Knee Cartilage Repair Without Surgery: What Actually Works

Ranked by the strength of the evidence behind each and how widely it applies to knees with cartilage loss.

The ordering is the opposite of how these are usually marketed. The two options that cost nothing and require the most effort sit at the top, and the injectable treatments presented as regenerative sit at the bottom.

This is a documentary review of published guidance and trial evidence rather than advice about your knee. Five options were chosen because they cover everything genuinely available without an operation, including the treatments people are most often offered privately.

The sources read were NICE guideline NG226 on osteoarthritis in over 16s; the RESTORE randomised clinical trial published in JAMA in 2021, which randomised 288 participants to platelet-rich plasma or saline placebo; and Messier and colleagues in Arthritis & Rheumatism on the relationship between body weight and knee joint load. Evidence checked in August 2026.

The main limitation is that most of the high-quality evidence concerns knee osteoarthritis, which is the commonest reason for cartilage loss and is not the only one. A focal traumatic cartilage defect in a 25-year-old is a different problem with a different surgical case, and the trial evidence summarised here does not settle that situation. Damage to the meniscus, which is a separate structure from the articular cartilage discussed here, follows its own rules and is covered in our guide to what the meniscus is. This page distinguishes the two rather than treating them as one.

No prices are quoted, because clinic, brace and injection costs could not be verified for this review.

Start with what the knee cannot currently do. Stairs, standing from a chair, walking distance and kneeling each point at different deficits, and the one that limits you most is the one worth measuring progress against.

Ask whether the loss is diffuse or focal. Widespread thinning across a compartment behaves like osteoarthritis and responds to the options below. A single deep defect with exposed bone in a younger person is the situation where a surgical opinion genuinely belongs.

Check whether one compartment is doing the suffering. Pain confined to the inner side of the knee, particularly with a bow-legged alignment, is the pattern where offloading with a brace or wedge insole has something specific to offer.

Be clear what each treatment is claiming. There is a difference between an intervention that reduces pain for a few months and one that changes the joint. Injections belong in the first category, and they are frequently sold as the second.

Decide what you can sustain. The two highest-ranked options require ongoing effort over months. An option you will actually keep doing beats a stronger option you will abandon in three weeks.

What it involves

Resistance training for the quadriceps, hamstrings and hip muscles, built up gradually and continued indefinitely. Typical content is sit-to-stand work, step-ups, leg press or wall sits, and hip abduction work, performed two to three times a week with the load increasing as tolerance improves.

What the evidence supports

NICE guideline NG226 makes tailored therapeutic exercise the focus of osteoarthritis management, including local muscle strengthening and aerobic fitness, supported by education. NICE also notes that exercise may increase joint pain at first and that this diminishes with regular consistent activity, going on to improve physical function and quality of life.

Why it helps a joint with less cartilage

Strong muscles absorb force that would otherwise pass through the joint surface, and they control the knee so that load lands where the joint tolerates it. The cartilage does not thicken. The knee becomes able to do more with the cartilage it has, which is the outcome that matters day to day.

Who it suits

Almost everyone with cartilage loss, at every stage, including people who have been told their knee is bone on bone. Strength work is also what produces a better outcome if a joint replacement eventually happens.

Limits and cautions

The early increase in pain that NICE describes is real and it puts people off in the first fortnight. Progress is slow and measured in months, and the gains reverse if the training stops.

Why it ranks first

It is the core treatment in the national guideline, it applies to the widest group, and it is the only option here that increases what the knee can tolerate rather than reducing what is asked of it.

What it involves

Stationary cycling with the seat high and the resistance low, swimming, water-based exercise and regular walking. Twenty to thirty minutes most days is a typical target, built up from whatever the knee currently tolerates.

Why movement feeds cartilage

Cartilage has no blood supply and takes its nutrition from joint fluid. Repeated gentle compression and release pushes waste out and draws nutrient-rich fluid in, which keeps the remaining cartilage cells viable. A joint that is not moved is a joint that is not fed.

What the evidence supports

Aerobic fitness sits alongside muscle strengthening within the therapeutic exercise that NICE makes the focus of management. Cycling is particularly useful early because it produces the movement without the impact.

Who it suits

Everyone, and especially anyone whose knee is too irritable for weight-bearing exercise. Cycling is usually the first thing tolerated after a flare, and it maintains fitness while the strength work builds.

Limits and cautions

Low resistance matters. Grinding up hills on a heavy gear turns a joint-friendly activity into a demanding one. Breaststroke leg kick provokes many knees, and front crawl or a pool walking session usually does not.

Why it ranks second

It is well supported, universally applicable and easy to sustain, and it supports the joint rather than strengthening the limb, which is why it sits below strength training.

What it involves

Losing weight where there is weight to lose, through diet alongside the activity in the first two entries.

What the evidence supports

Messier and colleagues measured knee joint loading in overweight and obese older adults with knee osteoarthritis and found each unit of weight lost was associated with roughly a four-unit reduction in knee joint forces during daily walking. NICE advises supporting weight loss for people who are overweight or obese, and states that any amount of weight reduction is beneficial.

Why the multiplier matters

The four-to-one relationship means small changes produce load reductions that are larger than they sound. A few kilograms lost removes a meaningful amount of force from the joint on every step, accumulated across thousands of steps a day.

Who it suits

People carrying extra weight, which is a large share of people with knee osteoarthritis. It has no application to someone at a healthy weight, and it should never be the only advice given.

Limits and cautions

This is difficult, it is slow, and it is frequently delivered as a one-line instruction with no support attached, which is why it so often achieves nothing. NICE frames it as something to support people with rather than something to tell them.

Why it ranks third

The mechanical evidence is strong and the guidance is clear, and it applies to a subset of people rather than to everyone, which is the only reason it sits below the first two.

What it involves

An unloader knee brace designed to shift load away from the affected side of the joint, or a wedged insole that changes the alignment through the foot.

Who it suits

People with cartilage loss concentrated in one compartment, most commonly the inner side of the knee, and particularly where the leg alignment contributes. Someone with even wear across the whole joint has less to gain.

How it works

The effect is purely mechanical. Changing the alignment of the limb shifts where force lands within the joint, moving some of it from the worn surface onto the surface that is intact. The benefit lasts exactly as long as the device is worn.

What the evidence supports

The case rests on measurable changes in joint loading rather than on a strong recommendation in national guidance. It is a reasonable option to trial where the wear pattern is compartmental, judged on whether it helps within a few weeks.

Limits and cautions

Braces are bulky, they are frequently abandoned for that reason, and they can rub. An insole is easier to live with and produces a smaller effect. Neither changes what the knee can tolerate when the device comes off.

Why it ranks fourth

It offers something specific to a defined subgroup, and it changes nothing about the joint or the limb once it is removed.

What they are

Corticosteroid, hyaluronan and platelet-rich plasma, all injected into the joint, and the latter two are frequently marketed as regenerative treatments that restore cartilage.

Corticosteroid

NICE positions intra-articular corticosteroid injections as a short-term option where other treatments have not worked or to support therapeutic exercise, noting that the evidence is inconsistent, that effects last around 2 to 10 weeks, and that no benefit is apparent beyond three months. Used to open a window for rehabilitation, that is a genuine role.

Hyaluronan

NICE advises that intra-articular hyaluronan injections should not be offered for osteoarthritis, citing inconsistent benefits and potential harms, notably at the hip. That is a direct recommendation against a treatment still widely sold privately.

Platelet-rich plasma

The RESTORE trial randomised 288 people with mild to moderate knee osteoarthritis to three weekly injections of leukocyte-poor platelet-rich plasma or saline placebo. At 12 months the change in knee pain was 2.1 points with PRP against 1.8 with placebo, and medial tibial cartilage volume changed by 1.4 per cent against 1.2 per cent. The authors concluded that the findings do not support the use of PRP for the management of knee osteoarthritis.

Limits and cautions

Repeated corticosteroid injections into the same joint are generally avoided. Where an injection is offered as a way to regrow cartilage, the trial evidence above is the thing worth asking about before paying for it.

Why it ranks fifth

One of the three has a defined short-term role, one is recommended against, and one failed its largest placebo-controlled trial on both pain and cartilage volume. Together that puts the category at the bottom of the list.

The fourth column is where the marketing and the evidence diverge most sharply. Nothing on this list regrows hyaline cartilage, and the top three entries produce knees that work better regardless.

What each option asks of you, which is the axis on which these actually differ.

The inverse relationship between effort and evidence runs right through this table, and it explains why the injectable options sell well. Fifteen minutes in a clinic is an easier proposition than three months of strength training, and the trial evidence sits with the harder one.

If you have been told your knee is bone on bone, strength training is still the first thing to do. Radiographic severity corresponds poorly with symptoms, and plenty of severely worn knees become comfortable with stronger muscles around them.

If your knee hurts most on the inner side and your legs bow outwards, an unloader brace or a wedge insole has something specific to offer alongside the strengthening.

If pain is too high to start exercising at all, a corticosteroid injection to open a 2 to 10 week window is a reasonable use of one, provided the rehabilitation is booked to start inside it.

If you have been offered PRP to regenerate your cartilage, ask about the RESTORE trial. Three weekly injections produced no significant difference from saline in either pain or cartilage volume at 12 months.

If you are under 30 with a single deep cartilage defect after an injury, this page is the wrong reference point. That is a focal defect rather than osteoarthritis, and it warrants a surgical opinion.

If the knee locks or gives way, something mechanical is happening beyond cartilage wear, and it needs assessing before a loading programme starts.

Where it is still unclear whether wear is the cause at all, our guide to knee pain covers how the common causes are told apart.

The belief that brings people in is that cartilage can be regrown, and that somewhere there is an injection that does it. It is an easy belief to acquire. Private clinics advertise regenerative knee treatment, the language around platelet-rich plasma is built on growth factors and healing, and the alternative on offer is usually described as just physio.

What the evidence shows is a cleaner picture than either side of that. Adult hyaline cartilage does not come back, and the largest placebo-controlled trial of platelet-rich plasma in knee osteoarthritis found no significant difference from saline in pain or in cartilage volume at 12 months. Meanwhile the knees that improve most in clinic are the ones attached to people who spent three months getting stronger. The improvement is real, it shows up in walking distance and stairs and getting out of a chair, and none of it involves the cartilage changing at all. What changes is how much load reaches the worn surface and how well the limb controls it.

At PhysioHub in Uckfield, that reframing is usually the first useful thing that happens, because it moves the target from a scan finding to something that can actually be improved. Our 1:1 rehab and strength training is where those three months of getting stronger are planned and progressed.

Can knee cartilage repair itself without surgery?
Adult articular cartilage has no blood supply and does not regrow into its original structure once lost. Symptoms and function commonly improve a great deal without surgery, through strengthening, low-impact activity and weight management, so a knee can become pain-free with the cartilage loss still present.

Does PRP regrow knee cartilage?
The RESTORE trial randomised 288 people with knee osteoarthritis to three weekly platelet-rich plasma injections or saline placebo. At 12 months pain changed by 2.1 points with PRP against 1.8 with placebo, and medial tibial cartilage volume by 1.4 per cent against 1.2 per cent. The authors concluded the findings do not support using PRP for knee osteoarthritis.

Are hyaluronic acid injections recommended for knee cartilage loss?
NICE guideline NG226 advises that intra-articular hyaluronan injections should not be offered for osteoarthritis, citing inconsistent benefits and potential harms. They remain available privately, and the national guidance recommends against them.

How does cycling help knee cartilage?
Cartilage has no blood supply and is nourished by fluid moving in and out of it. The repeated gentle compression and release of low-resistance cycling pumps waste out and draws nutrient-rich joint fluid in, while producing the movement without impact. Keep the seat high and the resistance low.

How much weight do I need to lose to help my knee?
Research measuring knee joint loading found each unit of body weight lost was associated with roughly a four-unit reduction in knee joint force during walking, and NICE states that any amount of weight reduction is beneficial. Small losses therefore produce larger load reductions than they appear to.

Will exercise wear my knee cartilage out faster?
No, and NICE addresses this directly, noting that exercise may increase joint pain initially and that this diminishes with regular consistent activity while function and quality of life improve. Therapeutic exercise is the core of osteoarthritis management rather than a risk to the joint.

Do knee braces work for cartilage loss?
An unloader brace helps where wear is concentrated in one compartment, usually the inner side of the knee, by shifting load onto the intact surface. The effect is mechanical and lasts only while the brace is worn, so it works best alongside strengthening.

When is surgery the right answer for knee cartilage?
A focal full-thickness defect with exposed bone in a younger person, a loose fragment causing the knee to lock, or end-stage joint damage with night pain after conservative treatment has been properly tried. Diffuse mild to moderate cartilage loss is generally managed without an operation.