PBM Research Review: Best of 2026 So Far
Share
Photobiomodulation research: the best of 2026 so far
I'm changing up a little this week as I head towards the summer solstice here in Norway. Instead of a weekly update, I'm going to look at an overview of what has been published this year (2026) that is actually relevant to you.
Every week I go through my PubMed alerts and pick out the studies that I think actually matter: the ones that change how I think about protocols, or confirm what we suspected, or open up a new direction I hadn't been paying attention to. Yeah it happens, I am human and have my own personal interests that I tend to focus on which I deem to be more "sexy". However, these updates do get me looking at other areas and that is important as well. This is that post for the first half of 2026: what has been published and what sparked my interest.
A quick note on what's in here and what isn't. I've only included papers from 2026. I've focused on positive outcomes because that's what's useful for people trying to understand what this intervention is actually capable of. That doesn't mean null results don't exist out there, they do, and I'll flag a couple of those in the notable papers section at the end. But first, the good stuff.
One thing that strikes me looking across these papers is the wavelength range. We're not just talking about 660nm and 850nm, the standard panel combination. Some of the most interesting 2026 results involve 590nm yellow, 808nm near-infrared, and multiwavelength approaches that combine several ranges at once. That matters when you're thinking about what device is right for what purpose, and I'll come back to it at the end.
Paper 1: The dry AMD result that should be in every conversation about red light therapy
Jaffe GJ et al. Long-term efficacy and safety of photobiomodulation in dry age-related macular degeneration. Retina. 2026 May 1;46(5):783-795. https://pubmed.ncbi.nlm.nih.gov/41791029/
This is LIGHTSITE III's 24-month analysis, and it's the strongest piece of clinical evidence to come out of the photobiomodulation field in some years. Dry age-related macular degeneration (AMD) is a condition that affects the central vision and has, until recently, had essentially no treatment beyond vitamin supplementation. This trial ran at ten US retinal centres, enrolled 100 subjects with early to intermediate dry AMD, and followed them for two years with a randomised, double-masked, sham-controlled design.
The results: a statistically significant 6.2-letter gain in best corrected visual acuity (BCVA) in the PBM group compared to sham at 21 months (p = 0.0036). At that timepoint, 61.5% of treated eyes showed a gain of five or more letters. More clinically significant, though, was the disease progression data. In the sham group, 21.6% of eyes progressed to geographic atrophy, the later, more destructive stage of the disease. In the PBM group, only 5.7% did. The difference was statistically significant (p = 0.003) with an odds ratio of 9.3.
The device used was the LumiThera Valeda system, delivering three wavelengths simultaneously: 590nm (yellow), 660nm (red), and 850nm (near-infrared). Treatment was delivered nine sessions per series, three times per week, with a series every four months across the 24-month period. No phototoxicity was observed.
Ask yourself: if someone you knew was told they had dry AMD and there was nothing their optometrist could do about it, would you want them to know this existed? This is that kind of result.
The multiwavelength approach here is worth noting. 590nm, which sits in the yellow range, is not in most home panels. The 660nm and 850nm components are standard. The fact that the combination produced these results is evidence for the multiwavelength hypothesis: that different chromophores at different tissue layers respond to different wavelengths, and using several simultaneously reaches more biology.
Paper 2: Alzheimer's and the metabolism question
Gaggi NL et al. A potential metabolic basis for brain activity changes after transcranial photobiomodulation in Alzheimer's disease. Photonics. 2026 June 4;13(6):551. https://doi.org/10.3390/photonics13060551
This is from the NYU Grossman School of Medicine group, funded by the Alzheimer's Association and the NIA. It's preliminary work: eleven subjects: but it's genuinely interesting methodology that I haven't seen applied to PBM before.
The design combined phosphorus magnetic resonance spectroscopy (31P-MRS) with resting-state fMRI in participants with mild cognitive impairment or early Alzheimer's disease. The 31P-MRS measured baseline brain energy metabolism: specifically PCr/Pi ratios and brain pH, both of which decline as mitochondrial function deteriorates in neurodegeneration. The fMRI during transcranial PBM application (808nm, delivered over bilateral frontal sites for approximately 12 minutes) measured real-time changes in brain activity as a proxy for metabolic engagement.
The finding that stands out: the more metabolically compromised the brain region at baseline: the lower the PCr/Pi and the lower the pH: the greater the increase in brain activity during PBM. In the right frontal pole specifically, the model explained 99.2% of variance in the response. The authors describe this as evidence that transcranial PBM may preferentially engage brain regions where metabolism is most impaired, rather than stimulating uniformly.
This is the cellular rescue hypothesis applied to the brain. A brain running on compromised mitochondria responds more strongly to the mitochondrial stimulus that PBM provides. The researchers are careful to call it hypothesis-generating, and it is. But the direction is exactly what the basic science would predict.
The ongoing clinical trial from this group will follow a larger cohort with repeated sessions. Watch this space.
Paper 3: The comprehensive brain mapping: from stroke to depression
Published in Neuromodulation in February 2026 (doi: 10.1016/j.neurom.2025.12.006), this scoping review from the Global Brain Health Institute is the most comprehensive single document on transcranial PBM in humans published to date. It covers every condition for which human data exist: stroke, traumatic brain injury, chronic traumatic encephalopathy, Alzheimer's and Parkinson's diseases, autism spectrum disorder, ADHD, Down syndrome, anxiety, depression, insomnia, sexual dysfunction, and multiple sclerosis.
Two things from this paper worth singling out.
First, the confirmation that transcranial PBM exhibits a biphasic dose response: both understimulation and overstimulation are possible. The review explicitly states that both under- and overstimulation can occur. This is something I've been saying about panels for years, and it's validating to see it formally confirmed for brain applications too. More is not better. The right amount, delivered correctly, is what matters.
Second, the observation that optimal outcomes appear to depend on multiwavelength, multi-target approaches with pulsed delivery. The review authors note this is likely because different wavelengths complement each other mechanistically and pulsed delivery may prevent excessive ROS production. This aligns with the AMD findings above.
The full range of human conditions now supported by at least some clinical evidence is broader than most people realise. This paper maps that territory properly.
Paper 4: Knee pain: what a meta-analysis of 340 people actually found
Alayat MS et al. The effectiveness of photobiomodulation therapy on pain and function in patients with patellofemoral pain syndrome: a systematic review and meta-analysis. Journal of Clinical Medicine. 2026 Jan; 15(1):20. https://pubmed.ncbi.nlm.nih.gov/41517270/
Eight randomised controlled trials, 340 participants, PRISMA guidelines. Patellofemoral pain syndrome is the kind of knee pain that affects a huge number of people: the grinding discomfort under the kneecap when climbing stairs, sitting for long periods, or getting up from a low chair. It's common, it's often under-treated, and standard physiotherapy helps some people more than others.
The pooled analysis found that PBM combined with exercise significantly reduced pain (SMD = -0.83; 95% CI -1.40 to -0.27) and improved functional outcomes (SMD = 0.68; 95% CI 0.08 to 1.27) compared to control.
The honest caveat from the authors: substantial heterogeneity across studies (I² = 83%), five of the eight included studies were rated high-risk on the Cochrane bias tool, and GRADE rated the overall evidence quality as very low. The authors conclude that PBM as an adjunct to exercise shows promise but that larger, better-designed trials with standardised dosing are needed.
I'm including this because it's real-world relevant. Patellofemoral pain is one of the most common musculoskeletal complaints I hear about from customers. The effect sizes are meaningful even if the evidence quality has room to grow.
Paper 5: Wound healing after childbirth: an area nobody talks about
Published in Journal of Clinical Medicine in January 2026 (doi: 10.3390/jcm15030964), this systematic review and meta-analysis looked at PBM for perineal pain and wound healing following episiotomy: the incision sometimes made during vaginal delivery to aid the birth.
Eight studies were included in the systematic review, six in the meta-analysis. The review found that PBM reduced perineal pain and supported wound healing in this population, with results across multiple outcome measures including pain intensity, tissue repair assessment, and patient-reported discomfort.
I'm including this because it's an application most people don't associate with red light therapy. Postpartum recovery is a period of real physical vulnerability where analgesic options are limited (particularly for breastfeeding mothers), recovery time matters enormously for a new parent's functioning, and the tissue involved responds well to exactly the mechanisms PBM targets. This is a genuinely useful clinical application that deserves more attention.
Paper 6: The burn wound: dual NIR wavelength approach
Published in Archives of Rehabilitation Research and Clinical Translation in 2026 (ScienceDirect DOI: 10.1016/j.arrct...2025.001708), this study examined combined pulsed 810nm and superpulsed near-infrared light on transdermal burn wound repair. The combined NIR approach accelerated wound closure through anti-inflammatory, pain-relieving, and redox-regulating mechanisms.
Burns are one of the areas where wound healing research is genuinely urgent: standard protocols are often inadequate for anything beyond minor burns, and the combination of pain, infection risk, and prolonged recovery makes adjunct therapies highly relevant. Near-infrared at 810nm penetrates significantly deeper than visible red wavelengths, reaching tissue layers that surface-level treatments cannot. The superpulsed delivery modality in this study delivered higher peak power in short pulses, which is thought to increase penetration depth while limiting thermal effects.
The anti-inflammatory and redox components here connect directly to the cytokine research we've been discussing: reduced TNF-α, reduced IL-1β, and modulated ROS at the wound site.
Paper 7: LED vs laser: does the device type actually matter?
Published in Cell Biochemistry and Function in January 2026 (doi: 10.1002/cbf.70161), this systematic review compared LED and laser photobiomodulation devices directly for cutaneous wound healing, pulling together both in vitro and in vivo data.
The headline conclusion: both modalities enhanced cell viability, proliferation, migration, inflammatory modulation, collagen deposition, and angiogenesis. The review found no consistent evidence that laser produces meaningfully superior outcomes to LED when matched for wavelength and dose. The key variables are wavelength, power density, fluence, and pulse parameters: not the specific light source technology.
This is practically significant for everyone reading this site, because the devices I sell are LED-based. The question of "is a laser better than LEDs" comes up regularly. This review, which represents the best synthesis of the current evidence, says the answer is: not inherently. What matters is what the light is doing at the tissue level.
Notable papers: worth knowing about but not in this review's scope
A few 2026 papers that didn't make the main list but are worth being aware of:
The dual wavelength RCT for dental third molar extraction post-operative recovery (Journal of Clinical Medicine, March 2026): red-only versus combined red-plus-infrared in a split-mouth randomised trial. The combined approach showed superior anti-inflammatory outcomes. Reinforces the multiwavelength principle.
The LED-based diabetic wound healing mechanistic study (IJMS, 2026, doi: 10.3390/ijms26062...): detailed pathway work showing LED PBM enhanced wound closure in diabetic mice via focal adhesion-mediated cell migration and extracellular matrix remodelling, with macrophage infiltration modulation and reduced myofibroblast activation. Good mechanistic grounding for the diabetic wound application.
The scoping review of transcranial PBM for spasticity in paediatric cerebral palsy (Brain Sciences, February 2026): the authors conclude the evidence is too weak for clinical use yet, but the direction is promising. Worth knowing it's on the radar.
The Wound Repair and Regeneration meta-analysis on venous leg ulcers (March 2026): this one was not positive. Eleven trials, 615 patients, and no statistically significant reduction in ulcer area. High heterogeneity, multiple methodological concerns. I'm including it in the notable section because honest science means noting where results are not there yet. Venous leg ulcers remain an area where the evidence is genuinely unclear.
What the first half of 2026 actually tells us
A few things I take from this as a pattern, not just individual results.
The multiwavelength story is getting stronger. The AMD result used 590nm, 660nm, and 850nm together. The transcranial brain review points toward multiwavelength and pulsed approaches as likely optimal. The dental split-mouth trial showed combined red-plus-infrared outperforming red alone. If you're thinking about what panel to use or recommend, the single-wavelength argument is getting harder to make.
The brain is moving from promising to evidenced. The transcranial application has been in the background for years. The 2026 scoping review is the most comprehensive map of human clinical evidence yet, and the Alzheimer's metabolic targeting paper from NYU/NKI is the most sophisticated mechanistic investigation we've seen. This is a direction that will continue developing.
The AMD result is the one that deserves more general attention. A disease with no previous treatment options, a randomised trial at multiple top-tier US retinal centres, statistically significant slowing of progression to geographic atrophy with an odds ratio of 9.3. That is a meaningful clinical result by any standard.
The evidence base for what most people actually use panels for: pain, recovery, general wellbeing: continues to grow incrementally. The patellofemoral meta-analysis adds to the joint pain body of evidence. The burn wound and episiotomy papers extend wound healing into new clinical contexts.
I'll do the same review at the end of the year when the full 12 months are in.
Disclaimer: This post is for educational and informational purposes only and does not constitute medical advice. The studies discussed are peer-reviewed research; they are not claims about specific devices. If you have a medical condition, speak to a qualified healthcare provider before starting any new therapy.
FAQ
What is the most significant photobiomodulation study published in 2026 so far?
The 24-month analysis of the LIGHTSITE III trial, published in Retina in May 2026, is the strongest piece of clinical evidence published in the PBM field in some years. It showed a statistically significant 6.2-letter visual acuity gain in dry age-related macular degeneration patients treated with multiwavelength photobiomodulation over two years, and crucially, a statistically significant reduction in disease progression to geographic atrophy compared to sham treatment. This is notable because dry AMD previously had no treatment capable of slowing visual decline beyond vitamin supplementation.
Hva betyr den nye AMD-studien fra 2026 for vanlige brukere av rødlysterapi?
LIGHTSITE III-studien brukte en kombinasjon av tre bølgelengder: 590nm (gul), 660nm (rød) og 850nm (nær-infrarød): levert i kliniske behandlingsserier over to år. For hjemmebrukere viser resultatene at flerbolgelengde-tilnærminger gir biologiske fordeler som enkeltbølgelengde-enheter kanskje ikke gjenskaper fullt ut. Det betyr ikke at rødlysterapi hjemme er nytteløst: det betyr at enhetens bølgelengdeprofil er viktigere enn mange leverandører innrømmer.
Does the 2026 research suggest that LED panels are less effective than laser for photobiomodulation?
A systematic review comparing LED and laser devices for wound healing, published in Cell Biochemistry and Function in January 2026, found both modalities produced comparable outcomes across the key biological parameters including cell viability, collagen production, inflammatory modulation, and tissue repair. The determining variables are wavelength, power density, and dose delivery, not the light source technology itself. For home and wellness use, well-designed LED panels operating at appropriate therapeutic parameters produce the relevant biology.
Is there any 2026 research on photobiomodulation and mental health or depression?
A comprehensive scoping review published in the journal Neuromodulation in February 2026 mapped all existing human clinical evidence for transcranial photobiomodulation across neurological and psychiatric conditions. Depression and anxiety are included in the conditions with existing clinical trial data, alongside conditions like Alzheimer's disease, Parkinson's disease, traumatic brain injury, and insomnia. The review notes that optimal outcomes appear to require appropriate dosing, with both under- and overstimulation possible, and that multiwavelength and pulsed approaches may produce better results than single-wavelength continuous wave delivery.
References
- Jaffe GJ et al. Long-term efficacy and safety of photobiomodulation in dry age-related macular degeneration (LIGHTSITE III: 24-month analysis). Retina. 2026 May;46(5):783-795. https://pubmed.ncbi.nlm.nih.gov/41791029/
- Gaggi NL et al. A potential metabolic basis for brain activity changes after transcranial photobiomodulation in Alzheimer's disease. Photonics. 2026;13(6):551. https://doi.org/10.3390/photonics13060551
- Transcranial photobiomodulation for neuromodulation of brain disorders: a perspective. Neuromodulation. 2026 Feb 5. doi: 10.1016/j.neurom.2025.12.006. https://www.sciencedirect.com/science/article/abs/pii/S1094715925012036
- Alayat MS et al. The effectiveness of photobiomodulation therapy on pain and function in patients with patellofemoral pain syndrome. J Clin Med. 2026;15(1):20. https://pubmed.ncbi.nlm.nih.gov/41517270/
- Photobiomodulation for perineal pain and wound healing after episiotomy: systematic review and meta-analysis. J Clin Med. 2026;15(3):964. https://doi.org/10.3390/jcm15030964
- LED vs laser for cutaneous wound healing: systematic review. Cell Biochem Funct. 2026 Jan 15;44(1):e70161. https://pmc.ncbi.nlm.nih.gov/articles/PMC12809106/