- Email: [email protected]
Assessment of epidermal growth factor in the healing process of clean full-thickness skin wounds
Assessment of epidermal growth factor in the healing process of clean full-thickness skin wounds Alfredo J. Jijon, MD, Donald G. Gallup, MD, M. Ali Behzadian, PhD, and William P. Metheny, PhD Augusta, Georgia Epidermal growth factor is a potent stimulant of epithelialization. However, the usefulness of topical applications of epidermal growth factor in accelerating wound healing in full-thickness skin wounds with a large panniculus adiposus has not been clear. Four full-thickness skin incisions were made in the back of 10 female pigs that treated twice a day for 14 days with 2 ml of epidermal growth factor (300 ng/ml) or 2 ml of Ringer's lactate solution in a single-blind, randomized fashion. Two pigs received only epidermal growth factor, two pigs received only Ringer's lactate solution, and six pigs were treated with both solutions. The original skin plug was weighed to ensure similarity of groups. Photographs and measurements of each incision were taken every 7 days. The mean surface areas of the incisions treated with epidermal growth factor were 8.45, 7.50, and 2.30 cm2; in the incisions treated with Ringer's lactate solution the measurements were 8.42, 8.16, and 2.37 cm 2 on observation days 1, 7, and 14, respectively. Although a trend toward a faster healing rate was noted in the incisions treated with epidermal growth factor, this difference was not statistically significant. With the doses and the time interval used between treatments, minimal benefit was obtained with epidermal growth factor when compared with Ringer's lactate solution. (AM J OSSTET GYNECOL 1989;161 :1658-62.)
Key words: Epidermal growth factor, wound healing
The different properties of epidermal growth factor have been established. It is a potent stimulant of epithelialization both in vivo and in vitro. It has been noted to increase the thickness of the epidermis, increase the epithelial cell proliferation and keratinization, I and accelerate wound contraction." Thus its use generally results in decreased wound contracture and scarring. 3 Epidermal growth factor also decreases the thickness of the skin by decreasing the amount of fat, I and stimulates protein synthesis.' Most studies on the woundhealing properties of epidermal growth factor have been done in split- or partial-thickness skin wounds in animals that have a small panniculus adiposus. 3 5 h-8 Our study was designed to evaluate the in vivo effectiveness of epidermal growth factor in accelerating epithelial regeneration in full-thickness skin wounds in an experimental animal model with a large panniculus adiposus to determine the potential application of epidermal growth factor in superficial abdominal wound breakdown. If wound healing is accelerated, the use of epidermal growth factor could potentially decrease From the Department of Obstetrirs and Gynecology, Medical College of Georgia. Presented at the Thirty-sixth Annual Meetzng of the SOCIety for G.vnecologzc InvestigatIOn, San Diego, Californza, March 15-18,
1989.
Reprznt requests: Donald G. Gallup, MD, Section of Gynecologic Oncology, CK-166, Medical College of Georgia, August, GA
30912.
6/6/15662
1658
hospitalization time, costs, and discomfort for affected patients. Its application could be useful in patients with infection, hypoproteinemia, and diabetes, not uncommon conditions occurring in some patients in our specialty.
Material and methods Ten female pigs of approximately 2 to 3 months of age served as animal models. Pigs were chosen because of their large panniculus adiposus. Four full-thickness skin incisions, including the panniculus carnosus, were made with a sterile knife in the back of each animal, two on the right side and two on the left side. Each incision was 5 cm long and 1.5 cm wide. The cuts were made with a sterile technique and with the animals under general anesthesia. Permission for this study was obtained from the Committee on Animal Use for Research and Education at the Medical College of Georgia; all guidelines for the care and use of animals at our institution were followed. The tissue removed was weighed. Each wound was assigned a number, measured, and photographed from a standard distance of 50 cm. Two solutions were prepared; one contained 300 ng of epidermal growth factor per milliliter of Ringer's lactate solution, and the other contained only Ringer's lactate solution. A laboratory assistant, blinded to the intent of the study, marked the two solutions as red or green.
Epidermal growth factor and wound healing
Volume 161 l'\umber 6. Part I
The ten pigs were divided into three groups: group I, two pigs with eight incisions that were treated only with green solution; group 2, two pigs with eight incisions treated only with red solution; group 3, six pigs in which the incisions on the left received the green solution and the incisions on the right received the red solution. Thus a total number of 40 incisions were studied; 20 were treated with the green solution and 20 with the red solution. Treatments of the incisions were done twice a day after the pigs were sedated. Wet-todry dressing changes with 2 ml of either the green or the red solution were applied. Each pig was caged separately and supplied with regular food and water. The duration of the project was 14 days. Measurements of the incisions were done every 7 days with the pigs under heavy sedation, for a total of three measurements of each incision. The maximum and minimum length, depth, and width were obtained, and from these values a maximum, minimum, and mean area and volume were calculated. Each incision was photographed every 7 days with a ruler placed on its side to more accurately measure the surface of the wound. Color slides were obtained. Each individual slide was projected onto sheets of standardized batches of paper from exactly the same distance so that the length of the ruler was always kept constant. The margins of each wound picture were traced, cut, and weighed. As a sideline to this experiment, we compared the correlations among the three different measurements used to assess wound healing. We examined these correlations to determine whether all three measurements are necessary for future studies in this line of investigation. A strong correlation (>0.80) would suggest that only one assessment method is necessary. Before the analysis the code was broken, and it was established that the green solution contained the epidermal growth factor and the red solution contained the placebo. The results were analyzed with an independent t test and a repeated-measures analysis of variance test with a single covariate, and the Pearson Product-Moment Lorrelation coefficient. The a-value was set at 0.05 with a one-tailed test. We considered that the sample size of 20 incisions per condition would provide sufficient power to detect significant differences, judging from the large effect size (>0.08) yet low power «0.30) as a result of small samples observed in previous studies."' OJ. 0-9
Results During each observation means and standard deviations were calculated for each of the three variables: surface area, wound volume, and weight of the tracing paper. To ensure that the individual values did not
1659
Table I. Initial values (mean ± SD) of wounds per treatment condition Rmger's lactate solutIOn and epIdermal gruwth fart01
RmKer's lactate solutIOn and j)larebo
Skin plug (gm) Surface area (em') Wound area
3,43 ± 086 8,45 ± 0,41
3.04 ± 0.68 8,41 ± 0,48
3.67 ± 0.98
4.06 ± 0.93
Weight of paper tracing (gm)
0.82 ± 0.07
0.84 ± 0.08
A.I.lessment measll1emen/1
(ems)
make a difference, we used the weight of the skin plug as covariate in our analysis. The weight of the skin plug removed at the beginning of the experiment was nearly equivalent for both groups, and the values were not statistically different according to an independent t test. The values of the two groups with regard to surface area, wound volume, and the weight of paper tracing were roughly equivalent, and no statistically significant differences were found between them (Table I). We subsequently performed three separate analyses and used the weight of the skin plug removed from each pig at the beginning of the experiment as a control for each. In the first analyses we compared the mean surface area for the 3 days of observation. On observation day 1 (initiation of the study) the mean surface area was 8.45 cm' in the epidermal growth factor group and 8.41 cm' in the placebo group. On observation day 7 the area was 7.50 cm' (± 2.1) and 8.16 cm' (± 2.16) in the epidermal growth factor and placebo groups, respectively. On observation day 14 the area was 2.30 cm 2 (± 1.15) and 2.37 cm' (± 1.18) in the epidermal growth factor and placebo groups respectively (Fig. 1). An identical analysis was conducted for the weight of the paper tracing. On observation day 1 the weight of the paper tracing of the projected slides of the incisions was 0.82 gm. and 0.84 mg in the epidermal growth factor and placebo groups, respectively. On observation day 7 the weight was 0.75 gm (±0.26) and 0.80 gm (± 0.23) in the epidermal growth factor and placebo groups, respectively. On observation day 14 the weight was 0.24 gm (± 0.11) and 0.25 gm (± 0.12) in the epidermal growth factor and placebo groups, respectively (Fig. 2). In the third analysis the mean wound volume only was compared during the first two observations because it was thought that at the end of the experiment all incisions could have a depth of zero. On observation day 1 the wound volume was 3.67 cm I and 4.06 cm" in the epidermal growth factor and placebo groups, respectively. On observation day 7 the volume was
1660 Jijon et al.
December l'lH9
Am
5
10
-
----a---
'"E u
.::
. :." .
8 7
RINGERS + EGF RINGERS + PLACEBO
'"E
"
-----0--
4
I:
4>
6
E :>
3
'":>
2
a>
5
u
I:
!!
5
J Obstet (;vncwl
RINGERS + EGF RINGERS + PLACEBO
0 ~
.
~
". ~
I:
4>
:;
:E
O~----r----------r--------~
Day 1
Day 7
Day 1
Day 14
Fig. 1. Plotted comparison of two treatments for wound healing measured by mean surface area in square centimeters at three observation points.
0.9 0.8
"C,
0.7
E
:: :E
'"
;; ~
'" 'u ~
~
l;
~
Q.
RINGERS + EGF RINGERS + PLACEBO
0.6 0.5 0.4 0.3 0.2 0.1 0.0 Day 1
Day 7
__----
Day 14
Observations
Observations
~
Day 7
Day 14
Observations
Fig. 2. Plotted comparison at two treatments tor wound healing measured by mean weight in grams from paper tracing' of surface area photographs at three observation points.
0.72 cm' (± 1.28) and 0.92 cm' (± 1.33) in the epidermal growth factor and placebo groups, respectively. On observation day 14 both groups of incisions had a volume of 0 (Fig. 3). Although a faster healing rate was noted in lI1cislOns treated with epidermal growth factor compared with those treated with placebo, in none of the three analyses was this difference found to be statistically significant. Nevertheless, all three analyses demonstrated a time effect in the healing rate, independent of group. Most healing occurred during the first week. The incisions were then analyzed according to the initial three groups of treatment: group 1, those incisions made in the two pigs that only received epidermal growth factor; group 2, those incisions made in the two pigs that only received placebo; and group 3, those incisions made in the six pigs that received both epi-
Fig. 3. Plotted companson ot two treatment;, tor wound healing measured by mean volume in cubic centimeters at three observation points.
dermal growth factor and placebo. This group was divided for the purpose of the analysis into two groups: group 3.1, incisions on the left side of the pigs that received only epidermal growth factor; group 3.2, incisions on the right side of the pigs that received only placebo. In the study of the incisions by treatment groupings, we noted that the groups were statistically equivalent (see Figs. 4 and 5) at the onset of the experiment. However, by observation day 7 the incisions in both groups 1 and 2 had healed faster than those in both groups 3.1 and 3.2. The groups were roughly equivalent by observation day 14, since no statistically significant differences appeared. These results indicate that incisions on animals treated with only one solution, either epidermal growth factor or placebo, healed faster than incisions on animals that received epidermal growth factor on the left side and placebo on the right side. As a third component of this study, we examined the correlations among the three different methods used to assess wound healing. The correlation of 0.92 between surface area and mean wound volume indicates that these two methods are equivalent. Surface area also correlated significantly with the weight of the paper tracing measure (r = 0.75). The paper tracing measure showed a weaker relationship with mean wound volume (r = 0.65).
Comment Controversy still exists concerning the actual benefit of epidermal growth factor in accelerating wound healing. Some have found epidermal growth factor to be usefuV 3 j.6 " whereas others have found no significant effect. 7 " Because of this controversy we used three different assessment methods of wound healing to eval-
Epidermal growth factor and wound healing
Volume Hi! I'umber 6. Part I
uate epidermal growth factor in the pig model. Thus confirmation of our values was attained . Analyses of the correlation of the different assessment methods showed that surface area provides the optimal method for assessing wound healing. Many animals tend to lick their open wounds. This observation and the documented presence of epidermal growth factor in the saliva led Niall et al. fi to evaluate its use in open granulating wounds. They noted that topical applications of epidermal growth factor enhanced wound closure and demonstrated that the healing rate was faster in both intact or castrated male rats than in female rats. They postulated an androgendependent cell sensitivity to epidermal growth factor but noted that such sensitivity does not require the continuous presence of androgens for its expression. Hutson et al. 2 demonstrated that mice caged in groups healed faster than those caged separately; however, sialoadenectomized mice caged in groups did not. To avoid the potential androgenic influence on the activity of epidermal growth factor, we used only female pigs. Each pig was caged separately to avoid licking of the incisions. Franklin and Lynch" performed full-thickness incisions in rabbits' ears and demonstrated that the wound defects treated with epidermal growth factor filled sooner with more cellular elements, thus leaving less space for collapse by wound contracture. Brown et al." also found that epidermal growth factor significantly accelerates epidermal regeneration in partial-thickness skin burns of miniature pigs. Greaves et al. 7 were unable to see any effect in otherwise healthy skin wounds of five human volunteers. Thornton et al." used once-aday treatments in partial-thickness skin wounds and obtained similar results. We chose twice-a-day treatments, a regimen frequently used in wound breakdowns in our patients. Epidermal growth factor has two types of cellular receptors. The high-affinity receptors seem to play the major role for epidermal growth factor stimulation of cell division, and these receptors are stimulated at low concentrations of epidermal growth factor. Lowaffinity receptors are stimulated with higher concentrations of epidermal growth factor, a process of downregulation of the receptors is produced, and an inhibitory effect of cell division is seen. 1U. 11 Behzadian et aLII demonstrated that blocking low-affinity receptors by a specific monoclonal antibody does not decrease the cellular response to epidermal growth factor and the subsequent stimulation of deoxyribonucleic acid synthesis. They concluded that low concentrations of epidermal growth factor (100 to 200 ng / ml), which mainly stimulate high-affinity receptors, are ideal for cellular proliferation. Our choice of 300 ng/ml was based on this information. However, we recognize that the topical
.E
10
1661
GROUP 1
---0--
GROUP 2
8
.. . u
GROUP 3.1
.:
GROUP 3.2
~
6
!l
5
~
4
c:
:
3
2
Day 1
Day 7
Day
14
Ob •• rvltlonl
Fig. 4. Plotted comparison of two treatments. grouped by condition within subject. for wound healing measured by mean surface area in centimeters at three observation points: group I, incisions treated with epidermal growth factor only; group 2, incisions treated with placebo only; group 3.1. incisions on left side treated with epidermal growth factor; group 3.2, incisions on right side treated with placebo.
1.0
---0--
0.9
E
0.8
.E
'"
0.7
'" .2'
0.6
J
0.5
'"c:
0.4
~
.
~ ...;; :.
GROUP 1 GROUP 2 GROUP 3.1 GROUP 3.2
0.3 0.2 0.1 0.0
Day 1
Day 7
Day 14
Ob.ervatlons
Fig. 5. Plotted comparison of two treatments, grouped by condition within subject. for wound healing measured by mean weight in grams from paper tracing of surface area photographs at three observation points: Group I. incisions treated with epidermal growth factor only; group 2. incisions treated with placebo only; group 3.1 . incisions on left side treated with epidermal growth factor ; group 3.2, incisions on right side treated with placebo.
application of the ligand on a live animal could never be as precise and uniform as noted in in vitro studies. We applied 2.0 ml of the solution to each wound, since this amount completely moistened the gauze. When the analysis was done by groups of treatment, pigs with half of their incisions treated with either epidermal growth factor or placebo were statistically different from those pigs with all four of their incisions treated with either epidermal growth factor or placebo. The healing rate was faster when only one solution was used on the same animal. However, no significant dif-
1662 Jijon et al.
ference was noted between the incisions on the animals that received either only one solution or both solutions. These results suggest that a systemic regulation of epidermal growth factor may be present. In summary, a trend toward faster healing in the incisions treated with epidermal growth factor was noted. The benefit of using epidermal growth factor to accelerate the healing of full-thickness skin wounds in the pig model was not demonstrated in this study. Further research is needed to assess the potential of epidermal growth factor in wound healing. REFERENCES 1. Carpenter G, Cohen S. Epidermal growth factor. Annu Rev Biochem 1979;48: 193-216. 2. Hutson JM, Niall M, Evans D, et al. Effects of salivary glands on wound contraction in mice. Nature 1979; 279 :793·5. 3. Franklin JD, Lynch JB. Effects of topical applications of epidermal growth factor on wound healing. An experimental study on rabbit ears. Plast Reconstr Surg 1979; 64:766-70.
December 1989 Am J Obstet Gynecol
4. Serrero G. EGF inhibits the differentiation of adipocyte precursors in primary cultures. Biochem Biophys Res Commun 1987;146:194-202. 5. Hiramatsu, M, Hatakeyama K, Minami N, et al. Increase in collagen synthesis of cotton pellet granuloma in rats by epidermal growth factor. Jpn J Pharmacol 1982;32: 198201. 6. Niall M, Graeme BR, O'Brien BMcC. The effect of epidermal growth factor in wound healing in mice. J Surg Res 1982;33:164-9. 7. Greaves MW. Lack of effect of topically applied epidermal growth factor (EGF) on epidermal growth in man in vivo. Clin Exp DermatoI1980;5:101-3. 8. Thornton JW, Hess CA, Cas singham V, et al. Epidermal growth factor in the healing of second degree burns: a controlled animal study. Burns 1981;8:156-60. 9. Brown GI, Curtsinger L, Brightwell JR, et al. Enhancement of epidermal regeneration by biosynthetic epidermal growth factor.J Exp Med 1986;163:1319-4. 10. King AC, Cuatrecasas P. Resolution of high and low affinity epidermal growth factor receptors. J Bioi Chern 1982;257:3053-60. 11. Behzadian MA, Shimizu N. Monoclonal antibody that immunoreacts with a subclass of human receptors for epidermal growth factor. Cell Struct Funct 1985; 10:219.
Key words: Epidermal growth factor, wound healing
The different properties of epidermal growth factor have been established. It is a potent stimulant of epithelialization both in vivo and in vitro. It has been noted to increase the thickness of the epidermis, increase the epithelial cell proliferation and keratinization, I and accelerate wound contraction." Thus its use generally results in decreased wound contracture and scarring. 3 Epidermal growth factor also decreases the thickness of the skin by decreasing the amount of fat, I and stimulates protein synthesis.' Most studies on the woundhealing properties of epidermal growth factor have been done in split- or partial-thickness skin wounds in animals that have a small panniculus adiposus. 3 5 h-8 Our study was designed to evaluate the in vivo effectiveness of epidermal growth factor in accelerating epithelial regeneration in full-thickness skin wounds in an experimental animal model with a large panniculus adiposus to determine the potential application of epidermal growth factor in superficial abdominal wound breakdown. If wound healing is accelerated, the use of epidermal growth factor could potentially decrease From the Department of Obstetrirs and Gynecology, Medical College of Georgia. Presented at the Thirty-sixth Annual Meetzng of the SOCIety for G.vnecologzc InvestigatIOn, San Diego, Californza, March 15-18,
1989.
Reprznt requests: Donald G. Gallup, MD, Section of Gynecologic Oncology, CK-166, Medical College of Georgia, August, GA
30912.
6/6/15662
1658
hospitalization time, costs, and discomfort for affected patients. Its application could be useful in patients with infection, hypoproteinemia, and diabetes, not uncommon conditions occurring in some patients in our specialty.
Material and methods Ten female pigs of approximately 2 to 3 months of age served as animal models. Pigs were chosen because of their large panniculus adiposus. Four full-thickness skin incisions, including the panniculus carnosus, were made with a sterile knife in the back of each animal, two on the right side and two on the left side. Each incision was 5 cm long and 1.5 cm wide. The cuts were made with a sterile technique and with the animals under general anesthesia. Permission for this study was obtained from the Committee on Animal Use for Research and Education at the Medical College of Georgia; all guidelines for the care and use of animals at our institution were followed. The tissue removed was weighed. Each wound was assigned a number, measured, and photographed from a standard distance of 50 cm. Two solutions were prepared; one contained 300 ng of epidermal growth factor per milliliter of Ringer's lactate solution, and the other contained only Ringer's lactate solution. A laboratory assistant, blinded to the intent of the study, marked the two solutions as red or green.
Epidermal growth factor and wound healing
Volume 161 l'\umber 6. Part I
The ten pigs were divided into three groups: group I, two pigs with eight incisions that were treated only with green solution; group 2, two pigs with eight incisions treated only with red solution; group 3, six pigs in which the incisions on the left received the green solution and the incisions on the right received the red solution. Thus a total number of 40 incisions were studied; 20 were treated with the green solution and 20 with the red solution. Treatments of the incisions were done twice a day after the pigs were sedated. Wet-todry dressing changes with 2 ml of either the green or the red solution were applied. Each pig was caged separately and supplied with regular food and water. The duration of the project was 14 days. Measurements of the incisions were done every 7 days with the pigs under heavy sedation, for a total of three measurements of each incision. The maximum and minimum length, depth, and width were obtained, and from these values a maximum, minimum, and mean area and volume were calculated. Each incision was photographed every 7 days with a ruler placed on its side to more accurately measure the surface of the wound. Color slides were obtained. Each individual slide was projected onto sheets of standardized batches of paper from exactly the same distance so that the length of the ruler was always kept constant. The margins of each wound picture were traced, cut, and weighed. As a sideline to this experiment, we compared the correlations among the three different measurements used to assess wound healing. We examined these correlations to determine whether all three measurements are necessary for future studies in this line of investigation. A strong correlation (>0.80) would suggest that only one assessment method is necessary. Before the analysis the code was broken, and it was established that the green solution contained the epidermal growth factor and the red solution contained the placebo. The results were analyzed with an independent t test and a repeated-measures analysis of variance test with a single covariate, and the Pearson Product-Moment Lorrelation coefficient. The a-value was set at 0.05 with a one-tailed test. We considered that the sample size of 20 incisions per condition would provide sufficient power to detect significant differences, judging from the large effect size (>0.08) yet low power «0.30) as a result of small samples observed in previous studies."' OJ. 0-9
Results During each observation means and standard deviations were calculated for each of the three variables: surface area, wound volume, and weight of the tracing paper. To ensure that the individual values did not
1659
Table I. Initial values (mean ± SD) of wounds per treatment condition Rmger's lactate solutIOn and epIdermal gruwth fart01
RmKer's lactate solutIOn and j)larebo
Skin plug (gm) Surface area (em') Wound area
3,43 ± 086 8,45 ± 0,41
3.04 ± 0.68 8,41 ± 0,48
3.67 ± 0.98
4.06 ± 0.93
Weight of paper tracing (gm)
0.82 ± 0.07
0.84 ± 0.08
A.I.lessment measll1emen/1
(ems)
make a difference, we used the weight of the skin plug as covariate in our analysis. The weight of the skin plug removed at the beginning of the experiment was nearly equivalent for both groups, and the values were not statistically different according to an independent t test. The values of the two groups with regard to surface area, wound volume, and the weight of paper tracing were roughly equivalent, and no statistically significant differences were found between them (Table I). We subsequently performed three separate analyses and used the weight of the skin plug removed from each pig at the beginning of the experiment as a control for each. In the first analyses we compared the mean surface area for the 3 days of observation. On observation day 1 (initiation of the study) the mean surface area was 8.45 cm' in the epidermal growth factor group and 8.41 cm' in the placebo group. On observation day 7 the area was 7.50 cm' (± 2.1) and 8.16 cm' (± 2.16) in the epidermal growth factor and placebo groups, respectively. On observation day 14 the area was 2.30 cm 2 (± 1.15) and 2.37 cm' (± 1.18) in the epidermal growth factor and placebo groups respectively (Fig. 1). An identical analysis was conducted for the weight of the paper tracing. On observation day 1 the weight of the paper tracing of the projected slides of the incisions was 0.82 gm. and 0.84 mg in the epidermal growth factor and placebo groups, respectively. On observation day 7 the weight was 0.75 gm (±0.26) and 0.80 gm (± 0.23) in the epidermal growth factor and placebo groups, respectively. On observation day 14 the weight was 0.24 gm (± 0.11) and 0.25 gm (± 0.12) in the epidermal growth factor and placebo groups, respectively (Fig. 2). In the third analysis the mean wound volume only was compared during the first two observations because it was thought that at the end of the experiment all incisions could have a depth of zero. On observation day 1 the wound volume was 3.67 cm I and 4.06 cm" in the epidermal growth factor and placebo groups, respectively. On observation day 7 the volume was
1660 Jijon et al.
December l'lH9
Am
5
10
-
----a---
'"E u
.::
. :." .
8 7
RINGERS + EGF RINGERS + PLACEBO
'"E
"
-----0--
4
I:
4>
6
E :>
3
'":>
2
a>
5
u
I:
!!
5
J Obstet (;vncwl
RINGERS + EGF RINGERS + PLACEBO
0 ~
.
~
". ~
I:
4>
:;
:E
O~----r----------r--------~
Day 1
Day 7
Day 1
Day 14
Fig. 1. Plotted comparison of two treatments for wound healing measured by mean surface area in square centimeters at three observation points.
0.9 0.8
"C,
0.7
E
:: :E
'"
;; ~
'" 'u ~
~
l;
~
Q.
RINGERS + EGF RINGERS + PLACEBO
0.6 0.5 0.4 0.3 0.2 0.1 0.0 Day 1
Day 7
__----
Day 14
Observations
Observations
~
Day 7
Day 14
Observations
Fig. 2. Plotted comparison at two treatments tor wound healing measured by mean weight in grams from paper tracing' of surface area photographs at three observation points.
0.72 cm' (± 1.28) and 0.92 cm' (± 1.33) in the epidermal growth factor and placebo groups, respectively. On observation day 14 both groups of incisions had a volume of 0 (Fig. 3). Although a faster healing rate was noted in lI1cislOns treated with epidermal growth factor compared with those treated with placebo, in none of the three analyses was this difference found to be statistically significant. Nevertheless, all three analyses demonstrated a time effect in the healing rate, independent of group. Most healing occurred during the first week. The incisions were then analyzed according to the initial three groups of treatment: group 1, those incisions made in the two pigs that only received epidermal growth factor; group 2, those incisions made in the two pigs that only received placebo; and group 3, those incisions made in the six pigs that received both epi-
Fig. 3. Plotted companson ot two treatment;, tor wound healing measured by mean volume in cubic centimeters at three observation points.
dermal growth factor and placebo. This group was divided for the purpose of the analysis into two groups: group 3.1, incisions on the left side of the pigs that received only epidermal growth factor; group 3.2, incisions on the right side of the pigs that received only placebo. In the study of the incisions by treatment groupings, we noted that the groups were statistically equivalent (see Figs. 4 and 5) at the onset of the experiment. However, by observation day 7 the incisions in both groups 1 and 2 had healed faster than those in both groups 3.1 and 3.2. The groups were roughly equivalent by observation day 14, since no statistically significant differences appeared. These results indicate that incisions on animals treated with only one solution, either epidermal growth factor or placebo, healed faster than incisions on animals that received epidermal growth factor on the left side and placebo on the right side. As a third component of this study, we examined the correlations among the three different methods used to assess wound healing. The correlation of 0.92 between surface area and mean wound volume indicates that these two methods are equivalent. Surface area also correlated significantly with the weight of the paper tracing measure (r = 0.75). The paper tracing measure showed a weaker relationship with mean wound volume (r = 0.65).
Comment Controversy still exists concerning the actual benefit of epidermal growth factor in accelerating wound healing. Some have found epidermal growth factor to be usefuV 3 j.6 " whereas others have found no significant effect. 7 " Because of this controversy we used three different assessment methods of wound healing to eval-
Epidermal growth factor and wound healing
Volume Hi! I'umber 6. Part I
uate epidermal growth factor in the pig model. Thus confirmation of our values was attained . Analyses of the correlation of the different assessment methods showed that surface area provides the optimal method for assessing wound healing. Many animals tend to lick their open wounds. This observation and the documented presence of epidermal growth factor in the saliva led Niall et al. fi to evaluate its use in open granulating wounds. They noted that topical applications of epidermal growth factor enhanced wound closure and demonstrated that the healing rate was faster in both intact or castrated male rats than in female rats. They postulated an androgendependent cell sensitivity to epidermal growth factor but noted that such sensitivity does not require the continuous presence of androgens for its expression. Hutson et al. 2 demonstrated that mice caged in groups healed faster than those caged separately; however, sialoadenectomized mice caged in groups did not. To avoid the potential androgenic influence on the activity of epidermal growth factor, we used only female pigs. Each pig was caged separately to avoid licking of the incisions. Franklin and Lynch" performed full-thickness incisions in rabbits' ears and demonstrated that the wound defects treated with epidermal growth factor filled sooner with more cellular elements, thus leaving less space for collapse by wound contracture. Brown et al." also found that epidermal growth factor significantly accelerates epidermal regeneration in partial-thickness skin burns of miniature pigs. Greaves et al. 7 were unable to see any effect in otherwise healthy skin wounds of five human volunteers. Thornton et al." used once-aday treatments in partial-thickness skin wounds and obtained similar results. We chose twice-a-day treatments, a regimen frequently used in wound breakdowns in our patients. Epidermal growth factor has two types of cellular receptors. The high-affinity receptors seem to play the major role for epidermal growth factor stimulation of cell division, and these receptors are stimulated at low concentrations of epidermal growth factor. Lowaffinity receptors are stimulated with higher concentrations of epidermal growth factor, a process of downregulation of the receptors is produced, and an inhibitory effect of cell division is seen. 1U. 11 Behzadian et aLII demonstrated that blocking low-affinity receptors by a specific monoclonal antibody does not decrease the cellular response to epidermal growth factor and the subsequent stimulation of deoxyribonucleic acid synthesis. They concluded that low concentrations of epidermal growth factor (100 to 200 ng / ml), which mainly stimulate high-affinity receptors, are ideal for cellular proliferation. Our choice of 300 ng/ml was based on this information. However, we recognize that the topical
.E
10
1661
GROUP 1
---0--
GROUP 2
8
.. . u
GROUP 3.1
.:
GROUP 3.2
~
6
!l
5
~
4
c:
:
3
2
Day 1
Day 7
Day
14
Ob •• rvltlonl
Fig. 4. Plotted comparison of two treatments. grouped by condition within subject. for wound healing measured by mean surface area in centimeters at three observation points: group I, incisions treated with epidermal growth factor only; group 2, incisions treated with placebo only; group 3.1. incisions on left side treated with epidermal growth factor; group 3.2, incisions on right side treated with placebo.
1.0
---0--
0.9
E
0.8
.E
'"
0.7
'" .2'
0.6
J
0.5
'"c:
0.4
~
.
~ ...;; :.
GROUP 1 GROUP 2 GROUP 3.1 GROUP 3.2
0.3 0.2 0.1 0.0
Day 1
Day 7
Day 14
Ob.ervatlons
Fig. 5. Plotted comparison of two treatments, grouped by condition within subject. for wound healing measured by mean weight in grams from paper tracing of surface area photographs at three observation points: Group I. incisions treated with epidermal growth factor only; group 2. incisions treated with placebo only; group 3.1 . incisions on left side treated with epidermal growth factor ; group 3.2, incisions on right side treated with placebo.
application of the ligand on a live animal could never be as precise and uniform as noted in in vitro studies. We applied 2.0 ml of the solution to each wound, since this amount completely moistened the gauze. When the analysis was done by groups of treatment, pigs with half of their incisions treated with either epidermal growth factor or placebo were statistically different from those pigs with all four of their incisions treated with either epidermal growth factor or placebo. The healing rate was faster when only one solution was used on the same animal. However, no significant dif-
1662 Jijon et al.
ference was noted between the incisions on the animals that received either only one solution or both solutions. These results suggest that a systemic regulation of epidermal growth factor may be present. In summary, a trend toward faster healing in the incisions treated with epidermal growth factor was noted. The benefit of using epidermal growth factor to accelerate the healing of full-thickness skin wounds in the pig model was not demonstrated in this study. Further research is needed to assess the potential of epidermal growth factor in wound healing. REFERENCES 1. Carpenter G, Cohen S. Epidermal growth factor. Annu Rev Biochem 1979;48: 193-216. 2. Hutson JM, Niall M, Evans D, et al. Effects of salivary glands on wound contraction in mice. Nature 1979; 279 :793·5. 3. Franklin JD, Lynch JB. Effects of topical applications of epidermal growth factor on wound healing. An experimental study on rabbit ears. Plast Reconstr Surg 1979; 64:766-70.
December 1989 Am J Obstet Gynecol
4. Serrero G. EGF inhibits the differentiation of adipocyte precursors in primary cultures. Biochem Biophys Res Commun 1987;146:194-202. 5. Hiramatsu, M, Hatakeyama K, Minami N, et al. Increase in collagen synthesis of cotton pellet granuloma in rats by epidermal growth factor. Jpn J Pharmacol 1982;32: 198201. 6. Niall M, Graeme BR, O'Brien BMcC. The effect of epidermal growth factor in wound healing in mice. J Surg Res 1982;33:164-9. 7. Greaves MW. Lack of effect of topically applied epidermal growth factor (EGF) on epidermal growth in man in vivo. Clin Exp DermatoI1980;5:101-3. 8. Thornton JW, Hess CA, Cas singham V, et al. Epidermal growth factor in the healing of second degree burns: a controlled animal study. Burns 1981;8:156-60. 9. Brown GI, Curtsinger L, Brightwell JR, et al. Enhancement of epidermal regeneration by biosynthetic epidermal growth factor.J Exp Med 1986;163:1319-4. 10. King AC, Cuatrecasas P. Resolution of high and low affinity epidermal growth factor receptors. J Bioi Chern 1982;257:3053-60. 11. Behzadian MA, Shimizu N. Monoclonal antibody that immunoreacts with a subclass of human receptors for epidermal growth factor. Cell Struct Funct 1985; 10:219.