Do Knee Replacements Really Only Last
10 to 15 Years?

A gentleman in his late fifties came to my clinic at Aster MIMS Kannur recently, walking with a stick and clearly in pain, yet hesitating over surgery. When I asked why, he repeated something he had read online and heard from a relative. "Doctor, these knees only last ten or fifteen years. I am too young. I will need it done again, and the second one is worse." He had decided, on the strength of a number, to keep suffering. That number has probably talked more people out of a good decision than almost anything else I hear.

So let me deal with it properly, because it matters. Where did "ten to fifteen years" come from, is it still true, and what actually determines how long your particular knee replacement will serve you? I will give you the real figures, name where they come from, and be straight about the parts we still cannot promise.

Where the "ten to fifteen years" figure came from

It was not invented out of thin air. Decades ago, when knee replacement was a younger operation, the implants, the plastic bearing surfaces and the techniques were simply not as good as they are now. Surgeons counselling patients in the 1980s and 1990s were quoting the survival they had actually seen with the implants of that era. For those implants, that era and those patients, the caution was reasonable.

The problem is that the figure stuck, long after the thing it described had changed. The plastic has improved. Fixation has improved. The instruments and the understanding of alignment and balance have improved. Quoting the old survival of an old implant to a patient being offered a modern one is a bit like quoting the fuel efficiency of a 1990 car to someone buying a new model. The headline stayed in circulation while the engineering moved on underneath it.

What the best current evidence actually shows

Here is where we get away from anecdote and into real numbers, because we now have something we did not have years ago: enormous national registries that track almost every joint replacement done in a country, for decades, and tell us exactly how many are still working.

The landmark study here was published in The Lancet in 2019 by a Bristol research group, and it asked exactly our question as its title: how long does a knee replacement last? The authors pooled data from national joint registries with more than fifteen years of follow-up. Their headline finding, drawn from fourteen registries, was that around 82 percent of total knee replacements were still in place and working at 25 years. Roughly eight in ten last a quarter of a century. Not ten to fifteen years. Twenty-five.

A separate large population study from the United Kingdom, published in The Lancet in 2017, fills in the earlier part of the curve. It found that about 96 percent of total knee replacements were still in place at 10 years, and just under 90 percent at 20 years. And modern registry reports tell the same story for current implants: the American Joint Replacement Registry, now the largest in the world, reports that only around 2 percent of common cemented total knee replacements have needed to be redone by 10 years.

So the honest, evidence-based answer I gave that gentleman with the walking stick was this. For most people, a knee replacement is far more likely to outlast the old fifteen-year figure than to fall short of it.

So why does anyone still need a second operation?

Because no implant is forever, and I would be misleading you to suggest otherwise. The Lancet study that gave us the reassuring 25-year figure made the same point plainly in its own conclusion: given enough time, all knee replacements will eventually fail. The question is not whether failure is possible. It is how likely it is for you, and over what time frame.

When a knee replacement does need redoing, there are a handful of usual reasons. Studies that examine why knees are revised find the main culprits to be infection, the implant gradually loosening from the bone, the knee becoming unstable or stiff, wear of the plastic bearing, and fractures in the bone around the implant. One detailed analysis from a specialist centre found infection to be the single most common reason for revision, followed by loosening and then fracture around the implant. That centre sees a concentrated, complex caseload, so its proportions are higher than a whole-country average, but the list of reasons is the one every knee surgeon recognises.

Two things are worth pulling out of that list. First, infection is one of the most important early reasons a knee fails, and it is partly preventable. That is precisely why we are so particular about antiseptic precautions, antibiotics at the right moment, blood-sugar control and treating infections elsewhere in the body before surgery. Second, most of these problems are uncommon. They are the exceptions that make the survival curve dip, not the fate of the typical patient.

What actually shortens a knee replacement's life

This is the part patients most want, and the part where I have to be most careful to separate what the evidence shows from what sounds intuitive. Some of the things people fear turn out to matter less than they think, and some genuinely do count.

Age at surgery is the big one, and it is mostly arithmetic. A younger knee has more years ahead of it in which to wear or loosen, and tends to be worked harder. The 2017 Lancet study quantified this in a way I find genuinely useful in clinic. For someone having a knee replacement after the age of 70, the lifetime chance of ever needing it revised was only about 5 percent. For a man having the same operation in his early 50s, that lifetime risk rose toward 35 percent. This is not because a 50 year old gets a worse operation. It is because they live, and load the knee, for far longer afterwards.

A previous fracture in the same knee raises the risk. A large registry study from Denmark found that knees replaced because of arthritis following an old fracture had roughly one and a half to two and a half times the early and medium-term revision risk of knees replaced for ordinary arthritis. If your arthritis grew out of an old injury, that is worth knowing and planning for.

Weight is more nuanced than the lectures suggest. For years patients were told, bluntly, that being heavy would wear the knee out. The recent data are kinder. A large analysis from the American Joint Replacement Registry found no clear association between a higher body mass index and the implant loosening from the bone. I still encourage a healthy weight, because it helps your other joints, your heart and your recovery, and because very high weight carries its own surgical risks. But I no longer tell patients that their weight alone has doomed the implant, because the evidence does not support saying that.

And ordinary activity is not the enemy I was once taught it was. A recent study of recreational sport after hip and knee replacement found that patients doing intermediate or high-impact activity did not have higher revision rates than the sedentary ones at around five years of follow-up. That is reassuring, and it fits what I want for my patients, which is to use the new knee, walk, swim, cycle, garden and live, not to wrap it in cotton wool. I draw the line at violent, repetitive impact and at risky activities where a fall could fracture the bone around the implant, but a normally active life is part of the point of the operation.

Is the "second one is always worse" part true?

This was the other half of my patient's fear, and it deserves an honest answer rather than false comfort. Revision knee replacement, redoing a knee, is a bigger and more demanding operation than the first one. The surgeon has to remove the old components, deal with whatever bone has been lost, and rebuild. It generally asks more of the patient and the surgeon than a straightforward first replacement.

But, and this is the point that should lift the weight off, this is not a reason to refuse a first operation you genuinely need now. It makes no sense to endure years of pain and lost function today in order to avoid a possible, less likely than you fear, second operation decades from now. If and when a revision is ever needed, it is a well-established operation with good techniques behind it. The right way to think about it is the way I think about it: do the first operation well, at the right time, with the right implant, and most patients will simply never need the conversation about a second.

What this means for you, in plain terms

If you are weighing up a knee replacement and the "ten to fifteen years" line is sitting on your shoulder, here is what I would want you to take away. That figure is out of date. The best evidence we have, from the largest registries in the world, says roughly eight in ten knee replacements are still doing their job at 25 years, and the great majority sail through the first decade untouched.

Younger patients carry a higher lifetime chance of needing it redone someday, which is a fair thing to discuss when we decide on timing, but it is not a reason to suffer needlessly in your fifties. Infection and serious injury are the threats worth guarding against. And living a normal, active life is something to aim for, not to fear. The decision about when to have a knee replacement should rest on how much your knee is limiting your life and whether non-surgical measures have run their course, not on a number that describes an operation we have already left behind.