[Skip to Content]
Access to paid content on this site is currently suspended due to excessive activity being detected from your IP address 54.163.129.96. Please contact the publisher to request reinstatement.
[Skip to Content Landing]
Download PDF
Figure 1.

 Left vocal fold immobility in subject 37. Computed tomographic images during inspiration (A) and phonation (B), and coronal multiplanar reconstructed images during inspiration (C, left) and phonation (C, right). A and B, Upper left, oral view; upper right, tracheal view; lower left, left hemilarynx viewed from right side; and lower right, right hemilarynx. R indicates right; L, left. Mild staircase motion artifacts were seen in the coronal multiplanar reconstruction image during phonation. Note the abduction and thinning of the left vocal fold during phonation (paradoxical movement).

Left vocal fold immobility in subject 37. Computed tomographic images during inspiration (A) and phonation (B), and coronal multiplanar reconstructed images during inspiration (C, left) and phonation (C, right). A and B, Upper left, oral view; upper right, tracheal view; lower left, left hemilarynx viewed from right side; and lower right, right hemilarynx. R indicates right; L, left. Mild staircase motion artifacts were seen in the coronal multiplanar reconstruction image during phonation. Note the abduction and thinning of the left vocal fold during phonation (paradoxical movement).

Figure 2.

 Sulcus vocalis and normal vocal fold movement in a control subject (55-year-old man). A-C show computed tomographic endoscopic and coronal multiplanar reconstruction images arranged in the same manner as in Figure 1. R indicates right; L, left.

Sulcus vocalis and normal vocal fold movement in a control subject (55-year-old man). A-C show computed tomographic endoscopic and coronal multiplanar reconstruction images arranged in the same manner as in Figure 1. R indicates right; L, left.

Figure 3.

 Left vocal fold immobility in subject 19. A-C show computed tomographic endoscopic and coronal multiplanar reconstruction images arranged in the same manner as in Figure 1. Although paradoxical movement was not apparent in the images viewed from the oral and tracheal sides, the vocal fold on the affected side became thinner during phonation than during inspiration. Overadduction of the right vocal fold during phonation was seen. R indicates right; L, left.

Left vocal fold immobility in subject 19. A-C show computed tomographic endoscopic and coronal multiplanar reconstruction images arranged in the same manner as in Figure 1. Although paradoxical movement was not apparent in the images viewed from the oral and tracheal sides, the vocal fold on the affected side became thinner during phonation than during inspiration. Overadduction of the right vocal fold during phonation was seen. R indicates right; L, left.

Figure 4.

 Left vocal fold immobility in subject 34. The top and bottom rows show images during inspiration and phonation, respectively. The left and right columns show images viewed from the oral and tracheal sides, respectively. Note the prominent overadduction of the right vocal fold during phonation. R indicates right; L, left.

Left vocal fold immobility in subject 34. The top and bottom rows show images during inspiration and phonation, respectively. The left and right columns show images viewed from the oral and tracheal sides, respectively. Note the prominent overadduction of the right vocal fold during phonation. R indicates right; L, left.

Figure 5.

 Left vocal fold immobility in subject 2. A, Computed tomographic endoscopic and coronal multiplanar reconstruction (MPR) images during inspiration (upper row) and phonation (lower row): Left, oral view; center, tracheal view; right, coronal MPR images. Note the presence of paradoxical movement of the left vocal fold (the abduction and thinning of the left vocal fold during phonation). B, Stroboscopic images of the larynx during inspiration (left) and phonation (right). R indicates right; L, left.

Left vocal fold immobility in subject 2. A, Computed tomographic endoscopic and coronal multiplanar reconstruction (MPR) images during inspiration (upper row) and phonation (lower row): Left, oral view; center, tracheal view; right, coronal MPR images. Note the presence of paradoxical movement of the left vocal fold (the abduction and thinning of the left vocal fold during phonation). B, Stroboscopic images of the larynx during inspiration (left) and phonation (right). R indicates right; L, left.

Table 1.  
Profiles of the 37 Subjects
Profiles of the 37 Subjects
Table 2.  
Computed Tomography Endoscopic and Coronal Multiplanar Reconstruction Image Analysis Results
Computed Tomography Endoscopic and Coronal Multiplanar Reconstruction Image Analysis Results
Table 3.  
Videostroboscopic Analysis Results*
Videostroboscopic Analysis Results*
1.
Isshiki  NOkamura  HIshikawa  T Thyroplasty type I (lateral compression) for dysphonia due to vocal cord paralysis or atrophy. Acta Otolaryngol.1975;80:465-473.
PubMed
2.
Isshiki  NTanabe  MSawada  M Arytenoid adduction for unilateral vocal cord paralysis. Arch Otolaryngol.1978;104:555-558.
PubMed
3.
Isshiki  NTaira  TKojima  HShoji  K Recent modifications in thyroplasty type I. Ann Otol Rhinol Laryngol.1989;98:777-779.
PubMed
4.
Maragos  N Type I thyroplasty: pitfalls of modifying the Isshiki approach. J Voice.1997;11:470-473.
PubMed
5.
Remacle  MDujardin  JMLawson  G Treatment of vocal fold immobility by glutaraldehyde-cross-linked collagen injection: long-term results. Ann Otol Rhinol Laryngol.1995;104:437-441.
PubMed
6.
Shaw  GYSzewczyk  MASearle  JWoodroof  J Autologous fat injection into the vocal folds: technical considerations and long-term follow-up. Laryngoscope.1997;107:177-186.
PubMed
7.
Tucker  H Laryngeal reinnervation for unilateral vocal cord paralysis. Ann Otol Rhinol Laryngol.1981;90:457-459.
PubMed
8.
Crumley  RLIzdebski  KMcMicken  B Nerve transfer versus Teflon injection for vocal cord paralysis: a comparison. Laryngoscope.1988;98:1200-1204.
PubMed
9.
Yumoto  ESanuki  THyodo  M Three-dimensional endoscopic images of vocal fold paralysis by computed tomography. Arch Otolaryngol Head Neck Surg.1999;125:883-890.
PubMed
10.
Inagi  KKhidr  AAFord  CNBless  DMHeisey  DM Correlation between vocal functions and glottal measurements in patients with unilateral vocal fold paralysis. Laryngoscope.1997;107:782-791.
PubMed
11.
Newmark  GMConces Jr  DJKopecky  KK Spiral CT evaluation of the trachea and bronchi. J Comput Assist Tomogr.1994;18:552-554.
PubMed
12.
Silverman  PMZeiberg  ASSessions  RBTroost  TRZeman  RK Three-dimensional imaging of the hypopharynx and larynx by means of helical (spiral) computed tomography: comparison of radiological and otolaryngological evaluation. Ann Otol Rhinol Laryngol.1995;104:425-431.
PubMed
13.
Zeiberg  ASSilverman  PMSessions  RBTroost  TRDavros  WJZeman  RK Helical (spiral) CT of the upper airway with three-dimensional imaging: technique and clinical assessment. AJR Am J Roentgenol.1996;166:293-299.
PubMed
14.
Yumoto  ESanuki  THyodo  MYasuhara  YOchi  T Three-dimensional endoscopic mode for observation of laryngeal structures by helical computed tomography. Laryngoscope.1997;107:1530-1537.
PubMed
15.
Morinaka  SMiyamoto  SHidaka  A Helical CT scanning of laryngeal deviation. Auris Nasus Larynx.2001;28:151-159.
PubMed
16.
Isshiki  NIshikawa  T Diagnostic value of tomography in unilateral vocal cord paralysis. Laryngoscope.1976;86:1573-1578.
PubMed
17.
Hong  KHJung  KS Arytenoid appearance and vertical level difference between paralyzed and innervated vocal cords. Laryngoscope.2001;111:227-232.
PubMed
18.
von Leden  HMoore  P The mechanics of the cricoarytenoid joint. Arch Otolaryngol.1961;73:541-550.
PubMed
19.
Woodson  GEHengesteg  ARosen  CAYeung  DChen  N Changes in length and spatial orientation of the vocal fold with arytenoid adduction in cadaver larynges. Ann Otol Rhinol Laryngol.1997;106:552-555.
PubMed
20.
Kumakawa  KYamaguchi  HHirose  H Image analysis of the contour of paralyzed vocal cord during phonation. Nippon Jibiinkoka Gakkai Kaiho.1990;93:61-70.
PubMed
21.
Hiroto  IHirano  MTomita  H Electromyographic investigation of human vocal cord paralysis. Ann Otol Rhinol Laryngol.1968;77:296-304.
PubMed
22.
Blitzer  AJahan  AFKeidar  A Semon's law revisited: an electromyographic analysis of laryngeal synkinesis. Ann Otol Rhinol Laryngol.1996;105:764-769.
PubMed
23.
Crumley  RL Laryngeal synkinesis revisited. Ann Otol Rhinol Laryngol.2000;109:365-371.
PubMed
24.
Omori  KSlavit  DHKacker  ABlaugrund  SM Quantitative criteria for predicting thyroplasty type I outcome. Laryngoscope.1996;106:689-693.
PubMed
25.
Tanaka  SChijiwa  KHirano  M Study on over-adduction of unaffected vocal fold in unilateral recurrent laryngeal nerve paralysis. Larynx Jpn.1993;5:135-141.
Original Article
August 2004

Three-dimensional Characteristics of the Larynx With Immobile Vocal Fold

Author Affiliations

From the Departments of Otolaryngology–Head and Neck Surgery (Drs Yumoto, Oyamada, and Nakano) and Radiology (Drs Nakayama and Yamashita), Graduate School of Medicine, Kumamoto University, Kumamoto, Japan. The authors have no relevant financial interest in this article.

Arch Otolaryngol Head Neck Surg. 2004;130(8):967-974. doi:10.1001/archotol.130.8.967
Abstract

Objectives  To evaluate the 3-dimensional (3-D) characteristics of the laryngeal lumen in patients with unilateral vocal fold immobility (UVFI) during phonation with the aid of multislice helical computed tomography (MSHCT).

Design  A retrospective study.

Setting  University hospital.

Subjects  Thirty-seven patients with UVFI.

Interventions  Each subject was asked to sustain the vowel /a/ and then to inhale slowly. The region over the larynx was scanned using MSHCT during each maneuver for 5 seconds; 3-D endoscopic images and coronal multiplanar reconstruction images were produced and evaluated. Thirty-two subjects underwent videostroboscopy within 2 weeks of the MSHCT.

Main Outcome Measures  Presence of thinning and paradoxical movement of the affected vocal fold, overadduction of the healthy fold, and vertical positional difference between the vocal folds during phonation were assessed based on 3-D and multiplanar reconstruction images.

Results  During phonation, the affected vocal fold was thinner in 31 subjects and was situated in a higher position in 21 subjects than the healthy fold. In 4 subjects, the affected vocal fold showed paradoxical movement and 3 other subjects had probable paradoxical movement. Overadduction of the healthy vocal fold occurred during phonation in 15 subjects. Videostroboscopy detected paradoxical movement in 2 of the 3 subjects with abduction of the affected vocal fold during phonation based on 3-D images, and overadduction in all 13 subjects examined.

Conclusions  The combination of 3-D endoscopy with coronal multiplanar reconstruction images enables description of the 3-D characteristics of the unilaterally immobile larynx and supplements videostroboscopic findings exemplified by differences in vertical position and thickness between the vocal folds.

Patients with unilateral vocal fold paralysis complain of breathy dysphonia and mild aspiration. Both vocal folds must meet at the midline and be of equal thickness and tension to produce symmetrical oscillation during phonation. However, a paralyzed vocal fold may be fixed at a midline, paramedian, or intermediate position. In addition, a paralyzed vocal fold is often thinner and less tense than a healthy vocal fold. Furthermore, a difference in the vertical position of the 2 vocal folds usually exists during phonation. Various phonosurgical procedures18 aim at correcting such asymmetry between the 2 vocal folds. Since electromyography was not performed to confirm the presence of vocal fold paralysis, the situation is called unilateral vocal fold immobility (UVFI) in the present study.

Videostroboscopy has been the method of choice for observing the vocal folds before planning phonosurgery. In most patients, this examination reveals the horizontal position of the immobile vocal fold and the presence or absence of flaccidity of the vocal fold on the affected side, and a glottal gap during phonation. However, it cannot assess the vertical difference in the position of the 2 vocal folds or their thicknesses (bulk). In addition, overadduction of the ventricular folds and anteroposterior shortening of the glottis may prevent visualization of the vocal fold from the oral side.

Using helical scanning high-resolution computed tomography (CT) with a 3-dimensional (3-D) reconstruction technique, Yumoto et al9 described the characteristics of 3-D CT endoscopic images of the unilaterally paralyzed larynx and the changes in the affected vocal fold after phonosurgery. Although they reported the 3-D characteristics of the unilaterally paralyzed larynx, they did not observe the larynx during phonation because their patients were required to hold their breath for approximately 30 seconds for scanning. The recent development of multislice helical CT (MSHCT) allows high-speed data acquisition without affecting final image resolution. Therefore, this study was designed to analyze the laryngeal lumen in 3-D during phonation in patients with UVFI. This study assessed the differences in the 3-D characteristics of the vocal folds, and the presence or absence of abduction of the vocal fold on the affected side ("paradoxical movement") and overadduction of the vocal fold on the healthy side during phonation. We evaluated 3-D endoscopic images combined with coronal images produced by the multiplanar reconstruction (MPR) method. Finally, the 3-D assessment of the immobile larynx was compared with videostroboscopic findings.

METHODS
SUBJECTS

The subjects included 24 men and 20 women with UVFI who underwent MSHCT examination during inspiration and phonation between April 1999 and March 2001 to produce 3-D endoscopic images. In addition, a 55-year-old man with sulcus vocalis who did not have UVFI was retrospectively selected to serve as a control. All but 3 subjects (4, 8, and 10) had no history of surgical intervention for UVFI. During the examination, each subject was asked to inhale slowly for 5 seconds and then sustain the vowel /a/ for 5 seconds after a short rest. Since patients with UVFI consume a greater amount of air to phonate, they often cannot produce sustained phonation for 5 seconds. In such cases, we asked them to phonate in as quiet a voice as possible throughout the entire scan. Even with practice, 4 men and 3 women could neither inhale nor phonate during the scan and were excluded from the study. Therefore, the subjects consisted of 20 men and 17 women ranging in age from 31 to 85 years with a mean age of 62.6 years. The time since the onset of UVFI ranged from 3 weeks to 20 years with a mean of 34.6 months. The clinical profiles of the subjects are shown in Table 1.

PRODUCTION OF 3-D ENDOSCOPIC AND CORONAL MPR IMAGES

The subjects lay in the supine position with a pillow under their necks to avoid overlapping of the mandible above the larynx. All scans were performed with an MSHCT (HiSpeed Advantage QX/i; General Electric) using 200 mA, 120 kV, 1.25-mm slice thickness, 1 x-ray tube rotation/0.8 seconds, and 7.5 mm/0.8 seconds table speed. Each subject was scanned during phonation and inspiration. The scan covered the range from the protrusion of the superior thyroid notch to 37.5 mm below this point and included the ventricular fold, ventricle, vocal fold, and subglottic region. All CT parameters and 3-D rendering techniques were determined in a conference with radiologists (Y.N. and Y.Y.).

Axial images were generated at 1-mm intervals by overlapping each adjacent image by 0.5 mm, and were then sent to a workstation (Advantage, General Electric) together with voxel data. A volume-rendering technique was used with 3-D endoscopic mode. In our study, 2 voxels with −100 or lower Hounsfield units served as the surface of the laryngeal lumen. Voxels other than the selected voxels were eliminated. Views from the tracheal and oral sides, and 2 vertically split hemilaryngeal images were displayed. Coronal 2-mm-thick MPR images perpendicular to the glottic axis were systematically built in all subjects.

VIDEOSTROBOSCOPY

Videostroboscopic examination was performed during sustained phonation of the vowel /e/ or /i/ at a comfortable pitch and loudness in 32 subjects within 2 weeks of the CT scan. The remaining 5 subjects failed to visit the outpatient clinic within 2 weeks of the CT scan. A laryngostroboscope (LS-3A; Nagashima, Tokyo, Japan) with an oblique-view rigid endoscope (SFT-1; Nagashima) connected to a charge-coupled device camera (MV-596; Stryker, Santa Clara, Calif) was used. The images were recorded on digital videotapes (DVCPRO AJ-D230H; Panasonic, Tokyo).

ANALYSIS

The presence or absence of "staircase" motion artifacts was evaluated on CT endoscopic images to exclude subjects whose images could not be used for further analysis. Two of us (E.Y. and Y.O.) evaluated the CT endoscopic and coronal MPR images of each patient in random order for the presence or absence of thinning and paradoxical movement of the affected vocal fold, and overadduction of the healthy vocal fold during sustained phonation. We also evaluated vertical positional differences between the 2 vocal folds during phonation. Thinning was defined as a decrease in the vertical thickness of the affected fold during phonation in comparison with the healthy side. Paradoxical movement was defined as abduction and thinning of the affected fold during phonation, and adduction and thickening of the affected fold during inspiration. Abduction and adduction were evaluated from images viewed from the oral and tracheal sides, while thickness was evaluated from coronal MPR images. Overadduction was defined as adduction of the healthy vocal fold beyond the glottic axis during phonation. Figure 1A-C shows CT endoscopic and coronal MPR images of subject 37, which show staircase motion artifacts during phonation, and thinning and paradoxical movement of the affected vocal fold.

Two of us (E.Y. and Y.O.) evaluated the video recordings of 32 subjects in random order for the presence or absence of overadduction of the ventricular folds and anteroposterior shortening of the glottis during phonation, for paradoxical movement of the affected vocal fold, and for overadduction of the healthy vocal fold during phonation.

Subsequently, the 3-D assessment of the unilaterally immobile larynx was compared with videostroboscopic findings.

RESULTS

Figure 2A-C shows CT endoscopic and coronal MPR images of the control subject who did not have UVFI. Abduction of the vocal fold during inspiration and adduction during phonation were apparent on the CT endoscopic images. The vocal processes met at the midline during phonation. Coronal MPR images showed symmetrical thickness of the vocal folds during phonation.

The results of the CT endoscopic and coronal MPR image analysis are summarized in Table 2. In all patients who could sustain their phonation during scanning, artifacts due to motion or beam hardening were not seen in axial images. Although minimal staircase motion artifacts (eg, Figure 1) were seen in 5 subjects on MPR images, such artifacts did not interfere with further evaluation of these subjects. The vocal fold on the affected side during phonation was thinner in 31 subjects than that on the healthy side, while the thickness of both vocal folds was equal in the remaining 6 subjects. Compared with the healthy side, the vocal fold on the affected side was located at a higher position during phonation in 21 subjects, at a lower position in 8 subjects, and at the same vertical position in 8 subjects.

Paradoxical movement was detected in 4 of the 37 subjects. Coronal MPR images of 2 other subjects (subjects 15 and 19) showed that the vocal fold on the affected side became thinner during phonation than during inspiration, suggesting the presence of paradoxical movement, although no such movement was apparent on oral and tracheal views (Figure 3). By contrast, CT endoscopic images of subject 33 showed abduction of the affected vocal fold during phonation, although the thickness of the affected vocal fold was not increased during inspiration based on coronal MPR images. Overadduction of the vocal fold on the healthy side over the midline during phonation was seen in 15 subjects. Figure 3 shows slight overadduction of the vocal fold on the healthy side. Subject 34 showed prominent overadduction of the vocal fold on the healthy side during phonation, as illustrated in Figure 4.

The results of the videostroboscopic analysis of 32 patients are summarized in Table 3. Although overadduction of the ventricular fold, anteroposterior shortening of the glottis, or both occurred during videostroboscopy in 20 subjects, such strangulation of the laryngeal inlet was only severe enough to interfere with further evaluation in 1 of these subjects (subject 7). Paradoxical movement could not be assessed in subjects 10 and 37 because the vocal fold and arytenoid mound on the healthy side overlapped those on the affected side, interfering with visualization of the vocal fold on the affected side during and at the beginning of phonation, respectively. Two of 29 subjects (subjects 33 and 35) showed paradoxical movement of the affected vocal fold. Thirteen of 31 subjects showed overadduction of the vocal fold on the healthy side during phonation.

COMMENT

Videostroboscopy has been the standard means of evaluating vocal fold vibration in patients with UVFI. This examination reveals the horizontal position of the paralyzed vocal fold (deviation from the glottic axis), shape and size of the glottal gap during phonation, degree of the bowing of the vocal fold, and presence or absence of compensatory overadduction of the healthy vocal fold during phonation. Although the vertical thickness and subglottal shape of the vocal fold are also important in determining the convergence of the vocal folds during phonation,10 videostroboscopy through an endoscope is unable to provide depth information. In addition, the larynx of some patients cannot be observed through an endoscope because of overadduction of the ventricular fold, anteroposterior shortening of the glottis, or both. In our series, such strangulation of the laryngeal inlet occurred in 20 of 32 subjects and the paralyzed vocal fold was not visible in 1 (subject 7). In 2 other subjects (subjects 10 and 37), paradoxical movement was not assessed because of lack of visualization of the affected vocal fold during phonation (Table 3).

Three-dimensional reconstruction techniques have been used to demonstrate lesions of luminal organs, such as the lower and upper airways.1113 However, these applications were limited to detecting the growth of neoplasms and cicatricial strictures. Yumoto et al14 first succeeded in depicting the laryngeal structures in 3-D, including the vocal fold, ventricular fold, and ventricle. Subsequently, they proved that 3-D CT endoscopy enables description of the 3-D characteristics of unilateral vocal fold paralysis,9 although they observed the laryngeal lumen while the patient held his or her breath. Morinaka et al15 observed 3-D endoscopic laryngeal images together with 3-D laryngeal framework images in patients with deviation of the larynx. These reports did not describe the characteristic behavior of the laryngeal lumen in relation to phonation. In this study, we scanned the whole larynx for 5 seconds, and could produce 3-D and coronal MPR images during both inspiration and phonation in patients with UVFI. In CT endoscopic images, artifacts due to motion, beam hardening, and partial volume averaging may degrade image quality and potentially lead to misinterpretation. However, thanks to high-speed data acquisition (<1 second for gantry rotation and 5 seconds for total data acquisition) and thin slice thickness on MSHCT, these artifacts were negligible. The patients were asked to practice phonating in as quiet a voice as possible to sustain the phonation for 5 seconds. However, 7 of the 44 enrolled patients could not sustain phonation for 5 seconds, or could not inhale slowly during the scan. Consequently, 37 patients were included in the study.

Multislice helical computed tomography is an invasive technique because radiation is given to a patient. The CT dose index (CTDi) was calculated to estimate an approximate dose of irradiation during the scan of the larynx in the present study. The total of CTDi values of the laryngeal scanning during inspiration and phonation was 25.76 mGy. The value was compared with that of scanning of other regions in the head and neck. The CTDi values of the scannings over paranasal sinuses and temporal bone based on the settings at the Kumamoto University Hospital were 30.25 mGy and 63.57 mGy, respectively. The scanning of the temporal bone required as thin a slice as 0.5-mm thickness and its CTDi was more than twice as that of the laryngeal scanning in the present study. Although CTDi of the laryngeal scanning is less than that of paranasal sinuses and temporal bone, we should be careful in applying the laryngeal scanning to a patient with UVFI.

We examined the 3-D characteristics of the unilaterally paralyzed larynx during phonation with the aid of MSHCT. Although minimal staircase motion artifacts were seen in 5 subjects, such artifacts did not interfere with further evaluation. Consequently, differences in the vertical thickness and location of both vocal folds during phonation, the presence of paradoxical movement of the affected vocal fold, and the presence of overadduction of the healthy vocal fold were evaluated (Table 2).

The vocal fold on the affected side was thinner in 31 subjects than that on the healthy side, while both vocal folds were of similar thickness in the remaining 6. Of these 6 subjects, subject 3 recovered from UVFI 1 month after the scan; subjects 11 and 14 showed weak movement of the affected vocal fold; and subjects 9 and 21 had undergone resection of meningioma in the jugular foramen and thyroid cancer 10 months and 19 years before scanning, respectively. Although electromyography was not performed in these 2 subjects, some regeneration of nerve fibers in the recurrent laryngeal nerve is possible after onset of UVFI. The left vocal fold in subject 17 was immobile for 2 months and scanning was performed the day before thyroidectomy. Although the surgical record indicated that her left recurrent laryngeal nerve was trapped in the tumor, some of the nerve fibers might have been functional. Electromyography should be performed to determine the neurophysiological situation in these exceptional cases.

The vocal fold on the affected side was located in the higher position during phonation in 21 subjects (57%), and in the lower position in 8 subjects. In 8 subjects, there was no vertical difference in position. Using conventional tomography during phonation, Isshiki and Ishikawa16 reported that the paralyzed vocal fold was higher than the intact fold in 26 (46%) of 56 patients with UVFI, while there was no difference in the remaining patients. Hong and Jung17 reported that videostroboscopy revealed that the paralyzed vocal fold was not higher than the innervated vocal fold in most cases during phonation. There are marked differences in the vertical position of the vocal fold on the affected side compared with that on the healthy side during phonation within patients. Based on an anatomical study, von Leden and Moore18 reported that the vocal process lowers with adduction of the vocal cord. Using cadaver larynges, Woodson et al19 demonstrated that the vocal process consistently moved caudally following arytenoid adduction, by an average of 3.5 mm. The vertical level of the vocal process on the affected side depends on the position, rotated status (medial or lateral), and rocked status (anterior or posterior) of the arytenoid cartilage. Using a digital radiography system, Kumakawa et al20 evaluated the vertical difference in the vocal folds of 35 patients with UVFI whose affected vocal folds were fixed in either the median or paramedian position. They found that a cranial shift of the affected vocal fold occurred in 16 patients (46%) during phonation compared with its vertical level during inspiration, and suggested that subglottal air pressure is one of the factors causing the higher position of the vocal fold on the affected side during phonation.

Paradoxical movement of the paralyzed vocal fold was observed in 4 subjects based on CT endoscopic and coronal MPR image analysis. Three other subjects (15, 19, and 33) had probable paradoxical movement. Coronal MPR images of subjects 15 and 19 showed that the vocal fold on the affected side was thicker during inspiration than during phonation, although CT endoscopic images viewed from the oral and tracheal sides did not indicate apparent abduction of the affected vocal fold during phonation. On the contrary, although the thickness of the affected vocal fold was not increased during inspiration based on coronal MPR images, CT endoscopic images of subject 33 showed abduction of the affected vocal fold during phonation. With an increase in subglottal air pressure during phonation, the tensionless vocal fold on the affected side might have been blown upward as well as laterally. Therefore, we defined paradoxical movement as abduction and thinning of the affected fold during phonation, and adduction and thickening of the affected fold during inspiration.

Hiroto et al21 reported that the electrical potentials in immobile vocal folds suggested the presence of synkinesis. Blitzer et al22 obtained electromyograms from 14 patients with persistent vocal fold immobility, and found that 7 patients with good voice had a relatively normal arytenoid position and evidence of synkinesis, while those with poor voice had their arytenoid tipped into the laryngeal inlet and denervation or poor reinnervation. Crumley23 classified the functional status of laryngeal synkinesis following recurrent laryngeal nerve injury into 4 categories: from type 1 with satisfactory voice and airway to type 4 with hyperabducted vocal folds, poor voice, and possible aspiration. In types 3 and 4 synkinesis, the affected vocal fold may abduct during phonation and the membranous portion may bulge medially (increase the vertical thickness of the vocal fold) during inspiration. He considered synkinetic imbalance between the adductor and abductor muscles to be the most likely cause of a posterior glottal gap during phonation. Omori et al24 reported that vocal function following type 1 thyroplasty was significantly worse in patients with UVFI who had a relatively larger posterior glottal gap during phonation than in other patients. Therefore, the size of the glottal gap and the presence of paradoxical movement are important factors that must be assessed before phonosurgical treatment.

Overadduction of the vocal fold on the healthy side over the midline during phonation was observed in 15 (40%) of our 37 subjects based on CT endoscopic and coronal MPR image analysis. Tanaka et al25 examined 120 patients with UVFI stroboscopically and found that 51 (42.5%) had overadduction of the healthy vocal fold during phonation. They also reported that closure of the posterior glottis was imperfect in all patients with overadduction, and that the vocal function was worse in patients with overadduction of the healthy vocal fold than in those without overadduction. They concluded that overadduction of the healthy vocal fold is not a compensatory behavior, and that it tends to occur when the affected vocal fold is fixed away from the midline. In our study, the vocal fold on the affected side in 13 of the 15 subjects in whom overadduction of the healthy vocal fold occurred was thinner during phonation than that on the healthy side, and 7 of the 15 subjects showed apparent or probable paradoxical movement on the affected side. The presence of overadduction of the healthy vocal fold during phonation needs to be assessed before phonosurgical treatment because type 1 thyroplasty alone may not significantly improve hoarseness.

Multislice helical computed tomography detected paradoxical movement of the affected vocal fold in 4 subjects and probable paradoxical movement in 3 other subjects. In these 7 subjects, varying degrees of strangulation of the laryngeal inlet occurred during stroboscopic observations, and the affected vocal fold could not be visualized during phonation in subjects 10 and 37. Two of the remaining 5 subjects (subjects 33 and 35) were assessed as having paradoxical movement of the affected vocal fold in the videostroboscopic examination. The result that stroboscopic observations did not indicate the presence of paradoxical movement in subjects 15 and 19 corresponded to the results of CT endoscopic and coronal MPR image analysis as described previously. Figure 5 shows CT endoscopic and coronal MPR images, and stroboscopic images of subject 2. Adduction of the left vocal fold during inspiration and abduction during phonation were apparent on the CT endoscopic images. The coronal MPR images showed that vertical thickness of the left vocal fold was greater during inspiration than during phonation. On the contrary, careful repeated observation of stroboscopic recording was not able to detect the presence of paradoxical movement of the left vocal fold in this patient. One possible reason for this discrepancy is that the cartilaginous part of the vocal fold including the vocal process was not visualized during phonation under conventional laryngeal observation. However, 3-D CT endoscopic images seen from the oral side are produced without the arytenoid mound and the vocal process is always depicted in 3-D CT endoscopic images seen from the oral and tracheal sides. Another reason is that fluctuation and jittery motion of the laryngeal images under conventional laryngeal observation may interfere with accurate evaluation of relatively small paradoxical vocal fold movement. Electromyography should be performed to conclude the presence of paradoxical movement of the affected fold. Overadduction of the healthy vocal fold during phonation was detected in 15 subjects in the MSHCT examination. The vocal fold on the healthy side was not visualized under endoscopic observation in 1 (subject 7) of the 15 subjects and another (subject 34) did not undergo videostroboscopy. The vocal fold on the healthy side of the remaining 13 subjects was assessed as being overadducted during phonation. Therefore, overadduction of the healthy vocal fold during phonation can be detected by videostroboscopy as long as the vocal fold is visualized endoscopically.

In conclusion, the use of 3-D endoscopy combined with coronal MPR images enables description of the 3-D characteristics of the unilaterally immobile larynx, and supplements videostroboscopic findings as exemplified by differences in vertical position and thickness between both vocal folds.

Back to top
Article Information

Correspondence: Eiji Yumoto, MD, Department of Otolaryngology–Head and Neck Surgery, Department of Radiology, Graduate School of Medicine, Kumamoto University, 1-1-1 Honjo, Kumamoto 860-8556, Japan (yumoto@kumamoto-u.ac.jp).

Submitted for publication August 22, 2003; final revision received December 18, 2003; accepted January 24, 2004.

References
1.
Isshiki  NOkamura  HIshikawa  T Thyroplasty type I (lateral compression) for dysphonia due to vocal cord paralysis or atrophy. Acta Otolaryngol.1975;80:465-473.
PubMed
2.
Isshiki  NTanabe  MSawada  M Arytenoid adduction for unilateral vocal cord paralysis. Arch Otolaryngol.1978;104:555-558.
PubMed
3.
Isshiki  NTaira  TKojima  HShoji  K Recent modifications in thyroplasty type I. Ann Otol Rhinol Laryngol.1989;98:777-779.
PubMed
4.
Maragos  N Type I thyroplasty: pitfalls of modifying the Isshiki approach. J Voice.1997;11:470-473.
PubMed
5.
Remacle  MDujardin  JMLawson  G Treatment of vocal fold immobility by glutaraldehyde-cross-linked collagen injection: long-term results. Ann Otol Rhinol Laryngol.1995;104:437-441.
PubMed
6.
Shaw  GYSzewczyk  MASearle  JWoodroof  J Autologous fat injection into the vocal folds: technical considerations and long-term follow-up. Laryngoscope.1997;107:177-186.
PubMed
7.
Tucker  H Laryngeal reinnervation for unilateral vocal cord paralysis. Ann Otol Rhinol Laryngol.1981;90:457-459.
PubMed
8.
Crumley  RLIzdebski  KMcMicken  B Nerve transfer versus Teflon injection for vocal cord paralysis: a comparison. Laryngoscope.1988;98:1200-1204.
PubMed
9.
Yumoto  ESanuki  THyodo  M Three-dimensional endoscopic images of vocal fold paralysis by computed tomography. Arch Otolaryngol Head Neck Surg.1999;125:883-890.
PubMed
10.
Inagi  KKhidr  AAFord  CNBless  DMHeisey  DM Correlation between vocal functions and glottal measurements in patients with unilateral vocal fold paralysis. Laryngoscope.1997;107:782-791.
PubMed
11.
Newmark  GMConces Jr  DJKopecky  KK Spiral CT evaluation of the trachea and bronchi. J Comput Assist Tomogr.1994;18:552-554.
PubMed
12.
Silverman  PMZeiberg  ASSessions  RBTroost  TRZeman  RK Three-dimensional imaging of the hypopharynx and larynx by means of helical (spiral) computed tomography: comparison of radiological and otolaryngological evaluation. Ann Otol Rhinol Laryngol.1995;104:425-431.
PubMed
13.
Zeiberg  ASSilverman  PMSessions  RBTroost  TRDavros  WJZeman  RK Helical (spiral) CT of the upper airway with three-dimensional imaging: technique and clinical assessment. AJR Am J Roentgenol.1996;166:293-299.
PubMed
14.
Yumoto  ESanuki  THyodo  MYasuhara  YOchi  T Three-dimensional endoscopic mode for observation of laryngeal structures by helical computed tomography. Laryngoscope.1997;107:1530-1537.
PubMed
15.
Morinaka  SMiyamoto  SHidaka  A Helical CT scanning of laryngeal deviation. Auris Nasus Larynx.2001;28:151-159.
PubMed
16.
Isshiki  NIshikawa  T Diagnostic value of tomography in unilateral vocal cord paralysis. Laryngoscope.1976;86:1573-1578.
PubMed
17.
Hong  KHJung  KS Arytenoid appearance and vertical level difference between paralyzed and innervated vocal cords. Laryngoscope.2001;111:227-232.
PubMed
18.
von Leden  HMoore  P The mechanics of the cricoarytenoid joint. Arch Otolaryngol.1961;73:541-550.
PubMed
19.
Woodson  GEHengesteg  ARosen  CAYeung  DChen  N Changes in length and spatial orientation of the vocal fold with arytenoid adduction in cadaver larynges. Ann Otol Rhinol Laryngol.1997;106:552-555.
PubMed
20.
Kumakawa  KYamaguchi  HHirose  H Image analysis of the contour of paralyzed vocal cord during phonation. Nippon Jibiinkoka Gakkai Kaiho.1990;93:61-70.
PubMed
21.
Hiroto  IHirano  MTomita  H Electromyographic investigation of human vocal cord paralysis. Ann Otol Rhinol Laryngol.1968;77:296-304.
PubMed
22.
Blitzer  AJahan  AFKeidar  A Semon's law revisited: an electromyographic analysis of laryngeal synkinesis. Ann Otol Rhinol Laryngol.1996;105:764-769.
PubMed
23.
Crumley  RL Laryngeal synkinesis revisited. Ann Otol Rhinol Laryngol.2000;109:365-371.
PubMed
24.
Omori  KSlavit  DHKacker  ABlaugrund  SM Quantitative criteria for predicting thyroplasty type I outcome. Laryngoscope.1996;106:689-693.
PubMed
25.
Tanaka  SChijiwa  KHirano  M Study on over-adduction of unaffected vocal fold in unilateral recurrent laryngeal nerve paralysis. Larynx Jpn.1993;5:135-141.
×