Woman reweaving her portrait tapestry with rose and gold threads beside a sunlit workshop window.
Back to the library

Recovery

Red light and skin

Skin is the use with the most consistent human evidence for red light, and it is also the tissue the light actually reaches. Controlled trials have reported denser dermal collagen on ultrasound, smoother skin and shorter wrinkles after courses of two or three sessions a week for one to fifteen weeks. The effects reported are modest, build over weeks, and several of the larger trials were funded by the companies that make the devices. Nothing here is a treatment for a skin condition, and none of it happens after a single session.

Evidence so far
Moderate
Reading timeAbout 12 minutes

Start here

Why this is being studied

Red light for skin means exposing bare skin to red light at around 630 to 660 nm at doses far too low to heat it. Red light is absorbed strongly in the outer layers of skin, which is the one place where a panel reliably delivers the sort of dose used in published trials. The working idea is that light absorbed by an enzyme in the mitochondria of skin cells nudges collagen-producing fibroblasts back towards more active repair, and this is the use where human trials most often line up with that idea.

Where the evidence stands

Flag text

The mechanism

What is going on, and why this might help

What the light is absorbed by

Red light at roughly 630 to 660 nm is absorbed inside skin cells by cytochrome c oxidase, the final enzyme in the mitochondrial chain that hands electrons to oxygen so energy can be made. Fibroblasts, the cells that build collagen, sit in the dermis a fraction of a millimetre to a couple of millimetres down, which is comfortably within reach of red light. This is the main reason the skin evidence is the tidiest in the field: the dose that arrives is close to the dose intended.

Nitric oxide, ATP and signalling

In stressed or ageing cells, nitric oxide can sit on cytochrome c oxidase and slow respiration. The leading model is that absorbed photons displace that nitric oxide, so oxygen use and ATP production briefly rise, nitric oxide released locally relaxes small blood vessels and increases blood flow, and a small burst of reactive oxygen species acts as a signal rather than as damage. Downstream, transcription factors that govern repair switch on, and laboratory work reports more procollagen made and less activity from the enzymes that break collagen down.

The biphasic dose response

Light does not behave like a dose of something where more is stronger. The published curve rises from no effect below a threshold, plateaus, then falls away and can reverse at higher doses. Skin trials cluster in a narrow window of roughly 5 to 20 joules per square centimetre per session, and the 2025 frequency trial found three sessions a week no better than two. Standing closer for longer is not a way to accelerate anything.

How far the light actually goes

Tissue is relatively transparent between about 600 and 1,100 nm, but red light is still scattered and absorbed by haemoglobin and melanin within the first few millimetres. For skin this is an advantage rather than a limitation. It also means melanin matters: members with darker skin absorb proportionally more red light nearer the surface, which is a physical fact rather than a studied clinical question. Sunscreen, make-up and thick moisturiser sit in the light path and reduce what gets in.

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

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 · Controlled trial with an untreated control group, n=136

Read the paper

What they asked

Most red light skin research up to this point used small facial devices on a few dozen people. This German trial asked whether large-area units of the kind used in wellness settings could change how skin looks and feels, and whether any change could be measured by instruments rather than only by opinion. It also compared a narrow red band with a broader source that added near-infrared, to see whether the extra wavelengths mattered.

Who took part

One hundred and thirty-six volunteers enrolled and 128 finished. Fifty-seven used the narrow red units, forty-eight the broader-spectrum units, and twenty-three formed an untreated comparison group who came in for the same measurements but received no light. They were broadly healthy adults across a wide age range with no acute skin disease, which makes them reasonably close to the sort of person who would use a panel for cosmetic reasons.

What they did

Treated volunteers had thirty sessions, twice a week for fifteen weeks. Exposure times were set so each unit delivered roughly 9 joules per square centimetre per session at about 6 to 13 milliwatts per square centimetre. At the start, midway and at the end the team took standardised photographs, measured skin roughness using silicone casts, and used high-frequency ultrasound to measure collagen density in the dermis. Assessors grading the photographs did not know who had been treated.

What they found

Both light groups improved on every measure across the fifteen weeks, while the untreated group did not and drifted slightly worse on roughness and wrinkle scores. Ultrasound collagen density rose in the treated groups and fell slightly in the comparison group. Blinded assessors judged roughly seven in ten treated faces improved, against about one in twenty-five of the untreated. The narrow red and the broader-spectrum units performed much the same, which suggests the red band was doing most of the work.

What it means

This is the largest controlled trial of large-area red light for skin, and its most persuasive number is the ultrasound one, because a machine measuring collagen density does not know what anyone hoped for. It supports the idea that a course of red light at a modest dose may support the skin directly under the light. It says nothing about deeper tissue, and it describes an appearance change over months rather than anything clinical.

Limits

The comparison group received no sham light, so everyone knew whether they were being treated, and expectation could have shaped the self-rated and photographic results. The trial was funded by the manufacturer of the units and the lead investigator was paid by the sponsor. Follow-up was short, so how long the changes last is unknown, and the dose used belongs to those specific units rather than to panels in general.

Paper 2

Role of photobiomodulation application frequency in facial rejuvenation: randomized, sham-controlled, double-blind, clinical trial

Bragato et al. · Lasers in Medical Science · 2025 · Randomised sham-controlled double-blind trial, n=95

Read the paper

What they asked

Device makers often suggest daily or near-daily use, and members reasonably want to know whether more sessions produce more benefit. This Brazilian team asked two questions at once: does a red LED mask reduce facial wrinkles compared with a sham mask over four weeks, and does using it three times a week beat using it twice. It is one of the few red light trials designed around frequency rather than just presence or absence of light.

Who took part

Ninety-five women took part, randomly allocated between active and sham masks and between the two frequencies. Both the women and the assessors measuring their skin were kept unaware of which mask was which, which is the main reason this trial carries more weight than a larger but unblinded one. Participants were recruited for facial wrinkles rather than any skin disease.

What they did

Active masks delivered 660 nm light at about 6.4 milliwatts per square centimetre for a dose of roughly 8 joules per square centimetre per session. Sham masks were identical in appearance. Sessions ran either two or three times a week for four weeks. Wrinkle length in the glabellar area between the brows and around the eyes was measured from standardised images before and after the course.

What they found

Researchers found shorter glabellar and periorbital wrinkles in the active groups compared with sham after four weeks. Three sessions a week produced no advantage over two. No meaningful adverse effects were reported.

What it means

This is useful in two ways. It is a properly blinded trial reporting a measurable skin effect from red light, which strengthens a literature otherwise full of unblinded studies. It also puts a number on the plain rule that more is not automatically better: two sessions a week did the same job as three, which is consistent with how dose appears to behave across this field.

Limits

Four weeks is short and there was no follow-up to show whether anything persisted. The changes measured are small distances on an image rather than something a member would necessarily notice in a mirror. It studied women only, used a close-fitting face mask rather than a panel at a distance, and tested a single wavelength and dose, so it cannot say what other settings would do.

Paper 3

A prospective, randomized, placebo-controlled, double-blinded, and split-face clinical study on LED phototherapy for skin rejuvenation

Lee et al. · Journal of Photochemistry and Photobiology B · 2007 · Randomised placebo-controlled double-blind split-face trial

Read the paper

What they asked

The question here was whether visible changes in skin after LED light are accompanied by changes in the skin's actual structure. Earlier work could show that people liked how their skin looked, but not what had happened underneath. The team asked whether red light at 633 nm, near-infrared at 830 nm, or the two combined would improve wrinkles and elasticity, and whether biopsies would show more collagen and less of the enzymes that break collagen down.

Who took part

Adult volunteers with periorbital wrinkles took part in a split-face design, meaning each person had one side of the face treated and the other side given a placebo light. That design is powerful because every participant acts as their own comparison, removing differences in age, skin type, sun exposure and lifestyle that otherwise muddy small studies. Both the participants and the assessors were kept unaware of which side was which.

What they did

Participants were randomised to one of several LED arms: 633 nm, 830 nm, a combination of both, or placebo, over a course of sessions spanning several weeks. Outcomes included profilometry of wrinkle depth, measured skin elasticity, melanin and clinical photographs. A subset of participants gave small skin biopsies before and after, which were examined for collagen and for matrix metalloproteinases, the enzymes that degrade it.

What they found

The study reported improved wrinkle appearance and better skin elasticity on the treated sides compared with placebo sides. The biopsies showed increased collagen in the dermis and reduced matrix metalloproteinase activity in treated skin. Both the red and near-infrared arms showed effects, with the combination performing well.

What it means

This is the study most often cited for the idea that red light may support collagen rather than just make skin look temporarily better. Because the comparison sat on the other half of the same face and the tissue itself was examined, it is harder to explain the findings by expectation alone. It remains the mechanistic anchor beneath the larger appearance trials.

Limits

The sample was small, as biopsy studies usually are, and it is now nearly two decades old. Light in a split-face design can scatter across the midline, so the placebo side may not have been entirely untreated. Histology from a handful of samples is sensitive to where the biopsy was taken. The findings apply to facial skin under close LED exposure, not to deeper tissue or to a whole-body panel at a distance.