ARTICLE IN PRESS

Biomaterials 25 (2004) 1559–1567

Porcine wound models for skin substitution and burn treatment

E. Middelkoopa,b,*, A.J. van den Bogaerdtb, E.N. Lammec, M.J. Hoekstrab,
K. Brandsmad, M.M.W. Ulrichb
a
Burns Center, Red Cross Hospital, Vondellaan 13, P.O. Box 1074, Beverwijk 1940 EB, The Netherlands
b
Research Department, Dutch Burns Foundation, Zeestraat 29, Beverwijk 1941 AJ, The Netherlands
c
Dermatology, UMC St. Radboud, Geert Grooteplein-Zuid 10, Nijmegen 6525 GA, The Netherlands
d
Institute for Animal Exp. (GDIA), University of Amsterdam, Academic Medical Center, Meibergdreef 9,
Amsterdam Zuidoost 1105 AZ, The Netherlands

Abstract

Skin regeneration is an important field of tissue engineering. Especially in larger burns and chronic wounds, present treatments
are insufficient in preventing scar formation and promoting healing. Initial screening of potentially interesting products for skin
substitution is usually done by in vitro tests. Before entering the clinic, however, in vivo studies in immunocompetent animals are
necessary to prove efficacy and provide information on safety aspects.
We have obtained extensive experience using the domestic pig as test animal for studies on skin replacement materials, including
tissue engineered skin substitutes, and burn wound treatment.
Two models are described: an excisional wound model for testing of dermal and epidermal substitutes and a burn wound model
for contact and scald burns, which allows testing of modern wound dressings in comparison to the present gold standards in burn
treatment. The results of these experiments show that in vivo testing was able to reveal (dis)advantages of the treatments which were
not detected during in vitro studies.
r 2003 Elsevier Ltd. All rights reserved.

Keywords: Animal model; Burn; Co-polymer; Dermis; Fibroblast; Scaffold; Honey

1. Introduction * There is hardly any limitation to collection of tissue

The healing of full thickness skin defects, such as in
chronic wounds (leg ulcers) or deep burns, comprises
complex processes leading to the formation of new
tissue. Wound contraction and scar formation are still
unavoidable components of the healing process. In
chronic wounds the misbalance between synthesis and
degradation of extracellular matrix is one of the
characteristics leading to non-healing [1,2]. Since so
many processes are involved which cannot easily be
controlled in a clinical situation, the use of animal
models in the design and testing of new therapeutic
approaches may be helpful. Some of the advantages of
the use of animal models are:
* Different experimental and control treatments can be
compared in the same animal.
*Corresponding author. Burns Center, Red Cross Hospital,
Vondellaan 13, P.O. Box 1074, Beverwijk 1940 EB, The Netherlands.
Tel.: +31-251-265283; fax: +31-20-6479813.
E-mail address: bwcemk@rkz.nl (E. Middelkoop).
samples apart from the limitation set by the size of
the animal.
* Less variation than in the clinical situation.
* Timing of treatment can be controlled.
The choice for a specific animal model depends largely
on the specific research questions to be answered, as well
as local circumstances.
Small mammals, such as mouse, rat, rabbit, guinea
pig, are suitable for studies that require large numbers of
animals or specific characteristics such as availability of
knockouts or transgenic animals (mainly available in
mice) or a compromised immune system (athymic mice).
It should be realised, however, that wound healing
characteristics in rodents may differ from the human
situation considerably, as wounds in these animals
mainly heal through wound contraction rather than
migration of epidermal cells [3].
For studies on gene function or the function of
specific enzymes or proteins in wound healing [4], the
use of genetically modified mice has been very valuable

0142-9612/$ - see front matter r 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/S0142-9612(03)00502-7
 ARTICLE IN PRESS
1560 E. Middelkoop et al. / Biomaterials 25 (2004) 1559–1567
(see [5] for a review). Similarly, identification of human
cells in the demonstration of cellular transfer, or even in
the use of gene therapy [6], was possible through the use
of athymic mice. In these studies the demonstration of
wound healing concepts was of greater relevance than
exact similarity between the animal model and the
human situation.
Another important aspect in the choice of an animal for
wound healing studies is whether the model should reflect
acute or chronic wound healing. Most animal wound
healing models described today reflect acute or delayed
acute rather than chronic (non-)healing. Nevertheless,
costs associated with chronic wound healing are nowadays
realised to have a major impact on society [7,8].
Several animal models have been developed to mimic
chronic wound healing. Wound healing problems
associated with diabetes can be studied in diabetic mice
[9–11], radiation-impaired wounds have been described
in rats [12] as well as pigs [13]. Chemicals have also been
used to develop a model of delayed wound healing in the
pig [14,15]. All of these models have their limitations, as
most of them reflect delayed healing rather than truly
impaired healing. Nevertheless, these models can be
used very efficiently during product development of new
wound healing drugs, e.g. in safety, dose-finding and
efficacy studies for wound debridement [16].
For studies on skin regeneration, which aim at
improvement of the quality of healing of the present
gold standards in wound healing therapies, similarity
between the skin of the test animal and human skin is
important and relevant. Examples of important out-
come parameters are: wound contraction, take of a
(autologous) graft, rate of epithelialisation, adverse
reactions directed against implanted material, dermal
scarring (which can be quantified by measuring histo-
logical parameters [17]). Since these studies are usually
undertaken in preparation of clinical studies, a high level
of agreement between results in animal studies and
human studies is desirable. Sullivan et al. [3] have
recently performed an extensive comparison of results in
wound healing studies in humans, pigs, small mammals
and in vitro studies. The concordance between the
results in humans versus pig was 78%, versus small
mammals 53% and versus in vitro studies 57%. These
data demonstrate that the pig can be a good test animal
when a high level of similarity with human skin is
indicated.
Specific advantages of the use of the domestic pig as
animal for wound healing studies are:
* There is a great similarity between human and
porcine skin, in dermal architecture and the absence
of a panniculus carnosus.
* Due to the large size of the animal several treatments
can be compared in the same animal, which reduces
variability.
However, there are also some disadvantages
related to the use of the domestic pig as test animal:
* Studies are expensive.
* Treatments with systemic effects cannot be compared
in the same animal. This drawback is more important
for large test animals such as the pig than for small
animals as the rat, since experiments on small animals
usually involve a relatively high number of animals.
* In immunohistochemistry cross-reactivity of mono-
clonal antibodies against human or rodent antigens is
often lacking, specific antibodies for porcine tissue
are required.
* In general, animal models lack clinical complexity.
Nevertheless, many aspects of the wound healing
process can be studied in animal models when keeping
the limitations of such a model in mind. This paper
describes the use of two different wound models in the
domestic pig, used for the development of skin
substitutes and new treatment modalities in burn wound
healing.
2. Skin substitutes: excisional wound model
To test the performance of scaffolds, to be used as
dermal replacement in the treatment of full thickness
burns, we used an excisional wound model. The
protocol was approved by the University of Amsterdam
Animal use Committee or the Animal use Committee of
the Burns Research Institute Beverwijk.
2.1. Materials
Biological as well as synthetic materials were tested as
dermal substitutes. Biological materials usually con-
sisted of collagen in different forms (crosslinked with
chemical or physical methods), combined with several
different extracellular matrix molecules [18].
Synthetic materials consisted of different variants
from Polyether Urethane and Polyglactin [19] and lately
several new variants of co-polymer (Polyactives, Isotis
NV, Bilthoven, the Netherlands) were tested. This
material consists of a poly(ethyleneglycol-terephthalate)
(55%)/poly(butylene-terephthalate) 45%(PEGT/PBT)
copolymer with a weight ratio of 55/45, and PEG
having a MW of 300 Da [20–22].
2.2. Operation procedure
Up to 14 full thickness surgical wounds (3
3 cm2)
were created on the back of female Yorkshire pigs,
weighing approximately 20 kg at arrival. The wounds
were created using an electrical dermatome to a depth of
2.7 mm. The area surrounding the wounds was marked
by tattoos, to facilitate quantification of the remaining
 ARTICLE IN PRESS
E. Middelkoop et al. / Biomaterials 25 (2004) 1559–1567
Fig. 1. Bandage technique for pigs with full thickness wounds.
Photograph taken immediately post-operative.
wound area during the healing process. The dermal
substitute to be tested was placed in the wound bed and
immediately covered with a split skin mesh graft
(extension 1:3, 0.2 mm thickness). The wounds were
covered with an appropriate wound dressing, such as
Allevyn or Melolin (Smith & Nephew, Hull, UK) which
was kept in place by covering it with adhesive bandage
(Curafixs Lohmann & Rauscher, Neuwied, Germany)
and elastic stockings (Tubigrip) (see Fig. 1). All
procedures were carried out under proper anaesthesia
and pain relief.
2.3. Sedation/analgesia/anaesthesia
About 20–30 min before start of the procedure the
animals were sedated with azaperon (Stressnils, Jans-
sen, Gent, Belgium), 4 mg/kg, by means of intra-
muscular injection. For the transplantation procedure,
this was combined with administration of ketamin
(Nimateks, Eurovet Animal Health BV, Bladel, the
Netherlands), 10 mg/kg, by means of intra-muscular
injection.
For tattoo and biopsy procedures, anaesthesia was
performed using N2O+O2+Isoflurane via the mouth
cap, with guarding by pulse oxymeter. For the
transplantation procedures, artificial respiration with
N2O (60%)+O2 (40%) and Isoflurane was applied, with
guarding by cardio-cap.
Analgesia was obtained by administration of Fina-
dynes (Schering-Plough BV, Maarssen, the Nether-
lands) (2 ml/50 kg) by means of intra-muscular injection,
immediately upon completion of the procedures. For the
transplantation procedure, this was combined with
Sufentas (Janssen-Cilag BV, Tilburg, the Netherlands)
during the total time of anaesthesia and Temgesics
(Schering-Plough BV) after completion of the proce-
1561
dure. The day after the transplantation procedure
Finadynes was administered again (2 ml/50 kg). Admin-
istration of Finadynes the day after biopsy or tattoo
procedure was performed if necessary.
2.4. Study design
In previous work we evaluated the role of seeded
fibroblasts in biological collagen-based dermal substi-
tutes [23,24]. Here, we describe recent work with cell-
seeded and cultured synthetic matrices, based on co-
polymer [25].
Co-polymer matrices were seeded statically with
autologous porcine fibroblasts at high (0.75
106 fibro-
fibroblasts/cm2) and low (0.10
106 fibroblasts/cm2)
densities (10 wounds for each treatment) and compared
to non-seeded co-polymer (eight wounds).
In a second series of experiments, co-polymer
matrices were seeded statically with autologous fibro-
blasts (0.50
106 cells/cm2) (five wounds) and cultured
for 21 days in order to induce extracellular matrix
formation prior to implantation. For comparison, also
matrices were seeded with 0.75
106 cells/cm2 (similar to
the high cell density of the first series) (five wounds) the
day before implantation. Acellular co-polymer matrices
and split skin graft alone served as control treatments
(four wounds each). In total, three animals were used for
these studies.
2.5. Evaluation
Wound healing was followed in time during 4–6
weeks. During this time-period, the wounds were
macroscopically recorded by photography. Also con-
traction was recorded by measuring the wound area
during healing. Correction for growth of the animals
was accomplished by determining the change in a larger
area surrounding the wound, which was marked by a
tattooed grid.
Biopsies were taken at different time intervals to study
wound healing parameters as epithelialisation, extra-
cellular matrix formation, collagen remodelling, cell
density, presence of myofibroblasts, vascularity and
biomaterial properties such as biodegradation and
foreign body reactions (see also [26] for a detailed
(immuno)histochemical evaluation of skin substitutes).
2.6. Results: effect of seeding fibroblasts and pre-
culturing on wound healing
Macroscopic healing is illustrated for several wounds
in Fig. 2.
In all wounds treated with the synthetic co-polymer,
healing was not optimal. After 1 week outgrowth of the
split skin graft was retarded and the graft itself showed
no signs of revascularisation. After 2 weeks, the graft
 ARTICLE IN PRESS
1562 E. Middelkoop et al. / Biomaterials 25 (2004) 1559–1567

(A) (B)

(C) (D)

Fig. 2. (A) Co-polymer seeded with 0.75
106 cells/cm2 fibroblasts, 4 weeks post-operative. (B) Co-polymer without fibroblasts, 4 weeks post-
operative. (C) Control treatment (only split skin graft, meshed 1:3), 4 weeks post-operative. (D) Wound treated with collagen/elastin dermal
substitute seeded with fibroblasts (0.5
106 cells/cm2), 6 weeks post-operative.

was degenerated and the wounds were not closed.

Moreover, the wounds showed signs of hypergranulation
with a highly red appearance. After 3 weeks, the wounds
were barely epithelialised. No macroscopic differences
were observed between the different seeding densities
and the non-seeded co-polymer. After 4–6 weeks, the
wounds were closed but contracted (see also Fig. 3) and
showed formation of stiff scar tissue. Usage of co-
polymer resulted in a high contraction for all wounds. A
trend was observed indicating reduced contraction when
more cells were seeded in the co-polymer.
In Fig. 4 some representative pictures of histological
slides are depicted. Fig. 3. Wound contraction in time at different seeding densities.
Some differences were noted in reaction to the
fibroblast-seeded and non-seeded co-polymer matrices. structures (Fig. 4C and D). In the end, co-polymer
If the matrices were not seeded, the co-polymer seemed fragments in the upper-part of the dermis were extruded
to be present mainly in the lower granulation tissue and from the wounds. Seeding in low densities resulted in an
was fragmented (Fig. 4A and B). Seeding high densities intermediate situation where parts of the co-polymer
of autologous fibroblasts in co-polymer resulted in were extruded and other parts fragmented in the lower
localisation of the matrix in the higher granulation granulation tissue. Furthermore, co-polymer fragments
tissue and encapsulation of the co-polymer by epidermal were surrounded by an inflammatory region in which
 ARTICLE IN PRESS
E. Middelkoop et al. / Biomaterials 25 (2004) 1559–1567 1563

(A)
(A) (B)

(C) (D)

Fig. 4. (A) Wounds treated with co-polymer without cells: integration into the wound tissue at week 1. (B) Wounds treated with co-polymer without
cells: fragmentation in the deeper granulation tissue at week 6. (C) Wounds treated with co-polymer with fibroblasts: encapsulation in epidermal
structures at week 1. (D) Wounds treated with co-polymer with fibroblasts: encapsulation in epidermal structures at week 6.

giant cells were present (Fig. 5). No obvious differences

between pre-cultured and non-pre-cultured seeded ma-
trices could be noted with respect to co-polymer
fragmentation, degradation or wound healing parameters.

3. Burns: models for contact burns and scalds

A burn wound is a dynamic wound that can deepen in

time, thereby increasing the total tissue damage and the
risk of complications such as hypertrophic scarring. A
well-known cause of wound deepening is infection. This
may lead to loss of vital skin remnants and may change Fig. 5. Co-polymer fragments surrounded by giant cells during
a healing wound into a non-healing one. inflammation process (non-seeded co-polymer at week 4).
Less known is the fact that different types of burns
may lead to different levels of vascular damage. The
vascular damage greatly influences the ultimate level of * a periferal zone with increased capillary bleeding that
tissue damage. suffers superficial burn damage: the hyperaemic zone;
Three zones can be discriminated: * an area with retarded capillary filling: stasis zone;
 ARTICLE IN PRESS
1564 E. Middelkoop et al. / Biomaterials 25 (2004) 1559–1567

* an area with no capillary refill: the zone of coagulated

tissue.

Depending on the burn treatment the stasis zone may

turn into a coagulated zone by dessication.
Two different burn models have been developed in
pigs in order to investigate the similarities and differ-
ences between the levels of tissue damage caused by
contact burns or scald burns [27,28]. Scalds were made
by pouring 80 C hot water into a PVC cylinder that was
positioned on the back of the animal for 10–40 s, under (A)
proper inhalation anaesthesia and pain relief of the
animal as described above. In these burns the following
pattern of tissue damage was noted.
Throughout the whole dermal area, a mixture of vital
and damaged extracellular matrix was seen (Fig. 6A).
Depending on the time of contact between the skin and
the hot water, superficial to deep scalds were made. In
superficial scalds, no damage to the vascular structures
was detected. In deep scald burns, both the superficial
and the deep vascular structures were damaged. How-
ever, in scalds of intermediate depth, the superficial
vasculature could still be intact, whereas the deep
dermal plexus was damaged [28]. This deep dermal
vascular damage did not seem to hamper spontaneous
reepithelialisation, but caused an extensive inflamma-
tory response resulting in scar formation.
Contact burns were initiated by placing a heated
(170 C) brass block on the animal during 10 s to create a
superficial contact burn or during 20 s to inflict a deep
partial thickness burn. Three levels of tissue damage
could be discriminated: a superficial layer which was
fully coagulated, a small intermediate layer with a
mixture of damaged and still vital tissue and a virtually
intact deep dermal tissue (Fig. 6B). In this type of burn, (B)
the deep vascular structures remained intact when the
Fig. 6. (A) Scald at post-burn day 3. Intermingled pattern of intact
deep dermal tissue was unaffected. (blue) and damaged (red) collagen (Masson trichrome stain). (B)
This deep dermal burn model was used to study the Contact burn at post burn day 3. Clear demarcation between necrotic
effects of Silver Sulphadiazine cream 1% (SSD) versus (red) and viable (blue) collagen (Masson trichrome stain).
honey (source: lime tree) on macroscopic and microscopic
wound healing parameters such as crust formation, cream was used (Solvay, Weesp, the Netherlands) (six
inflammatory response, epithelialisation and granulation wounds) or honey from the lime tree (24 wounds).
tissue formation. Since ancient times, honey is said to be Excision biopsies were taken weekly during the total
favourable for wound healing. Therefore, it was con- follow-up of 6 weeks.
sidered relevant to perform a direct comparison with a
standard burn wound treatment such as SSD cream. 3.2. Results

3.1. Study design SSD cream treatment resulted in a slower re-

Twelve deep dermal burns (see also Fig. 7) of 50 cm2
each were inflicted as described above on the back of
Yorkshire pigs (n ¼ 3) by application of a heated brass
block (170 C) during 20 s. Wounds were treated daily
with fresh cream and new dressings until they had
healed (approximately 3 weeks for honey treatment, 4–5
weeks for SSD). Either the commercially available SSD
epithelialisation than the honey treatment: complete
re-epithelialisation took 4–5 weeks for SSD-treated
wounds versus 2–3 weeks for honey-treated wounds
(Fig. 8). Contraction of the wounds was evident in both
treatment groups, especially at later time points (5–6
weeks). No differences in wound contraction were
established between the groups. A slightly higher
bacterial load was seen in the superficial necrotic layers
 ARTICLE IN PRESS
E. Middelkoop et al. / Biomaterials 25 (2004) 1559–1567 1565

in the honey-treated wounds. SSD treatment showed
more inflammatory response, with thicker granulation
tissue and marked presence of myofibroblasts compared
to the honey treatment.
4. Discussion
4.1. Excisional wound model
The excisional wound model has proven to be very
valuable for the screening and in vivo testing of
Fig. 7. Silversulphadiazine (SSD) cream treatment, PBD 7, PAP
staining. Damaged vessels in deeper dermal layer.
prototypes for skin substitutes. Many positive results
were obtained with biological matrices based on
collagen and elastin [18,24,26]. Based on these results,
clinical evaluation of this material for treatment of
burns and reconstructive wounds has started [29,30].
Discriminative power of this wound healing model is
also indicated by the data presented here on synthetic
scaffolds. The tested version of the co-polymer scaffold
has shown problematic reactions in skin wound healing,
with the scaffold being removed by encapsulation and
subsequent extrusion, whereas much better data were
found in previous studies using the collagenous scaffolds
[18,24].
In the study presented here, only minor differences
occurred in the healing of wounds treated with co-
polymer alone or with seeded or seeded and pre-cultured
fibroblasts. In general, fragmentation of co-polymer
seemed to occur predominantly, but not exclusively, in
non-seeded matrices or scaffolds seeded with a low
density of cells. In the wounds treated with co-polymer
in which the higher cell density was seeded, or with those
with pre-cultured matrices, extrusion seemed to be the
most frequently observed way of removal of co-polymer.
This process was combined with fragmentation of the
material.
In conclusion, we can state that this synthetic
template, when overgrafted with split skin, was not able

(A) (B)

(C) (D)

Fig. 8. (A) Silversulphadiazine (SSD) cream treatment, post-burn day (PBD) 4: intact, soft and supple yellow-brown discoloured eschar with red rim
around the lesion. (B) SSD treatment, PBD 28: minimal crust remnants in a predominantly epithelialised wound with minimal signs of contraction.
(C) Honey treatment, PBD 4: stiff intact brown discoloured crust. (D) Honey treatment, PBD 28: near complete epithelialisation with cobble stone
aspect of the scar and contraction.
 ARTICLE IN PRESS
1566 E. Middelkoop et al. / Biomaterials 25 (2004) 1559–1567
to reduce wound contraction or scarring and was in fact
actively removed from the wound bed at an early stage
in the healing process. This process was accompanied by
excessive inflammation and granulation tissue forma-
tion.
4.2. Burns: models for contact burns and scalds
The interest in honey as an adequate treatment of
burns, donor site wounds and even skin ulcers has
increased in recent times [31–34]. Most prominent
beneficial effects noted in this experimental study in
contact burns were: a faster re-epithelialisation, less
inflammation and granulation tissue compared to the
SSD-treated wounds. In the literature, honey is also
described to have specific antibacterial properties [34].
Based on the evidence available in the literature today,
honey could contribute to improved wound care,
especially in the less developed world where ‘modern’
wound dressings are not readily available or too
expensive. A prerequisite for the marketing of a
therapeutic honey would be to characterise the product
in terms of sources and activities. The use of well-
characterised wound models is indispensable for such
studies.
5. Conclusion
The coming era promises great improvements in the
development of tissue-engineered skin. In vitro models
should remain the first choice for testing prototypes,
such as screening of suitable scaffolds for dermal as well
as epidermal cells. Parameters as cytotoxicity, stability,
ease of handling, mechanical properties can all be easily
tested in in vitro models.
When it comes to testing of efficacy, fine-tuning of
required cell densities and safety aspects relating to
immunoreactivity and biocompatibility, in vivo models
are warranted. This paper describes some of the models,
designed in the domestic pig, which can be used for
in vivo testing of prototypes of tissue engineered skin.
The models allow discrimination between scaffolds
that are biodegradable and those that provoke foreign
body reactions; they can be used to determine optimal
cell densities for dermal substitutes and for testing new
wound dressings.
Nevertheless, it should be realised that pigs are not
humans. However well-designed a model may be,
dissimilarities between humans and animals remain.
Therefore, a critical selection of the most appropriate
animal model should be performed before the onset of a
study. A careful examination of the results should
precede the interpretation and extrapolation towards the
human situation.
Acknowledgements
Part of this work was financially supported by grants
from the Dutch Burns Foundation and Isotis NV, the
Netherlands.
References
[1] Parks WC. Matrix metalloproteinases in repair. Wound Rep Reg
1999;7:423–32.
[2] Wall SJ, Bevan D, Thomas DW, Harding KG, Edwards DR,
Murphy G. Differential expression of matrix metalloproteinases
during impaired wound healing of the diabetic mouse. J Invest
Dermatol 2002;119:91–8.
[3] Sullivan TP, Eaglstein WH, Davis SC, Mertz P. The pig as model
for human wound healing. Wound Rep Reg 2001;9:66–76.
[4] Drew AF, Liu H, Davidson JM, Daugherty CC, Degen JL.
Wound-healing defects in mice lacking fibrinogen. Blood
2001;97:3691–8.
[5] Scheid A, Meuli M, Gassmann M, Wenger RH. Genetically
modified mouse models in studies on cutaneous wound healing.
Exp Physiol 2000;85:687–704.
[6] Campbell C, Hultman S, Cairns B, DeSerres S, Meyer A. Green
fluorescent protein-adenoviral constructs as a model for transient
gene therapy for human cultured keratinocytes in an athymic
mouse model. J Trauma 2003;54:72–9.
[7] Pompeo MQ. The role of ‘‘wound burden’’ in determining the
costs associated with wound care. Ostomy Wound Manage
2001;47:65–71.
[8] Harding K, Cutting K, Price P. The cost-effectiveness of wound
management protocols of care. Br J Nurs 2000;9:S6–24.
[9] Tsuboi R, Shi CM, Rifkin DB, Ogawa H. A wound healing model
using healing-impaired diabetic mice. J Dermatol 1992;19:673–5.
[10] Danno K, Mori N, Toda K, Kobayashi T, Utani A. Near-
infrared irradiation stimulates cutaneous wound repair: labora-
tory experiments on possible mechanisms. Photodermatol Photo-
immunol Photomed 2001;17:261–5.
[11] Kampfer H, Pfeilschifter J, Frank S. Expressional regulation of
angiopoietin-1 and -2 and the tie-1 and -2 receptor tyrosine
kinases during cutaneous wound healing: a comparative study of
normal and impaired repair. Lab Invest 2001;81:361–73.
[12] Gu Q, Wang D, Gao Y, Zhou J, Peng R, Cui Y, Xia G, Qing Q,
Yang H, Liu J, Zhao M. Expression of MMP1 in surgical and
radiation-impaired wound healing and its effects on the healing
process. J Environ Pathol Toxicol Oncol 2002;21:71–8.
[13] Bernatchez SF, Parks PJ, Grussing DM, Matalas SL, Nelson GS.
Histological characterization of a delayed wound healing model in
the pig. Wound Rep Reg 1998;6:223–33.
[14] Graham JS, Schomacker KT, Glatter RD, Briscoe CM, Braue
EH, Squibb KS. Efficacy of laser debridement with autologous
split-thickness skin grafting in promoting improved healing of
deep cutaneous sulphur mustard burns. Burns 2002;28:719–30.
[15] Mekkes JR, Le Poole IC, Das PK, Bos JD, Westerhof W.
Efficient debridement of necrotic wounds using proteolytic
enzymes derived from Antarctc krill: a double-blind, placebo-
controlled study in a standardized animal wound model. Wound
Rep Reg 1998;6:50–7.
[16] Mekkes JR, Zeegelaar JE, Westerhof W. Quantitative and
objective evaluation of wound debriding properties of collagenase
and fibriolysin/desoxyribonuclease in a necrotic ulcer animal
model. Arch Dermatol Res 1998;290:152–7.
[17] Van Zuijlen PPM, de Vries HJC, Lamme EN, Coppens JE, van
Marle J, Kreis RW, Middelkoop E. Reliability of collagen
 ARTICLE IN PRESS
E. Middelkoop et al. / Biomaterials 25 (2004) 1559–1567
orientation assessment by observers and Fourier analysis.
J Pathol 2002;198:284–91.
[18] De Vries HJC, Middelkoop E, Mekkes JR, Dutrieux RP,
Wildevuur CHR, Westerhof W. Dermal regeneration in a native
non cross-linked collagen sponge with different extracellular
matrix molecules. Wound Rep Reg 1994;2:37–47.
[19] De Vries HJC, Mekkes JR, Middelkoop E, Hinrichs WLJ,
Wildevuur CHR, Westerhof W. Dermal substitutes for full
thickness wounds in a one stage grafting model. Wound Rep
Reg 1993;1:244–52.
[20] Beumer GJ, van Blitterswijk CA, Ponec M. Biocompatibility of a
biodegradable matrix used as a skin substitute: an in vivo
evaluation. J Biomed Mater Res 1994;28:545–52.
[21] Van Dorp AG, Verhoeven MC, Koerten HK, Van Der Nat-Van
Der Meij TH, Van Blitterswijk CA, Ponec M. Dermal regenera-
tion in full-thickness wounds in Yucatan miniature pigs using a
biodegradable copolymer. Wound Rep Reg 1998;6:556–68.
[22] van Dorp AG, Verhoeven MC, Koerten HK, van Blitterswijk CA,
Ponec M. Bilayered biodegradable poly(ethylene glycol)/
poly(butylene terephthalate) copolymer (Polyactive) as substrate
for human fibroblasts and keratinocytes. J Biomed Mater Res
1999;5(47):292–300.
[23] Lamme EN, van Leeuwen RTJ, Jonker A, van Marle J,
Middelkoop E. Living skin substitutes: survival and function of
fibroblasts seeded in a dermal substitute in experimental wounds.
J Invest Dermatol 1998;111:989–95.
[24] Lamme EN, Van Leeuwen RTJ, Brandsma K, Van Marle J,
Middelkoop E. Higher numbers of autologous fibroblasts in an
artificial dermal substitute improve tissue regeneration and
modulate scar tissue formation. J Pathol 2000;190:595–603.
[25] Van den Bogaerdt AJ, van Galen M, Lamme EN, Middelkoop E.
In vitro and in vivo evaluation of a synthetic copolymer. Wound
Rep Reg 2001;9:419.
1567
[26] Lamme EN, de Vries HJC, van Veen H, Gabbiani G, Westerhof
W, Middelkoop E. Extracellular matrix characterization during
healing of full thickness wounds treated with a collagen/elastin
dermal substitute shows improved skin regeneration in pigs.
J Histochem Cytochem 1996;44:1311–22.
[27] Hoekstra MJ, Hupkens P, Dutrieux RP, Bosch MM, Brans TA,
Kreis RW. A comparative burn wound model in the New
Yorkshire pig for the histopathological evaluation of local
therapeutic regimens: silver sulfadiazine cream as a standard. Br
J Plast Surg 1993;46:585–9.
[28] Brans TA, Dutrieux RP, Hoekstra MJ, Kreis RW, du Pont JS.
Histopathological evaluation of scalds and contact burns in the
pig model. Burns 1994;20(Suppl 1):S48.
[29] Van Zuijlen PPM, Van Trier AJM, Vloemans AFPM, Groenevelt
F, Kreis RW, Middelkoop E. Graft survival and effectiveness of
dermal substitution in burns and reconstructive surgery in a one-
stage grafting model. Plast Reconstr Surg 2000;106:615–23.
[30] Van Zuijlen PPM, Vloemans AFPM, van Trier AJM, Suijker
MH, van Unen E, Groenevelt F, Kreis RW, Middelkoop E.
Dermal substitution in acute burns and reconstructive surgery: a
subjective and objective long term follow-up. Plast Reconstr Surg
2001;108:1938–46.
[31] Molan PC. Potential of honey in the treatment of wounds and
burns. Am J Clin Dermatol 2001;2:13–9.
[32] Misirlioglu A, Eroglu S, Karacaoglan N, Akan M, Akoz T,
Yildirim S. Use of honey as an adjunct in the healing of split-
thickness skin graft donor site. Dermatol Surg 2003;29:168–72.
[33] Lusby PE, Coombes A, Wilkinson JM. Honey: a potent agent for
wound healing? J Wound Ostomy Continence Nurs 2002;29:
295–300.
[34] Moore OA, Smith LA, Campbell F, Seers K, McQuay HJ, Moore
RA. Systematic review of the use of honey as a wound dressing.
BMC Complement Altern Med 2001;1:2.