Signs of a Failed Knee Replacement (and What to Do Next)

Published 11 June 2026.

Most knee replacements never fail: in national registry data, about 82% are still in place at 25 years [1]. But new pain after a long good stretch, swelling and warmth that will not settle, a knee that gives way, motion you are steadily losing, or any drainage from the incision deserves a real workup, and drainage or fever is urgent.

I evaluate painful knee replacements every week, and most of those worried patients turn out to have a knee that is structurally fine. A minority have a genuinely failing implant, and for them waiting has a real cost: small problems tend to become bigger operations the longer they smolder. Here are the signs I take seriously, the ones I usually do not, and what to do next.

How Often Do Knee Replacements Actually Fail?

National joint-registry data covering nearly 300,000 total knee replacements show that about 93% are still in place at 15 years, about 90% at 20 years, and about 82% at 25 years [1].

I put that number first on purpose. Search this topic online and you will find injury law firms and clinics selling unproven injections, both with a financial interest in making failure sound common. It is not: a modern total knee replacement is one of the most reliable operations in orthopedics.

But uncommon is not never. Implants are mechanical devices living inside biology; a small percentage loosen, wear, become unstable, or get infected. The job is telling a knee that needs reassurance from a knee that needs a diagnosis.

What Are the Warning Signs of a Failed Knee Replacement?

In one large single-center series of 781 revision knee replacements, infection was the most common cause of failure within the first two years after surgery, and loosening was the most common cause later on [2].

That timing shapes my thinking: in the first couple of years I worry about infection first; later, loosening tops the list. Here are the signs that get my attention.

Pain that comes back after the honeymoon period

The single most telling pattern is change. A knee that recovered well, gave you years of good service, and has now developed a new, deepening ache, especially with the first steps after sitting or with weight-bearing, makes me think about loosening until the imaging says otherwise. Pain that has been identical since the day of surgery is a different, and usually less mechanical, problem.

Swelling and warmth that will not settle

Every replaced knee is swollen and warm in the early months. What concerns me is a knee that had quieted down and then starts swelling again, or warmth and repeated effusions that persist long past the point they should have faded. That pattern can reflect wear debris, instability, or a low-grade infection around the implant, which can smolder for months with mild symptoms.

A knee that gives way

Buckling on stairs or a sense that the knee shifts under load suggests the ligament balance or component position is no longer right. Patients often compensate quietly for months before they mention it. Instability rarely improves on its own, and it is worth an exam.

Stiffness, or motion you are steadily losing

There is a difference between a knee that never bent well after surgery and one losing motion it used to have. Progressive stiffness can mean scar tissue, what we call arthrofibrosis, but late loss of motion with pain or swelling is also one way infection announces itself, so I do not dismiss it.

New clunking or grinding

Soft, painless clicking is usually nothing; more on that below. A new, pronounced clunk or grind, particularly alongside pain or swelling, can signal wear of the plastic liner or a component problem, and it deserves an X-ray rather than a shrug.

Wound drainage, spreading redness, or fever: treat these as urgent

Fluid draining from the incision, redness tracking up the leg, or fever and chills in anyone with a joint replacement is urgent: call your surgeon’s office the same day, or go to urgent care or the emergency department. Deep infections are time-sensitive. Caught very early, some can be treated while keeping most of the implant; caught late, they almost always mean staged revision surgery. And do not start leftover antibiotics on your own; a partly treated infection is much harder to diagnose.

At a glance:

When Is It Not a Failed Knee Replacement?

In a systematic review of unselected patients, roughly 10% to 34% reported some persistent pain after knee replacement, about 20% in the best-quality studies, and persistent pain by itself is not the same thing as a failed implant [3].

Plenty of normal replaced knees click softly, especially coming down stairs; that is metal and plastic making contact, not a part coming loose. A numb patch beside the scar, discomfort with kneeling, and mild warmth for months are all compatible with a perfectly good result. A replaced knee is a very good knee, but it is rarely a silent one.

Timing matters too. I tell my patients to judge the knee at a year, not at three months: swelling, stiffness, and warmth at six or twelve weeks are recovery, not failure, and most knees keep improving for a full twelve months.

Then there is the hardest category: a knee that is well-fixed and uninfected on every test, but still hurts. That is exactly where I am most conservative, because revision gives its best results when we can name the problem we are fixing. Re-operating on a painful but well-fixed, non-infected knee is one of the least rewarding operations in my field, and a surgeon quick to offer it is worth a second opinion. If pain is your main symptom two or more years out, without the red flags above, I have written a separate guide to knee replacement pain after 2 years.

Why Do Knee Replacements Fail?

In that same revision series, the leading failure mechanisms were loosening (39.9%), infection (27.4%), instability (7.5%), periprosthetic fracture (4.7%), and arthrofibrosis (4.5%) [2].

Those percentages come from one large center, not a national registry, so treat them as a credible snapshot. They match the failure causes the American Academy of Orthopaedic Surgeons describes [5]:

• Loosening and wear. The implant-bone bond can fail over years, sometimes driven by microscopic plastic wear particles that provoke bone loss. This is the dominant late-failure mode.
• Infection. Bacteria can settle on the implant at surgery or reach it years later through the bloodstream. Implant infections almost never clear with antibiotics alone.
• Instability. If the ligaments stretch out or were never quite balanced, the components slide abnormally and the knee feels untrustworthy.
• Stiffness and arthrofibrosis. Aggressive scar tissue can lock up an otherwise sound implant.
• Periprosthetic fracture. A break in the bone around the implant, usually after a fall, often needing specialized fracture care or revision.

What Does the Workup for a Suspected Failure Involve?

The rule I practice by is simple: nobody should revise a knee replacement without being able to name exactly what is wrong with it.

A proper workup is methodical, and most of it is not surgery:

• History and exam. The pattern and timeline of your symptoms narrow the list faster than any single test.
• X-rays compared over time. Loosening usually declares itself as a widening line between implant and bone across serial films, so your old X-rays are genuinely valuable. Bring them, or have them sent.
• Blood tests. Simple inflammatory markers screen for infection.
• Joint aspiration. If infection is possible, drawing fluid for cultures and cell counts is the key test, and it should happen before anyone operates.
• Advanced imaging when needed. CT can assess component rotation and fixation when plain films do not explain the symptoms.

Two requests: bring your operative report, and your implant card if you have one. The exact make and sizes of your components determine whether parts can be exchanged or the whole construct must come out, and you are entitled to a copy of the report from your surgeon’s office.

And sometimes everything comes back normal. I treat the patient, not the X-ray; when a thorough workup is truly negative, the honest recommendation is usually surveillance rather than surgery on a knee we cannot explain.

What Should You Do Next If You Think Your Knee Replacement Is Failing?

Experience matters here: in a large Medicare study, patients whose knee replacement was performed by surgeons doing more than fifty per year had a lower risk of serious perioperative complications than patients of surgeons doing twelve or fewer [4].

Revision surgery is harder than first-time surgery; the volume-outcome data above is one proxy. The next steps:

• See a surgeon who does revision work regularly. Revision knee replacement is its own discipline, with its own implants and judgment calls; fellowship training in adult reconstruction is a reasonable thing to ask about directly. Our failed knee replacement page explains how we approach these evaluations.
• Gather your records first. Operative report, implant card, and prior X-rays improve the quality of the visit more than anything else.
• Treat a second opinion as routine. You do not need to go back to your original surgeon, and no good surgeon is offended by a second look at a possible failure. I cover the logistics in my guide to second opinions on knee replacement.
• Use a virtual review if travel is hard. A virtual second opinion based on your imaging and operative report can tell you whether an in-person revision consultation is worth the trip.
• Do not sit on red flags. Drainage, spreading redness, or fever is a same-day phone call.

One caution: no injection can re-fix a loose implant or sterilize an infected one. AAOS notes there is not yet strong evidence that stem cell treatments are effective; a failing knee replacement is a mechanical, sometimes infectious, problem that no syringe addresses [6].

Frequently Asked Questions About Failed Knee Replacements

Related Resources

These pages go deeper on each step:

• Failed Knee Replacement – How we evaluate one that is not working
• Revision Knee Replacement – What the operation involves
• Knee Replacement Pain After 2 Years – When late pain matters
• Second Opinions on Knee Replacement – How to get one
• Virtual Second Opinion – A records-based review
• Prosthetic Joint Infection – Why implant infections behave differently
• Total Knee Replacement – How the primary operation works

Revision Knee Evaluations in Long Beach, California

I am Dr. Leo Zalikha, fellowship-trained in adult reconstruction at Stanford, and complex revision hip and knee surgery is a core part of my Long Beach practice. I evaluate painful knee replacements no matter where the original surgery was performed; surgical cases are done at MemorialCare Long Beach Medical Center and the Surgery Center of Long Beach.

If your knee replacement is worrying you, the right first step is a structured evaluation, not a guess. If the workup shows a well-fixed, uninfected knee, I will say so plainly. If it shows a failing implant, you will leave knowing what is wrong, what your options are, and how urgent it is.

References

1. Evans JT, Walker RW, Evans JP, Blom AW, Sayers A, Whitehouse MR. How long does a knee replacement last? A systematic review and meta-analysis of case series and national registry reports with more than 15 years of follow-up. The Lancet. 2019;393(10172):655-663. PMID: 30782341. PubMed
2. Sharkey PF, Lichstein PM, Shen C, Tokarski AT, Parvizi J. Why are total knee arthroplasties failing today—has anything changed after 10 years? J Arthroplasty. 2014;29(9):1774-1778. PMID: 25007726. PubMed
3. Beswick AD, Wylde V, Gooberman-Hill R, Blom A, Dieppe P. What proportion of patients report long-term pain after total hip or knee replacement for osteoarthritis? A systematic review of prospective studies in unselected patients. BMJ Open. 2012;2(1):e000435. PMID: 22357571. PubMed
4. Katz JN, Barrett J, Mahomed NN, Baron JA, Wright RJ, Losina E. Association between hospital and surgeon procedure volume and the outcomes of total knee replacement. J Bone Joint Surg Am. 2004;86(9):1909-1916. PMID: 15342752. PubMed
5. American Academy of Orthopaedic Surgeons (OrthoInfo). Revision Total Knee Replacement. orthoinfo.aaos.org. AAOS
6. American Academy of Orthopaedic Surgeons (OrthoInfo). Use of Stem Cells in Orthopaedics. orthoinfo.aaos.org. AAOS

Medical Disclaimer: This content is provided for general educational purposes and does not constitute medical advice. Treatment recommendations vary based on individual diagnosis, imaging findings, and health history. Please consult Dr. Zalikha or a qualified orthopedic physician for an evaluation specific to your condition.