Interferon Alpha-2a Reduces Early Erythema AfterFull-Thickness Skin Graft in the Pig

JUN SIK KIM, MD, PHD,� DAEGU SON, MD, PHD,y TAE HYUN CHOI, MD, PHD,z

KIHWAN HAN, MD, PHD,y JUN HYUNG KIM, MD, PHD,y HYUN MI CHO, MSC,y WON HEE KIM, MSC,y

SANG-HYON KIM, MD,y NAM GYUN KIM, MD,� KYUNG SUK LEE, MD,� O. HYUN HWANG, MD,z

GU SEOB ROH, MD, PHD,J AND JUNGBIN PARK��

BACKGROUND Skin grafting is a commonly performed procedure, but studies of changes in the levelsof cytokines after skin grafting have not been reported.

OBJECTIVE We examined changes in cytokines and the degree of erythema after skin grafting in pigs inthe control group. Interferon alpha (IFN-a) was injected to reduce erythema, and subsequent changes incytokines and the degree of erythema were examined in the experimental group.

METHODS Vascular endothelial growth factor (VEGF), thrombospondin-1 (TSP1), and CD31 were ex-amined using Western blot analysis and immunohistochemistry. The degree of erythema was measuredat 2, 4, and 8 weeks using a chromometer.

RESULTS In the control group, VEGF increased at 2 weeks and decreased at 4 and 8 weeks. TSP1increased over time. CD31 increased to 4 weeks and decreased at 8 weeks. In the experimental group,VEGF was lower at 2 weeks and higher at 8 weeks than in the control group, TSP1 was higher at 2 weeksand lower at 8 weeks, and CD31 was lower at 4 and 8 weeks. Erythema in the experimental group waslower than that in the control group at 2 and 8 weeks.

CONCLUSION IFN-a may be one of the agents that reduces erythema by suppressing excessive re-vascularization.

The authors have indicated no significant interest with commercial supporters.

Skin grafting is one of the procedures that is per-

formed most commonly in the field of surgery.

Survival mechanisms of skin graft are known to be

serum imbibition and revascularization.1

Vascular endothelial growth factor (VEGF) has been

discovered to be a mitogen for vascular endothelial

cells, and its importance as a potent regulator of

angiogenesis has been shown.2 In the last decade,

improvements in wound healing and the viability of

skin grafts and skin flaps have been demonstrated

after application of VEGF.3,4 Thrombospondin-1

(TSP1) was the first endogenous inhibitor of angio-

genesis to be identified. TSP1 inhibits angiogenesis

by modulating the proliferation and migration of

endothelial and vascular smooth muscle cells. Isen-

berg and colleagues found that blocking TSP1

binding or suppressing CD47 expression dramati-

cally increased skin graft survival.5 Nonetheless, to

the authors’ knowledge, studies of alterations in

VEGF or TSP1 during the revascularization of skin

grafts have not been reported.

& 2009 by the American Society for Dermatologic Surgery, Inc. � Published by Wiley Periodicals, Inc. �ISSN: 1076-0512 � Dermatol Surg 2009;35:1514–1524 � DOI: 10.1111/j.1524-4725.2009.01267.x

1 5 1 4

Jun Sik Kim and O. Hyun Hwang contributed equally to this paper as first authors.

�Department of Plastic and Reconstructive Surgery and JDepartment of Anatomy, Institute of Health Sciences, Collegeof Medicine and Hospital, Gyeongsang National University, Jinju, Republic of Korea; yDepartment of Plastic andReconstructive Surgery and yDepartment of Internal Medicine, Keimyung University Dongsan Medical Center, Daegu,Republic of Korea; zDepartment of Reconstructive Plastic Surgery, Seoul National University Hospital, Seoul, Republicof Korea; zDepartment of Plastic and Reconstructive Surgery, Changwon Fatima Hospital, Changwon, Republic ofKorea; ��Daegu Science High School, Daegu, Republic of Korea.

Erythema develops in grafted skin during the early

period after surgery and is attributable to revas-

cularization. Because of the erythema, the grafted

skin appears distinctly different from the surround-

ing skin, and patients are often dissatisfied with the

aesthetic results.6

Interferon alpha (IFN-a) is a cytokine with pleio-

tropic cellular functions that include antiviral,

antiproliferative, immunomodulatory, and anti-

angiogenic activities. Clinical studies have demon-

strated that IFN-a treatment can induce impressive

responses in angioproliferative diseases, such as

Kaposi’s sarcoma and hemangiomas.7 Our hypoth-

esis was that the application of IFN-a, which has

antiangiogenic activities, would suppress the ery-

thema that develops after skin grafting and thus

improve patient satisfaction.

To examine the changes in VEGF, TSP1, and CD31

occurring during skin graft surgery in pigs, we per-

formed Western blot and immunohistochemistry

analyses. In an attempt to reduce the erythema de-

veloping in the early period after grafting, recombi-

nant IFN-a-2a was injected. After injection, changes

in the cytokines were examined, and to assess the

level of reduction in erythema, skin color was mea-

sured objectively using a chromometer.

Materials and Methods

Materials and Full-Thickness Skin

Graft Procedure

The experimental animals were Yorkshire pigs, aged

2 to 3 months old, weighing 20 to 30 kg, and with-

out dermatological diseases. They were fed with

standard pig feed and allowed to consume water

freely. The environment was maintained at one

atmosphere pressure, 201C to 231C, and 65%

humidity, with a 12-hour light–dark cycle.

Sedation was performed using an intramuscular injec-

tion of 2.2 mg/kg azaperone (Stresnil, Janssen, Beerse,

Belgium), and 37 mg/kg propofol (Diprivan-PFS,

AstraZeneca Korea, Seoul, Korea) was injected into

the external ear vein to induce anesthesia. Two iden-

tical experiments per pig were performed on the right

and left side. The hairs on the dorsum (recipient site)

were removed, and a 5� 5 cm wound was created

using a dermatome (Acculan, Aesculap Inc., Center

Valley, PA) with the depth adjusted to 0.8 mm. From

the inguinal area (donor site), full-thickness skin

5� 5 cm in size was harvested, and the defect area was

primarily closed with 2-0 nylon suture (Blue Nylon,

Ailee, Busan, Korea). The harvested skin was grafted

to the recipient site and immobilized using 3-0 chromic

catgut (Catgut chrom, B. Braun, Penang, Malaysia)

and a skin stapler (Reflex One and Reflex TL, Con-

med, Utica, NY). After surgery, a bolster dressing was

applied in the skin graft area to prevent injury. To

prevent infection, antibiotics were injected intramus-

cularly for 3 days. The Keimyung University School of

Medicine Institutional Animal Care and Use Commit-

tee approved the study.

Control Group: Four pigs received skin grafts using

the method described above.

Experimental Group: Three pigs received skin grafts

using the method described above. From 1 day after

surgery, recombinant human IFN-a-2a (Intermax

alpha, LG Life Sciences, Seoul, Korea) at a concen-

tration of 2,000,000 units/m2 was injected intra-

muscularly three times per week for 3 weeks.

Negative Control (Normal Skin): The negative con-

trol was a 5-� 5-cm area of skin harvested from

both sides of the back of the pigs that had not been

grafted.

Western Blot Analysis

Procedure: At 2, 4, and 8 weeks after surgery, an area

of the dorsal skin graft approximately 0.5� 1 cm in

size was harvested from control and experimental

pigs, and the following experiments were performed.

A sample of normal skin (negative control) was also

harvested, and identical experiments were performed.

Briefly, the tissues were transferred into five volumes

of RIPA buffer and homogenized in an ice bath with

3 5 : 1 0 : O C T O B E R 2 0 0 9 1 5 1 5

K I M E T A L

a tissue homogenizer. Five bursts of 15 seconds at

maximum speed with 45-second intervals of cooling

between each burst were applied. The homogenate

was then centrifuged at 2,000 g for 15 minutes at 41C.

The supernatant was recovered, and an aliquot was

used to evaluate the amount of protein. Fifty micro-

grams of protein were separated using 6% and 10%

sodium dodecyl sulfate polyacrylamide gel electro-

phoresis and then electrophoretically transferred to a

polyvinylidene fluoride membrane (Bio-Rad, Hercules,

CA) for immunoblot analysis according to standard

protocols. For protein detection, the membranes were

incubated with the primary antibodies (Abcam, Cam-

bridge, UK). Peroxidase-conjugated anti-mouse

immunoglobulins (KPL, Guelph, Ontario, Canada)

were used as secondary antibodies. The signals were

detected using the enhanced chemiluminescence West-

ern blot analysis system (Amersham Pharmacia,

Piscataway, NJ).

Quantitative data were determined as the mean ratio

of the optical density of the specific bands normal-

ized to that of b-actin. The densitometric analysis

was carried out using a Scion Image Data Analyzer

(SCION Corporation, Frederick, MD).

Analysis of Data: In each group, the change in ex-

pression of proteins with time was analyzed, and at

each (specific) time point, the difference in the ex-

pression of proteins between the control and exper-

imental groups was analyzed. In analysis of the

Western blot results, it was assumed that, as the

measurements were repeated, the measured values

followed a normal distribution, and the parametric

one-way analysis of variance (ANOVA) test was used.

Immunohistochemistry

The harvesting methods were as described for the

Western blot analysis. The harvested samples were

embedded in Optimal Cutting Temperature Com-

pound (Sakura, Tokyo, Japan) for frozen tissue

specimens and sectioned using a Cryocut Microtome

(Shandon, England) at 8mm thickness.

Cryosections were washed three times with phosphate

buffered saline (PBS) and fixed in methanol for 10

minutes at �201C. After being washed with PBS, sec-

tions were treated with 3% hydrogen peroxide for 20

minutes at room temperature and with antibody

diluent (Golden Bridge International Inc., Mukilteo,

WA) to block nonspecific reactions. The following

antibodies were used: VEGF (dilution 1/80; Abcam),

TSP1 (dilution 1/25; Abcam), and CD31 (dilution 1/

100; Chemicon, CA). After 1.5 hours incubation with

primary antibodies in a humidity chamber, a

DakoCytomation EnVision 1 System-horseradish

peroxidase–labelled polymer (DakoCytomation,

Glostrup, Denmark) was used for 30 minutes in

conjunction with all of the primary antibodies. A

DakoCytomation Liquid DAB Substrate Chromogen

System was used to label the antibodies. The slides

were counterstained with Mayer’s hematoxylin for

1 minute.

Chromometer Measurement and Analysis

of Results

Using a chromometer (Chromometer CR-300,

Minolta Co., Osaka, Japan), the redness of the re-

cipient site or skin graft was measured before surgery

and 2, 4, and 8 weeks after surgery in the control and

experimental groups. The area from which the skin

graft was taken was measured three times, and the

final value was defined as the average of the values

obtained.

In each group, change in redness of the skin graft

with time was analyzed, and at each (specific) time

point, the difference in redness between the control

and experimental groups was analyzed.

The Rate of Viability

The viability rate of the skin grafts was assessed 2

weeks after surgery. The viability rate was calculated as

the area of the skin graft that had taken as a percentage

of the original size (25 cm2). The viability rates of

the control and experimental groups were compared.

D E R M AT O L O G I C S U R G E RY1 5 1 6

I N T E R F E R O N -a - 2 A R E D U C E S E A R LY E RY T H E M A A F T E R S K I N G R A F T

Results

Changes in Levels of Cytokines According to

Western Blot Analysis

Changes in VEGF, TSP1, and CD31 in the Control

Group: The level of VEGF in the normal skin was

0.43; 2 weeks after surgery, it increased to 1.02, and

then 4 and 8 weeks after surgery decreased to 0.55

and 0.56, respectively. The values were significantly

different with time according to the one-way

ANOVA (po.001). Using the Tukey test, the value

was significantly higher than that of the normal skin

at 2 weeks (po.001), but at 4 and 8 weeks it was not

different from that of the normal skin (p = .08, .06).

The level of TSP1 in the normal skin was 0.34. At 2,

4, and 8 weeks after surgery, it increased to 0.49,

0.50, and 1.10, respectively. Using ANOVA, the

values were significantly different with times (po.001). Using the Tukey test, at 2, 4, and 8 weeks,

the levels were significantly higher than those in

normal skin (p = .008, .006, o.001).

The level of CD31 in the normal skin was 0.21. It

increased to 0.35 at 2 weeks and to 0.95 at 4 weeks

after surgery and then decreased slightly to 0.63 at 8

weeks. The values varied significantly with time

(ANOVA; po.001) and were significantly different

from those in normal skin at 2, 8, and 4 weeks

(Tukey; p = .01, o.001, o.001). In summary, the

level of CD31 increased continuously up to 4 weeks

and then decreased slightly at 8 weeks (Figure 1).

Changes in VEGF, TSP1, and CD31 in the Exper-

imental Group: The level of VEGF in the normal

skin was 0.43, and values at 2, 4, and 8 weeks after

surgery were 0.90, 0.53, and 0.68, respectively. Us-

ing one-way ANOVA, the values were significantly

different with time (po.001). The Tukey test showed

a significantly higher value at 2 weeks (po.001), and

at 8 weeks, the level was higher than in normal skin

(po.001). The level of VEGF therefore increased

maximally at 2 weeks, decreased at 4 weeks to the

normal skin level, and then increased again at 8

weeks.

The level of TSP1 in the normal skin was 0.34, and

the values at 2, 4, and 8 weeks after surgery were

1.07, 0.61, and 0.41, respectively. ANOVA showed

that the values differed significantly with time (po.001). At 2 and 4 weeks, the level was significantly

higher than in normal skin (Tukey; po.001,

p = .007). The level increased maximally at 2 weeks;

at 4 weeks, it decreased slightly but remained

higher than in normal skin, and by 8 weeks it had

decreased to the normal skin level.

The level of CD31 in the normal skin was 0.21, and

the values at 2, 4, and 8 weeks after surgery were

0.32, 0.86, and 0.37, respectively. According to

ANOVA, the values varied significantly with time (p

o.001). The Tukey test showed that, at 4 weeks, the

level of CD31 was significantly higher than in nor-

mal skin (po.001), but at 8 weeks it had decreased

to the normal skin level (Figure 1).

Comparison of the Control and Experimental

Groups: The control and experimental groups were

compared using an independent t-test. The level of

VEGF in the experimental group was lower than that

in the control group at 2 weeks (p = .03) but was not

different at 4 weeks (p = .74) and was higher at 8

weeks (p = .03). The level of TSP1 in the experimental

group was higher than that in the control group at 2

weeks (po.001), was not different at 4 weeks

(p = .16), and lower at 8 weeks (po.001). The level of

CD31 in the experimental group was not different

from the control group at 2 weeks (p = .56), but was

lower at 4 and 8 weeks (p = .04, .009) (Figure 1).

Expression of Cytokines According

to Immunohistochemistry

Vascular Endothelial Growth Factor: The dermis

and epidermis showed immunohistochemical stain-

ing, indicating VEGF protein expression. In the epi-

dermis, VEGF was expressed abundantly in

keratinocytes primarily, and in the dermis VEGF was

expressed abundantly in the endothelial cells of

blood vessels (Figure 2).

3 5 : 1 0 : O C T O B E R 2 0 0 9 1 5 1 7

K I M E T A L

Thrombospondin 1: Expression of TSP1 was mainly

observed in keratinocytes in the epidermis (Figure 3).

CD31: Because blood vessels are mainly present in

the dermis, CD31, which is expressed in the endo-

thelial cells of the vessels, was mainly observed in the

dermis (Figure 4).

The Effects of IFN-a on the Redness of

Grafted Skin

In the control group, the redness of the recipient

site was 6.85. At 2, 4, and 8 weeks after the skin

graft, the redness of the grafted skin was 18.95,

11.31, and 11.17, respectively. A significant differ-

ence between the four groups was shown according

to ANOVA (po.001). Using the Scheffe test, a sig-

nificant difference was shown between the redness of

the recipient site and the redness at 2, 4, and 8 weeks

(po.001 for all), although a significant difference

was not shown between the values at 4 and 8 weeks,

(p = 1.00). In summary, the redness was greatest at 2

weeks and decreased at 4 and 8 weeks; nonetheless,

the redness at 8 weeks remained different from that

at the recipient site (Figure 5).

In the experimental group, the redness of the recip-

ient site was 7.59. The redness of the grafted skin at

2, 4, and 8 weeks was 9.12, 10.70, and 9.36,

Figure 1. Expression of vascular endothelial growth factor (VEGF), thrombospondin 1 (TSP1), and CD31 according to West-ern blot analysis (above, left). The relative protein levels of VEGF (above, right), TSP1 (below, left), and CD31 (below, right)are shown. One-way analysis of variance was used to compare the multiple group means, followed by the Tukey test. AoBoCoD or A’oB’oC’, results of the Tukey test (compared with normal skin) in the control and experimental groups,respectively. Both groups were compared using an independent t-test (significance compared with control group, ypo.05).NS, normal skin; C, control group; E, experimental group; wk, weeks after skin graft.

D E R M AT O L O G I C S U R G E RY1 5 1 8

I N T E R F E R O N -a - 2 A R E D U C E S E A R LY E RY T H E M A A F T E R S K I N G R A F T

respectively. A significant difference between the four

groups was shown using ANOVA (p= .006). Using the

Scheffe test, a significant difference was found between

the recipient site and the graft at 4 weeks (p= .007),

although there was no significant difference between the

recipient site and the graft at 2 or 8 weeks (p= .29, .19).

In summary, at 4 weeks, the redness had significantly

increased, but it had decreased at 8 weeks and was then

not different from that at the recipient site (Figure 5).

Using the independent t-test, the redness of the skin

in the control and experimental groups was com-

Figure 2. Immunohistochemistry of vascular endothelial growth factor (VEGF). In the control group, the intensity ofimmunohistochemical staining of VEGF increased at 2 weeks (A) and returned to the normal skin level at 4 (B) and 8 (C)weeks. In the experimental group, it increased maximally at 2 weeks (D), decreased at 4 weeks (E), and increased again at 8weeks (F). Red and blue arrowheads indicate positive immunohistochemical staining in keratinocytes and vascular endo-thelial cells, respectively. (G) Normal skin. (H) High magnification of (D). Lens magnification: A–G, � 100; H, � 400.

3 5 : 1 0 : O C T O B E R 2 0 0 9 1 5 1 9

K I M E T A L

pared. The redness in the experimental group was

significantly less at 2 and 8 weeks than in the control

group (po.001). Before surgery and 4 weeks after

surgery, a difference between the two groups was not

shown (p = .29, .39) (Figure 5).

The Effects of IFN-a on the Viability of the Graft

The viability rate of the grafts in the control group

was 83.7%, and the viability rate in the experimental

group was 81.7%. The viability rate was not sig-

nificantly different in the two groups (t-test; p = .66).

Figure 3. Immunohistochemistry of thrombospondin 1 (TSP1). In the control group, the intensity of immunohistochemicalstaining of TSP1 increased at 2 (A) and 4 weeks (B) and increased maximally at 8 weeks (C). In the experimental group, itincreased maximally at 2 weeks (D), decreased slightly at 4 weeks (E), and decreased to the normal skin level at 8 weeks (F).Red arrowheads indicate keratinocytes. (G) Normal skin (H) High magnification of (D). Lens magnification: A–G, � 100; H,� 400.

D E R M AT O L O G I C S U R G E RY1 5 2 0

I N T E R F E R O N -a - 2 A R E D U C E S E A R LY E RY T H E M A A F T E R S K I N G R A F T

Discussion

The level of VEGF increased maximally at the period

of ‘‘take’’ of the skin graft, at 2 weeks, and then

decreased at 4 and 8 weeks, to return to the level in

normal skin. It is probable that the angiogenesis

stimulatory factor VEGF is expressed abundantly 2

weeks after surgery to ensure the survival of the

graft, and once survival is ensured, the level de-

creases naturally. In contrast, TSP1, which is known

to inhibit angiogenesis, increased continuously

Figure 4. Immunohistochemistry of CD31. In the control group, the intensity of immunohistochemical staining of CD31increased at 4 weeks (B) and decreased at 8 weeks (C). In the experimental group, it increased at 4 weeks (E) and decreasedto the normal skin level at 8 weeks (F). Blue arrowheads indicate vascular endothelial cells. Two weeks after surgery in thecontrol (A) and experimental (D) group, respectively. (G) Normal skin. (H) High magnification of (B). Lens magnification: A–G, �100; H, �400.

3 5 : 1 0 : O C T O B E R 2 0 0 9 1 5 2 1

K I M E T A L

after skin grafting and was expressed maximally at 8

weeks. In other words, it is probable that, during the

early period after skin grafting, VEGF increases and

thus stimulates revascularization for survival. At 4 to

8 weeks, after survival is ensured, VEGF decreases,

and the angiogenesis inhibitor factor TSP1 increases,

resulting in reduced revascularization, and thus ho-

meostasis is maintained. Thus, VEGF and TSP1 are

speculated to maintain a balance during revas-

cularization of the skin graft. Our results are similar

to those of the study reported by Yano and

colleagues on the effects of ultraviolet B (UVB) ir-

radiation. According to Yano and colleagues, skin

vascularity in the upper dermis increased greatly af-

ter a single dose of UVB irradiation until day 8 and

was associated with upregulation of VEGF messen-

ger ribonucleic acid (mRNA) expression and down-

regulation of TSP1 mRNA. After 13 days, skin

vascularity was normalized, with downregulation of

VEGF mRNA expression and upregulation of TSP1

mRNA expression.8 One point of difference is that,

in our study, the expression of TSP1 was not

downregulated during the early period after surgery.

To evaluate the revascularization of the skin graft

objectively, we measured CD31, which has been

used as a vascular endothelial cell marker. We also

measured the redness of the skin graft objectively

using a chromometer. The redness of the skin color is

mainly affected by the degree of vascularization.9

Consistent with the changes in VEGF and TSP1 de-

scribed above, CD31 increased up to 4 weeks after

surgery and then decreased gradually to 8 weeks.

The redness increased maximally at 2 weeks and

decreased gradually to 4 and 8 weeks. This result is

almost in full agreement with the report of Kim

and colleagues, which described the color changes in

skin grafts with time in human patients.10 One factor

to be considered here is that the level of CD31 was

decreased at 4 weeks after surgery, whereas the

redness decreased after 2 weeks. It could be that

blood flow in vessels generated during revas-

cularization decreases slowly, owing to unknown

factors from 2 weeks after skin grafting. This spec-

ulation is in agreement with the claim of Omori and

Kurata that, after the 10th day after a skin graft,

circulation increased to a level greater than that seen

in normal skin. Circulation in the skin graft nor-

malized on the 20th day.11

As mentioned earlier, because IFN-a is a cytokine

with antiangiogenic activities,7 it could be expected

that IFN-a would inhibit the expression of VEGF

and increase that of TSP1. Review of the literature

shows that IFN-a can downregulate VEGF expres-

sion, represented by mRNA and protein levels, in

MHCC97 cells,12 and that IFN-a dose-dependently

inhibits VEGF release in peripheral blood mononu-

clear cells.13 The above reports are in agreement

with our research. At 2 weeks after surgery, the ex-

pression of VEGF in the experimental group injected

with IFN-a was lower than in the control group. At 8

weeks, the level of VEGF had increased more than in

the control group, which may be due to a rebound

phenomenon after the withdrawal of IFN-a.

Reports on the relationship between IFN-a and TSP1

are rare. It has been reported that the increased level

of platelet surface-associated TSP in essential

thrombocythemia tended to be normalized during

IFN-a treatment.14 In contrast, in our research, the

Figure 5. Redness value of the grafted skin. One-way anal-ysis of variance was used to compare the multiple groupmeans, followed by the Scheffe test. AoBoC or A’oB’, re-sults of the Scheffe test (compared with recipient site) in thecontrol and experimental groups, respectively. Both groupswere compared using the independent t-test (significancecompared with control group, ypo.05). C, control group; E,experimental group; RS, recipient site; wk, weeks after skingraft.

D E R M AT O L O G I C S U R G E RY1 5 2 2

I N T E R F E R O N -a - 2 A R E D U C E S E A R LY E RY T H E M A A F T E R S K I N G R A F T

expression of TSP1 increased in the experimental

group at 2 weeks and decreased at 8 weeks as a result

of the rebound phenomenon. More studies are re-

quired on the relationship between IFN-a and TSP1.

Consistent with the changes in cytokines described

above, CD31 decreased at 4 and 8 weeks in the

grafted skin of the experimental group, and the

redness was less at 2 and 8 weeks than in the control

group.

One potential cause for concern is that, because of

the action of IFN-a in suppressing revascularization,

the graft survival rate could be reduced, but the

survival rate in the experimental group was not sig-

nificantly different from that in the control group. In

other words, without a difference in survival rate,

administration of IFN-a reduced erythema in the

skin graft.

There have been a few studies of drugs that decrease

erythema by suppressing excessive revascularization

during the early phase of skin grafting, but IFN-ais thought to be a candidate for use in clinics.

Although more studies on human patients are re-

quired, if IFN-a can ameliorate erythema after skin

grafting, it is anticipated that the aesthetic satisfac-

tion of patients would be much greater.

One of the advantages of our study is that pigs were

used as experimental animals. Because pig skin is

similar to human skin anatomically and physiolog-

ically, the pig makes an excellent animal model for

studies of human wound healing.15 In addition, to

evaluate the redness of the skin graft objectively, a

chromometer was used. Because redness is shown

numerically on this instrument, changes in redness

and treatment effectiveness could be evaluated ac-

curately and objectively.

The shortcomings of our study are that the studies

were performed only in vivo. In the future, it may be

necessary to investigate the relationship between

cells that are present in the skin, such as keratin-

ocytes, fibroblasts, and endothelial cells, and IFN-a

in vitro. In addition, as mentioned earlier, more

studies on the relationship between IFN-a and TSP1

are required.

Acknowledgments This work was supported by

the Korea Research Foundation Grant, funded by the

Korean Government (MOEHRD) (KRF-2006-331-

E00225) and a special clinical fund of Gyeongsang

National University Hospital.

References

1. Paletta CE, Pokorny JJ, Rumbolo PM. Skin grafts. In: Mathes SJ,

editor. Plastic Surgery. 2nd ed. Philadelphia: Saunders Elsevier;

2006. p. 293–316.

2. Celik-Ozenci C, Akkoyunlu G, Korgun ET, et al. Expressions of

VEGF and its receptors in rat corpus luteum during interferon

alpha administration in early and pseudopregnancy. Mol Reprod

Dev 2004;67:414–23.

3. Zhang F, Oswald TM, Lin L, et al. Improvement of full-thickness

skin graft survival by application of vascular endothelial growth

factor in rats. Ann Plast Surg 2008;60:589–93.

4. Richter GT, Fan CY, Ozgursoy O, et al. Effect of vascular endo-

thelial growth factor on skin graft survival in Sprague-Dawley

rats. Arch Otolaryngol Head Neck Surg 2006;132:637–41.

5. Isenberg JS, Pappan LK, Romeo MJ, et al. Blockade of

thrombospondin-1-CD47 interactions prevents necrosis of full

thickness skin grafts. Ann Surg 2008;247:180–90.

6. Cho SH, Yoo SI, Noh BK, et al. Color analysis of forehead flap

and full thickness skin graft in facial reconstruction. J Kor Soc

Plast Reconstr Surg 2008;35:35–40.

7. Marschall ZV, Scholz A, Cramer T, et al. Effects of interferon

alpha on vascular endothelial growth factor gene transcription

and tumor angiogenesis. J Natl Cancer Inst 2003;95:437–48.

8. Yano K, Kajiya K, Ishiwata M, et al. Ultraviolet B-induced skin

angiogenesis is associated with a switch in the balance of vascular

endothelial growth factor and thrombospondin-1 expression. J

Invest Dermatol 2004;122:201–8.

9. Han K, Choi T, Son D. Skin color of Koreans: statistical evalu-

ation of affecting factors. Skin Res Technol 2006;12:170–7.

10. Kim JS, Park SW, Choi TH, et al. The evaluation of relevant

factors influencing skin graft changes in color over time. Dermatol

Surg 2008;34:32–9.

11. Omori S, Kurata K. Experimental studies on the blood supply to

various types of skin grafts in rabbits using isotope P32. Plast

Reconstr Surg 1960;25:547–55.

12. Wu WZ, Sun HC, Shen YF, et al. Interferon alpha 2a downreg-

ulates VEGF expression through PI3 kinase and MAP kinase

signaling pathways. J Cancer Res Clin Oncol 2005;131:169–78.

13. Salven P, Anttonen K, Repo H, et al. Endotoxins induce and in-

terferon-a suppresses vascular endothelial growth factor (VEGF)

3 5 : 1 0 : O C T O B E R 2 0 0 9 1 5 2 3

K I M E T A L

production in human peripheral blood mononuclear cells. FASEB

J 2001;15:1318–20.

14. Legrand C, Bellucci S, Disdier M, et al. Platelet thrombospondin

and glycoprotein IV abnormalities in patients with essential

thrombocythemia: effect of a-interferon treatment. Am J Hematol

1991;38:307–13.

15. Sullivan TP, Eaglstein WH, Davis SC, Mertz P. The pig as a model

for human wound healing. Wound Rep Reg 2001;9:66–76.

Address correspondence and reprint requests to: Tae HyunChoi, MD, PhD, Department of Plastic and ReconstructiveSurgery, Seoul National University College of Medicine,Seoul National University Hospital, 101 Daehang-roJongno-gu, Seoul 110-744 Republic of Korea, ore-mail: psthchoi@snu.ac.kr

D E R M AT O L O G I C S U R G E RY1 5 2 4

I N T E R F E R O N -a - 2 A R E D U C E S E A R LY E RY T H E M A A F T E R S K I N G R A F T