Is the Patella Removed in a Total Knee Replacement?

The patella is not removed during a total knee replacement. Your kneecap stays in place throughout the surgery and afterward. What surgeons debate, sometimes heatedly, is whether to resurface the underside of the patella by shaving away a thin layer of bone and cartilage and capping it with a small plastic button. That decision has been one of the longest-running controversies in joint replacement surgery, with practice varying dramatically depending on where in the world you happen to have the operation.

What Actually Happens to the Patella During Surgery

In a total knee replacement (also called total knee arthroplasty), the surgeon replaces the worn surfaces of three bones: the bottom end of the femur gets a metal cap, the top of the tibia gets a metal tray with a plastic spacer, and the patella either gets a small polyethylene disc cemented onto its back surface or is left as-is. The kneecap itself is moved to the side during the procedure to give the surgeon access to the joint, then returned to its normal position before closing. It is never taken out of the body.

The patella matters because it acts as a lever for your quadriceps muscles. Modeling work has shown that the kneecap increases the effective force your thigh muscles can generate when straightening the leg, and its presence shifts how that force is distributed between rotating the thigh and extending the lower leg.¹ Removing it entirely would weaken the knee’s extension mechanism, which is why no modern procedure takes it out.

Resurfacing Versus Leaving It Alone

The real question your surgeon faces is whether to resurface the back of the patella. In resurfacing, the surgeon shaves a few millimeters of bone and damaged cartilage from the undersurface of the kneecap, then cements a dome-shaped polyethylene button in its place. This creates a smooth, artificial gliding surface that matches the groove in the new femoral implant. In non-resurfacing, the surgeon smooths any rough spots on the natural patellar cartilage but leaves the bone intact and does not attach an implant.²

There is also a third approach: selective resurfacing, where the surgeon decides intraoperatively based on what the cartilage looks like. Algorithms for this decision consider factors like preoperative kneecap pain scores, X-ray findings, and the degree of cartilage wear seen during surgery.³ In patients with rheumatoid arthritis, resurfacing tends to be done routinely because residual diseased cartilage can fuel ongoing inflammation. For degenerative arthritis, resurfacing is typically reserved for cases with severe cartilage loss and exposed bone.⁴

What the Evidence Says About Outcomes

If you are hoping for a clean verdict on which approach produces better knees, the research is frustratingly mixed. A large meta-analysis of 22 studies found no significant difference in overall knee function scores between resurfaced and non-resurfaced patients. Resurfacing did lower the odds of anterior knee pain by about 40%, but there was no meaningful difference in pain scales, range of motion, or Oxford Knee Scores.⁵ A prospective comparative study, on the other hand, found statistically better clinical and functional scores in the resurfacing group at one year, along with a greater reduction in pain.⁶ And yet another prospective study using selective resurfacing found no association between resurfacing and either pain or function at twelve months.⁷

One area where the data tilts more clearly is reoperation rates. A systematic review and meta-analysis found that resurfacing cut the risk of revision surgery substantially, with a reoperation rate of about 1% compared to roughly 7% when the patella was not resurfaced.⁸ A separate meta-analysis confirmed a lower revision risk with resurfacing and also found fewer complications overall.⁹ The revision argument is a strong one because a second surgery on the knee is a much bigger deal than the few extra minutes it takes to resurface the patella during the original procedure.

Isokinetic testing, which measures the actual force muscles can produce during controlled movement, found that patients whose patella was resurfaced had significantly stronger knee extension at six months and one year after surgery. Flexion strength was about the same in both groups.¹⁰

Why Practice Varies So Much Around the World

The resurfacing debate is not just academic — it translates into wildly different surgical practice depending on geography. A global analysis of national joint registries from 2004 to 2022 found that average resurfacing rates ranged from 4% in Sweden to 94% in the United States.¹¹ That means an American patient with the same knee condition as a Swedish patient is more than twenty times more likely to have the patella resurfaced. Australia saw a 40% increase in resurfacing rates over that period, while Scandinavian countries kept theirs extremely low.

Part of the explanation is that different national registries show different outcomes. In the US, Canada, Australia, and Switzerland, registry data showed a lower risk of revision when the patella was resurfaced. In Sweden, the opposite appeared to be true — resurfacing was associated with a higher revision risk.¹¹ Why the registries disagree is not entirely clear. It could reflect different implant designs, different patient populations, different surgical techniques, or a combination of all three. Earlier registry data from 2004 to 2014 showed the same pattern: rates ranged from 4% in Norway to 82% in the US, and outside the US, only about a third of primary knee replacements included patellar resurfacing.¹²

Anterior Knee Pain After Replacement

Pain at the front of the knee is one of the most common complaints after a total knee replacement, and it can happen regardless of whether the patella was resurfaced.¹³ This is worth understanding because many patients assume patellar resurfacing is what prevents that pain. The causes are more complex, involving both mechanical factors (how the implant components are aligned, soft tissue tension) and functional ones (muscle weakness, scar tissue formation, changes in how the kneecap tracks).

A large cohort study of patients who did not have their patella resurfaced found that about 16% reported anterior knee pain at follow-up. Those patients had substantially worse functional scores and satisfaction compared to pain-free patients. Interestingly, the severity of cartilage damage on the patella before surgery did not predict who would develop pain afterward.¹⁴ A separate study confirmed that patients with mild, moderate, and severe patellar cartilage degeneration all had similar pain outcomes when the patella was left unresurfaced.¹⁵ Obesity did appear to matter: patients with a BMI over 40 had significantly higher rates of anterior knee pain.¹⁴

Patellar Denervation as a Middle Ground

Some surgeons who choose not to resurface the patella use electrocautery to burn the tiny nerve endings around the rim of the kneecap, a technique called patellar denervation. The idea is to reduce pain signals from the patella without adding an implant. A meta-analysis of randomized controlled trials found that denervation reduced anterior knee pain both within the first year and beyond, without increasing complications or reoperations.¹⁶ Another meta-analysis similarly found improved pain relief, lower opioid use, and better range of motion in the denervation group.¹⁷

The picture is not perfectly tidy, though. A prospective comparative study found that circumferential denervation did not provide additional benefit in reducing anterior knee pain or improving function compared to doing nothing at all.¹⁸ So while the technique is popular and supported by pooled trial data, individual studies have reached conflicting conclusions. It remains a reasonable option but not a guaranteed fix.

If You Need Resurfacing Later

When a patient who did not have their patella resurfaced develops persistent anterior knee pain, one option is secondary patellar resurfacing — going back in to add the plastic button after the fact. This does help many people, but the results tend to be less predictable than having resurfacing done during the original surgery. A systematic review and meta-analysis of 604 knees found that about 53% of patients achieved improvement in patient-reported outcomes, and roughly 59% were satisfied with the procedure. The complication rate was low (around 2%), but about 10% ended up needing another revision surgery.¹⁹

An earlier study found that while clinical scores improved significantly after secondary resurfacing, the final results were still worse than what would be expected from primary resurfacing. The longer patients waited between the original knee replacement and the secondary resurfacing, the worse the outcome tended to be — suggesting that if secondary resurfacing is going to happen, it should happen sooner rather than later.²⁰ In a separate cohort, scores improved but about 14% of patients needed yet another revision, and some continued to have pain without a clear cause.²¹ This is one of the arguments resurfacing advocates lean on: if there is a reasonable chance the patient will need it eventually, it is better to do it up front than to subject them to a second operation with inferior outcomes.

Complications Specific to the Patella

Whether resurfaced or not, the patella can be a source of trouble after total knee replacement. Recognized complications include patellar fracture, loosening of the patellar component, instability (the kneecap dislocating or tilting), wear of the polyethylene button, and osteonecrosis (loss of blood supply to the bone).²² Earlier implant designs with metal-backed patellar components or cementless fixation had significantly higher complication rates. The loosening rate with cementless fixation was over 13%. Modern practice overwhelmingly uses cemented, all-polyethylene domed patellar components, which carry the lowest complication rates.²³

How the kneecap tracks — the path it follows in the groove of the femoral implant as you bend and straighten your knee — depends heavily on the rotational alignment of the femoral and tibial components, and also on the shape of the patella itself.²⁴ Computer modeling has shown that even a 5-degree internal rotation of the femoral component can tilt and rotate the patella laterally by about the same amount throughout the full range of knee bending.²⁵ This is why surgical precision in placing the implant matters just as much as the resurfacing decision itself.

Patellar Clunk Syndrome

One complication unique to certain implant designs is patellar clunk syndrome, where patients feel (and sometimes hear) a painful clunking sensation as the knee moves from a bent position to straight. It is caused by a nodule of fibrous tissue that forms at the top of the patella where it meets the quadriceps tendon. As the knee extends, this nodule catches on the edge of the femoral implant and then snaps free.²⁶ Posterior-stabilized implant designs, which have a raised post-and-cam mechanism in the center of the knee, are more prone to this than cruciate-retaining designs. Newer implant designs have reduced the size of the intercondylar box where this catching occurs, which has brought the incidence down.²⁷ When the syndrome does occur, the treatment is typically arthroscopic surgery to remove the fibrous nodule.

The Cost Question

The financial dimension is more nuanced than you might expect. A cost-effectiveness analysis concluded that over 14 years, resurfacing appeared to be the more cost-effective option, primarily because unresurfaced knees had higher revision rates and revision surgery is expensive.²⁸ But a large randomized controlled trial in the UK tracked costs over 20 years and found that while the resurfacing group had higher upfront implant and hospitalization costs, the non-resurfacing group accumulated more readmission costs over time. The net result was that 20-year total healthcare costs were virtually identical between the two groups.²⁹ So the cost argument, often raised in either direction, largely washes out over a long enough time horizon.

How Alignment Strategy Affects the Kneecap

Beyond the resurfacing question, how the surgeon aligns the implant components has its own effect on patellar behavior. Two philosophies dominate: mechanical alignment, which aims to position the implant along the mechanical axis of the leg, and kinematic alignment, which tries to reproduce the patient’s natural joint geometry. Research has found that these different strategies significantly influence how the kneecap tracks after surgery.³⁰ This is a growing area of research and an example of how the patellar question in knee replacement extends well beyond the binary of “resurface or not.” The kneecap’s long-term comfort and function depend on a web of decisions made during surgery: implant design, alignment philosophy, component rotation, and soft tissue balancing all play interconnected roles.