Low Level Laser Therapy For Chronic Shoulder Pain

I have had pretty good success using low level light therapy (LLLT) for a number of conditions. However, I have a generalized shoulder/scapular pain that is not well treated by the devices which I currently have. So I decided to take a look at some of the information relating to LLLT treatments of chronic shoulder and scapular pain. This article looks at low level laser therapy treatments for shoulder tendinopathy. I look for the best treatment protocols and the most appropriate devices. One treatment protocol led to a 7/10 improvement in pain by 3 months. Most of that improvement was seen in a few weeks.

Laser Treatment Of Chronic Shoulder Pain

A 2014 review of the scientific literature looked at the effect of low level laser therapy on shoulder tendinopathy. [*] Tendinopathy consists of either inflammation of the tendon or tears in the tendon without inflammation.
Some other terms used to describe shoulder tendinopathies in these articles:

  • Rotator cuff syndrome,
  • Rotator cuff tendinitis,
  • Shoulder tendonitis,
  • Subacromial impingement,
  • Unilateral shoulder pain,
  • Supraspinatus tendinoses,
  • Myofascial pain,
  • Shoulder periarthritis,
  • Cervical Osteoarthritis

Tendinopathy is generally considered to be a condition caused by overuse, in the case of shoulder tendinopathy this is often from reaching overhead. Trauma, arthritis, and age are also associated with tendinopathy. It’s a pretty common complaint in racket sports like tennis. The review included studies which met the following criteria:

“A systematic review with meta-analysis and primary outcome measures pain relief on 100-mm visual analogue scale (VAS) and relative risk for global improvement. Two independent assessors rated the included studies according to the PEDro scale. Intervention quality assessments were performed of LLLT dosage and treatment procedures according to World Association for Laser Therapy guidelines. The included trials were sub-grouped by intervention quality and use of other physiotherapy interventions.”

Visual Analogue Scale

A Visual Analogue Scale (VAS) is a system of assessing a characteristic that cannot be easily directly measured, like pain. VAS often uses a 10mm or 100 mm line with descriptor words at each end, for example “No pain”and “Severe pain”. The relative pain sensation can be marked between those two extremes on the line. The markings are measured and compared to assess pain and changes in pain. This provides a range of scores from 0-100. Sometimes regions are classified, for example: no pain (0–4 mm), mild pain(5-44 mm), moderate pain (45–74 mm), and severe pain (75–100 mm) (11). [*] [*]

Visual Analogue Scale

Relative Risk

Relative risk is used interpret relevance, magnitude and direction of effects between treatments. A relative risk of 1 indicates no difference. A relative risk of 2 indicates that the treatment had twice the probability of being effective, and so on. (Higher is better)

PEDro & Walt

The studies were also rated according to standards of the Physiotherapy Evidence Database PEDro [*] WALT is the World Association for Laser Therapy and they put out treatment guidelines for low level laser therapy for musculoskeletal disorders. The studies included in this review were rated against WALT guidelines, which can be found here. 780-860nm [*] 904nm [*]


The study found LLLT to be significantly and clinically effective in treating pain from shoulder tendinopathy. LLLT showed improvement of 20 VAS points weighted mean differences over placebo. Likelihood of improvement from LLLT alone over placebo was almost double, whereas improvement with LLLT and and physiotherapy was only half of that. The review looks at studies that combine exercise or physiotherapy with LLLT, my focus is on LLLT alone vs placebo which provided the most improvement. 


Significant and clinically important pain relief was found with weighted mean differences (WMD) over placebo, for LLLT as monotherapy at 20.41 mm (95% CI: 12.38 to 28.44) and as adjunct to exercise therapy at 16.00 mm (95% CI: 11.88 to 20.12). The WMD when LLLT was used in a multimodal physiotherapy treatment regime reached statistical significance over placebo at 12.80 (95% CI: 1.67–23.94) mm pain reduction on VAS. Relative risks for global improvement were statistically significant at 1.96 (95% CI: 1.25–3.08) and 1.51 (95% CI: 1.12–2.03), for laser as monotherapy or adjunctive in a physiotherapy regime, respectively.”

Function & Dose

LLLT does not seem to be as effective in improving shoulder function, at least with these protocols. Though there was some improvement. These studies were relatively short in time frame and that might be a factor in shoulder function. The paper reiterates the importance of dose parameters. Some papers were excluded for using the Roland IR 904 laser device, which has been shown to have a power output less than 1% of what the manufacturer states.

“Secondary outcome measures of shoulder function were only significantly in favour of LLLT when used as monotherapy. Trials performed with inadequate laser doses were ineffective across all outcome measures.“

Penetration Power And Wavelength

The studies included in this paper all used infrared, not red light. This near infrared portion of the spectrum used has better penetration than red light. Positive effects were seen in trials using 3-100mW powered lasers. Positive results were found at energy doses from 0.9 to 7.5 Joules (per application point). Most of the studies reporting no effect were on subacromial impingement, though there was one positive report on the same diagnosis [*]. Let’s take a look at the specifics of one of the more impressive studies cited in this review (in terms of VAS/pain improvement).

Low-level laser therapy treatment of trigger points in rotator cuff tendonitis

This double-blind randomized controlled trial included 60 people[*]. The placebo group were subjected to the treatment procedure without the laser being turned on. Infrared is invisible and there is no sensation from light at the doses used here. The device used in this study was a pulsed laser. The on/off pulsing was set to 5,000 Hz. The wavelength was 820 nm – which is in the lower range of wavelengths included in the review selection. Power output is listed at 100 mW – the highest included in the review. Where we have pulsing and stated power output without any clarification we have a problem of interpretation. Is the laser 100 mW peak power average power – remember this is a pulsed beam.


When devices are pulsed the manufacturer and reporting study should indicate whether they are stating the peak power (the actual power of the beam), or the average power. If a 100 mW beam will turn on and off once per second the peak power would be 100 mW but the average power would be only 50mw, assuming that the on/off cycle are the same length.


Further information is given inside the body of the study and we find that the reported 100 mW is the average power and not the peak power. So how can we calculate the peak power? The duty cycle is necessary. Duty cycle (aka work cycle) is the descriptor of the percentage of time which the beam spends on during pulsing. The study does not give a duty cycle. This often means that the duty cycle is 50% – an equal period on and off. If this assumption is correct then the peak power would be 200mW as it is on for half of the cycle. [*]

Duty Cycle Or Work Cycle

Laser Treatments

This study treated the three most painful trigger points in each of the patients. Each trigger point application lasted for 40 seconds, delivering a total dose  of 4 J per point. The size of the laser beam was 0.125cm2 (so the radiant exposure could be stated at 32J/cm2 on the “points” themselves. This treatment was applied three times per week for four weeks.


“The laser (Excel, Omega Universal Technologies Ltd, London, UK) parameters used were a wavelength of 820 nm, a power output of 100 mW, a frequency of 5000 Hz (modulated) and energy density of 32 J/cm2. The two groups received a course of 12 treatment sessions for four weeks (3 sessions per week).”

Reduction In Shoulder Pain

There was a massive reduction in pain as measured by the VAS in the LLLT group. Treatment produced a decrease of six points on a scale of 0 to 10 by the end of the trial period. Further improvements were seen three months later a follow-up examination. The higher peak power in this study may be one reason why it was so successful. This allows for deeper penetration. Pulsing seems to be required for a number of reasons. By the final assessment there was a reduction in pain of 7 points on a scale of 10, and pain was not necessarily a 10 to begin with.


“Pain, functional activities (as measured using the Shoulder Pain and Disability Index, SPADI), pressure pain threshold (PPT) and range of motion (ROM) were assessed pre and post treatment, with a three month follow-up assessment. Significant improvements in pain (p < 0.001) were observed for the laser group (6 cm median improvement on a 10 cm VAS) compared to the sham group (2 cm median improvement) immediately post treatment. The improvements in the laser group continued post treatment with a 7 cm median improvement observed at three month follow-up.


(A previous, similar study using constant wave (non-pulsed) 830 nm beam with 30mW output and only 1J per point was ineffective. [*]This study also showed improvements in range of motion, function and pain threshold using an algometer to prod the tissues in question. *TrPs are trigger points*

“Similar between group differences were observed for ROM (p < 0.01), functional activities (p ≤ 0.001) and PPT (p ≤ 0.05). The findings of the current study suggested that LLLT is effective in treating patients with TrPs associated with rotator cuff tendonitis, when using the parameters described. However, the mechanism of its action is not yet clear, and will require further investigation.”

Peak Or Average Power Reporting?

I decided to take a look at some of the other studies included in the review that could be classified as effective in the real world. A handful of studies show an improvement of 20-25/100 points of VAS improvement vs placebo. One of these studies has the lowest power included in the review (3mW).  I thought perhaps I could compare a device that was a little less expensive against a pulsed device. However, the same reporting problem is seen with this study. That study did not use a constant 3mW peak beam. In fact the laser device used in this study had a 10,000 mW peak power. [*] A 10,000 mW peak pulse will penetrate far deeper than a 3 mW constant beam. Reporting in low level laser therapy is very often misleading. The other relatively effective treatment vs placebo is listed as 40mW but this is also pulsed, at 5000 Hz, and the duty cycle is not given. If assumed to be 50% then the peak power would be 80mW, that study used 820nm. [*]

Peak/Pulse Effect?

It seems to me that these reviews and abstracts often ignore the peak power and so give misleading data points. Peak power is a key factor for penetration and penetration is key in musculoskeletal disorders. How important is the pulse rate? That’s not so clear, but I’ve seen a lot of positive results with 3,000 Hz to 10,000 Hz. Many of the Ukrainian and Russian studies on autoimmune diseases including hypothyroidism use >1,000mW peak power and pulsing upwards of 800 Hz, though they often combine other laser treatments with lower parameters.

What Devices Could Provide These Treatments?

Based on these data points it seems that around 820/830 nm pulsed laser with 50% or less duty cycle and pulse frequency range in thousands might be an optimal device for shoulder tendinitis. 80-10,000mW peak power. Treatments of 4J per point with 3+ points per session, 3x per week.


The B-Cure seems to be the cheapest device out there with published specifications that fit within the desired range. B-Cure produce a number of handheld devices which are all based around pulsed 810 nm, 250 mW peak power laser. There are different versions which have slightly different pulse rates, pulse durations and work/duty cycles. I suspect these small differences are irrelevant for most uses.  

B-Cure Classic, which is the cheapest device in the range, is available on Ebay and Amazon.co.ukThe B-Cure Vet and B-Cure Sport Pro are listed on Amazon.com.

The specification differences are as follows:

B-CURE Laser Classic

  • Laser Pulse frequency: 15 kHz\
  • Laser Pulse duration: 17 µS
  • Energy per minute: 3.75 J (Joules)
  • Work Cycle 30%


  • Laser Pulse frequency: 13 kHz
  • Laser Pulse duration: 26 µS
  • Energy per minute: 5 J (Joules)
  • Work cycle: 33 %


  • Laser Pulse frequency: 14 kHz
  • Laser Pulse duration: 22 µS
  • Energy per minute: 4.5 J (Joules)
  • Work cycle: 30 %

Sport Pro

  • Laser Pulse frequency: 13 kHz
  • Laser Pulse duration: 26 µS
  • Energy per minute: 5 J (Joules)
  • Work cycle: 33 %


  • Laser Pulse frequency: 14 kHz
  • Laser Pulse duration: 22 µS
  • Energy per minute: 4.5 J (Joules)
  • Work cycle: 30 %

Dental Pro

  • Laser Pulse frequency: 13 kHz
  • Laser Pulse duration: 26 µS
  • Energy per minute: 5 J (Joules)
  • Work cycle: 30 %

Vet Pro

  • Laser Pulse frequency: 13 kHz
  • Laser Pulse duration: 26 µS
  • Energy per minute: 5 J (Joules)
  • Work cycle: 33 %


Thorlaser produce top of the range laser and LED equipment that runs into tens of thousands of dollars for a full set. These devices are also relatively common for physiotherapists and other practitioners, so getting treatment with a Thorlaser may not cost an arm and a leg. The output parameters are controlled/limited by the control unit and probe. Thorlasers standard control unit is capable of 11 pulse frequencies from 2,500 Hz to 20,000 Hz. Thorlabs produce a probe with 810 nm wavelength laser at 200mW. So Thorlabs could produce parameters very close to an optimal treatment for shoulder tendinopathy provided the right probe is used. [*] 


I suspect the terraquant devices can provide the required specifications but I cannot find them online. I have contacted the manufacturer/supplier to request specs and will update if I receive them. I have found that manufacturers of low level laser/light therapy devices very rarely reply to enquiries. The terraquant solo puts out a peak power of 15,000mW at 904nm, so it’s certainly not underpowered.


There are significantly cheaper devices available which state that they have pulsing and >100 mW infrared beams. I bought a “Laspot GD-P1” which is a Chinese manufactured device sold under many names. It tested a similar output to the stated power, around 150mW for the infrared beam (808nm). The pulsing is only approximately 1Hz and while it is helpful for more superficial problems it doesn’t make a dent on my shoulder/scapula pain. This device looks identical to my “Laspot”.

Another easily available device may or may not be helpful. Though it appears under many listings I haven’t been able to compile full specs. The HD-Cure Max type device claims 510 mw, but this is the combined power of multiple beams. The most powerful 808nm beam is somewhere between 60-150mW by some claims and 450mW by other sellers. Some sellers claim the device has pulsing, parameters of which are unspecified. I haven’t been able to figure out who the manufacturer is in order to ask about specs and I’d avoid unless those specs show up. *Update – A seller stated this has a pulse rate of 5Hz, another person was told the same by a different seller.* I’ve never seen a pulse rate this low in a study, which makes me wonder if this is simply for marketing. It’s one I’d avoid.

The RG 500i may provide the required specs, with max peak output of 500mW. Some sellers claim it has pulsed settings. I’ve contacted a number of sellers about this device and some similar looking devices on aliexpress that do not have full parameters listed. I’ll update if I get any useful responses. It seems the B-Cure and thorlaser might be the best options.