurrence of the MU during phonation in 2 living human subjects. Yumoto et al20 observed the inferior aspect of the vocal fold during phonation with the aid of a rigid oblique-view endoscope inserted through a tracheostoma (ie, inferior glottoscopy). Inferior glottoscopy revealed the occurrence of the MU in 19 subjects during sustained phonation of the vowel "a" at a comfortable pitch and loudness (ie, easy phonation). During high-pitched phonation of the vowel, the vocal fold became longer. Use of a dilated blood vessel as a landmark in one subject showed the location of the MU during high-pitched phonation to actually shift medially toward the oral side compared with that during easy phonation. Thus, the infraglottic aspect of vocal fold vibration observed in the living human subjects was quite similar to that observed in the excised canine larynx, although different layer structures of the vocal fold in these species should be noted. The lamina propria of the canine vocal fold mucosa is thicker than that in man, which is devoid of the sparse deep layer of the lamina propria. Further studies are required to examine pliability of the human vocal fold along the vertical axis to clarify the relationship between the MU and pliability of the vocal fold mucosa.
Accepted for publication March 13, 1998.
Reprints: Eiji Yumoto, MD, Department of Otolaryngology–Head and Neck Surgery, School of Medicine, Ehime University, 454 Shitsukawa, Shigenobu-cho, Onsen-gun, Ehime, Japan 791-0295 (e-mail: email@example.com).
Create a personal account or sign in to: