Red Light Therapy Side Effects: What the Evidence Actually Shows

Written by the Lumiglo research team. Published July 2026. Every figure below is sourced, and where the evidence is thin we say so.

Red light therapy side effects are mostly mild and mostly avoidable. The problem is that almost every guide on this subject lists a single case report, a manufacturer's precaution and a debunked myth in the same bullet list, as if they carried equal weight.

If you are weighing whether the benefits justify the risks, we went through every claimed benefit and its evidence in a separate guide. This one is only about what can go wrong.

This guide separates them. Each risk below is labeled with how much evidence sits behind it, from a documented injury in a human being down to a myth that will not die.

What is in this guide

• Red, near-infrared and blue are not the same thing
• How we rated the evidence in this guide
• Common red light therapy side effects
• Eye safety: what is documented and what is assumed
• Burns, blisters and overdosing
• Melasma, hyperpigmentation and deeper skin tones
• Photosensitising medication: the most overlooked risk
• Who should not use red light therapy
• Contraindications that are not what people think
• At-home devices against professional treatments: which is riskier
• What we still do not know
• LED face mask safety: a practical checklist
• The honest summary
• Frequently asked questions

Red, near-infrared and blue are not the same thing

This distinction decides almost everything that follows, and almost no guide makes it properly.

• Red, roughly 630 to 660 nm. Visible, low energy per photon, penetrates a few millimetres. This is the band behind the skin research.
• Near-infrared, roughly 830 to 850 nm. Invisible, penetrates deeper. Same low-energy category.
• Blue, roughly 415 to 470 nm. Visible, higher energy per photon, and the band that carries the documented eye risk and most of the pigmentation concern.

All three are non-ionizing. None of them is ultraviolet, none damages DNA the way UV does, and none of them is a tanning bed. That much is consensus across every medical source on this subject.

But "it is not UV" is where most articles stop, and that is too early. Non-ionising does not mean inert. It means the risks are different ones: heat, photosensitisation, pigment response and, for blue specifically, the retina.

How we rated the evidence in this guide

Four levels, applied to every risk below.

If a guide tells you something is dangerous without telling you which of these four it is, it is not helping you decide anything.

Common red light therapy side effects, and what they mean

Across the medical sources, the consensus is that side effects are mild and uncommon when devices are used as directed. Here is what actually gets reported.

Redness, warmth and dryness

The most frequent, and usually the least worrying. A flush that fades within an hour after a session is a normal vascular response. Redness or swelling that persists beyond two hours is the signal to stop. That is the threshold dermatology-adjacent brands use as a practical test for light sensitivity, and it is a good one.

An initial breakout

Sometimes reported in the first weeks. Poorly documented, plausibly related to increased circulation and cell turnover. If it persists past three or four weeks, that is not a purge, that is a reaction.

Headaches and eye fatigue

Usually from light leaking around the eye cups over a long session. Almost always solved by closing your eyes and lying down. Some manufacturers list light-induced headaches as an outright contraindication, which is worth knowing if you get migraines with visual triggers.

Disrupted sleep

Rarely mentioned, occasionally reported after late-evening sessions. If you use a mask at eleven at night and then cannot settle, move the session earlier rather than assuming it is unrelated.

Eye safety: what is documented and what is assumed

This is where the entire category gets it wrong, and the correction matters.

The published case of retinal injury from a consumer mask, and it was blue light

A case report describes photochemical retinopathy in a 37-year-old woman after using an LED mask emitting 460 to 470 nm blue light, twenty minutes every other day for a month, with her eyes open, following the manufacturer's instructions. Imaging showed damage to the retinal pigment epithelium. She was treated and improved over four weeks, with some persistent change. For completeness: that mask also emitted red and infrared, so the blue attribution comes from the authors' analysis rather than from the device being blue-only.

The authors' own conclusion is the part nobody quotes: they recommend red or infrared light, at longer wavelengths, for retinal safety.

What happened when researchers pointed red light directly at the retina

Red light at 650 nm has been used deliberately and directly on the retina in myopia research in children. A systematic review of that work found no irreversible loss of visual function and no structural ocular damage. Two case reports of reduced acuity resolved completely within four months of stopping. The most common effect was an afterimage lasting under six minutes.

And the case that complicates the picture

Honesty requires this one too. JAMA Ophthalmology published a case of retinal damage after repeated low-level red-light exposure in 2023. It involved a device directing light into the pupil in a myopia protocol, not a face mask, and it is a single case against a literature that has otherwise been reassuring. But it exists, and an article claiming the risk is exclusively blue would be doing the selective quoting we are complaining about.

So the accurate statement is narrower than the one usually made: the documented eye injury from a consumer LED face mask involved blue light. Red light directed into the eye has its own small literature, mostly reassuring, with at least one reported injury. Guides that write "red light can damage your eyes" without distinguishing wavelength, device type and exposure route are collapsing three different questions into one.

What that means for your mask

• Never stare into the diodes, whatever the color. That is not a wavelength issue, it is a brightness issue.
• Close your eyes and lie down. It is more comfortable and it removes the leakage question.
• If your mask has a blue mode, treat eye protection as non-negotiable for that mode specifically.
• Ask the manufacturer about IEC 62471, the photobiological safety standard for lamps. A brand that has tested against it can tell you. A brand that has not should not be claiming its device is safe for the eyes.

Burns, blisters and overdosing

Evidence level: documented, but through misuse and hardware failure rather than normal use.

Reported burns in this category share a profile: sessions far longer than recommended, falling asleep with the device on, or devices with broken wiring or corrosion. The reported cases involve intensities and session lengths an order of magnitude beyond anything a consumer mask delivers.

The biphasic dose response, which is the real story

Photobiomodulation does not get better with more. Below a threshold, nothing happens. Above an optimum, the effect flattens and then reverses. This is well described in the literature and it is the single most useful thing to understand about using these devices.

We went through what each brand publishes on power and dose in our comparison of nine LED face masks, including the ones that publish nothing at all.

Practically, for a home mask, the risk of overuse is not permanent injury. It is diminished returns plus short-term sensitivity. Thirty minutes instead of ten has not been reported to cause lasting harm at consumer intensities. What it reliably does is waste twenty minutes and risk irritating your skin.

Signs you are overdoing it

• Redness that persists more than two hours after a session
• A warm or tight feeling that does not settle
• New dryness or flaking that was not there before you started

Any of those means reduce frequency or session length, not push through.

Melasma, hyperpigmentation and deeper skin tones

Evidence level: contradictory, and this is the worst-covered subject in the whole category. Several major medical pages on red light therapy side effects do not mention it at all.

The case for concern

High-energy visible light, roughly 400 to 450 nm, is documented to induce pigmentation in Fitzpatrick skin types III and above, and can worsen melasma and post-inflammatory hyperpigmentation. The American Academy of Dermatology recommends that people with deeper skin tones speak to a dermatologist before using an at-home red-light device, noting that dark spots can be more intense and longer lasting. Consumer Reports puts it plainly: the use of LED masks in people of color has not been extensively studied.

Heat may matter more than wavelength

An argument that deserves more attention: the most likely mechanism for aggravating pigmentation with a high-intensity LED device may be the heat it produces rather than the specific wavelength. That would make device design and session length more relevant than the color on the box.

And the case on the other side

A pilot study has reported that pulsed photobiomodulation improved dermal melasma. So the same technology appears in the literature as both a possible aggravator and a possible treatment.

Our honest position: the evidence is too thin to tell you which way it goes for your skin. If you have melasma or a history of pigmentation changes, this is the one situation where we would rather you asked a dermatologist before buying anything, from us or anyone else.

Photosensitising medication: the most overlooked risk

Evidence level: plausible, well established mechanistically, and the risk most likely to actually apply to you.

Several hundred substances are classified as photosensitizing. Most guides list four of them.

The honest nuance: photosensitisation is predominantly a UV-driven phenomenon. Whether it translates fully to visible red light at consumer intensities is not well established. That is a reason to test, not a reason to panic.

The patch test, in three steps

1. Run a full three-minute session on the inside of your forearm rather than your face.
2. Look at the area immediately afterwards.
3. If redness appears and lasts more than two hours, you have a light sensitivity. Stop, and speak to a healthcare professional before going further.

Do this once, before your first facial session. It costs three minutes.

Who should not use red light therapy

Not one of the highest-ranking medical pages on this subject has a section with this title, despite it being one of the most searched questions in the category.

Pregnancy and breastfeeding, evidence level: no data

Manufacturers contraindicate it almost universally, and the honest reason is that these devices have not been tested in pregnant or breastfeeding women. That is an absence of evidence, not a demonstration of harm, and it deserves to be stated that way rather than as a warning. Ask your doctor.

Epilepsy and light-triggered seizures, evidence level: plausible, and device-dependent

A minority of people with epilepsy have the photosensitive form, and the flash frequencies most often implicated sit in the range of a few to a few dozen per second. The point almost nobody makes: this concerns pulsed or flashing devices. A mask running in continuous-wave mode does not flash. If you have photosensitive epilepsy, ask the manufacturer whether the device pulses, and treat a non-answer as a no.

Photosensitive conditions, evidence level: plausible

Lupus, porphyria, albinism, xeroderma pigmentosum and photosensitive eczema. Speak to your doctor before using any light device.

Active cancer or undiagnosed lesions, evidence level: precautionary

Every manufacturer says do not treat over moles, warts, open sores or suspicious lesions, and the honest reason is rarely given: photobiomodulation stimulates cellular metabolism, and nobody knows what that does to cells that are already transformed. It is an unknown rather than a demonstrated danger, and unknown is enough.

Eye conditions and light-induced headaches

Glaucoma, cataract, recent eye surgery, inherited retinal conditions. And if bright light triggers your headaches, several manufacturers list that as an outright contraindication.

Contraindications that are not what people think

Five things you will read on forums and on badly researched product pages. None of these is a reason to override your own doctor, and where the honest answer is "not for the reason usually given" we say that rather than calling it a myth.

At-home devices against professional treatments: which is riskier

The usual answer is that clinic devices are more powerful, so home devices are weaker but safer. That is roughly right and it misses the more interesting point.

The clearest statement on this came from Stanford: efficacy depends on wavelength, duration and frequency, which is largely unknown when people buy tools for use at home. Harvard makes the same point from the other side, that professional versions can better control the dose across a greater range. The issue is not raw power. It is dose control.

And there is one area where a well-designed home mask may genuinely be safer. Clinic panels are not built to spare your eyes, which is why patients are given opaque shields. A mask that sits against the face, with molded eye cups, physically blocks more light than a panel aimed at you from a distance. Nobody writes this, and it is true.

The real home risk is neither power nor eyes. It is the absence of a diagnosis. A clinician looks at your skin before treating it. Buying a device online skips that step, which is why the pigmentation section above matters more than the burns section.

What we still do not know

Every medical source on this subject admits the following, usually in half a sentence buried near the end. It deserves its own section.

• Long-term safety. There is no long-term safety data on daily or weekly LED mask use over years. Not for our device, not for anyone's. The American Academy of Dermatology states plainly that the long-term effects on skin and hair are not known.
• Optimal dose. Published protocols vary by three orders of magnitude. Harvard notes that researchers have not pinned down how long a session should last or what dose suits which skin concern.
• Deeper skin tones. Under-studied, and the one area where the potential downside is a visible, lasting pigment change.
• What actually reaches your skin. Most brands publish an irradiance figure without the distance it was measured at, which makes the number close to meaningless.

And what "FDA-cleared" does not tell you

These devices are Class II and go through a 510(k) clearance, which establishes substantial equivalence to a device already on the market. The American Academy of Dermatology puts it better than we could: FDA cleared tells you nothing about how effective the device is. And the terms "FDA approved" or "FDA certified" have no meaning when used to describe this kind of device.

So a cleared device is not a proven one, and an uncleared device is not a dangerous one. It means the manufacturer did or did not file.

LED face mask safety: a practical checklist

1. Patch test on your forearm before the first facial session. Three minutes, then check at two hours.
2. Check your medication and your skincare against the table above. If anything on it is in your routine, ask a pharmacist.
3. Close your eyes. Every session, whatever the color. Never look into the diodes.
4. Clean, dry skin. Serums and moisturiser go on afterwards.
5. Ten minutes. Not thirty. On a device at around 27 mW/cm² that is roughly 16 J/cm², inside the range the published protocols used. Longer is not better and can irritate.
6. Stop if redness lasts beyond two hours. That is your threshold, not a general guideline.
7. Do not treat moles, lesions or broken skin. Get anything undiagnosed looked at first.
8. If you have melasma, deeper skin tone, epilepsy, a photosensitive condition, or you are pregnant, ask a professional before buying.

The honest summary

Sort everything above by evidence level and the picture is not complicated. Documented: one retinal injury from a consumer mask, and it was blue light. Plausible: interactions with photosensitizing medication, and pigment changes in deeper skin tones. No data: pregnancy, and long-term use of any of these devices by anyone. Not for the reason usually given: pacemakers, deep implants, tattoos.

Which means the risk you can actually act on today is not exotic. It is your medication list and your bathroom shelf. Check both, run the three-minute patch test, close your eyes, and keep sessions short.

Frequently asked questions

Is red light therapy safe?

For most people, yes, when used as directed. The reported side effects are mild and mostly avoidable: transient redness, warmth, occasional headaches from light leakage. The risks worth taking seriously are photosensitizing medication, pigment changes in deeper skin tones, and eye exposure with blue modes. What nobody can tell you is long-term safety, because no long-term data exists for any device in this category.

What are the risks and dangers of red light therapy?

Ranked by how much evidence sits behind them: a documented retinal injury from a blue-light mask, plausible interactions with photosensitizing medication and skincare actives, plausible pigment changes in Fitzpatrick types III and above, burns reported through misuse or faulty hardware rather than normal use, and a set of widely repeated contraindications such as pacemakers where the usual mechanism does not hold up. Long-term risk is unknown.

Is red light therapy bad for your eyes?

The one documented case of retinal injury from an LED mask in the published literature involved blue light at 460 to 470 nm, used with eyes open. The authors of that report specifically recommended red and infrared as safer for the retina. A systematic review of red light aimed directly at the retina in myopia research found no irreversible damage. Close your eyes anyway, and never stare into the diodes.

Who should not use red light therapy?

Anyone pregnant or breastfeeding, because it has not been tested rather than because harm is known. People with photosensitive epilepsy, particularly with pulsed devices. People with lupus, porphyria or other photosensitive conditions. Anyone with active cancer or undiagnosed lesions in the treatment area. Anyone taking photosensitizing medication. And anyone with melasma or a deeper skin tone should speak to a dermatologist first.

Can red light therapy cause cancer?

There is no known mechanism by which it could. Red and near-infrared light is non-ionizing and contains no ultraviolet, so it does not cause the DNA damage that drives UV-related skin cancer. This is one of the few points on which every medical source in this field agrees. What is contraindicated is treating over an existing suspicious lesion, because nobody knows what stimulating cellular metabolism does to cells that are already abnormal.

Can you overdo red light therapy?

Yes, but the consequence reported for home masks is diminished returns and temporary sensitivity rather than lasting harm. Photobiomodulation follows a biphasic dose response: below a threshold nothing happens, above an optimum the effect flattens and then declines. Thirty minutes does not do three times what ten minutes does. It mostly wastes twenty minutes.

Can red light therapy make melasma worse?

The evidence points both ways, which is the honest answer. High-energy visible light is documented to induce pigmentation in deeper skin tones and can worsen melasma, and the heat a device produces may matter more than its wavelength. At the same time, a pilot study reported that pulsed photobiomodulation improved dermal melasma. If you have melasma, speak to a dermatologist before buying any light device.

Is it safe to use red light therapy every day?

Daily use has not been reported to cause harm at these intensities, but it is not more effective either. Published protocols used two sessions a week over four to twelve weeks. Three to five is a reasonable practical ceiling. If you get redness that persists beyond two hours, reduce frequency rather than pushing through.

Do I need to wear goggles?

For red and near-infrared, closing your eyes and lying down is enough for most people, and goggles are a sensible addition if you are light-sensitive. For blue light modes, treat eye protection as non-negotiable, because that is the band carrying the documented risk. Check whether goggles are included, because with most brands they are not.

Are there long-term side effects of red light therapy?

Nobody knows, and any brand telling you otherwise is guessing. There is no long-term safety data on years of regular LED mask use, for any device on the market. The American Academy of Dermatology states that the long-term effects on skin and hair are not known and that more research is needed.

Is an LED mask with a blue mode more risky?

On eye safety specifically, yes. Blue light at 415 to 470 nm carries the only documented case of retinal injury from a consumer mask, and it is also the band associated with pigment induction in deeper skin tones. That does not make blue modes useless, they have evidence behind them for blemish-prone skin. It makes eye protection and caution around pigmentation genuinely important when you use them.

Does "FDA-cleared" mean an LED mask is safe and effective?

It means neither, exactly. A 510(k) clearance establishes substantial equivalence to a device already on the market under Class II. The American Academy of Dermatology puts it directly: FDA cleared tells you nothing about how effective the device is. And "FDA approved" has no meaning for this kind of device at all.

Keep reading

• Best LED face masks of 2026, the specs brands publish and the ones they hide.
• Dull skin: seven things that actually restore radiance, where LED fits in an evening routine.
• How red light works at the cellular level, the mitochondrial mechanism without the marketing.
• Red light and hair, the same wavelengths, a different application.

Sources: Wunsch A, Matuschka K. Photomed Laser Surg. 2014;32(2):93-100. Zein R, Selting W, Hamblin MR. J Biomed Opt. 2018;23(12):120901. Huang YY, Chen ACH, Carroll JD, Hamblin MR. Dose-Response. 2009;7(4):358-383. American Academy of Dermatology, red light therapy guidance. Stanford Medicine, February 2025. Harvard Health, October 2025. Consumer Reports. Case report of photochemical retinopathy following blue LED mask use, 2020. Systematic review of repeated low-level red light therapy in myopia, 2024. FTC Health Products Compliance Guidance, December 2022. Manufacturer specifications and public FDA 510(k) filings, consulted July 2026. This article is informational and is not medical advice. Lumiglo devices are general wellness devices and are not intended to diagnose, treat, cure or prevent any disease.