Skin Research and Technology 2008; 14: 436439
Printed in Singapore
All rights reserved
doi: 10.1111/j.1600-0846.2008.00308.x
Changes of human skin in subepidermal wound healing
process
Keiichi Sugata, Takashi Kitahara and Yoshinori Takema
Biological Science Laboratories, Kao Corporation, Tochigi, Japan
Background/purpose: The wound healing process involves
unexplained mechanisms. An aberration in this process is
known to cause dermal disorders such as keloid or hyper-
trophic scars, but the mechanism by which these scars are
formed remains to be elucidated. Here we examined the
usefulness of a non-invasive optical imaging device to
clarify mechanisms of wound healing and of scar formation.
Methods: An 8 mm experimental wound was made in
the forearms of six subjects by a suction blister method.
To observe chronological changes associated with wound
healing, horizontal cross-sectional images were non-inva-
sively obtained of the wounded area from the skin surface
down to 129 mm below at 21.5 mm intervals using in vivo
laser confocal scanning microscopy (LCSM).
Results: The wounds were covered with a new epidermis by
week 2, at which time the dermal papilla count decreased
while the thickness from the skin surface to the apex of the
dermal papilla increased. The count and the thickness
returned to the initial levels when the wound was healed.
D ERMAL WOUND healing requires continuous
modification of the complex cellular sup-
port matrix. The mechanism has not been clar-
ified yet, but it is believed that an imbalance
during wound healing causes fibrous tissue over-
growth known as keloid or hypertrophic scars.
Besides aesthetic problems, these disorders tend
to cause irritation or even pain. The mechanism
of these dermal disorders is likewise unknown
due to the lack of proper animal models. In
particular, there is no treatment for keloid that
is considered to be effective in all patients due to
the fact that an excision of the affected region of a
patient may induce the formation of similar or
even larger scars.
In this study, we non-invasively observed the
chronological changes in healing of superficial
wounds using in vivo laser confocal scanning
microscopy (LCSM), which provides us an image
of the human skin at any depth (1, 2). The aim of
this study was to assess the usefulness of this
436
r 2008 The Authors
Journal compilation r 2008 Blackwell Munksgaard
Skin Research and Technology
In two out of six subjects, fibrous tissues were observed in
the upper dermis, whereas in one other subject, melano-
cyte-like dendritic cells were observed in the epidermis
dermis border in later phases of wound healing.
Conclusion: This non-invasive method using in vivo LCSM
revealed chronological changes in the dermis and epidermis
during wound healing. In addition, although a scar was not
formed in any of study subjects, this microscopy revealed
aspects similar to the fibrous tissue overgrowth or to mela-
nocyte migration, both of which may relate to wound healing.
These results indicate the usefulness of this non-invasive
method in studies of wound healing and of scar formation.
Key words: wound healing scar in vivo confocal laser
scanning microscopy
& Blackwell Munksgaard, 2008
Accepted for publication 29 December 2007
non-invasive method in understanding the me-
chanism and variations of wound healing.
Materials and Methods
Subjects
The subjects used in this study were six healthy
males (mean age, 34 years). After explaining the
test to all subjects, their written consent was
obtained, and adequate consideration was given
to their human rights and safety throughout the
study.
Experimental wound
An experimental superficial wound was created
by the suction blister method (3, 4). Firstly, the
blistering device with a 6-mm diameter syringe
induced a blister using a vacuum pump (DA-
30D, ULVAC Inc., Kanagawa, Japan) for appro-
ximately 1.5 h. Secondly, the 6-mm diameter

syringe was replaced by an 8-mm diameter syr-
inge in order to raise the blister. Finally, the blister
roof (the epidermis) was removed from the arm
skin, after which the wound was covered with a
transparent dressing (3M Co. Ltd., St. Paul, MN,
USA).
Analysis of internal skin structure
As shown in Fig. 1, 10 visual fields (2.7  0.8 mm)
were fixed in the central-outer area sof each
wound. Using LCSM (Vivascope 1000 , Lucid
Inc., Rochester, NY, USA), a horizontal image
was taken in each visual field from the skin
surface to a depth of 129 mm at 21.5 mm intervals.
A surface image of the wound
s
was photographed
using a Minolta a 707 Si with 50 mm macro lens
(Konica Minolta Holdings Inc., Osaka, Japan).
Fig. 1. Representative image of the skin surface 2 weeks after removal
of the epidermis. Confocal images were captured from 10 fields
(2.7  0.8 mm) within the suction blister wound (dashed line).
Fig. 2. Representative confocal images of internal skin after removal of the epidermis. Image size: 400  540 mm.
Subepidermal wound healing process
These procedures were repeated at weeks 1, 2,
4, 6 and 8.
Results
Figure 2 shows the representative photographs at
each depth, taken before inducing the suction
blister wound and during the wound healing
process.
(1) Before inducing the suction blister wound: Der-
mal papillae were observed (21.5 mm). Fine
fibrous elements of the reticular layer were
observed (64.5 mm).
(2) Two weeks after creation of the suction blister
wound: The smooth shiny epidermis appeared
after a blood clot soughed (0 mm). Large
keratinocytes similar in size to granular cells
occupied the epidermis (43 mm). Vague out-
lines of dermal papillae appeared (64.5 mm).
(3) Four weeks: Outlines of the dermal papillae
were refined (43 mm).
(4) Eight weeks: The dermal papillae recovered to
initial levels (21.5 mm).
It was inferred that the observed chronological
change of dermal papillae may relate to wound
healing. To confirm this hypothesis, we counted
the number of dermal papillae (Fig. 3a) and then
measured the thickness between the skin surface
and the apex of dermal papillae (Fig. 3b). Results
were expressed as the mean value of each week
compared with the mean reference value of 100%
at week 0 (Fig. 3). The dermal papilla count
sharply decreased approximately 80% by week
1 and recovered to its initial level by week 8.
The thickness doubled by week 1, and gradually
recovered to its initial level by week 8.
437
Sugata et al.
In two out of six subjects, fibrous tissues were
observed at week 8 between the depths of 64.5
and 129 mm. These tissues were not determined
before the suction blister inducement (Fig. 4) and
were found to be irreversible. In those subjects
Fig. 3. Chronological changes in the number of dermal papillae: (a)
chronological changes in the thickness between the skin surface and the
apex of dermal papilla; (b) values were determined from the mean of
six subjects but at weeks 1 and 2, four subjects and one subject,
respectively, were excluded due to the difficulty in obtaining an image
through crusta.
Fig. 4. Confocal images of dermis taken at depths of 64.5 mm: (a) 86 mm (b) and 129 mm (c) before and 8 weeks after removal of the epidermis. *Dermal
papillae. Image size: 400  540 mm.
438
with the fibrous tissues, the dermal papilla count
was apparently smaller than the initial number.
In one other subject, many bright dendritic
cells were observed around the epidermisder-
mis border at week 2 (Fig. 5a). Hyperpigmenta-
tions were also seen in the surface image of this
subject (Fig. 5b).
Discussion
Wound healing is known to involve interactions
between the epidermis and the dermis. These
interactions are well organized, as if these layers
are communicating each other (5). In this study,
although the epidermis was carefully removed
from the dermis without damaging it, a promi-
nent change was observed in this undamaged
layer. The dermal papillae disappeared after in-
duction of the suction blister wound, followed
by a slow recovery period of more than 6 weeks.
The changes in the dermal layer seem to be an
event associated with, or contributing to the
recovery of the upper epidermal layer.
Previously, Suetake et al. detected a parameter
for evaluating the proliferative changes of the
dermis in scars. They confirmed that transepider-
mal water loss in the stratum corneum reflects the
abnormal conditions of the dermis in various
scars (6). In our study, fibrous tissues were
observed in two subjects. These tissues were
irreversible for the duration of this study, and
looked similar to a mild level scar. Interestingly,
the dermal papilla count was apparently smaller
in areas with fibrous tissues. Our result suggests
that the dermal papilla count may likewise reflect

Fig. 5. (a) A confocal image that shows dendritic cells. Image size:
400  540 mm. (b) A surface image taken from the same wound.
Several hyperpigmentations were seen (arrows).
the abnormal conditions of the dermis during
wound healing.
It is widely known that the incidence of keloid
or hypertrophic scars is higher in highly pigmen-
ted skin. This information led us to assume
melanocytes relate to scar formation. In this
study, melanocytes were removed from the ob-
served area together with the epidermis. At week
2, we found bright dendritic cells in images taken
at the epidermisdermis border in one subject.
Yamashita et al. (7) had previously determined
that melanocytes appear bright in a LCSM image,
and thus these bright cells are assumed to be
melanocytes that might have migrated from the
peri-wound margin or from hair follicles. LCSM
cannot however, fully differentiate melanocytes
from other cells; another image taken from the
surface of this wound shows hyperpigmentations
supporting our assumption.
In summary, LCSM enabled us to non-inva-
sively observe changes, presumably controlled
by interactions between the epidermis and
Subepidermal wound healing process
dermis during the wound healing. We could not
confirm the migration of melanocytes and its
contribution to scar formation, and thus further
investigation using an optical device with higher
resolution will be required. In vivo two-photon
excited fluorescence microscopy that permits the
identification of substances seems to be the most
promising tool for this investigation (8, 9). At any
rate, an optical device has the potential to the
primary tool in the study of dermal disorders,
especially in the absence of suitable animal models.
References
1. Rajadhyaksha M, Grossman M, Esterowitz D, Webb RH,
Anderson RR. In vivo confocal scanning laser micro-
scopy of human skin: melanin provides strong contrast.
J Invest Dermatol 1995; 104: 946952.
2. Huzaira M, Rius F, Rajadhyaksha M, Anderson RR,
Gonzalez S. Topographic variations in normal skin, as
viewed by in vivo reflectance confocal microscopy.
J Invest Dermatol 2001; 116: 846852.
3. Kiistala U, Mustakallio KK. In vivo separation of epider-
mis by production of suction blisters. Lancet 1964; 1:
14441445.
4. Kiistala U. Suction blister device for separation of viable
epidermis from dermis. J Invest Dermatol 1968; 50: 129
137.
5. Werner S, Krieg T, Smola H. Keratinocytefibroblast
interactions in wound healing. J Invest Dermatol 2007;
127: 9981008.
6. Suetake T, Sasai S, Zhen YX, Ohi T, Tagami H. Functional
analyses of the stratum corneum in scars. Sequential
studies after injury and comparison among keloids,
hypertrophic scars, and atrophic scars. Arch Dermatol
1996; 132: 14531458.
7. Yamashita T, Kuwahara T, Gonzalez S, Takahashi M.
Non-invasive visualization of melanin and melanocytes
by reflectance-mode confocal microscopy. J Invest Der-
matol 2005; 124: 235240.
8. Konig K, Riemann I. High-resolution multiphoton to-
mography of human skin with subcellular spatial reso-
lution and picosecond time resolution. J Biomed Opt
2003; 8: 432439.
9. Yeh AT, Kao B, Jung WG, Chen Z, Nelson JS, Tromberg
BJ. Imaging wound healing using optical coherence
tomography and multiphoton microscopy in an in vitro
skin-equivalent tissue model. J Biomed Opt 2004; 9: 248
253.
Address:
Keiichi Sugata
Biological Science Laboratories
Kao Corporation
2606, Akabane
Ichikai-machi
Haga-gun
Tochigi 321-3497
Japan
Tel: 181 285 687491
Fax: 181 285 687360
e-mail: sugata.keiichi@kao.co.jp
439