Imagined city inside a cell, with red light illuminating upper power stations while deeper districts remain dim.
Back to the library

Recovery

Red light and near-infrared (photobiomodulation)

Photobiomodulation uses red (630 to 660 nm) and near-infrared (800 to 850 nm) light at low, non-heating doses. Its best evidence is a set of international oncology guidelines for preventing oral mucositis, followed by a 2019 meta-analysis of 22 placebo-controlled trials reporting reduced knee osteoarthritis pain at recommended doses, and controlled trials reporting denser dermal collagen and fewer wrinkles. Evidence for sleep, mood, cognition and whole-body panels remains early, and results depend heavily on dose.

Evidence so far
Moderate
Reading timeAbout 12 minutes

Start here

Why this is being studied

Photobiomodulation, once called low-level laser therapy, is the use of red and near-infrared light at intensities too low to heat tissue. The light is absorbed by an enzyme in mitochondria, and the working theory is that this nudges under-performing cells back towards normal energy production, signalling and repair. It is studied because the same basic mechanism could plausibly matter in very different tissues: skin, joints, muscle, the lining of the mouth and even the brain. For members it offers a non-invasive, largely side-effect-free session; for clinicians it is a field with one area of guideline-level evidence, several areas of moderate evidence, and a persistent problem of inconsistent dosing that makes device-to-device comparison difficult.

Where the evidence stands

Flag text

The mechanism

What is going on, and why this might help

What is going on in the body

Every cell runs on energy made in its mitochondria, and the final step of that process is handled by an enzyme called cytochrome c oxidase. In tissue that is inflamed, injured or simply ageing, that enzyme can be partly blocked by nitric oxide, and the cell's energy output, signalling and repair capacity fall with it. Tissues under this kind of strain include sore joints, over-worked muscle, photo-aged skin and the lining of the mouth after chemotherapy or radiotherapy. The common thread in all of these is a cell that is still alive but working below its potential.

Where light comes in

Red light (roughly 630 to 660 nm) and near-infrared light (roughly 800 to 850 nm) sit in a narrow window where skin and water absorb relatively little, so photons can travel millimetres to a few centimetres into tissue. Cytochrome c oxidase absorbs light in exactly this window. The leading model, set out by Karu and by de Freitas and Hamblin, is that absorbed photons displace inhibitory nitric oxide from the enzyme, briefly raising oxygen use, ATP production and a small, signalling-level burst of reactive oxygen species. Downstream, cells switch on transcription factors that govern repair, growth factor release and the resolution of inflammation. Light-sensitive ion channels and effects on local blood flow are also proposed. The response follows a biphasic curve: too little light does nothing measurable, a moderate dose is where effects are reported, and higher doses can blunt or reverse them.

How solid is this

Mixed, and it depends heavily on the use. For preventing oral mucositis in specific cancer treatment settings, the evidence is strong enough that an international oncology body (MASCC/ISOO) issues formal recommendations with exact device settings. For knee osteoarthritis pain, a 2019 meta-analysis of 22 placebo-controlled trials reported meaningful pain reductions when doses matched published guidelines, and smaller or no effects when they did not. For skin appearance, controlled trials report improvements in collagen density and wrinkle scores, but several are industry-funded and use no-treatment rather than sham controls. Evidence for sleep, mood, cognition, thyroid function and body composition ranges from early to weak, and whole-body panel studies are few. Across the field, the biggest problem is inconsistent dosing and reporting, so a positive trial with one device does not automatically transfer to another.

Explainer illustration

The studies

Every study is laid out the same way: what the researchers wanted to find out, who took part, what they actually did, what they found, why it might work and where the limits are. No jargon, and a link to the original if you want to read it yourself.

Each study is laid out the same way: what the researchers asked, who took part, what they did, what they found, why it might work, and where the limits are. The clinician layer at the end of each one carries the numbers.

Paper 1

Efficacy of low-level laser therapy on pain and disability in knee osteoarthritis: systematic review and meta-analysis of randomised placebo-controlled trials

Stausholm et al. · BMJ Open · 2019 · Systematic review and meta-analysis of 22 randomised placebo-controlled trials, n=1,063

Read the paper

What they asked

Knee osteoarthritis is one of the commonest causes of chronic pain in adults, and most treatment guidelines do not recommend low-level laser or LED therapy for it, largely because earlier reviews found inconsistent results. The Bergen group behind this review suspected that the inconsistency came from dose rather than from the light itself. Some trials used very small energies, others used much larger ones, and few had been judged against published dosing recommendations. So they asked a sharper question: across all the placebo-controlled trials, does red or near-infrared light reduce pain and disability in knee osteoarthritis, and does the answer change when you separate trials that used the World Association for Photobiomodulation Therapy's recommended doses from those that did not?

Who took part

The review pooled 22 randomised trials involving 1,063 adults with a diagnosis of knee osteoarthritis. Participants were typically middle-aged or older, and many had lived with knee pain for years. Every included trial compared active light against a placebo device that looked and sounded identical but emitted no therapeutic light, which is the key design feature that lets researchers separate a genuine effect from the reassurance of being treated. Trials came from several countries and used a mix of laser and LED devices in the red and near-infrared range. The reviewers registered their protocol in advance on PROSPERO (CRD42016035587), which limits the temptation to change the question once the data are in.

What they did

The team searched the major medical databases for randomised placebo-controlled trials, extracted pain scores on a 0 to 100 mm visual analogue scale and disability scores, and assessed each trial for risk of bias. They then did something few earlier reviews had done: they read the dosing details of each trial and classified it as either following or not following the recommended dose window, which for knee osteoarthritis means 4 to 8 joules per treatment point at 785 to 860 nm, or 1 to 3 joules per point at 904 nm. Results were pooled separately for the end of treatment and for follow-up visits up to twelve weeks later, so that any lasting effect could be seen rather than only the immediate one.

What they found

Across all trials, pain was lower with active light than with placebo by about 14 mm on the 100 mm scale at the end of treatment and about 16 mm at follow-up. In the trials that used recommended doses the difference was larger: roughly 19 mm at the end of treatment, 23 mm across follow-ups, and a peak of about 32 mm two to four weeks after the course ended. Trials using doses outside the recommended window showed much smaller effects. Disability was also lower in the active groups, with a moderate effect at the end of treatment growing to a large effect at follow-up. No adverse events were reported in any of the included trials. The reviewers judged most trial methods to be adequate.

Why it might work

Osteoarthritic knees are not just worn cartilage; the joint lining is inflamed, the surrounding muscle is guarded and weak, and pain nerves are sensitised. Near-infrared light in the 800 to 900 nm range penetrates several millimetres to a few centimetres, enough to reach the joint capsule and the superficial parts of the joint. In laboratory and animal studies, light at these doses reduces inflammatory signalling molecules, lowers swelling, and appears to slow conduction in the small nerve fibres that carry pain. The finding that effects peaked weeks after the course ended, rather than during it, fits a biological repair process better than a simple painkilling one, and is one of the more interesting observations in this review.

Limits

Meta-analyses inherit the weaknesses of the trials they pool. Many individual trials were small, follow-up rarely went beyond three months, and the dose categories were applied after the fact, which is a reasonable approach but not the same as a trial designed to test dose head-to-head. The lead authors are long-standing researchers in this field and two were former board members of the association whose dose recommendations were being tested, though they report no funding from it. Pain scales are subjective, and blinding with light devices is imperfect because members can sometimes feel warmth from active units. The findings apply to knee osteoarthritis treated with targeted laser or LED probes at the knee, not to whole-body panels or to other joints.

Paper 2

A controlled trial to determine the efficacy of red and near-infrared light treatment in patient satisfaction, reduction of fine lines, wrinkles, skin roughness, and intradermal collagen density increase

Wunsch et al. · Photomedicine and Laser Surgery · 2014 · Randomised controlled trial with an untreated control group, n=136

Read the paper

What they asked

Most early skin studies of red light used small hand-held devices or facial panels on a few dozen people. This German trial asked whether large-area, full-body light units of the kind found in wellness settings could improve how skin looks and feels, and whether those changes could be measured objectively rather than only reported by the person in the light. Two lamp types were compared: one emitting a fairly narrow red band around 611 to 650 nm, and a broader polychromatic source spanning 570 to 850 nm that included both red and near-infrared. The researchers wanted to know whether either would raise collagen density in the dermis, smooth skin roughness, reduce wrinkle scores, and whether volunteers would notice the difference themselves.

Who took part

One hundred and thirty-six volunteers took part, and 128 completed the trial. Fifty-seven were treated with the red light units, forty-eight with the broader-spectrum units, and twenty-three formed an untreated control group who came in for the same measurements but received no light. Participants spanned a wide adult age range and had no acute skin disease. Because this was a trial of skin appearance rather than a medical condition, the volunteers were broadly healthy adults, which makes the results reasonably relevant to members using panels for cosmetic reasons. It is worth noting that the control group received no sham treatment, so participants knew whether they were being treated.

What they did

Treated volunteers stood in front of or lay beneath the light units twice a week for fifteen weeks, thirty sessions in total. Exposure times were adjusted so that each unit delivered a similar dose in the red band, roughly 9 joules per square centimetre per session at an irradiance of about 6 to 13 milliwatts per square centimetre. At the start, midway and at the end, the team photographed faces under standardised conditions, measured skin roughness with a silicone-cast profilometry technique, and used high-frequency ultrasound to measure the density of collagen in the dermis. Blinded assessors graded wrinkles from the photographs, and volunteers rated their own skin complexion and skin feeling on visual analogue scales.

What they found

Both light groups improved on every measure between the first and thirtieth session, with statistically significant gains in self-rated complexion and skin feeling, ultrasound collagen density, skin roughness and blinded wrinkle scores. The untreated control group did not improve and in fact drifted slightly worse on roughness and wrinkle status over the fifteen weeks. Collagen density scores rose by around six units in the treated groups against a small fall in controls. Blinded assessors judged about seven in ten treated faces as improved compared with about one in twenty-five controls. There was no meaningful difference between the narrow red and the broader-spectrum units, suggesting that the red band was doing most of the work. No serious adverse events occurred; a few volunteers noticed temporary visibility of small blood vessels.

Why it might work

Skin ageing involves slower collagen production by fibroblasts, more collagen breakdown by enzymes called matrix metalloproteinases, and a duller, less even micro-circulation. Red light at 630 to 660 nm is absorbed by cytochrome c oxidase in fibroblast mitochondria, and in laboratory studies this is followed by more ATP, more procollagen synthesis and less metalloproteinase activity. Improved local blood flow and a mild anti-inflammatory effect in the dermis are also proposed. The ultrasound finding of denser dermal collagen is the most convincing part of this trial because it does not depend on anyone's opinion, and it is consistent with what the cell biology would predict.

Limits

The most important limit is that the control group was untreated rather than given a sham light, so expectation could have influenced the self-rated and even the photographic outcomes, though not easily the ultrasound. The trial was fully funded by the manufacturer of the light units, and the principal investigator was paid by the sponsor to run it. The follow-up was short, so how long the changes last is unknown. Randomisation and allocation concealment are not described in detail. The dose used, about 9 J/cm² twice a week, is specific to these units, and results from a different panel with a different irradiance and spectrum cannot be assumed to be identical.

Paper 3

Systematic review of photobiomodulation for the management of oral mucositis in cancer patients and clinical practice guidelines

Zadik et al. · Supportive Care in Cancer · 2019 · Systematic review underpinning international guidelines, multiple randomised trials

Read the paper

What they asked

Oral mucositis is the painful ulceration of the mouth lining that affects most people having high-dose chemotherapy before a stem cell transplant or radiotherapy to the head and neck. It can be severe enough to stop people eating, require opioid painkillers or feeding tubes, and interrupt cancer treatment. The Multinational Association of Supportive Care in Cancer and the International Society of Oral Oncology (MASCC/ISOO) periodically review every intervention for mucositis. This paper is their dedicated review of photobiomodulation. It asked whether red or near-infrared light applied inside the mouth can prevent or reduce mucositis, in which cancer treatment settings, and at what exact device settings, so that a clear recommendation could be written for oncology teams.

Who took part

This is a systematic review rather than a single trial, so the participants are those of the trials it gathered: adults and children with cancer who were about to receive, or were receiving, treatments known to cause mucositis. The three main groups were adults having a haematopoietic stem cell transplant after high-dose chemotherapy with or without total body irradiation, adults having radiotherapy to the head and neck without chemotherapy, and adults having combined chemotherapy and radiotherapy to the head and neck. Earlier pooled analyses in this area covered more than a thousand people across eighteen randomised trials, and the evidence base has grown since. Children were considered separately because dosing and safety data were thinner.

What they did

A multinational panel of oral medicine specialists, oncologists and photobiology researchers searched the literature for all studies of light therapy in mucositis, graded each for quality and risk of bias using a standard framework, and then judged the overall strength of evidence for each cancer treatment setting. Crucially, they did not lump all light devices together. For each recommendation they specified the wavelength, power density, energy per site, spot size, number of intraoral sites and the treatment schedule that the supporting trials had used. Where studies disagreed or used incomparable settings they said so and declined to recommend. The review fed directly into the 2020 MASCC/ISOO clinical practice guidelines published in the journal Cancer.

What they found

The panel concluded that the evidence supports using specific settings of photobiomodulation to prevent oral mucositis in three populations: adults having a stem cell transplant after high-dose chemotherapy, adults having head and neck radiotherapy alone, and adults having combined head and neck chemo-radiotherapy. In the 2020 guideline these became formal recommendations, two of them at the highest evidence level, with detailed protocols using 632.8 nm, 650 nm or 660 nm light at 1 to 6 joules per square centimetre per site. For the treatment of mucositis once it has already developed, and for mucositis from chemotherapy without transplant, the evidence was judged insufficient. An earlier meta-analysis of eighteen randomised trials had estimated that preventive light therapy cut the risk of severe mucositis by roughly 60 per cent.

Why it might work

Chemotherapy and radiotherapy generate a burst of reactive oxygen species and inflammatory signalling in the fast-dividing cells of the mouth lining, which then break down into ulcers. Red light at 630 to 660 nm absorbed by mitochondria in those cells appears to support their energy production and survival, damp down the inflammatory cascade before ulcers form, and speed re-epithelialisation of tissue that has already broken down. It also seems to reduce pain signalling locally. Because the mouth lining is only a fraction of a millimetre thick and the light is applied directly to it, dose delivery is far more predictable here than in deeper tissues, which is one reason the evidence is more consistent than in almost any other use of light therapy.

Limits

The recommendations are tightly specific: particular wavelengths, energies, spot sizes and schedules applied inside the mouth by trained staff. They cannot be extrapolated to consumer panels or to applying light to the cheeks from outside. The evidence is strongest for prevention and much weaker for treating established ulcers. Long-term follow-up for tumour safety is still limited, and the panel advised caution when the tumour itself lies in the mouth. Individual trials varied in blinding quality and some were small. Several panel members research or consult in this field. For members of Breathe, the relevance is not that panels will do this, but that this is the setting where light therapy has been most rigorously tested and adopted into mainstream oncology care.