Keywords
vocal folds - rabbits - FGFs - fibroblasts - grafting - fascia
Introduction
The human larynx enables interpersonal interaction. No other organ of the human body
presents similar structural arrangement like the vocal folds, especially the flexibility
and vibratory features, which are related to sound production.[1]
Several conditions promote structural changes of the vocal fold lamina propria, with
the formation of scar tissue (increased deposition of type I collagen and fibronectin,
decreased elastin, decorin, and hyaluronic acid), decreasing the vibration, leading
to a poorer voice quality.[2]
[3]
[4] Since the 1990s, many studies focused on reducing postoperative fibrosis and scarring,
and to improving the flexibility of the vocal folds.[5]
[6]
[7]
[8]
[9]
[10]
[11] However, many of these techniques are not yet available for clinical usage.
Fibroblast growth factors (FGFs) are potent regulators of cell proliferation, differentiation
and function and are critically important to normal development, tissue maintenance,
wound repair and angiogenesis. Fibroblast growth factors are also linked with several
pathological conditions.[12] Therefore, we aim to determine whether FGFs may influence tissue integration of
grafted fascia into rabbits' vocal folds.
Method
Ethics
The Institutional Review Board approved this study (CEP 1033–1).
Specimens
Twenty New Zealand rabbits of both genres were used in this experiment, weighing between
two and three kilograms. The number of male and female animals was the same. All the
animals were kept with water and food “ad libitum,” from the moment of the procedure
to the moment of their sacrifice.
The rabbits were divided into two groups. Group A, with animals maintained under observation for 12 months after the surgical procedure,
and group B, which was observed for 1 month. The animals were randomly assigned to each group.
Animals with congenital malformations of the larynx and female rabbits that got pregnant
during the experiment were excluded from this study.
Surgical Procedure
The animals underwent intramuscular anesthesia with ketamine (25 mg/kg/dose) and xylazine
(3 mg/kg/dose) and kept with spontaneous breathing.
A 4-cm cervical longitudinal incision was performed at the level of the thyroid cartilage.
The subcutaneous tissue was exposed by using scissors, and a fragment of superficial
fascia of the neck, of ∼ 10 × 10 mm, was excised. The fascia was extended on a metal
surface and left to dehydrate spontaneously.
A 2-mm longitudinal incision was made along the vocal process of the vestibular face
of the right and left vocal folds, close to the glottic aperture. A 45 degree-angled
dissector was introduced into the incision to create a submucosal pouch, parallel
to the medial face of the vocal fold, ∼ 2 mm long anteroposteriorly, in both right
and left vocal folds.
After dehydration, a fragment of 1 × 1 mm of the fascia was inserted into the submucosal
pouch. The right vocal fold was injected with 5 µL of FGFs solution (concentration
of 1 ng/μL, diluted in sodium chloride) immediately after the fascia was implanted.
The left vocal fold received no injection of FGFs and remained as a control for histological
changes due to grafting the fascia.
Histological Analysis
After the animals were sacrificed, their larynxes were obtained and maintained in
a 10% formalin solution for 24 hours. The vocal folds were then resected with the
thyroid cartilage, the segments were histologically processed (3 μm of thickness for
each section) and embedded in paraffin. Every section contained both vocal folds for
comparison. The slides were stained with hematoxylin and eosin (H and E), picrosirius
red, and Masson trichrome and then examined under light microscopy (in different magnifications)
to assess the following parameters: a) number and type of inflammatory cells (neutrophils,
lymphocytes, and macrophages); b) presence or absence of fibrosis; c) neovascularization;
d) the density of collagen within the grafted area; and e) presence or absence of
foreign body reaction.
The inflammatory reaction was semiquantitatively evaluated by comparing the small
portion of the surrounding of the implanted fascia with the lamina propria beyond
this area. It was then classified as:
0–absence of mononuclear cells;
I –mild (1 to 10 mature mononuclear cells were observed);
II –moderate (11 to 20 mature mononuclear cells were observed); and
III –severe (more than 20 mature mononuclear cells).
Additionally, the inflammatory reaction was classified as acute when there was presence of polymorphonuclear cells, and as chronic when there was presence of a foreign body reaction with active macrophage predominance.
Fibrosis and the density of collagen within it were analyzed using a Sony DXC-101/Trinitron
CCD Camera System (Sony Electronics Inc., San Diego, CA, USA), scanned trough an OCULUS
TCX System (Teledyne DALSA, Waterloo, ON, Canada), and analyzed with the image editing
software Image Plus 4.5 for Windows (Astra Image, Kawanishi City, Japan). Fibrosis
was also semiquantitatively assessed and graded as:
0–absence of collagen fibers;
I – mild (few foci of collagen fibers);
II – moderate (continued collagen fibers around the grafted area); or
III – severe (bigger area of continued collagen fibers around the grafted area, consisting
a fibrous capsule).
Neovascularization was defined by the presence of thin wall blood vessels within the
grafted area.
Statistical Analysis
We calculated and compared the data obtained from the analysis of the inflammatory
reaction, presence of neovascularization, and fibrosis using the Fisher exact test.
Statistical analysis obtained from the density of collagen fibers was performed by
non-parametric t-test, Wilcoxon test, and Kruskal-Wallis algorithm, comparing both the left and right
sides, and in both groups (A and B). Findings were considered statistically significant
when p values were < 0.05.
Results
Our final groups comprised of 6 rabbits in group A (12-months survival) and 6 rabbits in group B (1-month survival). All data obtained from the histological analysis are shown in
[Table 1].
Table 1
Histological findings in both groups
|
Case No./Group
|
Intensity of inflammatory reaction
|
Cell type
|
Fibrosis
|
Chronic reaction
|
Neovas-cularization
|
|
RVF
|
LVF
|
RVF
|
LVF
|
RVF
|
LVF
|
RVF
|
LVF
|
RVF
|
LVF
|
|
1/A
|
I
|
I
|
LH
|
0
|
II
|
0
|
0
|
0
|
I
|
0
|
|
2/A
|
I
|
I
|
LH
|
0
|
I
|
0
|
0
|
0
|
I
|
0
|
|
3/A
|
II
|
I
|
LH
|
0
|
I
|
0
|
0
|
0
|
I
|
0
|
|
4/A
|
I
|
I
|
LH
|
0
|
I
|
0
|
0
|
0
|
I
|
0
|
|
5/A
|
I
|
I
|
L
|
0
|
I
|
0
|
0
|
0
|
I
|
0
|
|
6/A
|
I
|
I
|
L
|
0
|
I
|
0
|
0
|
0
|
I
|
0
|
|
1/B
|
II
|
0
|
H
|
LH
|
0
|
0
|
0
|
0
|
0
|
0
|
|
2/B
|
II
|
I
|
LH
|
L
|
0
|
0
|
0
|
0
|
0
|
0
|
|
3/B
|
III
|
I
|
L
|
LH
|
0
|
0
|
0
|
0
|
0
|
0
|
|
4/B
|
II
|
II
|
LH
|
LH
|
0
|
0
|
0
|
0
|
0
|
0
|
|
5/B
|
III
|
I
|
LH
|
LH
|
0
|
0
|
0
|
0
|
0
|
0
|
|
6/B
|
III
|
I
|
LH
|
LH
|
0
|
0
|
0
|
0
|
0
|
0
|
Abbreviations: H, histiocytic pattern; L, lymphocytic pattern; LH, lymphohistiocytic
pattern; LVF, left vocal fold; RVF, right vocal fold.
Note: 0, absence; I, mild; II, moderate; III, severe.
Histological analysis of cellular predominance in both vocal folds (injected with
FGFs or control) revealed predominant lymphohistiocytic infiltration with no differences
between sides or groups. Additionally, the lymphohistiocytic pattern was statistically
significant when compared with the other types of cells within the inflammation (p = 0.008).
There was a statistical difference between both groups when we compared the presence
of inflammatory reaction and the side of the vocal fold. The intensity of inflammatory
reaction in the right vocal fold (with FGFs injection) showed frequent association
with a “moderate” or “severe” reaction in group B; however, in group A, it was related to an “absent” or “mild” reaction (p = 0.011) ([Figs. 1] and [2]). The intensity of inflammatory reaction in the left vocal fold (considered as a
control group) showed no differences between both groups (p = 0.573).
Fig. 1 Right vocal fold of a rabbit from group A with grafted fascia within the lamina propria and injected fibroblasts growth factor.
Mild inflammatory reaction (arrow) (hematoxylin and eosin x20).
Fig. 2 Right vocal fold of a rabbit from group B with grafted fascia within the lamina propria and injected fibroblasts growth factor.
Moderate inflammatory reaction (arrow) (hematoxylin and eosin x20).
The intensity of fibrosis was statistically significant when we compared the right
vocal fold in both groups (p = 0.014). It was frequently found in group A, and considered as “mild” in most of the samples. The left vocal fold in both groups
showed no fibrosis between both groups ([Figs. 3] and [4]).
Fig. 3 Right vocal fold of a rabbit from group A with grafted fascia within the lamina propria and injected fibroblasts growth factor.
Mild fibrosis and irregular deposition of collagen fibers in the subepithelial space
(arrow) (Trichrome Massonx20).
Fig. 4 Right vocal fold of a rabbit from group B with grafted fascia within the lamina propria and injected fibroblasts growth factor.
Mild fibrosis and irregular deposition of collagen fibers in the subepithelial space
(arrow) (Trichrome Massonx20).
Neovascularization was only observed in the right vocal folds (injected with FGFs)
of animals from group A. However, it was not statistically significant between the groups.
None of the vocal folds among both groups showed signs of chronic inflammatory reaction,
presence of giant cells, or foreign body reaction.
Discussion
Many techniques aim to restore the functional mechanism and provide a better oscillation
of the vocal fold's mucosa, such as: vocal therapy, injections of collagen or hyaluronic
acid, axial incisions of the mucosa, excision of the scarring tissue; additionally,
another option consists in the insertion of foreign tissue, such as fat or muscular
fascia between the epithelium and the vocal ligament.[13]
[14]
[15]
[16]
[17] Fascia graft has been used for decades in otologic and plastic surgeries without
complications and with reproducible results. Fascia presents a similar consistency
to that of collagen and low rate of metabolic activity. Rihkanen[18] proposed the use of minced fascia in the surgery of volumetric augmentation of vocal
fold. Reijonen et al[19] injected minced fascia into the vocal muscle and observed that the graft remained
identifiable histologically after 1 year of follow-up. Tsunoda et al[20]
[21] used intact fascia graft within the Reinke's space for correction of glottal incompetence
in human model, mainly arising from sulcus vocalis, with promising results.
The correction of extracellular matrix components distribution is essential to restore
the structure of the vocal folds' lamina propria. These components are primarily synthesized
by fibroblasts, and fascia is rich in this type of cells. Hirano et al[22] demonstrated that the growth factor is a strong inducer of fibroblasts proliferation
and that it stimulates the production of hyaluronic acid from fibroblasts culture.
Therefore, we must assume that, when the fibroblasts are stimulated by the addition
of growth factors, there is an induction of hyaluronic acid production, key for tissue
repair. This study, to the best of our knowledge, is the first to histologically evaluate
the use of superficial cervical fascia, associated with FGFs, when grafted into the
Reinke's space of the vocal folds of a rabbit model.
The presence of vocal fold scars due to injury or inflammation can alter the ultrastructure
of the lamina propria, disrupt the normal mucosal wave, and may result in glottic
incompetence and dysphonia, affecting the quality of life.[23]
[24] Several substances have been used to prevent this problem and improve glottal closure,
from synthetic to biological materials: Teflon, Gelfoam (Pfizer Inc., New York, NY,
USA), intralesional steroids, bovine collagen, homologous collagen, muscular fascia,
and fat.[23]
[24]
[25] Our study is the first to evaluate the fascia fragment integration into the lamina
propria with the addition of FGFs.
First, the presence of a lymphohistiocytic pattern was observed in both groups (after
1 and 12 months of procedure) of grafted vocal folds (left and right). It is also
important to note that the absence of macrophages among all the samples may be related
to an absent activating factor, triggering macrophage activity. Robins et al[26] demonstrated that the presence of macrophage is related to foreign body chronic
inflammatory reaction. However, Reijonen et al[19] identified that no intense inflammatory reaction was fostered until up to 12 months
after muscular fascia injection in paralyzed canine vocal folds. The isolated presence
of the fascia did not trigger inflammatory reaction, as demonstrated by Duke et al,[27] who evaluated vocal folds that have undergone crushed fascia lata injection.
Secondly, in vitro studies showed that the FGFs led to an increased production of
type I collagen and hyaluronic acid.[5]
[28] Yet, Scapini et al[29] demonstrated that their therapeutic interventions in an animal model of rabbits
statistically increased the density of collagen around the graft. However, Akdogan
et al[23] found significant difference in a group of vocal folds from rabbits treated with
vitamin A compared with an untreated group (lesser deposition of collagen and fibroblasts
in the first group); and Krishna et al,[25] studying in vitro and in vivo vocal fold scar model of rabbits with hepatocyte growth
factor exposure, found a decreased collagen secretion by fibroblasts. Our study demonstrated
that the group of vocal folds injected with FGFs seemed to trigger an increased inflammatory
reaction within 30 days after the procedure, compared with the group of vocal folds
with no FGFs injection. Furthermore, the density of collagen observed within the vocal
folds is bigger in rabbits from group A than in animals from group B, in both sides. Our observations are similar to others found in the literature, since
healing and scarring processes take more than 1 month, and mature collagen was also
present during those studies.[30]
[31]
Interestingly, the grafted fascia injected with FGFs seemed to be associated to an
increased rate of neovascularization, which may indicate a positive effect of grafted
tissue integration on the receiver site. No preliminary work evaluated the neovascularization
process in grafted vocal folds.
Our findings also suggest that FGFs may have triggered an acute inflammatory reaction.
Because there was a statistical difference between both groups and both sides of the
vocal folds, we may assume that the presence of grafted fascia was not the only factor
to trigger the inflammatory reaction. Only one study had evaluated the presence of
inflammatory reaction after grafting fascia into vocal folds,[27] with similar findings.
This is the first study to evaluate the cervical superficial fascia fragment integration
into the vocal folds of rabbits with the addition of FGFs. All specimens in the longer
observation group showed frequent signs of fibrosis that may have been triggered by
FGFs. This may suggest that fascia grafting, used isolatedly, may become a good option
for surgical procedures because it did not induce an acute or chronic inflammatory
reaction or trigger fibrosis.
Conclusion
Fibroblast growth factors may have induced fibrosis and neovascularization when injected
in the vocal fold of rabbits. Fibroblast growth factors alone do not represent a good
therapeutic option in phonosurgery, since we observed higher levels of fibrosis in
the vocal fold lamina propria. Laryngologists should be aware of this when performing
phonosurgery.
