DOI:10.2214/AJR.07.2778
AJR 2008; 190:1376-1379
© American Roentgen Ray Society
Usefulness of Laryngeal Phonation CT in the Diagnosis of Vocal Cord Paralysis
Bum Soo Kim1,
Kook Jin Ahn2,
Young Hak Park3 and
Seong Tai Hahn2
1 Department of Radiology, Kangnam St. Mary's Hospital, College of Medicine, The
Catholic University of Korea, Seoul, South Korea.
2 Department of Radiology, St. Mary's Hospital, College of Medicine, The
Catholic University of Korea, 62, Yeouido-dong, Youngdeungpo-gu, Seoul,
150-713, South Korea.
3 Department of Otolaryngology, St. Mary's Hospital, College of Medicine, The
Catholic University of Korea, Seoul, South Korea.
Received July 6, 2006;
accepted after revision November 12, 2007.
Address correspondence to K. J. Ahn
(ahn-kj{at}catholic.ac.kr).
Abstract
OBJECTIVE. The purpose of this study was to determine the
effectiveness of laryngeal phonation CT for the diagnosis of vocal cord
paralysis by examining the physiologic and functional changes in the larynx
during vowel phonation in patients with vocal cord paralysis.
SUBJECTS AND METHODS. For the control study, three healthy
volunteers underwent laryngeal phonation CT while vocalizing the vowels /hee/,
/ih/, and /ah/, and reconstructed coronal images of the larynx were obtained.
After the control study, 28 patients with unilateral vocal cord paralysis
underwent laryngeal phonation CT during /hee/ phonation, which was chosen as a
most appropriate vowel for this purpose. Changes in the paralyzed and normal
vocal cords were evaluated quantitatively and qualitatively on coronal
reconstruction images.
RESULTS. On the coronal reconstructed images from the healthy
volunteers, the normal cords had a shoulder formation appearance, and the
cords lay within 1 mm lateral to the midline during phonation. For patients
with vocal cord paralysis during /hee/ phonation, the average angle formed by
the long axis of the vocal cord and the midline was 71.67° on the affected
side and 92.21° on the normal side (p = 0.001). The vocal cord
edges lay 1.5 mm lateral to the midline on the affected side and 0.44 mm
lateral to the midline on the normal side (p = 0.003). In the
qualitative study, the two observers found the coronal reconstructions of the
laryngeal phonation CT scans yielded a higher detection rate than did
conventional axial CT.
CONCLUSION. Laryngeal phonation CT proved more useful for evaluating
vocal cord paralysis than did conventional CT and can be used as a primary
diagnostic tool when vocal cord paralysis is suspected.
Keywords: larynx • phonation CT • vocal cord paralysis
Introduction
Coronal radiography and simple tomography have been used for years to see
changes in the larynx in many pathologic states, but CT has become fast enough
to replace these techniques. While laryngoscopy and stroboscopy are commonly
used for the diagnosis of vocal cord paralysis in patients with sustained
hoarse ness [1,
2], CT is mainly used to
identify the cause of paralysis. This is because conventional axial CT could
not capture dynamic changes in the vocal cords, although changes in the vocal
cords during phonation, inspiration, and expiration are crucial to the
diagnosis of vocal cord paralysis. The advent of MDCT scanners has enabled
acquisition of images of the larynx during phonation as one volume image, and
coronal reconstruction of the volume images makes it possible to see changes
in the vocal cords during phonation
[3–6].
This technique would also be useful for determining the type and amount of
materials needed for injection laryngoplasty, which is used to manage vocal
cord paralysis [7]. The aim of
this study was to determine the usefulness of coronal reconstruction of volume
images from laryngeal phonation CT in the diagnosis of vocal cord
paralysis.
Subjects and Methods
Subjects
For identification of the normal appearance of the larynx, three volunteers
who did not have laryngeal disorders were recruited to undergo phonation CT
during normal breathing and during sustained phonation of three tones:
sustained falsetto high-pitched /hee/, lowest-pitch /ih/, and the most
comfortable pitch and intensity of /ah/. After the volunteer study, the most
appropriate vowel for use in the diagnosis of vocal cord paralysis was
identified in a pilot study of laryngeal phonation CT of 15 patients with
vocal cord paralysis confirmed at laryngoscopy and stroboscopy. The pilot
study was performed to reduce the total amount of CT radiation exposure by
decreasing the frequency of phonation CT to one examination of each patient.
The 15 patients were divided into three groups of five, and each group was
instructed to vocalize one of the three vowels for 10 seconds during CT.
After the one vowel was chosen on the basis of the results of the pilot
study, 28 patients with vocal cord paralysis were recruited (19 men, nine
women; mean age, 55 years; range, 31–77 years), including the 15
patients who had been included in the pilot study. These patients performed
sustained phonation of the vowel /hee/ during 10 seconds of CT. Most of the
cases of vocal cord paralysis were associated with surgery, such as thyroid
surgery (17 patients); the other cases were idiopathic (three patients) or due
to trauma (three patients) or other causes (two patients). The symptom
duration was more than 1 year for 14 patients, 6 months–1 year for 10
patients, and less than 6 months for four patients. The study protocol was
approved by the institutional board for clinical investigation, and informed
consent was obtained from each subject participating in the study.
CT
An MDCT scanner (Somatom Volume Zoom, Siemens Medical Solutions) was used
with the following parameters: tube current, 120 mAs; voltage, 120 kV;
detector collimation, 5 x 1 mm; table speed, 4 mm/rotation (pitch, 1);
rotation time, 0.75 second. Image recon struction was performed with a
standard kernel; re constructed slice thickness was 1 mm; and the
reconstruction interval was 0.5 mm. The CT range included the distance between
the hyoid bone and the cricoid cartilage in the larynx. The slice planes were
as close to parallel to the vocal cords as possible to obtain axial images of
the larynx. The region of the larynx (hyoid bone through cricoid cartilage)
was reconstructed with a small field of view (70 mm) for quiet breathing and
phonation. During one pho nation CT examination, each patient received an
average of 88 mGy dose–length product (CT dose index, 10.80) absorbed
radiation dose.
Analysis
Quantitative analysis—On the coronal reconstructed images,
the distance between the tip of the vocal cord and the midline of the larynx
was measured to evaluate vocal cord mobility during ad duction
(Fig. 1A). The angles between
the vocal cords and the long axis of the larynx were measured to evaluate the
degree of vocal cord angle formation during phonation. To measure vocal cord
angle formation, a presumptive median line was drawn between the superior and
inferior marginal surfaces of the vocal cord, and that line was used as a
vocal cord axis. The angle between the vocal cord axis and the midline of the
laryngeal airway was measured (Fig.
1B). The Student's t test was performed to assess the
difference in measurements between the paralyzed vocal cords and the normal
vocal cords. Values of p < 0.05 were considered statistically
significant.
View larger version (108K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1A —37-year-old woman with left vocal cord paralysis. Coronal
reconstructed image shows distance between cord tip and midline of laryngeal
airway during phonation. Straight line is drawn along middle of laryngeal
airway, and distances (gray arrows a and b) between each vocal cord
and midline of laryngeal airway are measured. Black arrows indicate stretched
vocal cord during phonation.
|
|
View larger version (110K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1B —37-year-old woman with left vocal cord paralysis. Coronal
reconstructed image shows measurement of vocal cord angle formation during
phonation. Median line is drawn between superior and inferior marginal lines
of vocal cord and is used as vocal cord axis. Angles (a and b) between vocal
cord axis and midline of laryngeal airway are measured and used as measure of
vocal cord angle formation.
|
|
Qualitative analysis—To ensure the difference in
effectiveness between the coronal images reconstructed from laryngeal
phonation CT and conventional axial images, two general radiologists not
experienced in the diagnosis of vocal cord paralysis with routine axial neck
CT were recruited for the qualitative study. These radiologists were given the
axial images and the reconstructed images of 21 patients with vocal cord
paralysis (11 with left-sided vocal cord paralysis, 10 with right-sided vocal
cord paralysis) randomly selected from the images of the 28 patients and were
instructed to make a diagnosis of vocal cord paralysis by analyzing each
image.
View larger version (121K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2A —29-year-old woman in good health. Coronal reconstructed CT
image of larynx during normal breathing shows larynx at level of vocal cords
appears to be flat, without vocal cord protrusion.
|
|
View larger version (115K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2B —29-year-old woman in good health. Coronal reconstructed CT
image of larynx during phonation of /hee/ shows both vocal cords stretched
across opening of larynx, forming acute angle with midline known as shoulder
formation (arrows).
|
|
Results
Appearance of Normal Larynx
The reconstructed coronal images of the normal larynx showed that the vocal
cords were flat and without protrusion during normal breathing
(Fig. 2A). When the three
volunteers were asked to phonate the vowels /hee/, /ih/, and /ah/, the vocal
cords stretched across the opening of the larynx, forming an acute angle with
the midline called the shoulder formation
(Fig. 2B).
Pilot Study
In the pilot study with 15 patients with vocal cord paralysis, the largest
differences in the average angle and the distance between the affected side
and the normal side were observed during phonation of /hee/
(Table 1). On the basis of this
finding, an additional 23 patients with vocal cord paralysis were later
instructed to phonate the vowel /hee/ during laryngeal phonation CT.
View this table:
[in this window]
[in a new window]
|
TABLE 1: Measured Angles of Vocal Cords and Distances Between Vocal Cord Tip and
Midline in /hee/, /ih/, and /ah/ Phonation Groups
|
|
View larger version (123K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3A —49-year-old man with right vocal cord paralysis for 4 months.
Coronal reconstructed CT image of larynx during normal breathing shows relaxed
appearance of unaffected left vocal cord in contrast to protruding appearance
of affected right vocal cord (arrow).
|
|
View larger version (118K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3B —49-year-old man with right vocal cord paralysis for 4 months.
Coronal reconstructed image during phonation of /hee/ shows unaffected left
vocal cord has definite protrusion to midline of laryngeal airway
(arrow). No definite change is evident at affected right vocal
cord.
|
|
/hee/ Phonation in Patients with Vocal Cord Paralysis
Among the 28 patients, the average angle between the long axis of the vocal
cord and the midline was 71.67° on the affected side and 92.21° on the
normal side, a significant difference (p = 0.001). The vocal cord
tips lay 1.48 mm lateral to the midline on the affected side and 0.43 mm
lateral to the midline on the normal side; this difference also was
significant (p = 0.003) (Figs.
3A,
3B and
4A,
4B).
View larger version (114K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 4A —65-year-old woman with left vocal cord paralysis for 2 years.
Coronal reconstructed CT image of larynx during normal breathing shows relaxed
appearance of both vocal cords without protrusion (arrows).
|
|
View larger version (120K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 4B —65-year-old woman with left vocal cord paralysis for 2 years.
Coronal reconstructed image during phonation of /hee/ shows unaffected right
vocal cord protruding to midline of laryngeal airway with definite shoulder
formation (black arrow). Affected left vocal cord (white
arrow) has relaxed, obtuse appearance.
|
|
Qualitative Analysis
One of the two radiologists made accurate diagnoses in 17 (81%) of 21 cases
by referring to the conventional axial images. The same radiologist made a
correct diagnosis in 20 (95%) of 21 cases using the coronal reconstructed
images. The other radiologist made the correct diagnosis in 18 (86%) of 21
cases using the conventional axial images and in 20 (95%) of 21 cases using
the coronal reconstructed images. The interobserver variability (kappa value)
was 0.83.
Discussion
Chin et al. [8] reported
that for making the diagnosis of vocal cord paralysis, dilatation of the
ipsilateral pyriform sinus, medial positioning and thickening of the
ipsilateral aryepiglottic fold, and dilatation of the ipsilateral laryngeal
ventricle are the most sensitive CT findings in the neck. These findings are
valuable clues in the diagnosis of vocal cord paralysis. Even with these
useful diagnostic findings, however, a radiologist must be experienced to
diagnose vocal cord paralysis with CT. In addition, owing to the ambiguity of
the CT findings, no radiologist makes an accurate diagnosis 100% of the time,
as was found in the qualitative analysis in our study. Thus a new imaging
method is needed to improve the ease and accuracy of the diagnosis of vocal
cord paralysis. It should be possible for a radiologist to make an accurate
diagnosis of vocal cord paralysis with CT of the neck, even when CT is
requested for the diagnosis of a disease other than vocal cord paralysis.
CT also can be used to examine the vocal cords and the surrounding soft
tissues to estimate the amount of material needed for injection laryngoplasty
in the management of vocal cord paralysis
[9]. Injection laryngoplasty is
a continuously evolving procedure whereby a plastic material, such as
polytetrafluoroethylene (Teflon, DuPont) or acellular dermis (AlloDerm,
LifeCell) is injected into the vocal cords under indirect mirror laryngoscopic
or flexible videolaryngoscopic guidance for visualization of the vocal cords.
Depending on the material used, the injection must be made into the
paraglottic space or the medial or lateral aspect of the thyroarytenoid muscle
to move an immobile cord to the median. Among the various approaches,
transcutaneous techniques are most popular
[7]. During injection of the
plastic material, the target can vary depending on the symptoms and the
condition of the diseased vocal cord. In these cases, preoperative phonation
CT of the larynx and coronal reconstruction can be valuable in understanding
the 3D shape of the diseased vocal cords and determining the amount of
material to inject [5,
7,
9]. Further study of the role
of preoperative phonation CT of the larynx before injection laryngoplasty is
necessary.
In this study, coronal reconstructed images of phonation CT scans were used
to discern the normal appearance of the vocal cords in three healthy persons
during normal respiration and phonation. During normal respiration, the vocal
cords abducted, and the inter-cartilaginous part became wider to promote easy
inspiration and expiration. On the coronal reconstructed images, the vocal
cords were flat; there were no protruding parts
[10,
11]. During phonation, the
vocal cords lay within 3 mm of the midline owing to activation of the lateral
cricoarytenoid and interarytenoid muscles. This change in the vocal cords was
coupled with activation of the cricothyroid and vocalis muscles, leading to
changes in the pitch of the voice
[10,
11]. The CT findings regarding
changes in the vocal cords during phonation were well documented in a study by
Gamsu et al. [12]. On the
coronal reconstructed images, the vocal cords were stretched across the
laryngeal airway, forming an acute angle with the midline during phonation.
This acute angulation with the midline of the larynx is called the shoulder
formation.
In the pilot study, as expected, the vocal cords were most distinctly
visualized during phonation of the high-pitched sound /hee/. To phonate
high-pitched tones such as /hee/, the vocal cords must be thinner than during
phonation of other tones. To make the vocal cords thinner, the subglottis of
the larynx should be elevated, resulting in a more acute shoulder formation.
As the results of the pilot study show, the /hee/ phonation is better than
other sounds for evaluating the vocal cords.
In the patients with vocal cord paralysis, the lateral cricoarytenoid and
interarytenoid muscles were not activated, so there was no acute angle with
the midline, and adduction of the vocal cords was incomplete. As a result,
there was no gross change on the coronal reconstructed images between regular
respiration and phonation. In the qualitative analysis, two radiologists read
both the conventional axial images and the reconstructed coronal images from
phonation CT to assess the effectiveness of the two imaging methods for making
a diagnosis of vocal cord paralysis. One radiologist had an accuracy of 81%
and the other an accuracy of 86% with the conventional axial images, but both
radiologists had 95% accuracy with the coronal reconstructed images. Thus the
reconstructed images helped them to diagnose vocal cord paralysis.
Patient cooperation is important for imaging of the vocal cords, and
cooperation is not easy during phonation CT because the imaging must be
performed during sustained phonation for more than 10 seconds. Given the
ongoing advances in CT technology, this inconvenience is expected to be
overcome in the near future.
We conclude that laryngeal phonation CT is more useful than conventional CT
in the evaluation of vocal cord paralysis. The technique can be used as a
primary diagnostic tool in the care of patients with suspected vocal cord
paralysis.
References
- Harries ML, Morrison M. The role of stroboscopy in the management
of a patient with a unilateral vocal fold paralysis. J Laryngol
Otol 1996; 110:141
–143[Medline]
- Omori K, Kacker A, Slavit DH, Blaugrund SM. Quantitative
videostroboscopic measurement of glottal gap and vocal function: an analysis
of thyroplasty type I. Ann Otol Rhinol Laryngol1996; 105:280
–285[Medline]
- Baum U, Greess H, Lell M, Nomayr A, Lenz M. Imaging of head and
neck tumors: methods—CT, spiral-CT, multislice-spiral-CT. Eur
J Radiol 2000; 33:153
–160[CrossRef][Medline]
- Lell MM, Gress H, Hothorn T, Janka R, Bautz WA, Baum U. Multiplanar
functional imaging of the larynx and hypopharynx with multislice spiral CT.
Eur Radiol 2004;14
:2198
–2205[CrossRef][Medline]
- Jun BC, Kim HT, Kim HS, Cho SH. Clinical feasibility of the new
technique of functional 3D laryngeal CT. Acta
Otolaryngol 2005; 125:774
–778[Medline]
- Yumoto E, Oyamada Y, Nakano K, Nakayama Y, Yamashita Y.
Three-dimensional characteristics of the larynx with immobile vocal fold.
Arch Otolaryngol Head Neck Surg 2004;130
: 967–974[Abstract/Free Full Text]
- O'Leary MA, Grillone GA. Injection laryngoplasty.
Otolaryngol Clin North Am 2006;39
: 43–54[CrossRef][Medline]
- Chin SC, Edelstein S, Chen CY, Som PM. Using CT to localize side
and level of vocal cord paralysis. AJR2003; 180:1165
–1170[Abstract/Free Full Text]
- Laccourreye O, Bely N, Crevier-Buchman L, Brasnu D, Halimi P.
Computerized tomography of the glottis after intracordal autologous fat
injection. J Laryngol Otol 1998;112
: 971–972[Medline]
- Standring S, Ellis H, Berkovitz BKB, eds. Larynx. In:
Gray's anatomy: the anatomical basis of clinical
practice, 39th ed. Philadelphia, PA: Elsevier, 2005:633
–646
- Jiang J, Lin E, Hanson DG. Vocal fold physiology.
Otolaryngol Clin North Am 2000;33
: 699–718[CrossRef][Medline]
- Gamsu G, Mark AS, Webb WR. Computed tomography of the normal larynx
during quiet breathing and phonation. J Comput Assist
Tomogr 1981; 5:353
–360[Medline]
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
|
|
 
V.V. Wear, J.W. Allred, D. Mi, and M.K. Strother
Evaluating "Eee" Phonation in Multidetector CT of the Neck
AJNR Am. J. Neuroradiol.,
June 1, 2009;
30(6):
1102 - 1106.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
