The knee joint's deepest structures sit beneath layers of skin, fat, muscle, and connective tissue: articular cartilage, meniscus, subchondral bone, cruciate ligaments, and the nerve fibers threaded through the joint capsule. Red light at 630–660nm drives the cellular and inflammatory mechanisms described in Red Light Benefits for Knee Pain, but its energy attenuates rapidly as it passes through that tissue. Near-infrared light at 808–905nm delivers energy deeper, reaching the structures where knee osteoarthritis, meniscal degeneration, and tendon pathology originate. For the disease these wavelengths target, see red light therapy for osteoarthritis. A 2024 network meta-analysis of 13 RCTs put a number on the advantage: Near-infrared light ranked as the most effective wavelength for knee pain reduction, outperforming sham treatment. What that energy does once it reaches the deep joint structures is what determines whether the knee continues to degrade or begins to repair.
Key Takeaways
- Near-infrared wavelengths (808–905nm) deliver therapeutically relevant energy to knee structures that red light alone cannot reach effectively. The majority of near-infrared energy deposits within the first 10mm of tissue, with residual energy reaching 15–20mm, a range that encompasses the articular cartilage, synovial membrane, and superficial meniscal tissue of the knee.
- A 2024 network meta-analysis of 13 RCTs identified near-infrared light as the most effective wavelength for knee pain (SUCRA 86.9%). A 2023 double-blind RCT directly demonstrated that 808nm infrared produced significantly greater knee extensor strength improvement than 660nm red light in OA patients.
- Near-infrared wavelengths have produced structural and biological outcomes beyond pain relief in knee-specific studies: reduced cartilage-degrading enzymes in patients, a 50% reduction in cartilage degradation scores in OA models, meniscal stem cell activation, and objectively measured post-surgical swelling reduction.
Why Penetration Depth Determines Knee Outcomes
The articular cartilage of the femoral condyle, the medial and lateral menisci, the cruciate ligaments, and the subchondral bone plate all sit beneath the patella, patellar tendon, joint capsule, and the overlying skin, subcutaneous fat, and quadriceps musculature. Light energy passing through those layers follows exponential decay: each millimeter of tissue absorbs and scatters a fraction of the remaining energy, so small differences at the surface compound into large ones at depth.
That is where the wavelength gap opens up. Ash et al. (2017) modeled the interaction computationally and found that 850nm near-infrared reaches roughly 2.4–2.5mm into skin tissue versus 2.0mm for 660nm red light. A 20% edge at the first layer becomes a much larger one after several. Kaub & Schmitz (2022) put bounds on where the energy ends up: over 90% is absorbed within the first 10mm, with residual energy reaching 15–20mm depending on tissue composition. For the knee, the anterior compartment structures fall inside that window, the patellar cartilage surface, the suprapatellar bursa, and the anterior portions of the medial and lateral compartments. The deeper posterior structures receive less, which is the practical reason treatment protocols apply light from several angles around the joint rather than one.
The Wavelength That Works Best for the Knee
The 2024 Fan et al. network meta-analysis, published in Aging Clinical and Experimental Research, isolated the wavelength variable across 13 RCTs and 673 knee OA patients. The 904–905nm and 785–850nm ranges came out on top, significantly outperforming sham. The practical reading is that the specific wavelength chosen for knee treatment moves the odds of a clinically significant outcome, rather than being a detail buried in the spec sheet.
Jankaew et al. (2023) tested that directly. In a double-blind RCT, 47 knee OA patients were assigned to 808nm infrared, 660nm red, or sham. The 808nm group gained significantly more knee extensor muscle strength than either the sham or the 660nm group, and pain and WOMAC scores improved in the infrared group while the red-only group showed no such gain. The penetration physics predicted exactly this: for a weight-bearing joint where the pathology runs deep, near-infrared produces functional outcomes that visible red light does not match.
How Near-Infrared Light Reaches and Protects Knee Structures
Reaching and Protecting Articular Cartilage
The clinical prize in knee OA is slowing the cartilage destruction that makes the disease progressive. Near-infrared light reaches the articular cartilage surface and the chondrocytes embedded within it, engaging the enzyme-suppression and repair-support mechanisms covered in the Red Light Benefits article, this time at the depth where the cartilage actually sits.
The strongest human evidence here is biochemical rather than dimensional. Nambi et al. (2017) measured significant reductions in the cartilage-degrading enzymes MMP-3, MMP-8, and MMP-13, along with the breakdown marker CTX-II, in knee OA patients after photobiomodulation. The Ferreira et al. (2026) pilot confirmed MMP-3 and MMP-13 reductions directly in synovial fluid. At the tissue level, the Fan et al. (2025) in-vivo study documented a 50% reduction in OARSI cartilage degradation scores and collagen II upregulation using near-infrared LED. That is modality-correct, knee-specific evidence that photobiomodulation slows the enzymatic destruction of cartilage.
Direct imaging of cartilage thickness is the weaker and more mixed part of the picture, and it is worth being exact about it. In its low-level laser arm, Şen et al. (2025) recorded a within-group increase in femoral cartilage thickness on ultrasound, but that trial had no placebo arm, so the change cannot be separated from the concurrent exercise. The one placebo-controlled low-level laser trial that measured this exact endpoint, Stausholm et al. (2022), found no significant effect on femoral cartilage thickness. So the evidence that photobiomodulation reduces cartilage degradation is strong, while the evidence that it measurably increases cartilage thickness in humans is mixed and unconfirmed under placebo control. Thickness gain is the next threshold the research has to cross, not a result already in hand.
Shaikh-Kader and Houreld (2022) reviewed the evidence across the three cell types most relevant to knee integrity: chondrocytes, osteoblasts, and tenocytes. For chondrocytes, near-infrared photobiomodulation prevented degradation and improved tissue organization, with effects that were dose- and wavelength-dependent and near-infrared protocols showing particular relevance for deeper connective-tissue targets. For a joint where the relevant cells are buried in dense matrix beneath several tissue layers, that is the confirmation that matters: the energy arriving at those cells is enough to switch on repair pathways.
Subchondral Bone and the OA Feedback Loop
Beneath the articular cartilage lies the subchondral bone plate, and in osteoarthritis it undergoes pathological remodeling, thickening, microfracture, and altered vascularity, that both results from and accelerates cartilage loss. The interplay between cartilage degradation and subchondral bone change is now treated as a central feature of OA progression rather than a downstream consequence.
Near-infrared light reaches that bone, and the evidence shows it can push both sides of the remodeling balance. Bossini et al. (2012) found that 830nm laser significantly improved bone repair in osteoporotic animals, with both dose levels producing more new bone formation, better-organized collagen, and higher markers of osteoblast development and new blood vessel formation at the repair site. The Shaikh-Kader and Houreld (2022) review adds that photobiomodulation promotes osteoblast survival, proliferation, migration, and gene expression while stimulating stem cell differentiation into bone-forming cells. Working the other side of the equation, Lim et al. (2013) showed that 635nm LED can suppress osteoclast formation by disrupting the actin cytoskeleton those bone-resorbing cells need to function. Supporting bone formation while limiting resorption speaks to a structural dimension of OA that anti-inflammatory treatments leave untouched, though it is worth saying plainly that this evidence is preclinical.
Meniscal Tissue: Stem Cell Activation and Clinical Response
Degenerative meniscal tears drive knee pain in millions of adults, and the meniscus heals poorly on its own because its inner two-thirds have almost no blood supply. Light that can reach the meniscus and switch on its repair machinery is therefore worth taking seriously.
Tong et al. (2025) looked at this at the cellular level and found that 1064nm near-infrared light most effectively stimulated proliferation of meniscus-derived stem cells, working through TRPV1 channel activation, calcium influx, and downstream ROS signaling; the 700–710nm range was also effective. That gives a biological basis for near-infrared light activating a repair population the meniscus carries but does not normally deploy.
The clinical side rests on Malliaropoulos et al. (2013), still the only placebo-controlled RCT conducted specifically on meniscal pathology. Of the patients treated with near-infrared light, 87.5% responded, with significant pain improvement and Lysholm functional gains holding at 6 months and 1 year, and the authors concluded the therapy should be considered for meniscal tears in patients who want to avoid surgery. Nakamura et al. (2014) adds a smaller series: 35 patients with chronic knee pain from OA-related degenerative meniscal tears, significant pain improvement after 4 weeks of 830nm near-infrared treatment.
Tendons Surrounding the Knee: Patellar and Quadriceps Pathology
The patellar tendon, quadriceps tendon, and iliotibial band are common sources of anterior and lateral knee pain, and all of them are poorly vascularized and slow to heal. Their depth beneath the skin is exactly why near-infrared reaches them better than red.
What makes the tendon evidence interesting is the mechanism. Marcos et al. (2025) showed in a tendinopathy model that photobiomodulation upregulated ALX lipoxin receptors and TGF-β, actively driving inflammation resolution rather than mere suppression. That matters because tendinopathy is now understood less as an active inflammatory problem and more as a failed healing response, tissue stuck in a disorganized repair state. Engaging the pro-resolving pathways that finish the healing cycle goes at the pathology where it actually stalls. Lim et al. (2025) found that 630/880nm LED expanded the tenocyte population (SCX+ cells), reduced fibrosis, lowered disorganized collagen III and TGF-β1, and shifted macrophages toward the M2 phenotype in a tendon-healing model, with histological recovery inside 14 days. Ryu et al. (2022) measured the cell movement underneath all of that: 630nm LED tripled human tendon fibroblast migration versus control, the migration repair depends on. For the patellar tendon specifically, Morimoto et al. (2013) treated 41 patients across a range of sports injuries and found jumper's knee among the highest-response conditions, within a 65.9% overall effectiveness rate.
Post-Surgical Knee Recovery: Swelling, Range of Motion, and Opioid Reduction
Two RCTs have measured what near-infrared photobiomodulation does for the knee in the acute window after surgery, and the numbers are concrete.
Bahrami et al. (2023) enrolled 45 patients after total knee arthroplasty into a three-arm prospective RCT. The laser group reached 116.8° of knee flexion at 3 months against 92.3° for controls. Twenty-five degrees of range is the difference between limited mobility and being able to climb stairs, kneel, and rise from a chair without help. Swelling was significantly lower at 3 months, opioid consumption dropped, and KSS functional scores were higher at every follow-up. Chia et al. (2025) applied photobiomodulation in the first five days after TKA and tracked swelling with bioimpedance spectroscopy, an objective measurement rather than a clinician's visual estimate. The treatment group showed significantly less tissue edema and walked nearly twice as far, 27m versus 16m, on the 2-minute walk test. It is the first post-TKA trial to move past subjective swelling assessment to a quantifiable biophysical measure.
For anyone planning or recovering from knee replacement, these two trials establish that photobiomodulation speeds the recovery milestones that decide whether the surgery delivers its functional payoff. None of the trials reported treatment-related adverse events, consistent with the broader Photobiomodulation Safety for Knee Pain record.
Conclusion
For the knee, penetration isn't an abstract physics question. The cartilage, the meniscus, the subchondral bone plate, those are the structures where the disease actually lives, and they sit deep. What persuades me is that two different kinds of evidence converge. The human enzyme data shows reduced MMP-3 and MMP-13 in knee OA patients, and the in-vivo work shows reduced cartilage degradation. Both point the same way, toward a therapy that slows the destruction rather than only quieting the pain. The Fan wavelength ranking tracks the physics: deeper penetration reaches the target tissue. If you're choosing a device for knee OA, I'd tell you the wavelength and power numbers matter more than anything on the marketing page.— Dr. William Carter, MD
