Full text of DOI:10.1097/00005537-200107000-00009

The Laryngoscope
Lippincott Williams & Wilkins, Inc., Philadelphia
© 2001 The American Laryngological,
Rhinological and Otological Society, Inc.
An Investigation of Airway Wound Healing
Using a Novel in vivo Model
Jose Hardillo, MD; Christoph Vanclooster, MD; Pierre R. Delaere, MD, PhD
Objective: To study the amount of wound contrac-
tion and reepithelialization occurring in the healing
process of full-thickness mucosal defects treated with
and without mitomycin. Study Design: A new wound
healing model was developed in which the tracheal
mucosa was exteriorized without interference with
the blood supply or with the cartilage support of the
trachea. This was done by: 1) orthotopic tracheal re-
vascularization in vascularized fascia; 2) isolation of
revascularized segment after 14 days; 3) posterior lon-
gitudinal incision of revascularized segment; 4) exte-
riorization of tracheal mucosa with formation of an-
terior full-thickness mucosal defect; and 5) closure of
posterior tracheal incision and reimplantation in the
airway. This model was used to study airway wound
healing in three groups of animals: 1) controls (revas-
cularization, exteriorization, reimplantation) (N ؍ 6);
2) full-thickness mucosal defect: patch defect (N ؍ 5),
circumferential defect (N ؍ 3); and 3) full-thickness
mucosal defect after topical mitomycin application:
patch defect (N ؍ 7), circumferential defect (N ؍ 3).
The animals were followed for periods varying from 2
to 4 weeks or until signs of dyspnea. The surface areas
of the wounds before and after follow-up were mea-
sured. Wound healing was studied histologically on
axial and longitudinal sections. Results: Group 1: All
the animals survived for 1 month. No significant dif-
ference existed between surface area of isolated tra-
chea and of reimplanted trachea after follow-up.
Group 2: Five animals (patch defects) survived for 1
month. Full-thickness mucosal defects healed by
reepithelialization and by a surface area reduction of
58.9% (mean - standard deviation ؍ 10.5). The animals
with the circumferential defects showed dyspnea af-
ter an average follow-up of 14 days as a result of
excessive granulation tissue formation. Group 3: Mit-
omycin reproducibly inhibited wound closure, yield-
ing wounds that on average closed 56% less than con-
trols by day 14 (P <.001). Histologic comparisons
showed that mitomycin blocks angiogenesis during
wound healing. Conclusions: A wound healing model
based on tracheal revascularization, isolation, and re-
implantation was developed in rabbits. This model
allowed us to study the healing of full-thickness mu-
cosal defects inside the airway. Key Words: Respira-
tory, epithelium, mitomycin, migration, circumferen-
tial, mucosa, healing.
Laryngoscope, 111:1174–1182, 2001
INTRODUCTION
In comparison with the myriad of studies investigat-
ing various aspects of cutaneous wound healing, there is a
real dearth of basic science research investigating muco-
sal wound healing of the airway. Currently, there is no
effective way to study intraluminal events of reepithelial-
ization and wound contraction after airway injury. The
amount of wound contraction and reepithelialization for
full-thickness mucosal airway defects is unknown. This
information would be of interest to better understand the
etiology of airway stenosis and to give insight into the
principles leading to successful airway reconstruction.
The difference in knowledge between mucosal and skin
defects is because of the inaccessibility of the airway tis-
sue for direct observation. In vivo wound healing studies
of the larynx and trachea are based on endoscopic mod-
els.1,2 The problem with wounds that are made endoscop-
ically is the difficulty in standardization of the depth and
dimensions of the wound.
A solution to this problem may be found in studying
in vivo airway wound healing by using an invasive, open
approach. After opening of the airway lumen, airway
wounding can be performed in a controlled manner. The
problem with an open approach, however, is the interfer-
ence with the airway’s blood supply after anterior incision
of the airway wall. An in vivo experimental model that
interferes with the airway’s vascularization is not accept-
able because of the importance of an intact blood supply
for normal tissue healing. Another problem after anterior
incision of the airway wall is the interference with the
cartilaginous support, which might also influence the
healing process of mucosal wounds.
Presented at the 104th Annual Meeting of the American Laryngo-
logical, Rhinological and Otological Society, Inc., Palm Desert, CA, May 15,
2001.
From the Department of Oto-Rhino-Laryngology Head and Neck
Surgery, University Hospital St. Rafael, K.U. Leuven, Kapucijnenvoer 33,
B-3000 Leuven, Belgium.
Editor’s Note: This Manuscript was accepted for publication April
13, 2001.
Send Correspondence to Pierre R. Delaere, MD, Department of Oto-
Rhino-Laryngology Head & Neck Surgery, University Hospital St. Rafael,
K.U. Leuven, Kapucijnenvoer 33, B-3000 Leuven, Belgium. E-mail:
pierre.delaere@uz.kuleuven.ac.be
We developed an invasive in vivo model without in-
terfering with the airway’s blood supply and without in-
Laryngoscope 111: July 2001
1174
Hardillo et al.: Airway Wound Healing

terfering with the airway’s cartilaginous support. The
model consists of a complete isolation of tracheal segments
after previous tracheal revascularization. With this
model, the internal site of the trachea could be exterior-
ized and the trachea could be reimplanted after wounding
without interfering with the blood supply or with the
cartilaginous support. This model allowed us to study the
healing of full-thickness mucosal defects and the influence
of mitomycin application on mucosal wound healing. Re-
cent publications have shown that mitomycin prevents
scarring and fibrosis after tracheal wounding.3,4 In these
studies, the effect of mitomycin application was evaluated
in a model of subglottic stenosis. We wanted to evaluate
the effect of mitomycin on normal mucosal wound healing.
MATERIAL AND METHODS
Experimental Model: Control Animals
Six New Zealand white rabbits weighing 3 kg were used to
control the reliability of the experimental model. One animal was
used to document the amount of revascularization after 2 weeks.
Five animals were used to study the feasibility of tracheal reim-
plantation after exteriorization of the tracheal mucosa.
Stage 1: Vascular induction through staged tracheal
transfer. A staged vascular induction technique was used (Fig.
1). The concept is that a fascial flap can become a “vascular
carrier” and can be induced to provide an alternate blood supply
through revascularization of a tracheal segment after a relatively
short staging period in the orthotopic position.
In the first stage, the rabbits were premedicated with 0.3
mL/kg HypnormR (Janssen Pharmaceuticals, Belgium) (10 mg
fluanisone/0.2 mg fentanyl per mL) intramuscularly and anesthe-
sia was maintained with halothane through mask ventilation.
The left thoracic skin was incised and the left lateral thoracic
fascia flap was isolated on the lateral thoracic vessels. The vas-
cular anatomy and dissection technique of this fascial flap were
described previously.5 The cervical trachea was reached through
a midline neck incision. The trachea was dissected from the
underlying esophagus and cleared from most of the connective
tissue without incising the trachea. The trachea was still per-
fused in situ but was stripped of its connective tissue envelope
containing part of the blood supply. The ischemic cervical trachea
was wrapped over 2 cm by the transposed lateral thoracic fascial
flap. The lateral thoracic and cervical skin was closed with inter-
rupted sutures.
Stage 2: Tracheal isolation, exteriorization of luminal
site, tracheal re-implantation. After 14 days, the trachea on
its fascial vascular carrier becomes transferable by vascular in-
duction. In the second stage, the neck was reopened and the
cervical trachea with a length of 2 cm was dissected with its new,
thin fascial envelope.
One animal was used to document the amount of revascu-
larization after 14 days. This animal was killed with an overdose
of pentobarbital. The laryngotracheal complex was isolated from
the airway tract on the vascularized fascia flap and was opened
posteriorly. The left axillary region was opened, and the left
lateral thoracic artery was cannulated with a 14-g catheter. The
vasculature was flushed through the cannula with 10 mL of
heparinized normal saline solution (10 IU of heparin per millili-
ter) followed by 5 mL of blue silicone dye (MicrofilR; Canton
BioMedical Products Inc., Boulder, CO). Following these injec-
tions, the cervical trachea could be inspected macroscopically and
the amount of revascularization could be evaluated.
Fig. 1. Revascularization of the rabbit trachea. The left lateral
thoracic fascial flap is dissected and transferred to the neck region.
A 2-cm segment of cervical trachea is wrapped by the fascia flap.
After 2 weeks of fascial wrapping, the lateral thoracic artery was
injected with blue Microfil (arrow) in one rabbit. In the other animals,
the fascial-enwrapped segment of trachea was isolated from the
airway (black lines).
surface areas of the isolated and the reimplanted tracheal trans-
plant. In these animals, the fascial enwrapped tracheal segment
was removed from the airway tract with preservation of the
vascular pedicle of the fascia flap. After isolation, the segment
was opened posteriorly by longitudinal incision of the distal site of
the fascia flap and by longitudinal incision of the membranous
trachea (Fig. 2A). After posterior incision, the tracheal patch was
pinned onto cardboard in a stretched position with exposition of
the mucosal lining (Fig. 2B). After taking a photograph with a
ruler in place, the tracheal patch was made into a tube by closure
of the membranous trachea and the incised fascial flap using
Prolene 6–0. The tracheal tube was reintroduced into its original
position in the airway defect. The animals breathed spontane-
ously during the entire operation. The cervical skin was closed
with interrupted stitches. Antibiotics were administered for 4
postoperative days in both operation stages. Rabbits were housed
in separate quarters in an environment-controlled facility and
were fed a standard rabbit diet.
Stage 3: Control of tracheal segment. The fascia-
enwrapped segment was controlled 4 weeks after reimplantation.
The rabbits (N ϭ 5) were anesthetized, the left axillary region was
opened, and the left lateral thoracic artery was cannulated with a
14-g catheter. The vasculature was flushed through the cannula
with 10 mL of heparinized normal saline solution followed by 5 mL
In the other five control animals, the validity of our new
wound healing model was evaluated by comparing the mucosal
Laryngoscope 111: July 2001
Hardillo et al.: Airway Wound Healing
1175

of blue silicone dye (Microfil). Following these injections, the larynx
and cervical trachea were removed and opened dorsally to allow for
a macroscopic inspection of the reimplanted tracheal segment. The
rabbits were killed with an overdose of pentobarbital. The tracheas
were re-photographed with a ruler in place.
Full-Thickness Mucosal Defects
Anterior patch mucosal defect (N ؍ 5). The first stage
was identical as for the control animals except for the length of
the revascularized trachea (Fig. 2C). A revascularized segment of
trachea with a length of 1.5 cm was used to study the healing of
Fig. 2. Experimental model. (A) Isolation of revascularized tube. A
2-cm segment of cervical trachea is isolated on the fascial vascular
pedicle (arrow) and incised posteriorly. d, distal site of fascia flap; p,
proximal site of fascia flap. (B) Isolated tracheal patch. The tracheal
patch is pinned onto cardboard with exteriorization of the respira-
tory mucosa. Note the normal appearance of the mucosal blood
vessels. No mucosal defect was made in control animals. The
anterior mucosal defect with a width of 0.6 cm is indicated with
white dotted lines. The two mucosal defects (0.6 ϫ 0.6 cm) used to
study the effect of mitomycin are indicated with black boxes. The
upper mucosal defect (U) was treated with mitomycin, whereas the
lower mucosal defect (L) was treated with saline. (C) Schematic
presentation of anterior mucosal defect: 1). The isolated tracheal
segment wrapped with fascia is incised posteriorly. 2) An anterior
mucosal patch of 0.6 cm is removed from the exteriorized mucosal
lining. 3) The tracheal patch is transformed into a tube by suturing of
the incised fascial flap.
Laryngoscope 111: July 2001
1176
Hardillo et al.: Airway Wound Healing

full-thickness mucosal wounds. During the second stage, the re-
vascularized trachea was isolated in an identical way as in con-
trol animals. An anterior mucosal wound was made during the
second operation after exposition of the airway lumen. An ante-
rior defect with a width of 6 mm was made using a scalpel for
mucosal incision and a microscissors for mucosal dissection. He-
mostasis of the exposed cartilage was achieved by even compres-
sion with sterile gauze. Closure and replacement of the trachea
was done in an identical way as for the control animals. Animals
were followed for 4 weeks or until they developed respiratory
distress. After follow-up, the animals were anesthetized and their
vascular pedicle was injected with blue Microfil. The cervical
trachea was removed, opened posteriorly, and photographed. To
measure wound areas, photographic slides taken immediately
after wounding and 4 weeks later were projected. The included
ruler equalized their sizes so that all images were at the same
magnification. Wound areas were determined from tracings by
computer video graphics analysis. Data were expressed as
“wound area” in square millimeters. The wounded trachea was
incised axially for macroscopic and histologic (hematoxylin &
eosin [H&E]) evaluation.
RESULTS
Experimental Model: Control Animals (N ؍ 6)
Documentation of the revascularization process
(N ؍ 1). The mucosal lining of the fascia-enwrapped seg-
ment of trachea was intensely colored with blue silicone
after injection of the lateral thoracic artery. The blue
coloration represents the amount of revascularization af-
ter 14 days of fascial wrapping and after patch formation
(Fig. 3).
Exteriorization of respiratory mucosa (N ؍ 5).
After follow-up, the surface area of the previously isolated
segment could be identified because of the blue coloration
of the mucosal lining (Fig. 4A). No significant difference
was found between the surface area of the patch during
isolation (mean value, 11,500 mm2 - standard deviation
[SD]ϭ 541 mm2) and the patch after follow-up (mean
value, 11,744 mm2 - SD ϭ 555 mm2).
Full-Thickness Mucosal Defects
Circumferential defect (N ؍ 3). The same surgical steps
as for the anterior patch mucosal defect were made. The mucosal
lining of the isolated patch was completely removed during the
second operation stage. After wounding, the tracheal patch was
transformed into a tube. Animals were followed until signs of
dyspnea became apparent. After follow-up, the animals were
euthanized and the tracheas were incised longitudinally for his-
tological (H&E) evaluation.
Anterior patch mucosal defect (N ؍ 5). The sur-
face area of the isolated trachea showed a mean of 8736
Full-Thickness Mucosal Defect After Mitomycin
Application
Anterior patch mucosal defect after mitomycin appli-
cation (N ؍ 7). During the second operation, two full-thickness
mucosal squares (0.6 cm ϫ 0.6 cm) were excised (Fig. 2B). Hemo-
stasis was achieved by even compression with sterile gauze.
The upper mucosal excision was treated with a single topi-
cal application of 0.5 mL (0.2 mg/mL) mitomycin, whereas the
lower mucosal excision was treated with 0.5 mL of saline. Drug
delivery was through a soaked cotton pledget directly applied to
the operative site for a duration of 5 minutes. Five animals were
followed during 2 weeks for morphometric and histologic evalua-
tion. To measure wound areas, photographic slides taken imme-
diately after wounding and 2 weeks later were projected. Wound
areas were determined from tracings by computer video graphics
analysis. Data are expressed as “wound area” in square millime-
ters. The wounded trachea was incised longitudinally for histo-
logic (H&E) evaluation.
Two animals were followed during 1 month for histologic
evaluation. The vascular pedicle (lateral thoracic artery) of the
fascia flap was injected with blue Microfil before tracheal harvest
in all animals.
Circumferential mucosal defect after mitomycin ap-
plication (N ؍ 3). In this group, the completely denuded tra-
cheal segments were treated with 1 mL mitomycin (0.2 mg/mL).
Animals were followed until signs of dyspnea became apparent.
After follow-up, the tracheas were incised longitudinally for his-
tologic (H&E) evaluation.
Differences between treatment groups were analyzed using
the Student t-test or a one-way analysis of variance. Statistical
significance was defined as P Ͻ.05.
All procedures were carried out using facilities and proto-
cols approved by the University of Leuven Institutional Animal
Care and Use Committee.
Fig. 3. Amount of revascularization after 14 days (control animal 1).
The laryngotracheal complex is visible after posterior, longitudinal
incision and after Microfil injection of the vascular pedicle. The
blue-colored areas of the fascial-enwrapped trachea are perfused
by the fascial flap (asterisk).
Laryngoscope 111: July 2001
Hardillo et al.: Airway Wound Healing
1177

Fig. 4. Macroscopy of reimplanted trachea after 1 month. (A) Control
animals. The vascular pedicle of the fascia flap was injected with blue
Microfil 2 weeks after reimplantation of the exteriorized patch. At that time,
the laryngotracheal complex was removed and incised posteriorly with
visualization of the initial isolated patch (completely colored by Microfil). (B)
Anterior mucosal defect. The upper half of the tracheal transplant (opened
posteriorly) and the lower end of the larynx are visible. The intact mucosal
lining (intensely colored with blue Microfil) and the anterior scar (between
arrowheads) is visible. The submucosal tissue is thickest in the center of
the scar (asterisk). The surface area of the wound is reduced by thickening
of the lamina propria and by wound contraction acting on the cartilage
support of the wounded area. (C) Anterior mucosal defect treated with
mitomycin. Macroscopy of treated tracheal segment 14 days after reim-
plantation. The upper, mitomycin-treated defect is denuded over the ma-
jority of the initial defect area, whereas the lower defect is nearly completely
healed. At the site of the upper defect, granulation tissue formation is
completely blocked at the intercartilaginous ligaments.
mm2 (SD, 557 mm2), whereas the surface area after heal-
ing showed a mean of 6908 mm2 (SD, 595 mm2). The
surface area of the defect exhibited a mean of 4392 mm2
(SD, 418 mm2), whereas the surface area after healing
was reduced to 1779 mm2 (SD, 354 mm2). This represents
a 58.9% (mean - SD ϭ 10.5) wound contraction (Fig. 5).
Macroscopic and histologic examination showed that this
major amount of surface area reduction could be attrib-
uted to formation of granulation tissue and to wound
contraction forces acting in the axial axis of the wound
(Figs. 4B and 7A). The thickest layer of granulation tissue
was seen in the middle of the scar.
14, and 17 days. The animals showed respiratory distress
because of excessive granulation tissue formation with
obstruction of the airway lumen. Histologic evaluation
showed excessive granulation tissue formation at the sites
of the intercartilaginous ligaments (Fig. 7B). No reepithe-
lialization from the anastomotic regions was detected.
Full-Thickness Mucosal Defect After Mitomycin
Application
Anterior patch mucosal defect after mitomycin
application (N ؍ 7). As shown in Figures 4C and 7C,
untreated wounds are consistently smaller than the
wounds treated with mitomycin. This result is highly re-
Circumferential defect (N ؍ 3). The three animals
with the circumferential tracheal defects survived for 12,
Laryngoscope 111: July 2001
1178
Hardillo et al.: Airway Wound Healing

Fig. 5. Wound contraction of anterior
mucosal defect. Sqmm, square millime-
ter; ant. def., anterior defect.
producible; in five separate experiments, a single applica-
tion of mitomycin produced wounds that on average were
250% larger than controls, representing a 56% decrease in
closure relative to that seen in control healing (P Ͻ.001)
(Fig. 6). Histologic examination of the control patches
showed a reepithelialized wound with a lamina propria
that was slightly thicker than the lamina propria of the
normal trachea (Fig. 7C). Histologic examination of the
mitomycin-treated patches showed a blockage of angio-
genesis at the sites of the intercartilaginous ligaments.
The bare cartilage rings, which remained exposed to the
airway lumen, underwent necrosis with loss of support.
Two animals were followed during 1 month to evaluate
cartilage viability after longer follow-up. Histologic exam-
ination in these animals revealed cartilage necrosis with
respiratory epithelium between the necrotic cartilage
rings. In contrast to the process of angiogenesis, migration
of respiratory epithelial cells was not inhibited by the
mitomycin application (Fig. 7D).
Circumferential mucosal defect after mitomycin
application (N ؍ 3). The three animals with the circum-
ferential tracheal defects survived for 10, 13, and 14 days.
The animals showed respiratory distress because the bare
cartilage underwent necrosis with loss of support. No
granulation tissue was formed at the luminal site of the
mitomycin-treated cartilage segments.
DISCUSSION
The loss of airway lumen during secondary healing of
full-thickness mucosal defects is extensive. Granulation
tissue formation is the main reason for the loss of airway
lumen in tracheal wounds with preserved vascularization
and cartilaginous support. A process of angiogenesis act-
ing through the intercartilaginous ligaments forms the
Fig. 6. Wound contraction with and
without mitomycin application. Sqmm,
square millimeter.
Laryngoscope 111: July 2001
Hardillo et al.: Airway Wound Healing
1179

Fig. 7. Histology of tracheal wound healing. (A) Anterior defect. The different healing mechanisms of the full-thickness mucosal wound are
visible: 1) Proliferation of granulation tissue formation is blocked after relining with epithelial cells. The thickest layer of granulation tissue is seen
in the middle of the scar. 2) The reepithelialization process is still incomplete in the center of the wound after 1 month. The newly formed
epithelial cells (leading edge) are flattened and cuboidal. 3) Wound contraction forces (small arrows) are generated in the granulating wound
with displacement of the cartilage toward the center of the scar (H&E, original magnification ϫ10). (B) Full circumferential defect. Denuded
tracheal transplant on longitudinal section after 14 days of reimplantation. The place of anastomosis between intact trachea and denuded
trachea is visible (black line). The intact trachea shows a normal mucosal architecture with a lamina propria showing silicone-injected blood
vessels (arrows). The cartilage rings of the denuded trachea are intact and lined with a thick layer of granulation tissue. This granulation tissue
is formed by angiogenic induction from the surrounding fascia flap. Vascularized tissue is growing through the intercartilaginous ligaments
toward the airway lumen. The granulation tissue is not lined with epithelial cells and is thicker at the sites of the intercartilaginous ligaments
(asterisks). F, fascia flap (H&E, original magnification ϫ10). (C) Anterior defect—mitomycin. Histology on longitudinal section 14 days after
reimplantation. The lamina propria of the control defect3 contains smaller blood vessels and is slightly thicker than the lamina propria of the
native trachea.2 The control defect is completely covered by epithelial cells. The regenerated epithelium is one layer thick and has a squamoid
differentiation. The mitomycin-treated defect1 has preserved its initial surface area and the cartilage rings are directly exposed to the airway
lumen (H&E, original magnification ϫ4). Note that the external side of the cartilage rings are more eosinophilic (early cartilage necrosis) than
the internal side of the cartilage rings. (D) Anterior defect—mitomycin, detail after a 1-month healing period. Longitudinal section of
mitomycin-treated defect in animal surviving 1 month. The bare cartilage rings underwent partial necrosis and destruction (external side) with
loss of support (prolapse inside the lumen). The intercartilaginous space between two cartilage rings (asterisk) is lined with respiratory
epithelium. 1 ϭ mitomycin-treated defect, 2 ϭ native trachea (H&E, original magnification ϫ10).
Laryngoscope 111: July 2001
1180
Hardillo et al.: Airway Wound Healing

granulation tissue. The proliferation of vascularized con-
nective tissue is blocked when the regenerated lamina
propria becomes covered with respiratory epithelial cells.
Coverage by respiratory epithelium occurred relatively
fast in the small (0.6 ϫ 0.6 cm) control mucosal defect of
the mitomycin-treated tracheas. These wounds were com-
pletely reepithelialized and the regenerated lamina pro-
pria was only slightly thickened in comparison to the
lamina propria of the undamaged trachea. This healing
pattern, which is characterized by migration and prolifer-
ation of respiratory epithelium, is well known in the lit-
erature.6,7 The relatively fast reepithelialization can be
explained by an epithelial migration process that origi-
nates from the four (left, right, upper, lower) sides of a
relatively small defect.
The healing pattern is different when dealing with
larger mucosal defects. Regeneration and migration of
respiratory epithelium was much slower in the anterior
(0.6 ϫ 1.5 cm) airway defect. In these defects, migrating
epithelium has to come from two (left, right) sides of the
defect. Epithelial regeneration from the upper and lower
anastomotic sides will only be possible for the upper and
lower extremities of the defect. The center of the scar will
have the thickest lamina propria because this area is
lastly covered with respiratory epithelium. Loss of airway
lumen is the result of granulation tissue formation and
wound contraction forces acting in the direction of the
upper surface of the granulating wound. Wound contrac-
tion is a basic biological process and occurs in an incom-
pletely epithelialized defect. It has been defined as the
mechanism by which the edges of a wound are drawn
toward the center resulting from forces generated within
that wound. Wound contraction has been studied inten-
sively for skin wounds.8 In full-thickness rat skin wounds,
it was shown that the wound area diminishes 90% when
the wound is left to granulate, whereas wounds with full-
thickness skin grafts preserve there wound surface area.
Contraction is a vital aspect of open wound healing. The
size of the tissue defect is reduced so that a lesser degree
of connective tissue deposition and epithelialization is re-
quired. Myofibroblasts are the cells thought to be respon-
sible for this phenomenon.8 The wound contraction forces
acting during healing of the anterior mucosal wound (0.6
ϫ 1.5 cm) resulted in a reduction of the curvature of the
cartilage rings which resulted in a further loss of airway
lumen.
have a greater impact on the reduction of the airway
lumen than the length of the defect. The loss of airway
lumen resulting from healing of the anterior mucosal de-
fect (wound measuring 0.6 ϫ 1.5 cm) was extended over a
length of 1.5 cm and had no or minimal influence on the
respiration of the animals. The loss of airway lumen of a
circumferential defect of 0.6 cm would, however, inevita-
bly lead to respiratory distress because the loss of airway
lumen would occur over only 0.6 cm.
The problematic migration of respiratory cells from
the wound margins seen in the anterior (0.6 ϫ 1.5 cm) and
in the circumferential mucosal defect is in contrast to
optimistic reports in the literature concerning the possi-
bility of respiratory epithelial cell migration.11 Optimistic
reports on migration of respiratory epithelium are based
on the fast recovery of epithelium in small defects, on the
fast recovery of epithelium in superficial mucosal wound
healing with preservation of the basement membrane, and
on in vitro studies.12
In our model, topical application of mitomycin
blocked the angiogenic process and inhibited granulation
tissue formation at the internal site of the defect. Mito-
mycin is known to act as an agent for scar inhibition and
has gained wide acceptance in ophthalmology13,14 and
otolaryngology3,4,15 for procedures in which scarring is
problematic. In our experimental model we were able to
show that the drug acts by blocking angiogenesis but
without inhibition of the migration of respiratory epithe-
lium. In our model, mitomycin had a negative influence on
airway healing because the bare cartilage rings developed
necrosis with loss of support. It seems that mitomycin can
safely and effectively be applied on scar tissue in the
treatment of airway stenosis. However, warning is neces-
sary when dealing with bare cartilage because of the risk
for cartilage necrosis when granulation tissue formation is
blocked.
CONCLUSION
The following conclusions were made concerning the
etiology of airway stenosis and the requirements for la-
ryngotracheal repair:
● Full-thickness damage of the respiratory epithe-
lium leads to narrowing and eventual stenosis of the air-
way lumen.
● Long-term segmental tracheal reconstruction with
denuded vascularized tracheal autografts is impossible
because of granulation tissue formation in the untreated
transplants and because of cartilage necrosis after mito-
mycin application.
● Restoration of the mucosal component is important
in airway reconstruction. Secondary healing leads to
reepithelialization but the resulting granulation tissue
formation and wound contraction will have a negative
impact on the airway lumen.
In the circumferential mucosal defect, reepithelial-
ization is only possible by new growth of epithelium from
the upper and lower anastomosis. Animals with a circum-
ferential defect survived only a short period of 14 days
because of excessive granulation tissue formation without
signs of reepithelialization.
It has been reported that damage of the perichon-
drium and the cartilage is necessary for an airway steno-
sis to develop.9,10 From the results obtained in this study,
we conclude that a full-thickness mucosal tracheal defect
will be sufficient for obtaining a loss of airway lumen.
Whether the healing process will lead to a stenosis de-
pends on the length and the circumferential extent of the
defect. A long, circumferential defect will result in airway
stenosis. The circumferential extent of the wound will
Acknowledgments
This work was supported by a grant of the Research
Council from the University of Leuven, Belgium.
Laryngoscope 111: July 2001
Hardillo et al.: Airway Wound Healing
1181

8. Rudolph R. Contraction and the control of contraction. World
J Surg 1980;4:279–287.
9. Eliashar R, Eliachar I, Gramlich T, Esclamado R, Strome M.
Improved canine model for laryngotracheal stenosis. Oto-
laryngol Head Neck Surg 2000;122:84–90.
10. Borowiecki B, Croft CB. Experimental animal model of sub-
glottic stenosis. Ann Otol 1977;86:835–840.
11. Cheng ATL, Backer CL, Holinger LD, Dunham ME, Mavrou-
dis C, Gonzalez-Crussi F. Histopathologic changes after
pericardial patch tracheoplasty. Arch Otolaryngol Head
Neck Surg 1997;132:1069–1072.
12. Zahn JM, Chevillard M, Puchelle E. Wound repair of human
surface respiratory epithelium. Am J Respir Cell Mol Biol
1991;5:242–248.
13. Chen CW. Trabeculectomy with simultaneous topical appli-
cation of mitomycin-C in refractory glaucoma. J Ocul Phar-
macol Ther 1990;6:175–182.
14. Singh G, Wilson MR, Foster CS. Mitomycin eye drops as
treatment for pterygium. Ophthalmology 1988;95:
813–821.
15. Ward RF, April MM. Mitomycin-C in the treatment of tra-
cheal cicatrix after tracheal reconstruction. Int J Pediatr
Otorhinolaryngol 1998;44:221–226.
BIBLIOGRAPHY
1. Borowiecki B, Croft CB. Experimental animal model of sub-
glottic stenosis. Ann Otol Rhinol Laryngol 1977;86:
835–840.
2. Charous SJ, Ossoff RH, Renisch L, Davidson JM. Canine
subglottic stenosis as a model for excessive fibrosis: a pilot
histologic and immunohistochemical analysis. Wound Re-
pair and Regeneration 1996;4:444–453.
3. Correa AJ, Reinsch L, Sanders DL, et al. Inhibition of sub-
glottic stenosis with mitomycin C in the canine model. Ann
Otol Rhinol Laryngol 1999;188:1053–1060.
4. Eliachar R, Eliachar I, Esclamado R, Gramlich T, Strome M.
Can topical mitomycin prevent laryngotracheal stenosis?
Laryngoscope 1999;109:1594–1600.
5. Delaere PR, Liu ZY, Feenstra L. Tracheal autograft revascu-
larization and transplantation. Arch Otolaryngol Head
Neck Surg 1994;103:212–215.
6. Zahn JM, Kaplan H, Herard AL, et al. Cell migration and
proliferation during the in vitro wound repair of the respi-
ratory epithelium. Cell Motil Cytoskeleton 1997;37:33–43.
7. Zahn JM, Pierrot D, Chevillard M, Puchelle E. Dynamics of
cell movement during the wound repair of human surface
respiratory epithelium. Biorheology 1992;29:459–465.
Laryngoscope 111: July 2001
1182
Hardillo et al.: Airway Wound Healing