Partial or total knee replacement: how the two operations compare

Written by Dr Sean Leo, a consultant orthopaedic surgeon whose clinical focus is the knee, sports injuries and the lower limb.

Last updated: 9 October 2026.

A partial knee replacement replaces only the damaged compartment of the knee ^[4]. In one randomised trial of 528 patients, the two operations showed no statistically significant difference in knee scores at five years ^[1][2] and similar clinical outcomes at 10 years ^[8].

A partial knee replacement replaces only the damaged compartment of the knee ^[4]. Where the arthritis sits, the ligaments and the alignment of the leg are among the factors described in one review for whether it suits a patient ^[4]. In one randomised trial of 528 patients, the two operations showed no statistically significant difference in knee scores at five years ^[1][2] and similar clinical outcomes at 10 years ^[8].

What is the difference between a partial and a total knee replacement?

A partial knee replacement, also called unicompartmental knee arthroplasty (UKA), replaces only the damaged compartment of the knee ^[4]. The operation it is most often compared with is total knee replacement (total knee arthroplasty, TKA).

The knee has three main compartments. The medial compartment is the inner side. The lateral compartment is the outer side. The patellofemoral compartment is where the kneecap sits.

• The main trial enrolled people whose arthritis was confined to the inner (medial) compartment ^[1].
• Candidacy depends on the other compartments being healthy ^[4].

What’s covered on this webpage

• What is the difference between a partial and a total knee replacement?
• Which knees may suit a partial replacement?
• What did the main trial of the two operations find?
• How do revision rates compare?
• What do the trial and reviews report about hospital stay, pain and function?
• What about returning to sport and activity?
• When is urgent help needed?
• What should I do next?

Which knees may suit a partial replacement?

One review lists unicondylar osteoarthritis, which means arthritis in one part of the knee, among the current indications for a partial replacement ^[4].

• Criteria described for a partial replacement include a functionally normal medial collateral ligament (MCL, the ligament on the inner side of the knee), intact full-thickness cartilage in the lateral compartment, and a patellofemoral joint with no lateral grooving and bone loss (no grooves or loss of bone on the outer side of that joint) ^[4].
• A knee with primary traumatic damage to the anterior cruciate ligament (ACL, one of the ligaments inside the knee) and secondary arthritis at the back of the inner compartment should not be considered a candidate ^[4]. A knee with primary anteromedial arthritis and secondary degenerative ACL deficiency is now accepted as a possible candidate ^[4].
• A frontal deformity (a bow-legged or knock-kneed alignment) or a flexion contracture (a knee that cannot fully straighten) of more than 15 degrees has been thought to make a knee less ideal for a partial replacement ^[4].

More on how selection has changed

Under the older Kozinn and Scott selection criteria (an older set of rules for choosing which patients should have a partial replacement), only about 5% of knees would be regarded as suitable for a partial replacement if the criteria were strictly followed ^[4].

What did the main trial of the two operations find?

The trial, named TOPKAT, randomly assigned 528 patients with medial compartment knee osteoarthritis to a partial or a total knee replacement ^[1][2]. It was run at 27 sites across the UK ^[1].

• At five years there was no statistically significant difference in Oxford Knee Score between the groups ^[1][2]. The Oxford Knee Score is a questionnaire about knee pain and function.
• The 5-year Lancet report concluded that both operations are effective and offer similar clinical outcomes ^[1].
• At 10 years, the trial reported similar clinical outcomes, reoperation rates and revision rates for the two operations ^[8].
• Over the five years, partial replacement was estimated to give 0.24 more quality-adjusted life-years (QALYs, a measure that combines length and quality of life) and to cost £910 less than total replacement, in UK 2017 prices ^[1]. The 95% confidence interval (a range showing how precise the estimate is) was 0.046 to 0.434 for the QALYs and -£1,503 to -£317 for the cost difference ^[1]. The cost analysis was not designed to account for the possible cost of later revisions ^[1].
• Overall satisfaction at five years did not differ between the groups ^[1]. More patients in the partial group said their knee was better than before the operation and that they would have the operation again ^[1]. Patients could not be blinded to which operation they had, so those answers could have been influenced by that ^[2].

The trial enrolled selected patients and experienced surgeons, so its results may not apply to everyone ^[1].

More on how the trial was run

Surgeons in the trial had to have appropriate training and to have performed their operation at least ten times in the previous year ^[1]. The trial authors list the lack of masking, some missing health-care resource and EQ-5D (quality-of-life questionnaire) data (12% across all time periods) and some non-compliance with the allocated treatment as weaknesses ^[1].

How do revision rates compare?

A revision is a further operation to replace the original implant with a new device, usually because of device failure, infection or persistent pain ^[1]. A re-operation is any further operation, which can be a revision or another kind of surgery ^[1]. The trial and the wider data give a mixed picture.

• At five years, the trial found no difference in the number of re-operations and revisions between the groups, and its authors note that it had low power ^[1].
• At 10 years, the trial’s abstract (the short summary of the paper) reports similar reoperation rates and revision rates for the two operations ^[8].
• The same authors note that nearly all global registries (records kept of patients who have had joint replacements), which hold non-randomised observational data, report a substantially larger number of revisions after partial than after total replacement ^[1].
• A 2019 systematic review (a study that finds and combines the results of many earlier studies) in the BMJ found that revision rates at five years remained higher after partial replacement in all three study groups it analysed (randomised trials, national joint registries and database studies, and cohort studies), and that the risk of revision surgery was lower for total replacement ^[3]. It also noted that results varied depending on the level of data, and that findings were not always significant ^[3].
• One review reports that the higher failure and revision rates of partial replacement are mainly seen in the first five years, and are often due to incorrect indications and/or surgical errors ^[4].
• Arthritis can also progress in the parts of the knee that were not replaced, which is one of the recognised causes of a partial replacement failing ^[9].
• When a partial replacement is revised to a total replacement, the operation is generally simpler than revising a total replacement, and can be done with primary total knee components in most cases ^[7]. It is still a second operation, with its own recovery.

What do the trial and reviews report about hospital stay, pain and function?

In this trial, the operation time was similar in both groups, and the hospital stay was shorter after partial replacement ^[1].

• The BMJ review found significantly shorter hospital stays after partial than after total replacement ^[3].
• It found no significant difference in pain on patient-reported outcome measures, which are questionnaires completed by patients ^[3].
• In both groups of non-trial studies it found statistically significantly better functional scores after partial replacement, but the differences were small (mean difference -0.58, 95% confidence interval -0.88 to -0.27, and -0.32, 95% confidence interval -0.48 to -0.15, respectively) ^[3].
• These two groups were not randomised, so the patients having each operation may have differed ^[3].

What about returning to sport and activity?

A 2026 systematic review found that the overall rate of return to physical activity or sport was high after knee replacement, and it rated the certainty of the evidence as very low ^[6]. The review’s title and conclusions place this high rate of return particularly with low-impact activities ^[6].

• The same review found that the likelihood of returning to physical activity or sport was significantly lower after total than after partial replacement ^[6].
• Participation in high-impact sports often declines, largely due to psychological factors that may limit participation beyond actual physical capability ^[6].
• A 2016 review reported that sports participation seems more likely after partial than after total replacement ^[5]. Its authors note that the quality of the included studies was limited, mainly because confounding factors were inadequately taken into account in most studies ^[5].

When is urgent help needed?

The list below covers the signs that should not wait for a routine appointment, including signs after knee surgery.

What should I do next?

For anyone weighing a partial against a total knee replacement, an assessment is a sensible next step. Persistent knee pain or reduced function that limits daily life is a reason to arrange one. The assessment brings together the patient’s story, an examination and any tests that help, so that the patient and the surgeon can decide together. It looks at the structure involved, the patient as a whole person and the goal they want to return to. Where both operations are suitable, the choice is made together with the surgeon, because each carries its own trade-offs. An appointment supports an informed decision. It is not a commitment to surgery.

• Read about partial knee replacement surgery and total knee replacement surgery.
• Arrange a knee assessment when you are ready.
• Browse the Knee Hub for other knee problems, or more articles.
• Read more about Dr Sean Leo, who is named as the author of this page.

References

1. Beard DJ, Davies LJ, Cook JA, et al. The clinical and cost-effectiveness of total versus partial knee replacement in patients with medial compartment osteoarthritis (TOPKAT): 5-year outcomes of a randomised controlled trial. Lancet. 2019;394:746-756. doi:10.1016/S0140-6736(19)31281-4. PMID:31326135.
2. Beard DJ, Davies LJ, Cook JA, et al. Total versus partial knee replacement in patients with medial compartment knee osteoarthritis: the TOPKAT RCT. Health Technol Assess. 2020;24(20):1-98. doi:10.3310/hta24200. PMID:32369436.
3. Wilson HA, et al. Patient relevant outcomes of unicompartmental versus total knee replacement: systematic review and meta-analysis. BMJ. 2019;364:l352. doi:10.1136/bmj.l352. PMID:30792179.
4. Vasso M, Antoniadis A, Helmy N. Update on unicompartmental knee arthroplasty: current indications and failure modes. EFORT Open Rev. 2018;3(8):442-448. doi:10.1302/2058-5241.3.170060. PMID:30237902.
5. Witjes S, et al. Return to sports and physical activity after total and unicondylar knee arthroplasty: systematic review and meta-analysis. Sports Med. 2016;46(2):269-292. doi:10.1007/s40279-015-0421-9. PMID:26744336.
6. Braz R, Gomes E, Valente C, et al. High rate of return to low-impact physical activity or sports after total and unicompartmental knee arthroplasty: A systematic review with meta-analysis. Knee Surg Sports Traumatol Arthrosc. 2026;34(9):3415-3428. doi:10.1002/ksa.70267. PMID:41527885.
7. Lombardi AV Jr, et al. Revision of unicompartmental knee arthroplasty to total knee arthroplasty: is it as good as a primary result? J Arthroplasty. 2018;33(7S):S105-S108. doi:10.1016/j.arth.2018.03.023. PMID:29691172.
8. Beard DJ, Davies LJ, Cook JA, et al. Assessing clinical and cost effectiveness of total versus partial knee replacement (TOPKAT): 10-year follow-up of a multicentre, randomised controlled trial. Lancet Rheumatol. 2026;8(2):e116-e126. doi:10.1016/S2665-9913(25)00250-4. PMID:41270774.
9. Vasso M, Corona K, D’Apolito R, et al. Unicompartmental knee arthroplasty: modes of failure and conversion to total knee arthroplasty. Joints. 2017;5(1):44-50. doi:10.1055/s-0037-1601414. PMID:29114630.