Lasers in Surgery and Medicine 39:403–413 (2007)

Skin Changes Following Minor Trauma

Lise Lyngsnes Randeberg, PhD,1* Andreas M. Winnem, MSc,1 Neil E. Langlois, MD,6
Eivind L.P. Larsen, MSc,1 Rune Haaverstad, MD, PhD,4 Bjørn Skallerud, PhD,5
Olav A. Haugen, MD, PhD,2,3 and Lars O. Svaasand, PhD1
1
Department of Electronics and Telecommunications, Norwegian University of Science and Technology,
N-7491 Trondheim, Norway
2
Department of Laboratory Medicine, Children’s and Women‘s Health, Norwegian University of Science and
Technology, N-7006 Trondheim, Norway
3
Department of Pathology, Trondheim University Hospital, N-7006 Trondheim, Norway
4
Department of Cardiothoracic Surgery, Trondheim University Hospital, N-7018 Trondheim, Norway
5
Department of Structural Engineering, Norwegian University of Science and Technology, N-7491 Trondheim, Norway
6
Westmead Department of Forensic Medicine, Westmead Hospital, Wentworthville, NSW 2145, Australia

Background and Objective: Bruises are currently INTRODUCTION

evaluated by visual inspection, and little is known about Minor trauma like bruises can be important evidence in
the first phase after injury. The temporal development of cases of domestic violence or child abuse. So far, few
fresh injuries must be accurately described to be able to age quantitative descriptions of such minor injuries have been
bruises in a reliable manner. Color changes in a bruise published. Most of the current literature is considering
caused by hemoglobin breakdown products will depend on patterns of bruising or aging of bruises using visual
the severity of the trauma, and thus on the local immune observations, see for example [1–5]. The methods based
response in the skin. It is therefore important to relate the on visual inspection have been criticized for their subjective
nature of the impact to the temporal tissue responses. manner and low reliability [5,6]. Maguire et al. [5], stated
Materials and Methods: Controlled injuries were that there is a need for more objective methods like
inflicted on anesthetized domestic pigs. Trauma was spectroscopy to age bruises in a reliable manner.
induced either by a pendulum device, or by paintballs A correct description of the temporal development of an
released using pressurized air. The speed of the projectiles injury is essential to be able to age bruises. A bruise is a non-
was recorded using a high speed camera. Biopsies and penetrating injury, characterized by a hematoma caused by
reflection spectra (400–850 nm) were collected from normal blunt force to the skin. The hematoma is a result of bleeding
and bruised skin. The experiments were approved by the from smaller vessels that have been ruptured by the
national animal research authority. impact. The vessels may continue to bleed some time after
Results: The temporal development of the injury was the impact, and the bruise develops during the first 24–
found to depend strongly on the weight and speed of the 48 hours [1]. The biochemistry and immediate tissue
object. Low speed, blunt objects did not cause persistent responses to mechanical irritation are nicely described by,
skin changes. However, deep muscular bleeding could be for example Bauer et al. [7]. Mechanical irritation causes
found in most cases. High speed, light weight objects caused pain receptor stimulation and compression of mastocytes in
a rapidly developing bruise. These bruises were fully the dermis. This compression leads to histamine release in
developed within 15–20 minutes. No deep muscular the affected region. Nervous signals induced by pain
hemorrhages were observed in those cases. White blood receptors cause release of neuropeptides in a larger area
cells (neutrophilic granulocytes) could be found in biopsies surrounding the stimulated site. Histamine and neourpep-
from high speed injuries. The amount of white blood cells tides cause vasodilatation by decontraction of smooth
depended on the time between injury and collection of the muscle. The effect of histamine is instantaneous while the
biopsies. neoropeptide effect occurs after 20–30 seconds. Szczesny
Conclusion: Further investigations utilizing a larger et al. [8] showed that mechanical trauma caused extra-
range of object weight and velocities are required to be vasation, leakage of macromolecules, increased percentage
able to fully classify minor traumatic injuries. Preliminary of leucocytes and increased lymph flow. This is supported
results indicate that this can be achieved by controlled
experiments using a porcine model. Reflectance spectro-
scopy was found to be a useful tool to study immediate skin *Correspondence to: Lise Lyngsnes Randeberg, PhD, Depart-
ment of Electronics and Telecommunications, Norwegian
reactions to the trauma. Lasers Surg. Med. 39:403–413, University of Science and Technology, O.S. Bragstads Plass 2A,
2007. ß 2007 Wiley-Liss, Inc. N-7491 Trondheim, Norway. E-mail: Lise.Randeberg@iet.ntnu.no
Accepted 12 February 2007
Published online 23 May 2007 in Wiley InterScience
Key words: blunt trauma; diffuse reflectance spectro- (www.interscience.wiley.com).
scopy; optical diagnosis DOI 10.1002/lsm.20494

ß 2007 Wiley-Liss, Inc.

404 RANDEBERG ET AL.

by the observation that neutrophils and macrophages are Biopsies were important to be able to identify possible
recruited from the vascular system as a part of the connections between the spatial distribution of the hemor-
inflammatory reaction initiated by the trauma [9,10]. rhage and the impact.
Macrophages and neutrophils engulf both erythrocytes
and free hemoglobin molecules and initiate the heme MATERIALS AND METHODS
oxygenase system to break down the hemoglobin and
produce bilirubin and hemosiderin [11]. Presence of Animal Model
bilirubin will cause bruises to appear yellow [1]. The The experiments were carried out at the animal research
temporal development of a bruise is thus dependent on facilities at Trondheim University Hospital. Four female
the presence of macrophages and neutrophils in the injured Norwegian domestic pigs (25–40 kg) were intubated and
zone. anesthetized. Premedication: Stresnil (15 mg/kg), Stesolid
As sited above, several authors have described biochem- (0.375 mg/kg); induction of general anesthesia: Ketalar
ical changes in the skin following minor trauma. However, (12.5 mg/kg), Thiopenthon (5 mg/kg). The anesthesia was
the findings have not been related to the macroscopic skin maintained by administering Fentanyl (7.5 mg/kg/h), Mid-
and tissue changes that can be observed following mechan- azolam (0.5 mg/kg/h), and Isoflurane (gas 0.5
1%). Follow-
ical irritation or trauma. The visual appearance of a bruise ing the experiments the animals were sacrificed by giving
will be dependent on the time after impact, and optical an overdose pentobarbital. No anticoagulants were admi-
methods can be used to monitor and characterize the nistered. A constant skin temperature (328C) was
observed changes. Reflection spectroscopy has been applied achieved using heating blankets to avoid temperature loss.
to characterize bruises with good results [12–15]. Bohnert Three of the pigs had no visible skin pigmentation, while
et al. [12] published a relation between reflectance spectra, one pig was lightly pigmented. The injury sites were
observed bruise color and depth of injury in post mortem carefully shaved using a haircutter prior to measurements.
cases. They found that superficial hemorrhages cause No skin erythema was observed following shaving. The
bright red bruises, while deep subcutaneous bruises appear animal heart rate and blood pressure was monitored
blue or purple. Hughes et al. [13] used reflectance spectro- throughout the experiments. The protocol was approved
scopy to quantify bilirubin in bruises of various age, and by the national animal research authority.
Randeberg et al. [14] identified hemoglobin content in the
skin, hemoglobin oxygenation, and bilirubin content as Trauma Induction
potential indicators of bruise age. Randeberg et al. [15] Trauma was introduced at the shoulder and hip regions
later published a mathematical model to predict the age of a at both sides of the animals. A high speed camera (Phantom
bruise from a reflectance measurement. This method is v.5, Vision Research Inc, Wayne, New Jersey), was used to
based on the hypothesis that most bruises occur in the monitor the impacts and measure the speed of the objects.
junction between dermis and the subcutaneous fat layer. Trauma was introduced using two basic methods:
The model predicts the time from trauma to visible bruise, The first method was intended to give an injury
depending on the skin thickness and the hemoglobin comparable with a punch from a boxer, but less violent.
diffusion speed within dermis. The model allows individual The aim in this study was to create hemorrhage in the skin,
correction for skin properties like pigmentation and and underlying tissue, not to hurt the animals severely.
hemoglobin absorption in normal skin. This model has so Professional boxers have been found to hit their target with
far only been tested on limited, homogenous patient groups a fist speed of about 10 m/second, and an effective arm mass
[15]. To improve this model further it is important to study of about 2–3 kg. The average duration of a typical punch
how hemorrhages develop in an early phase, and to identify has been found to be approximately 10–15 milliseconds,
relations between the extent of trauma and how the trauma and the force onto the tissue was about 2500–3500 N [18].
is inflicted. The relation between the extent of trauma and A rig with a pendulum was constructed to create such a
the resulting injury have been thoroughly described for blunt force injury from a slowly moving object. The design of
major and often fatal trauma like head injuries or trauma to the pendulum device allowed variation of speed, weight,
inner organs caused by, for example, car crashes. Non- and shape of the object. The pendulum was released from
penetrating minor trauma are only described by a few different angles with respect to the skin surface to allow
authors, see for example [7,8,16,17]. Roberts et al. [16] different velocities. Positions where the arm (and attached
presented a finite element model to evaluate the effect of weight) had angles of 45, 60, and 90 degrees with respect to
non-penetrating ballistic impacts. However, this model the skin surface were tested. The length of the arm was
only considers the effects on inner organs and not skin approximately 0.9 m. Six different blunt objects were used
bruises. to induce injuries. The presented results were achieved
The current study was initiated to investigate immediate using a 2.5 kg weight attached to the pendulum. Two
skin reactions following minor trauma and provides differently shaped weights were used; a rounded tip and a
spectroscopic data from fresh injuries. It was important to square shape. The pendulum weight is shown in Figure 1.
identify parameters that could be used to describe the The acceleration and retardation of the hitting object was
relation between the impact and the degree of damage in monitored by accelerometers (B12/500, HBM, Darmstadt,
later studies, and to correlate reflectance measurements Germany) mounted on the weight. A typical maximal
with mechanical parameters such as velocity and mass. retardation was measured to be 80–100 m/second2, and the
 MINOR SKIN TRAUMA 405

energy per area was also comparable for the two methods,
with Ek/A ¼ 4.5104 J/m2 for the pendulum and Ek/
A ¼ 6.1104 J/m2 for the paintballs.
Ultrasound
The blunt trauma sites on the two first pigs were scanned
with an ultrasound probe (12 MHz) prior to and after
induction of the trauma. (GE Vingmed Vivid VII scanner,
GE Vingmed, Horten, Norway).

Biopsies and Tissue Examination

Punch biopsies, using sterile 5 mm needles (Biopsy
Punch, Stiefel1, Stiefel Laboratorium, Offenbach am
Main, Germany) were taken from normal skin and from
injured sites. The biopsies were fixed in buffered neutral 4%
formaldehyde solution and processed further for paraffin
embedding. From all specimens sections 5 mm slices were
cut and stained with haematoxylin-erythrosin-saffron.
Fig. 1. Device for blunt trauma induction. The front end of the The skin was removed after the animal was sacrificed,
pendulum had two interchangeable mounts; a rounded and a and examined for subcutaneous and muscular hemorrhage
square with rounded edges (shown in the photo). The weight caused by the trauma. The skin thickness was recorded
was mounted on a pendulum attached to a steel rig. One of the directly using a caliper. Skin thickness could also be
accelerometers can be seen on top of the weight. [Figure can be extracted from the ultrasound images, and to some degree
viewed in color online via www.interscience.wiley.com.] from the biopsies. The average total skin thickness
measured was approximately 2 mm at the injured sites.
impact lasted for approximately 50 milliseconds. The
highest speed of the object was measured to be 4.1 m/ Spectroscopy and Photography
second at impact.
Reflection spectra were collected in the 400–850 nm
The second method was chosen to give an injury
wavelength range using an integrating sphere setup (ISP-
comparable to a hard blow from a cane or a whip. Whips
REF, SD2000, Ocean Optics, Duiven, The Netherlands).
and canes can create severe skin damage. The tip of the
The integrating sphere was modified to a 20 mm aperture to
whip can reach very high, even supersonic velocities
avoid loss of (red) light backscattered from deeper skin
[19,20]. Such injuries are caused by high speed, low weight
layers. Measurements were collected prior to trauma
objects, and was thought to be comparable to a skin injury
induction, immediately after and then repeated at various
caused by a paintball hitting the skin at high velocity. In
times after injury. Spectra were collected from the central
this case the energy is deposited within a short time, and on
region of the injury and from the edge. The last measure-
a limited area, and the damage mechanisms are thought to
ments were collected 2–5 hours after injury. Each spec-
be comparable to a kick or blow from an athlete doing
trum was collected as the average of 10 subsequent scans of
martial arts [21–23]. In karate the speed of the arm or foot
the same area, and each measurement was repeated three
will be approximately 10–14 m/second at the impact, and
times. The integrating sphere was lifted off the skin
the hitting area will often be limited to a part of the hand.
between the three subsequent measurements. Typical
The impacts have been found to last approximately 5–
standard deviations for such measurements will be
20 milliseconds, with a total impact force as large as more
between 0.01 and 0.001 as a function of wavelength [24].
than 3,200 N [21–23]. The high speed injuries in this study
The integrating sphere was held gently against the skin
were created using 3.15 g paint filled, marble sized gelatin
during measurement to avoid blanching due to pressure
paintballs (d ¼ 1.71 cm). The skin was covered with a thin
from the sphere. Digital photos and video recordings were
plastic film to avoid spilling paint directly onto the skin.
also used to document the hemorrhage dynamics.
Paintballs were chosen due to the reproducibility and small
area of the injury. The localized trauma could easily be
monitored by spectrometry. The projectiles were released Spectral Data Analysis
by pressurized air at a distance of approximately 0.5 m from Hemoglobin and melanin are the most important
the skin surface. The exit speed of the projectile was chromophores in normal skin. The visual appearance of a
measured to be 104 m/second at 1 m distance. The impact fresh bruise is mainly due to hemorrhage from ruptured
lasted for approximately 0.3–0.5 milliseconds. vessels within the skin, and hemoglobin is therefore the
In this case the force applied to the skin was comparable predominant absorber. Bilirubin, methemoglobin, hemosi-
for the two methods. The average applied force was derin, and other hemoglobin breakdown products might
approximately 500 N for the pendulum and 600 N for the be found in older bruises [1–4,13–15]. However, only fresh
paintballs (for a 0.5 second impact). This was found by bruises were included in this study, and no traces of
dividing momentum at impact by impact duration. Kinetic methemoglobin, hemosiderin or bilirubin could be identified
406 RANDEBERG ET AL.

in any of the spectra. These chromophores were thus not injuries created with a rounded tip. Oxygenation, dermal
considered as problems in the spectral data analysis. blood volume fraction, and erythema index showed only
Hemoglobin oxygenation was calculated from the reflec- small changes with time after injury, see Fig. 2, panels (a, c,
tance spectra using Eq. 1 [15,25,26]. e), Inj1–Inj4. No differences could be observed between the
Rðl2 Þ
spectra measured in the central region of the injury and
mHb ðl1 Þ
mHbO2 ðl2 Þ  Rðl1Þ the measurements collected on the edge of the injury.
OS ¼ ð1Þ
mHb ðl1 Þ
mHbO2 ðl1 Þ Therefore only data from the central region of the injury is
presented for Inj1–Inj4. A time series of reflection spectra
where OS is the oxygen saturation, R is the measured in the wavelength range 400–850 nm from Inj4 is shown in
absorption coefficient, and mHb and mHbO2 are the specific Figure 3a. The spectra show decreased reflectance, espe-
absorption coefficients of deoxygenated and oxygenated cially for wavelengths shorter than 600 nm, indicating an
blood, respectively [27,28]. The wavelengths l1 and l2 are increased blood volume fraction in the most superficial skin
a non-isosbestic and an isosbestic point, respectively. In layers. The oxygenation was increased with respect to the
this case l1 ¼ 660 nm and l2 ¼ 805 nm was used. The spectra collected prior to the trauma. This can be seen from
calculated oxygenation will depend on the local penetra- the increased oxyhemoglobin absorption at 542 nm and
tion depth in the skin. The wavelengths used here were 576 nm. The skin reaction seemed to be fully developed
chosen to allow as deep probing into the tissue as possible. within 4 minutes after injury, see Figure 4. The skin
The penetration depth in human skin will be approxi- biopsies revealed only minor injuries in the subcutaneous
mately 0.5–1 mm for the current wavelengths (using layer. An example of such an injury from pig 3 is shown in
absorption- and reduced scattering coefficients from Figure 5a. However, deep hemorrhages were found in
Tuchin [29]). It is thus important to realize that the muscle tissue underlying the impact site. The typical depth
calculated oxygenation value will only reflect an ‘‘average’’ of these injuries were 1.5–3 cm, see Figure 6. All blunt force
hemoglobin oxygenation in the skin down to approxi- injuries in all animals showed similar trends. The shape
mately two penetration depths. The contribution from and weight of the object did have impact on the observed
each depth will be proportional to the amount of reflected injury. Three different velocities were tested, using
light from each depth that reaches the detector. It is different pendulum elevations. However, no quantitative
possible to use several sets of wavelengths with different measurements were done to record the amount of muscular
penetration depths [26], to gather further information hemorrhage. Based on visual inspection square shapes
about the spatial distribution of the hemoglobin oxygena- (even with rounded edges) seemed to create more damage
tion. However, such calculations were not included in this than hemispherical shapes. The speed of the object was the
work. most important factor to predict if muscular bleeding would
The dermal blood volume fraction was determined from occur. Increased speed gave a higher risk of hemorrhage,
the reflectance spectra as previously described [15]. The even with the same weight and shape of the object. Sites
photon transport model used to derive the dermal blood with a thin muscular layer (3 cm or less) on top of
volume fraction was modified to fit the skin of a young pig underlying bone seemed more vulnerable than sites with a
(3–4 months) by adjusting the scattering coefficient thick muscular layer.
(ms ¼ 1.089104 @ 600 nm). Erythema and melanin indices The high speed injuries could be clearly separated from
were determined from the spectra as described in [30]. the blunt force injuries. These injuries developed fully
within 15 minutes after impact. The reaction to the trauma
RESULTS caused a central whitening of the skin with a surrounding
Data from pig 4 was chosen for presentation, except from erythema, see Figure 7a. A hemorrhage in the upper skin
the blunt trauma biopsy results which were taken from layers became visible within 1–2 minutes after injury.
pig 3. Four blunt trauma injuries were induced on the These reactions were followed by red, blue and purple rings
left side of pig 4 using a 2.5 kg weight, square tip with forming simultaneously with different radii around the
rounded edges (Inj2, Inj4) and hemispherical tip (Inj1, central zone (where the projectile hit the skin), see Figure
Inj3). Five high speed injuries were inflicted at the right 7b,c. These injuries caused only superficial damage. No
side of the animal (Inj5–Inj9). All presented results refer subcutaneous or deep muscle injuries could be found.
to the injury numbers. One of the data series (Inj8) from Microscopically, biopsies from injuried sites (at the red
the high speed injuries contained only two measure- edge) showed extensive subepidermal vascular congestion
ments, and was therefore omitted from the analysis. All and extravasation of red blood cells, see Figure 5b. The
presented injuries were induced using an angle of deepest focus showing hemorrhage was close to 2 mm. Inj5
90 degrees between the skin surface and the pendulum, was the first high speed injury inflicted and the biopsies
unless explicitly stated. were collected approximately 4.5 hours after impact. In the
Blunt force injuries did not create any visible damage to subcutis of this injury the vessels showed large numbers of
the skin, except from an immediate wheal and flare white blood cells (neutrophilic granulocytes), many of
reaction. This reaction was established within 20– which also were found around capillaries and in the fatty
30 seconds after injury, and disappeared within minutes. tissue. Similar changes were absent from the central
However, the skin reaction lasted a minute or two longer for region, and far less striking in Inj7 which was inflicted
the injuries inflicted with the square edges, than for the 2 hours prior to collecting the biopsies. The epidermis also
 MINOR SKIN TRAUMA 407

a b
Inj5
0.88 0.88 Inj6
Inj7
0.86 0.86 Inj9

0.84 0.84

0.82 0.82

oxy
oxy

0.8 0.8

0.78 0.78
Inj1
Inj2
0.76 0.76
Inj3
Inj4
0.74 Inj5 0.74
Inj6
0.72 Inj7 0.72
Inj9
0.7 0.7
0 10 20 30 40 50 60 0 10 20 30 40 50 60
Time (minutes) Time (minutes)

c d
0.045 0.045
Inj1 Inj5
Inj2 Inj6
0.04 Inj3 0.04 Inj7
Inj4 Inj9
0.035 Inj5
0.035
Inj6
Inj7
0.03 Inj9 0.03
Bd

Bd

0.025 0.025

0.02 0.02

0.015 0.015

0.01 0.01

0.005 0.005
0 10 20 30 40 50 60 0 10 20 30 40 50 60
Time (minutes) Time (minutes)
e f
120 120
Inj1 Inj5
Inj2 Inj6
110 Inj3 110 Inj7
Inj4 Inj9
Inj5
100 Inj6 100
Inj7
Inj9
90 90
Ec

Ec

80 80

70 70

60 60

50 50
0 10 20 30 40 50 60 0 10 20 30 40 50 60
Time (minutes) Time (minutes)

Fig. 2. Temporal development of skin parameters derived from reflectance spectra. Inj8 was
omitted due to the low number of data points. The time zero is the time of injury. The value in
this point was calculated from spectra measured at the same site prior to injury. Panels (a–b):
Oxygenation derived from the reflectance values at 660 nm and 805 nm; Panels (c–d): Dermal
blood volume fraction, Bd. Panels (e–f): Erythemaindex, Ec. Data in panel (a,c,e) were
measured at the center of the low speed injury, while the data plotted in panels (b,d,f) were
measured at the center of the red edge of the high speed injury. [Figure can be viewed in color
online via www.interscience.wiley.com.]
408 RANDEBERG ET AL.

a b
60 60

50 50
Reflectance (%)

Reflectance (%)
40
40
30
30 1 min
20 2 min
0 min 17 min
20 4 min
10 65 min
5h
Normal
Normal
10 0
400 500 600 700 800 400 500 600 700 800
Wavelength (nm) Wavelength (nm)

Fig. 3. Reflectance spectra as a function of time after injury. Panel (a): Spectra measured on a
blunt trauma site. The data corresponds to Inj4 in Figure 2; Panel (b): Spectra measured
centrally in an injury caused by a high speed projectile. The spectra correspond to Inj5 in
Figure 2.

showed focal abrasion and partly detachment from the showed different trends for the two regions. In the central
basal membrane. No muscular hemorrhages were found region the oxygenation showed a fast drop after injury,
when the skin was opened after sacrificing the animal, see followed by a slow recovery toward normal values after 5–
Figure 8. 12 minutes. The oxygenation at the edge dropped to lower
All measured skin parameters reflected the visually values, and started to recover against normal values after
observed reactions, see Figure 2, Inj5–Inj7 and Inj9. approximately 15–20 minutes. The dermal blood volume
Hemoglobin oxygenation, erythema index, and dermal fraction showed similar behavior in the two regions.
blood volume fraction show fast changes after the impact, Normal values prior to injury ranged from 0.5% to 1.0%.
followed by a slow recovery toward normal values. After injury the blood value fraction increased up to about
Differences could be observed between the center and the 4%. The increase occurred within the 20 first minutes after
red edge in the high speed injuries. Fig. 2, panels (a,c,e) injury in both cases. However, the recovery toward normal
show data from the center of the injury while panel (b,d,f) values was faster at the red edge. The erythema index
show data from the red edge. Especially the oxygenation showed a rapid increase in the center of the injury, and a
somewhat slower increase at the red edge. A maximum
value could be observed after approximately 10 minutes at
the edge, and after 20 minutes in the central zone. Series 9
was measured on the abdomen of the pig. The measurement
site had underlying soft tissue, while all other measure-
ments were performed on sites with a muscular layer on top
of bone. Series 9 compare with the low speed injuries with
respect to oxygenation both in the center and on the edge of
the injury, and blood volume fraction and erythema index
measured on the red edge. In all other cases the behavior
was similar to the high speed injuries.
Reflection spectra from injury 5 were chosen to visualize
how the skin changes can be seen directly from spectra
(Fig. 3b). The measured reflectance first shows an increase
with respect to normal skin due to the whitening reaction,
this is followed by a decrease in reflectance within 2
minutes after injury. The first spectra indicate an increase
Fig. 4. Blunt force injury site after trauma. Injury 3 (S3) was in deoxyhemoglobin, which can be seen from the prominent
induced 1 hour and 19 minutes before the photo was taken. deoxyhemoglobin absorption at 760 nm. The dermal blood
Injury 4 (S4) was induced 9 minutes before the photo was volume fraction is increased at all measurements after the
taken. The injury was induced by the object shown in Figure 1. first minute after injury. The oxygenation increases with
[Figure can be viewed in color online via www.interscience. time after injury, and the measurement performed 1 hour
wiley.com.] post injury shows the highest oxygenation in this series.
 MINOR SKIN TRAUMA 409

Fig. 6. Same site as in Figure 4, approximately 5 hours after

injury. Deep muscular hemorrhage was observed down to a
depth of approximately 3 cm into the tissue. [Figure can be
viewed in color online via www.interscience.wiley.com.]

DISCUSSION
Due to the nature of the experiments it was difficult to
get data from human subjects even for minor injuries. It is a
problem to get controlled data from human volunteers
who have been bruised accidentally, and it would be
unethical to create controlled bruises in non-anestethized
humans. Fortunately, the skin of young pigs is quite similar
to human skin with respect to pigmentation, hair growth,
and thickness, and is a frequently used animal model in
dermatology. A porcine model was chosen in this study
Fig. 5. Histology of biopsies. Panel (a) Sample from a blunt to allow controlled experiments and to gain knowledge
trauma injury created using a pendulum with a square front about the trauma mechanisms. Further work will include
end. A hardly visible extravascular red blood corpuscle can be studies on bruising in paintball players who have been
seen in the subcutis at a depth of approximately 0.4 mm; Panel accidentally hit.
(b) Sample from a high speed injury showing dilated capillaries The trauma induction techniques used in this study were
with red blood cells and interstitial bleeding. The sample was chosen to be comparable to, but less violent than injuries
taken from Injury 5, at the red edge, see Figure 7c. The from a knock using the fist, a blow from an athlete doing
included scale is 1 mm. [Figure can be viewed in color online via martial arts or a blow from a cane or whip. Kicks in martial
www.interscience.wiley.com.] arts are designed to break bone, and in this study it was
desirable to create only minor injuries. The strokes in for
example karate can be characterized by a fast forward
The trends described here could be seen in all high speed movement of the arm followed by a rapid pullback from the
injuries. target, while a punch from a boxer is characterized by low
The high speed camera shoved massive wave motions in speed and a prolonged impact where the fist of the boxer
the skin following the impact for both high- and low speed keeps in contact with the target for a longer time [18–23]. In
injuries. However, the pendulum was in contact with the this study the impact was kept clearly below values given in
skin for a longer period of time, and thereby preventing the the literature to avoid unnecessary injury to the animals.
tissue oscillations to develop freely. In the high speed The images from the high speed camera verified that the
injuries the paintballs were crushed against the skin, and pendulum stayed attached to the skin for a longer period of
the impact created intense oscillations in the skin time like the fist of a boxer, while the paintball hit a small
surrounding the impact site. area and was crushed against the skin. The crushing of the
Ultrasound images revealed little information on mus- paintball allowed massive oscillations to occur in the tissue.
cular hemorrhages. Doppler ultrasound showed somewhat Oscillations in the target has been suggested as the damage
increased blood flow after impact, but no reservoirs of blood mechanism in karate blows, where the impact zone is small
could be observed in the muscular tissue. and the time of impact is short [21–23].
410 RANDEBERG ET AL.

Fig. 7. Temporal development of a representative high speed injury (Inj5). The photos were
taken after: (a): 30–60 seconds after inflicting the injury; (b): 5 minutes after injury; (c): 2
hours and 5 minutes after injury. Reflectance data, oxygenation, dermal blood volume, and
erythema index is given in Figures 2 and 3b. [Figure can be viewed in color online via
www.interscience.wiley.com.]
 MINOR SKIN TRAUMA 411

indicating a weakly increased blood volume in the most

superficial skin layers. The penetration depth for green
light can be estimated to be maximum 0.3–0.5 mm in skin.
If the overall dermal blood volume increased, the reflec-
tance would have shown a decrease for wavelengths longer
than 600 nm as well, due to the longer penetration depth for
red light. The oxygenation will depend on the probing depth
of the light, which is dependent on the optical penetration
depth. The only clear indication of injury in these sites were
deep muscular hemorrhages, see Figures 4, 5a and 6. The
hemorrhages were more prominent in sites with a thin
muscular layer, where the vasculature was crushed more
severely against the bone underneath (Inj3 and Inj4). For
constant speed and weight, a square shape (Inj1 and Inj3)
gave larger muscular hemorrhages than a hemispherical
shape (Inj2 and Inj4). However, the visual inspection of the
surface or the reflectance spectroscopy did not reveal
Fig. 8. High speed injury. The skin was removed to check for
significant differences. Deep hemorrhages could not be
muscular hemorrhage. No such hemorrhages were revealed for
predicted by spectroscopy in such a short time span
the injuries caused by light objects. The injury can be seen as a
(hours). The correlation between the shape, weight and
dark shadow on the skin flap. The spectral data from this site
speed of an object and the resulting injury will be studied
were omitted from the figures due to a low number of
further by finite element modeling of the impacts.
measurements.
The pigs were under general anesthesia during the
The transmitted power and the kinetic energy per area experiments, which might explain the missing skin
were the same order of magnitude for the two trauma response. The current experiments were comparable to
induction methods. Both were clearly below the values hitting a sleeping or unconscious person with low muscle
reported in the literature [18–23]. The small mass of the tension. Due to the deep anesthesia, the animals had a
paintballs cause the energy to be dissipated near the skin reduced heart rate and thereby a lowered perfusion rate in
surface and thus create superficial damage (and superficial the skin. Some of the more superficial muscular injuries
oscillations). Due to the large mass, the pendulum caused would probably have developed into visible bruises if the
energy dissipation and tissue damage at larger depths into animals had been kept alive for some days after injury. This
the tissue. These injuries were more like crushing injuries was not possible with the current protocol due to animal
in the deeper muscular layers. The degree of trauma is care regulations. Up to date, few clinical observations
dependent on the velocity and the mass of the hitting object. describing the first phase after injury have been published.
Further investigations must be undertaken to be able to Patients are seldom examined by a physician within the
predict the source of damage based on clinical findings. The first minutes after injury, therefore information describing
objective of this study was to study the trauma developed this first phase is valuable.
after to very different trauma induction methods, and to The high speed injuries created clearly visible bruises
describe immediate skin reactions in an objective manner. within minutes. The first reaction to a high speed trauma
In clinical cases there might be any combination of the was a wheal and flare reaction that developed within 20–
above mentioned injury mechanisms, based on the velocity 30 seconds after injury. A white zone could be seen in the
and mass of the hitting object. The minimum velocity to central part of the injury surrounded by erythema, see
induce skin damage for various objects will be investigated Figure 7a. The blanching caused the measured skin
in further studies, where a series of trauma is induced to reflectance to increase, see Figure 3b, probably due to
identify the minimal values necessary to create hemor- edema formation causing increased scattering and blocking
rhage for various tissues. However, the current results can of the capillary bed due to increased tissue pressure. Some
be used to derive some quantitative information about the of the reaction might also be caused by the impact itself
hitting object in clinical and forensic evaluation of non- blocking off vessels in the underlying tissue. Findings in
penetrating injuries. this work are consistent with the description of tissue
Blunt force injuries introduced by slowly moving objects response given by Bauer et al. [7]. The biopsies from the red
were not found to create any visible skin effects that lasted edge (seen in Fig. 7c) showed extensive subepidermal
more than a few minutes. Only minor damage could be seen vascular congestion and extravasation of red blood cells,
in the biopsy material. The traumas were induced with down to approximately 2 mm depth. A large number of
maximum retardation of 10 G, which is clearly below white blood cells (neutrophilic granulocytes) could be found
the values given for major blunt trauma by Grande [17]. in the subcutis, around the capillaries and in the fatty
The spectroscopic analysis of the impact site showed only tissue of the injured sites approximately 4 hours after
weak variations in a short period of time after impact, see injury. The number of white blood cells increased with time
Figure 2, Inj1–Inj4. The spectra shown in Figure 3a, show after injury. Epidermal abrasion and partly detachment
decreased reflectance for wavelengths shorter than 600 nm, from the basal membrane was also found, see Figure 5b.
412 RANDEBERG ET AL.

These findings are supported by Szczesny et al. [8]. They possible. Thanks to Tore Wisth, Olav Haldorsen, and Johan
showed that mechanical trauma to the hind leg of J. Sandnes for helping out with the rig construction and the
mice caused extravasation, leakage of macromolecules, acceleration measurements. Thanks to Andreas Gurk for
increased percentage of leucocytes and increased lymph help with the high speed camera measurements. Thanks to
flow. These results are important to determine the onset of Hans Torp for lending us the ultra sound scanner and to
neutrophil and macrophage activity in the injury zone. Bjarne Bergheim and Lasse Løvstakken for helping us with
Such immune cells are responsible for local hemoglobin the ultrasound measurements. Thanks to Magnus B.
breakdown, and thereby bilirubin formation in the trau- Lilledahl for his assistance in the lab. Thanks to Rolf Tore
matized area. To be able to model and date bruises Randeberg for proofreading the manuscript.
correctly, it is important to know when the bilirubin
production is initiated. The time scale found in this study REFERENCES
is consistent with the findings by Thornton [10]. The 1. Langlois NEI, Gresham GA. The aging of bruises; a review
and study of the colour changes with time. Forensic Sci Int
current results will serve as a basis to improve the existing 1991;50:227–238.
model for temporal development of a bruise. 2. Stephenson T, Bialas Y. Estimation of the age of bruising.
The spectroscopic results show a large hemoglobin Arch Dis Child 1996;74:53–55.
deoxygenation, and an increase in dermal blood volume 3. Schwartz AJ, Ricci LR. How accurately can bruises be aged in
abused children? Literature review and synthesis. Pediatrics
fraction which reaches a maximum after approximately 1996;97:254–257.
10–15 minutes in the central zone of the bruise, see Figures 4. Altemeier WA. Interpreting bruises in children. Pediatr Ann
2 and 3b. The injury seems to be fully established within 2001;30:517–520.
5. Maguire S, Mann MK, Sibert J, Kemp A. Can you age bruises
approximately 15–20 minutes, see Figures 2, 3b and 7c. accurately in children? A systematic review. Arch Dis Child
The blood volume fraction and erythema index developed 2005;90:187–189.
differently on the red edge of the injuries. This is expected 6. Cabinum-Foeller E, Frasier L. Bruising in children. Lancet
due to the visually observed differences. However, it should 2005;365(9468) 1369–1370.
7. Bauer D, Grebe R, Ehrlacher A. A three-layer continous
be noted that the measurements performed on the red model of porous media to describe the first phase of skin
edge might include some reflected light from the normal irritation. J Theor Biol 2005;232:347–362.
skin adjacent to the injury, due to the aperture size of 8. Szczesny G, Veihelmann A, Nolte D, Messmer K. Changes in
the local blood and lymph microsirculation in response to
the integrating sphere. Injury 9 behaved different from the direct mechanical trauma applied to leg: In vivo study in an
other high speed injuries. This was probably due to the animal model. J Trauma 2001;51(3) 508–517.
more flexible underlying tissues on the abdomen of the 9. Hume DA, Ross IL, Himes R, Sasmono RT, Wells CA, Ravasi
T. The mononuclear phagocyte system revisited. J Leukoc
pig. No muscular hematomas were found in any of the high Biol 2002;72:621–627.
speed trauma sites, see Figure 8. 10. Thornton RN, Jolly RD. The objective interpretation of
Further studies will include keeping the animal alive for histopathological data: An application to the aging of ovine
some days after injury to monitor long-time effects. bruises. Forensic Sci Int 1986;31:225–239.
11. Gemsa D, Schmid R. Hämoglobinstoffwechsel und bilirubin-
Collection of several biopsies at various times after injury bildung. Klin Wschr 1974;52:609–616.
will make it possible to study the temporal development of 12. Bohnert M, Baumgartner R, Pollak S. Spectrophotometric
the injury and the inflammation reaction. Finite element evaluation of the colour of intra and subcutaneous bruises.
Int J Legal Med 2000;113:343–348.
modeling of the impact for various hitting objects will also 13. Hughes VK, Ellis PS, Burt T, Langlois NE. The practical
be conducted to describe the damage mechanisms. application of reflectance spectrophotometry for the demon-
stration of haemoglobin and its degradation in bruises. J Clin
CONCLUSION Pathol 2004;57:355–359.
14. Randeberg LL, Winnem AM, Blindheim S, Haugen OA,
It has been shown that there are large differences Svaasand LO. Optical classification of bruises. Proceedings of
between blunt trauma induced by a pendulum device and SPIE 5312:54–64.
15. Randeberg LL, Haugen OA, Haaverstad R, Svaasand LO. A
high speed injuries induced by a paint filled gelatin novel approach to age determination of traumatic injuries by
projectile. The blunt trauma resulted in deep muscular reflectance spectroscopy. Lasers Surg Med 2006;38(4) 277–
hemorrhage without visible damage to the skin, whereas 289.
16. Roberts JC, O’Connor JV, Ward EE. Modeling the effect of
the high speed trauma caused instantaneous visible injury non-penetrating ballistic impact as a means of detecting
to the skin. Reflection spectroscopy was found to be a behind-armor blunt trauma. J Trauma 2005;58:1241–1251.
valuable method to monitor skin reactions following non- 17. Grande CM. Mechanisms and patterns of injury: The key to
penetrating trauma. However, deep muscular hemor- anticipation in trauma management. Crit care clin 1990;6(1)
25–35.
rhages could not be detected at an early stage using this 18. Walilko TJ, Viano DC, Bir CA. Biomechanics of the head for
technique. Further studies are required to establish a Olympic boxer punches to the face. Br J Sports Med 2005;39:
general relationship between the degree of damage and the 710–719.
19. McMillen T, Goriely A. Whip waves. Physica D 2003;184:
impact. The current results indicate that a porcine model is 192–225.
suitable for such experiments. 20. Goriely A, McMillen T. Shape of a cracking whip. Phys Rev
Letters 2002;88(24) 244301.
21. Wilk SR, McNair RE, Feld MS. The physics of karate. Am J
ACKNOWLEDGMENTS Phys 1983;51(9) 783–790.
22. Walker JD. Karate strikes. Am J Phys 1975;43(10) 845–849.
Thanks to Kirsten Rønning for helping out with the 23. Blum H. Physics and the art of kicking and punching. Am J
animal work, and to Dr. Ronald Mårvik for making it Phys 1977;45(1) 61–64.
 MINOR SKIN TRAUMA 413

24. Randeberg LL. Diagnostic applications of diffuse reflectance Norwegian University of Science and Technology, Trond-
spectroscopy. Doctoral thesis, Norwegian University of Science heim, Norway, 1999. Fys. El. rapport 1999; 50.
and Technology, Trondheim, Norway. http://urn.ub.uu.se/ 28. Zijlstra WG, Buursma A, van Assendelft OW. Visible and
resolve?urn¼urn:nbn:no:ntnu:diva-691 (2006-10-18). near infrared absorption spectra of human and animal
25. Spott T, Svaasand LO, Anderson RE, Schmedling PF. haemoglobin determination and application, The Nether-
Application of optical diffusion theory to transcutaneous lands: VSP, Zeist; 2000.
bilirubinometry. SPIE Europto Series 1997;3195:234–245. 29. Tuchin V. Tissue Optics. Light scattering methods and
26. Randeberg LL, Roll EB, Norvang Nilsen LT, Christensen T, Instruments for medical diagnosis. Bellingham, USA: SPIE
Svaasand LO. In vivo spectroscopy of newborn skin reveals Press; 2000.
more than a bilirubin index. Acta Paediatrica 2005;94:65–71. 30. Dawson JB, Barker DJ, Ellis DJ, et al. A theoretical and
27. Spott T. Characterization of layered tissue structures with experimental study of light absorption and scattering by in
diffusely propagating photon density waves. Doctoral thesis, vivo skin. Phys Med Biol 1980;25:695–709.