- Email: [email protected]
Comparison between thyroplasty type I andArytenoid rotation: a study of vocal fold vibration using excised human larynges
Comparison Between Thyroplasty Type I and Arytenoid Rotation: A Study of Vocal Fold Vibration Using Excised Human Larynges *Domingos H. Tsuji, *Edigar R. de Almeida, *Luiz Ubirajara Sennes, *Ossamu Butugan, and †Silvia M. R. Pinho Sa˜o Paulo, Brazil
Summary: The purpose of this paper was to compare the vibration of the vocal fold submitted to Isshiki thyroplasty type I (TPI) to that of the contralateral one adducted by the arytenoid rotation (AR) technique. The vocal folds of ten human fresh excised larynges were medialized by TPI on one side and by rotation of the arytenoid on the contralateral side. Laryngeal vibration was artificially produced and was recorded by videostroboscopy. The images were subjectively and objectively analyzed. Subjective analysis included periodicity of vibratory cycles, features of the mucosal wave present on the TPI side, amplitude of vibration, and profile of free border of each vocal fold during the opening phase. Objective analyses were carried out on frame-by-frame digitalized images to determine amplitudes of vibrations and phase differences between the folds in three glottic regions (anterior, middle, and posterior). Subjective analysis revealed regular periodicity in 100% of the larynges, a decrease in the mucosal wave on the TPI side in 70%, reduction in amplitude in 30%, and a sigmoid profile of the free border on the TPI side in 80%. Objective analysis showed mean amplitude in the posterior glottic region on the TPI side significantly larger than that on the arytenoids rotation side and phase asymmetry in 90% of the larynges. Key Words: Thyroplasty type I—Laryngoplasty—Arytenoid rotation—Vocal fold vibration.
Address correspondence and reprint requests to Domingos H. Tsuji, Rua Peixoto Gomide, 515, cj. 145, Cerqueira Ce´sar, Sa˜o Paulo, Capital, Brazil, CEP: 01409-001. Journal of Voice, Vol. 17, No. 4, pp. 596–604 쑕 2003 The Voice Foundation 0892-1997/2003 $30.00⫹0 doi:10.1067/S0892-1997(03)00071-7
Accepted for publication January 10, 2003. An outline of this paper was presented at the 2nd World Voice Congress and 5th International Symposium on Phonosurgery, Sa˜o Paulo, Brazil, February 8-11, 1999. From the *Department of Otolaryngology, Sao Paulo University Faculty of Medicine, Sao Paulo, Brazil; †CEFAC, Clinical Speech and Language Pathology and Audiology Specialization Courses, Sa˜o Paulo, Brazil.
596
THYROPLASTY TYPE I AND ARYTENOID ROTATION INTRODUCTION The aim of all surgical treatment of unilateral vocal fold (VF) paralysis consists of bringing the paralyzed fold to the midline of the glottis, thus obtaining better glottic closure during phonation, a condition that, from a physiological viewpoint, is fundamental for sound production of good quality. Among the different techniques available, the currently most widely used methods include injection techniques such as autologous fat injection1 or collagen injection2 performed by endoscopy, and external approaches such as thyroplasty type I (TPI) and arytenoid rotation (AR) as described by Isshiki.3,4 Many centers have established TPI as the surgery of choice, including ours. The reasons for this tendency are the good clinical results compared to some unfavorable results previously obtained with Teflon injection (mainly employed in the past), such as extralaryngeal leakage,5,6 excessive application and improper localization,7 late vocal deterioration, difficulty in reversing surgery, and granuloma formation.8–10 The good functional results obtained with TPI have been mainly attributed to the fact that the cordal structures do not suffer histological or anatomical alterations,11,12 thus permitting a better vibratory pattern of the manipulated fold. Despite the great popularity of this technique, studies on the vibratory characteristics of the paralyzed fold after surgery are still rare and most of them are based on clinical cases in which numerous variables may influence the outcome. In order to exclude or, at least, minimize some possible variables such as muscular atrophy, residual and/or regenerative innervation, and compensatory neuromuscular mechanisms, which might be present in a clinical study and somehow influence the final outcome of VF vibration, we conducted an experimental study using fresh excised human larynges artificially submitted to vibration. The objective of the present study was to compare the vibration of the vocal fold submitted to Isshiki thyroplasty type I to that of the contralateral one adducted by the arytenoid rotation technique. MATERIAL AND METHODS This experimental study was conducted on ten normal larynges excised from human cadavers
597
within a maximum period of 24 hours after death. Larynges in which the vocal folds were not in the paramedian position were excluded. One of the vocal folds was medialized by AR as described by Isshiki.4 TPI was carried out on the contralateral VF according to the basic technique and measures suggested by the same author,13 only differing from this method in the removal of the cartilage window. A silicone block of Shore 20 hardness grade was introduced through the window with careful adjustment of its depth in order to obtain a precise VF medialization. Further fine adjustments were also made during the induced vibration in order to achieve a stable and continuous vibration. In three larynges, the vocal processes were connected with a suture stitch to adjust the vocal processes to the same level in order to promote a better closure of the posterior glottis. Thus, both vocal folds were at the same medial position and level, with the main difference between them being the presence of a silicone block on the TPI side. Experimental vibration of the vocal folds was obtained using an apparatus modified from that employed by Van Den Berg and Tan.14 Stroboscopic images of VF vibration were obtained using a Bru¨el and Kjaer type 4914 stroboscopic source (Bru¨el and Kjaer A/S, Naerum, Denmark), and the images were captured and saved using a Machida laryngeal telescope (model LY-CS30) (Toshiba Corporation, Tokyo, Japan) connected to a Toshiba CCD camera (model IK-M41A) (Machida Endoscope Co., Ltd., Tokyo, Japan). The images were saved on S-VHS videotapes using a Panasonic model NV-FS90 videotape recorder (NTSC system) (Matsushita Electric Industrial Co., Ltd., Tokyo, Japan). Small adjustments in airflow were made during vibration to obtain the best possible vibration. The images obtained and saved were initially submitted to subjective analysis by two observers who had significant experience in this type of evaluation and then digitalized in order to obtain objective vibration data. For subjective analysis, periodicity was classified as regular, irregular, or inconsistent. The mucosal wave of the fold (defined as undulation of the mucosal layer traveling from a medial to a lateral direction along the superior surface of the vocal fold during vibration) submitted to TPI was classified as the same, larger, or smaller compared to the mucosal Journal of Voice, Vol. 17, No. 4, 2003
598
DOMINGOS H. TSUJI ET AL
FIGURE 1. Free border profile observed during opening phase. (A) sigmoid type. (B) semi-spindle-shaped type.
wave of the AR fold. The amplitude of vibration (defined as the lateral excursion of the vocal fold edge during its displacement away from the maximum medial point reached by the edge at the closed phase) was also classified as same, larger, or smaller compared to the AR side. The free border profile observed during progression of the opening phase of each VF was classified as semi-spindle-shaped or sigmoid (Figure 1). For objective analysis of vibration, five sequential vibratory cycles, was selected and digitalized using a Creative Video Blaster (Adobe Systems Incorporated, San Jose, CA) image capture plate and the Adobe Premiere LE image capture program (Creative Technology Ltd., Scotts Valley, CA) installed on a PC computer. The stored digitalized images of each larynx were analyzed frame-byframe, obtaining amplitude measurements at three points of the vocal folds (anterior, middle, and posterior) using the Corel Draw program. A glottic midline was traced between the posterior and anterior vertices of the glottis observable during the opened phase. This line was taken as the reference point zero to measure the position of the VF free borders in each frame. The values obtained on the right (R) and left (L) sides of the line were considered positive and negative, respectively (Figure 2). The values obtained as measurement Journal of Voice, Vol. 17, No. 4, 2003
FIGURE 2. Three glottic regions where the vibration amplitudes were measured. TPI on right (R)side and AR on left (L).
units were introduced into the Excel program to obtain a representative vibration graph used for the analysis of amplitudes and vibration phases (Figure 3). Amplitude was defined as the lateral excursion measurement of the free border in relation to the maximum medialization point (Figure 4). In order to visualize the morphological changes provoked by the TPI on the VF, coronal and axial CT sections were consecutively taken from the five larynges used during the experiment. The Fisher exact test was used to compare the presence of sigmoid profile on the TPI side and its occurrence on the AR side. The mean amplitudes of the TPI side (considering all the larynges) were compared to those of the AR side by the nonparametric Wilcoxon method.
RESULTS Subjective analysis of vibration The results of subjective analysis of cord vibration with respect to periodicity, mucosal wave, and amplitude are shown in Table 1. The subjective analysis of the free border profile during progression of the opening phase is presented in Table 2.
THYROPLASTY TYPE I AND ARYTENOID ROTATION
599
FIGURE 3. Graphic representation of the amplitude measurements obtained at tree points of vocal folds - larynx number 2, thyroplasty on right side.
Journal of Voice, Vol. 17, No. 4, 2003
600
DOMINGOS H. TSUJI ET AL
FIGURE 4. Amplitude corresponds to the lateral excursion measured in relation to the maximum medialization point of the free border.
Objective analysis of vibration Comparison between mean amplitudes of each glottic region The arithmetic mean and standard deviation of the amplitudes of each glottic region are shown in Table 3. No statistically significant differences in mean amplitudes of the anterior and middle regions were observed between the two sides. However, the mean amplitude of the posterior region was significantly larger in the folds submitted to TPI than in the AR folds. Study of the vibration phases in each larynx Asymmetry of vibration phases was observed in nine of ten (90%) larynges. Computed tomography (CT) images Coronal CT scans showed increased thickening of the fold on the TPI side in all five larynges
studied (Figure 5). No differences between folds were observed on the axial plane.
DISCUSSION Several variables may contribute to the vibratory patterns observed after VF medialization procedures in paralyzed larynges, such as muscular atrophy, residual innervation, degree of reinnervation, and compensatory vocal posture.15 Therefore, clinical observations are not sufficiently precise to determine the impact of these procedures on VF vibration. Hence, we used excised human larynges as an experimental model to demonstrate the vibratory behavior of the vocal folds under TPI and AR, without the influence of neuromuscular activities. The slow-motion effect on the vibratory movement obtained with a stroboscope depends on the relationship between the frequency of the light
TABLE 1. Subjective Analysis of Vibration Periodicity
Mucosal wave
Amplitude
Regular ⫽ 10 (100%) Irregular ⫽ 0 (0%) Inconsistent ⫽ 0 (0%)
(T ⬍ AR) ⫽ 7 (70%) (T ⫽ AR) ⫽ 2 (20%) (T ⬎ AR) ⫽ 1 (10%)
(T ⬎ AR) ⫽ 4 (40%) (T ⫽ AR) ⫽ 3 (30%) (T ⬍ AR) ⫽ 3 (30%)
T: VF submitted to TPI; AR: VF submitted to arytenoids rotation. Journal of Voice, Vol. 17, No. 4, 2003
TABLE 2. Profile of the Free Borders TPI Border profile Sigmoid Semi-spindle Total
AR
No. of larynges
(%)
No. of larynges
(%)
8 2 10
80.0 20.0 100.0
0 10 10
0.0 100.0 100.0
NOTE: Fisher exact test: P ⫽ 0.0007.
THYROPLASTY TYPE I AND ARYTENOID ROTATION
601
TABLE 3. Amplitude of the Three Glottic Regions Group
Mean ⫾ SD
A: Anterior TPI AR z ⫽ 1.46; P ⫽ 0.1440.
0.17 ⫾ 0.04 0.19 ⫾ 0.07
B: Middle TPI AR z ⫽ 0.19; P ⫽ 0.8510.
0.22 ⫾ 0.08 0.22 ⫾ 0.07
C: Posterior TPI AR z ⫽ 4.17; P ⫽ 0.0001.
0.23 ⫾ 0.13 0.18 ⫾ 0.06
impulses emitted by the stroboscopic source and the frequency of VF vibration.16 The absence of adequate synchronization between the light pulses and successive vibratory cycles results in the loss of sequential arrangement of the vibration phases with disturbance of the sharpness of the videostroboscopic images.17 We obtained videostroboscopic images that were characterized by a high degree of sharpness and synchronization between the light source and the vibratory cycles. This fact may be interpreted as clear evidence of the regularity of the vibratory cycle periods and was observed in all (100%) studied larynges submitted to the medialization procedures, even in the presence of a silicone block in the TPI. According to Moore and Thompson,18 the degree of periodic irregularity is directly proportional to the degree of hoarseness. The periodicity observed in our series proves that this characteristic is restored or preserved after TPI and AR, and that it may be considered as one of the determining factors of good vocal quality reported by different authors.19–22 Isshiki23 reported that hoarseness may be caused by both the absence of complete closure of the glottis and its excessive closure where the contact between the folds presents a certain degree of compression (initial glottic area ⬍0). In the present experiment, aperiodic vibratory cycles were noted when the depth of the silicone block was too great, causing excessive medialization of the VF. That was the
FIGURE 5. CT image obtained after thyroplasty type I on left side.
reason why further fine adjustments of the silicone block depth were made during the vibration. This fact should be emphasized in clinical practice when vocal adjustment during surgery is essential for preventing hypermedialization of the VF, which may result in aperiodic vibratory cycles and, consequently, in undesirable vocal quality. Based on the present observations, whenever possible the use of a videostroboscope through a nasofiberscope should be recommended during surgery to confirm cycle periodicity and to optimize vocal outcome. According to Hirano’s theory of body and cover,24 the difference in tissue consistency between the lining mucosa and the musculoligamentous body is of fundamental importance for the formation of the mucosal wave during vocal folds vibration. Fukuda et al25 experimentally demonstrated that muscle stimulation (with an increase in muscle rigidity) results in an increase in the mucosal vibration amplitude, confirming the importance of larger rigidity of the musculoligamentous body in the mucosal wave motion. As expected, the CT images obtained from five consecutively analyzed larynges clearly demonstrated that TPI provoked a significant increase in VF thickness (Figure 5) due to the compression of the thyroarytenoide muscle by the silicone block. The compression of the structures and the increase in VF volume lead to transverse stretching of the mucosa, resulting in alterations such as increased relative mass and higher degree of tension of the structures. Although the mechanism involved in this Journal of Voice, Vol. 17, No. 4, 2003
602
DOMINGOS H. TSUJI ET AL
transverse stretching is different from that involved in longitudinal stretching observed during emission of higher frequency sounds, the superficial layer of the lamina propria (Reinke’s space) should also become thinner due to the stretching and compression to which it is submitted. These alterations characterized by a higher tension at the mucosal level and by a decrease in the thickness of Reinke’s space may explain, in a way similar to what occurs during longitudinal stretching, the reduction in the mucosal wave observed in 70% of vocal folds submitted to TPI in comparison to the contralateral folds submitted to AR. However, one of the larynges showed an increased mucosal wave on the TPI side. This finding seems to be a paradox in view of the predominance of wave reduction in the other larynges, and we do not have a definitive explanation for it, but we may raise the hypothesis that the silicone implant might have increased the difference in rigidity between the body and the coverage. Tanabe et al26 and Isshiki et al27 observed VF vibration in the presence of tension asymmetry between folds characterized by the presence of phase asymmetry but without a difference in the frequencies between them. According to Isshiki et al,27 phase asymmetry is the most important sign of asymmetric tension. These findings have been confirmed by Moore et al28 in experimental studies on dogs and by Sercarz et al29 in clinical studies on 20 patients with VF paralysis. Isshiki30 reported that the same type of vibration, with phase asymmetry, can be observed in the presence of mass asymmetry between folds. In the present study, we observed phase asymmetry in 90% of the larynges. These findings seem to support our previous assumption that structural alterations caused by the silicone block in TPI result in tension and mass changes in the VF structure, a fact that does not occur, at least at the same intensity, in the AR. A rigid body present in the paraglottic space, medially displacing the VF structure, may behave as a blockage against lateral displacement during the opening phase of the vibratory cycle, reducing the vibration amplitude. However, subjective comparison of the vibration amplitude showed greater amplitude on the TPI side in 40% of the larynges, the same amplitude in 30%, and a smaller amplitude in only 30%. These results may appear to be even more Journal of Voice, Vol. 17, No. 4, 2003
incoherent if we consider that the mucosal wave on the TPI sides was found to be smaller or equal to the side of arytenoid rotation in nine of the ten larynges (smaller in seven larynges and equal in two). This apparent incoherence is very probably due to the difficulties met by the examiners during subjective analysis of amplitude. Isshiki et al23 also reported difficulties in defining the vibration amplitude in the presence of coaptation line displacement occurring during tension and phase asymmetry between folds. This difficulty was also encountered by Thompson et al21 when evaluating the vibration amplitude in patients submitted to TPI. To clarify these findings, we carried out objective analyses of the medial lateral displacement of the free border of the fold during vibration. No statistically significant differences in mean amplitudes were observed in the anterior and middle regions of the two vocal folds. However, the amplitude of the posterior region on the TPI side was significantly larger than that observed for the AR side. Very probably, this greater amplitude in the posterior region influenced the subjective analysis of the examiners at the time when they classified the amplitude on the TPI side as being larger than that on the AR side in 40% of the larynges. The larger amplitude detected in the posterior region also explains the results observed by subjective analysis of the free border of the vocal folds where eight larynges with a sigmoid profile were found on the TPI side and ten with a semi-spindle-shaped profile on the AR side. Lateral displacement of the VF with a sigmoid profile may be described as larger lateral movement of the posterior region of the VF, which even occurs after the anterior region has reached its maximum lateralization. We presumed that both the sigmoid profile on the fold submitted to TPI and the larger mean amplitude observed in its posterior region are the consequence of a larger airflow directed to the posterior region of the VF on the TPI side due to an increased VF resistance in its anterior two-thirds caused by the rigidity of the silicone block. As emphasized by different authors,24,31,32 the mucosal layer (cover) plays a fundamental role in VF vibration. Alterations in its characteristics result in vibratory disturbances with a consequent impact on sound production. Studies by Gray et al33 and Lu
THYROPLASTY TYPE I AND ARYTENOID ROTATION et al34 have demonstrated that, despite a significant vocal improvement upon TPI, some parameters differed significantly from normal patterns. Gray et al33 observed that vocal characteristics such as the degree of tension, breathiness, hoarseness, harshness, and unsteadiness were significantly altered compared to normality. Although the objective of the present study was not to establish a correlation between each type of vibratory alteration observed and the vocal acoustic changes, we believe that alterations such as strain, hoarseness, and unsteadiness observed by Gray et al33 in patients after TPI are the result of some vibratory abnormality such as a decrease in mucosal wave and asymmetry in vibration amplitude and phase.
CONCLUSION Based on the present experimental study, we conclude that TPI and AR are able to preserve the periodicity of VF vibration. However, the implanted silicone block in TPI, leading to increased vocal folds thickness and stretching of the mucosal membrane, causes more alterations in vibration parameters such as mucosal wave motion, profile of VF free border during the opening phase, and amplitude and phase symmetry than in the AR procedure.
REFERENCES 1. Hill DP, Meyers AD, Harris J. Autologous fat injection for vocal cord medialization in the canine larynx. Laryngoscope. 1991;101:344–348. 2. Ford CN, Martin DW, Warner TF. Injectable collagen in laryngeal rehabilitation. Laryngoscope. 1984;94:513–518. 3. Isshiki N, Morita H, Okamura H, Hiramoto M. Thyroplasty as a new phonosurgical technique. Acta Otolaryngol (Stockh). 1974;78:451–457. 4. Isshiki N, Tanabe M, Sawada M. Arytenoid adduction for unilateral vocal cord paralysis. Arch Otolaryngol. 1978; 104:555–558. 5. Stone JW, Arnold GE, Stephens CB. Intracordal polytef (Teflon) injection. Histologic study of three further cases. Arch Otolaryngol. 1970;91:568–574. 6. Wenig BM, Heffner DK, Oertel YC, Johnson FB. Teflonomas of the larynx and neck. Human Pathol. 1990;21: 617–623.
603
7. Rubin HJ. Misadventures with injectable polytef (Teflon). Arch Otolaryngol. 1975;101:114–116. 8. Ossoff RH, Koriwchak MJ, Netterville JM, Duncavage JA. Difficulties in endoscopic removal of Teflon granulomas of the vocal fold. Ann Otol Rhinol Laryngol. 1993; 102:405–412. 9. Kasperbauer JL, Slavit DH, Maragos NE. Teflon granulomas and overinjection of Teflon: a therapeutic challenge for the otorhinolaryngologist. Ann Otol Rhinol Laryngol. 1993;102:748–751. 10. Varvares MA, Montgomery WW, Hillman RE. Teflon granuloma of the larynx: etiology, pathophysiology, and management. Ann Otol Rhinol Laryngol. 1995;104:511–515. 11. Lablance GR, Maves MD. Acoustic characteristics of postthyroplasty patients. Otolaryngol Head Neck Surg. 1992; 107:558–563. 12. D’Antonio LL, Wigley TL, Zimmerman GJ. Quantitative measures of laryngeal function following Teflon injection or thyroplasty type I. Laryngoscope. 1995;105:256–262. 13. Isshiki N, Taira T, Kojima H, Shoji K. Recent modifications in thyroplasty type I. Ann Otol Rhinol Laryngol. 1989; 98:777–779. 14. Van Den Berg J, Tan TS. Results of experiments with human larynxes. Pract Otorhinolaryngol. 1959;21:425–450. 15. Woodson GE. Configuration of the glottis in laryngeal paralysis I: clinical study. Laryngoscope. 1993;103:1227–1234. 16. Kallen LA. Laryngostroboscopy in the practice of otolaryngology. Arch Otolaryngol. 1932;16:791–807. 17. Hirano M, Bless DM. Vocal fold vibration. In: Hicano M, Bless DM, eds. Videostroboscopic examination of the larynx. San Diego, CA: Singular Publishing Group Inc; 1993:23–36. 18. Moore P, Thompson CL. Comments on physiology of hoarseness. Arch Otolaryngol. 1965;81:97–102. 19. Isshiki N, Okamura H, Ishikawa T. Thyroplasty type I (lateral compression) for dysphonia due to vocal cord paralysis or atrophy. Acta Otolayngol (Stockh). 1975;80:465–473. 20. Koufman JA. Laryngoplasty for vocal cord medialization: an alternative to Teflon. Laryngoscope. 1986;96:726–731. 21. Thompson DM, Maragos NE, Edwards BW. The study of vocal fold vibratory patterns in patients with unilateral vocal fold paralysis before and after type I thyroplasty with or without arytenoid adduction. Laryngoscope. 1995;105: 481–486. 22. Bielamowicz S, Gupta A, Sekhar LN. Early arytenoid adduction for vagal paralysis after skull base surgery. Laryngoscope. 2000;110:346–351. 23. Isshiki N, Tanabe M, Ishizaka K, Broad D. Clinical significance of asymmetrical vocal cord tension. Ann Otol Rhinol Laryngol. 1977;86:58–66. 24. Hirano M. Morphological structure of the vocal cord as a vibrator and its variations. Folia Phoniatr. 1974;26:89–94. 25. Fukuda H, Saito S, Kitahara S, Isogai Y, Makino K, Tsuzuki T, Kogawa N, Ono H. Vocal fold vibration in excised larynges viewed with an X-ray stroboscope and an ultrahigh-speed camera. In: Titze ER, Scherer RC, eds. Vocal Journal of Voice, Vol. 17, No. 4, 2003
604
26.
27.
28.
29.
DOMINGOS H. TSUJI ET AL
fold physiology. San Diego, CA: College-Hill Press; 1983: 238–252. Tanabe M, Isshiki N, Kitajima K. Vibratory pattern of the vocal cord in unilateral paralysis of the cricothyroid muscle. An experimental study. Acta Otolaryngol. 1972; 74:339–345. Isshiki N. Functional surgery of the larynx. Department of Otolaryngology, Kyoto University Medical School; 1977:207. [Special report presented at 78th Congress of Japan Otolaryngology Society, Fukuoka, Japan, 1977]. Moore DM, Berke GS, Hanson DG, Ward PH. Videostroboscopy of the canine larynx: the effects of asymmetric laryngeal tension. Laryngoscope. 1987;97:543–553. Sercarz JA, Berke GS, Ming Y, Gerratt BR, Natividad M. Videostroboscopy of human vocal fold paralysis. Ann Otol Rhinol Laryngol. 1992;101:567–577.
Journal of Voice, Vol. 17, No. 4, 2003
30. Isshiki N. Physiology of speech production. In: Isshiki N, ed. Phonosurgery: theory and practice. Tokyo: SpringerVerlag; 1989:5–21. 31. Saito S, Fukuda H, Isogai Y, Ono H. X-ray stroboscopy. In: Stevens KN, Hirano M, eds. Vocal fold physiology. Tokyo: University of Tokyo Press; 1981:95–106. 32. Tsuzuki T. A basic study on the vocal fold vibration from the viewpoint of the layer system of the vocal fold. Otologia (Fukuoka). 1984;30:131–152. 33. Gray SD, Barkmeier J, Jones D, Titze I, Druker D. Vocal evaluation of thyroplastic surgery in the treatment of unilateral vocal fold paralysis. Laryngoscope. 1992;102: 415–421. 34. Lu FL, Casiano RR, Lundy DS, Xue JW. Longitudinal evaluation of the vocal function after thyroplasty type I in the treatment of unilateral vocal paralysis. Laryngoscope. 1996;106:573–577.
Summary: The purpose of this paper was to compare the vibration of the vocal fold submitted to Isshiki thyroplasty type I (TPI) to that of the contralateral one adducted by the arytenoid rotation (AR) technique. The vocal folds of ten human fresh excised larynges were medialized by TPI on one side and by rotation of the arytenoid on the contralateral side. Laryngeal vibration was artificially produced and was recorded by videostroboscopy. The images were subjectively and objectively analyzed. Subjective analysis included periodicity of vibratory cycles, features of the mucosal wave present on the TPI side, amplitude of vibration, and profile of free border of each vocal fold during the opening phase. Objective analyses were carried out on frame-by-frame digitalized images to determine amplitudes of vibrations and phase differences between the folds in three glottic regions (anterior, middle, and posterior). Subjective analysis revealed regular periodicity in 100% of the larynges, a decrease in the mucosal wave on the TPI side in 70%, reduction in amplitude in 30%, and a sigmoid profile of the free border on the TPI side in 80%. Objective analysis showed mean amplitude in the posterior glottic region on the TPI side significantly larger than that on the arytenoids rotation side and phase asymmetry in 90% of the larynges. Key Words: Thyroplasty type I—Laryngoplasty—Arytenoid rotation—Vocal fold vibration.
Address correspondence and reprint requests to Domingos H. Tsuji, Rua Peixoto Gomide, 515, cj. 145, Cerqueira Ce´sar, Sa˜o Paulo, Capital, Brazil, CEP: 01409-001. Journal of Voice, Vol. 17, No. 4, pp. 596–604 쑕 2003 The Voice Foundation 0892-1997/2003 $30.00⫹0 doi:10.1067/S0892-1997(03)00071-7
Accepted for publication January 10, 2003. An outline of this paper was presented at the 2nd World Voice Congress and 5th International Symposium on Phonosurgery, Sa˜o Paulo, Brazil, February 8-11, 1999. From the *Department of Otolaryngology, Sao Paulo University Faculty of Medicine, Sao Paulo, Brazil; †CEFAC, Clinical Speech and Language Pathology and Audiology Specialization Courses, Sa˜o Paulo, Brazil.
596
THYROPLASTY TYPE I AND ARYTENOID ROTATION INTRODUCTION The aim of all surgical treatment of unilateral vocal fold (VF) paralysis consists of bringing the paralyzed fold to the midline of the glottis, thus obtaining better glottic closure during phonation, a condition that, from a physiological viewpoint, is fundamental for sound production of good quality. Among the different techniques available, the currently most widely used methods include injection techniques such as autologous fat injection1 or collagen injection2 performed by endoscopy, and external approaches such as thyroplasty type I (TPI) and arytenoid rotation (AR) as described by Isshiki.3,4 Many centers have established TPI as the surgery of choice, including ours. The reasons for this tendency are the good clinical results compared to some unfavorable results previously obtained with Teflon injection (mainly employed in the past), such as extralaryngeal leakage,5,6 excessive application and improper localization,7 late vocal deterioration, difficulty in reversing surgery, and granuloma formation.8–10 The good functional results obtained with TPI have been mainly attributed to the fact that the cordal structures do not suffer histological or anatomical alterations,11,12 thus permitting a better vibratory pattern of the manipulated fold. Despite the great popularity of this technique, studies on the vibratory characteristics of the paralyzed fold after surgery are still rare and most of them are based on clinical cases in which numerous variables may influence the outcome. In order to exclude or, at least, minimize some possible variables such as muscular atrophy, residual and/or regenerative innervation, and compensatory neuromuscular mechanisms, which might be present in a clinical study and somehow influence the final outcome of VF vibration, we conducted an experimental study using fresh excised human larynges artificially submitted to vibration. The objective of the present study was to compare the vibration of the vocal fold submitted to Isshiki thyroplasty type I to that of the contralateral one adducted by the arytenoid rotation technique. MATERIAL AND METHODS This experimental study was conducted on ten normal larynges excised from human cadavers
597
within a maximum period of 24 hours after death. Larynges in which the vocal folds were not in the paramedian position were excluded. One of the vocal folds was medialized by AR as described by Isshiki.4 TPI was carried out on the contralateral VF according to the basic technique and measures suggested by the same author,13 only differing from this method in the removal of the cartilage window. A silicone block of Shore 20 hardness grade was introduced through the window with careful adjustment of its depth in order to obtain a precise VF medialization. Further fine adjustments were also made during the induced vibration in order to achieve a stable and continuous vibration. In three larynges, the vocal processes were connected with a suture stitch to adjust the vocal processes to the same level in order to promote a better closure of the posterior glottis. Thus, both vocal folds were at the same medial position and level, with the main difference between them being the presence of a silicone block on the TPI side. Experimental vibration of the vocal folds was obtained using an apparatus modified from that employed by Van Den Berg and Tan.14 Stroboscopic images of VF vibration were obtained using a Bru¨el and Kjaer type 4914 stroboscopic source (Bru¨el and Kjaer A/S, Naerum, Denmark), and the images were captured and saved using a Machida laryngeal telescope (model LY-CS30) (Toshiba Corporation, Tokyo, Japan) connected to a Toshiba CCD camera (model IK-M41A) (Machida Endoscope Co., Ltd., Tokyo, Japan). The images were saved on S-VHS videotapes using a Panasonic model NV-FS90 videotape recorder (NTSC system) (Matsushita Electric Industrial Co., Ltd., Tokyo, Japan). Small adjustments in airflow were made during vibration to obtain the best possible vibration. The images obtained and saved were initially submitted to subjective analysis by two observers who had significant experience in this type of evaluation and then digitalized in order to obtain objective vibration data. For subjective analysis, periodicity was classified as regular, irregular, or inconsistent. The mucosal wave of the fold (defined as undulation of the mucosal layer traveling from a medial to a lateral direction along the superior surface of the vocal fold during vibration) submitted to TPI was classified as the same, larger, or smaller compared to the mucosal Journal of Voice, Vol. 17, No. 4, 2003
598
DOMINGOS H. TSUJI ET AL
FIGURE 1. Free border profile observed during opening phase. (A) sigmoid type. (B) semi-spindle-shaped type.
wave of the AR fold. The amplitude of vibration (defined as the lateral excursion of the vocal fold edge during its displacement away from the maximum medial point reached by the edge at the closed phase) was also classified as same, larger, or smaller compared to the AR side. The free border profile observed during progression of the opening phase of each VF was classified as semi-spindle-shaped or sigmoid (Figure 1). For objective analysis of vibration, five sequential vibratory cycles, was selected and digitalized using a Creative Video Blaster (Adobe Systems Incorporated, San Jose, CA) image capture plate and the Adobe Premiere LE image capture program (Creative Technology Ltd., Scotts Valley, CA) installed on a PC computer. The stored digitalized images of each larynx were analyzed frame-byframe, obtaining amplitude measurements at three points of the vocal folds (anterior, middle, and posterior) using the Corel Draw program. A glottic midline was traced between the posterior and anterior vertices of the glottis observable during the opened phase. This line was taken as the reference point zero to measure the position of the VF free borders in each frame. The values obtained on the right (R) and left (L) sides of the line were considered positive and negative, respectively (Figure 2). The values obtained as measurement Journal of Voice, Vol. 17, No. 4, 2003
FIGURE 2. Three glottic regions where the vibration amplitudes were measured. TPI on right (R)side and AR on left (L).
units were introduced into the Excel program to obtain a representative vibration graph used for the analysis of amplitudes and vibration phases (Figure 3). Amplitude was defined as the lateral excursion measurement of the free border in relation to the maximum medialization point (Figure 4). In order to visualize the morphological changes provoked by the TPI on the VF, coronal and axial CT sections were consecutively taken from the five larynges used during the experiment. The Fisher exact test was used to compare the presence of sigmoid profile on the TPI side and its occurrence on the AR side. The mean amplitudes of the TPI side (considering all the larynges) were compared to those of the AR side by the nonparametric Wilcoxon method.
RESULTS Subjective analysis of vibration The results of subjective analysis of cord vibration with respect to periodicity, mucosal wave, and amplitude are shown in Table 1. The subjective analysis of the free border profile during progression of the opening phase is presented in Table 2.
THYROPLASTY TYPE I AND ARYTENOID ROTATION
599
FIGURE 3. Graphic representation of the amplitude measurements obtained at tree points of vocal folds - larynx number 2, thyroplasty on right side.
Journal of Voice, Vol. 17, No. 4, 2003
600
DOMINGOS H. TSUJI ET AL
FIGURE 4. Amplitude corresponds to the lateral excursion measured in relation to the maximum medialization point of the free border.
Objective analysis of vibration Comparison between mean amplitudes of each glottic region The arithmetic mean and standard deviation of the amplitudes of each glottic region are shown in Table 3. No statistically significant differences in mean amplitudes of the anterior and middle regions were observed between the two sides. However, the mean amplitude of the posterior region was significantly larger in the folds submitted to TPI than in the AR folds. Study of the vibration phases in each larynx Asymmetry of vibration phases was observed in nine of ten (90%) larynges. Computed tomography (CT) images Coronal CT scans showed increased thickening of the fold on the TPI side in all five larynges
studied (Figure 5). No differences between folds were observed on the axial plane.
DISCUSSION Several variables may contribute to the vibratory patterns observed after VF medialization procedures in paralyzed larynges, such as muscular atrophy, residual innervation, degree of reinnervation, and compensatory vocal posture.15 Therefore, clinical observations are not sufficiently precise to determine the impact of these procedures on VF vibration. Hence, we used excised human larynges as an experimental model to demonstrate the vibratory behavior of the vocal folds under TPI and AR, without the influence of neuromuscular activities. The slow-motion effect on the vibratory movement obtained with a stroboscope depends on the relationship between the frequency of the light
TABLE 1. Subjective Analysis of Vibration Periodicity
Mucosal wave
Amplitude
Regular ⫽ 10 (100%) Irregular ⫽ 0 (0%) Inconsistent ⫽ 0 (0%)
(T ⬍ AR) ⫽ 7 (70%) (T ⫽ AR) ⫽ 2 (20%) (T ⬎ AR) ⫽ 1 (10%)
(T ⬎ AR) ⫽ 4 (40%) (T ⫽ AR) ⫽ 3 (30%) (T ⬍ AR) ⫽ 3 (30%)
T: VF submitted to TPI; AR: VF submitted to arytenoids rotation. Journal of Voice, Vol. 17, No. 4, 2003
TABLE 2. Profile of the Free Borders TPI Border profile Sigmoid Semi-spindle Total
AR
No. of larynges
(%)
No. of larynges
(%)
8 2 10
80.0 20.0 100.0
0 10 10
0.0 100.0 100.0
NOTE: Fisher exact test: P ⫽ 0.0007.
THYROPLASTY TYPE I AND ARYTENOID ROTATION
601
TABLE 3. Amplitude of the Three Glottic Regions Group
Mean ⫾ SD
A: Anterior TPI AR z ⫽ 1.46; P ⫽ 0.1440.
0.17 ⫾ 0.04 0.19 ⫾ 0.07
B: Middle TPI AR z ⫽ 0.19; P ⫽ 0.8510.
0.22 ⫾ 0.08 0.22 ⫾ 0.07
C: Posterior TPI AR z ⫽ 4.17; P ⫽ 0.0001.
0.23 ⫾ 0.13 0.18 ⫾ 0.06
impulses emitted by the stroboscopic source and the frequency of VF vibration.16 The absence of adequate synchronization between the light pulses and successive vibratory cycles results in the loss of sequential arrangement of the vibration phases with disturbance of the sharpness of the videostroboscopic images.17 We obtained videostroboscopic images that were characterized by a high degree of sharpness and synchronization between the light source and the vibratory cycles. This fact may be interpreted as clear evidence of the regularity of the vibratory cycle periods and was observed in all (100%) studied larynges submitted to the medialization procedures, even in the presence of a silicone block in the TPI. According to Moore and Thompson,18 the degree of periodic irregularity is directly proportional to the degree of hoarseness. The periodicity observed in our series proves that this characteristic is restored or preserved after TPI and AR, and that it may be considered as one of the determining factors of good vocal quality reported by different authors.19–22 Isshiki23 reported that hoarseness may be caused by both the absence of complete closure of the glottis and its excessive closure where the contact between the folds presents a certain degree of compression (initial glottic area ⬍0). In the present experiment, aperiodic vibratory cycles were noted when the depth of the silicone block was too great, causing excessive medialization of the VF. That was the
FIGURE 5. CT image obtained after thyroplasty type I on left side.
reason why further fine adjustments of the silicone block depth were made during the vibration. This fact should be emphasized in clinical practice when vocal adjustment during surgery is essential for preventing hypermedialization of the VF, which may result in aperiodic vibratory cycles and, consequently, in undesirable vocal quality. Based on the present observations, whenever possible the use of a videostroboscope through a nasofiberscope should be recommended during surgery to confirm cycle periodicity and to optimize vocal outcome. According to Hirano’s theory of body and cover,24 the difference in tissue consistency between the lining mucosa and the musculoligamentous body is of fundamental importance for the formation of the mucosal wave during vocal folds vibration. Fukuda et al25 experimentally demonstrated that muscle stimulation (with an increase in muscle rigidity) results in an increase in the mucosal vibration amplitude, confirming the importance of larger rigidity of the musculoligamentous body in the mucosal wave motion. As expected, the CT images obtained from five consecutively analyzed larynges clearly demonstrated that TPI provoked a significant increase in VF thickness (Figure 5) due to the compression of the thyroarytenoide muscle by the silicone block. The compression of the structures and the increase in VF volume lead to transverse stretching of the mucosa, resulting in alterations such as increased relative mass and higher degree of tension of the structures. Although the mechanism involved in this Journal of Voice, Vol. 17, No. 4, 2003
602
DOMINGOS H. TSUJI ET AL
transverse stretching is different from that involved in longitudinal stretching observed during emission of higher frequency sounds, the superficial layer of the lamina propria (Reinke’s space) should also become thinner due to the stretching and compression to which it is submitted. These alterations characterized by a higher tension at the mucosal level and by a decrease in the thickness of Reinke’s space may explain, in a way similar to what occurs during longitudinal stretching, the reduction in the mucosal wave observed in 70% of vocal folds submitted to TPI in comparison to the contralateral folds submitted to AR. However, one of the larynges showed an increased mucosal wave on the TPI side. This finding seems to be a paradox in view of the predominance of wave reduction in the other larynges, and we do not have a definitive explanation for it, but we may raise the hypothesis that the silicone implant might have increased the difference in rigidity between the body and the coverage. Tanabe et al26 and Isshiki et al27 observed VF vibration in the presence of tension asymmetry between folds characterized by the presence of phase asymmetry but without a difference in the frequencies between them. According to Isshiki et al,27 phase asymmetry is the most important sign of asymmetric tension. These findings have been confirmed by Moore et al28 in experimental studies on dogs and by Sercarz et al29 in clinical studies on 20 patients with VF paralysis. Isshiki30 reported that the same type of vibration, with phase asymmetry, can be observed in the presence of mass asymmetry between folds. In the present study, we observed phase asymmetry in 90% of the larynges. These findings seem to support our previous assumption that structural alterations caused by the silicone block in TPI result in tension and mass changes in the VF structure, a fact that does not occur, at least at the same intensity, in the AR. A rigid body present in the paraglottic space, medially displacing the VF structure, may behave as a blockage against lateral displacement during the opening phase of the vibratory cycle, reducing the vibration amplitude. However, subjective comparison of the vibration amplitude showed greater amplitude on the TPI side in 40% of the larynges, the same amplitude in 30%, and a smaller amplitude in only 30%. These results may appear to be even more Journal of Voice, Vol. 17, No. 4, 2003
incoherent if we consider that the mucosal wave on the TPI sides was found to be smaller or equal to the side of arytenoid rotation in nine of the ten larynges (smaller in seven larynges and equal in two). This apparent incoherence is very probably due to the difficulties met by the examiners during subjective analysis of amplitude. Isshiki et al23 also reported difficulties in defining the vibration amplitude in the presence of coaptation line displacement occurring during tension and phase asymmetry between folds. This difficulty was also encountered by Thompson et al21 when evaluating the vibration amplitude in patients submitted to TPI. To clarify these findings, we carried out objective analyses of the medial lateral displacement of the free border of the fold during vibration. No statistically significant differences in mean amplitudes were observed in the anterior and middle regions of the two vocal folds. However, the amplitude of the posterior region on the TPI side was significantly larger than that observed for the AR side. Very probably, this greater amplitude in the posterior region influenced the subjective analysis of the examiners at the time when they classified the amplitude on the TPI side as being larger than that on the AR side in 40% of the larynges. The larger amplitude detected in the posterior region also explains the results observed by subjective analysis of the free border of the vocal folds where eight larynges with a sigmoid profile were found on the TPI side and ten with a semi-spindle-shaped profile on the AR side. Lateral displacement of the VF with a sigmoid profile may be described as larger lateral movement of the posterior region of the VF, which even occurs after the anterior region has reached its maximum lateralization. We presumed that both the sigmoid profile on the fold submitted to TPI and the larger mean amplitude observed in its posterior region are the consequence of a larger airflow directed to the posterior region of the VF on the TPI side due to an increased VF resistance in its anterior two-thirds caused by the rigidity of the silicone block. As emphasized by different authors,24,31,32 the mucosal layer (cover) plays a fundamental role in VF vibration. Alterations in its characteristics result in vibratory disturbances with a consequent impact on sound production. Studies by Gray et al33 and Lu
THYROPLASTY TYPE I AND ARYTENOID ROTATION et al34 have demonstrated that, despite a significant vocal improvement upon TPI, some parameters differed significantly from normal patterns. Gray et al33 observed that vocal characteristics such as the degree of tension, breathiness, hoarseness, harshness, and unsteadiness were significantly altered compared to normality. Although the objective of the present study was not to establish a correlation between each type of vibratory alteration observed and the vocal acoustic changes, we believe that alterations such as strain, hoarseness, and unsteadiness observed by Gray et al33 in patients after TPI are the result of some vibratory abnormality such as a decrease in mucosal wave and asymmetry in vibration amplitude and phase.
CONCLUSION Based on the present experimental study, we conclude that TPI and AR are able to preserve the periodicity of VF vibration. However, the implanted silicone block in TPI, leading to increased vocal folds thickness and stretching of the mucosal membrane, causes more alterations in vibration parameters such as mucosal wave motion, profile of VF free border during the opening phase, and amplitude and phase symmetry than in the AR procedure.
REFERENCES 1. Hill DP, Meyers AD, Harris J. Autologous fat injection for vocal cord medialization in the canine larynx. Laryngoscope. 1991;101:344–348. 2. Ford CN, Martin DW, Warner TF. Injectable collagen in laryngeal rehabilitation. Laryngoscope. 1984;94:513–518. 3. Isshiki N, Morita H, Okamura H, Hiramoto M. Thyroplasty as a new phonosurgical technique. Acta Otolaryngol (Stockh). 1974;78:451–457. 4. Isshiki N, Tanabe M, Sawada M. Arytenoid adduction for unilateral vocal cord paralysis. Arch Otolaryngol. 1978; 104:555–558. 5. Stone JW, Arnold GE, Stephens CB. Intracordal polytef (Teflon) injection. Histologic study of three further cases. Arch Otolaryngol. 1970;91:568–574. 6. Wenig BM, Heffner DK, Oertel YC, Johnson FB. Teflonomas of the larynx and neck. Human Pathol. 1990;21: 617–623.
603
7. Rubin HJ. Misadventures with injectable polytef (Teflon). Arch Otolaryngol. 1975;101:114–116. 8. Ossoff RH, Koriwchak MJ, Netterville JM, Duncavage JA. Difficulties in endoscopic removal of Teflon granulomas of the vocal fold. Ann Otol Rhinol Laryngol. 1993; 102:405–412. 9. Kasperbauer JL, Slavit DH, Maragos NE. Teflon granulomas and overinjection of Teflon: a therapeutic challenge for the otorhinolaryngologist. Ann Otol Rhinol Laryngol. 1993;102:748–751. 10. Varvares MA, Montgomery WW, Hillman RE. Teflon granuloma of the larynx: etiology, pathophysiology, and management. Ann Otol Rhinol Laryngol. 1995;104:511–515. 11. Lablance GR, Maves MD. Acoustic characteristics of postthyroplasty patients. Otolaryngol Head Neck Surg. 1992; 107:558–563. 12. D’Antonio LL, Wigley TL, Zimmerman GJ. Quantitative measures of laryngeal function following Teflon injection or thyroplasty type I. Laryngoscope. 1995;105:256–262. 13. Isshiki N, Taira T, Kojima H, Shoji K. Recent modifications in thyroplasty type I. Ann Otol Rhinol Laryngol. 1989; 98:777–779. 14. Van Den Berg J, Tan TS. Results of experiments with human larynxes. Pract Otorhinolaryngol. 1959;21:425–450. 15. Woodson GE. Configuration of the glottis in laryngeal paralysis I: clinical study. Laryngoscope. 1993;103:1227–1234. 16. Kallen LA. Laryngostroboscopy in the practice of otolaryngology. Arch Otolaryngol. 1932;16:791–807. 17. Hirano M, Bless DM. Vocal fold vibration. In: Hicano M, Bless DM, eds. Videostroboscopic examination of the larynx. San Diego, CA: Singular Publishing Group Inc; 1993:23–36. 18. Moore P, Thompson CL. Comments on physiology of hoarseness. Arch Otolaryngol. 1965;81:97–102. 19. Isshiki N, Okamura H, Ishikawa T. Thyroplasty type I (lateral compression) for dysphonia due to vocal cord paralysis or atrophy. Acta Otolayngol (Stockh). 1975;80:465–473. 20. Koufman JA. Laryngoplasty for vocal cord medialization: an alternative to Teflon. Laryngoscope. 1986;96:726–731. 21. Thompson DM, Maragos NE, Edwards BW. The study of vocal fold vibratory patterns in patients with unilateral vocal fold paralysis before and after type I thyroplasty with or without arytenoid adduction. Laryngoscope. 1995;105: 481–486. 22. Bielamowicz S, Gupta A, Sekhar LN. Early arytenoid adduction for vagal paralysis after skull base surgery. Laryngoscope. 2000;110:346–351. 23. Isshiki N, Tanabe M, Ishizaka K, Broad D. Clinical significance of asymmetrical vocal cord tension. Ann Otol Rhinol Laryngol. 1977;86:58–66. 24. Hirano M. Morphological structure of the vocal cord as a vibrator and its variations. Folia Phoniatr. 1974;26:89–94. 25. Fukuda H, Saito S, Kitahara S, Isogai Y, Makino K, Tsuzuki T, Kogawa N, Ono H. Vocal fold vibration in excised larynges viewed with an X-ray stroboscope and an ultrahigh-speed camera. In: Titze ER, Scherer RC, eds. Vocal Journal of Voice, Vol. 17, No. 4, 2003
604
26.
27.
28.
29.
DOMINGOS H. TSUJI ET AL
fold physiology. San Diego, CA: College-Hill Press; 1983: 238–252. Tanabe M, Isshiki N, Kitajima K. Vibratory pattern of the vocal cord in unilateral paralysis of the cricothyroid muscle. An experimental study. Acta Otolaryngol. 1972; 74:339–345. Isshiki N. Functional surgery of the larynx. Department of Otolaryngology, Kyoto University Medical School; 1977:207. [Special report presented at 78th Congress of Japan Otolaryngology Society, Fukuoka, Japan, 1977]. Moore DM, Berke GS, Hanson DG, Ward PH. Videostroboscopy of the canine larynx: the effects of asymmetric laryngeal tension. Laryngoscope. 1987;97:543–553. Sercarz JA, Berke GS, Ming Y, Gerratt BR, Natividad M. Videostroboscopy of human vocal fold paralysis. Ann Otol Rhinol Laryngol. 1992;101:567–577.
Journal of Voice, Vol. 17, No. 4, 2003
30. Isshiki N. Physiology of speech production. In: Isshiki N, ed. Phonosurgery: theory and practice. Tokyo: SpringerVerlag; 1989:5–21. 31. Saito S, Fukuda H, Isogai Y, Ono H. X-ray stroboscopy. In: Stevens KN, Hirano M, eds. Vocal fold physiology. Tokyo: University of Tokyo Press; 1981:95–106. 32. Tsuzuki T. A basic study on the vocal fold vibration from the viewpoint of the layer system of the vocal fold. Otologia (Fukuoka). 1984;30:131–152. 33. Gray SD, Barkmeier J, Jones D, Titze I, Druker D. Vocal evaluation of thyroplastic surgery in the treatment of unilateral vocal fold paralysis. Laryngoscope. 1992;102: 415–421. 34. Lu FL, Casiano RR, Lundy DS, Xue JW. Longitudinal evaluation of the vocal function after thyroplasty type I in the treatment of unilateral vocal paralysis. Laryngoscope. 1996;106:573–577.