How Long Does a Knee Replacement Last? What Registry Data Really Shows

There’s a question patients often save for the end of the consultation, half-joking, hand already on the door: “Will this knee outlive me?” It’s a fair thing to want to know. A knee replacement is not a repair: it’s a machine, a construct of metal and medical-grade plastic installed in a joint that will bend more than a million times over the next decade.

For years, the standard answer was a shrug and a range: fifteen years, maybe twenty. That number came from small, aging studies of implants designed in the 1970s and 80s. It stuck around long after the evidence moved on.

The better answer now comes from national joint registries, databases in countries like Australia, the UK, Sweden, and Finland that track nearly every artificial joint implanted, sometimes for decades. What they show is more encouraging, more specific, and more honest about who wears an implant out and why.

How long does a knee replacement last, according to registry data?

The most rigorous answer comes from a 2019 analysis published in The Lancet, which pooled long-term case series with national registry data covering more than 290,000 total knee replacements. The headline finding: about 82% of total knee replacements were still functioning 25 years after surgery. Not 15 years. Not 20. A quarter of a century, for four out of five knees.

The numbers step down gradually rather than falling off a cliff:

Years after surgery	Total knee replacements still in place
10 years	Roughly 96%
15 years	About 93%
20 years	About 90%
25 years	About 82%

Two caveats keep this honest. First, these are population averages: your personal odds shift with age, weight, activity, and surgical factors, which we’ll get to. Second, the 25-year figures describe implants placed in the 1990s. Today’s designs and techniques may do better or, in a few cases, worse; only time and the registries will tell.

Still, the direction of the evidence is clear. For most people who get a knee replacement at the typical age, the late 60s, per Cleveland Clinic, the implant has a genuinely good chance of lasting the rest of their life.

Where did the old "15 to 20 years" number come from?

Blame small studies and slow news. Before national registries matured, longevity estimates came from single hospitals following a few hundred patients, often with implant designs that were already obsolete by the time the results were published. Those studies also suffered from a quiet statistical problem: patients who moved away, stopped returning calls, or died were lost to follow-up, and nobody knew whether their knees had failed or thrived.

Registries fixed much of this. Because countries like Australia and the UK record nearly every joint replacement performed nationwide, the UK’s National Joint Registry alone holds millions of procedures, a failed implant that gets revised anywhere in the country shows up in the data. That completeness is why registry-based survivorship estimates are considered the gold standard.

The “15 to 20 years” figure wasn’t wrong for its era; it described first- and second-generation implants in an age before modern polyethylene processing and refined surgical technique. It simply hardened into folklore. You’ll still hear it in waiting rooms, and it still shapes decisions, some people in their 50s delay surgery for years, enduring real pain, because they fear “using up” their one implant too early.

The current evidence justifies a more nuanced conversation. Age at surgery does matter, as we’ll see. But the arithmetic of that conversation should start at 25 years, not 15.

What does "lasting" actually mean? Survivorship, explained

When researchers say an implant “survived,” they mean one specific thing: it hasn’t been surgically revised. Revision, an operation to replace or adjust part or all of the implant, is the endpoint registries count, because it’s unambiguous and reliably recorded.

That definition has an honest wrinkle worth understanding. An implant can technically “survive” while its owner is unhappy with it. Studies summarized by the NHS and Mayo Clinic consistently find that most people get meaningful pain relief and better function after knee replacement, but somewhere around one in ten remain dissatisfied, usually because of lingering discomfort, stiffness, or a knee that never quite feels natural. Those knees still count as survivors in the registry.

The reverse is also true. Some implants that are loosening or wearing don’t get revised, because the patient is elderly, medically fragile, or simply prefers to live with the symptoms rather than face another operation. So true mechanical failure rates sit somewhere near, but not exactly on, the revision numbers.

Why does this matter to you? Because “how long will it last” is really two questions: how long before the implant needs another surgery, and how long before it stops serving you well. Registry data answers the first with real precision. The second depends on the fit between the implant, the surgery, and your expectations, which is exactly the conversation to have with your surgeon before the operation, not after.

Do partial knee replacements wear out faster?

Yes, and this is one of the most useful facts registry data has given patients. A partial (unicompartmental) knee replacement resurfaces only the damaged compartment of the knee, usually the inner side, leaving the rest of the natural joint alone. The same 2019 Lancet analysis that found 82% of total knees lasting 25 years found roughly 70% of partial knees lasting that long.

Put differently: at 25 years, a partial knee is about twice as likely to have been revised as a total knee.

Before you cross partials off your list, consider what they offer in exchange:

• A smaller operation with less bone removed and typically a quicker recovery
• A knee that many patients say feels more natural, because both cruciate ligaments are usually preserved
• A generally more straightforward revision if one is eventually needed, since more original bone remains

There’s also a subtlety in the data: some partial knees are revised not because the implant failed, but because arthritis progressed in the untreated compartments: the implant did its job while the rest of the joint aged. And because converting a partial to a total knee is a relatively contained procedure, surgeons may set a lower threshold for revising a partial than they would for redoing a total knee.

The honest framing: a partial knee trades some longevity for a lighter surgical footprint. For the right knee, arthritis confined to one compartment, intact ligaments, that can be a reasonable trade. It’s a decision to make with imaging and a surgeon, not a statistic alone.

What happens 20 years after a knee replacement?

For most people, remarkably little, and that’s the point. Registry data indicates that about 90% of total knee replacements are still in place at the 20-year mark. The typical 20-year-old implant is doing what it did in year two: bearing weight, bending through stairs and car seats, and staying quiet.

That said, two decades of use leave a signature. The polyethylene liner, the plastic spacer between the metal components, is the implant’s wear surface, and it slowly sheds microscopic particles with every step, the way a car tire sheds tread. In most knees this wear is modest and harmless. In some, the accumulated particles trigger an immune response called osteolysis, in which the body’s own cells gradually thin the bone around the implant. That thinning can loosen the components’ grip.

This is why surgeons often recommend periodic follow-up X-rays even when the knee feels fine, commonly at intervals of a few years in the second decade. Osteolysis and early loosening are frequently visible on an X-ray before they cause pain, and catching them early can mean a simpler revision, sometimes just exchanging the worn liner rather than rebuilding the whole joint.

The other change at 20 years is you. The muscles, balance, and bone density surrounding the implant matter as much as the hardware. Staying active, keeping quadriceps strong, and protecting bone health are the maintenance schedule for the human half of the machine.

How do you know when your knee replacement is wearing out?

A failing implant usually announces itself gradually, and the signals are worth memorizing, partly so you can act on them, and partly so you don’t panic over normal sensations like occasional clicking, which is common and typically benign.

The patterns that genuinely warrant attention, per Cleveland Clinic and Mayo Clinic guidance:

• New pain after a long quiet period. The most telling symptom of loosening is a deep, aching pain, often felt when you first put weight on the leg after rest, in a knee that had been comfortable for years.
• Swelling that returns and lingers. Recurrent warmth or fluid in the joint without an obvious injury can signal wear debris, loosening, or low-grade infection.
• Instability. A sense that the knee gives way, buckles, or won’t be trusted on stairs suggests the components or supporting ligaments are no longer working together.
• New stiffness or lost motion. A knee that used to straighten fully but no longer does, or a shrinking range of bend.
• Grinding or clunking that’s new. Not the familiar soft click, but a changed sound or sensation accompanied by discomfort.

None of these proves the implant is failing, arthritis in nearby joints, tendon irritation, and back problems can all masquerade as implant trouble. But all of them justify an evaluation, usually starting with an exam and X-rays, sometimes blood tests to rule out infection. The mistake to avoid is waiting: a loosening implant that grinds against bone for years can destroy the very bone a surgeon would need for a sturdy revision.

Why do knee replacements fail?

“Wearing out” is really shorthand for several distinct failure mechanisms, and the mix changes over time.

In the first two years, the leading causes of revision are infection and instability. Infection deserves its own paragraph: it complicates roughly 1% of knee replacements, and it’s the failure mode least related to the implant itself. Bacteria can reach the joint during surgery or seed it later through the bloodstream. Because bacteria form protective films on metal and plastic that the immune system can’t clear, a deep implant infection almost always requires surgery to resolve, sometimes staged over months.

In the medium term, instability and stiffness dominate: the components are intact, but the knee’s soft-tissue balance isn’t right, producing a joint that wobbles or won’t bend.

In the long term, aseptic loosening takes over as the leading cause. “Aseptic” means no infection is involved; the bond between implant and bone simply fails, often driven by the osteolysis process described earlier, in which microscopic polyethylene wear particles provoke bone loss. Fractures of the bone around an implant, often after a fall in later life, account for another meaningful share.

Understanding the mechanisms clarifies what you can influence. You can’t control every variable, but prompt treatment of infections anywhere in the body, fall prevention, bone health, and sensible activity choices each target a real failure pathway. The rest, component alignment, ligament balancing, fixation technique, is decided in the operating room, which is a strong argument for choosing an experienced surgical team.

Does your age at surgery change how long the implant lasts?

More than any other single factor, yes, but not for the reason most people assume. The implant doesn’t know how old you are. What matters is arithmetic: a 55-year-old asks the implant to survive perhaps 30 years of relatively vigorous use, while a 75-year-old may ask for 15 gentler ones.

Registry-based analyses in the UK have quantified this bluntly. For patients having a knee replaced in their early 70s, the lifetime risk of ever needing a revision is only around 5%. For men in their early 50s, published estimates run as high as roughly one in three. Younger patients live longer with the implant, load it harder, and, being healthier surgical candidates, are more readily offered revision when problems arise.

This is the genuine tension behind the timing decision, and it deserves a fair hearing on both sides:

• The case for waiting: every year of delay is a year subtracted from the implant’s required service life, and first revisions have somewhat less predictable results than first replacements.
• The case against waiting too long: years spent avoiding a painful knee erode muscle, balance, heart health, and mood. Evidence summarized by Johns Hopkins and others suggests patients who arrive at surgery weaker and more limited tend to recover less function.

There’s no universal right answer. The modern registry data mostly removes false urgency in both directions: a knee replaced at 60 has a realistic chance of lasting for life, and delaying at heavy personal cost to “save” the implant is often a bad bargain.

Does weight or activity level shorten an implant's life?

Both matter, though the evidence is more nuanced than the finger-wagging version you may have heard.

Start with the biomechanics, because they’re striking: the knee carries roughly one and a half times body weight with each walking step, and several times body weight on stairs or rising from a chair. Every additional pound a person carries is multiplied at the joint surface, which means load on the polyethylene liner, the implant’s consumable part, scales with body weight. Registry data does show higher revision rates, particularly for infection and loosening, at higher body weights.

Here’s the nuance: studies also consistently show that people across a wide weight range get substantial pain relief and improved function from knee replacement. Major guidelines have moved away from rigid weight cutoffs toward individualized risk discussions. Weight is a modifier of risk, not a verdict, and framing it with shame helps no one. What the evidence supports is straightforward: any sustainable reduction in load, even modest, reduces stress on both the implant and the opposite knee.

Activity type matters as much as amount. The wear science distinguishes between cyclic low-impact loading, walking, cycling, swimming, which implants handle well for decades, and repetitive high-impact loading, which accelerates liner wear and raises loosening risk. Counterintuitively, being sedentary isn’t protective: inactivity weakens the muscles and bone that support the implant and raises fall risk. The sweet spot, supported by surgeon consensus and long-term follow-up data, is regular low-impact activity, most days, indefinitely.

What can you never do after a knee replacement?

The honest answer: the list of true “nevers” is shorter than the internet suggests, and much of it is surgeon judgment rather than hard evidence, because no one has run a trial randomizing implant patients to marathon running.

What most surgeons and institutions like Cleveland Clinic and Mayo Clinic consistently discourage long-term:

• Repetitive high-impact sport: distance running, competitive basketball or soccer, and sports built on jumping and hard pivoting. The concern is twofold, accelerated wear of the plastic liner and a higher chance of falls or fractures around the implant.
• Activities with high fall or collision risk, judged individually: think mogul skiing, not walking your dog on ice with reasonable care.

What’s generally encouraged: walking, cycling, swimming, golf, doubles tennis, hiking, elliptical training, dancing, and strength work. Many surgeons green-light skiing or singles tennis for experienced athletes who already have the skills.

Two everyday questions deserve straight answers. Kneeling: it is not dangerous to a well-fixed implant, but studies find around half of patients avoid it because it feels uncomfortable or strange over the incision. With padding and practice, many regain it. Squatting deeply: most implants tolerate the bend you can achieve; the limit is usually your motion, not a rule.

The framing that best matches the evidence isn’t a forbidden list: it’s a budget. Your liner has a finite number of high-load cycles in it. Spend them on the activities that make your life worth the surgery, and let the low-impact habits carry the daily mileage.

How many times can a knee be replaced?

There’s no fixed maximum, but each revision is a bigger engineering problem than the last, and the practical limit is usually bone.

A first revision is a well-established operation. Surgeons remove the original components, address whatever failed, worn liner, loose component, infection, instability, and implant revision components, which typically have longer stems that anchor deeper into the thigh and shin bones, sometimes with metal augments or bone graft to fill defects. Outcomes are good but measurably less predictable than first-time replacement: registry data shows revision implants themselves get re-revised at a higher rate, and patients on average recover somewhat less function.

A second revision is harder still. Every operation removes bone; every episode of loosening or infection erodes more. By a third operation, surgeons may be working with substantial bone loss, scarred soft tissue, and compromised ligaments, sometimes turning to hinged implants that supply their own stability because the knee’s ligaments no longer can.

In practice, most people never test these limits. Recall the registry arithmetic: with roughly 82% of total knees lasting 25 years, the majority of patients, especially those operated on after age 65, will need zero revisions. Among those who do need one, most need only one.

The takeaway for younger patients is not fear but planning. If you’re 52 and facing a knee replacement, it’s reasonable to assume one revision in your lifetime and to structure decisions, implant choice, activity habits, follow-up schedule, to make that revision as late and as simple as possible.

Do newer implants and robotic surgery last longer?

This is where an honest article has to slow down, because the marketing has outrun the evidence.

What has demonstrably improved: polyethylene. Modern processing produces liners that resist wear far better in laboratory testing than the plastic of the 1980s and 90s, and in hip replacements, where the evidence is longest, highly cross-linked polyethylene has clearly reduced wear-related revisions. For knees, lab data and mid-term results are promising, but knees load plastic differently than hips do, and 25-year human data on the newest materials doesn’t exist yet, for the simple reason that 25 years haven’t passed.

What remains unproven for longevity: robotic assistance, patient-specific cutting guides, and computer navigation. These tools measurably improve the precision of component placement, and precision is plausibly linked to durability. But registry analyses so far have not shown a consistent, meaningful reduction in revision rates compared with conventional technique in experienced hands. That may change as data accumulates; it hasn’t yet.

There’s a paradox worth naming: the implants with the best proven track records are, by definition, older designs, because only they have had decades to prove themselves. Several registries publish implant-specific survivorship, and well-established designs with 15-plus years of strong data are a defensible, even wise, choice. A brand-new design is a bet that it will beat a benchmark it hasn’t yet been measured against.

The variable with the most consistent evidence behind it isn’t the tool at all: it’s surgical and hospital experience. Higher-volume teams have lower complication and revision rates across studies. That’s the question worth asking.

When should you see a doctor about your knee replacement?

Two different timelines apply here, and both matter for how long the implant ultimately lasts.

Contact your surgical team promptly, within days, not weeks, if you notice:

• Increasing redness, warmth, or swelling around the knee, especially with fever or chills, possible infection, which is most treatable when caught early
• Drainage or an opening in the incision at any point, even years later
• New pain in the knee that follows an illness, a dental infection, or a skin infection elsewhere, bacteria can travel through the bloodstream to an implant
• Sudden inability to bear weight, a fall onto the leg, or a new deformity, possible fracture around the implant
• Calf pain, swelling in one leg, or shortness of breath in the weeks after surgery, which need urgent evaluation for a blood clot

Schedule a non-urgent evaluation if you notice: gradually returning pain in a previously comfortable knee, recurrent swelling without injury, a new sense of instability or giving way, shrinking range of motion, or new grinding sensations. These are the classic signals of wear or loosening described earlier, and X-rays usually clarify them quickly.

And even with no symptoms at all, keep the periodic follow-up visits your surgeon recommends, often every few years once the implant is past its first decade. Loosening and bone loss frequently show on X-ray before they hurt, and the earlier they’re found, the simpler the fix tends to be. An implant checkup is cheap insurance on a joint you’d like to keep for 25 years.

How to help your knee replacement last 25 years or more

You control roughly half the variables. Here’s where the evidence says your effort pays off.

Finish the rehabilitation, then never really stop. The implant is only as stable as the muscles around it. Strong quadriceps and hips reduce abnormal shear on the components and, just as important, reduce falls, a major cause of late fractures around implants. Twice-weekly strength work is a reasonable lifelong habit.

Choose low-impact as your default. Walking, cycling, swimming, and elliptical work deliver the health benefits of exercise while spending the liner’s wear budget slowly. Save high-impact activities for the moments that matter to you.

Guard against infection for life. Treat skin infections, urinary infections, and dental problems promptly rather than waiting them out; bacteria in the bloodstream can settle on an implant years after surgery. Tell every dentist and doctor you have an artificial joint, and ask your surgical team what precautions they recommend around invasive procedures, recommendations vary, and the evidence on routine preventive measures is genuinely mixed, so this is a personalized call.

Manage weight with self-respect, not perfectionism. Because the knee multiplies body weight several times over during daily activities, even modest, sustained changes meaningfully reduce load on the liner.

Protect your bones. Bone density around the implant determines how well it stays anchored and how it survives a fall. Weight-bearing exercise and age-appropriate bone health screening both serve the implant.

Show up for follow-up X-rays even when nothing hurts. Silent problems caught early are small problems.

Frequently asked questions

How long does a knee replacement last on average?

Registry data pooled in a 2019 Lancet analysis found about 93% of total knee replacements still in place at 15 years, roughly 90% at 20 years, and 82% at 25 years. These figures describe implants placed decades ago; modern designs may perform comparably or better. Individual results vary with age at surgery, weight, activity level, and surgical factors, so treat these as population averages rather than a personal guarantee.

What happens 20 years after a knee replacement?

Most 20-year-old implants, about nine in ten, are still working normally. The main long-term change is slow wear of the plastic liner between the metal components. In some knees, microscopic wear particles trigger gradual bone thinning around the implant, which can eventually cause loosening. That’s why surgeons recommend periodic X-rays even for comfortable knees: these problems are usually visible on imaging before they cause pain, when fixes are simpler.

How do you know when your knee replacement is wearing out?

The classic signals are new, deepening pain in a knee that had been comfortable for years, often worst when first bearing weight after rest, plus recurrent swelling without injury, a sense of instability or giving way, shrinking range of motion, or new grinding. Occasional painless clicking is common and usually harmless. Any of the warning signs justifies an exam and X-rays, since early detection of loosening typically means a simpler revision.

What can you never do after a knee replacement?

True prohibitions are few. Most surgeons discourage repetitive high-impact sports long-term, distance running, basketball, sports built on jumping and hard pivoting, because they accelerate liner wear and raise fracture risk. Low-impact activities like walking, cycling, swimming, golf, hiking, and doubles tennis are encouraged indefinitely. Kneeling isn’t dangerous to a well-fixed implant, though about half of patients find it uncomfortable. Specific limits should come from your own surgeon.

How many times can a knee be replaced?

There’s no fixed limit, but each revision is more complex because every operation removes bone and soft tissue quality declines. First revisions are well-established with generally good results, though less predictable than the original surgery. Second and third revisions are progressively harder and may require specialized implants. In practice, most patients never need any revision, and among those who do, most need only one in their lifetime.

Can a knee replacement last 30 years?

Some do. Registry data confirms 82% of total knees surviving 25 years, and a portion of those clearly continue beyond 30, long-term case series include implants functioning after three decades. There’s simply less data at that horizon, because few implants placed in the early 1990s have been tracked that long. For patients operated on at typical ages, in their mid-60s or later, the implant realistically may never need replacing.

Does being younger at surgery mean the implant wears out sooner?

It means the implant is more likely to be revised in your lifetime, partly because you’ll live with it longer and load it harder. UK registry-based analyses estimate a lifetime revision risk around 5% for patients in their early 70s, rising to as high as one in three for men in their early 50s. That’s a reason for younger patients to plan for a possible revision, not necessarily a reason to endure years of disabling pain.

Do partial knee replacements fail more often than total knees?

Yes. Registry data shows about 70% of partial (unicompartmental) knees surviving 25 years, versus about 82% of total knees. Some of that gap reflects arthritis progressing in untreated parts of the joint rather than implant failure, and surgeons may revise partials more readily because conversion to a total knee is relatively straightforward. Partials offer a smaller operation, faster recovery, and often a more natural-feeling knee in exchange.

Does running ruin a knee replacement?

There’s no definitive trial, but the wear science and most surgeons point the same direction: repetitive high-impact loading accelerates wear of the plastic liner and may increase loosening risk, so routine distance running is generally discouraged. Occasional jogging is a judgment call some surgeons permit for experienced runners. Low-impact endurance options, cycling, swimming, brisk walking, elliptical, deliver equivalent cardiovascular benefit while spending the implant’s wear budget far more slowly.

How can I make my knee replacement last longer?

Focus on what the evidence links to implant survival: keep the surrounding muscles strong to stabilize the joint and prevent falls, favor low-impact exercise as your daily default, treat infections anywhere in the body promptly since bacteria can seed an implant through the bloodstream, maintain bone health, manage weight sustainably because the knee multiplies body weight with every step, and attend periodic follow-up X-rays so silent wear or loosening is caught early.

References

• Knee Replacement (Arthroplasty): Cleveland Clinic
• Knee replacement: NHS
• Knee Replacement: MedlinePlus

This article is for general information only and is not a substitute for professional medical advice. Please consult a qualified doctor about your individual situation.