Many people told they "need" a knee replacement have real options to try first. Weight loss, targeted strengthening, and physical therapy have the strongest evidence and help most people with mild-to-moderate knee arthritis. Braces and injections, including PRP for the right knee, can buy meaningful time. But some knees are genuinely end-stage, and for those, surgery is the honest answer.
Here is the part most ads skip: no injection, brace, or supplement regrows a bone-on-bone joint. What the non-surgical ladder can do is reduce pain, restore function, and often delay or avoid replacement when arthritis is not yet end-stage. The trick is matching the option to the stage of your knee, and being honest about which options are well supported and which are oversold. Below is the full ladder, from least to most invasive, with the rough strength of evidence for each. If you want the deeper background on the disease itself, see our overview of knee pain and arthritis.
Start here: the foundation (strong evidence)
Before any needle or scalpel, this is the layer with the best science behind it. Every major osteoarthritis guideline, including those summarized across six professional societies, puts these first.
Weight management
Each pound of body weight translates to several pounds of load across the knee with every step. Losing even 5 to 10 percent of body weight measurably reduces pain and slows cartilage loss. It is unglamorous and it is the single most powerful lever many patients have. Weight loss combined with exercise produces better results than either alone (narrative review, Knee Osteoarthritis Rehabilitation, PMC 2024).
Activity modification and assistive devices
Swapping high-impact loading (running on pavement, deep squats under load) for low-impact movement (cycling, swimming, the pool) keeps you active without grinding the joint. A simple cane in the opposite hand can offload the painful knee enough to keep you walking. These are low-cost, low-risk, and underused.
Physical therapy and strengthening
Strong quadriceps and hips act like a built-in brace for the knee. Supervised exercise therapy, including resistance, aerobic, and neuromuscular training, has robust evidence for reducing pain and improving function. This is the cornerstone of non-surgical care, not an afterthought. If a treatment plan skips strengthening, it is skipping the most proven part.
If someone offers you an injection but never mentions weight, strength, or activity, be skeptical. The foundation has the best evidence of anything on this page, and it costs the least.
Symptom control: medication, bracing, and offloading (moderate evidence)
NSAIDs and topical anti-inflammatories
Oral NSAIDs (ibuprofen, naproxen) and topical NSAID gels (diclofenac) reliably reduce arthritis pain. Topicals are often preferred for the knee because they deliver relief with fewer stomach, kidney, and heart risks than pills, especially in older adults. These manage symptoms; they do not change the disease, and long-term oral use carries real side-effect risk worth discussing with your physician.
Unloader braces
For arthritis concentrated on one side of the knee (commonly the inner, medial compartment), an unloader brace applies gentle force to shift load off the worn area. The evidence is moderate, not overwhelming: the AAOS guideline supports bracing with moderate-quality evidence, and trials like the ROTOR randomized study (Scientific Reports, 2018) show benefit for selected patients. Fit matters, so these work best when prescribed and adjusted properly rather than bought blindly online.
Injections: where honesty matters most
This is where marketing and evidence diverge the most, so we will grade each one plainly.
Corticosteroid (cortisone) injections: fast, short, and not free of cost
Cortisone shots can quiet an inflamed, painful knee within days, which is genuinely useful for a flare or to get through an event. The catch is duration: relief usually fades within weeks to a few months. And repeated use is not harmless. A landmark randomized trial (McAlindon and colleagues, JAMA, 2017) found that triamcinolone injections every three months for two years produced greater cartilage loss than saline, with no better pain control. Translation: as an occasional rescue, cortisone has a place; as a routine, repeated strategy, it can work against the joint you are trying to save.
Hyaluronic acid (viscosupplementation): mixed evidence
Hyaluronic-acid "gel" injections aim to lubricate and cushion the joint. The honest verdict is that the evidence is mixed. Some patients get months of relief; large analyses and several guidelines find the average benefit modest and inconsistent, which is why some societies no longer routinely recommend it. It is reasonable to try for some patients, but it is not a reliable home run.
Platelet-rich plasma (PRP): favorable for the right knee, not a cure-all
PRP uses a concentrate of your own platelets to stimulate healing and calm inflammation. For knee osteoarthritis, the evidence is genuinely encouraging when the knee is the right candidate. Multiple meta-analyses of randomized, controlled trials find PRP outperforms hyaluronic acid, with roughly double the odds of a good outcome (odds ratio of about 2.19; systematic review and fragility-index meta-analysis, Arthroscopy, 2024). In moderate, Kellgren-Lawrence grade-3 arthritis, around 59 percent of patients did well when treated with three or more injections in pooled data.
The honest caveats matter. The American Academy of Orthopaedic Surgeons (AAOS, 2021) gives PRP only a limited recommendation, largely because preparation methods vary so widely between clinics that results are hard to compare. That is a fair criticism, and it is exactly why an ultrasound-guided, surgeon-performed protocol matters. Most important: PRP works best before a knee is bone-on-bone. In end-stage joints there is little cartilage left to influence, so it is not a substitute for replacement. We say so up front. For the deeper evidence breakdown, see is PRP worth it, and for pricing see our open cost page.
Less-invasive procedures (case-by-case)
Between conservative care and full replacement sit several targeted procedures. These are not for everyone, but they can fit specific situations.
- Osteotomy: the surgeon cuts and realigns the shin or thigh bone to shift load off the worn compartment. Often a good fit for younger, active patients with one-sided arthritis and good alignment potential.
- Partial (unicompartmental) knee replacement: resurfaces only the damaged compartment, preserving the rest of the joint. Smaller surgery and faster recovery than total replacement when only one part is worn out.
- Genicular nerve block or radiofrequency ablation: targets the nerves carrying pain signals from the knee. Does not treat the arthritis but can manage pain and help some patients delay or avoid replacement.
- Arthroscopy for mechanical problems: useful for a locking meniscus tear or loose fragment, but it is not an effective treatment for arthritis pain alone, a point most guidelines now make clearly.
The honest comparison
Here is the full ladder side by side. "Evidence" reflects the overall strength of high-quality research for arthritis pain, not a promise about your specific knee.
| Option | What it does | Evidence | Best for |
|---|---|---|---|
| Weight management | Reduces load and inflammation across the joint | Strong | Nearly everyone with knee OA |
| PT and strengthening | Builds muscle support around the knee | Strong | All stages, the foundation of care |
| NSAIDs / topicals | Reduces pain and inflammation (symptom control) | Moderate | Flares; topicals safer for older adults |
| Unloader brace | Shifts load off the worn compartment | Moderate | One-sided (medial) arthritis |
| Cortisone injection | Fast, short-term pain relief | Short-lived; repeat use can harm cartilage | Occasional flare rescue, not routine |
| Hyaluronic acid | Lubricates / cushions the joint | Mixed | Selected patients willing to try |
| PRP | Stimulates healing, calms inflammation | Favorable for mild-to-moderate OA | KL grade 2 to 3, before bone-on-bone |
| Osteotomy | Realigns bone to offload worn area | Good (selected cases) | Younger, active, one-sided arthritis |
| Partial replacement | Resurfaces only the damaged compartment | Strong (right candidate) | One worn compartment, rest healthy |
| Total replacement | Replaces the entire joint surface | Strong (end-stage) | Bone-on-bone, failed conservative care |
When you DO need surgery
Radical honesty cuts both ways. Just as some clinics oversell injections, it would be wrong to imply every knee can be saved. Both Johns Hopkins Medicine and Mayo Clinic, in their patient guidance on knee-replacement alternatives, are clear that conservative options are worth trying first, and equally clear that some knees genuinely need replacing.
Replacement is usually the right answer when several of these are true:
- End-stage, bone-on-bone arthritis on imaging, often with deformity (bowing) or instability.
- Pain that limits daily life and disrupts sleep, not just occasional ache.
- A fair trial of conservative care has failed, including weight loss, strengthening, bracing, and, where appropriate, injections.
- Function is severely restricted, you cannot walk the distances or do the activities that matter to your life.
Timing is its own honest conversation. Waiting can be reasonable while a knee is still salvageable, but waiting too long with severe deformity and muscle wasting can make surgery and recovery harder. And modern joint replacement, in the right patient, is one of the most reliable, life-changing operations in all of medicine. The point is not to avoid surgery at all costs; it is to make sure you are in the group that truly needs it.
Dr. Hock is a surgeon. He performed thousands of orthopedic operations before turning to regenerative medicine. That is exactly why he can tell you honestly whether you are someone who can try the non-surgical ladder, or someone whose knee genuinely needs replacing. A doctor who only offers injections has a reason to say yes. A surgeon who offers both has no reason to do anything but tell you the truth.
The bottom line
If you have been told you need a knee replacement, do not assume it is your only path, and do not assume an injection will rescue any knee. Start with the foundation: weight, strength, activity. Use bracing and medication for symptom control. Consider PRP if your arthritis is mild to moderate and not yet bone-on-bone. And if your knee is genuinely end-stage and conservative care has not given you a livable result, trust the surgeon who will say so. The honest first step is finding out which group you are in. Take the 2-minute assessment to see whether you are a candidate for non-surgical care.
Medically reviewed
Reviewed by Dr. David Hock, DO, Board-Certified Orthopedic Surgeon, affiliated with Hilo Medical Center. Published June 15, 2026 · Last reviewed June 15, 2026. This article is for education and is not a substitute for an individual medical evaluation. Sources include McAlindon et al. (JAMA, 2017) on corticosteroid and cartilage loss, randomized-trial meta-analyses of PRP versus hyaluronic acid (Arthroscopy, 2024), the AAOS knee osteoarthritis guideline (2021), and patient guidance from Johns Hopkins Medicine and Mayo Clinic.
Frequently asked questions
Can I avoid a knee replacement?
Sometimes, and often you can at least delay one for years. Foundational steps like weight loss, targeted strengthening, and physical therapy have the strongest evidence and help most people with mild-to-moderate knee arthritis. Injections such as PRP can help when arthritis is not yet end-stage. But if your knee is truly bone-on-bone with deformity and conservative care has failed, no injection or brace rebuilds the joint, and replacement may be the honest answer.
Does PRP work for bone-on-bone knees?
Usually not well. PRP has its best evidence in mild-to-moderate (Kellgren-Lawrence grade 2 to 3) arthritis, where it can reduce pain and outperforms hyaluronic-acid injections in pooled trials (odds ratio of roughly 2.19). In end-stage, bone-on-bone knees there is little cartilage left to influence, so results are limited. PRP works best before a joint is bone-on-bone, not after.
How long can I put off knee replacement surgery?
There is no fixed timeline. Many people delay surgery for years with weight management, strengthening, bracing, and well-chosen injections, especially when arthritis is mild to moderate. The goal is to control pain and keep function while the joint is still salvageable. The trade-off: waiting too long with severe deformity and muscle loss can make eventual surgery and recovery harder, so timing should be an honest, individual decision.
When do I actually need a knee replacement?
Replacement is usually appropriate when arthritis is end-stage (bone-on-bone), pain limits daily life and sleep, you have significant deformity or instability, and a fair trial of conservative care, including weight loss, therapy, and possibly injections, has not given you a livable result. A surgeon who also offers non-surgical options is well placed to tell you honestly which group you fall into.