Effects of laser power, wavelength, coat length, and coat color on tissue penetration using photobiomodulation in healthy dogs

Abstract

Photobiomodulation is an accepted regenerative medicine treatment modality used to stimulate tissue repair, mediate inflammation, and improve mobility in humans and animals. The objective of this study was to assess the influence of laser power and wavelength, coat length and color, and shaving on in-vivo photon delivery by therapeutic laser in dogs. Forty-seven dogs of various breeds and coat colors (17 black, 15 brown, and 15 white) and with varying coat lengths were assessed with 2 commercially available veterinary lasers. Photons were delivered to the lateral aspect of the inguinal fold and calcaneal tendon, with direct penetration through the dermis, as well as dermis and tendon, as measured with a thermopile laser sensor. Significant impacts on laser transmission were noted for laser power (P = 0.001), wavelength (P < 0.002), coat color (P < 0.001), and shaved coat (P < 0.001). Percent transmission was higher for a class IV 810/980 nm wavelength laser at 0.5 W than for a class IIIb 904 nm laser (P < 0.001). There was a significant difference between transmission of photons among white, brown, and black coats, with less transmission noted with increasing coat pigment (P < 0.001). Transmission was greater at higher power levels (3 W, 5 W) Results showed significant differences in laser transmission for all variables assessed, with the exception of coat length, which was not a significant predictor of laser transmission. As transmission was significantly reduced in darker and unshaved areas, higher power lasers may be necessary for darker pigmented dogs and shaving of hair is recommended before laser therapy.

Introduction

LASER therapy, an acronym for light amplification by stimulated emission of radiation, is an adjunctive modality that is increasingly used in the rehabilitative care of humans and animals. The earliest medical use of lasers occurred after a 694 nm ruby laser accelerated hair growth in shaved mice (1). Modern lasers have been recommended to treat a wide variety of conditions in dogs and animal models, such as healing of wounds (2,3), bone repair (4,5), spinal cord injury (6,7), preconditioning before orthopedic surgery to optimize post-operative recovery (8), and peripheral nerve regeneration (9).

Photobiomodulation, which is the influence of photons on tissues, is thought to occur through wavelength-dependent photon reception by cytochrome c oxidase in the mitochondria of treated cells. This accelerates the electron transport chain and results in an increased production of adenosine triphosphate (ATP) (10). Cellular dissociation of nitric oxide from cytochromes may also occur, thereby increasing metabolic turnover, vasodilation, and angiogenesis. Creation of reactive oxygen species during laser therapy may stimulate endogenous antioxidant production, progenitor cell differentiation, and cellular growth (11).

Most therapeutic lasers emit photons outside the visible spectrum that contribute to biochemical effects. A laser’s specific monochromaticity, collimation, and coherence prevent the divergence of the electromagnetic radiation to promote the efficient propagation of energy into tissue. The practitioner must set a number of parameters, such as power, time, and possibly wavelength, that contribute to the total emitted and absorbed photon dose (12). Most veterinary therapeutic lasers have an emission wavelength of 600 to 1000 nm to meet the absorption spectrum range of cytochrome c oxidase. Photons of lower wavelengths may be absorbed by hemoglobin and melanin, while higher wavelengths can be absorbed by water (11).

Medical lasers can be either therapeutic or surgical, or as commonly referred to, “cold” and “hot.” This terminology creates confusion as therapeutic lasers also generate thermal effects, but without the intent of tissue destruction or coagulation as seen in surgical lasers. Classes of lasers are commonly defined by power, with class IIIb producing 5 mW to 0.5 W and class IV producing > 0.5 W. When tolerated, a higher power class IV laser delivers a dose in less time than a class IIIb low-energy laser. The photon dose applied to the target tissue is critical as laser displays a therapeutic range between both a lower, ineffective dose cutoff and a higher cytotoxic dose limit (13).

An optimal tissue dose of approximately 3 to 10 J/cm² has been suggested for animals, but this depends on the therapeutic goal (14–18). In companion animals, while the color and density of the coat could impact laser penetration to the target tissue, few studies have evaluated tissue penetration in vivo. Previous equine studies reported that laser penetration was significantly diminished when transmitting through unshaved skin (19) and that penetration was reduced in ex-vivo skin samples with darker skin and coat color (20).

The objective of this study was to determine the impact of laser power and wavelength, coat length and color, and shaving of the coat on the transmission of a fixed wavelength laser at the inguinal fold, which represents 2 layers of dermis, and over the calcaneal tendon, which represents 2 layers of dermis bounding dense fibrous tissue, in healthy dogs. It was hypothesized that an intact and longer hair coat, more heavily pigmented hair, shorter wavelength, and lower power would all significantly reduce photonic transmission.

Materials and methods

Animals

Client-owned dogs with common coat and color patterns were recruited for participation, provided that coat color was consistent over the areas tested, including the inguinal fold and calcaneal tendon. A minimum weight of 10 kg was required to accommodate the laser sensor and probe on opposite sides of the dogs’ inguinal fold. Owners signed a consent form for research participation and the study was approved by the Institutional Animal Care and Use Committee of the University of Florida (IACUC protocol #201509167). Dogs were excluded if they had any comorbidities that could affect penetration, such as dehydration, peripheral or central edema, anemia, endocrinopathies, or dermatologic disease. Breeds with a mucinous subcutis were also excluded.

Measurement acquisition

Two laser therapy units were used, including a class IV laser (CTS Laser; Companion Animal Health, New Castle, Delaware, USA) and a class IIIb (2400XL; Respond Systems, Branford, Connecticut, USA). The class IIIb laser emitted a continuous beam of 904 nm, with an average power of 500 mW and a contact spot size of 1 cm². The class IV laser emitted a continuous dual wavelength beam of 810 nm/980 nm, with an adjustable power of 0.5 W to 15 W and a contact spot size of 2.21 cm². A low-profile thermopile sensor (PowerMax-RS PM10-19C; Coherent, Santa Clare, California, USA) was used, with a detection range of 190 to 11 000 nm and a power range of 10 mW to 10 W. In accordance with recommended safety methods, the investigator, the dog, and the assistant all wore protective goggles to prevent retinal injury from laser scatter. A digital Vernier caliper (Traceable; Fisher Scientific, Waltham, Massachusetts, USA) was used to measure the thickness of the inguinal fold and calcaneal tendon at the midpoint from musculotendinous transition to insertion on the calcaneus and the coat lengths over these respective areas. The general type of coat composition (smooth, wire, undercoat) and skin color was noted.

Each dog was manually restrained by a technician and the investigator in a private, windowless room with consistent lighting. The output of both lasers was tested, with the sensor held directly to the laser probe to ensure that anticipated output was within 5% of measured output. The Vernier caliper was used to measure the width of the inguinal fold and the calcaneal tendon, as well as coat length in millimeters over these respective areas. The laser probe was applied at the lateral aspect of both the inguinal fold and calcaneal tendon, with the sensor opposite it at the medial aspect. Power measurements were obtained over the inguinal fold and calcaneal tendon first with the hair intact and again after shaving with electric clippers (#40 blade). Each anatomic location was tested with the class IIIb laser (fixed 500 mW power) and the class IV laser at power levels 0.5 W, 1.5 W, 3 W, and 5 W. All measurements were done by the same investigator (LE) using light, firm, and perpendicular contact of the laser and sensor with the skin. The laser sensor and laser probe were held in position for approximately 3 s, or until penetration wattage readings plateaued at a maximum of 5 s, to obtain penetration readings (mW). This procedure was carried out in triplicate for each site and each laser power level, with the mean of these readings taken as the data point.

Statistical analysis

The dependent variable used for all statistical analyses was the transmission value expressed as a percentage of the laser energy delivered from each respective unit with the following equation:

$$\%\hspace{0.17em}\text{transmission}=\frac{[penetration\hspace{0.17em}1\hspace{0.17em}(mW)+penetration\hspace{0.17em}2\hspace{0.17em}(mW)+penetration\hspace{0.17em}3\hspace{0.17em}(mW)]/3}{laser\hspace{0.17em}energy\hspace{0.17em}setting\hspace{0.17em}(mW)}$$

The energy absorbed by the calcaneal tendon was calculated as % transmission inguinal fold — % transmission calcaneal tendon. The reported transmission value assumed that the energy not measured by the photosensor was absorbed by the treated tissue, as any possible refraction and reflection could not be quantified. Skewness and kurtosis statistics were run on each continuous distribution to check for the assumption of normality. If either statistic was above an absolute value of 2.0, the assumption of normality was not met and non-parametric statistics were used. The effect of laser power (5 W, 3 W, 1.5 W, and 0.5 W), coat color, and shave status on percent transmission was evaluated with Kruskall-Wallis tests with post-hoc, Bonferroni-corrected multiple comparisons. The effect of laser wavelength on transmission was evaluated with a Mann Whitney U-test. Across our population of differentially pigmented hair coats, body weight, and tissue width were compared with analysis of variance (ANOVA) and age and coat length with Kruskal-Wallis.

In order to assess the variables associated with transmission of photons, a backwards stepwise linear regression analysis was carried out on the log-transformed transmission percentages. For polychotomous categorical variables, including coat type, coat color, and laser power, the levels of the categories were dummy-coded for entry into the regression model. All variables, including skin color, coat color (white, brown, black), coat type (wire, smooth, undercoat), coat length, laser power (0.5 W, 1.5 W, 3 W, 5 W), shaving, and tissue thickness over the calcaneal tendon or inguinal fold, were forced into the model. Variables were taken out of the model if their individual significance was P > 0.2 to produce the best combination of variables that could account for significant variance in the log-transformed outcome of transmission. Variables were included in the model if the P-value associated with the variable at each iteration was below 0.10. Any P-value ≤ 5% was considered significant. All normality testing, Kruskal-Wallis, ANOVA, and Mann Whitney U-analyses were analyzed using a different software system (Statistical Software Version 17.3.1; Minitab, State College, Pennsylvania, USA) than the software used for the multivariable linear regression modeling analyses (SPSS Version 22; IBM Corporation, Armonk, New York, USA).

Results

Sample description

Forty-seven dogs were included in this study. Mean age was 6.0 y and mean weight was 25.2 kg. Twenty dogs were mixed breeds, 4 were Labrador retrievers, and 3 were Doberman pinschers. The remaining 20 dogs were of different breeds, with no more than 2 representatives of each breed. No significant differences were detected among color groups for age, weight, length of coat at inguinal fold and at calcaneal tendon, and tissue width over the inguinal fold (median: 3.39 mm, range: 1.39 to 7.38 mm) (Table I). However, brown dogs were found to have statistically significantly thicker calcaneal tendons (P = 0.01) than white and black dogs.

Table I.

Mean and standard deviation of general characteristics (age, body weight, regional skin thickness, coat length) of 47 dogs based on coat color categorization.

	Black (n = 17)	Brown (n = 15)	White (n = 15)	P-value
Age (y)	6.4 ± 3.1	5.4 ± 3.215	6.3 ± 3.7	0.78
Weight (kg)	23.7 ± 9.8	27.08 ± 9.7	24.6 ± 8.5	0.42
Inguinal fold coat length	32.2 ± 22.2	35.1 ± 38.6	49.1 ± 51.9	0.92
Inguinal fold width	3.1 ± 1.1	3.3 ± 0.9	3.9 ± 1.6	0.12
Calcaneal tendon coat length (mm)	20.8 ± 18.1	27.5 ± 36.1	41.1 ± 42.2	0.64
Calcaneal tendon width (mm)	6.2 ± 1.5	7.4 ± 1.7	6.4 ± 1.6	0.07

No significance was observed between groups of dogs assessed.

Table II.

Unstandardized and standardized beta coefficients for all variables of significance (power, coat color, tissue thickness, hair shaved) for photon transmission in backwards stepwise regression model.

Predictors	β (SE)	β	P-value
Constant	−3.47 (0.30)	—	< 0.001
Brown coat	0.71 (0.17)	0.24	< 0.001
White coat	0.99 (0.17)	0.34	< 0.001
1.5 W	−0.29 (0.17)	−0.09	0.082
3.0 W	−0.43 (0.16)	−0.13	0.007
5.0 W	−0.32 (0.17)	−0.09	0.065
Shaved coat	0.81 (0.13)	0.27	< 0.001
Inguinal fold width (mm)	−0.23 (0.05)	−0.20	< 0.001
Calcaneal tendon width (mm)	−0.14 (0.04)	−0.16	< 0.001

β — unstandardized beta coefficient; SE — standard error; β — standardized beta coefficient.

Statistical findings

Statistical assumptions for the backwards stepwise regression model were met after assessing the pertinent statistics and plots. A total of 7 iterations of the model were run to yield the best-fit model when predicting for the log-transformed transmission outcome. Brown coat color, white coat color, laser powers of 1.5 W, 3.0 W, 5.0 W, shaving, and width of the inguinal fold and calcaneal tendon were all included in the final model, with black coat color and a power of 0.5 W representing statistically significant reference categories. The unstandardized beta coefficients, standard errors, standardized beta coefficients, and P-values associated with the final model are presented in Table II and in the following equation:

$$\begin{array}{l}\%\hspace{0.17em}\text{transmission}=-3.47+0\hspace{0.17em}(\text{black\hspace{0.17em}coat})+0.71\hspace{0.17em}(\text{brown\hspace{0.17em}coat})+\\ 0.99\hspace{0.17em}(\text{white\hspace{0.17em}coat})+0\hspace{0.17em}(\text{low\hspace{0.17em}power})-0.43\hspace{0.17em}(\text{high\hspace{0.17em}power})+0.81\\ (\text{shaved})+0\hspace{0.17em}(\text{unshaved}).\end{array}$$

The effect of laser power (5 W, 3 W, 1.5 W, and 0.5 W), coat color, and shaving on percent transmission, as well as the effect of laser wavelength on transmission, are listed in Table III. Percent transmission values were higher at 0.5 W for the class IV laser with a wavelength of 810/980 nm than for the class IIIb laser with a wavelength of 904 nm (P < 0.001). Statistically significant differences were noted between the 0.5 W transmission values compared to the 3 W setting when using the class IV laser (P = 0.001). When all black dogs were removed from the data set, a significant difference was also noted between the 5 W and 0.5 W groups (P = 0.001). There was a significant difference between transmission values of white, brown, and black coats at both locations (all P < 0.001). Additionally, graphical representations of penetration values by the inguinal fold and calcaneal tendon width showed decreasing penetration measurements with increasing widths (P < 0.001, Figure 1A, B).

Table III.

Variables that contribute significantly to the emission of photons through tissue as median and range percentages.

Parameter	Median % transmission	Range	P-value
Class IIIb laser (904 nm)	0.0a	0.0 to 12.1	< 0.002
Class IV laser (810/980 nm)	0.34b	0.0 to 19.1
White coat	0.9a	0.0 to 19.1
Brown coat	0.3b	0.0 to 12.3	< 0.001
Black coat	0.0c	0.0 to 15.7
0.5 W	0.0a	0.0 to 15.4
1.5 W	0.4	0.0 to 17.1	= 0.001
3.0 W	0.5b	0.0 to 19.1
5.0 W	0.6b	0.0 to 17.3
Shaved coat	0.7a	0.0 to 19.1	< 0.001
Unshaved coat	0b	0.0 to 10.9

^a–cWithin each group, values with different letters are significantly different. The Durbin-Watson statistic was used to test for autocorrelations in the data. Residual analysis was used to assess normality and model fit. Homoscedasticity and linearity were assessed using a p-p plot. Tolerance and variance inflation factor (VIF) statistics were used for multicollinearity.

Figure 1.

Photon tissue penetration (mW) by A) inguinal fold and B) calcaneal tendon width at 3 W in unshaved dogs. Linear regression based on coat color as identified by key (white, brown, and black).

Discussion

One of the biggest challenges facing the advancement of laser therapy into mainstream medical practice is thought to be the conflicting recommendations for treatment parameters in the literature. There are few studies that investigate laser penetration through different types of tissue. Most of these are ex vivo through varying skin thicknesses (21–24). Mathematical models additionally describe the penetration of light in biological tissue (25). These studies each support the fact that laser penetrates only through a few millimeters of tissue. In this study design, the laser penetration over the inguinal fold likely best represents skin penetration, whereas over the calcaneal tendon best represents deep penetration. The results of this study indicate that, without proper preparation and settings, laser therapy may fail to deliver a therapeutic dose to the target tissue in dogs.

In this study, significant differences in laser transmission were identified among coat colors, with transmission decreasing as pigmentation increased. This finding is consistent with previous equine studies (19,20) that described greatest energy penetration with light-colored coats, although this has not been previously reported in dogs. Laser wavelengths used therapeutically are in the near-infrared or visible-red spectrum (600 to 1070 nm). This range is ideal for tissue penetration, as wavelengths shorter than 600 nm are increasingly absorbed by melanin and hemoglobin and wavelengths longer than 1070 nm are instead largely absorbed in water. Even within this therapeutic range, however, melanin significantly impacts absorbance, with a lesser effect observed at higher wavelengths (12).

Higher power levels (3 W, 5 W) significantly increased laser transmission, as did the use of a split or mixed wavelength laser compared to a fixed long-wavelength laser. Although the depth of photon penetration is mainly determined by wavelength, the precise effect of power has yet to be elucidated. Anecdotally, power is thought to govern photonic saturation at the target depth via an increased wattage at the treatment surface, although a recent study showed that penetration depth in biologic tissue did not depend on wattage (26). While higher power densities with shorter treatment times deliver total desired dosage (J) more efficiently in a laser treatment, they carry increased risk of thermal damage. Additional studies, potentially using implantable photosensors, would be required to assess laser transmission at different tissue depths.

Shaving significantly increased laser transmission, which is consistent with the results of previous studies, presumably due to decreased scatter and reduced photon absorption by melanin (19,20). Application of laser through a coat can result in reflection, hair absorption, and scatter of photons, which subsequently decreases or inhibits their delivery to the target tissue (27). Interestingly, coat length did not appear to significantly affect transmission. The study may have been underpowered to detect a significant difference, as several factors could explain these findings, including coat thickness, hair density, and how the hair lies against the skin. While it was noted during data collection whether or not an undercoat was present, it was difficult to reliably ascertain hair thickness and relative coat density from dog to dog. If the coats were long, but the hair was thin, sparse, and without an undercoat, less scatter of the laser light might have occurred. However, the backwards stepwise regression model convincingly suggests minimal effects of coat length on transmission. Although no differences were observed, further studies assessing coat length, hair density, and the diameter of hair shaft using a trichometer for objective data assessment as it relates to photon transmission are warranted due to weak trends observed in our data set. A future study should individually assess coat length, hair density, and hair shaft diameter objectively measured with a cross-section trichometer (28).

Previous equine studies similarly observed that clipping hair before treatment improved transmission compared to the unprepared state in vivo (19) and ex vivo (20). Differences were also noted related to the subjectively perceived pigmentation of the horses, with the greatest penetration occurring in more lightly pigmented coats. Coat color of the dogs in this study significantly influenced light penetration, with black and brown unshaved coats having a median value of 0 mW penetration at both the inguinal fold and calcaneal tendon. Median penetration values increased after shaving, with a significantly increasing trend from black dogs with the subjectively identified darkest underlying skin, to brown dogs, to white dogs. The effects of skin color were not investigated in this study, except for subjectively identifying that the darker coat colors had generally darker underlying skin as skin color was difficult to fully appreciate after shaving in many dogs.

The laser application and sensor measurement technique were generally well-tolerated. Increased patient reactivity to the laser treatment without gross appearance of thermal injury was noted at the 5 W setting with the class IV laser while obtaining early measurements in the black dogs after shaving. This was subsequently not assessed in the remaining black dogs for the shaved measurement at 5 W. Therefore, only the 3 W, 1.5 W, and 0.5 W with the class IV laser and class III laser measurements were obtained on the remaining black dogs after shaving had occurred. This could in part explain the differences in penetration among the coat colors after shaving, as well as the laser-elicited thermal sensitivity that warranted exclusion of the black dogs after shaving at the 5 W power setting. A possible explanation of this is that the laser energy was absorbed by melanin in the skin, which resulted in more rapid heat generation at the higher power or less probe movement than the constant movement that is recommended during therapeutic treatment. Additionally, the manufacturers of the class IV laser now recommend using a larger probe head when employing a power level greater than 3 W.

While tissue thickness was also not a specific parameter of interest in this study, the results of the backwards stepwise regression found that the thickness of both the inguinal fold and the calcaneal tendon impacted laser penetration. Measurements of penetration at the inguinal fold exceeded those at the calcaneal tendon. Clinically, it is thought that tissue thickness is an inconsequential variable as it would not be feasible to ultrasonographically measure the depth of the patient’s skin at the treatment site before calculating therapeutic laser dosage. Additionally, previous studies have found that the effect of skin thickness was minimal compared with the effect of pigmentation (20).

Another factor to consider in extrapolating the findings of this study is that the inguinal fold consists of laterally and medially facing layers of skin separated by loose connective tissue (29). Testing for laser penetration in this region was ideal for study design in vivo with minimal underlying tissue. However, the subcutaneous layer is likely thicker in this region compared with that overlying other potential treatment sites and a second layer of skin was therefore unavoidable. This should be viewed as a limitation of this study as therapeutic laser is not intended to penetrate 2 cutaneous layers and this cannot be estimated in the present study.

In human skin, laser penetration is significantly attenuated, specifically in the first 1 mm (23). Optical properties of tissue are extremely varied among in-vivo, ex-vivo, and in-vitro studies, however, as well as between species and tissue type (30). These results should be exclusively applied to dogs with the understanding that dosing may be dependent on local tissue matrices and epidermal thickness, which can vary depending on anatomical location. For example, the fibrous, collagen-rich tissue of the calcaneal tendon may have appreciably disrupted the photonic transmission and does not reflect other tissues such as muscle, fat, or nerve. Anecdotally, when examining a specific subset of dogs thought to be model candidates for laser penetrance (shaved white dogs using a laser power of 5 W), there was a 4-fold reduction in laser transmission through the calcaneal tendon versus the same thickness on average of the inguinal fold (data not shown). Although this was not the primary focus of this study, further studies looking at the specific influence of tissue type on laser penetration are warranted, ideally including transplanted thermopile sensors.

To the authors’ knowledge, the transmission differences by laser wavelength have not been independently studied in dogs. The class IIIb laser used in this study has a power level of 500 mW and was compared to the class IV laser set to the same power. The class IV laser showed significantly higher transmission than the class IIIb laser. Given that the power output of the 2 assessed laser units was equivalent, this discrepancy in transmission was potentially attributed to the different wavelengths, with the class IIIb at 904 nm and class IV energy divided between 810 nm and 980 nm.

The inability to evaluate the impact of the mixed wavelengths is a limitation of this study. The effect of wavelength on laser penetration is controversial with conflicting evidence. There may be a direct correlation between increasing laser wavelength and greater depth of penetration in a human skin model (31), as opposed to greater depth of penetration with shorter wavelengths in bovine tissue samples (32). Further investigation of the effects of wavelength on laser transmission is warranted, ideally using a sensor with the ability to detect both photons and wavelength.

Another limitation of the present study was the variety of dog breeds studied, which resulted in a myriad of coat types and differences in pigment. Although the investigator examined each dog for general health and coat qualifications before inclusion in the study, a variety of dog breeds was accepted, each with potentially different hair densities in order to obtain sufficient participation in the study. A follow-up study that controls for this inherent variation could use exclusively 1 dog breed with multiple coat phenotypes, i.e., Labrador retrievers, and use a trichometer to objectively measure coat characteristics. Even with this potential improvement in study design, the subtle differences in subjective coat color within general pigment groups could still introduce variability in the data. A previous equine study, however, showed that subjective perception of the skin color was adequate to reflect measured skin pigmentation via a spectrophotometer (20).

As this investigation was carried out in healthy dogs, another limitation to clinical application of its findings is the possible difference in laser penetration in diseased or inflamed tissue. The presence of cytokines or proteinaceous edema, either superficial to or within the target tissue, may modify photon penetration or optical properties and subsequent response to laser therapy. Additionally, while the results of this study provide information regarding laser penetration and transmission as a factor of power output, it cannot be concluded that all laser energy that attenuates with the target tissue is absorbed (33).

The backwards stepwise regression equation, although likely impractical for clinical use, indicates that, overall, higher doses may be needed than those frequently recommended by laser manufacturer algorithms. Higher dosages than previously reported, which were used in recent clinical canine research, have been shown to be effective (34). In an effort to standardize treatment protocols for consistency between dogs, a simplified algorithm that takes into account the effects of laser power and patient pigment is necessary to provide effective treatment of laser light to the tissue of interest. This was not entirely achievable with the results of this study given the 2 cutaneous layers and lack of data on overall depth of photon penetration. However, such a simplified algorithm is considered to be an important next step in laser research and clinical use.

The results of this study strongly suggest that darker pigmented dogs should be shaved in order to achieve photon transmission to the subcutis and that black-coated areas on dogs may not be ideal areas for treatment due to lack of this transmission. Black-coated areas of treatment must be shaved and should be monitored regularly during treatment for epidermal thermal injury due to heat generation. Using brisk probe movements and power settings below 5 W should eliminate this risk. In addition, transmission through tissue with a heavy collagen matrix is feasible, making it possible to deliver photons to areas in close proximity to the subcutis. Future studies using higher dosing are warranted. As approximately 20% of veterinary hospitals in the United States offer laser therapy (35), these findings can assist in the formulation of both an individualized laser treatment in the clinical setting and also in the design of future research investigations. Pending further research, shaving of target areas, especially in nonwhite dogs, is strongly recommended to ensure adequate photon delivery.