Introduction

Osteoarthritis (OA) is a common condition in both dogs and cats.^1,2 In patients with OA, the primary therapeutic focus of medical management is on pain control, with nonsteroidal anti-inflammatory drugs being the most commonly prescribed medication.³ However, in recent years there has been an increasing interest in alternative therapies for the treatment of OA pain. One such alternate is low-level laser therapy (LLLT).

The purported biologic effects of LLLT include activation of respiratory chain enzymes (cytochrome C oxidase), oxygen production, formation of proton gradients across cell and mitochondrial membranes, adenosine triphosphate production, deoxyribonucleic acid production, cell proliferation, reduced cyclooxygenase, and prostaglandin E2 production.^4–10 The generation of these biologic effects in veterinary patients are dependent on laser power and energy delivery to the tissues.¹¹ In the clinical setting, many factors such as hair coat, hair and skin color, and depth of the site to be treated can reduce the laser energy delivered to the target tissues.^10,12 However, uniform treatment recommendations and standardized treatment protocols for varying conditions are lacking in the veterinary literature.^10,13

Recently, a study found that 79% of veterinarians refer patients for rehabilitation therapy. Of those respondents who refer patients for rehabilitation, 59% referred patients with OA for rehabilitation therapy in general, whereas 28% specifically referred patients for laser therapy.¹⁴ Despite the frequent use of LLLT for the treatment of dogs with OA, there is a paucity of evidence supporting its efficacy. Treatment and dose recommendations for LLLT use in dogs with OA is inconsistent.^10,13,15,16 There are no veterinary clinical trials evaluating the efficacy of LLLT for the treatment of OA.

The objective of this study was to evaluate the use of LLLT for the treatment of OA in dogs by Missouri veterinarians and to document aspects of general LLLT use in small animal patients in Missouri veterinary hospitals. We hypothesized that OA would be the most commonly treated condition and that training level and OA treatment protocols would be inconsistent.

Materials and Methods

This e-mail-based survey was designed with a targeted population of Missouri veterinary practices. At the time the survey was initiated, there were 818 licensed veterinary facilities in the state of Missouri.¹⁷ With the aid of internal resources at the University of Missouri, as well as targeted internet searches, 867 e-mail addresses were located for veterinarians at these licensed veterinary facilities in Missouri. An electronic survey was developed to investigate the use and economic impact of LLLT for the treatment of OA in dogs and cats. Prior to study initiation, survey validation was performed through an initial pilot survey that was developed and deployed to veterinarians within the authors’ veterinary facility familiar with LLLT. That questionnaire formed the basis of the completed survey in this study. The finalized survey (supplementary Appendix 1) was revised based on their comments and approved by the institutional review board of the University of Missouri. It contained a total of 19 questions that included multiple-choice questions, polar questions (yes or no), and 3 open-ended question. The survey was designed and distributed and data were collected through use of commercial software^a. Clopper and Pearson analysis was used to calculate the 95% confidence intervals for binomial proportions^b. The target population of the survey was Missouri veterinary practices with an LLLT unit and where LLLT treatment was being performed on small animal patients (e.g., dogs and cats) for the treatment of osteoarthritis. The target population was isolated through questions within the survey. Respondents were asked to provide geographic information (state and county) regarding the location of their veterinary practice. A response indicating the practice was not in Missouri resulted in survey conclusion. Respondents were then asked to provide information regarding LLLT unit ownership and use for the treatment of small animal patients. For respondents without an LLLT unit or who do not treat small animal patients with LLLT, questions regarding referral for LLLT were supplied and then the survey concluded. For respondents with an LLLT unit who treat small animal patients, information regarding LLLT unit class, model, and energy source were collected. Additional treatment information (training level, treatment personnel, conditions treated, and patient preparation) was collected. For respondents that indicated OA as a treated condition, additional information was collected (treated joints, treatment frequency and duration, number of patients treated, and treatment cost).

The survey was initiated and responses collected between November 16, 2016, and January 17, 2017. An e-mail explaining the content and purpose of the survey was sent to all e-mail addresses along with a link that directed the respondent to the online survey. Restrictions within the survey software^a were in place to prevent multiple responses from a single veterinarian. Respondents were asked to only complete one survey per hospital in practices with more than one veterinarian. Completed survey results were tabulated and data exported for evaluation^a,c. Surveys were considered to be complete when respondents had navigated through all questions on the survey. To provide the greatest response rate, respondents were not required to provide an answer for all questions in order to advance through the survey.¹⁸ The results from the completed surveys were evaluated and descriptive statistics were produced using computerized software.^c

Results

In this study, 867 electronic surveys were distributed by electronic mail and 89 were completed. This equates to a 10% response rate (89/867; confidence interval: 8–12%; margin of error: 2%). The average time for survey completion was 4 min (range: 0.1–44 min). The Missouri county in which the clinic is located was provided in 99% (88/89) of completed surveys. Completed surveys were obtained from 35% (40/115) of Missouri counties. The largest number of completed surveys came from St. Louis County (14%; 12/88), Boone County (9%), St. Charles County (9%), Franklin County (6%), Greene County (6%), and St. Louis City (5%). Respondents from Callaway, Christian, and Jefferson Counties each made up 3% of completed surveys. Respondents from Cole, Jasper, Jackson, Lafayette, Morgan, and Saline Counties each made up 2% of completed surveys. Respondents from each of the remaining Missouri counties made up ≤1% of completed surveys (Figure 1).

View Figure

• Download as Powerpoint
• Download Figure

Almost one-half of Missouri veterinary hospitals surveyed (43%; 38/89) responded that they had an LLLT unit. All of these respondents (100%; 38/38) used LLLT for the treatment of small animals. In the 57% of hospitals without an LLLT unit, 20% (10/50) referred small animal patients to other hospitals for LLLT treatment. For those hospitals that indicated use of LLLT for small animal patients, additional survey questions were provided.

Information regarding LLLT training was provided in 100% (38/38) of surveys completed by respondents using LLLT for treatment of OA in small animals. Survey respondents provided a total of 69 responses. The majority of respondents indicated that training was received in-house from a representative of the LLLT unit manufacturer (76%; 29/38) or through an off-site continuing education (CE) course (58%). Other methods of instruction in the use of LLLT included training by a coworker (18%), self-taught or self-directed learning (16%), through online CE training (5%), during rehabilitation certification training (3%), or during an internship or residency (3%). A single method of training was indicated by 37% (14/38) of respondents. Of these, 64% (9/14) received in-house training from a representative of the LLLT unit manufacturer, 14% through an off-site CE course, 14% were self-taught, and 7% received training from an online CE course.

Information regarding the LLLT unit manufacturer was provided in 87% (33/38) of completed surveys from respondents using LLLT in small animals. The majority of respondents (58%) used an LLLT unit produced by one manufacturer^d. Respondents also used LLLT units produced by other manufacturers in (27%)^e, (6%)^f, (3%)^g, (3%)^h, (3%)ⁱ, and (3%)^j.

Information regarding laser class was provided in 100% (38/38) of completed surveys from respondents using LLLT in small animals. The most commonly reported class of laser was class 4 (68%; 26/38), followed by class 3 (8%) and 3b (8%), then class 2 lasers (3%). Laser class was reported as “unknown” in 13% of surveys. Information regarding energy source was provided in 97% (37/38) of completed surveys. The energy source (electrons and photons) of the laser was unknown by a majority of respondents (95%; 35/37). For the 5% of respondents that did know the energy source, it was equally divided between helium–neon (2.5%) and gallium–aluminum–arsenide (2.5%).

Recommendations for preparation of the area to be treated with LLLT are variable and include no preparation, parting the hair, clipping the hair, or clipping the hair and cleaning the skin with alcohol.^10,12,16 Information regarding treatment site preparation was provided in 97% (37/38) of completed surveys from respondents using LLLT in small animals. A single response was required for this question, although the option “other” with a blank text box was offered. Lack of treatment site preparation was reported by the majority of respondents (84%). For respondents that did perform treatment site preparation, 11% parted the hair and 3% clipped the hair. One respondent (3%) selected “other” with the explanation of “dark fur is wetted.” Hair clipping followed by skin cleaning with alcohol prior to LLLT treatment was not reported by any respondents.^10,12,16

Information regarding who administers the LLLT treatments was requested. Options provided included veterinarian, licensed or registered veterinary technician, veterinary assistant or unlicensed veterinary technician, and “other” with a blank text box. No responses were recorded for this question.

Information regarding medical conditions treated with LLLT was provided in 100% (38/38) of surveys completed by respondents that use LLLT in small animals. Survey respondents provided a total of 179 responses. All survey respondents were asked to provide information on all medical conditions treated with LLLT. The most common medical condition treated with LLLT was osteoarthritis (97%; 37/38). Respondents also reported treating the following conditions with LLLT: postoperative wounds (90%), chronic open wounds (92%), tendon/muscle injuries (90%), and intervertebral disc disease or neuropathic pain (74%). Other conditions treated with LLLT therapy reported by respondents included otitis (8%), urinary (urinary obstruction, feline interstitial cystitis, cystitis; 8%), dermatologic/pyoderma (5%), resistant infections (3%), aural hematomas (3%), anal sacculitis (3%), post dental procedures (3%), and miscellaneous inflammatory conditions (3%).

Information regarding joints treated with LLLT was provided in 100% (37/37) of surveys in which OA was listed as a managed condition. Survey respondents provided a total of 231 responses regarding which joints were treated. The most commonly treated joints were the stifle (97%; 36/37) and hip (95%). Other joints receiving LLLT treatment for OA included the elbow (87%), spine (92%), shoulder (81%), carpus (68%), tarsus (49%), front limb digits (24%) and hindlimb digits (27%), and other unspecified joints (5%).

Information regarding treatment frequency for OA joints was provided in 100% (37/37) of completed surveys. The frequency of LLLT for OA joints was most commonly reported as either two or three times per wk (38% and 46% respectively). The remainder of respondents reported treating OA joints one time per wk (14%) or five or more times per wk (3%). The duration of LLLT treatment for OA joints was most commonly reported as 4 wk (51%), followed by 2 wk or less (32%), 8 wk (8%), 6 wk (5%), and 12 wk or more (3%).

Information regarding treatment dose was provided in 97% (36/37) of surveys. For this question, respondents were asked specifically about the treatment of an osteoarthritic joint in a large dog. Most practitioners (64%; 23/36) reported using their machine’s predetermined settings instead of a user-calculated dose. Among those surveys that did report specific dosages used, wide variation existed: <2 J/cm² (3%), 2–3 J/cm² (11%), 4–5 J/cm² (6%), 6–7 J/cm² (6%), 8–10 J/cm² (8%), and >11 J/cm² (3%).

Information regarding the average number of dogs treated weekly for OA was provided in 92% (34/37) of surveys and was reported to be four patients (range: 0.5–12). The average cost of a single LLLT treatment for a single OA joint was provided by 100% of respondents and was $28 (range: $15–45).

Discussion

The hypotheses in this study were supported. Osteoarthritis was the most commonly treated condition, and both the level of user training and treatment protocols were highly variable. This was expected owing to the lack of standardized treatment protocols and recommendations available to veterinarians.

Approximately half of the veterinarians surveyed had an LLLT unit. Of the clinics that do not have an LLLT unit, 20% referred small animal patients with OA for LLLT therapy. This is similar to a previous study in which 27.8% of respondents referred patients for laser therapy.¹⁴ That study showed that veterinarians in the Midwest United States do not refer patients for rehabilitation and physical therapy, which may include LLLT, as frequently as veterinarians in other areas of the United States, likely because of perceived cost and limited availability.¹⁴ However, the results of our study suggest that referral for LLLT therapy might only appear low as ownership of an LLLT unit is not uncommon.

The economic impact of LLLT in Missouri was shown to be substantial. Respondents reported that an average of four patients were treated per wk for OA and the majority of these patients were treated three times per wk. Therefore, the average Missouri veterinary hospital performing LLLT in small animals with OA is estimated to perform a total of 12 LLLT treatments per wk. At a cost of $28 per joint (range: $15–45), this results in an estimated average income of $336 per wk for treatment of a single OA joint in a Missouri veterinary hospital. Using the information collected in this survey, a full course of LLLT treatment for a single OA joint would cost $252. If patient numbers and treatments are consistent throughout the year, an average Missouri veterinary practice would gross $17,472 per year from LLLT treatment of a single OA joint. At the time of the study, there were 818 registered veterinary practices in Missouri, according to the Missouri Division of Professional Registration. Using data obtained in this study, if 43% of Missouri practices (352 clinics) own and use LLLT for the treatment of small animal OA (four patients per wk receiving three treatments per wk), then Missouri dog and cat owners are estimated to spend roughly $6.2 million annually on LLLT for the treatment of OA in their pets.

To the authors’ knowledge, there are no veterinary studies evaluating the economic impact of laser therapy. Therefore, a direct comparison to other states in the United States or other countries cannot be made at this time. The most comparable study evaluated the economic impact of cranial cruciate ligament injury in dogs in the United States, which is estimated at $1.32 billion.¹⁹ The cost of medications for the treatment of arthritis in the United States in 2005 was $130 million per year, with an annual increase of 13% yearly.²⁰ A study evaluating the economic impact of osteoarthritis in horses found an average annual cost of $10,000 per patient.²¹ In humans, osteoarthritis accounts for an estimated $185.5 billion dollars of insurer expenditures in the United States.²² A Canadian study revealed an average of $9,882 per person due to osteoarthritis-related care.²³ Similar studies are not available in veterinary medicine. Given these numbers and the prevalence of osteoarthritis in dogs, it is not surprising that up to $6.2 million are spent yearly for the treatment of OA with LLLT in a single Midwestern state. Additionally, this is the first report of demographics and economics related to LLLT for the treatment of OA in small animals. These results may not be reflective of other states or regions of the country, owing to regional economic and population demands that may contribute to LLLT use and availability in this diverse Midwestern state.

The energy source of the LLLT unit was unknown by 95% of respondents. Knowledge of and a thorough understanding of a laser’s energy source are critical for the appropriate application to clinical conditions. The energy source can affect the laser’s wavelength and have an effect on light penetration. For example, gallium–arsenide and gallium–aluminum–arsenide lasers have longer wavelengths and therefore scatter less in superficial tissues, making them able to penetrate tissues more effectively than lasers with shorter wavelengths, such as helium–neon lasers.^10,16 Depth of penetration is an important consideration when treating various conditions. For example, a study evaluating the depth of penetration of an 850 nm gallium–aluminum–arsenide laser through human skin showed that the majority of laser energy was absorbed within the first millimeter of skin.²⁴ Treatment of an osteoarthritic joint requires laser penetration through the skin and subcutaneous tissues to reach the joint capsule, synovium, and cartilage, where the laser energy is thought to have the most effect.^16,25 Considering that the average skin thickness in a Labrador retriever is approximately 5 mm, it is important to consider whether the target tissues are being reached.²⁶ Additionally, the amount of tissue overlying joints and subsequent laser penetration depth can differ significantly depending on anatomical site (e.g., tarsal joint versus coxofemoral joint). However, LLLT treatment of an open wound with no overlying tissue only requires direct laser application to reach the therapeutic target.¹²

Laser class is another important consideration that can affect energy delivery to tissues as well as have safety implications for the patient and operator. Laser class is determined by the potential to damage tissues and correlates with energy production (wattage), with class 4 being the most likely to cause tissue damage. The largest risk to the operator and the patient with improper laser use outside of the immediate treatment area is permanent retinal damage.^10,13 In this study, most LLLT units were class 4 lasers (68%). and it is recommended that laser application occur in an overlapping grid fashion to avoid heating the tissues.¹⁰ Class 3 and 3b lasers, that represented 8 and 5% of LLLT units in this study, respectively, do not produce heat, and if heating of the tissues is appreciated, it indicates a potential malfunction of the unit.¹⁰ Class 3 and 3b lasers also possess the ability to cause retinal damage, and proper eye protection is recommended with these, as well as with class 4 lasers.^10,16 Owing to differing properties between LLLT units of different classes, knowledge of the class of laser currently used to deliver in-hospital therapy is critical to ensure optimum patient and operator safety. Interestingly, 15% of the survey respondents did not know the class of laser in use at their hospital.

Preparation of the treatment site is an important consideration. In this study, the majority of hospitals (84%) reported performing no treatment site preparation prior to LLLT application, despite the current clinical recommendation for site preparation.^10,13,16 Only 2.5% performed hair clipping prior to LLLT therapy. However, it has been demonstrated that hair reduces the depth of penetration of the therapeutic laser by 50–99%.¹⁰ Additionally, a recent study showed that hair clipping and cleaning the skin with alcohol prior to therapy allows the laser light to reach deeper tissues.¹² This method of site preparation was not performed by any hospital in this survey. Theoretically, coat and skin color should affect the depth of laser penetration because of the ability of melanin to absorb shorter wavelengths of light (less than 830 nm), such as those produced by helium–neon lasers. However, an equine study found no significant difference in depth of penetration of laser energy in horses with lighter (gray) and darker (bay) colored hair coats.¹² Interestingly, in that study a class IIIb laser was used, with a wavelength of 810 nm, which would be more affected by melanin content in the skin or hair. Currently, there is a paucity of information available regarding the effect of treatment site preparation in animals prior to LLLT. What little information is available is conflicting, and the studied laser class and energy source are inconsistent.^{10,12,16,25,27–30} Additionally, within the companion animal field, there is a diverse array of skin and coat colors and hair types and lengths among animals, both within and among breeds. These differences have the potential to greatly affect LLLT. Additional study in this area is warranted.

Training on and knowledge of any therapeutic instrument prior to initiating clinical use is critical to ensure the best outcomes for and the safety of patients. In this study, the majority of respondents (95%) indicated some type of formal instruction, while 5% were self-taught. Interestingly, training during advanced postgraduate programs such as an internship, residency, or rehabilitation and physical therapy certification programs was only included as a source of LLLT training by 7% of respondents. This is likely reflective of the degree of advanced postgraduate training within the sample population. However, in the authors’ opinion, the frequent use of LLLT in veterinary practice and the inconsistency in treatment protocols, as demonstrated in this study, serves to emphasize the need for improved research as well as education about this modality during veterinary school. In this study, direct questions regarding LLLT training during formal veterinary school education were not asked. However, a previous study noted that more than 60% of the accredited US veterinary schools provide no course work in physical therapy and associated therapeutic modalities commonly used during physical therapy and rehabilitation, such as LLLT.¹⁴

Interestingly, information regarding who the primary person was performing LLLT treatment on patients in the hospitals surveyed was not provided in any completed survey. Respondents were provided with several options, which included veterinarian, licensed or registered veterinary technician, veterinary assistant or unlicensed veterinary technician, or other. Only a single response could be recorded for this question; however, the choice of “other” provided the option of a short answer. The reason for uniform exclusion of this information is unknown.

Study limitations include a low survey response rate of 10%. However, this response rate is similar to other studies and reflects the general downward trend in response rates to surveys.^14,31 The survey completion rate may have been improved with direct mailing of surveys, offering financial incentives for survey completion or performing follow-up phone communication to clinics where no response was obtained. Response bias must also be considered in a study of this nature, as those who regularly perform LLLT may be overrepresented given their interest in the treatment modality. Question structure is also a consideration, and it is important to note that not all questions required an answer to advance through the survey and some questions allowed multiple response. By not mandating answers, it is possible that some completed surveys had missing data points; however, it has been shown that mandating answers will result in respondents not completing the survey or providing false information in order to advance.¹⁸ Also, surveys were requested to be completed by one veterinarian per practice, which was confirmed by verifying that there were no duplications in internet protocol address. The objective of the survey was outlined and respondents were directed to provide answers that reflected the veterinary practice. However, in multidoctor practices where there are differences between veterinarians, those differences may not be fully represented. Additionally, it should be noted that financial estimations are based entirely on data extrapolated from this study, as no previously published information was available at the time of this publication.

Although it is outside the scope of this study, an important point of consideration is the efficacy of LLLT for the management of OA. Owing to the lack of veterinary studies, the majority of LLLT treatment recommendations are extrapolated from human studies and in vitro data. Clinical studies evaluating the efficacy of LLLT for the treatment of OA in small animals is warranted, given the prevalence of LLLT use and the financial impact it represents.

Conclusion

Overall, this study provides information regarding LLLT use and economic impact in a Midwestern state within the United States. In our study, LLLT was most commonly used for the treatment of OA in dogs, despite limited data to support its efficacy in the treatment of OA pain. Additionally, this study demonstrated inconsistency in treatment protocols and level of training, which is highlighted by the fact that 84% of practitioners did not perform any treatment site preparation prior to LLLT treatment. The data presented here also suggests that the economic impact of LLLT for the treatment of osteoarthritis in Missouri is substantial, with up to $6.2 million dollars of estimated revenue per year. Given the common use of LLLT in Missouri practices for the treatment of OA and the significant economic impact, these results show that further research is needed to determine the efficacy of the technique and define treatment recommendations and protocols.