Original Research – Technical Article

Optimization and validation of an ischemic wound model

LISA J. GOULD, MD, PhD; MIMI LEONG, MD; JOSEPH SONSTEIN, BS; SHELLY WILSON, BS

Localized tissue ischemia is a key factor in the development and poor prognosis of chronic wounds. Currently,
there are no standardized animal models that provide sufficient tissue to evaluate the effect of modalities that
may induce angiogenesis, and in vitro models of angiogenesis do not mimic the complexity of the ischemic
wound bed. Therefore, we set out to develop a reproducible ischemic model for use in wound-healing studies.
Male Sprague–Dawley rats underwent creation of dorsal bipedicle skin flaps with centrally located excisional
wounds. Oxygen tension, wound-breaking strength, wound area, lactate, and wound vascular endothelial
growth factor (VEGF) were compared in flaps measuring 2.5 and 2.0
11 cm with and without an underlying
silicone sheet. We found that the center of the 2.0 cm flap with silicone remains in the critically ischemic range
up to 14 days without tissue necrosis (3374 vs. 4976 mmHg in controls). Wound healing and breaking strength
were significantly impaired and tissue lactate from the center of this flap was 2.9 times greater than tissue from
either nonischemic controls and 2.5 cm flap (0.2370.05 mg/dL/mg sample vs. 0.0970.02 and 0.0870.02, re-
spectively). Vascular endothelial growth factor was 2 times greater than the nonischemic control. This ischemic
wound model is relatively inexpensive, easy to perform, reproducible, and reliable. The excisional wounds pro-
vide sufficient tissue for biochemical and histologic analysis, and are amenable to the evaluation of topical
and systemic therapies that may induce angiogenesis or improve wound healing. (WOUND REP REG
2005;13:576–582)

The complexity of the interactions between oxygen and

oxidants, and the effect on wound repair is not fully un-
derstood. Growing evidence is accumulating that proper
cellular function, and therefore optimal wound healing,
depends upon the cellular redox potential.1 The center
of the wound is characterized by hypoxia, elevated lac-
tate, and high oxidant flux. These three elements are
known to stimulate the production of angiogenic fac-
tors. Because angiogenesis reverses wound hypoxia and
provides oxygen for the high metabolic demands of
wound repair, it is critical to proper wound healing.
From the Division of Plastic Surgery, Department of Sur-
gery, The University of Texas Medical Branch,
Galveston, Texas.
Manuscript submitted: December 16, 2004
Accepted in final form: March 8, 2005
Reprint requests: Lisa J. Gould, MD, PhD, Department of
Surgery, The University of Texas Medical Branch,
301 University Boulevard, Galveston, TX 77555-
0724. Fax: (409) 772-1872; Email: ljgould@
utmb.edu
Copyright r 2005 by the Wound Healing Society.
ISSN: 1067-1927.
ELISA Enzyme-linked immunosorbant assay
PscO2 Subcutaneous oxygen tension
VEGF Vascular endothelial growth factor
The quandary is that despite the presence of hypoxia,
elevated lactate, and high oxidant flux, which should
stimulate angiogenesis, ischemic tissue heals poorly, and
has increased susceptibility to wound infection.2–4 In or-
der to test the hypothesis that there is a critical level of
molecular oxygen required to induce angiogenesis and
promote healing of ischemic wounds, we developed the
model described in this article.
Commonly used model systems to study angiogen-
esis include in vitro assays, such as endothelial cell pro-
liferation and migration, ex vivo systems, such as the
chicken chorioallantoic membrane assay and the rat
aortic ring assay, and in vivo tumor models.5–7 How-
ever, these models do not completely mimic the com-
plex interactions with the extracellular matrix and
signals from surrounding cells in the wound bed, nor
do they mimic the tissue reaction to impaired perfusion.
Standardized animal models of ischemic wounds that
are easy to perform, reliably reproduce tissue ischemia,
eliminate craniocaudal variation, and are amenable to

576
WOUND REPAIR AND REGENERATION
VOL. 13, NO. 6
studying therapeutic modalities are currently lacking.
The ischemic rabbit ear dermal ulcer model, while ele-
gant in design, requires use of an operating microscope
and heals in 7–10 days.8 A modification of this model,
developed to more closely simulate a chronic wound, re-
quired reoperating on the rabbits on days 5, 10, 17, and
21.9 Furthermore, this model depends on the large rabbit
ear and has not been successfully adapted to either rats
or mice, limiting the availability of molecular probes. We
have developed a longitudinally oriented, dorsal, biped-
icle flap model that addresses these criteria and will
prove to be a valuable model for studying tissue ischemia.
This model is a modification of the ischemic wound
model originally described by Schwartz et al.10 and sub-
sequently utilized in modified form by Chen et al.11 Full-
thickness excisional wounds are created within a biped-
icle dorsal skin flap in rats. Two critical changes have
been made, namely making the skin flap sufficiently
narrow, so the blood supply is random and the wounds
located in the midpoint of the flap are ischemic, and in-
serting a silicone sheet beneath the skin flap, which pre-
vents readherence and reperfusion of the flap from the
underlying tissue.
Rodents as animal models are practical because of
their low cost, ease of care and handling, and availability
of molecular probes for tissue analysis. However, ro-
dents have a subcutaneous panniculus carnosus muscle,
which has been shown to contribute to healing by al-
lowing substantial wound contraction, and which also
contributes collagen to the wound bed.12,13 In this mod-
el, we remove the panniculus carnosus muscle from the
wound bed by dissecting just above the muscle fascia.
Wound contraction is limited but not eliminated by tack-
ing the flap to the silicone sheet. The excisional wound
model has the advantage of providing sufficient tissue
for cellular and molecular analysis and can be used to
study the effect of topical or systemic agents.
To validate and document the degree of ischemia
achieved with this model, we have determined the tissue
oxygen tension at the level of the wound, measured
wound surface area, tissue lactate, and tensile strength.
These parameters have been compared in 2.0 and 2.5 cm
flaps with and without an intervening sheet of silicone
and normal skin. In addition, we show that the excisional
wounds provide sufficient tissue for molecular and cel-
lular analysis by creating extracts of the wound bed and
determining the amount of excised tissue, protein con-
tent, and vascular endothelial growth factor (VEGF) con-
tent by enzyme-linked immunosorbant assay (ELISA).
MATERIALS AND METHODS
Healthy male Sprague–Dawley rats (Charles River Labo-
ratory, Wilmington, MA) weighing 250–300 g were utilized.
All procedures were approved by the institution’s Animal
Care Committee and abided by all requirements of the
GOULD ET AL. 577
Animal Welfare Act. Rats were anesthetized with inhala-
tional isoflurane. Their backs were shaved and a template
was used to outline the flap with a surgical marking pen.
The template, used to reduce interanimal variability was
centered along the spinal column and placed between the
base of the scapulae and the iliac crest. Marks are also
made in the center of the flap at 5.0 cm from the top to
indicate planned wound placement. This placement is just
cranial to the vertical midline of the flap and corresponds
to the greatest delay in healing as noted in Figure 1 of
Chen et al.11 The rats were then sprayed liberally with
topical Betadine and the entire surgery performed under
aseptic technique. A single dose of ampicillin (15 mg/kg)
was administered subcutaneously prior to making the in-
cision. Thirty-nine rats were divided into five groups: con-
trol, no flap, n 5 6; 2.5 cm flap without silicone, n 5 6;
2.5 cm flap with silicone, n 5 11; 2.0 cm flap without sil-
icone, n 5 6; 2.0 cm flap with silicone, n 5 10.
Two adjacent 6 mm full-thickness excisional wounds
were created in the center of the flap using a sterile 6 mm
disposable biopsy punch. Control animals received exci-
sional wounds at the same craniocaudal location as those
in which flaps were performed. The depth of excision
was down to, but not through, the anterior fascia of the
panniculus carnosus. The full-thickness punch biopsy,
including skin and panniculus carnosus muscle, was re-
moved by dissecting with scissors in the plane between
the panniculus carnosus and the fascia. A dorsal, biped-
icle skin flap was raised in the craniocaudal direction
deep to the panniculus carnosus muscle. Precut and ster-
ilized nonreinforced 0.01 in. thickness Sil-Tec medical
grade sheeting (Technical Products Inc. of Georgia, De-
catur, GA) was then placed underneath the flap. The sil-
icone strips were precut to 2.5
10 cm for the 2.5 cm
wide flaps and 2.0
10 cm for the 2.0 cm wide flaps to
ensure proper fit under the skin flap without buckling or
folding. The skin flaps and silicone sheet were sutured to
the adjacent skin edges with interrupted nonabsorbable
sutures. Suturing the flap, silicone, and skin edge to each
other along the length of the flap prevents movement of
the silicone sheet. We have also noted that the silicone
sheet inhibits wound contraction. Internal control, non-
ischemic full-thickness wounds were created with the 6
mm biopsy punch, 0.5 cm lateral to the flap, at the same
craniocaudal location as the ischemic wounds (Figure 1).
A sterile occlusive dressing (Tegaderm, 3M Health Care,
St. Paul, MN) was applied to cover all four wounds. The
rats were awakened from general anesthesia, extubated,
and placed in individual cages. They were allowed stand-
ard laboratory rat chow and water ad libitum.
Determination of subcutaneous oxygen tension
Twenty-four hours prior to measurement, the rats were
briefly anesthetized with inhalational isoflurane. Utili-
zing aseptic technique, a 16-gauge angiocatheter was

578 GOULD ET AL.
Tissue for Lactate
Ischemic Wounds
Internal Control Wounds
Tissue for Breaking Strength
11 cm
2 cm
FIGURE 1. Schematic diagram of the ischemic wound-healing
model. A bipedicle skin flap measuring 2.5 or 2.0
11 cm was
centered on the dorsum of the rat from the base of the scap-
ulae to the iliac crest. Two full-thickness, 6 mm excisional wounds
were created in the center of the flap. The panniculus carnosus
fascia was left intact in the base of the wounds. Insertion of a
silicone sheet beneath the skin flap prevented readherence
and reperfusion of the flap from the underlying tissue. Control
animals had excisional wounds at the same craniocaudal lo-
cation without creation of the flap. Two 6 mm full-thickness
wounds adjacent (lateral) to the flap served as internal, non-
ischemic controls. The wounds were covered with an occlusive
dressing for the first 7 days.
threaded above the panniculus carnosus such that the
tip was located between the two wounds on the flap.
This was flushed with 0.2 mL of sterile saline, capped,
and sutured in place. On day 14, a polarographic oxygen
electrode and a temperature probe (Licox, Haut Medical
Technology, Sea Cliff, NY) were passed through the pre-
viously placed angiocatheter to provide continuous
oxygen measurements for a minimum of 15 minutes.
Anesthesia was not required during this measurement.
Tissue sampling
With the rats under inhalational isoflurane anesthesia,
the wounds were traced to determine surface area. As
illustrated in Figure 1, one ischemic wound was har-
vested in a longitudinal strip for tensiometry. The other
ischemic wound and one nonischemic internal con-
trol wound were excised from each animal and snap
frozen in liquid nitrogen for VEGF analysis. Tissue for
lactate analysis was harvested from the skin bridge be-
tween the two ischemic wounds and from nonischemic
skin lateral to the flap. Blood for lactate analysis was
collected by cardiac puncture and the animals were
euthanized.
Wound-breaking strength
Wounds were cut in longitudinal strips with the wound
in the center of the strip. The width and thickness of
each strip were measured. The strips were pulled to
failure using an Instron Materials Testing device 4201
WOUND REPAIR AND REGENERATION
NOVEMBER–DECEMBER 2005
(Instron Corp, Canton, MA) with a 50 N load cell at a
crosshead speed of 10 cm/min.
Lactate analysis
Tissue stored at 801C was homogenized in 3 M HClO4
using a handheld glass tissue grinder and transferred
to tubes in a 8 to 101C ice/salt bath and agitated un-
til mixed. One milliliter H2O/0.3 mL HClO4 was added
and the suspension mixed at 41C for 10 minutes. Fol-
lowing centrifugation at 11.500 rpm for 10 minutes at
41C, the supernatant was removed and stored at 701C.
L-lactate was determined spectrophotometrically at 340
nm using a commercially available kit (Sigma-Aldrich,
St. Louis, MO) which follows the production of reduced
nicotinamide adenine dinucleotide (NADH) according
to the following reaction: pyruvate1NADH # NAD11
lactate.
VEGF analysis
Tissue stored at 801C was homogenized in tissue lysis
buffer (10 mM Tris, pH 7.4, 150 mM NaCl, 1% Triton).
Following centrifugation at 14,000 r.p.m. for 20 minutes,
the supernatant was removed and stored at 701C.
VEGF levels were determined using a commercially
available ELISA kit (R&D Systems, Minneapolis, MN)
according to the manufacturer’s instructions. Briefly,
samples were diluted (1 : 5) with calibrator diluent as
per manufacturer’s instructions and then loaded onto 96-
well plates. Recombinant mouse VEGF was used to cre-
ate the standard curve. Following addition of the conju-
gate and substrates, the reaction was stopped with the
stop solution and the optical density read on a Bio-Tek
EL800 microplate reader at 450 nm (correction wave-
length set at 540 nm; Winooski, VT). Protein levels in the
tissue homogenates were determined using a commer-
cially available kit according to the BCA (bicinchoninic
acid) method (Pierce Biotechnology, Rockford, IL).
VEGF levels were expressed as picograms per milligram
protein.
Statistical analysis
Data are presented as mean7SEM. One-way analysis of
variance with post hoc comparisons using Tukey’s high-
ly significant difference test was used to determine the
statistical significance between groups using InStat ver-
sion 3.0 (Graphpad Software, San Diego, CA). A p-value
of less than 0.05 was considered significant.
RESULTS
The subcutaneous oxygen tension (PscO2) at the level of
the wound was compared in 2.0 and 2.5 cm flaps with an
intervening sheet of silicone, in flaps without silicone,
and in normal skin (Figure 2A). Control, nonischemic
WOUND REPAIR AND REGENERATION
VOL. 13, NO. 6 GOULD ET AL. 579
A 80 Rate of wound healing and development of
70 breaking strength
60 Wound surface area at day 14 was determined by tracing
pO2 (mmHg)

50 the wound edge and converting the tracings to digital

images with Sigma Scan (Sigma Scan/Image Image Anal-
40
ysis, Version 1.20.09, Jandel Scientific, San Rafael, CA).
30
The results show that wounds on the 2.0 cm flap with
20 silicone remain significantly larger at 2 weeks compared
10 with all other groups, including the 2.0 cm flap without
0 silicone (0.12470.019 vs. 0.03970.018 cm2, po0.01),
Control 2.5 cm 2.5 cm 2.0 cm 2.0 cm and that there is minimal difference between control
flap flap flap flap
(no silicone) (no silicone) and the wounds on the 2.5 cm flaps (Figure 3).
Wound-breaking strength was measured in one is-
B 50 chemic wound per rat. As depicted in Figure 4, the
breaking strength decreased as the tissue was made
40 more ischemic by placement of the silicone sheet and
narrowing the flap. The breaking strength of the 2.0 cm
flap wounds was 50 percent that of nonischemic control
pO2 (mmHg)

30 wounds (4.370.7 vs. 8.970.6 N, po0.001).

20
10 With Silicone
Without Silicone
0 ischemic wounds (Figure 5). The 2.5 cm flap tissue lac-
Day 3 Day 7 Day 10 Day 14
Change in pO2 Over Time
FIGURE 2. PscO2 levels in the ischemic wounds. (A) Day 14 ox-
ygen tension. Oxygen tension was measured by placing a po-
larographic electrode in the subcutaneous tissue between the
two wounds. Anesthesia was not required during this measure-
ment. n 5 6 for control (no flap), n 5 6 for 2.5 cm flap without
silicone, n 5 7 for 2.5 cm flap with silicone, n 5 6 for 2.0 cm flap
without silicone, and n 5 6 for 2.0 cm flap with silicone. (B)
Change in pO2 over time. The temporal nature of the flap is-
chemia was examined by comparing PscO2 in the 2.0 cm flap
with (n 5 3) and without silicone (n 5 6). Addition of the inter-
vening silicone decreases day-to-day variability and increases
Tissue lactate and VEGF content
Blood lactate from anesthetized rats was 0.03 mg/dL.
This was compared with lactate levels from non-
ischemic tissue and from the tissue bridge between the
tate was not significantly different from the nonischemic
controls (0.0970.02 vs. 0.0870.02 mg/dL/mg sample).
The mean tissue lactate in the 2.0 cm flap with silicone
was 2.9 times greater than in the 2.5 cm flap (0.2370.05
vs. 0.0870.02 mg/dL/mg sample, p 5 0.063) and 1.5
times greater than in the 2.0 cm flap without silicone
(0.2370.05 vs. 0.1570.02 mg/dL/mg sample).
0.25
*†‡§
0.20
Surface Area (cm2)

ischemia.
0.15

0.10
tissue pO2 was unchanged between days 7 and 14 (un-
published data). The PscO2 at the level of the wounds in 0.05
the 2.5 cm flap without silicone was equivalent to con-
trol. Placement of the silicone sheet reduced this by 7 0.00
Control 2.5 cm 2.5 cm 2.0 cm 2.0 cm
mmHg, and narrowing the flap to 2.0 cm reduced the flap flap flap flap
oxygen tension to the equivalent of what would be ex- (no silicone) (no silicone)
pected in the unwounded tissue of an ischemic extrem- FIGURE 3. Day 14 wound surface area. The ischemic flap
ity (TcPO2o39 mmHg)14 and correlates well with the wounds are compared with nonischemic controls (no flap)
PscO2 values reported for the rabbit ear model (non- and with flaps without the intervening silicone sheet. (po0.05
ischemic ear 5 5378 mmHg, ischemic ear 5 2779 compared with control) n 5 6 for control, n 5 7 for 2.5 cm
flap without silicone, n 5 5 for 2.5 cm flap with silicone, n 5 6
mmHg at day 14).15 While there was no significant
for 2.0 cm flap without silicone and n 5 8 for 2.0 cm flap with sil-
change in oxygen tension when silicone was placed un- icone.  5 vs. Control po0.001, w 5 vs. 2.5 cm with silicone po
der the 2.0 cm flap, there was less variability in tissue 0.05, z 5 vs. 2.5 cm flap po0.05, y 5 vs. 2.0 cm without silicone
pO2 for the duration of the experiment (Figure 2B). po0.01.
 WOUND REPAIR AND REGENERATION
580 GOULD ET AL. NOVEMBER–DECEMBER 2005

12 120

10 100
Tensile Strength (N)

VEGF (pg/mg protein)

8 *† 80
*†
6
60

4
40

2
20

0
Control 2.5 cm 2.5 cm 2.0 cm 2.0 cm 0
flap flap flap flap 2.5 cm 2.5 cm 2.0 cm 2.0 cm 2.0 cm
(no silicone) (no silicone) Control flap Control flap flap
(no silicone)
FIGURE 4. Wound tensile strength decreases with increasing is-
chemia. Tensiometry of ischemic flap wounds is compared with FIGURE 6. Effect of increasing tissue ischemia on wound vascu-
wounds on control animals that did not have a flap raised. As lar endothelial growth factor (VEGF) content. VEGF content of
the flap is made more ischemic, by addition of a silicone sheet nonischemic internal control wounds is compared with ischemic
and narrowing the flap, the tensile strength decreases. (po wounds on the 2.5 and 2.0 cm flaps. n 5 8 for nonischemic in-
0.001 vs. control, wpo0.05 vs. 2.5 cm flap without silicone) n 5 6 ternal control wounds, n 5 11 for 2.5 cm flaps, n 5 8 for 2.0 cm
for control, n 5 6 for 2.5 cm flap without silicone, n 5 7 for 2.5 cm control wounds, n 5 5 for 2.0 cm flap without silicone, and n 5 10
flap with silicone, n 5 6 for 2.0 cm flap without silicone, and n 5 6 for 2.0 cm flap with silicone. VEGF levels are expressed as mean
for 2.0 cm flap with silicone.  5 vs. Control po0.01, w 5 vs. 2.5 cm pg/mg protein7SEM.
with silicone po0.05.

was equal to that in the nonischemic internal control

VEGF levels were measured in wound homogenates
by commercial ELISA. Wounds in the 2.5 cm flap with
silicone were compared with the 2.0 cm flap with and
without silicone and also compared with their corre-
sponding nonischemic control wounds (Figure 6). At 14
days, the mean VEGF level in the 2.5 cm flap wounds
0.30
*†**
0.25
Lactate (mg/dl/mg sample)
wounds. In contrast, the mean VEGF level in the 2.0 cm
flap with silicone wounds was 84719 pg/mg compared
with 4175 pg/mg in the internal controls (p 5 0.095) and
48715 pg/mg in the 2.0 cm flaps without silicone.
DISCUSSION
The choice of animal models to mimic the human con-
dition is based on a compromise of cost, ease of use,
reproducibility, and reliability of the data. The rat model
has the advantages of ease of use and low cost. How-

ever, wound healing in rats has been subject to scrutiny

0.20 because of their ability to heal infected wounds and the
high rate of interanimal variability. The model presented
0.15 here, i.e., excisional wounds centered on a 2.0 cm bi-
** pedicle flap with an intervening silicone sheet, has been
* †
0.10 standardized to reduce interanimal variation. We have
trained medical students and undergraduates to perform
0.05 this model with good reproducibility and flap survival.
The rat skin wound model described by Chen et al.11
0.00 has a molecular profile similar to that of chronic human
Blood 2.5 cm 2.5 cm 2.0 cm 2.0 cm 2.0 cm wounds. This model uses a 2.5 cm bipedicle flap without
Internal flap Internal flap flap
Control Control (no silicone) a silicone interface and six paired wounds on the flap.
FIGURE 5. Tissue lactate levels at day 14 correlate with the de-
That there is a significant difference in the rate of heal-
gree of ischemia. Comparison of normal blood lactate was ing depending on the craniocaudal location is graphical-
made with nonischemic tissue and with the skin bridge between ly shown in their article.11 Our modifications have
the two wounds. Data are shown as mean mg/dL/mg tissue expanded upon this model to rigorously document and
homogenate7SEM. n 5 6 for nonischemic control tissue, n 5 10 reproduce the degree of ischemia in the wound bed. We
for 2.5 cm flap without silicone, n 5 11 for 2.5 cm flap with sili-
cone, n 5 6 for 2.0 cm flap without silicone, n 5 10 for 2.0 cm flap have demonstrated that the 2.5 cm flap without silicone
with silicone, and n 5 2 for blood lactate.  5 o 0.01,  5 po is not ischemic compared with controls, but does have a
0.05, w 5 po0.01. slower rate of healing. The addition of an intervening
WOUND REPAIR AND REGENERATION
VOL. 13, NO. 6
silicone sheet decreases tissue oxygen slightly, but does
not impact upon other parameters of wound healing.
Only by further narrowing the flap to 2.0 cm are we able
to document biochemical and mechanical evidence that
correlates with tissue ischemia.
It is well accepted that wound healing is impaired
and susceptibility to infection is increased when the tis-
sue pO2 is less than 40 mmHg.2,16 A transcutaneous
oxygen pressure of less than 20 mmHg has been shown
to be a strong predictor of failure, whereas between 20
and 40 mmHg healing may occur, but is unpredicta-
ble.17,18 The goal in creating this model was to develop a
flap that would remain viable while seriously impairing
wound healing. Five of the six 2.0 cm flaps with silicone
had PscO2 values in the critically ischemic range (19–40
mmHg) 14 days after the flaps were created; however,
no tissue necrosis was observed in any flap.
There are several technical points that are important
to achieve consistency with this model. We have found it
important to create the wounds prior to flap elevation.
This reduces trauma to the critically ischemic flap. The
technique of punch biopsy down to, but not through, the
panniculus carnosus fascia requires some practice, but is
easily accomplished and permits a viable wound bed
over the silicone. Finally, application of a dressing for at
least the first 3–5 days prevents contamination of the
wounds. By using Mastisol (Ferndale Laboratories, Inc.,
Ferndale, MI), we have achieved good dressing adher-
ence that prevents manipulation by the rats.
Nonetheless, there is still variability, particularly in
the biochemical analysis of the tissue extracts. This may
be because of several factors: variability in excision of
the wound bed with retention of some normal tissue,
variability in homogenization, and inherent interanimal
variability. Factors 1 and 2 are purely technical. We feel
that our excision is as precise as possible. Liquid nitro-
gen pulverization of tissue has ensured that homogeni-
zation is complete. That leaves inherent interanimal
variability, which is the most plausible explanation. In
humans, a pO2 of 30 mmHg is considered critically is-
chemic, but some patients heal and some do not.17,18 We
have found that making the flap narrower than 2.0 cm or
traumatizing the flap results in frank necrosis, indicating
that even minor variations, such as temperature or
stress level, could contribute to biochemically detecta-
ble variation.
Tissue lactate accumulation can occur as a result of
either active aerobic or anaerobic glycolysis by leuko-
cytes and macrophages in the wound bed. As tissue is-
chemia occurs, oxygen and glucose delivery is reduced.
This results in a switch to anaerobic metabolism result-
ing in the elevation of lactate. Lactate in wound cham-
bers has been shown to be 10–20 times that of venous
blood.16 We have measured tissue lactate in the tissue
adjacent to the wounds to show the degree of tissue is-
chemia in the peri-wound tissue. It is expected that lac-
GOULD ET AL. 581
tate levels in the actual wound bed would be even
higher.
Ischemia and elevated lactate stimulate the secre-
tion of angiogenic factors.19, 20 Analysis of wound tissue
homogenates provides a means to assay for these fac-
tors in the wound bed. We have demonstrated that at a
single time point (14 days) VEGF is elevated in the 2.0
cm flap with silicone wounds compared with the inter-
nal controls and to the 2.5 cm flap wounds. This corre-
lates with the degree of ischemia and elevated lactate.
Each excised wound weighs 40–50 mg and provides 400
ml of extract containing an average of 2.8 mg of protein.
This is enough to evaluate four to five different factors
by ELISA or Western blot. Likewise, one half of a wound
provides sufficient tissue for total RNA extraction.
These results show that the longitudinally oriented,
2.0 cm wide, bipedicle flap with intervening silicone is a
reliable model of prolonged tissue ischemia. The model
provides two ischemic wounds and two nonischemic
control wounds at the same craniocaudal location. The
amount of wound tissue is sufficient to correlate gross
changes in wound size and tensile strength with histo-
logic and biochemical changes in the wound bed. We
anticipate that this model can be adapted to examine the
effect of ischemia on aged animals and may be per-
formed in either mice or rats. The excisional wounds are
amenable to treatment with either topical or systemic
agents and will allow us to objectively and systemati-
cally evaluate the effect of putative angiogenic factors
on angiogenesis and wound repair.
ACKNOWLEDGMENTS
The authors thank Iosef Mushkudiani and Qing Chang
for expert technical assistance and Brent Bell for instal-
lation and calibration of the Licox system.
REFERENCES
1. Sen CK, Khanna S, Babior BM, Hunt TK, Ellison EC, Roy S.
Oxidant-induced vascular endothelial growth factor expression
in human keratinocytes and cutaneous wound healing. J Biol
Chem 2002;277:33284–90.
2. Hopf HW, Hunt TK, West JM, Blomquist P, Goodson WH III, Jensen
JA, Jonsson K, Paty PB, Rabkin JM, Upton RA, von Smitten K,
Whitney JD. Wound tissue oxygen tension predicts the risk of
wound infection in surgical patients. Arch Surg 1997;132:997–1005.
3. Dowd GS, Linge K, Bentley G. Measurement of transcutaneous
oxygen pressure in normal and ischaemic skin. J Bone Joint Surg
Br 1983;65:79–83.
4. Dowd GS. Predicting stump healing following amputation for pe-
ripheral vascular disease using the transcutaneous oxygen moni-
tor. Ann R Coll Surg Engl 1987;69:31–5.
5. Celerier J, Cruz A, Lamande N, Gasc JM, Corvol P. Angiotensino-
gen and its cleaved derivatives inhibit angiogenesis. Hypertension
2002;39:224–8.
6. Auerbach R, Kubai L, Knighton D, Folkman J. A simple procedure
for the long-term cultivation of chicken embryos. Dev Biol
1974;41:391–4.

582 GOULD ET AL.
7. Zhu WH, MacIntyre A, Nicosia RF. Regulation of angiogenesis by
vascular endothelial growth factor and angiopoietin-1 in the rat aorta
model: Distinct temporal patterns of intracellular signaling correlate
with induction of angiogenic sprouting. Am J Pathol 2002;161:823–30.
8. Ahn ST, Mustoe TA. Effects of ischemia on ulcer wound healing: A
new model in the rabbit ear. Ann Plast Surg 1990;24:17–23.
9. Bonomo SR, Davidson JD, Tyrone JW, Lin X, Mustoe TA. Enhance-
ment of wound healing by hyperbaric oxygen and transforming
growth factor beta3 in a new chronic wound model in aged rabbits.
Arch Surg 2000;135:1148–53.
10. Schwartz DA, Lindblad WJ, Rees RS. Altered collagen metabolism
and delayed healing in a novel model of ischemic wounds. Wound
Rep Reg 1995;3:204–12.
11. Chen C, Schultz GS, Bloch M, Edwards PD, Tebes S, Mast BA.
Molecular and mechanistic validation of delayed healing rat
wounds as a model for human chronic wounds. Wound Rep Reg
1999;7:486–94.
12. Davidson JM. Animal models for wound repair. Arch Dermatol Res
1998;290 (Suppl.):S1–11.
13. Cohen IK, Moore CD, Diegelmann RF. Onset and localization of
collagen synthesis during wound healing in open rat skin wounds.
Proc Soc Exp Biol Med 1979;160:458–62.
WOUND REPAIR AND REGENERATION
NOVEMBER–DECEMBER 2005
14. Padberg FT, Back TL, Thompson PN, Hobson RW II. Transcutane-
ous oxygen (TcPO2) estimates probability of healing in the
ischemic extremity. J Surg Res 1996;60:365–9.
15. Siddiqui A, Davidson JD, Mustoe TA. Ischemic tissue oxygen ca-
pacitance after hyperbaric oxygen therapy: A new physiologic
concept. Plast Reconstr Surg 1997;99:148–55.
16. Hunt TK. The physiology of wound healing. Ann Emerg Med
1988;17:1265–73.
17. Vella A, Carlson LA, Blier B, Felty C, Kuiper JD, Rooke TW. Cir-
culator boot therapy alters the natural history of ischemic limb
ulceration. Vasc Med 2000;5:21–5.
18. Burgess EM, Matsen FA III, Wyss CR, Simmons CW. Segmental
transcutaneous measurements of PO2 in patients requiring below-
the-knee amputation for peripheral vascular insufficiency. J Bone
Joint Surg Am 1982;64:378–82.
19. Crowther M, Brown NJ, Bishop ET, Lewis CE. Microenvironmental
influence on macrophage regulation of angiogenesis in wounds
and malignant tumors. J Leukoc Biol 2001;70:478–90.
20. Constant JS, Feng JJ, Zabel DD, Yuan H, Suh DY, Scheuenstuhl H,
Hunt TK, Hussain MZ. Lactate elicits vascular endothelial growth
factor from macrophages: A possible alternative to hypoxia.
Wound Rep Reg 2000;8:353–60.