Crow's feet are the tiny lines that fan out from the corners of your eyes when you smile, a bit like little branches. After a while, they start to show even when you are not smiling. They happen because the skin around your eyes is super thin and moves a lot every time you blink or smile. Over time, that skin loses some of its bouncy stretchy bits.
A red light mask helps your skin make more of the bouncy stretchy bits, right where the crow's feet are. In one big study, about 86 out of every 100 people who used a mask for three months had clearly smoother eye lines. Only 17 out of 100 in a pretend group got the same. That is a really big difference, and it is one of the strongest reasons to try a mask.
Crow's feet are one of the most strongly evidenced LED outcomes.
Park et al. 2025 (multi-center, double-blind, sham-controlled): 86.2% of users with measurable improvement at twelve weeks against 16.7% sham. Mota et al. 2023 (split-face, 137 women): 31.6% reduction in periocular wrinkle volume. The thin periocular dermis (roughly 0.3 to 0.5mm vs 1.5 to 2.0mm at the cheek) means the same biological collagen gain shows up as a larger visible result here than anywhere else on the face. LED supports the dermal scaffolding around the eye; it does not replace botox for the dynamic muscle component.
Why the periocular area responds the way it does
Periocular dermis runs roughly 0.3 to 0.5 mm thick. Cheek dermis is closer to 1.5 to 2.0 mm. The collagen layer that LED therapy targets sits closer to the surface here than anywhere else on the face, and the same biological gain reads as a bigger visible result.
Crow's feet form along two pathways. The first is dynamic. Small repetitive contractions of the orbicularis oculi muscle, from squinting and smiling and hours of screen reading, crease the surrounding skin tens of thousands of times a year. In young skin, the dermis recovers between contractions almost completely. As collagen and elastin density decline with age and accumulated UV exposure, recovery slows and incompleteness creeps in. The crease starts to hold its shape between contractions.
The second pathway is structural. Fibroblasts, the cells that build collagen, slow their output. Existing collagen fibers fragment under cumulative photoaging. The scaffolding underneath the surface skin gets thinner and more disorganised, which is what makes aging skin look tired before any specific wrinkle is visible.
Red light therapy targets the second pathway. Wavelengths in the 630 to 660nm red range, often paired with 830 to 850nm in the near-infrared, are absorbed by cytochrome c oxidase in the mitochondria of skin cells. That activates a cascade that lifts ATP production and signals fibroblasts to step up collagen synthesis. The mechanism is gradual rather than acute, which is why these protocols run over weeks. We've covered the mitochondrial side of the story in detail in our collagen guide.
The periocular region is also unusually well-perfused. The orbicularis oculi sits over a dense vascular network, which means the skin gets oxygen, nutrient delivery, and clearance of metabolic byproducts faster than thicker, less perfused areas. Combined with the thin dermis, that creates a region where biological turnover is active and visible, which is one of the leading hypotheses for why periocular endpoints show such consistent effect sizes across the trial literature.
The strongest evidence
Three trials carry most of the weight on crow's feet. We walk through each, then summarise the supporting literature behind them.
Park et al. 2025: 86.2% improvement at 12 weeks
This is the strongest periocular trial in the current literature. Park and colleagues (PMID 39960921) ran a multi-center, double-blind, sham-controlled study of a home-use LED mask combining 630nm red and 850nm near-infrared. Participants used the mask at home for 12 weeks. Neither they nor the dermatologists scoring the photographs knew which device was active and which was sham. Both looked identical. Only one delivered light.
At the 12-week readout, blinded scoring showed 86.2% of the active group with clinically meaningful crow's feet improvement. The sham group came in at 16.7%. The result was statistically significant and the safety profile was clean. What this trial proves is that a properly designed home-use mask delivering 630nm and 850nm at the right protocol produces visible crow's feet improvement at 12 weeks in roughly six users out of seven. What it does not prove is durability beyond 12 weeks, or how 630nm-only would compare to 850nm-only. Those are separate questions that have not been answered with the same rigor.
Mota et al. 2023: 31.6% reduction in periocular wrinkle volume
Mota and colleagues (PMID 36780572) ran a split-face randomized controlled trial in 137 women aged 40 to 65. Split-face means each participant received one wavelength on one side of her face and another on the other side, so the same person serves as both experiment and comparator. That design controls for genetics, sun history, and lifestyle in a way between-group studies never can.
Each participant received 10 sessions over five weeks. One side was treated with 660nm red, the other with 590nm amber. Wrinkle volume around the eye was measured by 3D imaging that quantifies the actual three-dimensional volume of skin surface depressions, not just length or perceived depth. The 660nm side showed a 31.6% reduction in periocular wrinkle volume. The 590nm side came in at 29.9%. Both reductions were statistically significant. Around 30% volume reduction in five weeks is fast for red light therapy and lines up with what's expected when a thin-skinned area gets concentrated dosing. The split-face design is what makes this trial particularly hard to argue with.
Couturaud et al. 2023: improvements that held after stopping
Couturaud and colleagues (PMID 37522497) tested a 630nm LED mask on 20 women using it twice weekly for three months. The sample size is small. The design was open-label rather than sham-controlled. Both facts limit how much weight this trial carries on its own. The reason it still matters is durability.
Wrinkle depth was measured periodically using profilometry, a surface-relief technique that quantifies actual crease depth, alongside blinded photographic scoring. Improvements built progressively across the 12 weeks rather than plateauing in the first month. Dermal density also rose. After the three-month course ended, participants were followed for another month with no LED treatment. The improvements held. This is one of the few published trials to look at what happens after a course ends, which is the question that sets sensible maintenance schedules. The honest read is that LED-induced collagen changes are biological rather than pharmacological. The body builds new collagen and that collagen does not vanish on day 91. How long it persists past one month is unstudied, but a maintenance schedule of one to two sessions per week reads as reasonable based on what's available.
Supporting evidence
Two earlier trials sit behind these three. Russell et al. 2005 (PMID 16414908) tested combined 633nm and 830nm LED in 36 patients and reported 81% of subjects with significant periorbital improvement after nine sessions across 12 weeks, an early demonstration of what Park 2025 confirmed two decades later with better methodology. Weiss et al. 2004 (PMID 15624743) used a 590nm yellow LED panel in 93 patients and showed periocular wrinkle reduction by digital microscopy alongside 90% photoaging improvement overall. Both trials are smaller and methodologically older, but together they extend the pattern: the periocular response holds across a range of wavelength categories. Our research database catalogues the rest.
The practical timeline
The trials converge on a consistent rhythm. Two to three sessions per week for 8 to 12 weeks is the cadence most home-mask protocols are built around.
In the first two weeks, expect very little visible change. The mitochondrial cascade and collagen synthesis upregulation are happening biologically but they don't show on the surface yet. Some users notice their skin looks fresher or more hydrated early on. That's microcirculation and water content shifting, not structural change.
Take a baseline photograph at the start. Same lighting, same angle, same expression, no makeup. Repeat it every two weeks. Subtle gains accumulate slowly enough that most people miss them when they're looking in the mirror daily, then notice them clearly at the four-week comparison.
Weeks four to six are when texture and tone changes start to appear. Skin reads slightly smoother. Photographs taken in the same lighting often show a subtle evenness shift. Crow's feet themselves usually have not visibly reduced in depth at this point. The collagen reinforcement is happening underneath, but the surface relief takes longer to remodel.
Weeks 8 through 12 is when the periocular endpoint shows up in most published trials. Park 2025 measured at 12 weeks. Mota 2023 hit 30% wrinkle volume reduction in five weeks at concentrated dosing. Russell 2005 reported their periorbital number at week 12. The 8 to 12 week window is when most properly designed home-use trials set their primary endpoints because it is when the dose-response curve is closest to its plateau. Beyond 12 weeks, the published data thins. Our week-by-week timeline guide covers the cadence in more detail.
What it won't fix
Red light therapy has clear boundaries.
Deep dynamic crow's feet, the kind that are visible only when you smile or squint and flatten completely at rest, are driven by muscle contraction. LED therapy doesn't relax muscle. Botox does. The two work on different parts of the same problem and many dermatologists now recommend them together: botox for the dynamic component at quarterly intervals, LED for the dermal canvas continuously.
Lower-lid fat-pad changes that drive under-eye bags are structural. As the orbital septum (the membrane holding the orbital fat pad in place) weakens with age, the fat herniates forward and creates the puffy under-eye look. That is connective-tissue and structural anatomy, not collagen-density work, and LED does not address it. Surgical or filler-based interventions are the available options.
Severe lid laxity from age-related connective tissue loss falls in the same category. Mild laxity in earlier stages may respond to LED via dermal density gains. Past that point, LED is not the right tool.
Periocular dark circles caused by visible vasculature under thin skin are vascular rather than pigmentary, and they don't respond to LED either. For under-eye specific concerns, see our under-eye guide.
How our mask fits in
We built our mask around the wavelength categories represented in the trials above. It runs 633nm in the red range, 850nm and 1072nm in the near-infrared range, plus 590nm yellow and 415nm blue across six preset modes, including an Anti-Aging mode that pairs red with near-infrared, the same wavelength category combination used in Park 2025 and Russell 2005.
Three hundred and sixty medical-grade LEDs cover the full mask surface, with particular attention to the periocular zone. A contoured silicone shell directs light onto the surrounding skin while keeping direct exposure to the closed eyelid minimal. Sessions run 10 minutes. Sixty-day money-back guarantee. Two-year warranty. Free express shipping AU-wide.
Cited studies
Park SH, et al. · Medicine (Baltimore) · 2025 · PMID 39960921
“Clinical study to evaluate the efficacy and safety of home-used LED and IRED mask for crow's feet”
86.2% of active-treatment participants showed improvement in crow's feet wrinkles at 12 weeks versus only 16.7% in sham group; safe and well-tolerated.
View on PubMed →Couturaud V, et al. · Skin Research and Technology · 2023 · PMID 37522497
“Reverse skin aging signs by red light photobiomodulation”
Twenty women using a 630nm LED mask twice weekly for 3 months showed progressive reductions in crow's feet depth, improved dermal density, and improvements persisted up to one month after treatment ended.
View on PubMed →Mota LR, et al. · Photobiomodulation, Photomedicine, and Laser Surgery · 2023 · PMID 36780572
“Photobiomodulation Reduces Periocular Wrinkle Volume by 30%: A Randomized Controlled Trial”
In 137 women aged 40-65, 10 sessions of red (660nm) LED reduced periocular wrinkle volume by 31.6% and amber (590nm) LED by 29.9% compared to controls.
View on PubMed →Russell BA, et al. · Journal of Cosmetic Laser Therapy · 2005 · PMID 16414908
“A study to determine the efficacy of combination LED light therapy (633 nm and 830 nm) in facial skin rejuvenation”
52% of subjects showed 25-50% improvement in photoaging scores by week 12; 81% reported significant improvement in periorbital wrinkles after 9 sessions.
View on PubMed →Weiss RA, et al. · Journal of Drugs in Dermatology · 2004 · PMID 15624743
“A novel non-thermal non-ablative full panel LED photomodulation device for reversal of photoaging”
93 patients showed photoaging improvement in 90%; digital microscopy confirmed reduced periocular wrinkles and improved texture with no side effects.
View on PubMed →
See our full research database for the complete catalogue of peer-reviewed studies.
FAQ
How well does red light therapy work for crow's feet specifically?
Park et al. 2025 (PMID 39960921) reported 86.2% of users in the active treatment group with measurable crow's feet improvement at 12 weeks, against 16.7% in the sham group. That gap is one of the largest and cleanest in the home-use LED literature. Mota et al. 2023 (PMID 36780572) added a 31.6% periocular wrinkle volume reduction in 137 women using a rigorous split-face design. Across decades of published work, periocular endpoints regularly outperform other facial endpoints, likely because thin periocular skin shows collagen gains more readily than the thicker skin elsewhere on the face.
Is it safe to use red light therapy near the eyes?
Sessions are done with eyes closed. Properly designed home-use masks direct light onto the periocular skin while reducing direct exposure to the eyelid and cornea through a contoured silicone fit. Across the published mask trials referenced in this article, including Park 2025's multi-center double-blind study, no serious eye-related adverse events were reported in the home-use protocols. Users with hereditary eye conditions, recent eye surgery, severe dry eye, or who take photosensitising medications should consult an ophthalmologist before starting LED therapy.
Does red light therapy replace botox for crow's feet?
No. The two work on different layers of the same problem. Botox temporarily relaxes the orbicularis oculi muscle that creates dynamic crow's feet, with a fast and visible effect on the dynamic crease. LED therapy supports dermal collagen synthesis in the surrounding skin, a slower and more cumulative effect that improves the canvas rather than the muscle itself. Many dermatologists now recommend the two together: botox at standard quarterly intervals plus continuous LED for the surrounding skin texture and density.
How long do results last after stopping?
Couturaud et al. 2023 (PMID 37522497) followed participants for one month after a three-month LED course ended and the gains held across that month. Beyond that one-month follow-up window, no published data exists on six-month or twelve-month durability. The biological reasoning suggests gradual decline as fibroblast activity returns to its underlying age-related baseline, at a rate measured in months rather than days. A maintenance schedule of one to two sessions per week, after the initial 8 to 12 week course, is what we recommend based on what the literature supports.
Can I use retinol or other actives alongside LED?
No published trial has tested the combination cleanly. Most studies asked participants to avoid retinoids and acids during the trial to isolate the LED effect. Mechanistically there are reasons to expect retinol and LED to be complementary rather than competing. Retinol drives epidermal turnover. LED supports fibroblast collagen synthesis in the dermis. Many dermatologists recommend running both, with retinol applied at night and LED at a separate time of day. If your skin signals irritation in the first two weeks, reduce the frequency of one or the other.
What if I do not see results at week 12?
First, check cadence honestly. Most trials enrolled participants at two to three sessions per week, and sporadic use produces a lower accumulated dose than what the published research actually tested. Second, response rates were not 100% in any trial referenced here. Park 2025's 86.2% headline means roughly one in seven users did not show measurable change, even with the right device used correctly. Our 60-day money-back guarantee exists because of that gap. If at week 12 there is genuinely no change despite consistent use, returning the mask is the right call.
Related guides
- Red light therapy for wrinkles
Effect sizes from the strongest randomized trials across the full face.
- Red light therapy for under-eye bags
Honest read on what the research supports for the under-eye area.
- Red light therapy for collagen
The mechanism behind every periocular change discussed here.
- Customer reviews
Real periocular outcomes from users of the mask in Australia.
Designed for the area where the evidence is strongest.
Red Light Rejuve covers 633nm red, 590nm yellow, 415nm blue, and dual near-infrared at 850nm and 1072nm across six preset modes including Anti-Aging. 360 medical-grade LEDs, 10-minute sessions, 60-day money-back guarantee, two-year warranty.