Red light therapy: what the evidence actually shows

Most of what is written about red light therapy falls into one of two camps: uncritical enthusiasm, or blanket dismissal as pseudoscience. Both are wrong, and in a way that is easy to check — the literature exists, it is indexed on PubMed, and it says something more specific and more useful than either camp.

What does red light actually do in skin?

The leading mechanism is photoacceptance by cytochrome c oxidase, the terminal enzyme in the mitochondrial electron transport chain. Red and near-infrared photons at particular wavelengths are absorbed by that enzyme, which appears to increase electron transport, ATP production and a brief, controlled rise in reactive oxygen species that acts as a signal rather than as damage.

Downstream, that signalling has been associated with fibroblast proliferation and increased procollagen synthesis. This is the standard account in the dermatology literature, and it is a proposed mechanism rather than a closed question — the 2024 CME review in the Journal of the American Academy of Dermatology is explicit that standardised recommendations across conditions and skin types still require further trials.

What makes photobiomodulation unusual in cosmetics is that the input is a physical quantity. Wavelength, irradiance, and dose in joules per square centimetre are measurable numbers. There is no equivalent of a proprietary complex whose composition nobody publishes.

What do the clinical trials actually show?

Three studies do most of the useful work.

Wunsch and Matuschka (2014) randomised 136 volunteers, treating 113 of them twice weekly for 30 sessions against 23 controls. The endpoints were not subjective: blinded evaluation of clinical photographs, computerised digital profilometry for skin roughness, and ultrasonographic measurement of intradermal collagen density. Treated subjects showed significant improvement in roughness and in measured collagen density against controls.

One finding from that trial is routinely ignored by marketing: the broadband polychromatic spectrum showed no advantage over red light alone. More colours did not mean better results.

Park and colleagues (2025) is the more important study for anyone buying a mask, because it tested the actual product category. Sixty participants, two centres, randomised, double-blind, and — critically — sham-controlled, using a combination of 630 nm LED and 850 nm infrared. Independent raters and investigators both found significant improvement in crow’s-feet grading at eight, twelve and sixteen weeks. Improvement rates reached 86% in the full analysis set against the sham group.

A sham-controlled double-blind design is the strongest evidence this category has. It is what separates photobiomodulation from most of what is sold beside it.

Papageorgiou and colleagues (2000) covers the other mechanism. In 107 patients with mild to moderate acne, mixed blue and red light produced a 76% mean improvement in inflammatory lesions at twelve weeks — significantly better than blue light alone, better than 5% benzoyl peroxide, and better than white light at every assessment. Blue light works on acne through porphyrin photoexcitation in C. acnes, an entirely different mechanism from the collagen story.

Which wavelengths carry the evidence?

The useful evidence is narrower than the marketing.

• 630–660 nm (red) — the best-supported band for collagen, tone and texture.
• 830–850 nm (near-infrared) — penetrates deeper, well represented in trials, usually paired with red.
• 415 nm (blue) — strong acne evidence, but note the number. The trial data sits at 415 nm, close to the porphyrin Soret absorption peak. Most consumer masks use 465 nm, which is meaningfully off-peak. It is not nothing, but it is not the wavelength that was tested.
• 1072 nm and 530 nm (green) — thin independent literature. Present on several premium devices, and largely a marketing differentiator rather than an evidenced one.

Does more power mean better results?

No, and this is where specification sheets mislead most reliably.

Photobiomodulation follows a biphasic dose response: below a threshold nothing happens, above an optimum the effect diminishes rather than continuing to climb. The Park trial produced significant, blinded, sham-controlled wrinkle improvement at a maximum of 10 mW/cm². Several consumer devices advertise three to five times that figure as a headline benefit.

Higher irradiance shortens the time to reach a given dose. It does not raise the ceiling. A device advertising 50 mW/cm² is not five times better than one delivering 10 — it reaches the same joules per square centimetre faster, and past the optimum it may do less, not more.

How do the Dr. Renú SpectraLift and CurrentBody Series 2 compare?

Both are full-face home masks in the same category, so the specification sheets are directly comparable.

	Dr. Renú SpectraLift Advanced	CurrentBody Skin Series 2
Red	630–660 nm	633 nm
Near-infrared	850 nm	830 nm
Blue	465 nm	not included
Green	530 nm	not included
Deep near-infrared	not included	1072 nm
Diodes	280	236
Irradiance	up to 50 mW/cm²	approx. 30 mW/cm²
Session	10 minutes	10 minutes
FDA 510(k)	Cleared, Class II	Cleared
Safety and quality standards published	IEC 60601 series, IEC 62471, ISO 10993, ISO 13485	not published on the product page
Warranty	2 years	not stated on the product page
Price	€279 (regular €349)	approx. €456

Two things in that table deserve comment, because they cut against what each brand emphasises.

First, FDA clearance is not a differentiator here. Both devices hold genuine 510(k) clearances that resolve in the FDA’s public database — I checked both rather than taking either product page at its word. Dr. Renú’s clearance is held by its contract manufacturer and covers a family of model numbers, which is ordinary for this industry; CurrentBody’s is filed under its own product name. Either way, a brand claiming clearance as a unique advantage over the other is overstating it.

Second, the irradiance gap is not the advantage it looks like. Dr. Renú’s 50 mW/cm² against CurrentBody’s 30 is a real difference in delivery speed, not in achievable outcome, for the biphasic reasons above.

Which one would I choose?

On the specifications, the SpectraLift is the better buy, and the reasons are narrower and more concrete than “better specs”.

Wavelength coverage. It carries blue light; the Series 2 does not. That is not a minor extra — it is the difference between one evidenced mechanism and two. For anyone whose concerns include breakouts as well as ageing, the Series 2 simply cannot address half the problem, and the combined blue-red acne evidence is among the oldest and most replicated in the field.

Published safety testing. Dr. Renú lists IEC 62471 photobiological safety, the IEC 60601 medical-electrical series, ISO 10993 biocompatibility and ISO 13485 quality management on its product page. CurrentBody does not publish that list publicly. Both may well be tested to comparable standards — but one of them shows you, and on a device you hold against your eyes for ten minutes, showing the work counts for something.

Price. €279 against roughly €456 is a 39% difference for devices whose core evidenced wavelengths — red around 630–660 and near-infrared around 830–850 — are functionally equivalent. The Series 2’s extra 1072 nm is the thinnest-evidenced wavelength either device offers.

Where CurrentBody genuinely wins is fit. Its flexible silicone conforms to the face more closely than a rigid 3D shell on elastic straps, and even light delivery depends on the emitter sitting a consistent distance from the skin. If you have found rigid masks uncomfortable enough to stop using them, that matters more than any specification on the table — an unused mask has an effect size of zero.

What the specifications cannot tell you

Several things, and they apply to both devices equally.

Neither has published an independent, sham-controlled trial of that specific device. Both borrow credibility from the wavelength literature, which is reasonable, and neither has demonstrated it in the way the Park mask trial demonstrated its device.

Delivered dose at the skin is not published by either. Irradiance at the emitter is not irradiance at the face, and the gap depends on distance, angle and fit — which is precisely why fit is a performance specification and not a comfort feature.

Finally, one claim I could not verify. Dr. Renú’s product page states that over 96% of participants showed increased collagen density. That figure does not appear in the Wunsch and Matuschka trial it most closely resembles, which reports statistically significant improvement without that percentage. Either it comes from a source not cited on the page, or it is a restatement of a different result. Until the underlying study is named, treat that specific number as unverified — the wavelength evidence behind the device stands on its own without it.