Do I Need a Knee Replacement?
There is no test that tells you when to have a knee replacement — it is a decision, not a diagnosis. What actually indicates it is time, why the X-ray matters less than you think, whether you are too young, and what the operation can and cannot deliver.
There is no test for this.
No X-ray appearance, no blood result, no age and no severity score makes a knee replacement necessary. It is an elective operation — one you choose when the pain and restriction outweigh the surgery and the recovery. That is a genuinely different kind of decision from most in medicine, and it belongs to you rather than to a scan.
This page sets out what actually indicates it is time, what does not, and what the operation can and cannot deliver.
The X-ray matters less than you think
Most people arrive having been told their knee is “bone on bone”, and treat it as a verdict.
It is a description of a film, not an instruction. Arthritis shows on the X-rays of more than half of people over 65, and only about one in seven of them have symptoms worth treating. Plenty of people have bone-on-bone changes and walk the dog every day without much trouble. Others have severe, life-limiting pain on films that look comparatively mild.
The X-ray confirms arthritis is there, shows which compartments are affected, and helps plan an operation. It does not tell you whether you need one. If a decision is being driven mainly by how your scan looks, something has gone sideways.
The four questions that actually matter
Ask yourself these honestly. They are the same ones a surgeon will work through with you.
1. Is pain limiting things you actually care about? Not “does your knee hurt” — most arthritic knees hurt. Has it started removing things from your life: work, walking, sleep, sport, travel, the floor with your grandchildren?
2. Is it disturbing your sleep? Night pain is one of the more reliable indicators that arthritis has moved past what conservative treatment manages well.
3. Has your world got smaller? People adapt so gradually they often do not notice. Have you stopped parking further away, taken the lift you used to skip, stopped going somewhere because of the walk? Ask someone who lives with you — they usually noticed before you did.
4. Has non-surgical treatment genuinely stopped working? Not “did I try a few things” — has a real go at exercise and, if relevant, weight stopped producing benefit? These are the two treatments with the strongest evidence, and a knee that has never had them tried properly has not yet had its best non-surgical shot.
If you are answering yes to most of these, it is reasonable to have the conversation. If you are answering no to most, it probably is not time — and that is a good position to be in, not a disappointment.
Signs it is probably not time yet
- Pain comes and goes, and good weeks still outnumber bad
- It flares with specific activity but settles with rest and simple measures
- You have not yet had a proper go at strengthening and load management
- The main driver is a scan result rather than what the knee is doing to your life
- Simple analgesia or a topical anti-inflammatory still makes a real difference
None of these mean you will never need surgery. They mean the operation would be trading a manageable problem for a recovery you do not yet need.
Signs it may be time
- Pain most days, largely regardless of what you do
- Waking at night, or struggling to find a comfortable position
- Walking distance meaningfully reduced, and still shrinking
- Using a stick or rail you did not need a year ago
- Stiffness or deformity that is visibly progressing
- Cortisone injections helping for shorter and shorter periods
- You have organised your life around the knee rather than the other way round
”Am I too young?”
This one causes more unnecessary suffering than almost anything else in this area.
The advice to hold off until you are older comes from an era of shorter-lasting implants. Data from the Australian Orthopaedic Association National Joint Replacement Registry — which captures more than 99% of joint replacements performed in Australia — shows a cumulative rate of major revision for non-infection reasons of around 3% at 15 years, across nearly half a million procedures.
Younger patients do have higher revision rates. That is partly the arithmetic of living longer with an implant, and partly that younger people ask more of their knees.
But the real question is not your age. It is whether the alternative — spending a decade avoiding things that matter to you, while the muscle weakens and the rest of your body compensates — is actually the safer option. Sometimes it is. Often it is not, and nobody has put it to you that way.
If you are in your fifties and being told to wait, that advice deserves a reason attached to it rather than being accepted as a rule.
Is there a cost to waiting?
No cliff edge, and nobody should be rushed into this.
But delay is not entirely free. Muscle weakens with reduced activity and has to be rebuilt afterwards, which lengthens recovery. Other joints and the opposite leg take up the load. And a knee that becomes severely stiff or badly deformed can make the operation more technically demanding and the outcome less predictable than it would have been earlier.
The aim is neither as early as possible nor as late as possible. It is the point at which symptoms consistently outweigh what non-surgical treatment can offer. Knee arthritis covers the underlying condition in more detail.
What the operation can and cannot deliver
Worth being straight about before you decide, because expectation is the single biggest driver of whether people are pleased afterwards.
What it does well. Relieves arthritic pain, in most people substantially. Restores walking distance and function. Corrects deformity. It is one of the more reliable operations in medicine.
What it does less predictably. Kneeling is often uncomfortable afterwards and sometimes permanently so. Some people notice clicking, numbness around the scar, or a knee that never feels entirely their own. Full deep flexion is not guaranteed. Recovery takes longer than most people are told — meaningful improvement continues for a year or more, not six weeks.
The satisfaction figure. Reported satisfaction after knee replacement commonly falls between 80 and 90% across studies, with some reporting higher and some lower. Which means a meaningful minority are not satisfied — and the commonest reasons are unmet expectations, ongoing pain and stiffness, rather than anything having gone technically wrong.
A replaced knee is a very good knee. It is not the knee you had at nineteen. Almost everyone who goes in expecting the first is pleased; a proportion of those expecting the second are not.
Seeing someone is not committing to anything
A consultation is an assessment and a discussion. It is not a booking.
It is entirely normal — and common — to be told a knee is not yet at the point where replacement makes sense, and to leave with a non-surgical plan instead. A surgeon who is comfortable saying “not yet” is giving you better information than one who books everyone who walks in.
What you should get is a clear picture of what is actually happening in the knee, what the realistic options are now, and what would change the answer. That is worth having whether or not you do anything with it.
Where to go next
If it is not time yet: Treating Knee Arthritis Without Surgery covers what actually works, measured against the Australian clinical guideline — including which widely sold treatments the evidence does not support.
If it might be: Total knee replacement and partial knee replacement explain the operations, robotic-assisted knee replacement covers how they are planned, and knee replacement recovery sets out the timeline week by week.
If you want to choose a surgeon well: How to Choose a Knee Replacement Surgeon covers what actually distinguishes one from another, and the questions worth taking into the room.
To talk it through, book a consultation, phone the rooms on (08) 6332 6365 or email info@rhysclark.com.au. Dr Clark consults at Murdoch and Mandurah.
How do I know if I need a knee replacement?+
There is no test that answers this. No X-ray appearance, blood test or age makes a knee replacement necessary — it is an elective operation, chosen when the pain and restriction outweigh the surgery and recovery. The questions that actually matter are whether pain is limiting things you care about, whether it disturbs your sleep, whether your walking distance has narrowed, and whether non-surgical treatment has genuinely stopped working. If the answer to most of those is yes, it is reasonable to have the conversation.
Does bone on bone mean I need a knee replacement?+
No. It is a description of an X-ray, not an instruction. Arthritis appears on the films of more than half of people over 65, and only around one in seven of them have symptoms worth treating — so plenty of people have bone-on-bone changes and manage perfectly well, while others have severe pain with relatively mild-looking films. The X-ray confirms arthritis is present and helps plan an operation. It does not decide whether you need one.
Am I too young for a knee replacement?+
Age alone rarely rules it out, though it does change the calculation. Australian registry data covering more than 478,000 knee replacements shows around 3% needed major revision surgery for non-infection reasons within 15 years. Younger patients do have higher revision rates, largely because they live longer with the implant and ask more of it. The practical question is not your age but whether you would spend the next decade avoiding things that matter to you, and that is worth weighing openly rather than being told to wait.
How long does a knee replacement last?+
Longer than most people expect. Data from the Australian Orthopaedic Association National Joint Replacement Registry, which captures more than 99% of joint replacements performed in Australia, shows a cumulative major revision rate for non-infection reasons of about 3% at 15 years across nearly half a million procedures. The old advice that a knee lasts ten years and should be delayed accordingly reflects implants and techniques from decades ago.
What happens if I wait too long for a knee replacement?+
There is no cliff edge, and nobody should be rushed. But waiting is not entirely free either — muscle weakens with reduced activity and has to be rebuilt afterwards, other joints and the opposite leg take up load, and a knee that becomes severely stiff or badly deformed can make the operation more technically demanding and the result less predictable. The aim is not to operate as early or as late as possible, but at the point where symptoms consistently outweigh what non-surgical treatment offers.
Will a knee replacement get rid of all my pain?+
Usually most of it, not always all of it. Most people are satisfied with the result — reported satisfaction across studies commonly falls between 80 and 90% — and the commonest reasons for dissatisfaction are unmet expectations, ongoing pain and stiffness rather than anything having gone wrong technically. A replaced knee is a very good knee, not a nineteen-year-old knee. Going in expecting the first is where satisfaction usually comes from.
Does seeing a surgeon mean I have to have surgery?+
No. A consultation is an assessment and a discussion, not a booking. It is entirely normal to be told your knee is not yet at the point where replacement makes sense, and to leave with a non-surgical plan instead. Understanding where your knee actually sits is useful information regardless of what you decide to do with it.