What drives the choice between partial and total knee replacement

For many patients approaching knee replacement, the first question is whether they might avoid the full operation altogether. The honest answer depends entirely on how far the arthritis has spread — not on how severe the pain feels, and not on patient or surgeon preference alone.

The knee has three compartments: medial (inner side), lateral (outer side), and patellofemoral (the kneecap joint). Arthritis rarely erodes all three equally. Partial knee replacement (UKR) resurfaces the single compartment that is damaged, leaving healthy bone, cartilage, and ligaments untouched. Total knee replacement (TKR) resurfaces all three compartments simultaneously. They address different disease patterns — not different grades of the same one — which is why they are not interchangeable options for the same patient.

Neither operation carries a universal advantage over the other. Each is the right choice when matched to the correct anatomical picture. NICE QS206 (2022) enshrines this principle in clinical guidance, requiring that adults with isolated medial compartmental osteoarthritis who are proceeding to knee replacement be offered a genuine, informed choice between partial and total replacement. The rest of this article explains who each operation suits — and why that fit matters.

Who qualifies for a partial knee replacement

Candidacy for partial knee replacement rests on a defined anatomical checklist — each criterion exists for a mechanical reason, not as an administrative filter.

Single-compartment arthritis confirmed on weight-bearing X-ray. The medial (inner) compartment is the most commonly affected. The X-ray must be taken standing, under load: cartilage that looks present on a lying-down image can disappear entirely when the joint bears weight. UKR is designed to replace one compartment; if two or three are involved, the partial procedure simply cannot address the source of pain.

An intact anterior cruciate ligament (ACL). The ACL shares the load in a partial replacement, providing stability that the implant itself does not supply. In mobile-bearing designs, an absent ACL allows the bearing to shift abnormally — leading to dislocation and early failure. This requirement is non-negotiable for mobile-bearing UKR; some specialist centres consider combined ACL reconstruction alongside UKR on a case-by-case basis, but this remains selective practice.

Angular deformity within correctable limits. Fixed varus deformity exceeding 10°, or fixed valgus exceeding 5°, indicates structural change that a single-compartment implant cannot adequately correct. Beyond these thresholds, total knee replacement is the appropriate choice.

Adequate range of motion. A minimum of 110° of knee flexion and no fixed flexion contracture (ideally less than 15°) are required — partly to allow the operation itself, partly because the knee needs functional arc to recover well.

No inflammatory arthritis. Rheumatoid, psoriatic, or other inflammatory conditions affect the joint lining systemically; all three compartments are at risk, which makes whole-joint replacement the standard approach.

Classically, only around 6% of knee arthroplasty candidates meet every criterion without exception. Robotic-assisted surgery — by improving implant positioning accuracy to within a fraction of a millimetre — has expanded realistic eligibility to approximately 25% of patients with knee arthritis, though the eligibility criteria themselves remain unchanged. Mild patellofemoral wear is not an automatic bar to partial replacement; this is assessed case by case at consultation. Patients who do not qualify for UKR are not at a disadvantage: total knee replacement, when correctly indicated, is a highly durable solution in its own right.

When total knee replacement is the right path

Total knee replacement is not the larger version of a partial replacement — it is a categorically different procedure designed for a categorically different disease pattern. When arthritis has spread beyond a single compartment, or when the underlying condition affects the joint as a whole, resurfacing all three compartments is both necessary and appropriate.

The clearest indication is multi-compartmental osteoarthritis — damage confirmed across two or more of the medial, lateral, and patellofemoral compartments. Replacing only part of a joint that is widely affected would leave substantial pain unaddressed. Similarly, inflammatory arthritis — including rheumatoid and psoriatic arthritis — is a systemic process that alters the joint lining as well as the cartilage; the whole joint requires resurfacing, and partial replacement is contraindicated in these conditions.

Significant patellofemoral arthritis tips the balance firmly towards TKR. Where the kneecap compartment is heavily involved, a partial medial replacement cannot relieve that component of pain. Structural deformity that lies beyond the correctable range for a partial implant is also appropriately addressed by TKR, which can realign the whole joint surface simultaneously. Prior procedures that have substantially altered the knee's internal anatomy — such as extensive meniscectomy accompanied by disease progression into other compartments — likewise point towards total replacement.

During TKR, the anterior cruciate ligament is removed as part of the resurfacing process; the joint's post-operative mechanics rely on implant geometry rather than the native ligamentous architecture that a partial replacement preserves.

How recovery and day-to-day function compare

Recovery after the two procedures follows a notably different curve — and for well-selected patients, that difference often carries as much weight as the surgical decision itself.

Partial replacement (UKR). The smaller incision and limited soft-tissue dissection mean most patients are walking on the same day or the day following surgery, with hospital stays typically shorter than for total replacement. Because the cruciate ligaments remain in place, the operated knee moves in a pattern closer to its natural arc — patients commonly describe the result as feeling more like their own joint than a resurfaced one. Per-procedure blood loss is lower, and the risk of periprosthetic infection is reduced compared with the larger operation.

Total knee replacement (TKR). All three compartment surfaces are resurfaced simultaneously, which is a more extensive procedure by design. Recovery reflects that scope: structured physiotherapy over roughly three to six months provides the standard framework for regaining strength, range of motion, and confidence on the knee. Most patients reach reliable daily independence within that window, though pace varies considerably between individuals.

The trajectory tends to diverge early. UKR patients typically reach walking independence and a return to light daily activities sooner; TKR recovery consolidates more gradually across the physiotherapy course. Age, pre-operative muscle strength, BMI, and consistency with rehabilitation all shape individual progress in both groups — which is why realistic expectations are best set at a specialist consultation rather than extrapolated from population averages. The kinematic advantage of UKR — that more natural-feeling joint — applies specifically to patients whose anatomy and ligament integrity make them genuine candidates for the partial procedure.

What the long-term survival data shows

Survival figures for knee implants tend to appear in one of two forms — specialist centre data and national registry averages — and the gap between them is meaningful enough that patients benefit from understanding both.

At the high-volume end, a 2016 Bone & Joint analysis of 1,084 Oxford UKR procedures recorded 10-year cumulative survival of 93.2% (95% CI 86.1–100); the original designer series report figures of 96–98% over the same period. National registry data tells a different story: Foissey et al. (2024) found 10-year survival of only 84.1%, with the discrepancy attributed to broader case selection and lower average surgical volume rather than any intrinsic problem with the implant itself. New Zealand Registry data corroborates the volume relationship directly — surgeons performing more than ten UKAs a year recorded a 6% revision rate, with higher figures among lower-volume surgeons. This is a reasonable question to raise at a pre-operative consultation: how many of these procedures does the surgical team perform each year?

For implants that reach ten years intact, the outlook beyond that point is encouraging. Oxford series data suggest that second-decade attrition is low, making 20-year survival a realistic prospect for unrevised implants.

When revision becomes necessary, conversion to total knee replacement is the standard pathway. Because UKR removes far less bone than TKR, native bone stock is largely preserved — an advantage that makes the conversion technically more straightforward than revising a failed primary TKR. Published series indicate satisfactory mid-to-long-term outcomes after such revision, though results are generally slightly inferior to a primary TKR performed in equivalent circumstances.

Reaching a decision and what an assessment involves

The decision between partial and total replacement should not feel predetermined. Under NICE QS206 (2022), adults with isolated medial compartmental osteoarthritis are entitled to have both options presented as genuine choices — with documented discussion of risks, benefits, and uncertainties before any recommendation is made. That is not a courtesy; it is the expected standard.

A thorough pre-operative assessment draws on several sources. Weight-bearing X-rays of all three compartments establish the degree of joint-space narrowing and overall limb alignment. Clinical examination tests ligament integrity — the ACL in particular — and quantifies any fixed deformity. Inflammatory markers and BMI review help exclude conditions that alter the surgical path entirely. Where cartilage status or cruciate integrity remains uncertain on plain imaging, MRI provides a more granular picture; AI-assisted analysis can refine compartmental assessment further where that capability is available.

Practice varies considerably from surgeon to surgeon. Beard et al. (2012) documented wide divergence in decision-making between surgeons and centres, driven by training background, case mix, and differing thresholds for each procedure. A consultant who carries out both UKR and TKR at meaningful volume is better placed to give guidance free of single-technique bias — and patients are entitled to confirm that both options will be formally considered rather than one assumed from the outset.

Lincolnshire Knee, part of the MSK Doctors group, offers consultant-led knee assessments at Sleaford NG34 and Grantham NG31, with no GP referral required — a practical starting point for patients seeking that conversation with a team experienced across both procedures. Appointments can be booked at lincolnshireknee.co.uk.