Effects of Light Therapy on Osteoarthritis and Its Sequelae in Aging and Older Adults

A Systematic Narrative Review

Abstract

Objective: 

To assess the effectiveness of therapeutic phototherapy including laser, light-emitting diode, and supraluminous diode energy, in the management of osteoarthritis in aging and older adults.

Methods: 

A systematic review of PubMed and CINAHL was completed using the following search terms and their associated abbreviations: laser, supraluminous diode (SLD), or light-emitting diode (LED) combined with therapy, intervention, or treatment. We included randomized controlled trials published between January 1, 2008, and December 31, 2018, written in English, and included human subjects who had a central tendency for age greater than or equal to 50 years.

Results: 

Twenty-two randomized controlled trials met our search criteria. Multiple pathologies were evaluated including knee osteoarthritis (n = 20), total hip arthroplasty (n = 1), and hand osteoarthritis (n = 1). Of these 22 studies, 2 compared laser to control, 8 compared laser to sham or placebo, 1 compared laser to therapeutic exercise, 1 compared laser with exercise to control, 3 compared laser paired with exercise to exercise only, 4 compared laser with exercise to placebo with exercise, 3 compared laser to other modalities, 3 compared laser with other modalities to other modalities, 1 compared laser with other modalities to exercise with other modalities, 1 compared laser with exercise and glucosamine/chondroitin sulfate (GCS) to GCS and exercise as well as placebo and exercise, 1 compared laser intensities, and 2 compared laser used with ultrasound (US) to placebo as well as combination laser/US with exercise to placebo and only the combination laser/US unit. As demonstrated by this variability, a meta-analysis was not appropriate. Studies on knee osteoarthritis varied slightly in their results, but the majority of groups that received laser showed significant improvements in pain, function, and quality of life when compared with placebo or control. High-intensity laser demonstrated larger improvements when compared with low-level laser therapy. Histological studies found significant changes in some of the markers associated with tissue healing favoring the laser treatment groups.

Conclusion: 

The studies included all used a form of laser in their light therapy treatment protocols. There is some evidence that light therapy may provide statistically significant improvements in pain, function, and quality of life; however, results varied from study to study. Further research is needed to better evaluate the parameters associated with laser when used to treat conditions associated with osteoarthritis that affect aging and older adults.

There are various types of light therapy that are used in physical rehabilitation settings. This includes light-emitting diodes (LEDs), supraluminous diodes (SLDs), and lasers. One type of light-based treatment available to practitioners is laser therapy, especially low-level laser therapy (LLLT), which may be used to treat many different conditions. The use of therapeutic lasers was originally pioneered by Endre Mester for the treatment of chronic and recalcitrant wounds. Since his original work, the use of LLLT has been proposed for diagnoses ranging from bacterial infections to arthritis and many other pathologies a clinician may encounter.¹

The type of laser used in the clinical setting in the United States is a Class IIIb, which is designated to have a power output of 5 to 500 mW, sometimes called LLLT. One mechanism to create this beam of energy is by passing an electrical current through a gas. This electrical current causes electrons in the gas to jump to a higher energy level. A photon is released when these electrons fall from the higher energy level to a lower one. These photons will bounce from mirrors on one end of the unit to the other and continue to impact the gas atoms. As they bounce back and forth, they release identical photons from additional atoms and then combine to eventually escape through the semi-mirrored end of the tube. Safety precautions are needed with Class IIIb lasers because they can cause permanent eye injuries with brief exposure and minor skin burns with prolonged exposure.²

More commonly now, photodiodes are used to create the laser. These diodes consist of 2 types of materials, one that is more positively charged and the other that is more negatively charged. When the electrons fall from the negatively charged material to the positively charged material, the photons are emitted, creating the laser beam.² The photons emitted will exit at different wavelengths that are determined by the medium used. LLLT wavelengths range from 600 to 904 nm. Different gases that create different wavelengths (in nm) are used including helium-neon lasers (632.8 nm), gallium-aluminum-arsenide (860-780 nm), and gallium-arsenide (904 nm).¹ Lasers with wavelengths of 600 to 780 nm have a penetration depth less than 1 cm and those with wavelengths of 780 to 904 nm penetrate less than 5 cm.¹

The wavelength of the laser unit not only plays into the depth of penetration, but also targets specific chromospores, the light-absorbing parts of a molecule that give it color by absorbing specific wavelengths. It is suggested that the physiological and therapeutic effects of LLLT are based upon the chromospores of different tissues and that the absorption of the light energy by the chromospores increases the oxidative metabolism in mitochondria.¹ The increased output of the mitochondria is proposed to lead to analgesia, anti-inflammation, and increased protein collagen synthesis to promote soft tissue healing.¹ In conjunction with wavelength and power, other parameters of laser units include the power density (mW/cm²), energy (J), and energy density (J/cm²). It is helpful to document these parameters so that the treatment provided is most accurately described.¹ Furthermore, it is considered best practice to document the specific application area(s) and treatment technique (eg, skin contact vs noncontact and stationary vs manual scanning motion).

The use of lasers has been proposed for many diagnoses. A primary use of LLLT in the physical rehabilitation setting is for soft tissue and bone healing that originally started from seeing positive healing effects for wounds when lasers were first started to be used clinically; more recently, it has been used for osteoarthritis (OA). LLLT is also thought to improve postmastectomy lymphedema, neurological conditions like carpal tunnel syndrome, diabetic peripheral neuropathy, postherpetic neuralgia, and potentially after surviving a stroke (if applied within 24 hours).² The final, and important, impact of lasers is the proposed effects on reducing pain associated with neuromusculoskeletal conditions ranging from back or neck pain, lateral epicondylitis, trigger points, and delayed muscle onset soreness.²

Other forms of light therapy include LEDs and SLDs, both of which are semiconductors that create light when electrical currents are passed through them. The light created by LEDs is of lower power (1-5 mW) and is less focused than light being emitted from SLDs (5-35 mW), and lasers (5-500 mW for Class IIIB lasers).² The energy beam from lasers is the most focused of the 3 light sources; SLD is more focused than LED but less than laser. Due to lower intensities, multiple LEDs are included in the applicator and the treatment time is typically longer.² LED therapy applicators typically target a larger treatment area and offer a wider frequency range that is more suitable for treating superficial tissues.² The total power of the LED applicator is determined by the number of individual LEDs of that specific applicator. SLD devices offer a smaller range of available wavelengths than LEDs, but not as limited as lasers, and are best suited for treating superficial to moderately deep tissues.² As our search for this review focused on OA, it is expected that the majority of the studies used laser as their primary light source because of the depth of the joint tissues that are typically associated with OA. Some units combine 2 or 3 types of light devices to be used simultaneously during the treatment session to target superficial and deep tissues and to aid the user in aiming the phototherapy beam, as light from the laser will not be seen if appropriate safety equipment is being used.

For this narrative systematic review, we searched randomized controlled trials (RCTs) that used lasers, supraluminous diodes (SLD), or light emitting diodes (LED) in the treatment of OA in older adults. The majority of the articles identified focused on OA of the knee. We choose to also include high-intensity laser therapy (HILT) as there is growing utilization in countries outside the United States. These lasers are considered Class IV lasers that have a power greater than 500 mW and a typical wavelength of 1064 nm.³ Due to the increased power and longer wavelength, they are typically pulsed to moderate the total energy applied to the body.

METHODS

Search strategy and study design

A systematic review of PubMed and CINAHL was completed using the following search terms and their associated abbreviations: laser, supraluminous diode (SLD), or light-emitting diode (LED) combined with therapy, intervention, or treatment. We included RCTs published between January 1, 2008, and December 31, 2018, written in English, and included human subjects who had a central tendency for age greater than or equal to 50 years.

After using the provided search terms and screening titles and abstracts for relevance to geriatric rehabilitation, 30 articles were identified. Of these, 3 were systematic reviews,^4–6 1 was a case study,⁷ 2 were descriptive studies,^8,9 and 24 were randomized controlled trials (RCTs). The reference lists from systematic reviews were scanned for additional resources and no new RCTs were identified. We chose to focus our review on the RCTs; thus, the systematic reviews, case study, and descriptive studies were excluded. Two RCTs were excluded since they did not meet our mean, median, or mode age requirement of 50 years or more.^10,11 Therefore, 22 RCTs were included in our systematic review.^12–33 A flow diagram of our article selection process can be seen in Figure 1.

Figure 1.: Article selection process.

Physiotherapy Evidence Database (PEDro) scores, which have demonstrated validity³⁴ and reliability,³⁵ were calculated for each of the 22 RCTs that were reviewed for methodological quality. Two authors independently scored each RCT using the PEDro scale; when there was disagreement, the third author was used to reach a consensus. Similar to another systematic review performed by Laufer and Dar,³⁶ we interpreted the PEDro scores using the following convention: 9 to 10 is excellent, 6 to 8 is good, 4 to 5 is fair, and less than 4 is poor. For the purposes of this review, PEDro score was not used as an exclusion criterion.

RESULTS

The 22 RCTs included in this systematic review were initially clustered based on the type of physical agent used: LLLT (n = 16), HILT (n = 3), LLLT and HILT (n = 1), and LLLT combined with ultrasound (US) (n = 2). Due to the overlapping nature of research designs used, the 22 RCTs were then organized based on unique pairs of comparison groups. This organization strategy identified 13 different group comparisons and allowed the authors to further cluster studies that had similar research questions into 20 subgroups (Table 1). Several studies included more than 2 groups and were therefore added to multiple summaries of group comparisons. The largest comparison group examined laser versus sham/placebo and synthesized 8 studies that used LLLT, HILT, or laser applied at acupuncture points. There were 6 different comparison groups that only included a single study in its summary. Of note, the 2018 study by Alfredo et al¹⁵ was a long-term follow-up study of the same participants from their study in 2011.¹⁴ The characteristics of each RCT are summarized in Table 2, including demographics, comparison groups, diagnostic criteria, physical agent used and treatment parameters, outcome measures, and main findings.

TABLE 1 - Article Groupings by Research Question

Group	Subgroup(s)	Authors	PEDro Score or Mean Score and Range
Laser vs control	LLLT vs control (no treatment)	Barabás et al18 de Matos Brunelli Braghin et al19	Range: 3-6; mean: 4.5
Laser vs sham/placebo	LLLT vs sham/placebo	Alghadir et al16 Fukuda et al23 Langella et al27 Nambi et al29 Youssef et al33	Range: 5-10; mean: 8.2
	HILT vs sham/placebo	Angelova and Ilieva17	4
	Laser applied to acupuncture points vs sham/placebo	Al Rashoud et al12 Mohammed et al28	Range: 6-6; mean: 6.0
Laser vs exercise	LLLT vs exercise	de Matos Brunelli Braghin et al19	6
Laser + exercise vs control	LLLT + exercise vs control (no treatment)	de Matos Brunelli Braghin et al19	6
Laser + exercise vs exercise	LLLT + exercise vs exercise	de Matos Brunelli Braghin et al19 de Paula Gomes et al22	Range: 6-9; mean: 7.5
	HILT + exercise vs exercise	Nazari et al30	6
Laser + exercise vs placebo + exercise	LLLT + exercise vs placebo + exercise	Alfredo et al14,15 de Paula Gomes et al22 Kheshie et al25	Range: 7-9; mean: 7.7
	HILT + exercise vs placebo + exercise	Kheshie et al25	7
Laser vs other modalities	LLLT vs other modalities	de Oliveira Melo et al20 de Oliveira Melo et al21	Range: 7-7; mean: 7.0
	HILT vs other modalities	Kim et al26	4
Laser + other modalities vs other modalities	LLLT + other modalities vs other modalities	Ip24 de Oliveira Melo et al20 de Oliveira Melo et al21	Range: 5-7; mean: 6.3
Laser + other modalities vs exercise + other modalities	HILT + other modalities vs exercise + other modalities	Nazari et al30	6
Laser + GCS + exercise vs GCS + exercise	HILT + GCS + exercise vs GCS + exercise	Alayat et al13	6
Laser + GCS + exercise vs placebo + exercise	HILT + GCS + exercise vs placebo + exercise	Alayat et al13	6
Laser intensities compared	LLLT + exercise vs HILT + exercise	Kheshie et al25	7
Combo LLLT + US	Combo (LLLT + US) vs placebo	Paolillo et al31 Paolillo et al32	Range: 3-4; mean: 3.5
	Combo (LLLT + US) + exercise vs placebo	Paolillo et al31 Paolillo et al32	Range: 3-4; mean: 3.5
	Combo (LLLT + US) + exercise vs Combo (LLLT + US)	Paolillo et al31 Paolillo et al32	Range: 3-4; mean: 3.5

Abbreviations: GCS, glucosamine/chondroitin sulfate; HILT, high-intensity laser therapy; LLLT, low-level laser therapy; US, ultrasound.

TABLE 2 - Summary of Articles Reviewed

Author Country of Origin	Study Demographics (N): Gender, Age (Mean ± SD and/or Range), y Group Demographics (n): Gender, Age (mean ± SD), y	Body Region/Condition Diagnostic and Inclusion Criteria	Laser Manufacturer (Model) and Parameters	Outcomes Measures	Main Findings
Al Rashoud et al,12 England	N = 49, both genders, age range 54 ± 10 Group 1—active laser group (n = 26), age 52 ± 9 Group 2—placebo (n = 23), age 56 ± 11	Knee OA 1. ACR criteria 2. Pain VAS ≥3 cm 3. Ability to perform all movements in evaluation forms	Enraf Nonius (G22 Endolaser 476) Wavelength: 830 nm Power: 30 mW Power density (mW/cm2): Energy: 6 J Number of points: 5 Energy per point: 1.2 J Energy density: 4 J/cm2 Treatment frequency: 9 sessions	• VAS for pain • SKFS • Active ROM • Patient satisfaction	Statistically significant within-group improvements in VAS were found for the active laser group at all assessment periods. Placebo within-group improvements in VAS were statistically significant at all assessment periods except at 6 mo. Statistically significant within-group improvements in SKFS were found for both groups at all assessment periods. Statistically significant between-group improvements in VAS, SKFS, active ROM, and patient satisfaction were found at 6 mo favoring active laser treatment.
Alayat et al,13 Saudi Arabia	N = 67, all males, mean age 53.85 ± 4.39 Group 1—HILT + GCS + exercise (n = 23), age 55.0 ± 4.41 Group 2—GCS + exercise (n = 22), age 53.64 ± 3.54 Group 3—placebo + exercise (n = 22), age 52.86 ± 5.03	Knee OA 1. Grade II-III KL classification	ASALaser (HIRO 3.0) Wavelength: 1064 nm Power: average 10.5 W (very high peak power 3kW) Power density (mW/cm2): Energy: 750 J to anterior knee, 750 J to posterior knee Energy per point (J): Energy density: high levels of fluency (510-1780 mJ/cm2) Treatment frequency: 2 times/wk for 6 wk	• VAS for pain • Arabic WOMAC • US for synovial thickness and fibrocartilage thickness	Statistically significant within-group improvements in VAS and WOMAC were found for all 3 groups at 6 wk. HILT + GCS + exercise showed a significant within-group decrease in synovial thickness at 6 wk and 3 mo. Statistically significant between-group improvements in VAS and WOMAC at 3 mo favored HILT + GCS + exercise compared with the other 2 groups. The HILT + GCS + exercise group showed a significant decrease in synovial thickness at 6 wk and 3 mo compared with the other 2 groups. There were no significant within- or between-group changes in femoral cartilage thickness.
Alfredo et al,14 Brazil	N = 40, both genders, age range 50-75 Group 1—LLLT + exercise (n = 20, 75.0% female), age 61.15 ± 7.52 Group 2—placebo + exercise (n = 20, 80.0% female), age 62.25 ± 6.87	Knee OA 1. Grade II-IV KL classification 2. Age 50-75 y 3. Pain and functional disability ≥3 mo 4. Meet ACR criteria	Irradia (not specified) Wavelength: 904 nm Power: 60 mW average Power density (mW/cm2): Energy: 27 J Energy per point: 3 J Number of points: 9 Energy density (J/cm2): not reported Pulse duration: 4.3 ms Treatment frequency: 3 times/wk for 3 wk	• VAS for pain • Lequesne questionnaire • Knee ROM • Dynamometry for isometric strength • WOMAC	Significant within-group improvements in VAS, Lequesne, WOMAC (pain, function, and total), and ROM were found in the LLLT group. No significant within-group improvements were found in the placebo group. Significant between-group improvements were found in pain and activity (WOMAC for pain, function, and total scores) favoring the laser treatment. No significant between-group differences were found in Lequesne, stiffness subscale of the WOMAC, or dynamometry.
Alfredo et al,15 Brazil	N = 40, both genders, age range 50-75 Group 1—LLLT + exercise (n = 20, 75.0% female), age 61.15 ± 7.52 Group 2—placebo + exercise (n = 20, 80.0% female), age 62.25 ± 6.87	Knee OA 1. Grade II-IV KL classification 2. Age 50-75 y 3. Pain and functional disability ≥3 mo 4. Meet ACR criteria	Irradia (not specified) Wavelength: 904 nm Power: 60 mW average Power density (mW/cm2): Energy: 27 J Energy per point: 3 J Number of points: 9 Energy density (J/cm2): not reported Pulse duration: 4.3 ms Treatment frequency: 3 times/wk for 3 wk	• VAS for pain • Lequesne questionnaire • Knee ROM • Dynamometry for isometric strength • WOMAC	No significant between-group differences were found at 3 and 6 m in any outcome measures. The LLLT group used fewer analgesics over the follow-up period.
Alghadir et al,16 Saudi Arabia	N = 40, both genders, age range 45-65 Group 1—LLLT (n = 20, 50.0% female), age 55.2 ± 8.14 Group 2—placebo (n = 20, 40.0% female), age 57 ± 7.77	Knee OA 1. Grade II-III KL classification	Chattanooga (Intelect) Wavelength: 850 nm Power: 100 mW Power density (mW/cm2): Energy: 6 J Number of points: 8 Energy per point: 6 J Energy density: 48 J/cm2 Treatment frequency: 2 times/wk for 4 wk	• VAS for pain • WOMAC • Time required to walk 50 ft “as fast as possible”	Both groups demonstrated statistically significant within-group improvements in all outcomes. The laser group also demonstrated statistically significant between-group improvements postintervention when compared with the placebo group in pain at rest, pain with activity, walking time, WOMAC pain, and WOMAC function.
Angelova and Ilieva,17 Bulgaria	N = 72, both genders, age range 39-83 Group 1—HILT (n = 37, gender data not provided), age 65.11 ± 1.40 Group 2—sham (n = 35, 68.6% female), age 64.71 ± 1.98	Knee OA 1. Grade II-III KL classification	BTL (not specified) Wavelength: 1064 nm Power (mW): maximal power of 12 W Power density (mW/cm2): Energy: 300 J (first 3 sessions) to 3000 J (next 4 sessions) Number of points: Energy per point (J): Energy density: 12 J/cm2 (first 3 sessions) to 120 J/cm2 (next 4 sessions) Treatment frequency: 7 sessions of treatment for both groups	• VAS for pain (at rest, on palpation, on movement) • Fisher dolorimeter for pain • RS Footscan System for pedobarometric assessment of gait	Statistically significant within-group improvements in all measures of pain were found at the end of treatment for both groups. Statistically significant between-group reductions in the “dynamics of pain” for all measures of pain favored the laser group after therapy, at 1 mo, and 3 mo. Reductions in pressure under the heel and contact surface area of the affected and unaffected legs favored the laser group.
Barabás et al,18 Hungary	N = 6, gender data not provided, age range 54-79 Group 1—LLLT-irradiated synovial membrane (n = 6) Group 2—nonirradiated synovial membrane (n = 6)	Knee OA 1. Meet ACR criteria	Laseuropa Ltd (KLS-500) Wavelength: 807-811 nm Power: 448 mW Power density (mW/cm2): not reported Energy (J): not reported Energy per point (J): not reported Number of points: 1 Energy density: 25 J/cm2 Treatment: laser 8 cm away	• Protein expression	Statistically significant between-group increase (improvement) in expression of mitochondrial heat-shock 60 kD protein 1 variant 1 was found in the laser group. Statistically significant between-group reductions (improvements) in expression of calpain small subunit 1, tubulin α-1C and β-2, vimentin variant 3, annexin A1, annexin A5, cofilin 1, transgelin, and collagen type VI α-2 chain precursor were found favoring the laser group.
de Matos Brunelli Braghin et al,19 Brazil	N = 60, both genders, age range 40-70 Group 1—control (n = 15, 80.0% female), age 60.8 ± 9.2 Group 2—LLLT (n = 15, 86.7% female), age 58.20 ± 7.97 Group 3—exercise (n = 15, 66.7% female), age 58.57 ±7.42 Group 4—LLLT + exercise (n = 15, 80.0% female), age 64.6 ± 5.24	Knee OA 1. Grade I-III KL classification	DMC (Photon Lase III) Wavelength: 808 nm Power: 100 mW Power density (mW/cm2): Energy: 56 J Energy per point: 5.6 J Number of points: 10 Energy density: 200 J/cm2 Treatment frequency: twice a week, 2 mo	• WOMAC • Gait variables: speed, cadence, step length, duration of support phase, and duration of single-limb support	Statistically significant within-group changes in exercise group changes in pain, function, and total WOMAC scores after 8 wk of treatment. No between-group differences were found at 8 wk for WOMAC. Statistically significant improvements in gait speed favored the laser, exercise, and exercise with laser groups compared with the control. Cadence statistically improved for the laser and exercise group when compared with the control. The laser and exercise group statistically improved right single-support phase duration.
de Oliveira Melo et al,20 Brazil	N = 45, all females, age range 66-75 Group 1—NMES (n = 15) Group 2—LLLT (n = 15) Group 3—NMES + LLLT (n = 15)	Knee OA 1. Grade II-III KL classification	THOR (DD2 Control Unit) Wavelength: 810 nm, continuous wave Power: 200 mW Power density (mW/cm2): 0.0364 Energy: 36 J Number of points: 6 (3 anteromedial and 3 anterolateral) Energy per point: 4-6 J Number of points: 6 Energy density (J/cm2): 0.218 Treatment frequency: 2 times/wk for 8 wk (for e-stim; 1x laser and 1x estim for combo group, laser per recommendations cited)	• VAS for pain • Vastus lateralis US images • 6MWT • TUG	Statistically significant within-group improvements in pennation angle and muscle thickness for both NMES groups. Significant within-group improvements for pain, 6MWT, and TUG for all 3 treatment groups.
de Oliveira Melo et al,21 Brazil	N = 44, all female Group 1—NMES (n = 15), age 69.3 ± 5.5 Group 2—LLLT (n = 15), age 67.7 ± 4.7 Group 3—NMES + LLLT (n = 14), age 69.6 ± 4.7	Knee OA 1. Grade II-III KL classification	Thor (DD2) Wavelength: 810 nm Power: 200 mW Power density (mW/cm2): 0.218 J/cm2 Energy: 36 J (first 4 wk) to 24 J (last 4 wk) Number of points: 6 Energy per point: 6 J (first 4 wk) to 4 J (last 4 wk) Energy density (J/cm2): Treatment frequency: LLLT was administered twice a week over a period of 8 wk, with a minimum interval of 48 h between sessions	• Ultrasonography for muscle thickness and cross-sectional area • EMG for muscle activity • Dynamometry for isometric strength • WOMAC	All 3 groups demonstrated significant within-group improvements in muscle torque, muscle activity, and all 3 subscales of the WOMAC postintervention. Only the NMES and combined (NMES + LLLT) groups showed a significant improvement in muscle thickness and cross-sectional area postintervention.
de Paula Gomes et al,22 Brazil	N = 60, both genders, age range 40-80 Group 1—exercise (n = 20, 90% female), age 64.75 ± 3.76 Group 2—LLLT and exercise (n = 20, 90.0% female), age 65.15 ± 4.9 Group 3—placebo and exercise (n = 20, 95.0% female), age 67.20 ± 3.95	Knee OA 1. Grade II-III KL classification 2. Meet ACR criteria 3. Knee pain in the previous 6 mo	Multi Radiance Medical (PainAway/PainCure) 9-diode cluster: 1 laser, 8 LED diodes Laser wavelength: 905 nm Power: 0.9 mW average; 8.5 W peak Power density (mW/cm2): not reported 4-diode LED wavelength: 875 nm Power: 17.5 mW mean per diode 4-diode LED wavelength: 640 nm Power: 15 mW mean per diode Total energy: 23.55 J Energy per point: 7.85 J per quadrant Number of points: 3 quadrants Energy density (J/cm2): not reported Treatment frequency: twice a week, 10 sessions	• WOMAC • LEFS • NRPS for pain • Dynamometry for isometric strength • Functional Reach Test	Statistically significant between-group improvements in NRPS scores were found for exercise with laser when compared with exercise alone and exercise with placebo laser.
Fukuda et al,23 Brazil	N = 47, both genders, age range 50-78 Group 1—LLLT (n = 25, 80.0% female), age 63.0 ± 9.0 Group 2—placebo (n = 22, 63.6% female), age 63.0 ± 8.0	Knee OA 1. Grade II-IV KL classification	Irradia (not specified) Wavelength: 904 nm, 700 Hz frequency Power: 60 mW mean (peak power 20 W) Power density (mW/cm2): Energy: 27.0 J Number of points: Energy per point: 3.0 J (5 points medial and 4 points lateral) Energy density (J/cm2): Treatment frequency: 3 times/wk for 3 wk	• TUG • Knee flexion goniometry • Dynamometry for isometric strength • NRS for pain • Lequesne questionnaire	Statistically significant within-group improvement in pain during ADLs, TUG, and goniometry for the laser group. Statistically significant within-group improvements for both groups in pain and dynamometry. Significant between-group improvement in the Lequesne questionnaire and pain favoring laser.
Ip,24 Hong Kong	N = 100, both genders (male:female = 1.0:1.5), mean age 65, age range 60-77 Group 1—standard physical therapy (age and gender data not provided) Group 2—standard physical therapy + LLLT (age and gender data not provided)	Knee OA 1. Tricompartmental radiographic change	Raycome (Sundom) Wavelength: 810 nm Power: 20 mW/cm2 Power density: 20 mW/cm2 Energy (J): Number of points: Energy per point (J): Energy density: 3.6 J/cm2 Treatment frequency: The treatment regime consisted of 3 sessions of treatment per week for 12 consecutive weeks. Each treatment session lasted 180 s per point.	• WOMAC • Treatment failure that necessitated joint replacement surgery	Statistically significant between-group improvement in pain levels on the WOMAC subscale in favor of the laser group. In addition, the laser group required significantly fewer joint replacement surgical procedures.
Kheshie et al,25 Saudi Arabia	N = 53, all males, mean age 54.6 ± 8.49 Group 1—HILT + exercise (n = 20), age 52.1 ± 6.47 Group 2—LLLT + exercise (n = 18), age 56.56 ± 7.86 Group 3—placebo + exercise (n = 15), age 55.6 ± 11.02	Knee OA 1. Grade II-III KL classification 2. No ROM limitation except minimum knee tightness 3. Pain ≥6 mo	HILT: ASA (HIRO 3) Wavelength: 1064 nm Power (mW): Power density (mW/cm2): Energy: 1250 J per session (3 phases [500 J + 500 J + 250 J]) Number of points: n/a—unit was moved in scanning fashion Energy per point (J): Energy density (J/cm2): phase 1: 710 and 810 mJ/cm2, phase 2: 610 mJ/cm2, phase 3: 610 mJ/cm2 Treatment frequency: 2 times/wk for 6 wk	• VAS for pain • WOMAC	Statistically significant within-group improvements were seen for all 3 treatment groups in VAS and WOMAC scores.
			LLLT: BTL-5000 Wavelength: 830 nm, 1 KHz frequency, 80% duty cycle Power: 800 mW Power density (mW/cm2): Energy density: 50 J/cm2 Energy: 1250 J Number of points: Energy per point (J): Energy density (J/cm2): Treatment frequency: 2 times/wk for 6 wk		Statistically significant between-group improvements in VAS and WOMAC pain scores were found favoring HILT over LLLT and both treatment groups improved more compared with placebo when combined with exercise. Significant between-group improvement in the WOMAC stiffness subscale was seen between HILT and the other 2 groups.
Kim et al,26 South Korea	N = 20, no gender data provided, age range 55-75 Group 1—conservative physical therapy (n = 10), age 65.5 ± 4.0 Group 2—HILT (n = 10), age 65.3 ± 4.2	Knee OA 1. Radiographic findings	United Technology Inc (HEALTRON) Wavelength (nm): Power (mW): Power density (mW/cm2): Energy (J): Number of points: Energy per point (J): Energy density: 1500 mJ/cm2 Treatment frequency: 3 times/wk for 4 wk	• VAS for pain • Korean WOMAC	Statistically significant within-group improvements in pain and function for both groups postintervention. Statically significant between-group improvements in VAS and WOMAC (total score) in favor of the HILT group postintervention.
Langella et al,27 Brazil	N = 18, both genders Group 1—LLLT (n = 9, 44.5% females), age 69 ± 5.6 Group 2—placebo (n = 9, 66.7% females), age 67 ± 6.4	Post-total hip arthroplasty (8-12 h) due to OA	Multi Radiance Medical (PainAway/PainCure) 9-diode cluster: 1 laser, 8 LED diodes Laser wavelength: 905 nm Power: 2.7 mW average; 8.5 W peak Power density (mW/cm2): not reported 4-diode LED wavelength: 875 nm Power: 17.5 mW mean per diode 4-diode LED wavelength: 640 nm Power: 15 mW mean per diode Total: Energy: 39.8 J Energy per point (J): Number of points: 5 Energy density (J/cm2): not reported Treatment frequency: 8-12 h immediately after surgery, single application	• VAS for pain • Cytokines—interleukin-6 and -8 and tumor necrosis factor α	No significant between-group difference was found for IL-6. Statistically significant between-group improvements in pain, IL-8, and TNF-α were found favoring the LLLT group when compared with placebo.
Mohammed et al,28 Egypt	N = 40, gender data not provided Group 1—“laser acupuncture” (n = 20), age 54.0 Group 2—placebo (n = 20), age 54.5	Knee OA 1. Grade II KL classification	Petrolaser Ltd (Soft-laser 202) Wavelength: 808 nm Power: 90 mW for each point Power density: 2.8 W/cm2 Energy: 70.2 J per knee Number of points: Energy per point: 5.4 J to each acupoint, 21.6 J to most tender points Energy density (J/cm2): Treatment frequency: 3 times/wk for 4 wk	• VAS for pain • Serum β-endorphin • Substance P	Statistically significant improvements in VAS were found in both groups at the end of treatment. Significant decrease (improvement) in substance P was seen in the “laser acupuncture” group only. Statistically significant between-group difference in VAS favoring the “laser acupuncture” group. No significant between-group differences in VAS or substance P.
Nambi et al,29 Saudi Arabia	N = 34, both genders, age range 45-65 Group 1—LLLT (n = 17), age 58 ± 6 Group 2—placebo laser (n = 17), age 60 ± 8	Knee OA 1. Grade II-IV KL classification 2. Unilateral knee OA 3. Pain VAS ≥4 cm	Fisiolaser Scan hp4 (EL12079-A01) Wavelength: 905 nm Power: 25 mW, pulsed mode Power density (mW/cm2): 2.8 W/cm2 Energy: 12 J Number of points: 8 Energy per point: 1.5 J Energy density (J/cm2): Treatment frequency: 3 times/wk for 4 wk	• VAS for pain • Joint space width in standing • Collagen-II telopeptide • Matrix metalloprotinease-3, -8, and -13	Statistically significant within-group improvements were seen in all measures for the laser group at the end of the trial. No within-group difference for the placebo group.
Nazari et al,30 Iran	N = 90, age range 50-75 Group 1—HILT + exercise (n = 30, 56.7% female), age 61.5 ± 3.90 Group 2—conventional physical therapy + exercise (n = 30, 53.3% female), age 62.4 ± 3.14 Group 3—exercise alone (n = 30, 53.3% female), age 62.24 ± 3.87	Knee OA 1. Grade II-III KL classification	Wavelength: 1064 nm, 30 Hz frequency, 70% duty cycle Power (mW): 5 W peak Power density (mW/cm2): Energy (J): 2400 J per session Number of points: Energy per point (J): Energy density (J/cm2): 60 J/cm2 Treatment frequency: 3 times/wk for 4 wk	• VAS for pain • Knee ROM • TUG • 6MWT • WOMAC	Statistically significant within-group improvements were seen in all 3 groups for all outcomes. Statistically significant between-group differences for VAS and WOMAC function subscale favored HILT compared with conventional PT. Significant between-group differences in VAS, knee flexion ROM, TUG, 6MWT, and WOMAC favored HILT and conventional PT group compared with placebo.
Paolillo et al,31 Brazil	N = 43, all female, age range 60-80 Group 1—placebo (n = 11), age 72 ± 6 Group 2—US + LLLT (n = 13), age 69 ± 5 Group 3—US + LLLT + exercise (n = 13), age 68 ± 6	Hand OA Diagnostic criteria not specified	MM Optics Ltda (prototype device) Wavelength: 808 nm Power: 100 mW peak Power density: 40 mW/cm2 avg Energy: 360 J Number of points: Energy per point: 72 J Energy density: 28 J/cm2 per point Treatment frequency: 3 times/wk for 4 wk US parameters: 1 MHz frequency, 1.0 W/cm2 intensity, pulsed mode 50% duty cycle, 3.5 cm2 effective radiating area	• Grip strength • PPT	Statistically significant within-group improvements in PPT were seen in both groups receiving US + LLLT but not the placebo group. No significant within-group differences in grip strength were observed in any group. Statistically significant between-group difference in PPT favored both treatment groups compared with placebo.
Paolillo et al,32 Brazil	N = 35, all female, age range 60-80 Group 1—placebo (n = 10), age 67.7 ± 4.7 Group 2—US + LLLT (n = 13), age 69.3 ± 5.5 Group 3—US + LLLT + exercise (n = 12), age 69.6 ± 4.7	Knee OA Diagnostic criteria not specified	Prototype device Wavelength: 808 nm, continuous noncontact scanning (∼1 cm distance) Power: 100 mW peak Power density (mW/cm2): Energy: 360 J Number of points: Energy per point: 72 J Energy density: 28 J/cm2 per region Treatment frequency: 1 time/wk for 3 mo US parameters: 1 MHz frequency, 1.0 W/cm2 intensity with 3.5 cm2 effective radiating area, 15 min per knee	• PPT • Sit-to-stand (30 s)	Statistically significant within-group improvements in PPT were seen in both groups receiving US + LLLT but not the placebo group. Significant within-group improvements in sit-to-stand were observed in all groups. Statistically significant between-group differences in PPT and sit-to-stand favored both groups receiving US + LLLT compared with placebo; no between-group differences were observed between treatment groups.
Youssef et al,33 Saudi Arabia	N = 51, both genders Group 1—LLLT total dose 48 J (n = 18, 61.1% female), age 67.3 ± 2.9 Group 2—LLLT total dose 27 J (n = 18, 66.7% female), age 67.5 ± 2.5 Group 3—placebo (n = 15, 66.7%), age 66.3 ± 3.2	Knee OA 1. Grade II-III KL classification 2. Meet ACR criteria	Irradia (not specified) Wavelength: Group 1: 880 nm or Group 2: 904 nm Power: Group 1: 50 mW or Group 2: 60 mW Power density (mW/cm2): Energy: Group 1: 48 J or Group 2: 27 J Number of points: Group 1: 8 or Group 2: 9 Energy per point: Group 1: 6 J or Group 2: 3 J Energy density: Group 1: 6 J/cm2 or Group 2: 3 J/cm2 Treatment frequency: 2 sessions/wk × 8 wk for 16 sessions in total	• VAS for pain • WOMAC • Knee ROM • Dynamometry for isometric strength	Statistically significant within-group improvements in pain and WOMAC post for all groups postintervention; improvements in VAS and WOMAC were greatest for Group 1, followed by Group 2, and the placebo group was the lowest. Statistically significant within-group improvements in isometric strength and ROM for all groups postintervention; improvements in isometric strength and ROM were greatest improvements for Group 1, followed by Group 2, and the placebo group was the lowest.

Abbreviations: 6MWT, 6-Minute Walk Test; ACR, American College for Rheumatology; ADL, activities of daily living; EMG, electromyography; GCS, glucosamine/chondroitin sulfate; HILT, high-intensity laser therapy; KL, Kellgren-Lawrence; LEFS, Lower Extremity Functional Scale; LLLT, low-level laser therapy; NMES, neuromuscular electrical stimulation; NRPS, Numeric Rating Pain Scale; NRS, Numeric Rating Scale; OA, Osteoarthritis; PPT, pain pressure threshold; ROM, range of motion; SKFS, Saudi Knee Functional Scale; TUG, Timed Up and Go; VAS, visual analog scale; US, ultrasound; WOMAC: Western Ontario and McMaster Universities Osteoarthritis Index.

PEDro scores are presented in Table 3. PEDro scores ranged from 3 to 10, with a median of 6.0 and a mean of 6.3. The most common scores (modes) were 6 and 7 (n = 5 for both). Methodical quality was poor for 2, fair for 5, good for 12, and excellent for 3 of the studies reviewed. No single criterion was met by all studies. Criteria 1, 2, 4, and 10 were met by 21 of the 22 studies. Criterion 6 was the least commonly met and was only found in 3 of the 21 studies.

TABLE 3 - Summary of PEDro Scores

Author	Eligibility Criteria Specified (Criteria 1)	Random Allocation of Subjects to Groups (Criteria 2)	Concealed Allocation (Criteria 3)	Baseline Similarity (Criteria 4)	Blinding: Subjects (Criteria 5)	Blinding: Treating Therapist (Criteria 6)	Blinding: Assessor (Criteria 7)	Measures of ≥1 Outcome for ≥85% of Subjects (Criteria 8)	All Subjects Received Treatment as Allocated or ITT Was Performed (Criteria 9)	Between-group Comparison for ≥1 Outcome (Criteria 10)	Point Measures and Variability for ≥1 Outcome (Criteria 11)	Total Score
Al Rashoud et al12	Yes	+	−	+	+	+	−	−	−	+	+	6
Alayat et al13	No	+	−	+	−	−	+	+	−	+	+	6
Alfredo et al14	Yes	+	+	+	−	−	+	+	−	+	+	7
Alfredo et al15	Yes	+	+	+	−	−	+	+	−	+	+	7
Alghadir et al16	Yes	+	+	+	+	−	−	+	+	+	+	8
Angelova and Ilieva17	Yes	+	−	+	+	−	−	−	−	+	−	4
Barabás et al18	Yes	−	−	−	−	−	−	−	+	+	+	3
de Matos Brunelli Braghin et al19	Yes	+	+	+	−	−	+	−	−	+	+	6
de Oliveira Melo et al20	Yes	+	+	+	−	−	+	+	−	+	+	7
de Oliveira Melo et al21	Yes	+	+	+	−	−	+	+	−	+	+	7
de Paula Gomes et al22	Yes	+	+	+	+	−	+	+	+	+	+	9
Fukuda et al23	Yes	+	+	+	+	+	+	+	+	+	+	10
Ip24	Yes	+	+	+	−	−	−	−	−	+	+	5
Kheshie et al25	Yes	+	−	+	+	−	−	+	+	+	+	7
Kim et al26	Yes	+	−	+	−	−	−	−	−	+	+	4
Langella et al27	Yes	+	+	+	+	+	+	+	+	+	+	10
Mohammed et al28	Yes	+	−	+	−	−	+	+	−	+	+	6
Nambi et al29	Yes	+	+	+	−	−	+	+	+	+	+	8
Nazari et al30	Yes	+	−	+	−	−	+	+	−	+	+	6
Paolillo et al31	Yes	+	−	+	−	−	−	−	−	+	−	3
Paolillo et al32	Yes	+	−	+	−	−	−	−	−	+	+	4
Youssef et al33	Yes	+	+	+	+	−	−	−	−	−	+	5
Number met	21/22	21/22	12/22	21/22	8/22	3/22	12/22	13/22	7/22	21/22	20/22

Abbreviations: ITT, intention to treat; PEDro, Physiotherapy Evidence Database.

Individual article summaries in alphabetical order

Al Rashoud et al¹²

Al Rashoud et al¹² examined patients with knee OA and the effects of LLLT applied to acupuncture points. Aging and older adults were randomly assigned to 1 of 2 groups: (1) (active) LLLT applied to acupuncture points or (2) a sham/placebo group that received the same procedures but the device was inactive. The laser device for both groups was applied to 5 acupuncture points commonly used to treat knee OA. Both groups also received supportive advice on coping with OA and exercise. Outcome measures examined included the visual analog scale (VAS) for pain, the Saudi Knee Functional Scale (SKFS), and active range of motion (ROM). The researchers measured across multiple times: baseline, fifth treatment session, last treatment, 6 weeks after intervention, and 6 months after intervention. At 6 weeks (mean difference −1.3) and 6 months (mean difference −1.8) after intervention, the active laser group reported significant improvements in pain compared with the sham group. Similarly, at the last treatment (mean difference −15) and 6 months (mean difference −21) after intervention, the active laser group reported significant improvements in SKFS scores compared with the sham group. Active ROM was also significantly improved for the active laser group at 6 months when compared with the sham group. The authors concluded that LLLT, when applied to acupuncture points and combined with advice and exercise, is effective at reducing pain and improving quality of life in patients with knee OA.

Alayat et al¹³

The effects of HILT on knee OA were examined by Alayat et al.¹³ Aging and older adults, all male, were randomly assigned to 1 of 3 groups: (1) HILT + glucosamine/chondroitin sulfate (GCS) + exercise (EX), (2) GCS + EX, or (3) placebo laser (PL) + EX. HILT was administered for 7 minutes, twice per week for 6 weeks, to the anteromedial, posteromedial, and lateral surfaces of the knee. Anterior and posterior knee surfaces were scanned in 2 subphases (initial and final) with 3 fluency levels. Outcome measures examined included the VAS for pain, an Arabic version of the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) for knee and lower limb function, and synovial thickness and femoral cartilage thickness quantified by US. Outcomes were examined pretreatment, after 6 weeks of treatment, and at a 3-month follow-up visit. Pain VAS and WOMAC scores significantly improved in all 3 groups after 6 weeks of treatment. Moreover, the HILT + GCS + EX group demonstrated significantly better VAS (2.73 ± 0.69) and WOMAC function scores (14.78 ± 1.2) at 3-month follow-up compared with the GCS + EX (VAS: 4.05 ± 0.79; WOMAC function: 21.82 ± 1.8) and PL + EX (VAS: 4.05 ± 0.79; WOMAC function: 20.63 ± 2.46) groups. Significant improvements in synovial thickness with HILT at 6 weeks and 3 months were also found when compared with the other 2 groups. With respect to medial and lateral femoral cartilage thickness, no significant differences were observed by group or time. The researchers concluded that, overall, the combination of HILT + GCS + EX was more effective in the treatment of knee OA than GCS + EX or PL + EX.

Alfredo et al¹⁴

In a randomized, double-blind study, Alfredo et al¹⁴ investigated the effects of laser paired with therapeutic exercises compared with placebo laser and the same exercises. The study participants were diagnosed with knee OA and related pain and functional disability for at least 3 months. Both groups were seen 3 times per week for their exercise and laser or placebo treatment for 3 weeks and then both groups only performed the exercises for 8 more weeks without the laser or placebo treatment. The laser group was treated over the medial knee lateral sides of the affected knee. The exercises that both groups performed were divided into 3 phases. All sessions started with a 10-minute warm-up and then moved into activities that progressed between each phase that focused on ROM, motor learning, and balance training. All participants were assessed at baseline, 3 weeks, and at 11 weeks when treatment was concluded. A blinded therapist completed all assessments. Outcome measures examined included pain using the VAS, function using the Lequesne questionnaire, ROM, muscular strength using a portable dynamometer, and activity level from the WOMAC. The laser treatment group showed statistically significant improvement in the pain, function, and total scores of the WOMAC at the 3- and 11-week follow-ups when each was compared with baseline. No other variables showed statistical significance between groups. The laser treatment group showed significant within-group improvements in pain and activity between baseline and 3 weeks (pain: 5.32 ± 3.55 to 3.336 ± 3.47; function: 11.88 ± 3.98 to 10.78 ± 4.62) as well as from 3 to 11 weeks of treatment (pain: 3.36 ± 3.47 to 2.58 ± 3.27; function: 10.78 ± 4.62 to 8.37 ± 4.27). There were no significant within-group changes in the placebo group. The authors stated that the limitations to the study included a small number of patients in each treatment group, absence of a control group, and the absence of follow-up after treatment. They concluded that the application of LLLT 3 times per week can assist in the performance of exercises in patients with knee OA and the combination of laser and exercise can improve pain, function, and activities in subjects with knee OA.

Alfredo et al¹⁵

In 2018, Alfredo and colleagues¹⁵ had the same 40 patients from their 2011 study return to determine the long-term benefits of adding LLLT to a typical exercise treatment protocol for knee OA. The same outcome measures that were used for pain, function, ROM, strength, and activity level were compared to the results found in the 2011 study. In addition to these outcomes that were used in the first study, the assessment of rescue medication (paracetamol) was included in this follow-up study. At the 3- and 6-month follow-ups, there were no significant differences in the original outcome measures between the treatment and placebo groups. However, the consumption of the rescue medication was significantly lower in the LLLT group at 3 months (laser group: 1 ± 1.12; placebo group: 2.8 ± 1.90) and 6 months (laser group: 0.45 ± 1.27; placebo group: 3.55 ± 2.16). From these results, the authors concluded that the use of LLLT with exercise reduces the need for long-term medication use in subjects with knee OA. They also concluded that the improvements in the treatment group found in the original study were maintained after treatment for 6 months.

Alghadir et al¹⁶

Alghadir et al¹⁶ examined the effects of LLLT on knee OA. Aging and older adults were randomly assigned to 1 of 2 groups: (1) LLLT or (2) placebo. All patients were first treated with hot packs to their affected knee(s) for 20 minutes. LLLT was then applied to 3 points on the medial aspect of the knee, 3 points on the lateral aspect, and 2 points in the popliteal fossa; each point was treated for 60 seconds. All patients were also given a home exercise program in addition to physical therapy (PT) visits twice per week for 4 weeks. Outcome measures examined included the VAS for pain, an Arabic version of the WOMAC for knee and lower limb function, and the time required to walk 50 ft as fast as possible for ambulation activity (performance-based). Outcomes were examined at baseline and following 4 weeks of intervention. The researchers found significant postintervention within-group improvements in VAS, WOMAC, and walking time scores for both treatment groups. The LLLT group had significantly better VAS (at rest [3.0 ± 1.25] and during activity [4.45 ± 1.19]), WOMAC (pain [3.25 ± 2.61] and function subscales [10.0 ± 7.39]), and walking time scores (18.8 ± 2.35 s) compared to the placebo group (VAS at rest: 4.4 ± 1.27; VAS during activity: 6.05 ± 1.35; WOMAC pain: 5.5 ± 2.5; WOMAC function: 18.2 ± 9.0; walking time 20.65 ± 2.99 s). The researchers concluded that LLLT seemed to be effective for improving short-term pain and function in patients with knee OA.

Angelova and Ilieva¹⁷

The effects of HILT on knee OA were also examined by Angelova and Ilieva.¹⁷ Middle-aged to older adults (gender distribution not specified) were randomly assigned to 1 of 2 groups: (1) HILT or (2) control (sham laser). Both groups received 7 treatment sessions. The first 3 sessions for the HILT group were analgesic and treatment was applied to the medial and lateral sides of the knee. The last 4 sessions for the HILT group used biostimulating parameters and treatment was applied to the medial side of the knee. Outcome measures examined included the VAS and dolorimetry for pain and pedobarometric assessment for static and dynamic analysis of gait (ie, contact surface area and maximum pressure under the heel). Patient outcome measures were examined before treatment, at the end of treatment, 1 month after treatment conclusion, and 3 months after treatment conclusion. Both groups demonstrated significant improvements in pain at the end of treatment. However, compared with the control group, the HILT group demonstrated significantly greater improvements in pain dynamics (ie, reduction of pain percentage) at the end of treatment, 1 month after treatment, and 3 months after treatment. For example, 3 months after treatment, the HILT had greater improvements in pain on the VAS at rest (−80.95% vs −11.37%), on palpation (−74.92% vs −3.83%), during movement (−68.51% vs −10.82%), pain using the dolorimeter in kg (6.73% vs 1.45%), static pressure in N/cm² (−19.94 vs 7.79), and dynamic surface area in cm² (−22.44 vs 16.32), when compared with the control group. With respect to pedobarometric assessment data, the HILT group demonstrated significantly decreased pressure under the heel and contact surface area between the affect and unaffected limbs, suggesting greater functional recovery. The researchers concluded that HILT may reduce pain and improve function immediately (after 7 sessions) and long-term (up to 3 months after treatment) in patients with knee OA.

Barabás et al¹⁸

In a study completed by Barabás et al,¹⁸ the authors investigated the effects of LLLT on the synovial membrane of knee joints that have been diagnosed with knee OA. One of the proposed mechanisms for the use of LLLT in treating arthritis is the effects it has on the tissue proteins. This study included 6 patients whose synovial membranes were surgically removed. Two pieces of tissue from each sample were removed and half of the samples received radiation from a laser. The remaining half of the samples were kept in the same conditions as the samples treated with the laser. The samples were then homogenized, treated with US, acetone, and then centrifuged to then separate the proteins by 2-dimensional differential gel electrophoresis. The tissues were then eventually evaluated using liquid chromatography-mass spectrometry analysis. Through this analysis, 11 proteins exerted differential expression in the LLLT-treated synovial samples. It was found that there was a statistically significant increase in the expression of mhsp60 and decreased expression of calpain small subunit 1, annexin, A5, vimentin variant 3, and collagen type VI α-2 chain precursor. The authors propose that these protein expressions may result in the suppression of cartilage degradation in OA. Although the focus of this systematic review is on clinical changes, this study was included as it fit our search criteria and provided additional information regarding the histological effects of laser treatment.

de Matos Brunelli Braghin et al¹⁹

In a controlled, randomized, factorial, and single-blind trial, de Matos Brunelli Braghin et al¹⁹ compared the effects of laser, laser and therapeutic exercise, therapeutic exercise, and an untreated control group. Participants were 40 to 70 years old diagnosed with knee OA, and randomly assigned to 1 of 4 groups: (1) control group, (2) laser group, (3) exercise group, and (4) laser and exercise group. The treatments were done twice per week for 2 months for a total of 15 sessions. The laser was applied near the lateral and medial epicondyles of the tibia and femur, in the joint line of the lateral and medial knee, the popliteal fossa, and on the patellar tendon. The exercises performed by the participants in the exercise group and the laser with exercise group included a warm-up, lower-extremity strengthening and then stretching. These activities progressed in the second and third phases where sit-to-stands from a chair, circuit training, gait training, and balance training were performed. The WOMAC and select spatial and temporal gait parameters were evaluated. The exercise group showed significant changes in the pain (from 24.64 ± 26.18 to 8.00 ± 10.99), function (from 19.22 ± 19.14 to 6.57 ± 8.28), and total WOMAC scores (from 19.56 ± 18.59 to 7.14 ± 7.81) after 8 weeks of treatment. There were significant changes between groups in gait speed between the laser (1.19 ± 0.21 m/s) and control group (1.0 ± 0.13 m/s), the exercise (1.18 ± 0.14 m/s) and control group (1.0 ± 0.13 m/s), and laser with exercise (1.22 ± 0.11 m/s) and the control group (1.0 ± 0.13 m/s). At the 8-week assessment, there was a significant change in cadence between the laser and exercise group (118.6 ± 7.58 steps/min) and the control group (105.45 ± 6.38 steps/min). The duration of the right foot support phase showed significant within-group changes for all groups receiving treatment and only the exercise with laser (61.42% ± 0.86 seconds) and exercise (61.62% ± 1.68 seconds) groups demonstrated significant changes over the control group. The duration of the right single-support phase demonstrated statistical improvement between the laser with exercise group (38.92% ± 1.27 seconds) compared with the control group (36.93% ± 1.74 seconds). The authors concluded that the group treated with only exercise showed improvements in pain and function (WOMAC total and subscale scores). All treatment groups showed improvement in gait speed compared with the control and exercise combined with laser provided the best results for the other gait variables.

de Oliveira Melo et al²⁰

de Oliveira Melo et al²⁰ examined females diagnosed with knee OA in a randomized, assessor-blinded, clinical trial. Using block randomization, participants were evenly assigned to 1 of 3 groups: (1) LLLT, (2) neuromuscular electrical stimulation (NMES), or (3) LLLT + NMES. Participants then received their assigned intervention twice per week for 8 weeks. During the first 4 intervention weeks, LLLT was administered to 6 points (3 anteromedial and 3 anterolateral over the intercondylar notch). Over the second 4 weeks, the dose was decreased by 30%. NMES was administered to the quadriceps motor point and just proximal to the patellar border, perpendicular to the longitudinal axis of the thigh, and stimulation was delivered in a pulsed symmetric biphasic rectangular current at the maximum intensity level tolerated by a subject. Those in the combined treatment received LLLT prior to NMES as per the same parameters described earlier. Outcomes were assessed at baseline, following a 4-week control period, and following the 8-week intervention period. Measures of interest were quadriceps muscle architecture (pennation angle, muscle thickness, and fascicle length), Timed Up and Go (TUG) test, 6-minute walk test (6MWT), and pain VAS postfunctional testing. All 3 groups demonstrated statistically significant improvements in 6MWT distance before to after intervention (LLLT = 430 ± 80 m to 470 ± 80 m, NMES = 400 ± 50 m to 420 ± 60 m, and LLLT + NMES = 380 ± 70 m to 410 ± 80 m), but there were no between-group differences in distance or pain relief. Similarly, there were no between-group differences in performance or pain relief during the TUG test. There were increases in pennation angle (PA) and muscle thickness (MT) but not fascicle length in the NMES group (PA: 12.2° ± 2.7°; MT: 1.9 ± 0.3 cm) and the LLLT + NMES group (PA: 12.9° ± 2.9°; MT: 1.8 ± 0.3 cm), but without differences between groups; there were no significant changes in the LLLT group (PA: 12.2° ± 2.7°; MT: 1.5 ± 0.5 cm). The authors concluded that NMES promoted incremental changes in muscle structure and function but the addition of LLLT did not result in any additional benefits.

de Oliveira Melo et al²¹

de Oliveira Melo et al²¹ examined the effects of NMES and LLLT on knee OA. Older adults, all female, were randomly assigned to 1 of 3 groups: (1) NMES, (2) LLLT, or (3) combined NMES + LLLT. All 3 groups started with a 4-week control period in which no intervention was provided, followed by 8 weeks of intervention administered twice per week with a minimum of 48 hours between sessions. For the LLLT group, treatment was applied to 3 anteromedial and 3 anterolateral points over the knee. During the first 4 weeks, LLLT settings were used to optimize the analgesic and anti-inflammatory effects for 30 seconds per point; during the last 4 weeks, LLLT settings were used to facilitate cartilage regeneration for 20 seconds per point. For the NMES + LLLT, participants received LLLT using the same protocol from the LLLT (only) group, followed by NMES to the quadriceps. Outcome measures examined included MT and cross-sectional area (CSA), knee extensor muscle activity (electromyography [EMG]), torque (isokinetic dynamometer), and health status (WOMAC). Outcome measures were examined at precontrol, preintervention, and postintervention time points. There were no significant differences in outcome measures between precontrol and preintervention for any of 3 groups. Postintervention, all 3 groups demonstrated a significant increase in torque and quadriceps EMG. MT and CSA significantly increased postintervention for the NMES (MT: 4.90 ± 0.93; CSA: 3.37 ± 1.30) and combined groups (MT: 4.75 ± 0.71; CSA: 2.99 ± 1.00) and were statistically significantly greater compared with the LLLT group (MT: 3.89 ± 0.88; CSA: 2.19 ± 0.75). All 3 groups demonstrated significant postintervention improvements in WOMAC pain, stiffness, and function. The authors concluded that LLLT improved EMG activity and health status for patients with knee OA, but had no effect of muscle mass. NMES used alone or in combination with LLLT resulted in statistically significant improvements in all of the outcomes examined.

de Paula Gomes et al²²

In a randomized, blinded, and placebo-controlled study by de Paula Gomes et al,²² the effects of laser combined with LED (phototherapy) and therapeutic exercise were examined. The 60 participants were 40 to 80 years in age with knee pain for the previous 6 months and a diagnosis of unilateral knee OA. The participants were randomly allocated to 1 of the following groups: exercise only, exercise and phototherapy, or exercise and placebo phototherapy. All treatments were performed twice per week for 5 consecutive weeks. The laser was applied over 3 quadrants (medial, lateral, and posterior) of the knee. The exercises performed by all groups included a 10-minute treadmill warm-up, squats, seated knee extension, side-lying clam, standing weight transfers side to side, and holding the toes flexed in a standing position. The outcome measures included the WOMAC, Lower Extremity Functional Scale, Numeric Rating Pain Scale (NRPS), pressure pain threshold (PPT), and Functional Reach Test. Of these outcome measures, the only significant differences were found between the exercise with phototherapy group and the exercise only group (mean difference 2.75, 95% confidence interval [CI] 3.17-2.32) as well as the exercise with phototherapy group compared with the exercise and placebo phototherapy group (mean difference 2.38, 95% CI 2.79-1.96) in their NRPS scores. Overall, the authors concluded that the combination of laser and LED (ie, phototherapy) is more effective in reducing pain than exercise alone or exercise combined with placebo phototherapy. Limitations of this study include a small sample size, the patients were from a single therapy center, and that the majority of patients were female.

Fukuda et al²³

In a double-blind randomized clinical trial, Fukuda et al²³ examined the short-term efficacy of LLLT in treating males and females who were diagnosed with knee OA. Participants were randomly assigned to 1 of 2 groups: (1) LLLT and (2) placebo. Those in the LLLT group received laser application to 5 points on the medial face of the knee and 4 on the lateral face. Those in the placebo group received identical duration, but no light administration. Both groups received 3 treatments per week for 3 weeks, a total of 9 sessions. Outcomes of interest were TUG test, knee flexion goniometry, knee extension dynamometry at 60° of knee flexion, 11-point pain VAS, and Lequesne algofunctional questionnaire. Compared with baseline measures, the laser group showed statistically significant within-group improvements in group means ± standard deviation for all outcome measures, including pain VAS during activities of daily living (from 6.1 ± 2.6 to 4.4 ± 2.9) and at rest (from 4.1 ± 3.1 to 1.9 ± 2.6), the Lequesne questionnaire (from 11.0 ± 4.4 to 7.8 ± 4.7), and TUG (from 9.2 ± 3.4 to 8.0 ± 1.9). The placebo group showed improvement only in knee extension dynamometry. Statistically significant between-group differences were present for pain VAS at rest and Lequesne, but not for other outcome measures. The authors concluded that LLLT resulted in short-term pain reduction and improved functional ability in patients with knee OA.

Ip²⁴

Ip²⁴ examined the effects of adding LLLT to PT for patients with knee OA. Older adult males and females were randomly assigned to 1 of 2 groups: (1) conventional PT or (2) PT + LLLT. Both groups received 3 treatment sessions per week for 12 weeks. LLLT was applied to multiple points for 180 seconds per point: medial and lateral epicondyle of the tibia and femur, medial and lateral knee joint gap, and to the biceps femoris and semitendinosus muscles. The main outcome of interest was subsequent knee joint replacement; secondary outcomes of interest were pain and stiffness WOMAC scores. The researchers measured participants after 12 weeks of intervention and followed-up with the cohort for a mean of 6 years (range 5.5-6.5 years). At 12 weeks, the PT-only group reported statistically significant improvements in WOMAC pain (mean = 7, range 6-8) and stiffness (mean = 4, range 3-5) scores compared with baseline values; the PT + LLLT group also reported significant improvements in WOMAC pain (mean = 4, range 3-5) and stiffness (mean = 3, range 2-4) scores compared with baseline values. At 6 years, the PT group reported an increase in WOMAC pain (mean = 11, range 10-12) but had the same stiffness values (mean = 4, range 3-5). The PT + LLLT group also reported an increase in WOMAC pain (mean = 6, range 5-7) and similar stiffness values (mean = 4, range 3-5). Pain scores on the WOMAC were significantly different between groups. Furthermore, only 1 of the 100 patients in the PT + LLLT group subsequently underwent knee joint replacement surgery compared with 9 of the 100 patients in the PT-only group, a statistically significant difference. The authors concluded that LLLT should be incorporated into the PT treatment for patients with knee OA to reduce short- and long-term pain and to reduce the need for joint replacement surgery.

Kheshie et al²⁵

Kheshie et al²⁵ investigated the effects of pulsed HILT, LLLT, and placebo laser treatment when combined with exercise. There were a total of 53 male participants diagnosed with knee OA. The subjects were randomly assigned to 3 groups. Group 1 included HILT and each session included 3 phases of treatment that focused on the joint line of the knee. Subjects in group 2 were treated using LLLT. The third group included placebo laser treatment. All groups utilized the same exercise program, which consisted of active ROM activities, muscle strengthening, and flexibility exercises. The exercises were also repeated at home with encouragement to maintain compliance; the home program frequency was not reported. The laser and placebo treatment groups had sessions twice per week for 6 weeks to total 12 visits. The authors found significant improvement in VAS and WOMAC scores for all treatment groups compared with their baseline measurements. Significant differences were found at 6 weeks between the 2 treatment groups, favoring the HILT group, in the following scores: VAS (HILT and exercise: 2.15 ± 0.75, LLLT and exercise: 2.97 ± 0.848), WOMAC pain (HILT and exercise: 3.15 ± 1.136, LLLT and exercise: 4.77 ± 1.11), and WOMAC function subscales (HILT and exercise: 13.90 ± 1.86, LLLT and exercise: 16.88 ± 2.11). The WOMAC stiffness subscale showed improvement between the HILT group (1.60 ± 0.68) and the low-level laser treatment group (2.22 ± 0.732) and the HILT group compared with the placebo group (2.40 ± 0.63). The authors concluded that LLLT or HILT combined with exercise were more effective than placebo laser with exercise and that HILT is more effective than LLLT in decreasing pain and increasing function when paired with exercise.

Kim et al²⁶

The effect of HILT for patients with knee OA was examined by Kim et al.²⁶ Aging and older adults of unspecified gender were randomly assigned to 1 of 2 groups: (1) conservative PT or (2) PT + HILT. Both groups received 3 sessions of treatment per week for 4 weeks. Conservative PT consisted of hot pack treatment, interferential current (IFC) therapy, and diathermy. HILT was applied to the tibia and femoral epicondyles for 5 minutes. Outcome measures examined included the VAS for pain and the Korean-WOMAC for function. The researchers measured participants before intervention and 4 weeks after. Preintervention scores for both groups were similar. Postintervention, both groups demonstrated statistically significant improvements in pain and function. In addition, the PT + HILT group had significantly lower scores (VAS = 3.1 ± 0.7, Korean-WOMAC = 17.2 ± 6.2) compared with the PT group (VAS = 5.8 ± 1.1 Korean-WOMAC = 31.9 ± 6.5) after intervention. The authors concluded that HILT is an effective treatment for patients with knee OA to reduce pain and improve function. One limitation of this study is the exclusive use of modalities to define conservative PT, which is inconsistent with contemporary management strategies for knee OA.

Langella et al²⁷

In a randomized, triple-blind, placebo-controlled trial, the effect of light therapy on pain and tissue repair was investigated for individuals after total hip arthroplasty (THA) due to hip OA.²⁷ The 18 participants were randomly separated into 2 groups, 1 that was treated with a light therapy device that included 1 super-pulsed laser, 4 red LEDs, and 4 infrared LEDs (ie, photobiomodulation therapy), and a placebo group. The photobiomodulation therapy was focused at 5 sites along the incision at 2-cm intervals. The evaluated outcomes included pain via the VAS and levels of cytokines including interleukin (IL) 6, IL 8, and tumor necrosis factor alpha (TNF-α). These were assessed at baseline and 10 minutes after irradiation. Pain ratings were similar at baseline and demonstrated significant reduction in the treatment group when compared to the placebo group after treatment. The cytokines were measured using enzyme-linked immunosorbent assay. The levels of IL-6 showed no statistically significant difference between the treatment groups at all time points. The IL-8 values were homogeneous at baseline between the groups and there was a significant reduction in IL-8 values in favor of the treatment group. TNF-α values were also homogeneous at baseline and showed a significant reduction in favor of the treatment group after treatment. The authors report that the use of photobiomodulation therapy using laser and different LEDs when used immediately after a THA due to OA provides a reduction in pain and IL-8 values.

Mohammed et al²⁸

A single-blinded randomized case controlled study by Mohammed et al²⁸ examined the effects of laser on participants with knee OA. Participants were separated into 2 equally sized treatment groups: (1) laser and (2) sham laser. Those in the laser group underwent a total of 12 “laser acupuncture” sessions over the course of 4 weeks (3 times per week). During each session, LLLT was directed to each of 5 acupoints that were selected according to traditional Asian medicine. LLLT was used on each acupoint for 1 minute. The Ashi points (most tender points) received the same average energy over 2 points per knee. Those in the control group received sham laser. The outcomes of interest were pain VAS, β-endorphin serum levels, and substance P levels. Substance P is proposed to have a role in increased pain perception and neurogenic inflammation. Groups were shown to be similar at baseline for all measures. There was a statistically significant between-group difference in median pain scores after intervention (LLLT = 2 vs sham = 7). Similarly, there was a statistically significant between-group difference in the median decrease in substance P (laser = −10 vs sham = −2). The authors concluded that LLLT is effective for pain control immediately following treatment sessions in individuals with grade 2 knee OA.

Nambi et al²⁹

Using laser as a treatment technique, Nambi et al²⁹ examined participants with unilateral knee OA. Participants were randomized into 2 equally size treatment groups: (1) LLLT + exercise and (2) placebo LLLT + exercise. Both groups participated in treatment 3 times per week for 4 weeks, which consisted of the respective laser assignment, conventional exercises, and the application of kinesiotape. Exercises included a warm-up followed by strengthening and stretching for the lower extremities; further details were not provided regarding kinesiotape application. LLLT was applied over the medial and lateral knee joint lines, medial and lateral epicondyles of the tibia and femur, and the tendons of the biceps femoris and semitendinosus muscles. The placebo group received identical procedures but with power output set minimally to 0.5 mW. Outcome measures of interest included pain using the VAS and biochemical analysis using fasting morning urine, were measured at baseline, following 4 weeks of treatment, and 8-week follow-up. There were statistically significant within-group improvements in pain for the LLLT group at 8 weeks (VAS: 7.8 ± 1.2 to 1.2 ± 0.2) but not in the placebo group (VAS: 7.6 ± 1.5 to 6.8 ± 1.3), resulting in a significant between-group difference. There were also statistically significant between-group differences for biochemical analysis, which included CTX-II, MMP-3, MMP-8, and MMP-13. The authors concluded that, in patients with knee OA, LLLT was an effective treatment for short-term improvements in pain, joint space width, and various biochemical variables.

Nazari et al³⁰

Nazari et al³⁰ performed an assessor-blinded RCT in males and females with a diagnosis of knee OA. Participants were randomly assigned to 1 of 3 groups (n = 30 in each group): (1) HILT + exercise, (2) conventional PT + exercise, or (3) exercise alone. HILT was performed in contact with the knee joint line using slow manual scanning in longitudinal and perpendicular directions over the medial and lateral aspects of the knee. The conventional PT care included transcutaneous electrical stimulation (TENS) and US, alternating between modalities at each session. The exercise protocol was the same for all participants and was designed to be performed at home following education at the first treatment session and an instruction booklet. Exercises included walking and gentle hamstring and calf stretches for warm-up followed by 9 exercises were to be performed in 3 sets of 10 repetitions. All groups participated in a total of 12 treatment sessions provided 3 times per week; the exercise-only group performed their protocol as a home-based exercise program without any modalities. Outcomes of interest were pain VAS, knee flexion ROM, TUG test, 6MWT, and WOMAC (overall, pain subscale, and stiffness subscale); all outcomes were assessed at baseline, after 4 weeks of treatment, and 12 weeks after baseline. All groups demonstrated statistically significant within-group improvements for all outcome measures at both 4 and 12 weeks after baseline with the HILT group demonstrating the largest within-group improvements. Both the HILT and conventional PT groups demonstrated statistically significant differences compared with the exercise-only group. However, the only statistically significant differences between the HILT and conventional PT groups existed for pain at 4 and 12 weeks (HILT: 3.79 ± 0.54 and 4.14 ± 0.68; conventional PT: 4.28 ± 0.56 and 5.21 ± 0.59) and WOMAC function subscale (HILT: 13.66 ± 1.42 and 16.43 ± 1.22; conventional PT: 15.50 ± 1.27 and 18.56 ± 1.19), both favoring HILT. The authors concluded that the application of HILT and exercise therapy in patients with knee OA was more beneficial than conventional PT and exercise.

Paolillo et al³¹

Paolillo et al³¹ examined the use of a prototype US + LLLT device in the treatment of hand OA and compared its effectiveness, alone and in combination with therapeutic exercise, to placebo treatment. Caucasian women with hand OA were randomly assigned to 1 of the 3 treatment groups: (1) prototype device + therapeutic exercise (n = 13), (2) prototype device alone (n = 13), and (3) placebo (n = 11). The device used included 4 ellipse-shaped diode laser beams arranged in a square around a US transducer. It was moved slowly and continuously in a circular pattern for 3 minutes at 24 treatment points. Both hands were treated at each session once per week for 3 months. This protocol was identical for all treatment groups, but a “null dose” was applied in the placebo group. Those in the exercise treatment group performed 10 grip movements at each session prior to receiving the prototype device; resistance progressed from 5 to 10 to 16 to 23 lb over the course of 2 months. Primary outcomes were grip strength of the dominant limb using a hand dynamometer and PPT using an electronic algometer. Outcomes were assessed at baseline and at the conclusion of the 3-month study. The authors reported no statistically significant within- or between-group differences for grip strength, but significantly increased average PPT for both groups receiving the nonplacebo US + LLLT modality, indicating reduced pain sensitivity; results were reported graphically. The authors concluded that the LLLT and US combination device was beneficial for long-term pain relief in women with hand OA.

Paolillo et al³²

In a similar study, Paolillo et al³² examined the use of the same prototype US + LLLT device in the treatment of knee OA in Caucasian women. Participants were randomly assigned to 1 of the 3 groups: (1) prototype device + therapeutic exercise (n = 12), (2) prototype device alone (n = 13), and (3) placebo (n = 10). The device included 4 ellipse-shaped diode laser beams arranged in a square around a US transducer. Treatment was provided to a total of 5 regions for a total treatment time of 15 minutes per knee. Treatment protocol was identical for all groups, with the placebo group receiving a “null dose.” The exercise program consisted of resisted hip flexion-extension in sitting and standing, knee flexion-extension in standing, and hip abduction-adduction in standing, performed in 3 sets of 10 repetitions. Ankle weights and elastic bands were used for resistance, which was progressed once per month. Outcomes were 30-second sit-to-stand test and PPT using an electronic algometer; these were assessed at baseline and at 3 months. The authors reported statistically significant within-group changes for PPT in the US + LLLT group (Δ = 13 ± 15 N) and US + LLLT + therapeutic exercise (Δ = 13 ± 12 N), but not the placebo group. Both groups receiving active US + LLLT were statistically significantly improved compared with placebo, but no difference was found between the group that performed exercise and the group that did not. The number of sit-to-stands improved significantly within all groups (US + LLLT + exercise from 9 ± 3 to 14 ± 3, US + LLLT from 9 ± 2 to 13 ± 2, and placebo from 9 ± 3 to 11 ± 3); as with PPT, there were statistically significant differences between placebo and both groups receiving active US + LLLT, but no difference between the latter 2 groups. The authors concluded that the combination of LLLT with US reduced pain and improved functional performance in women with knee OA.

Youssef et al³³

Youssef et al³³ examined the effects of LLLT combined with an exercise training program (stretching and strengthening) for patients with knee OA. In this study, the researchers examined the effects of different intensities (doses) of LLLT on pain (VAS), physical function (WOMAC), knee mobility (goniometry), and muscle strength (handheld dynamometer). Aging and older adult males and females (inclusion criteria of 60-72 years) were randomly assigned to 1 of 3 groups: (1) 6 J/cm² (total dose of 48 J on 8 points each session), (2) 3 J/cm² (total dose of 27 J on 9 points each session), or (3) placebo. All patients received treatment twice per week for 8 weeks (16 sessions in total). Outcome measures were examined before and after 8 weeks of intervention. All 3 groups demonstrated statistically significant improvements in pain, physical function, strength, and knee mobility after PT intervention compared with baseline. Post hoc analysis revealed significant differences between groups after intervention: patients in group 1 (total dose of 48 J) had the best outcomes, followed by patients in group 2 (total dose of 27 J), with the patients in group 3 having the worst outcomes. For example, knee flexion ROM was the greatest for group 1 (125.83° ± 3.3°) compared with group 2 (122.29° ± 2.4°) and group 3 (123.5° ± 2.8°). The authors concluded that the addition of LLLT to an exercise training program improved outcomes for patients with knee OA. Furthermore, the improvements seen by adding LLLT to an exercise program may be dose dependent.

Comparison group summaries

The articles found during our search were organized based upon their study design and comparison groups; groupings and descriptive statistics for subgroupings are presented in Table 1. The PEDro score means for subgroupings ranged from 3.5 (all subgroups for LLLT combined with US) to 8.2 (LLLT vs sham/placebo). The PEDro means for the majority (70%) of subgroups were good (≥6) or better. Because some articles included more than 2 comparison groups, articles may be included in multiple comparison summaries next.

Laser versus control

Based on 2 studies,^18,19 it appears that LLLT may be more beneficial than no treatment. For those receiving LLLT, Barabás et al¹⁸ identified potential histological benefits, which may suppress the cartilage degeneration in OA; those in the control group did not show these same changes. Similarly, de Matos Brunelli Braghin et al¹⁹ identified improvements in gait speed that were present in those who received LLLT but were not present in the control group.

Laser versus sham/placebo

Eight studies compared the effects of laser therapy on knee OA^{12,16,17,23,28,29,33} or hip arthroplasty²⁵ to a sham/placebo group using different methodologies. For all 8 studies, the sham/placebo group underwent the same protocol as the LLLT group, but the laser was not turned on or a sealed pen was used, thus there was no emission of energy. Five^{16,23,27,29,33} of the 8 studies examined LLLT, 1 study examined HILT,¹⁷ and 2 studies examined laser applied at acupuncture points.^12,28 As seen in Table 2, there was a high degree of variability in the modality settings, procedures, and treatment frequency for groups that received laser therapy. Outcomes examined also varied and included clinically based (eg, pain, joint mobility, self-report and performance-based function, and quality of life) and laboratory-based measures (eg, biochemical, spectrometry, and pedobarometric analysis). Across these 8 studies, patients with knee OA or hip arthroplasty who received laser therapy had significantly better outcomes than the sham/placebo groups.

Laser versus exercise

A single study, de Matos Brunelli Braghin et al,¹⁹ compared laser only to therapeutic exercise only and found no significant differences between groups at the 8-week assessment.

Laser with exercise versus control

de Matos Brunelli Braghin et al¹⁹ was the only study that compared LLLT with exercise to a control (untreated) group. They found significant improvements in the laser with therapeutic exercise treatment group in gait measures, but there were no significant differences between the groups in overall and subgroup WOMAC scores.

Laser with exercise versus exercise

Three studies compared the use of light therapy used in conjunction with therapeutic exercise to therapeutic exercise on its own; 2 investigated the use of LLLT^19,22 and 1 examined HILT.³⁰

Results regarding the effects of LLLT with exercise were variable. In the group receiving LLLT and exercise, de Matos Brunelli Braghin et al¹⁹ found that the exercise-only group improved their WOMAC total, pain, and functional scores. In contrast, de Paula Gomes et al²² did not identify significant differences in WOMAC scores, but did in one pain measure (NRPS). It should be noted that there were differences in the phototherapy parameters, as de Paula Gomes et al used laser with LEDs and de Matos Brunelli Braghin et al used laser alone, which may explain the differing results.

With regard to HILT, Nazari et al³⁰ identified between-group differences in pain VAS, flexion ROM, TUG, 6MWT, and WOMAC stiffness, function, and total scores favoring the group receiving HILT.

Laser with exercise versus placebo with exercise

Four different studies^14,15,21,23 reported statistically significant improvements in pain and self-reported function for groups receiving laser and exercise compared with placebo and exercise. As reported by Alfredo et al,¹⁴ statistically significant improvements seen in the group receiving LLLT and exercise were not present in the group receiving placebo and exercise. Furthermore, these improvements were maintained at long-term follow-up many years later as well as lower medication usage in the group receiving LLLT.¹⁵ de Paula Gomes et al²² reported similar improvements in their group receiving phototherapy (combo LLLT + LED) and exercise compared with placebo and exercise. The results provided by Kheshie et al²⁵ are in agreement, indicating that pain and functional scores were statistically significantly better in groups receiving laser (LLLT or HILT) and exercise compared with placebo and exercise.

Laser versus other modalities

Based on these studies,^20,21,26 it appears that LLLT does not result in clinical benefits or alterations in muscle structure compared with other modalities, namely NMES. Two studies completed by de Oliveira Melo et al^20,21 examined the effects of LLLT whereas a single study by Kim et al²⁶ investigated HILT. The 2015 study by de Oliveira Melo et al²⁰ identified structural changes in muscle in the group receiving NMES, but no clinical differences between groups. In 2016, de Oliveira Melo et al²¹ found increased EMG activity in a group receiving LLLT, but did not identify functional improvements compared with NMES. With regard to HILT, Kim et al²⁶ reported that the addition of HILT to other PT modalities resulted in statistically significant improvements in pain and function.

Laser with other modalities versus other modalities

Three studies combined LLLT with other modalities in the treatment of knee OA.^20,21,24 Ip²⁴ combined LLLT with a battery of modalities including US, TENS, and short-wave diathermy (SWD). They reported statistically significant improvements in WOMAC pain and stiffness scores 6 years later in patients receiving LLLT combined with US, TENS, and SWD when compared with those who received US, TENS, and SWD alone (without laser). Additionally, only 1 of 100 patients receiving LLLT underwent total knee arthroplasty within 6 years compared with 9 of 100 in the group that did not receive laser. Both studies by de Oliveira Melo et al^20,21 combined LLLT with NMES. In contrast with the findings of Ip,²⁴ both studies published by de Oliveira Melo et al failed to identify any between-group differences related to LLLT. In addition, neither study revealed differences in terms of functional improvements, pain, nor muscle mass.

Laser with other modalities versus exercise with other modalities

Based on the work of Nazari et al,³⁰ HILT combined with exercise was more beneficial than other modalities (TENS and US) combined with exercise in terms of long-term pain and function.

Laser with glucosamine chondroitin sulfate and exercise versus glucosamine chondroitin sulfate with exercise

Only 1 study, Alayat et al,¹³ compared the effects of HILT + GCS + exercise (EX) with GCS + EX. There were significant within- and between-group differences that support the addition of HILT to GCS and EX. Pain VAS and WOMAC scores significantly improved in both groups after 6 weeks of treatment. There were also significant between-group differences in favor of the group that received HILT for pain, function, and synovial thickness at 6 weeks and 3 months and for stiffness at 6 weeks.

Laser with glucosamine chondroitin sulfate and exercise versus placebo with exercise

Alayat et al¹³ also compared the effects of HILT + GCS + EX to placebo HILT + EX. Significant within- and between-group differences that support the use of HILT with GCS and EX when compared with placebo laser and EX were reported. Pain VAS and WOMAC scores significantly improved in both groups after 6 weeks of treatment. There were also significant between-group differences for pain, function, and synovial thickness at 6 weeks and 3 months and for stiffness at 6 weeks in favor of the group that received HILT.

LLLT with exercise versus HILT with exercise

We were only able to identify a single study that compared HILT and LLLT. Kheshie et al²⁵ found that, when combined with an exercise program, HILT was more effective than LLLT for decreasing pain and improving function.

Laser combined with ultrasound

Two studies completed by Paolillo et al^31,32 examined the effects of a prototype unit that combined LLLT with typical US. This was used individually and in conjunction with therapeutic exercise to treat hand and knee OA and compared to placebo. Overall, participants with either condition who received application of this combo unit showed increased PPT, whether or not it was combined with therapeutic exercise. There were no significant differences when therapeutic exercise with the combo unit was compared to using the unit alone.

DISCUSSION

In this review, the effects of light therapy were examined by several RCTs that focused on OA and its sequelae and commonly associated treatments experienced by aging and older adults. Many of the studies included both genders and focused primarily on knee OA, with some investigating the effects on hand OA and postoperative THA. We focused on more recent literature due to the advancements in laser technology and usage.

Many of the author groups reported findings indicating that the groups receiving light therapy, whether independently or in conjunction with exercise, as a component of their treatment for knee OA demonstrated significant improvements in various clinical outcomes, gait parameters, and tissue proteins compared with those receiving control, sham, and placebo treatment. However, compared with those receiving exercise, groups receiving light therapy, used independently or in combination with exercise, results indicated some improvement to no differences.

When LLLT was used with NMES, there was no significant improvement in pain or function. There were significant improvements in pain and function when HILT was used in conjunction with SWD, IFC electrical stimulation, US, and hot packs compared with those modalities used together but without light therapy. The use of HILT with US and TENS proved better than TENS and US with exercise. Overall, it appears the additional use of laser with other modalities resulted in significant improvement compared with the other modalities; the exception was NMES when treating knee OA.

The use of HILT with GCS showed more significant improvements when compared with GCS and exercise as well as placebo combined with exercise for treating knee OA.

HILT was found to be more effective than LLLT and a prototype combination unit with LLLT and US showed significant improvement over placebo regardless if it was combined with exercise or not for knee and hand OA.

When light therapy is used over surgical incisions immediately after THA, there were significant differences in pain and interleukin markers favoring the LLLT treatment group when compared with sham/placebo.

Overall, there appears to be a majority of evidence that supports the use of laser therapy when treating knee OA, hand OA, and surgical incisions post-THA for older and aging adults in improving pain and functional outcomes; however, those that compared light therapy to exercise showed some varying results. The results were impacted by laser parameters and intensity (HILT vs LLLT), if laser was used in conjunction with LEDs, and potentially gender.

Two recent systematic reviews that included meta-analysis examined LLLT and used similar, but slightly different, inclusion criteria. Both Huang et al⁴ and Rayegani et al⁵ compared LLLT to placebo laser in the treatment of knee OA and included studies that were outside our study's date range. Huang et al⁴ were more selective, requiring PEDro scores greater than 5 and excluding studies that had additional treatments like exercise and education. Rayegani et al⁵ included articles that were written in Persian as well as English and included treatments that used LLLT along with other interventions. Huang et al⁴ found no significant differences in immediate or 12-week reassessment in pain VAS or WOMAC pain subscale; there were also no significant differences in the WOMAC stiffness and function subscales. Rayegani et al⁵ found significant differences favoring LLLT in pain VAS and the WOMAC function, stiffness, and total scores; they did not find significance in the WOMAC pain scores.

A single systematic review with meta-analysis investigated the effects of HILT on musculoskeletal disorders (Song et al⁶), not just OA as done in this study. This review included placebo comparisons as well as other treatments such as US and TENS. They included RCTs written in English and Korean with mean age ranges from 32 to 58 years, thus there was some inconsistency compared with our inclusion criteria. Compared with a control group, there was significant improvement seen in the HILT treatment groups for pain located in the neck, back, and arms/hands but not for the shoulder. When assessing various disability outcome measures specific to each region, groups receiving HILT significantly improved compared with a control group for the neck, back, and shoulder but not for the arm/hand group. It was also found that disability scores improved significantly compared with those reported by a placebo group, but no difference compared with active treatment control groups.

Future recommendations

Based upon the findings of our systematic review, we offer the following recommendations for future examination:

1. Compare different laser wavelength, total dosage, and inclusion of LEDs to determine optimal treatment delivery. There was variability among the studies reviewed with regard to treatment parameters, including laser specifications, namely wavelength (the parameter that theoretically impacts which tissues are targeted), total energy applied to the treatment area, and the inclusion of a second type of diode, all of which potentially impact the overall treatment.
2. Provide sufficient information on unit settings and parameters to allow replication in practice and research.
3. Further comparison of light therapy to other modalities. The studies that we found compared laser treatment to groupings of pain modulating modalities. In some treatment settings, the grouping of modalities is not typical practice; it would be beneficial to compare laser to independent use of TENS/IFC, SWD, or US to determine whether one is more effective in pain control and/or functional outcomes.
4. Identify the most effective type of phototherapy (laser, SLDs, and LEDs) for treating arthritic changes. Because each of the 3 devices has various parameters and associated effects, it may be beneficial to first find the most effective parameters for each type, then compare that to the most effective parameters of a different diode type (ie, compare the most effective laser parameters with the most effective LED parameters).
5. Compare the effects of phototherapy to, or when combined with, orthopedic manual PT.
6. Compare the effects of light therapy, a conservative treatment with few associated side effects, to pharmacological agents in reducing pain in patients with arthritis.
7. Examine the effects of light therapy on performance-based measures of function and quality of life. The majority of studies that examined changes in function used self-reported measures.
8. Related to the theoretical effects of phototherapy on tissues, additional research that examines the association between histological changes and changes in clinical outcome measures including pain, function, and quality of life is warranted.
9. Explore long-term clinical benefits from using light therapy.
10. Investigate other body regions or arthritic pathologies that commonly affect older adults. The majority of the studies we found investigated the effects of laser therapy on individuals with knee OA.

Limitations

This narrative systematic review had several limitations. There were challenges in grouping our interpretation of results due to the diversity of study design and grouping of treatments within studies. We elected not to perform meta-analytic analyses due to the high degree of variability (heterogeneity) in protocols, unit settings and types (LLLT, HILT, LLLT with LED, LLLT with US, etc), treatment philosophies, and outcome measures among studies within each cluster. While we used 2 of the most commonly searched databases, this strategy may have resulted in a selection bias. Some studies included in this review, while in English, originated from countries outside the United States where the use of lasers in rehabilitation may differ from the United States. By limiting our search to English, we may have excluded relevant studies published in other languages.

Clinical recommendations

Phototherapy appears to provide significant improvements in pain and functional outcomes compared with control, sham, or placebo treatments. There are varied results when phototherapy is compared with exercise, but there are some studies that found it is beneficial when used in conjunction with an exercise program. Overall, it appears that laser may be an effective treatment option for controlling pain, improving function, quality of life, and stimulating interleukins related to healing. It is also apparent that exercise continues to be an important treatment consideration for knee OA. As there are many parameters and settings associated with laser treatment, it is important to document them to further understand whether there are more effective settings to be used.