Abstract
Aims
Whether medial unicompartmental knee arthroplasty (mUKA) or total knee arthroplasty (TKA) are more suitable for patients with anteromedial osteoarthritis (AMOA) remains debated. We aimed to compare the development over time in patient-reported outcome measures (PROMs) between patients receiving mUKA and TKA.
Methods
We conducted a matched cohort study of patients receiving TKA or mUKA between March 2018 and February 2020. TKA patients were evaluated based on preoperative radiographs and excluded if not deemed eligible for mUKA. The PROMs (Oxford Knee Score (OKS), Forgotten Joint Score (FJS), and Activity and Participation Questionnaire (APQ)) were completed preoperatively, and at three, six, and 12 months postoperatively. Patients were propensity score matched in a variable 1:2 ratio using BMI, sex, age, and preoperative PROM scores. Area under the curve (AUC) was calculated using the trapezium rule to quantify the change from preoperative PROM scores to scores at three, six, and 12 months postoperatively. AUC differences were analyzed using linear regression.
Results
A total of 618 patients (242 mUKA, 376 TKA) were included. The AUC was significantly lower for TKA patients compared with mUKA patients across all three PROM scores (OKS: ∆AUC of -19% (95% CI -27% to -9 %); FJS: ∆AUC of -23% (95% CI -32% to -14%); APQ -22% (95% CI -32% to -12%). Median PROM scores at three months were 35 and 30 for OKS, 50 and 43 for FJS, 44 and 31 for APQ, for the mUKA and TKA groups, respectively. At six months, the median PROM scores increased to 39 and 35 for OKS, 62 and 50 for FJS, and 56 and 41 for APQ for the mUKA and TKA groups, respectively.
Conclusion
The AUC for PROM scores during the first year was 19% to 23% lower for TKA compared with mUKA. This contrast in development trajectories is present especially in the early recovery three to six months after surgery.
Cite this article: Bone Jt Open 2025;6(12):1588–1597.
Keywords: Knee, Osteoarthritis, Antero-medial osteoarthritis, Arthroplasty, Knee replacement, Medial unicompartmental knee arthroplasty, Total knee arthroplasty, Patient reported outcome measures, medial unicompartmental knee arthroplasty, patient-reported outcome measures (PROMs), total knee arthroplasty (TKA), Forgotten Joint Score (FJS), Oxford Knee Score, BMI, propensity score matching, preoperative radiographs, linear regression, anteromedial osteoarthritis
Introduction
Total knee arthroplasty (TKA) and medial unicompartmental knee arthroplasty (mUKA) are established surgical treatments for end-stage anteromedial osteoarthritis (AMOA) of the knee.1,2 Historically, factors like age, weight, and activity level have limited the use of mUKA. However, some studies have proposed evidence-based and validated criteria without these factors as contraindications.3,4 These studies indicate that up to 50% of knee arthroplasty patients may be eligible for mUKA.
mUKA has been associated with higher revision rates compared with TKA.1 This may, in part, reflect a lower threshold for revision compared with TKA.1,5-7 Revision rates for mUKA have declined over the last 20 years,8 and registry data suggest that mUKAs performed by surgeons with high usage (> 20%) have substantially improved implant survivorship.9-11 Some studies suggest mUKA offers advantages over TKA for eligible patients. mUKA patients have been shown to have shorter lengths of stay, to be more cost-efficient,12-16 and have half the risk of periprosthetic joint infections compared with TKA.14,17
Patient-reported outcome measures (PROMs) can provide additional support for treatment choice in value-based healthcare.1,18 While existing research, utilizing PROMs, has mainly concentrated on long-term functional outcomes (one to five years), predominantly finding comparable18-22 or clinically insignificant differences,18,23 some studies have found indications of a potential for better or faster early postoperative recovery for mUKA patients compared with TKA patients.15,24,25 These potential differences in early recovery could have meaningful clinical implications, for both patients and healthcare systems. From a patient’s perspective, improved early recovery might influence treatment preferences, despite similar long-term outcomes. Therefore, understanding the potential differences in early recovery following mUKA and TKA could be important to inform treatment choices through shared-decision-making.
In light of this, the objective was to compare the trajectories of development in PROM scores one year postoperatively between mUKA and TKA patients.
Methods
All patients undergoing primary TKA or mUKA between 12 March 2018 and 27 February 2020 were included. Previous surgery to the same knee was not an exclusion criterion; however, TKA patients with previous anterior cruciate ligament (ACL) reconstruction were excluded to increase comparability, as one of the criteria for mUKA is functionally intact ligaments.3 Surgeries were performed at a single centre, university hospital setup (Copenhagen University Hospital Hvidovre, Denmark). Patients with no available PROM data or simultaneous bilateral surgeries were excluded. TKA patients with missing preoperative radiographs were excluded. For patients with two surgeries in the inclusion period, only the second was included. TKA patients deemed ineligible for mUKA based on preoperative radiographs, were excluded to increase group comparability. Radiographs were evaluated twice by the corresponding author (ALEC), using the radiological decision aid by Hamilton et al.3 In cases of uncertainty, a consensus decision was reached by the co-authors. Overall, 57.8% of the primary TKAs were deemed radiologically eligible for mUKA. Patient characteristics (sex, age, height, weight) and the American Society of Anesthesiologists (ASA) Physical Status Classification Score26 were registered at preoperative physical examination. BMI was calculated from height and weight.
Surgical technique
Unicompartmental cases were performed using the uncemented medial Oxford partial knee (Zimmer Biomet, USA), with microplasty instrumentation. Indications for mUKA were bone-on-bone AMOA, functionally intact ligaments, a preserved lateral compartment, and no severe lateral facet patellofemoral osteoarthritis. TKA procedures involved implantation of the Nexgen CR, Persona CR, or MC knee systems (Zimmer Biomet), using standard instruments and a measured resection technique with ligament releases as necessary. TKAs were cemented on both tibia and femur. No computer navigation or robotic assistance was used. All TKAs were inserted with the aim to achieve neutral mechanical alignment (proximal tibia cut perpendicular to the long axis and a 5° to 6° valgus cut of the distal femur relative to the long axis). Patellar resurfacing was routinely performed. Surgeries were performed by 11 surgeons, all orthopaedic specialists with subspecialty training in knee arthroplasty: five surgeons performed both mUKA and TKA; while six performed only TKA.
Both groups followed the same standardized postoperative follow-up and rehabilitation protocol, including early mobilization with full weightbearing and access to the same public outpatient rehabilitation services.
Outcomes
Patients completed the following PROMs: Oxford Knee Score (OKS),27,28 Forgotten Joint Score (FJS),29 and the OKS Activity and Participation Questionnaire (APQ).27 These have demonstrated acceptable levels of validity, reliability, and responsiveness.27,29-32 Questionnaires were completed after preoperative physical examination, as well as at three, six, and 12 months postoperatively using Procordo Software (Procordo, Denmark).33 Follow-up questionnaires were sent twice by email. If there was no reply or an incomplete reply, a letter containing the questionnaire was sent. Patients without email received the questionnaires by letter once.
Multiple imputation
Missing values were assumed to be missing at random and handled with the multiple imputation method prior to propensity score matching, using predictive mean matching. Imputation was performed using the mice package in R, creating five imputed datasets.34 Missing values for BMI and PROM scores were imputed using surgery type, sex, age, BMI, preoperative PROM score, and three, six, and 12-month PROM scores. A random seed was used for the imputation, and the results presented are based on this original random seed. The imputation was tested using ten different random seeds to ensure consistency. Patients who underwent revision surgery within one year were included in the analysis with imputed data for missing PROM scores to ensure that the results accurately reflected the full range of outcomes.
Missingness before imputation and additional baseline characteristics for responder subgroups, and revised patients, are available in the Supplementary Material.
Propensity score matching
Patients were propensity score matched in a variable ratio 1:2 (mUKA:TKA) using BMI, sex, age, and preoperative PROM scores (Table I). Propensity scores were estimated using logistic regression. Nearest-neighbour matching without arthroplasty was performed using the MatchIt package in R, with a caliper of 0.2 on the logit scale and TKA as the reference group (MatchIt v. 4.5.3, CRAN; R Foundation for Statistical Computing, Austria). Standardized mean differences (SMDs) of ≤ 0.1 were considered sufficient balance between the groups.35
Table I.
Baseline characteristics.
| Variable | Unmatched | Variable 1:2 matched | ||||
|---|---|---|---|---|---|---|
| mUKA (n = 244) | TKA (n = 382) | SMD | mUKA (n = 242) | TKA (n = 376) | SMD | |
| Male sex, n (%) | 103 (42.2) | 165 (43.2) | 0.020 | 103 (42.6) | 160 (42.6) | < 0.001 |
| Mean age, yrs (SD) | 67.5 (9.3) | 67.6 (9.8) | 0.015 | 67.6 (9.3) | 67.7 (9.8) | 0.013 |
| Median BMI, kg/m2 (IQR) | 28.6 (25.3 to 32.8) | 30.1 (27.6 to 33.2) | 0.179 | 28.7 (25.4 to 33.0) | 30.0 (27.6 to 33.1) | 0.157 |
| ASA grade, n (%) | 0.110 | 0.123 | ||||
| 1 | 15 (6.1) | 33 (8.6) | 14 (6.8) | 33 (8.8) | ||
| 2 | 187 (76.6) | 277(72.5) | 186 (76.9) | 274 (72.9) | ||
| 3 | 42 (17.2) | 72 (18.8) | 42 (17.4) | 69 (18.4) | ||
| Mean OKS score (SD) | 23.2 (7.3) | 22.46 (6.9) | 0.105 | 23.1 (7.3) | 22.5 (6.8) | 0.090 |
| Median FJS score (IQR) | 14.6 (6.08 to 25) | 13.6 (4.17 to 25) | 0.053 | 14.6 (6.3 to 25.0) | 13.6 (4.2 to 25.0) | 0.050 |
| Median APQ score (IQR) | 6.3 (3.1 to 18.8) | 9.4 (3.1 to 18.8) | 0.048 | 6.3 (3.1 to 18.8) | 9.38 (3.1 to 18.8) | 0.022 |
Baseline patient characteristics, unmatched and propensity score matched in a variable ratio 1:2 using BMI, sex, age, and preoperative patient-reported outcome measure scores.
FJS 0 to 100, OKS 0 to 48, APQ 0 to 100.
APQ, activity and participation questionnaire; ASA, American Society of Anesthesiologists; FJS, Forgotten Joint Score; mUKA, medial unicompartmental knee arthroplasty; OKS, Oxford Knee Score; SMD, standardized mean difference; TKA, total knee arthroplasty.
AUC
Changes in PROM scores from baseline were calculated using the trapezium rule, as a measure of improvement over time.36,37 AUC, with time on the x-axis and PROM scores on the y-axis, represents the average change in PROMs over the follow-up period.38 The between-group differences in AUC were compared using linear regression, unadjusted and adjusted for preoperative scores, ASA, sex, and age. Estimates for the difference in AUC were expressed as a percentage, illustrating the proportional difference in AUC for TKA patients compared with mUKA patients.
Probability of negative PROM change
The proportion of patients with a negative treatment effect on PROM scores was calculated using Fisher’s exact test, stratified by surgery type. Negative treatment effect was defined as lower postoperative PROM scores compared with preoperative scores, or a negative AUC. The odds ratio for a TKA patient experiencing a negative change in PROM score was calculated relative to mUKA.
Statistical analysis
Baseline characteristics were assessed using histograms and quantile-quantile plots. Categorical data are displayed as n (%). Continuous data are summarized as mean (SD) or median (IQR). Statistical significance was defined as p-values < 0.05. Analyses were performed using R v. 4.1.0 (R Foundation for Statistical Computing).
Results
Data from 1,117 patients (304 mUKA, 813 TKA) were available. Figure 1 shows the inclusion and exclusion process. Ultimately, 618 patients (242 mUKA, 376 TKAs) were included in the analysis.
Fig. 1.
Flowchart of final study population. Included patients underwent primary total knee arthroplasty (TKA) or medial unicompartmental knee arthroplasty (mUKA) at Copenhagen University Hospital Hvidovre between 12 March 2018 and 27 February 2020. The exclusion criteria was simultaneous bilateral surgery, missing patient-reported outcome measure (PROM) data at all time points, TKA patients with previous anterior cruciate ligament (ACL) reconstruction, or with missing preoperative radiographs and TKA patients not eligible for mUKA. Additionally, the first surgery of patients with two surgeries in the inclusion period were excluded.
Prior to imputation, 274 patients had missing values for either BMI or PROM scores. Thus, 357 of the 631 patients had complete data, rendering a response rate of 56.6%. Proportion of missingness prior to imputation is shown in Supplementary Material.
Baseline characteristics, before and after matching, are shown in Table I. After matching, the groups had sufficient balance on all matched parameters, except BMI, where TKA patients had a higher median BMI, 30.0 kg/m2 (IQR 27.6 to 33.1), compared with mUKA patients, 28.7 kg/m2 (25.4 to 33.0).
AUC
The AUC was significantly lower for TKA patients compared with mUKA patients across all three PROMs, unadjusted and adjusted for preoperative PROM scores, ASA score, sex, and age (Table II, Figure 2). Mean AUC, calculated from individual AUCs for each patient, showed a greater area of development for mUKA patients in all three PROMs. Adjusted proportional differences in AUC, were -19% lower for TKA compared with mUKA in OKS (95% CI -27% to -9%); -23% lower for TKA in FJS (95% CI -32% to -14%), and -22% lower for TKA in APQ (95% CI -32% to -12%).
Table II.
Area under the curve regression results.
| Score | Mean mUKA, AUC (SD) | Mean TKA, AUC (SD) | ∆AUC estimate (crude) | 95% CI | p-value | ∆AUC estimate (adjusted)* | 95% CI | p-value |
|---|---|---|---|---|---|---|---|---|
| OKS | 130 (82.9) | 111 (84.3) | -19 | -32 to -5 | < 0.01 | -24 | -35 to -12 | < 0.001 |
| Percentage difference | -14.0 | -25% to -4% | -19.0 | -27 to -9 | ||||
| mUKA | 400 (258) | 317 (255) | -83 | -125 to -42 | < 0.001 | -92 | -129 to -55 | < 0.001 |
| Percentage difference | -21.0 | -31 to -11 | -23.0 | -32 to -14 | ||||
| APQ | 408 (296) | 325 (288) | -83 | -130 to -36 | < 0.001 | -89 | -131 to -47 | < 0.001 |
| Percentage difference | -20.0 | -32 to -9 | -22.0 | -32 to -12 |
Areas under the curve (AUCs) were calculated to measure improvement in patient-reported outcome measures (PROMs) over time. Estimates for AUC differences were compared using linear regression, both crude and adjusted for preoperative scores, sex, and age. Percentages illustrate the proportional difference in AUC for total knee arthroplasty patients compared with medial unicompartmental knee arthroplasty patients.
Adjusted for preoperative score, sex, age, and American Society of Anesthesiologists score.
APQ, activity and participation questionnaire; AUC, area under the curve; FJS, Forgotten Joint Score; mUKA, medial unicompartmental knee arthroplasty; OKS, Oxford Knee Score; TKA, total knee arthroplasty.
Fig. 2.
One-year trajectory of patient-reported outcome measure (PROM) scores (Oxford Knee Score (OKS), Forgotten Joint Score (FJS), activity and participation questionnaire (APQ)). This figure illustrates the mean change from preoperative PROM score, stratified for medial unicompartmental knee arthroplasty (mUKA) and total knee arthroplasty (TKA). The x-axis displays months after surgery, and the y-axis represents mean change from preoperative scores. Error bars denoting the 95% CI of the mean. a) OKS; b) FJS c) APQ.
A total of 15 patients (seven mUKA, eight TKA) underwent revision surgery within one year. Additional AUC analyses, excluding revised patients, and those with imputed data were consistent with the primary findings.
Median PROM scores
The median PROM scores at three months were 35 and 30 for OKS, 50 and 43 for FJS, 44 and 31 for APQ, for the mUKA and TKA groups, respectively. At six months, the median PROM scores increased to 39 and 35 for OKS, 62 and 50 for FJS, and 56 and 41 for APQ, for the mUKA and TKA groups, respectively. At 12 months median PROM scores increased to 39 and 38 for OKS, 63 and 56 for FJS, and 53 and 47 for APQ, for the mUKA and TKA groups, respectively (Figure 3). The most notable difference between mUKA and TKA was seen three to six months postoperatively, favouring mUKA.
Fig. 3.
Multivariate kernel density plot of patient-reported outcome measure (PROM) scores (Oxford Knee Score (OKS), Forgotten Joint Score (FJS), activity and participation questionnaire (APQ)). Density plots, showing the density of responses in PROM scores preoperatively, three to six and 12 months postoperatively, stratified for medial unicompartmental knee arthroplasty (mUKA) and total knee arthroplasty (TKA). Vertical lines represent median scores, for mUKA (red) and TKA (blue). a) OKS; b) FJS; c) APQ.
Probability of negative PROM change
The proportion of patients with a negative treatment effect on PROM scores was higher for TKA than mUKA in both AUC and at three, six, and 12 months after surgery for all three PROM scores, except APQ score at 12 months (Table III). However, at 12 months, the odds ratios for a TKA patient experiencing a negative treatment effect relative to a mUKA patient were close to 1 across all PROM scores, indicating minimal or no differences between groups at this time point.
Table III.
Probability of negative treatment effect.
| Probability of a negative treatment effect on OKS | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| ∆AUC | 3 months | 6 months | 12 months | |||||||||
| % | OR | 95% CI | % | OR | 95% CI | % | OR | 95% CI | % | OR | 95% CI | |
| UKA | 6.2 | Ref | 12.4 | Ref | 6.61 | Ref | 7.44 | Ref | ||||
| TKA | 8.78 | 1.46 | 0.75 to 2.95 | 18.4 | 1.59 | 0.98 to 2.62 | 11.2 | 1.8 | 0.98 to 3.44 | 8.24 | 1.12 | 0.59 to 2.18 |
| Probability of a negative treatment effect on FJS | ||||||||||||
| ∆AUC | 3 months | 6 months | 12 months | |||||||||
| % | OR | 95% CI | % | OR | 95% CI | % | OR | 95% CI | % | OR | 95% CI | |
| UKA | 5.37 | Ref | 8.26 | Ref | 7.02 | Ref | 8.68 | Ref | ||||
| TKA | 8.78 | 1.69 | 0.85 to 3.59 | 16.2 | 2.15 | 1.24 to 3.87 | 12 | 1.60 | 0.85 to 3.09 | 9.57 | 1.11 | 0.61 to 2.06 |
| Probability of a negative treatment effect on APQ | ||||||||||||
| ∆AUC | 3 months | 6 months | 12 months | |||||||||
| % | OR | 95% CI | % | OR | 95% CI | % | OR | 95% CI | % | OR | 95% CI | |
| UKA | 7.44 | Ref | 9.5 | Ref | 6.2 | Ref | 9.5 | Ref | ||||
| TKA | 9.31 | 1.28 | 0.68 to 2.46 | 12.8 | 1.39 | 0.80 to 2.47 | 11.7 | 2.0 | 1.06 to 3.97 | 8.78 | 0.92 | 0.51 to 1.68 |
Negative treatment effect defined as patients having lower postoperative patient-reported outcome measure (PROM) scores compared with preoperative PROM scores. Odds ratio (OR) was calculated as the odds of a total knee arthroplasty patient experiencing a negative treatment effect relative to a medial unicompartmental knee arthroplasty patient.
AUC, area under the curve; FJS, Forgotten Joint Score; OKS, Oxford Knee Score.
The biggest difference was seen at three and six months postoperatively. In OKS, 30 mUKA patients (12.4%) had a negative treatment effect at three months compared with 69 TKA patients (18.4%), and at six months, it was 16 (6.6%) for mUKA compared with 42 (11.2%) for TKA. In FJS, the proportion of patients with a negative treatment effect was 8.3% (n = 20) for mUKA compared with 16.2% (n = 61) for TKA at three months, and at six months the results were 7.0% (n = 17) for mUKA compared with 12.0% (n=45) for TKA. In APQ, 23 mUKA patients (9.5%) had a negative treatment effect at three months compared with 48 (12.8%) for TKA, and 15 (6.2%) for mUKA compared with 44 (11.7%) for TKA at six months. In contrast, the proportion of mUKA patients with a negative treatment effect on APQ at 12 months was slightly higher compared with TKA, with 23 (9.5%) for mUKA, and 33 (8.8%) for TKA. Statistically significant differences were, however, only found in FJS at three months and APQ at six months, both favouring mUKA.
Distribution of high scorers
A higher proportion of mUKA patients reached the highest possible scores compared to TKA on all three PROM scores. In OKS, 15 (6.1%) mUKA and 15 TKA patients (4%) reached the highest possible scores. In FJS, it was 11.6% (n = 28) for mUKA, and 6.4% (n = 24), for TKA patients. In APQ distribution, it was 18.6% (n = 45) of the mUKA, and 10.4% ( n= 39) of the TKA patients.
Discussion
This study showed mUKA patients had significantly better early functional recovery within 12 months compared with TKA, with TKA patients demonstrating 18% to 23% lower improvement in PROM scores, measured in AUC of OKS, FJS, and APQ. Additionally, mUKA patients had higher median PROM scores, especially at three and six months.
To our knowledge, this is the first comparison of the early one-year postoperative trajectory of mUKA and TKA surgery using difference in AUC, making it difficult to compare directly with other studies. The use of AUC may provide a more nuanced insight into the trajectory of recovery, compared with traditional single-point assessments, as it reflects a time-weighted average of PROM scores across follow-up time points. The TOPKAT study used AUC of mean OKS as a secondary outcome, showing a 1.54 points lower AUC of mean OKS score for TKA patients compared with mUKA patients, but this was calculated based on all data available from the five-year follow-up period.15 As most of the functional recovery appears to occur within the first year, a 12-month AUC approach may offer a more focused and clinically relevant insight into differences in early postoperative outcomes.
Both groups showed improvements in OKS from baseline that exceeded the minimal important change (MIC) thresholds for UKA and TKA at both three and 12 months.39 This indicates that both procedures led to patient-perceived meaningful change. Currently, there is no established minimal clinically important difference in AUC for OKS, FJS, or APQ. Still, our results are supported by findings of the TOPKAT study, which indicate a stagnation in functional recovery for both mUKA and TKA within the first year.15 Additionally, a 2021 randomized controlled trial showed faster functional recovery for mUKA patients, with a 6.2 mean between-group difference in OKS score two months after surgery, although no difference were found at two years.24 The findings of the present study, however, contrast with those of Sershon et al,20 who found no significant difference six weeks and six months after surgery in FJS, Knee Injury score, and Osteoarthritis Outcome Score (KOOS), and Knee Society Score (KSS).
While existing research suggests comparable long-term functional outcomes for mUKA and TKA,19-22 the present study focuses instead on short-term functional recovery. The results, indicating a faster recovery for mUKA patients, may add nuance to discussions of surgical treatment options. While the clinical implications of early PROM improvements remain to be fully understood, they could be relevant to certain patient groups prioritizing early postoperative function.
Being a retrospective cohort study, the absence of randomization may introduce imbalance in baseline characteristics. Selection-bias could be considered a potential limitation, as some argue mUKA is more likely to be chosen for healthier patients.6 After matching on BMI, sex, age, and preoperative PROM scores, the TKA group still had a 1.3 higher median BMI and a slightly higher proportion of ASA 1 patients. The AUC calculations were adjusted for preoperative PROM scores, ASA score, sex, and age, to minimize the impact of remaining imbalances. However, residual confounding from unknown variables cannot be definitively ruled out.40
Another limitation of this study is that all mUKAs performed during the inclusion period were uncemented Oxford implants, which may limit generalizability of the findings to cemented or other implant designs.
Variation in surgeon groups, with five performing both mUKA and TKA, and six performing only TKA, presents a potential source of bias. Matching was used to minimize this, but differences in surgeon experience and learning curves may still have influenced outcomes.
When using PROMs, missing response data is a limitation. Five patients were excluded due to missing data at all time points. The proportion with complete data was 56.6% and may cause some selection-bias. This was addressed by using multiple imputation. Comparing mean PROM scores before and after imputation showed a slight decrease in postoperative scores for both groups, suggesting a potential bias where patients with poorer outcomes are less inclined to respond to postoperative questionnaires. A 2022 study on THA, using the FJS questionnaire, identified a similar tendency.41 By using multiple imputation, we tried to mitigate this bias, although some degree of non-responder bias cannot be fully excluded due to the moderate response rate.
A higher proportion of mUKA patients reached the highest possible scores, on all three PROM scores. Specifically, 18.6% of mUKA patients reached the highest possible score in APQ, exceeding the 15% threshold commonly defined for problematic ceiling effect.32,42,43 Ceiling effect reduces the ability to accurately measure improvement, for those scoring at the highest levels.44 This study showed mUKA patients had the highest scores and improvement in APQ compared with TKA patients. Thus, the observed difference in APQ, favouring mUKA, may potentially be higher than what APQ can measure, as the highest scorers may be constrained by ceiling effect.
The findings of this study indicate that functional recovery predominantly happens within three to six months after surgery. Future research could therefore be recommended to focus on short-term recovery, alongside revision rates for mUKA and TKA, as there is a possibility for early differences that might be of importance to patients.
In conclusion, Oxford mUKA patients had significantly better early functional recovery within the first 12 months compared with TKA patients. Although previous studies have demonstrated comparable long-term outcomes, these findings of differences in early functional recovery may be relevant for surgeons to consider, when discussing surgical treatment options with patients.
Take home message
- Medial unicompartmental knee arthroplasty (mUKA) patients demonstrate significantly better early postoperative functional recovery compared to total knee arthroplasty (TKA) patients.
- The majority of functional recovery for both mUKA and TKA patients occurs within the first three to six months postoperatively, suggesting that short-term recovery may be a key differentiating factor between the two procedures.
Author contributions
A. L. Elkjær Christensen: Data curation, Formal analysis, Investigation, Methodology, Project administration, Visualization, Writing – original draft, Writing – review & editing
C. Bredgaard Jensen: Data curation, Formal analysis, Investigation, Methodology, Project administration, Supervision, Visualization, Writing – review & editing
C. Henkel: Methodology, Supervision, Writing – review & editing
L. H. Ingelsrud: Methodology, Supervision, Writing – review & editing
K. I. Bunyoz: Methodology, Supervision, Writing – review & editing
K. Gromov: Methodology, Writing – review & editing
A. J. Price: Conceptualization, Methodology, Writing – review & editing
A. Troelsen: Conceptualization, Data curation, Methodology, Resources, Supervision, Writing – review & editing
Funding statement
The author(s) received no financial or material support for the research, authorship, and/or publication of this article.
ICMJE COI statement
A. L. E. Christensen received an unrelated research scholarship during the study period (August 2023 to July 2024) from the Novo Nordisk Foundation (grant no. NNF21OC0072155). C. B. Jensen is a board member of the Young Orthopaedics Denmark Association.C. Henkel has no conflicts of interest to declare. L. H. Ingelsrud is co-chair of the International Society of Arthroplasty Registries (ISAR) Patient-Reported Outcome Measures Working Group (unpaid). K. I. Bunyoz reports support for attending meetings and/or travel from Zimmer Biomet and Clockwork Medical, unrelated to this study.K. Gromov reports unrelated research and institutional support from Zimmer Biomet. A. J. Price receives unrelated consulting fees from Zimmer Biomet.A. Troelsen reports research support, travel/accommodations/meeting expenses, consulting fees, and advisory board membership from Zimmer Biomet, and research support, consulting fees, and advisory board membership from Pfizer Denmark.
Data sharing
The data that support the findings for this study are available to other researchers from the corresponding author upon reasonable request.
Acknowledgements
We would like to thank the staff at Clinical Orthopaedic Research Hvidovre (CORH) for their support, and acknowledge the patients who contributed to the questionnaire-based arthroplasty database.
Ethical review statement
As this study is purely observational, ethics committee approval was not required according to Danish law. Patients provided informed consent to the questionnaire-based arthroplasty database. Institutional approval was granted to apply data from this database (P-2022 to 290). Access to routinely obtained preoperative x-rayradiographs was granted by the Danish Data Protection Agency (R-24028352).
Open access funding
The open acccess fee was self-funded by Clinical Orthopaedic Research Hvidovre, Department of Orthopaedic Surgery, Copenhagen University Hospital Hvidovre, Denmark.
Supplementary material
Additional tables providing baseline characteristics before and after imputation, missing data distribution, responder subgroups, high-scorer distributions, baseline characteristics of patients revised within one year, and supplementary area under the curve regression analyses, excluding patients revised within one year and excluding partial responders.
© 2025 Elkjær Christensen et al. This article is distributed under the terms of the Creative Commons Attributions (CC BY 4.0) licence (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium or format, provided the original author and source are credited.
Data Availability
The data that support the findings for this study are available to other researchers from the corresponding author upon reasonable request.
References
- 1. Price AJ, Alvand A, Troelsen A, et al. Knee replacement. Lancet. 2018;392(10158):1672–1682. doi: 10.1016/S0140-6736(18)32344-4. [DOI] [PubMed] [Google Scholar]
- 2. Murray DW, Parkinson RW. Usage of unicompartmental knee arthroplasty. Bone Joint J. 2018;100-B(4):432–435. doi: 10.1302/0301-620X.100B4.BJJ-2017-0716.R1. [DOI] [PubMed] [Google Scholar]
- 3. Hamilton TW, Pandit HG, Lombardi AV, et al. Radiological decision aid to determine suitability for medial unicompartmental knee arthroplasty: development and preliminary validation. Bone Joint J. 2016;98(10):3–10. doi: 10.1302/0301-620X.98B10.BJJ-2016-0432.R1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Pandit H, Jenkins C, Gill HS, et al. Unnecessary contraindications for mobile-bearing unicompartmental knee replacement. J Bone Joint Surg Br. 2011;93-B(5):622–628. doi: 10.1302/0301-620X.93B5.26214. [DOI] [PubMed] [Google Scholar]
- 5. Goodfellow JW, O’Connor JJ, Murray DW. A critique of revision rate as an outcome measure: re-interpretation of knee joint registry data. J Bone Joint Surg Br. 2010;92-B(12):1628–1631. doi: 10.1302/0301-620X.92B12.25193. [DOI] [PubMed] [Google Scholar]
- 6. Murray DW, Liddle AD, Judge A, Pandit H. Bias and unicompartmental knee arthroplasty. Bone Joint J. 2017;99-B(1):12–15. doi: 10.1302/0301-620X.99B1.BJJ-2016-0515.R1. [DOI] [PubMed] [Google Scholar]
- 7. Tay ML, Monk AP, Frampton CM, Hooper GJ, Young SW. A comparison of clinical thresholds for revision following total and unicompartmental knee arthroplasty. Bone Joint J. 2023;105-B(3):269–276. doi: 10.1302/0301-620X.105B3.BJJ-2022-0872.R2. [DOI] [PubMed] [Google Scholar]
- 8. Mikkelsen M, Price A, Pedersen AB, Gromov K, Troelsen A. Optimized medial unicompartmental knee arthroplasty outcome: learning from 20 years of propensity score matched registry data. Acta Orthop. 2022;93:390–396. doi: 10.2340/17453674.2022.2265. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Liddle AD, Pandit H, Judge A, Murray DW. Optimal usage of unicompartmental knee arthroplasty: a study of 41,986 cases from the National Joint Registry for England and Wales. Bone Joint J. 2015;97-B(11):1506–1511. doi: 10.1302/0301-620X.97B11.35551. [DOI] [PubMed] [Google Scholar]
- 10. Liddle AD, Pandit H, Judge A, Murray DW. Effect of surgical caseload on revision rate following total and unicompartmental knee replacement. J Bone Joint Surg Am. 2016;98-A(1):1–8. doi: 10.2106/JBJS.N.00487. [DOI] [PubMed] [Google Scholar]
- 11. Mohammad HR, Matharu GS, Judge A, Murray DW. The effect of surgeon caseload on the relative revision rate of cemented and cementless unicompartmental knee replacements: an analysis from the National Joint Registry for England, Wales, Northern Ireland and the Isle of Man. J Bone Joint Surg Am. 2020;102-A(8):644–653. doi: 10.2106/JBJS.19.01060. [DOI] [PubMed] [Google Scholar]
- 12. Liddle AD, Judge A, Pandit H, Murray DW. Adverse outcomes after total and unicompartmental knee replacement in 101,330 matched patients: a study of data from the National Joint Registry for England and Wales. Lancet. 2014;384(9952):1437–1445. doi: 10.1016/S0140-6736(14)60419-0. [DOI] [PubMed] [Google Scholar]
- 13. Hunt LP, Blom AW, Matharu GS, et al. Patients receiving a primary unicompartmental knee replacement have a higher risk of revision but a lower risk of mortality than predicted had they received a total knee replacement: data from the National Joint Registry for England, Wales, Northern Ireland, and the Isle of Man. J Arthroplasty. 2021;36(2):471–477. doi: 10.1016/j.arth.2020.08.063. [DOI] [PubMed] [Google Scholar]
- 14. Jensen CB, Petersen PB, Jørgensen CC, et al. Length of stay and 90-day readmission/complication rates in unicompartmental versus total knee arthroplasty: a propensity-score-matched study of 10,494 procedures performed in a fast-track setup. J Bone Joint Surg Am. 2021;103-A(12):1063–1071. doi: 10.2106/JBJS.20.01287. [DOI] [PubMed] [Google Scholar]
- 15. 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(10200):746–756. doi: 10.1016/S0140-6736(19)31281-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Hamilton TW, Rizkalla JM, Kontochristos L, et al. The interaction of caseload and usage in determining outcomes of unicompartmental knee arthroplasty: a meta-analysis. J Arthroplasty. 2017;32(10):3228–3237. doi: 10.1016/j.arth.2017.04.063. [DOI] [PubMed] [Google Scholar]
- 17. Bredgaard Jensen C, Gromov K, Petersen PB, Jørgensen CC, Kehlet H, Troelsen A. Short-term surgical complications following fast-track medial unicompartmental knee arthroplasty. Bone Jt Open. 2023;4(6):457–462. doi: 10.1302/2633-1462.46.BJO-2023-0054.R1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Wilson HA, Middleton R, Abram SGF, 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. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Meena A, Abermann E, Hoser C, et al. No difference in sports participation and patient-reported functional outcomes between total knee arthroplasty and unicompartmental knee arthroplasty at minimum 2-year follow-up in a matched control study. Knee Surg Sports Traumatol Arthrosc. 2023;31(8):3098–3105. doi: 10.1007/s00167-022-07166-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Sershon RA, Fricka KB, Hamilton WG, et al. Early results of a randomized controlled trial of partial versus total knee arthroplasty. J Arthroplasty. 2022;37(6):S94–S97. doi: 10.1016/j.arth.2022.02.076. [DOI] [PubMed] [Google Scholar]
- 21. Tveit M. The renaissance of unicompartmental knee arthroplasty appears rational - a radiograph-based comparative study on adverse events and patient-reported outcomes in 353 TKAs and 98 UKAs. PLOS One. 2021;16(9):e0257233. doi: 10.1371/journal.pone.0257233. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Lyons MC, MacDonald SJ, Somerville LE, Naudie DD, McCalden RW. Unicompartmental versus total knee arthroplasty database analysis: is there a winner? Clin Orthop Relat Res. 2012;470(1):84–90. doi: 10.1007/s11999-011-2144-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Annapareddy A, Mulpur P, Prakash M, Suhas Masilamani AB, Eachempati KK, Gurava Reddy AV. Partial versus total knee arthroplasty for isolated antero-medial osteoarthritis - an analysis of PROMs and satisfaction. SICOT J. 2023;9:11. doi: 10.1051/sicotj/2023006. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Knifsund J, Niinimaki T, Nurmi H, et al. Functional results of total-knee arthroplasty versus medial unicompartmental arthroplasty: two-year results of a randomised, assessor-blinded multicentre trial. BMJ Open. 2021;11(6):e046731. doi: 10.1136/bmjopen-2020-046731. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Mikkelsen M, Wilson HA, Gromov K, Price AJ, Troelsen A. Comparing surgical strategies for end-stage anteromedial osteoarthritis: total versus unicompartmental knee arthroplasty. Bone Jt Open. 2022;3(5):441–447. doi: 10.1302/2633-1462.35.BJO-2021-0174.R1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Saklad M. Grading of patients for surgical procedures. Anesthesiology. 1941;2(3):281–284. doi: 10.1097/00000542-194105000-00004. [DOI] [Google Scholar]
- 27. Dawson J, Beard DJ, McKibbin H, Harris K, Jenkinson C, Price AJ. Development of a patient-reported outcome measure of activity and participation (the OKS-APQ) to supplement the Oxford knee score. Bone Joint J. 2014;96-B(3):332–338. doi: 10.1302/0301-620X.96B3.32845. [DOI] [PubMed] [Google Scholar]
- 28. Murray DW, Fitzpatrick R, Rogers K, et al. The use of the Oxford Hip and Knee scores. J Bone Joint Surg Br. 2007;89-B(8):1010–1014. doi: 10.1302/0301-620X.89B8.19424. [DOI] [PubMed] [Google Scholar]
- 29. Behrend H, Giesinger K, Giesinger JM, Kuster MS. The “forgotten joint” as the ultimate goal in joint arthroplasty: validation of a new patient-reported outcome measure. J Arthroplasty. 2012;27(3):430–436. doi: 10.1016/j.arth.2011.06.035. [DOI] [PubMed] [Google Scholar]
- 30. Dawson J, Fitzpatrick R, Murray D, Carr A. Questionnaire on the perceptions of patients about total knee replacement. J Bone Joint Surg Br. 1998;80-B(1):63–69. doi: 10.1302/0301-620x.80b1.7859. [DOI] [PubMed] [Google Scholar]
- 31. Thomsen MG, Latifi R, Kallemose T, Barfod KW, Husted H, Troelsen A. Good validity and reliability of the forgotten joint score in evaluating the outcome of total knee arthroplasty: a retrospective cross-sectional survey-based study. Acta Orthop. 2016;87(3):280–285. doi: 10.1016/j.arth.2022.02.076. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Deng W, Shao H, Zhou Y, Li H, Wang Z, Huang Y. Reliability and validity of commonly used patient-reported outcome measures (PROMs) after medial unicompartmental knee arthroplasty. Orthop Traumatol Surg Res. 2022;108(8):103096. doi: 10.1016/j.otsr.2021.103096. [DOI] [PubMed] [Google Scholar]
- 33.No authors listed Procordo. [28 November 2025]. https://procordo.com/en/ date last. accessed.
- 34. Buuren S, mice G-O. Multivariate imputation by chained equations in R J Stat Softw 2011. 45 3 1 67 10.18637/jss.v045.i03 26889483 [DOI] [Google Scholar]
- 35. Kane LT, Fang T, Galetta MS, et al. Propensity score matching: a statistical method. Clin Spine Surg. 2020;33(3):120–122. doi: 10.1097/BSD.0000000000000932. [DOI] [PubMed] [Google Scholar]
- 36. Odgaard A, Madsen F, Kristensen PW, Kappel A, Fabrin J. The mark coventry award: patellofemoral arthroplasty results in better range of movement and early patient-reported outcomes than TKA. Clin Orthop Relat Res. 2018;476(1):87–100. doi: 10.1007/s11999.0000000000000017. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37. Matthews JNS, Altman DG, Campbell MJ, Royston P. Analysis of serial measurements in medical research. BMJ. 1990;300(6719):230–235. doi: 10.1136/bmj.300.6719.230. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38. Sloan JA, Dueck AC, Erickson PA, et al. Analysis and interpretation of results based on patient-reported outcomes. Value Health. 2007;10:S106–S115. doi: 10.1111/j.1524-4733.2007.00273.x. [DOI] [PubMed] [Google Scholar]
- 39. Harris LK, Troelsen A, Terluin B, Gromov K, Ingelsrud LH. Minimal important change thresholds change over time after knee and hip arthroplasty. J Clin Epidemiol. 2024;169:111316. doi: 10.1016/j.jclinepi.2024.111316. [DOI] [PubMed] [Google Scholar]
- 40. Freemantle N, Marston L, Walters K, Wood J, Reynolds MR, Petersen I. Making inferences on treatment effects from real world data: propensity scores, confounding by indication, and other perils for the unwary in observational research. BMJ. 2013;347:f6409. doi: 10.1136/bmj.f6409. [DOI] [PubMed] [Google Scholar]
- 41. Yamate S, Hamai S, Kawahara S, et al. Multiple imputation to salvage partial respondents: analysis of the forgotten joint score-12 after total hip arthroplasty. J Bone Joint Surg Am. 2022;104-A(24):2195–2203. doi: 10.2106/JBJS.21.01547. [DOI] [PubMed] [Google Scholar]
- 42. Saarinen A, Pakarinen O, Vaajala M, et al. Randomized controlled trials reporting patient-reported outcomes with no significant differences between study groups are potentially susceptible to unjustified conclusions-a systematic review. J Clin Epidemiol. 2024;169:111308. doi: 10.1016/j.jclinepi.2024.111308. [DOI] [PubMed] [Google Scholar]
- 43. Harris K, Dawson J, Gibbons E, et al. Systematic review of measurement properties of patient-reported outcome measures used in patients undergoing hip and knee arthroplasty. Patient Relat Outcome Meas. 2016;7(7):101–108. doi: 10.2147/PROM.S97774. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44. Eckhard L, Munir S, Wood D, et al. The ceiling effects of patient reported outcome measures for total knee arthroplasty. Orthop Traumatol Surg Res. 2021;107(3):102758. doi: 10.1016/j.otsr.2020.102758. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The data that support the findings for this study are available to other researchers from the corresponding author upon reasonable request.