260 / 412
samples from normophonic speakers. The authors strongly ad-
vised that caution should be used when determining the abnor-
mality of mucosal wave variations during clinical visualization
procedures. One concern in our study is that in 45.5% of the pa-
tients with histopathologically determined CIS, nonvibrating
segments were absent, which is a significant number.
Treatment involves removing the lesion with epithelium,
basal membrane, and lamina propria and, depending on the
type of cordectomy, deeper underlying structures. Vocal fold
scaring was examined on animal models. Rousseau et al
18,19
described the development of a vocal fold scar 6 months after
surgical injury in canine and rabbit models. As early as 2
months after the surgical removal of the epithelium and
lamina propria, no significant difference in collagen density
was noted, but at 6 months after injury, collagen density was
significantly increased in the surgically injured animals
compared with those with normal vocal folds. By 6 months,
the procollagen and elastin levels had achieved the densities
observed in normal vocal folds, although the elastin fibers
remain fragmented and disorganized. The basal layer of the
mucosal epithelium continues to experience remodeling in
the later stages of wound healing, whereas the intercellular
epithelial space undergoes remodeling earlier during the
acute stage of wound healing.
20
Kishimoto et al
21
investigated the maturation process of vo-
cal fold scarring after cordectomy in 10 patients (eight with
early laryngeal carcinoma and two with laryngeal dysplasia) us-
ing videostroboscopy. The patients were treated with cordec-
tomy types I–III. Improvements in amplitude of mucosal
wave were visible 6 months after the procedure and continued
to improve up to 14 months after the procedure. Twelve months
after the initial treatment was a reasonable time to assess the
treatment results in our study. Indeed, there were improvements
in phase symmetry, periodicity, amplitude of vocal fold vibra-
tions, and the regularity of mucosal wave. The number of pa-
tients with nonvibratory segment decreased. At the end of the
follow-up period, there were 23 (20.53%) patients with detected
nonvibrating segment. Four patients who developed invasive
carcinoma were among these patients. In other patients, this
result could be explained by the vocal fold scarring process,
particularly because in these patients, type II and type III cor-
dectomies were performed as a treatment of choice. This is
yet another limiting factor for stroboscopy use because it cannot
reliably distinguish the vocal fold process resulting from the ex-
istence of a nonvibrating segment.
Many voice disorders are marked by either aperiodicity or
fluctuating frequency and, therefore, cannot be visualized
with stroboscopy.
22
There are a growing number of articles
that emphasize the importance of different and more effective
methods in evaluating irregular vocal fold vibrations and the
propagation and existence of the mucosal wave, such as electro-
glottography, high-speed digital imaging, videokymography, or
digital kymography. Mucosal wave propagates in both vertical
and horizontal directions, and quantifying the vertical displace-
ment is crucial for understanding the effect of pathologies on
the mucosal wave. Stroboscopy, videokymography, and high-
speed digital imaging only provide a two-dimensional image
TABLE 6.
Multivariate Regression Analysis of Correlation of Dysplasia With Stroboscopic Signs and Type of Treatment
Cordectomy
Type
Glottic
Occlusion
Phase
Symmetry
Periodicity
Amplitude
Mucosal Wave Nonvibratory Segment
t
5.31
7.65
1.39
2.36
6.69
5.288
10.39
OR (95% CI) 1.22 (0.76 to 1.68) 1.88 (1.39 to 2.37) 0.80 ( 0.34 to 1.94) 1.24 (0.19 to 2.28) 2.03 (1.43 to 2.63) 1.68 (1.05 to 2.30) 2.950 (2.39 to 3.51)
Sig.
0.006*
0.791
0.076
0.265
0.474
0.636
0.000*
Abbreviations:
CI, confidence interval; OR, odds ratio.
*
P
< 0.05.
Vojko Djukic,
et al
Stroboscopy in Detection of Laryngeal Dysplasia
35