What usually comes next

Most people at this stage are not moving straight to a knee replacement, and routine arthroscopy is usually not the answer either. A more realistic next step is a better-structured non-operative plan, sometimes paired with a targeted symptom-control option. In contemporary guidelines and major reviews, the most strongly supported measures for knee osteoarthritis remain exercise, weight loss when relevant, self-management, bracing or a cane, topical or oral NSAIDs, and intra-articular glucocorticoid injection for symptom relief. Knee replacement is generally held back for people with advanced symptoms and structural damage after these measures are no longer enough.

• Exercise and activity programming: usually a specific, progressive plan rather than a few scattered sessions or a vague home sheet. This can still be worthwhile even after a previous round of “physio” that was brief, interrupted, or hard to keep going.
• Weight loss, if overweight or obese: guideline-backed and often one of the highest-yield steps for pain and function.
• Self-management, a cane, or tibiofemoral bracing: simple tools, but guideline summaries support them for the right pattern of knee OA.
• Topical or oral NSAIDs: among the best-supported medicines short of surgery, when they are appropriate for the individual.
• Corticosteroid injection: useful as a symptom-relief option, not a joint-restoring fix.

The usual aim is better pain control and more reliable day-to-day function, not a fully “normal” knee. Routine arthroscopic lavage or debridement is not usually presented as part of a standard stepped pathway for ordinary degenerative knee OA symptoms unless there is another specific mechanical problem to explain it.

Which option fits your knee and goals

Two knees labelled “osteoarthritis” can point to different next steps. Rather than running through treatments again, the useful sorting starts with the details that actually change the path: whether the wear is mainly medial, lateral or patellofemoral, how advanced it looks on X-ray or MRI, whether there is varus “bow-leg” alignment, body weight, what has already been tried, and whether any earlier injection helped for days, months, or not at all. Medial-compartment OA with varus alignment, for example, raises different joint-preserving questions from pain centred behind the kneecap.

Pain and pictures also need to be kept separate. A knee that hurts badly on stairs, wakes someone at 3 am, or cuts an 18-hole golf round short does not always have the most dramatic scan; the reverse can happen too. That is why a report saying “moderate” or “severe” OA is only one part of the decision. The real target may be walking tolerance, better sleep, more confidence on stairs, a return to tennis, or delaying replacement without making function worse.

So the purpose of assessment is usually to sort pattern, severity, alignment, BMI, treatment history, and goals before calling one option the “best” fit. Reassessment matters more when symptoms are worsening quickly over weeks, the knee truly “locks”, there is major giving-way, or there is real doubt that osteoarthritis is the main diagnosis.

Have you really exhausted non surgical care

A knee has not usually exhausted non-surgical care just because “physio” happened at some point. A more meaningful test is whether there has been a structured block of treatment aimed at function: in guideline summaries and major reviews, exercise and self-management are core measures, and in practice that means building quadriceps and hip strength, extending walking tolerance, improving stair confidence, and having a clear way to reduce and then reload activity after a flare. A few generic stretches, long spells of rest, or a stop-start home sheet do not really show what the knee can do.

Body weight can also be a treatment lever rather than a side issue. Guidelines strongly back weight loss for people who are overweight or obese, and a 68-week randomised trial found that semaglutide plus diet and activity counselling produced 13.7% mean weight loss versus 3.2% with placebo, alongside greater improvement in WOMAC pain. That does not make drug treatment appropriate in every case, but it does show that substantial weight reduction may materially change pain, walking distance, and what needs to happen next.

A genuinely optimised programme usually includes simple supports that are often skipped. Guideline summaries also support tibiofemoral bracing, cane use, and topical or oral NSAIDs in appropriate patients; in everyday terms that can mean an unloading brace for the right pattern of OA, a stick in the opposite hand, footwear changes, pacing a shopping trip or golf round, and a 24- to 48-hour flare plan rather than abandoning activity altogether. NSAIDs can be useful symptom-control tools when safe, but gastrointestinal, renal, and cardiovascular risk still matter. The aim is a knee that stays usable through good and bad weeks, not one that feels perfect every day.

Which injections or procedures are worth considering

For the procedural layer of treatment, the simplest ranking is by certainty rather than novelty. In guideline summaries and major reviews, intra-articular glucocorticoid injection is the clearest supported procedure for knee OA, and it is best framed as symptom relief rather than a durable fix. In practice, that makes a steroid injection the most defensible next procedural step when pain is disrupting sleep, blocking exercise progression, making travel difficult, or stalling rehabilitation. It may ease pain and improve function for a period, but it does not repair cartilage or reverse established OA.

• Most established for short-term relief: corticosteroid injection. The evidence and guideline support are stronger here than for the newer options, but the trade-off is that benefit is usually symptom control rather than restoration.
• More situational: hyaluronic acid, and in some practices PRP. Guidance is mixed, especially for hyaluronic acid: some recommendations advise against it, while other guidelines still allow it for symptom relief when other nonoperative options are ineffective.
• Reasonable second-line pain procedure: genicular nerve radiofrequency ablation. Guideline summaries describe it as a conditional option, and a 2025 systematic review of 2,218 participants found about 51% achieved at least 50% pain reduction at 6 months, falling to 43% at 12 months. That points to symptom control, not joint repair.
• More experimental: genicular artery embolisation. This remains an emerging option with a limited evidence base, so its role is less established.

When joint preserving surgery is realistic

A realistic pre-replacement surgical discussion usually starts with a pattern, not just a bad pain score. The clearest example is an “inside-worn, bow-legged” knee: pain mainly from the medial compartment, varus alignment on examination or X-ray, and enough activity goals left that preserving the native joint still matters. In that narrower group, high tibial osteotomy (HTO) is the main joint-preserving operation worth discussing because it aims to shift load away from the overloaded inner side of the knee rather than replace the joint outright.

That said, HTO is a selective option rather than a standard next step for knee OA. A 2024 systematic review covering 18 studies and 1,296 knees reported average 10-year survivorship of 74.6% even in advanced medial OA, but older age, higher BMI, more diffuse compartment involvement, and greater symptomatic disability all make results less predictable. A separate 20-year study found better survivorship in patients aged under 55 and with BMI under 30, which is why an older active adult needs particularly careful selection.

By contrast, arthroscopic lavage or debridement should not be framed as a routine halfway house before replacement. Surgical review is most useful when symptoms and imaging point to that specific mechanical pattern, rather than broad whole-knee wear.

When replacement becomes the better path

Replacement tends to become the better path when the knee is shrinking life on an ordinary week, not just flaring after an unusual one. Typical signs are pain on most days, pain at night, walking or stairs becoming reliably difficult, repeated flare-ups over months, and giving up valued activities such as golf, gardening or longer walks because the knee no longer settles back to an acceptable baseline. In a JAMA review, knee replacement was framed as the stage reached when symptoms are advanced and non-operative care is no longer giving adequate control of pain and function.

Being “not ready” is not a fixed category. At 65 or 75, someone may be emotionally not ready for an operation, functionally still coping well enough, or medically better served by waiting until other health issues are steadier. That can change over a 6- or 12-month period as symptoms, confidence and general health shift.

At that point, the useful conversation is usually quite focused:

• what has been tried properly and for long enough
• what only helped temporarily, such as a short-lived injection response
• what the knee anatomy now shows, including how widespread the wear is
• what matters most now: pain relief, walking distance, stairs, sleep, or returning to a specific activity

Choosing replacement after a genuine stepped approach has stopped delivering acceptable quality of life is not a failure. The clearest signal is often simple: when the effort of delaying starts to cost more than the operation is likely to. The next step is usually a targeted assessment that weighs symptoms, function and current imaging together, rather than reacting to one scan report or one bad week.

1. [1] Diagnosis and treatment of hip and knee osteoarthritis: A review. (2021). https://doi.org/10.1001/jama.2020.22171 https://doi.org/10.1001/jama.2020.22171