Ex vivo penetration of low-level laser light through
equine skin and flexor tendons
Katja F. Duesterdieck-Zellmer dr med vet, phd
Maureen K. Larson ms
Thomas K. Plant phd
Andrea Sundholm-Tepper dvm
Mark E. Payton phd
Received September 15, 2015.
Accepted October 30, 2015.
From the Department of Clinical Sciences, College of
Veterinary Medicine (Duesterdieck-Zellmer, Larson,
Sundholm-Tepper), and the School of Electrical Engi-
neering and Computer Science, College of Engineering
(Plant), Oregon State University, Corvallis, OR 97331;
and the Department of Statistics, College of Arts and
Sciences, Oklahoma State University, Stillwater, OK
74078 (Payton). Dr. Sundholm-Tepper’s present ad-
dress is Department of Veterinary Clinical Sciences,
College of Veterinary Medicine, Washington State
University, Pullman, WA 99164.
Address correspondence to Dr. Duesterdieck-Zellmer
(katja.zellmer@oregonstate.edu).
T he effectiveness of LLLT for the treatment of mus-
culoskeletal conditions such as tendon injuries or
osteoarthritis is not universally accepted, as studies1,2
have failed to unequivocally show clinical benefit.
Reported positive effects include decreased inflam-
mation,3 decreased pain perception,4,5 improvement
of delayed wound healing,6 and improvement in
healing of injured deeper tissues such as tendons.2
The cellular and molecular mechanisms underlying
these effects have not been established. However, it
is thought that absorption of light energy by cellular
components triggers a chain of chemical reactions,
resulting in altered cellular metabolism—a process
termed photobiostimulation or photobiomodula-
tion.7,8 For example, LLLT increased ATP production
ABBREVIATION
LLLT Low-level laser therapy
AJVR • Vol 77 • No. 9 • September 2016
OBJECTIVE
To measure penetration efficiencies of low-level laser light energy through
equine skin and to determine the fraction of laser energy absorbed by
equine digital flexor tendons (superficial [SDFT] and deep [DDFT]).
SAMPLE
Samples of skin, SDFTs, and DDFTs from 1 metacarpal area of each of 19
equine cadavers.
PROCEDURES
A therapeutic laser with wavelength capabilities of 800 and 970 nm was
used. The percentage of energy penetration for each wavelength was de-
termined through skin before and after clipping and then shaving of hair,
through shaved skin over SDFTs, and through shaved skin, SDFTs, and
DDFTs (positioned in anatomically correct orientation). Influence of hair
color; skin preparation, color, and thickness; and wavelength on energy
penetration were assessed.
RESULTS
For haired skin, energy penetration was greatest for light-colored hair and
least for dark-colored hair. Clipping or shaving of skin improved energy pen-
etration. Light-colored skin allowed greatest energy penetration, followed
by medium-colored skin and dark-colored skin. Greatest penetration of
light-colored skin occurred with the 800-nm wavelength, whereas greatest
penetration of medium- and dark-colored skin occurred with the 970-nm
wavelength. As skin thickness increased, energy penetration of samples
decreased. Only 1% to 20% and 0.1% to 4% of energy were absorbed by
SDFTs and DDFTs, respectively, depending on skin color, skin thickness,
and applied wavelength.
CONCLUSIONS AND CLINICAL RELEVANCE
Results indicated that most laser energy directed through equine skin was
absorbed or scattered by the skin. To achieve delivery of energy doses
known to positively affect cells in vitro to equine SDFTs and DDFTs, skin
preparation, color, and thickness and applied wavelength must be consid-
ered. (Am J Vet Res 2016;77:991–999)
and oxygen consumption in cells cultured in vitro,9,10
which may be explained by absorption of laser light
(wavelength, 600 to 1,000 nm) by cytochrome c,11,12
a component of the mitochondrial electron transport
chain. Further, absorption of light energy by cyto-
chrome b and flavoproteins of the cell membrane-
bound nicotinamide adenine dinucleotide phosphate
(reduced form [NADPH]) oxidase complex13 has
been suggested to induce subtle changes in cells’
redox potential by promoting the production of re-
active oxygen species,13–15 resulting in altered gene
expression.16
Although in vitro investigations in other species
have suggested possible benefits of LLLT in the treat-
ment of tendon injuries, such as promotion of fibroblast
proliferation,17 increased decorin and type I collagen
gene expression in tenofibrocytes,17 and stimulation of
tenocyte migration,18 LLLT failed to improve the his-
991
tologic appearance of experimentally induced tendon
lacerations in horses.19 Also, LLLT did not appear to
affect clinical outcome in racehorses with superficial
digital flexor tendon injuries.20 These negative results
may be attributable to energy doses (termed fluence
[measured in J/cm2]) of LLLT that were inappropriate.
The relationship between laser fluence and biological
response appears to be biphasic, with low fluence as-
sociated with biostimulatory effects and high fluence
associated with no or negative effects, emphasizing the
importance of appropriate fluence selection in LLLT.21
Alternatively, it is possible that LLLT is ineffective in
equids because of unknown species-specific factors.
When attempting to affect deeper tissues with
LLLT, translation of effective in vitro fluence to the in
vivo setting is hampered by loss of laser energy via re-
flection, absorption, and scattering by the skin. Prep-
aration of skin and hair has been shown to influence
low-level laser energy penetration through horse
limbs, with clipping of hair followed by cleaning the
skin with alcohol allowing greatest energy penetra-
tion through the tissues.22 However, it is unknown
how much laser energy is transmitted through equine
skin. In the only previous study22 to investigate the
penetration of laser light through equine tissues, mea-
surements were made with the laser beam penetrat-
ing the entire distal portion (from lateral to medial)
of the equine limbs. Thus, the objective of the study
reported here was to measure penetration efficiency
of low-level laser energy through equine skin and to
determine what fraction of delivered laser energy is
absorbed by the superficial and deep digital flexor
tendons of horses. We hypothesized that, similarly to
what has been shown previously in vivo, skin prepa-
ration influences penetration of laser energy through
skin ex vivo. We further hypothesized that hair color,
skin color, and skin thickness influence penetration
of laser energy through equine skin ex vivo.
Materials and Methods
Laser unit and measurement setup
A class IV therapeutic lasera with wavelength ca-
pabilities of 800 and 970 nm was used at its minimum
beam diameter (12 mm) and at a constant distance
(9.0 cm) from a photodetector. Results of preliminary
experiments indicated that the distance between the
photodetector and the laser did not influence energy
penetration efficiencies. The photodetector was a sili-
con PIN photodiodeb with an active region of 9.7 X
9.7 mm, reverse biased by a 9.3-V battery in series
with a 100-Ω load resistor. To assure a sample area
smaller than the active region of the photodetector,
a 5-mm-diameter aperture in a 0.5-mm-thick alumi-
num plate was centered over the photodetector. To
protect the detector from contamination by the tissue
samples, a film of plastic wrapc was placed over the
aluminum plate, and tissue samples were situated di-
rectly over the aperture onto the plastic wrap.
The optical signal was quantified by measuring
the voltage produced across the load resistor by the
992 AJVR • Vol 77 • No. 9 • September 2016
detector photocurrent. The voltage across the load
resistor was connected to the 1MΩ input of a direct
current-coupled 100-MHz oscilloscope.d With a time
sweep of 500 µs/major division and a reduced band-
width of 20 MHz, and by continuously averaging 16
sweeps to reduce the noise level, each measurement
was recorded on 3 occasions and a mean value calcu-
lated. The signal level for the 100% laser signal with
no tissue sample but with clean plastic wrapc and ap-
erture in place was maintained < 8 V to assure detec-
tor linearity at all signal levels.
Collection of tissue samples
Tissue samples were collected from a randomly
chosen forelimb of each of 19 horses euthanized
by means of an IV overdose (≥ 1 mL/4.5 kg of body
weight) of pentobarbital and phenytoine for rea-
sons unrelated to the investigation. Breeds included
American Quarter Horse (n = 11), Thoroughbred (4),
warmblood (3), and Miniature horse (1). All hors-
es were ≥ 2 years of age. One sample per horse of
grossly healthy-appearing skin with normal hair coat
and superficial and deep digital flexor tendons was
collected en bloc from the middle to distal third of
the metacarpal region (minimum proximal to distal
length, approx 7 cm), dissected to separate skin and
both tendons, and frozen at –20°C. All samples (n =
19) were allowed to thaw to room temperature (ap-
prox 21°C) before measurements were obtained.
Skin samples
Tissue thickness of shaved skin was determined
in the area of laser application with a dial caliper.f
Perceived colors of hair and skin were recorded sepa-
rately. For data analysis involving measurements with
hair present, samples were categorized into 3 groups
on the basis of hair color (light, medium, or dark). For
data analysis involving measurements of clipped or
shaved skin, samples were categorized into 3 groups
on the basis of subjectively perceived skin pigmenta-
tion (light, medium, or dark).
To ascertain that subjectively perceived skin col-
or was consistent with more objective measures of
skin pigmentation and to provide a continuous inde-
pendent variable for skin pigmentation, the percent-
age of scattered light at the 750-nm wavelength was
measured for all shaved skin samples. For these mea-
surements, the wavelength of 750 nm was selected
because it is the visible wavelength23 that is closest
to the wavelengths of the therapeutic laser used in
this study. The percentage of scattered light was de-
termined by directing a white-light quartz halogen
microscope illuminatorg onto the skin from near-nor-
mal incidence and a distance of 10 cm. The scattered
light spectrum from 400 to 800 nm was collected at
near-normal incidence with a 250-µm-diameter opti-
cal fiber and miniature spectrometer.h The 100% ref-
erence scattering spectrum was measured by use of a
sheet of bright white paper i instead of a skin sample.
Subsequently, skin samples were ordered (highest to
lowest) on the basis of the percentage of scattered
light at the 750-nm wavelength (Table 1). It was ap- determine the effects of skin color, skin thickness,
parent from this ordering that grouping based on and laser energy wavelength on penetration of laser
subjectively perceived skin color was adequate to energy.
reflect measured skin pigmentation at a wavelength
of 750 nm. The percentage of scattered light at a Laser energy absorption by digital flexor
wavelength of 750 nm for white-colored skin samples tendons
(group designated as light-colored skin; n = 4) was To calculate the percentages of laser energy ab-
66.0% to 48.9%. For dark brown–, gray-, and all dark sorbed by shaved skin, superficial digital flexor ten-
gray–colored skin samples but 1 (group designated dons, and deep digital flexor tendons, measurements
as medium-colored skin; n = 9), the percentage of were obtained from shaved skin, shaved skin with the
scattered light at a wavelength of 750 nm was 25.0% superficial digital flexor tendon arranged in anatomi-
to 9.6%. For 1 dark gray– and the black-colored skin cally correct orientation, and shaved skin with the
samples (group designated as dark-colored skin; n = superficial and deep digital flexor tendons arranged
6), the percentage of scattered light at a wavelength in anatomically correct orientation. The respective
of 750 nm was 4.7% to 3.2%. values were expressed as a percentage of reference
laser energy, which was measured without any tissue
Experimental conditions between laser and photodetector. For the purpose
Optical transmission measurements were made of these calculations (Appendix 1), it was assumed
for continuous wave laser power of 1.0 W at 800 nm, that all energy that did not penetrate tissue placed
at 980 nm, and then at 800 and 980 nm emitted simul- between the laser and the detector was absorbed by
taneously. Measurements at these wavelengths were the tissue or tissues of interest.
performed without any tissue between the laser and
the photodetector (100% reference laser energy) and Data analysis
then with skin with the natural hair cleaned with wa- Penetration of laser energy through tissues was
ter to determine the influence of hair color on en- determined as a percentage of the laser energy de-
ergy penetration through haired skin samples. Subse- tected without any tissue between the laser and the
quently, to determine the effect of skin preparation photodetector at the same laser settings. Values are
on energy penetration, measurements of skin-only reported as mean ± SEM to provide an estimate of
samples were repeated after clippingj the hair and the precision of the sample mean in relation to the
cleaning the skin with water and then after shavingk population mean. Statistical analysis to determine dif-
the hair and cleaning the skin with water in the same ferences in laser energy penetration through equine
location as previous measurements. Measurements skin under the various experimental conditions was
from shaved-skin-only samples were then used to performed by ANOVA and post hoc F-protected t
Table 1—Characteristics of 19 equine skin samples collected from 1 metacarpal area of each of 19 equine cadavers and used to
determine skin penetration of laser energy at 3 different wavelength settings (800 nm, 970 nm, and both 800 and 970 nm emitted
simultaneously).
Percentage
of scattered light
at 750 nm Assigned skin Perceived hair Assigned hair Skin thickness
for each sample Perceived skin color color group color color group (mm)
66.0 White Light White Light 2.6
64.2 White Light White Light 1.6
52.1 White Light White Light 2.8
48.9 White Light Palomino Light 2.7
24.9 Gray Medium Dark chestnut Medium 2.5
23.1 Gray Medium Gray Medium 3.5
21.1 Dark brown Medium Chestnut-sorrel Medium 2.4
18.2 Gray Medium Brown Medium 1.6
12.6 Gray Medium Palomino Medium 2.7
12.3 Gray Medium Light brown Medium 1.5

12.1 Dark gray Medium Light gray Light 1.9
9.7 Dark gray Medium Mixed light and dark brown Medium 1.7
9.6 Gray Medium Mixed black, gray, and white Medium 1.5
4.7 Dark gray Dark Black with single cream Dark 1.8
and brown-colored hairs
4.6 Black Dark Black Dark 1.6
4.1 Black Dark Black Dark 2.5
3.9 Black Dark Black Dark 2.0
3.6 Black Dark Black Dark 1.9
3.2 Black Dark Black Dark 2.4
Grossly healthy-appearing skin with normal hair coat and superficial and deep digital flexor tendons were collected en bloc from the middle to
distal third of the metacarpal region and then dissected to separate skin and both tendons. Subjectively perceived hair and skin colors of samples
were recorded, and then the percentage of scattered light at the 750-nm wavelength was measured for skin samples after shaving and cleaning with
water. Those values were ordered (highest to lowest) and confirmed that sample grouping based on subjectively perceived skin color was adequate
to reflect measured skin pigmentation at a wavelength of 750 nm.

AJVR • Vol 77 • No. 9 • September 2016 993


Figure 1—Mean ± SEM laser energy penetration (expressed
as a percentage of laser energy detected without any tissue
between laser and photodetector) of equine skin samples
with light-colored hair (n = 5; A), medium-colored hair (8;
B), or dark-colored hair (6; C) at 3 different wavelength set-
tings (800 nm, 970 nm, and both 800 and 970 nm emitted
simultaneously). A class IV therapeutic laser with wavelength
capabilities of 800 and 970 nm was used at its minimum beam
diameter (12 mm) and at a constant distance (9.0 cm) from
a photodetector. Measurements of laser energy penetrating
skin from 1 metacarpal area of each of 19 equine cadavers
were obtained first with hair present that was cleaned with
water (light gray bars), then after hair was clipped and skin
was cleaned with water (gray bars), and finally after hair was
shaved and skin was cleaned with water (black bars). Re-
peated measurements were performed in the same location
as previous measurements. *Value for samples with cleaned
hair present is significantly (P < 0.001) less than the value for
the same samples after clipping or shaving. †Value for samples
with cleaned hair present is significantly (P < 0.01) less than
the value for the same samples after clipping or shaving.
994 AJVR • Vol 77 • No. 9 • September 2016
tests.l Association between tissue thickness and la-
ser energy penetration was determined by multiple
linear regression analysis,m,n with wavelength and
percentage of scattered light at 750 nm in the model.
Significance was set at a value of P < 0.05.
Results
When the beam was directed through skin-only
samples prior to clipping or shaving, the subjectively
determined hair color significantly influenced energy
penetration at all 3 wavelength conditions (ie, con-
tinuous wave laser power of 1.0 W at 800 nm, at 980
nm, and then at 800 nm and 980 nm emitted simulta-
neously). Energy penetration through skin with light-
color haired was higher, compared with skin samples
with medium- or dark-colored hair (all P < 0.001).
For skin samples with light-colored hair, mean ±
SEM percentage laser energy penetration (compared
with laser energy penetration in the absence of tis-
sue) was 14.9 ± 3.6%, 11.3 ± 2.1%, and 13.3 ± 2.7% at
wavelengths of 800 nm and 970 nm and at both wave-
lengths emitted simultaneously, respectively. For skin
samples with medium-colored hair, mean percent-
age laser energy penetration was 0.3 ± 0.1%, 0.8 ±
0.2%, and 0.6 ± 0.2% at wavelengths of 800 nm and
970 nm and at both wavelengths emitted simultane-
ously, respectively. For skin samples with dark-colored
hair, mean percentage laser energy penetration was
0.04 ± 0.1%, 0.6 ± 0.6%, and 0.1 ± 0.1% at wavelengths
of 800 nm and 970 nm and at both wavelengths emit-
ted simultaneously, respectively.
Skin preparation had a significant effect on la-
ser energy penetration through equine skin samples
(Figure 1). For skin samples with subjectively deter-
mined light- or medium-colored hair, penetration was
increased after clipping or shaving of the hair, com-
pared with the effect of only cleaning the hair, in all
wavelength settings. However, for skin samples with
subjectively determined dark-colored hair, this effect
was detected at only the 970-nm wavelength setting.
For shaved skin samples, subjectively determined
skin color significantly influenced laser energy pen-
etration at all 3 wavelength settings (Figure 2). Laser
energy penetration of light-colored skin was greater
than that of medium- or dark-colored skin. Laser en-
ergy penetration of medium-colored skin was greater
than that of dark-colored skin.
Wavelength setting (800 nm, 980 nm, or 800 and
980 nm emitted simultaneously) had a significant ef-
fect on laser energy penetration in the shaved skin
samples (Table 2). For shaved light-colored skin
samples, laser energy penetration at a wavelength
of 800 nm was more efficient than that achieved at
a wavelength of 970 nm. For shaved medium- and
dark-colored skin samples, laser energy penetration
at a wavelength of 970 nm was more efficient than
that achieved at a wavelength of 800 nm. With the
combined wavelength setting, the efficiency of laser
energy penetration was intermediate for shaved skin
samples of all colors.
 Results of the multiple linear regression analysis
indicated that skin thickness, as well as skin pigmen-
tation measured as a percentage of scattered light at
750 nm, had a significant effect on laser energy pen-
etration through shaved skin samples (adjusted R2 =
0.87; P < 0.001; multiple regression equation: y = 8.6
– 3.5x1 + 0.38x2, with x1 being skin thickness in milli-
meters and x2 being the percentage of scattered light
at 750 nm). As skin thickness increased, laser energy
penetration decreased (P < 0.001). Furthermore, the
greater the percentage of scattered light at 750 nm,
the greater the laser energy penetration (P < 0.001).
Laser energy penetration was determined for
samples of shaved skin, shaved skin with the superficial
digital flexor tendon, and shaved skin with the superfi-
cial and deep digital flexor tendons, arranged in anatom-
ically correct fashion. Laser energy absorption by skin,
the superficial digital flexor tendon, and the deep digital
flexor tendon was calculated (Table 3).

Figure 2—Mean ± SEM laser energy penetration of the
equine skin samples in Figure 1 classified on the basis of skin
color (light-colored skin [n = 4; light-gray bars], medium-col-
ored skin [9; gray bars], or dark-colored skin [6; black bars])
at 3 different wavelength settings (800 nm, 970 nm, and both
800 and 970 nm emitted simultaneously). First, subjectively
perceived skin color (white, gray, dark gray, dark brown, or
black) of samples was recorded, and then the percentage of
scattered light at the 750-nm wavelength was measured for
samples after shaving and cleaning with water. The values
were ordered (highest to lowest) and confirmed that sample
grouping based on subjectively perceived skin color was ad-
equate to reflect measured skin pigmentation at a wavelength
of 750 nm. The group designated as the light-colored skin
sample group included samples with skin color perceived as
white. The group designated as the medium-colored skin
sample group included samples with skin perceived as dark
brown and gray. The group designated as the dark-colored
skin sample group included 1 sample perceived as dark gray
and the samples perceived as black. *,†,‡Within a given wave-
length setting, different superscript symbols indicate signifi-
cant (all P < 0.001) differences between the means. See Figure
1 for key.
Discussion
Low-level laser therapy has been shown to posi-
tively affect tendon-derived fibroblasts of some spe-
cies in vitro. The positive effects include cell activities
thought to be necessary for tendon healing, such as
increased cell proliferation,17,24 cell migration,18 and
expression of tendon matrix proteins.17,18 These ef-
fects were found to occur at a relatively narrow range
of low fluence, with 2 J/cm2 possibly having the great-
est effects. Translation of effective in vitro fluence to
an in vivo setting requires adjustment for losses of
laser energy via reflection, absorption, and scattering
by tissues overlying the injured tendon. In the pres-
ent study, the influence of subjectively assessed hair
color, skin pigmentation, skin preparation, and wave-
length on laser energy penetration through equine
skin was investigated. The intent was that the study
data would provide guidance for equine practitioners
with regard to adjustment of laser energy outputs for
individual patients to achieve delivery of appropriate
in vivo energy fluence for augmentation of healing of

Table 2—Mean ± SEM laser energy penetration (expressed as a percentage of laser energy de-
tected without any tissue between laser and photodetector) at 3 different wavelength settings
(800 nm, 970 nm, and both 800 and 970 nm emitted simultaneously) through shaved equine skin
samples in Table 1 that were classified as light-, medium-, or dark-colored skin.

Wavelength Mean percentage

Skin color (nm) laser energy penetration P value
Light (n = 4) < 0.001
800 25.1 ± 2.1a
970 19.1 ± 1.7b
800 and 970 combined 21.9 ± 1.6c
Medium (n = 9) < 0.001
800 5.9 ± 0.8a
970 9.0 ± 1.3b
800 and 970 combined 7.5 ± 1.0c
Dark (n = 6) 0.006
800 1.2 ± 0.4a
970 3.5 ± 0.9b
800 and 970 combined 2.4 ± 0.7a,b
A class IV therapeutic laser with wavelength capabilities of 800 and 970 nm was used at its minimum beam
diameter (12 mm) and at a constant distance (9.0 cm) from a photodetector.
a–cWithin a given skin color group, values with different letters are significantly different.

See Table 1 for key.

AJVR • Vol 77 • No. 9 • September 2016 995

Table 3—Mean ± SEM laser energy absorption (expressed as the percentage of the total laser energy detected without any tissue
between the laser and the photodetector) at 3 different wavelength settings (800 nm, 970 nm, and both 800 and 970 nm emitted
simultaneously) in samples of shaved skin, superficial digital flexor tendons, and deep digital flexor tendons collected en bloc from
1 metacarpal area of each of the 19 equine cadavers in Table 1 that were classified as having light-, medium-, or dark-colored skin.
Skin color group

Light (n = 4) Medium (n = 9) Dark (n = 6)

800 and 800 and 800 and

Tissue 800 nm 970 nm 970 nm 800 nm 970 nm 970 nm 800 nm 970 nm 970 nm

Skin 74.9 ± 2.1 80.9 ± 1.7 78.1 ± 1.6 94.1 ± 0.8 91.0 ± 1.3 92.5 ± 1.0 98.8 ± 0.4 96.5 ± 0.9 97.6 ± 0.7
Superficial 20.7 ± 2.7 16 ± 1.7 18.2 ± 1.9 5.2 ± 0.7 7.9 ± 1.1 6.6 ± 0.8 1.2 ± 0.4 3.3 ± 0.9 2.3 ± 0.6
digital flexor
tendon
Deep 4.1 ± 1.7 2.8 ± 1.4 3.4 ± 1.5 0.7 ± 0.2 1.1 ± 0.3 0.9 ± 0.2 0.1 ± 0.0 0.2 ± 0.0 0.1 ± 0.0
digital flexor
tendon

To calculate the percentages of laser energy absorbed by shaved skin, superficial digital flexor tendons, and deep digital flexor tendons, measure-
ments were obtained from shaved skin, shaved skin with the superficial digital flexor tendon, and shaved skin with the superficial and deep digital
flexor tendons, arranged in anatomically correct fashion. The respective values were expressed as a percentage of reference laser energy, which
was measured without any tissue between laser and photodetector.
See Tables 1 and 2 for key.

flexor tendon injuries in horses. The data obtained in
the present study suggested that without laser energy
output adjustment based on perceived equine skin
color and skin thickness, it is unlikely that effective
energy fluences are delivered to injured tendons.
Similar to findings in a previous study22 in horses,
penetration of laser energy through skin samples was
significantly less when the hair was not clipped or
shaved, although there was no difference between
the effect of clipping and shaving in the present
study. In the presence of natural hair, only white or
cream-colored hair allowed laser energy penetration
through equine skin, whereas gray, brown, chest-
nut, or black hair blocked > 99% of the laser energy
penetration. Thus, the data obtained in the present
study supported previous recommendations to clip
or shave and then clean the skin over the area to be
treated via LLLT.22 After clipping or shaving the skin,
penetration of laser energy through equine skin was
significantly affected on the basis of the subjectively
perceived skin color; laser energy penetration was
greatest for nonpigmented skin, followed by that for
moderately pigmented skin and then black-pigmented
skin. This was not surprising, given that skin color is
influenced by the type and amount of melanin within
the skin and light of a wavelength of 800 or 970 nm is
absorbed by melanin.21
In the present study, skin thickness was sig-
nificantly associated with laser energy penetration
through equine skin samples, but the effect was small,
compared with the effect of skin pigmentation. This
may be a result of the minor variability in thickness
of the skin samples used (thickness range, 1.5 to 3.5
mm). Correction of laser energy settings on the basis
of skin thickness would theoretically be possible by
ultrasonographically measuring skin thickness at the
region requiring treatment.25 However, correction of
laser energy settings on the basis of skin thickness
is apparently much less important than corrections
based on skin pigmentation.
It has been stated that longer-wavelength laser
light penetrates deeper into tissues, compared with
the effect of shorter-wavelength laser light,8 but in the
present study, this was dependent on skin pigmenta-
tion of the equine skin samples. In skin samples with
moderate or dark pigmentation, laser energy penetra-
tion was greater at a wavelength of 970 nm than at a
wavelength of 800 nm. However, laser energy pen-
etration in nonpigmented skin samples was greater
at a wavelength of 800 nm than at a wavelength of
970 nm. A possible explanation is that more laser ener-
gy was absorbed by melanin in the skin at the 800-nm
wavelength than at the 970-nm wavelength and that
more laser energy was absorbed by water at the 970-
nm wavelength than at the 800-nm wavelength.8 In
nonpigmented skin, the major chromophore may be
water because of the lack of melanin, which would
result in better skin penetration by laser energy at
the 800-nm wavelength, compared with the effects
of longer wavelengths.
In our opinion, the best estimate for a clinically
effective laser fluence delivered to tendon fibroblasts
in the area of a tendon lesion in vivo is the fluence
that has been shown to improve cellular functions in
vitro. After receiving a laser fluence of 2.0 to 2.5 J/cm2
(660 nm, 50 mW, and continuous mode), rat Achil-
les tenocytes in monolayer culture had increased
cell proliferation24 and cell migration.18 Furthermore,
administration of 1 to 3 J/cm2 (820 and 635 nm,
40 mW, and 50-Hz pulse frequency) was associated
with increased gene expression of decorin and col-
lagen type I in porcine Achilles tendon fibroblasts.17
Cell proliferation was enhanced at a fluence of 2.00
to 2.16 J/cm2, but not at higher fluences.17,26 On the
basis of these findings, a fluence of 2 J/cm2 may be
the most appropriate fluence in vitro to stimulate cel-
lular functions that possibly benefit tendon healing.
With as little as 1.9% of emitted laser energy penetrat-
ing shaved, black-pigmented equine skin in the pres-
ent study, the question arises whether it is feasible

996 AJVR • Vol 77 • No. 9 • September 2016

to deliver this energy fluence to injured flexor ten-
dons within a reasonable amount of time. When at-
tempting to treat the superficial digital flexor tendon
in a horse with black-pigmented skin, the prediction
based on the data obtained in the present study is
that use of a laser beam at a wavelength of 800 nm
would most likely result in only 1.9% of the laser en-
ergy penetrating the skin (Table 2). Also, in general,
1.8% of the laser energy is absorbed by the superfi-
cial digital flexor tendon (Table 3). Thus, with the la-
ser set at a 1-W power output and a spot diameter of
1.2 cm, it would take 10 minutes and 24 seconds of
treatment time to deliver a fluence of 2 J/cm2 to a ten-
don lesion that was determined to be 1 cm in width
and 5 cm in length (Appendix 2). However, even
the assumption that 1.8% of laser energy is absorbed
by the superficial digital flexor tendon is probably an
overestimation because additional energy is likely to
be reflected or scattered by the tendon. This energy
loss is very difficult, if not impossible, to assess27 and
subsequently has not been taken into account in the
example calculation.
When attempting to treat the deep digital flexor
tendon, it may be theoretically beneficial to avoid
having to penetrate skin and superficial digital flexor
tendon by directing the beam toward the lesion in
the lateral or medial plane, as opposed to the sagit-
tal plane, from the palmar or plantar aspect because
only approximately 1% of laser energy (at the 800-
nm wavelength) penetrated black-pigmented skin
and the superficial digital flexor tendon in the pres-
ent study. However, we did not attempt to determine
and thus do not know how much laser energy would
be absorbed by healthy tendons when treatment is
directed in this fashion.
A limitation of the present study was that we
used cadaver tissues with lack of blood flow. Hemo-
globin is a chromophore, and it is conceivable that
perfused equine skin attenuates the laser beam to a
greater extent than does nonperfused skin, resulting
in target tissues that receive even less laser energy
than what was calculated in this study. Also, absorp-
tion of laser energy by damaged areas of tendon, such
as core lesions, may be different from findings for
the normal tendons used in the present study. Fur-
thermore, it is unknown how much laser energy is
absorbed by tenocytes located at different depths
within the tendon. This illustrates the difficulty of
translating biostimulatory effects of LLLT in monolay-
er cultures to 3-D tissues. It is also unknown whether
tenocytes surrounded by tendon matrix would have
the same biostimulatory responses as tenocytes in
monolayer culture. Additionally, inflammatory media-
tors that are found in tendon lesions28 may also alter
tenocytes’ responses to LLLT. Finally, it is possible
that LLLT is ineffective in influencing tendon healing
in equids because of species-specific factors, such as
biomechanics or wound healing.
The data obtained in the present study have clear-
ly suggested that to deliver laser output equivalent to
AJVR • Vol 77 • No. 9 • September 2016
that which provides effective laser energy fluences
in vitro, adjustments to laser energy output must be
made depending on the preparation of the skin and
the subjectively determined skin color. Hair color in-
fluenced penetration of laser energy through equine
skin at wavelengths of 800 and 970 nm. Only hair of
light color (white or cream colored) allowed penetra-
tion of sufficient laser energy to treat an area without
clipping or shaving the overlying skin and without
excessively long treatment times. As has been shown
previously, penetration of laser energy through
equine skin was improved by clipping or shaving of
the hair, followed by cleaning of the skin. Laser en-
ergy penetration of equine skin was dependent on
wavelength. Penetration of nonpigmented skin was
better with a wavelength of 800 nm, compared with
the effect of a wavelength of 970 nm, and penetration
of pigmented skin was better with a wavelength of
970 nm, compared with the effect of a wavelength
of 800 nm. Laser energy penetration decreased with
increasing skin thickness, but it was influenced to
a much greater extent by skin pigmentation: the
darker the skin pigmentation, the less laser energy
penetration of the skin. Thus, we accept our hypoth-
eses that hair color, skin preparation, skin color, and
skin thickness influence penetration of laser energy
through equine skin ex vivo. We also conclude that
laser energy of 2 J/cm2 can be delivered to equine
superficial digital flexor tendons, even in horses with
black-pigmented skin, although adjustments must be
made to assure adequate duration of treatment. In our
opinion, delivery of that fluence to the deep digital
flexor tendon or the suspensory ligament appears to
be impossible if the laser beam is directed through
skin and the superficial digital flexor tendon.
Acknowledgments
Supported by the Department of Clinical Sciences, College of
Veterinary Medicine, Oregon State University.
Presented in part in poster form at the 2014 American College
of Veterinary Surgeons Surgery Summit, San Diego, Calif, October
2014.
Foonotes
a. K Series 1200, K-Laser, Franklin, Tenn.
b. FDS1010, Thorlabs, Newton, NJ.
c. Saran, SC Johnson, Racine, Wis.
d. TDS 220, Tektronix, Beaverton, Ore.
e. Beuthanasia-D Special, Merck Animal Health, Madison, NJ.
f. Series 505, Mitutoyo, Aurora, Ill.
g. L2, Leica, Buffalo Grove, Ill.
h. USB4000, Ocean Optics, Dunedin, Fla.
i. X-9, Boise White Paper, Boise, Ind.
j. 40 blade, Oster, McMinnville, Tenn.
k. Disposable single-blade shaver, Bic, Clichy, France.
l. SAS, version 9.4, SAS Institute Inc, Cary, NC.
m. Daniel’s XL Toolbox add in version 6.52 for Excel, Daniel
Kraus, Würzburg, Germany.
n. Excel 2010, Microsoft, Redmond, Wash.
References
1. Ramey DW, Basford JR. Laser therapy in horses. Compend
Contin Educ Pract Vet 2000;22:263–271.
2. Tumilty S, Munn J, McDonough S, et al. Low level laser treat-
997
 ment of tendinopathy: a systematic review with meta-analy-
sis. Photomed Laser Surg 2010;28:3–16.
3. Marcos RL, Leal ECP Jr, de Moura Messias F, et al. Infrared
(810 nm) low-level laser therapy in rat Achilles tendinitis:
a consistent alternative to drugs. Photochem Photobiol
2011;87:1447–1452.
4. Alfredo PP, Bjordal JM, Dreyer SH, et al. Efficacy of low
level laser therapy associated with exercises in knee osteo-
arthritis: a randomized double-blind study. Clin Rehabil
2012;26:523–533.
5. Bjordal JM, Johnson MI, Lopes-Martins RA, et al. Short-term
efficacy of physical interventions in osteoarthritic knee pain.
A systematic review and meta-analysis of randomised place-
bo-controlled trials. BMC Musculoskelet Disord 2007;8:51.
6. Hawkins D, Houreld N, Abrahamse H. Low level laser ther-
apy (LLLT) as an effective therapeutic modality for delayed
wound healing. Ann N Y Acad Sci 2005;1056:486–493.
7. Vladimirov YA, Osipov AN, Klebanov GI. Photobiological
principles of therapeutic applications of laser radiation. Bio-
chemistry (Mosc) 2004;69:81–90.
8. Chung H, Dai T, Sharma SK, et al. The nuts and bolts of low-
level laser (light) therapy. Ann Biomed Eng 2012;40:516–
533.
9. Karu T, Pyatibrat L, Kalendo G. Irradiation with He-Ne laser
increases ATP level in cells cultivated in vitro. J Photochem
Photobiol B 1995;27:219–223.
10. Benedicenti S, Pepe IM, Angiero F, et al. Intracellular ATP
level increases in lymphocytes irradiated with infrared
laser light of wavelength 904 nm. Photomed Laser Surg
2008;26:451–453.
11. Passarella S, Casamassima E, Molinari S, et al. Increase of pro-
ton electrochemical potential and ATP synthesis in rat liver
mitochondria irradiated in vitro by helium-neon laser. FEBS
Lett 1984;175:95–99.
12. Yu W, Naim JO, McGowan M, et al. Photomodulation of oxi-
dative metabolism and electron chain enzymes in rat liver
mitochondria. Photochem Photobiol 1997;66:866–871.
13. Lavi R, Shainberg A, Friedmann H, et al. Low energy visible
light induces reactive oxygen species generation and stimu-
lates an increase of intracellular calcium concentration in
cardiac cells. J Biol Chem 2003;278:40917–40922.
14. Grossman N, Schneid N, Reuveni H, et al. 780 nm low power
diode laser irradiation stimulates proliferation of keratino-
cyte cultures: involvement of reactive oxygen species. La-
sers Surg Med 1998;22:212–218.
15. Zhang J, Xing D, Gao X. Low-power laser irradiation activates
Appendix 1
Calculations used to determine absorption of laser energy by samples of shaved skin, SDFTs, and DDFTs obtained from the meta-
carpal areas of equine cadavers.
Variable
Value measured with no tissue between laser and detector
Value measured with skin only between laser and detector
Value measured with skin and SDFT between laser and detector
Value measured with skin, SDFT, and DDFT between laser and detector
When only skin is between laser and detector
When skin and SDFT are between laser and detector
When skin, SDFT, and DDFT are between laser and detector
Energy absorbed by skin
Energy absorbed by SDFT
Energy absorbed by DDFT
DDFT = Deep digital flexor tendon. SDFT = Superficial digital flexor tendon.
E = emitted energy. A = absorbed energy. T = transmitted energy.
998 AJVR • Vol 77 • No. 9 • September 2016
Src tyrosine kinase through reactive oxygen species-mediat-
ed signaling pathway. J Cell Physiol 2008;217:518–528.
16. Lubart R, Eichler M, Lavi R, et al. Low-energy laser irradia-
tion promotes cellular redox activity. Photomed Laser Surg
2005;23:3–9.
17. Chen CH, Tsai JL, Wang YH, et al. Low-level laser irradiation
promotes cell proliferation and mRNA expression of type I
collagen and decorin in porcine Achilles tendon fibroblasts
in vitro. J Orthop Res 2009;27:646–650.
18. Tsai WC, Hsu CC, Pang JHS, et al. Low-level laser irradiation
stimulates tenocyte migration with up-regulation of dyna-
min ii expression. PLoS ONE 2012;7:e38235.
19. Kaneps AJ, Hultgren BD, Riebold TW, et al. Laser thera-
py in the horse: histopathologic response. Am J Vet Res
1984;45:581–582.
20. Marr CM, Love S, Boyd JS, et al. Factors affecting the clini-
cal outcome of injuries to the superficial digital flexor ten-
don in National Hunt and point-to-point racehorses. Vet Rec
1993;132:476–479.
21. Huang YY, Chen AC, Carroll JD, et al. Biphasic dose
response in low level light therapy. Dose Response
2009;7:358–383.
22. Ryan T, Smith R. An investigation into the depth of penetra-
tion of low level laser therapy through the equine tendon in
vivo. Ir Vet J 2007;60:295–299.
23. Curtis H, Barnes NS. Photosynthesis, light, and life. In:
Curtis H, Barnes NS, eds. Invitation to biology. 5th ed. New
York: Worth Publishers, 1994;161–175.
24. Tsai WC, Cheng JW, Chen JL, et al. Low-level laser irradia-
tion stimulates tenocyte proliferation in association with in-
creased NO synthesis and upregulation of PCNA and cyclins.
Lasers Med Sci 2014;29:1377–1384.
25. Jain SM, Pandey K, Lahoti A, et al. Evaluation of skin and
subcutaneous tissue thickness at insulin injection sites in In-
dian, insulin naive, type-2 diabetic adult population. Indian
J Endocrinol Metab 2013;17:864–870.
26. Yu W, Naim JO, Lanzafame RJ. The effect of laser irradiation
on the release of bFGF from 3T3 fibroblasts. Photochem Pho-
tobiol 1994;59:167–170.
27. Hutchison AM, Beard DJ, Bishop J, et al. An investigation of
the transmission and attenuation of intense pulsed light on
samples of human Achilles tendon and surrounding tissue.
Lasers Surg Med 2012;44:397–405.
28. Hosaka Y, Kirisawa R, Yamamoto E, et al. Localization of cy-
tokines in tendinocytes of the superficial digital flexor ten-
don in the horse. J Vet Med Sci 2002;64:945–947.
Calculation
Etotal = 100% of emitted energy
Tskin = % of energy transmitted through skin
Tskin + SDFT = % of energy transmitted through skin and SDFT
Tskin + SDFT + DDFT = % of energy transmitted through skin, SDFT,
and DDFT
Etotal = (%)Askin (%) + Tskin (%)
Etotal = (%)Askin (%) + ASDFT (%) + Tskin+SDFT (%)
Etotal = (%)Askin (%) + ASDFT (%) + ADDFT (%) + Tskin + SDFT + DDFT (%)
Askin (%) = 100% – Tskin (%)
ASDFT (%) = 100% – Askin (%) – Tskin + SDFT (%)
ADDFT (%) = 100% – Askin (%) – ASDFT (%) – Tskin + SDFT + DDFT (%)
Appendix 2 appears on the next page
Appendix 2
Example calculation of treatment time with a therapeutic laser to provide LLLT in an SDFT in
metacarpal area of a horse.
Size of the tendon lesion (in the SDFT): 5 X 1 cm = 5 cm2
Attenuation of laser energy by black pigmented skin (through absorption and scattering): 98.2%
Target fluence to be delivered to tendon lesion: 2 J/cm2 = 10 J/5 cm2
Laser settings: power output = 1 W; spot size diameter = 12 mm

1) Calculation of energy emitted by laser:

1 W/1.13 cm2 = 0.89 W/cm2 = 0.89 J/s/cm2

2) Calculation of actual fluence delivery rate to the SDFT:

0.89 J/s/cm2 X 1.8% = 0.01602 J/s/cm2

3) Calculation of time necessary to deliver 2 J/cm2 to the SDFT:

2 J/cm2 / 0.01602 J/cm2 = 124.84 seconds
The treatment time required to deliver 2 J to 1 cm2 of an SDFT is 124.8 seconds. The treatment
time required to deliver 2 J to 5 cm2 of an SDFT is 624.2 seconds. Thus, the treatment time
required to deliver 10 J to 5 cm2 of SDFT is 10 minutes and 24 seconds.
SDFT = Superficial digital flexor tendon.

AJVR • Vol 77 • No. 9 • September 2016 999