Skip to main content
NCBI home page
Cureus logoLink to Cureus
. 2026 May 25;18(5):e109613. doi: 10.7759/cureus.109613

Assessment of Fixed Bearing and Rotating Platform Total Knee Arthroplasty Designs Using the Patient Knee Implant Performance Scoring System: Post-hoc Pooled Analysis of Data From Two Prospective, International, Multi-centre, Non-randomised Studies

Naseer Mohammed Abdul 1,2,✉, Thierry Bernard 3, Piers Yates 1
Editors: Alexander Muacevic, John R Adler
PMCID: PMC13293154  PMID: 42359216

Abstract

Background

The optimal choice of polyethylene insert in total knee arthroplasty (TKA) continues to cause contention. Rotating platform bearings (RP) have theoretical advantages but have failed to demonstrate a functional difference compared to fixed bearings (FB) in clinical studies. No studies have analysed the effect of body mass index (BMI) on functional scores between RP and FB TKA. No studies have investigated RP and FB TKA using the new Patient Knee Implant Performance (PKIP) patient-reported outcome measure (PROM).

Methods

This was a post-hoc analysis of the data from two prospective, non-randomised, multi-centre studies conducted across centres in Australia, the United Kingdom, and the United States by DePuy Synthes (West Chester, Pennsylvania, United States). Functional outcomes between RP TKA and FB TKA were compared. Baseline and two-year follow-up data were compared across multiple PROMs in both groups, including PKIP. The analysis included adjustment for BMI between groups.

Results

A total of 1,718 patients across 27 centres were included in the study. Of this, 959 participants underwent FB TKA and 759 participants underwent RP TKA. No significant difference was found between RP and FB in the change from baseline at two years across any parameter in all PROMs. When adjusted for BMI, statistical differences were seen between RP TKA and FB TKA at two years in several PROMs.

Conclusion

Both FB and RB showed excellent results at two years. Our data does not demonstrate a significant difference in RP TKA over FB TKA at two years in any PROM. Advantages were seen in some PROMs for RP when adjusted for BMI alone at two years.

Keywords: bmi, fixed bearing, proms, rotating platform, total knee arthroplasty

Introduction

Total knee arthroplasty (TKA) is a reliable and reproducible form of treatment for patients with end-stage osteoarthritis, with over 78,000 TKAs performed in Australia in the year 2024 alone [1]. Despite improved survivorship rates, aseptic loosening remains one of the major causes of revision of TKA [2]. In addition, there remains a high proportion of patients dissatisfied with their TKA, with reports ranging from 12.7% to 20%, despite ongoing innovation in implant design [3]. Whilst multiple factors may contribute to this, a component of the TKA that has continued to be debated is the design of the polyethylene insert.

Rotating platform bearings (RP) have a theoretical advantage when compared to fixed bearings (FB) by reducing contact stress between surfaces, leading to reduced rates of wear and loosening [4]. In RP, the insert can rotate around an axis and theoretically mimic the physiological kinematics of the knee, thus reducing force through the tibial prosthesis [5]. In vivo biomechanical studies have demonstrated an advantage in RP systems compared to FB systems with regard to wear rates [6]. RP is part of a larger group of ‘mobile bearings’, but the RP controls motion on both the upper and lower polyethylene interfaces.

Despite this theoretical advantage, no clinical studies thus far have demonstrated a difference in terms of functional or radiological outcome between RP TKA and FB TKA [7]. All randomised studies have been limited by their small sample size [8,9] and have utilised traditional patient-reported outcome measures (PROMs). Additionally, whilst the index theoretical advantage of RP TKA is reduced wear, no study has investigated the differences between RP TKA and FB TKA when adjusted for body mass index (BMI). A recent large registry-based analysis has, however, suggested higher rates of revision and mechanical loosening associated with MB designs when compared to FB TKA, particularly in longer-term follow-up from national registries [10]. This registry finding contrasts with smaller clinical studies focused on functional outcomes and highlights the need to better understand whether differences exist in patient-reported performance, particularly in specific patient subgroups.

PROMs are questionnaires completed by patients to assess the degree of difficulty or symptom severity. With regard to TKA, whilst some PROMs are generic overviews of a person’s health, such as the EuroQol five-dimensional questionnaire (EQ-5D) [11], specific PROMs exist for the assessment of knee joints. Commonly used PROMs in TKA studies include the Oxford Knee Score (OKS) [12], the Knee Injury Osteoarthritic Outcome Score (KOOS) [13], and the American Knee Society (AKS) score [14].

The conventional knee scoring systems have limitations in that they do not reflect outcomes or measure aspects important to patients, such as confidence whilst performing activities, the perceived stability of the knee, and overall satisfaction with their joint replacement. They also often have a ceiling effect, which reduces the ability to differentiate between better levels of knee performance, and are subject to observer bias [15,16]. New scoring systems have been developed to address these limitations, such as the Patient Implant Knee Performance (PKIP) [17]; however, this instrument has not yet been utilised in comparing RP TKA and FB TKA.

The primary objective of this study was to compare two-year patient-reported functional outcomes between RP TKA and FB TKA using the PKIP PROM. Secondary objectives were to compare functional outcomes across additional validated PROMs and to explore whether BMI influenced functional outcomes between implant designs

Materials and methods

This study was a post-hoc comparative observational analysis of data derived from two prospective, non-randomised, multi-centre clinical studies [18,19], which were run and conducted by DePuy Synthes (West Chester, Pennsylvania, United States) as part of routine product evaluation. Participants were recruited from across sites in the United States, the United Kingdom, and Australia from November 14, 2012, to May 27, 2015. Participants were followed for two years. Approval was obtained from each participating centre’s Institutional Review Board or ethics committee, and informed written consent was obtained from all participants before enrolment into the studies. The current study was approved by the Western Institutional Review Board (WIRB PRO NUM 20121213/20111599).

Participants were recruited if they were undergoing TKA with either the P.F.C. SIGMA™ Knee System (DePuy Synthes) or the original ATTUNE™ Primary Knee System (DePuy Synthes). Surgeons participating in this study were experienced with these implants and used them in their daily practice. Baseline demographics, preoperative interventions, and PROMs were collected within 90 days of surgery. All participants had standard postoperative care as per their institution and were followed up for two years. PROMs collected were: KOOS [13], OKS [12], EQ5D-3L [20], Knee Society Score (KSS) [21], and PKIP [22]. 

The PKIP Questionnaire was developed to address limitations of traditional knee PROMs. PKIP scores the performance of the knee prosthesis across five subscales [17]. These subscales assess the patient’s awareness of the implant, their confidence with the knee, their sense of stability during activities, the modification of any activities, and satisfaction with knee functions. It comprises 24 items which employ a 5-, 6-, or 11-point ordinal response scale, where higher scores indicate better knee function, and an overall score ranging from zero to 100. PKIP is a highly responsive PROM that has undergone psychometric validation and has demonstrated the capability of discriminating between groups with better or worse knee function [22].

All PROMs used in this study are validated instruments. The KOOS was available free at the time of our study (it was free till 2023), a license agreement was done with Oxford University Innovation Limited for the use of the OKS, and the KSS was also used with permission from the Knee Society. Use of the EQ-5D-3L was conducted in accordance with licensing requirements from the EuroQol Research Foundation. 

Data analysis

All data were analysed using SAS software version 9.4 (SAS Institute Inc., Cary, North Carolina, United States). Demographic characteristics were compared using the two-sided independent t-test for continuous variables and using Fisher’s exact test for categorical variables. Survival estimates and confidence intervals were produced using the Kaplan-Meier methodology, and p-values comparing FB to RP were produced using the log-rank test to analyze rates of revision of any component and rates of revision with removal of either the tibial and/or the femoral component.

The change from baseline (CFB) was calculated as follow-up minus the preoperative measurement. Adjusted changes from baseline were adjusted by using a propensity score methodology. Age, sex, BMI, country, product (PFC SIGMA/ATTUNE), and three baseline ligament indicators (previous treatments: Arthroscopy/Lateral Release/Ligament Reconstruction) were used to establish propensity scores, which predict the subjects’ design type (FB or RP). 

For each PROM separately, propensity scores as well as baseline PROM and design type were then used as covariates in a general linear model to estimate the adjusted change from baseline at two years. BMI was considered a potential confounder influencing functional outcomes following TKA. Therefore, BMI was incorporated into propensity score adjustment models, and a secondary analysis examining BMI-adjusted outcomes was also performed. Propensity scores were generated using logistic regression with implant design as the dependent variable and clinically relevant baseline covariates included in the model. Missing data were handled using complete-case analysis. Centre- or surgeon-level clustering adjustments were not performed.

Results

A total of 1,718 participants from 27 centres across Australia, the United States, and the United Kingdom were included in this analysis (Table 1). Of these, 959 participants underwent FB TKA and 759 participants underwent RP TKA; 85% of participants in FB TKA and 81% of participants in RP TKA completed two-year data sets. 

Table 1. Demographics and breakdown.

FB: fixed bearings; TKA: total knee arthroplasty; RP: rotating platform bearings

Characteristics FB TKA (n=959) RP (n=759)
Gender, n (%)
Female 564 (58.8%) 442 (58.2%)
Male 395 (41.2%) 317 (41.8%)
Age
n 959 759
Mean (SD) 66.1 (7.8) 64.3 (8.1)
Range 39, 80 28, 85
BMI
n 954 759
Mean (SD) 31.9 (5.9) 32.3 (6.2)
Range 16.7, 55.4 16.5, 56.1
BMI >30
Yes 459 (47.9%) 304 (40.1%)
No 500 (52.1%) 455 (59.9%)
BMI >40
Yes 94 (9.9%) 69 (9.1%)
No 860 (90.1%) 690 (90.9%)
Country-wise Distribution
Patients FB (n=959) RP (n= 759) Total
Australia 148 (15.4%) 65 (8.6%) 213 (12.4%)
United Kingdom 408 (42.5%) 39 (5.1%) 447 (26%)
United States 403 (42%) 655 (86.3%) 1058 (61.6%)
Sites FB (n=16) RP (n=11) Total
Australia 3 (18.8%) 1 (9.1%) 4 (14.8%)
United Kingdom 7 (43.8%) 1 (9.1%) 8 (29.6%)
United States 6 (37.5%) 9 (81.8%) 15 (55.6%)
Surgeons FB (n=27) RP (n=12) Total
Australia 9 (33.3%) 1 (8.3%) 10 (25.6%)
United Kingdom 13 (48.1%) 1 (8.3%) 14 (35.9%)
United States 5 (18.5%) 10 (83.3%) 15 (38.5%)
Open in a new tab

Demographics

There was no difference in baseline demographics, including age and gender (Table 1). The mean age of enrolment was 66.1 years in FB TKA (range, 39-80) and 64.3 years in RP TKA (range, 34-80). Gender was evenly distributed across both groups. Female patients comprised 58.8% of FB TKA and 58.2% of RP TKA. 

Mean BMI was found to be 31.9 in FB TKA and 32.3 in RP TKA. The FB TKA group had a slightly higher proportion of patients with a BMI >30 kg/m2 (47.9%) compared to the RP TKA group (40.1%). Patients with a BMI >40 kg/m2 accounted for 9.9% of FB TKA and 9.1% RP TKA. Other co-morbidities were similar between groups, with no statistically significant differences observed. 

There was no difference in average follow-up between groups. 

Preoperative interventions

Preoperative interventions for both FB TKA and RP TKA are displayed in Table 2. Slightly higher interventions were seen in the FB TKA, but overall, the groups were comparable. 

Table 2. Previous treatments for participants with RP TKA and FB TKA.

FB: fixed bearings; TKA: total knee arthroplasty; RP: rotating platform bearings

Previous Treatment FB RP
None 39.80% 33.50%
Intra-articular knee injections 31.90% 32.30%
Total Knee Arthroscopy 31.40% 27.90%
Meniscal 9.50% 10.00%
Total Knee 5.00% 5.00%
Other 4.00% 2.00%
Total Hip Arthroplasty 3.90% 2.90%
Debridement 3.10% 0.00%
Ligament reconstruction 1.60% 5.10%
Open in a new tab

PROMs

When adjusted for all confounders, no significant difference was found between RP TKA and FB TKA in change from baseline at two years across any parameter in all PROMs investigated (Table 3). With regard to PKIP, no difference was found in either group in any subscale at two years. There were similar improvements from baseline across both groups in all PROMs and their subscales.

Table 3. Baseline and two-year PROMs for RP TKA vs FB TKA.

FB: fixed bearing; RP: rotating platform; CFB: change from baseline; PROMs: patient-reported outcome measures; PKIP: Patient Knee Implant Performance questionnaire [22]; KOOS: Knee Injury and Osteoarthritis Outcome Score [13]; OKS: Oxford Knee Score [12]; EQ-5D-3L: EuroQol five-dimensional questionnaire [20]; KSS: Knee Society Score [21]; TKA: total knee arthroplasty

KOOS was available freely for academic research use at the time of our study (prior to 2023); OKS was used under an license agreement with Oxford University Innovation Limited; KSS was used with permission from The Knee Society. EQ-5D-3L was used in accordance with licensing requirements from the EuroQol Research Foundation. 

Parameters FB (N) RP (N) Baseline mean 2-year mean Change from baseline mean Adjusted change from baseline mean
FB RP Test t Statistic p-value FB RP Test t Statistic p-value FB CFB RP CFB Test t Statistic p-value FB Adjusted CFB RP Adjusted CFB Test F Statistic p-value
PKIP Overall 783 581 27.5 28.4 t-test 1.21 0.227 74.2 75.4 t-test 1.16 0.248 46.7 47.0 t-test 0.27 0.786 46.8 46.9 ANCOVA 0.00 0.983
PKIP Confidence 789 585 3.5 3.8 t-test 2.45 0.014 8.1 8.3 t-test 1.98 0.048 4.6 4.5 t-test 0.34 0.732 4.6 4.5 ANCOVA 0.20 0.655
PKIP Stability 795 594 3.2 3.5 t-test 2.88 0.004 8.3 8.5 t-test 2.38 0.018 5.1 5.0 t-test 0.38 0.703 5 5.1 ANCOVA 0.40 0.528
PKIP Satisfaction 794 596 2.1 2.2 t-test 0.82 0.414 8.1 8.2 t-test 0.53 0.596 6 6.0 t-test 0.08 0.939 6 6 ANCOVA 0.29 0.588
PKIP Modify Activities 796 597 4.1 3.7 t-test 2.80 0.005 6.4 6.6 t-test 1.12 0.264 2.2 2.9 t-test 2.66 0.008 2.4 2.7 ANCOVA 1.30 0.255
KOOS Function in Daily Living 796 598 47.1 50.3 t-test 3.33 0.001 87.1 88.4 t-test 1.61 0.108 40 38.1 t-test 1.74 0.082 39.3 39 ANCOVA 0.05 0.817
KOOS Pain 796 598 42.7 46.3 t-test 4.17 < .0001 87.2 88.1 t-test 1.10 0.273 44.5 41.8 t-test 2.47 0.014 43.4 43.4 ANCOVA 0.00 0.984
KOOS Symptoms 797 599 45.4 48.5 t-test 3.16 0.002 83.4 81.9 t-test 1.73 0.083 37.9 33.4 t-test 3.88 0 36.7 35 ANCOVA 3.19 0.074
KOOS Function in Sport and Recreation 765 554 18 19.1 t-test 0.98 0.327 62.5 67.8 t-test 3.39 0.001 44.4 48.7 t-test 2.51 0.012 46.2 46.3 ANCOVA 0.01 0.924
KOOS Quality of Life 797 599 22.3 23.9 t-test 1.84 0.066 74.8 74.1 t-test 0.54 0.586 52.5 50.1 t-test 1.66 0.097 52.3 50.5 ANCOVA 1.64 0.2
Oxford Knee Total Score (0 - 48) 797 592 21.6 23.7 t-test 5.04 < .0001 41 42 t-test 2.61 0.009 19.4 18.3 t-test 2.37 0.018 18.9 19 ANCOVA 0.05 0.829
EQ5D3L USA Index 797 592 0.6 0.7 t-test 2.48 0.013 0.9 0.9 t-test 3.20 0.001 0.2 0.2 t-test 0.05 0.958 0.2 0.2 ANCOVA 1.68 0.195
EQ5D3L Current Health 796 596 71.3 75.9 t-test 4.95 < .0001 79.6 82.3 t-test 3.13 0.002 8.3 6.4 t-test 1.82 0.069 7.4 7.6 ANCOVA 0.06 0.804
Knee Society Objective Knee Score 494 303 42.5 45.7 t-test 2.21 0.027 90.2 90.9 t-test 0.74 0.457 47.8 45.3 t-test 1.54 0.124 47.4 45.9 ANCOVA 2.51 0.113
Knee Society Satisfaction Score 507 318 11.8 13 t-test 2.57 0.01 34.1 34.4 t-test 0.60 0.552 22.3 21.5 t-test 1.26 0.209 22.1 21.7 ANCOVA 0.44 0.506
Knee Society Functional Activity Score 457 273 36.8 39.7 t-test 2.47 0.014 76.8 79.9 t-test 2.15 0.032 39.9 40.2 t-test 0.16 0.87 40.1 40 ANCOVA 0.00 0.999
Knee Society Expectation Score 510 319 14 14.3 t-test 3.30 0.001 10.7 10.4 t-test 1.49 0.136 -3.3 -4.0 t-test 2.73 0.007 -3.4 -3.8 ANCOVA 2.89 0.09
Open in a new tab

When adjusted for BMI only, there was a statistical difference favoring RP in several PROMs at two years, including PKIP overall (p=0.003), PKIP confidence (p=0.029), PKIP stability (p<0.013), PKIP modified activities (p<0.0001), OKS total score (p < 0.02), EQ5D3L current health score (<0.001), KSS functional activity (p<0.02), and KSS objective (p=0.021) scores (Table 4). In the same analysis, there was no demonstrated advantage to FB TKA when adjusted for BMI alone at two years. 

Table 4. Adjusted changes from baseline with BMI confounder only.

FB: fixed bearing; RP: rotating platform; CFB: change from baseline; PROMs: patient-reported outcome measures; PKIP: Patient Knee Implant Performance questionnaire [22]; KOOS: Knee Injury and Osteoarthritis Outcome Score [13]; OKS: Oxford Knee Score [12]; EQ-5D-3L: EuroQol five-dimensional questionnaire [20]; KSS: Knee Society Score [21]

KOOS was available freely for academic research use at the time of our study (prior to 2023); OKS was used under an license agreement with Oxford University Innovation Limited; KSS was used with permission from The Knee Society. EQ-5D-3L was used in accordance with licensing requirements from the EuroQol Research Foundation. 

Parameters FB (N) RP (N) Baseline mean 2-year mean Change from baseline mean Adjusted CFB for BMI
FB RP Test T Statistic p-value FB RP Test T Statistic p-value FB CFB RP CFB Test T Statistic p-value FB BMI Adjusted CFB RP BMI Adjusted CFB Test F Statistic p-value
PKIP Overall 783 581 27.5 28.4 t-test 1.21 0.227 74.2 75.4 t-test 1.16 0.248 46.7 47 t-test 0.27 0.786 46.3 47.5 ANCOVA 1.26 0.003
PKIP Confidence 789 585 3.5 3.8 t-test 2.45 0.014 8.1 8.3 t-test 1.98 0.048 4.6 4.5 t-test 0.34 0.732 4.5 4.6 ANCOVA 2.57 0.029
PKIP Stability 795 594 3.2 3.5 t-test 2.88 0.004 8.3 8.5 t-test 2.38 0.018 5.1 5 t-test 0.38 0.703 5 5.2 ANCOVA 3.45 0.013
PKIP Satisfaction 794 596 2.1 2.2 t-test 0.82 0.414 8.1 8.2 t-test 0.53 0.596 6 6 t-test 0.08 0.939 6 6.1 ANCOVA 0.30 0.149
PKIP Modify Activities 796 597 4.1 3.7 t-test 2.80 0.005 6.4 6.6 t-test 1.12 0.264 2.2 2.9 t-test 2.66 0.008 2.4 2.7 ANCOVA 1.81 < .0001
KOOS Function in Daily Living 796 598 47.1 50.3 t-test 3.33 0.001 87.1 88.4 t-test 1.61 0.108 40 38.1 t-test 1.74 0.082 38.9 39.5 ANCOVA 0.64 0.23
KOOS Pain 796 598 42.7 46.3 t-test 4.17 < .0001 87.2 88.1 t-test 1.10 0.273 44.5 41.8 t-test 2.47 0.014 43.2 43.5 ANCOVA 0.13 0.641
KOOS Symptoms 797 599 45.4 48.5 t-test 3.16 0.002 83.4 81.9 t-test 1.73 0.083 37.9 33.4 t-test 3.88 0 36.8 35 ANCOVA 4.95 0.148
KOOS Function in Sport and Recreation 765 554 18 19.1 t-test 0.98 0.327 62.5 67.8 t-test 3.39 0.001 44.4 48.7 t-test 2.51 0.012 44.1 49.2 ANCOVA 10.88 0.077
KOOS Quality of Life 797 599 22.3 23.9 t-test 1.84 0.066 74.8 74.1 t-test 0.54 0.586 52.5 50.1 t-test 1.66 0.097 51.9 50.9 ANCOVA 0.63 0.095
Oxford Knee Total Score (0-48) 797 592 21.6 23.7 t-test 5.04 < .0001 41 42 t-test 2.61 0.009 19.4 18.3 t-test 2.37 0.018 18.7 19.2 ANCOVA 1.95 0.033
EQ5D3L USA Index 797 592 0.6 0.7 t-test 2.48 0.013 0.9 0.9 t-test 3.20 0.001 0.2 0.2 t-test 0.05 0.958 0.2 0.2 ANCOVA 7.38 0.187
EQ5D3L Current Health 796 596 71.3 75.9 t-test 4.95 < .0001 79.6 82.3 t-test 3.13 0.002 8.3 6.4 t-test 1.82 0.069 6.8 8.3 ANCOVA 3.18 < .0001
Knee Society Objective Knee Score 494 303 42.5 45.7 t-test 2.21 0.027 90.2 90.9 t-test 0.74 0.457 47.8 45.3 t-test 1.54 0.124 46.6 47.2 ANCOVA 0.41 0.021
Knee Society Satisfaction Score 507 318 11.8 13 t-test 2.57 0.01 34.1 34.4 t-test 0.60 0.552 22.3 21.5 t-test 1.26 0.209 21.9 22.1 ANCOVA 0.07 0.238
Knee Society Functional Activity Score 457 273 36.8 39.7 t-test 2.47 0.014 76.8 79.9 t-test 2.15 0.032 39.9 40.2 t-test 0.16 0.87 39.3 41.3 ANCOVA 2.14 0.025
Knee Society Expectation Score 510 319 14 14.3 t-test 3.30 0.001 10.7 10.4 t-test 1.49 0.136 -3.3 -4 t-test 2.73 0.007 -3.4 -3.8 ANCOVA 2.29 0.181
Open in a new tab

The KSS PROM had a smaller sample size by 48.5% in the FB TKA and 58% in the RP TKA. This was due to two separate KSS versions (pre-2011 and post-2011) being used across centres, and analysis was only done on the newer KSS questionnaire. No other PROMs were affected.

Survival rate

Survival rates were comparable between groups at two years. There was a statistically significant difference between groups for revision for any reason; however, the majority of revisions on the RP TKA involved the exchange of the polyethylene insert only. Revisions for any reason at two years in the FB TKA were eight, with a survival rate of 99.1%, and 15 in the RP TKA with a survival rate of 97.9% (p-value: 0.0117) (Table 5). 

Table 5. Kaplan-Meier (KM) implant survivorship (removal of any TKA component for any reason).

TKA: total knee arthroplasty

  1 Year KM Survivorship (95% CI), N with Later Follow-up (Cumulative Revised) 2 Year KM Survivorship (95% CI), N with Later Follow-up (Cumulative Revised)
FB (N=959) 99.7% (99.0, 99.9), N = 925 (3 Revised) 99.1% (98.3, 99.6), N = 802 (8 Revised)
RP (N=759) 99.1% (98.0, 99.5), N = 720 (7 Revised) 97.9% (96.6, 98.7), N = 558 (15 Revised)
Open in a new tab

When comparing revisions involving the removal of the femoral and/or tibial component only, there was no statistically significant difference between groups. Revisions involving the removal of the femoral and/or the tibial component at two years in the FB TKA group were four, with a survival rate of 99.6%, and five in the RP TKA group, with a survival rate of 99.3% (p-value: 0.2973) (Table 6). 

Table 6. Kaplan-Meier (KM) implant survivorship estimates (removal of the tibial and/or femoral component only).

  1 Year KM Survivorship (95% CI), N with Later Follow-up (Cumulative Revised) 2 Year KM Survivorship (95% CI), N with Later Follow-up (Cumulative Revised)
FB (N=959) 99.9% (99.3, 100.0) N = 927 (1 Revised) 99.6% (98.8, 99.8) N = 806 (4 Revised)
RP (N=759) 99.9% (99.1, 100.0) N = 726 (1 Revised) 99.3% (98.3, 99.7) N = 567 (5 Revised)
Open in a new tab

Of the 15 revisions in the RP group, 10 involved isolated polyethylene insert exchange without removal of the femoral or tibial components. Indications for isolated polyethylene exchange included infection (n=3), crepitus (n=1), pain (n=3), and stiffness (n=3). In all cases, the polyethylene insert was replaced with the same bearing design, and no conversion between fixed-bearing and rotating platform systems occurred. Only two participants underwent revision for loosening, and both were in the RP TKA group. All revisions and reasons for revisions are presented in Table 7.

Table 7. Revision details.

b Subjects had a revision involving the removal of the insert prior to having another revision involving the removal of other components.

*P.F.C. SIGMA™ Knee System (DePuy Synthes, West Chester, Pennsylvania, United States); **ATTUNE™ Primary Knee System (DePuy Synthes, West Chester, Pennsylvania, United States)

FB: fixed bearings; RP: rotating platform bearings; F: femoral; T: tibial; I: insert; P: patella

FB (N=959) 8 Revisions
Revision Reason Product Time (years) at Revision Components removeda
Infection ATTUNE* 1.7 F, T, I, P
ATTUNE 0.6 P
Pain SIGMA** 1.8 I
SIGMA 1.8 F, T, I, P
SIGMA 1.7 P
SIGMA 1 I
1.5 F, T, I b
Stiffness ATTUNE 0.4 F, T, I
ATTUNE 0.9 I
RP (N=759) 15 Revisions
Revision Reason Product Time (years) at Revision Components removed
Bone Fracture ATTUNE 1.2 P
ATTUNE 0.3 F, T, I, P
Crepitus ATTUNE 0.9 I
Infection SIGMA 1.1 I
SIGMA 0.5 I
1.4 F, T, I, P b
SIGMA 1.5 F, T, I, P
ATTUNE 0 I
Loosening SIGMA 1.7 F, T, I, P
ATTUNE 1.9 F, T, I
Pain ATTUNE 1.8 I
ATTUNE 1.6 I
SIGMA 0.6 I
Stiffness ATTUNE 0.4 I
ATTUNE 0.4 I
ATTUNE 1.2 I
Open in a new tab

Discussion

The purpose of this study was to compare the functional outcomes of rotating platforms and fixed bearing TKA, including use of the novel PKIP PROM, in a post‑hoc pooled analysis of data from two large muti-centre studies. We found that both FB TKA and RP TKA provide excellent improvements in all PROMs over two years. In our initial analysis, adjusting for all confounders, we did not find a statistically significant difference between the rotational platform and fixed bearing prosthesis at baseline and two years across any PROM. When adjusting for BMI, we found that there were statistically significant differences between groups in favour of RP. 

Whilst many PROMs exist for use in TKA studies, PKIP is one of only eight PROMs that have undergone psychometric evaluation [17]. In a recent systematic review, PKIP is the only PROM that has been given an ‘adequate’ rating with regard to evidence in instrument development, whilst others were either ‘low’ or ‘indeterminate’ [23]. Our study is the first to compare RP TKA with FB TKA using PKIP; hence, there are no direct comparisons we can make with other studies. The PKIP allowed measurement of the TKA across several subscales. These included patients’ confidence, satisfaction, stability, and need for medication for activities. We found no difference in overall score or any subscale of PKIP between RP TKA and FB TKA at two years. The only statistically significant differences were seen when adjusting for only BMI.

Our findings concur with similar studies investigating these prostheses using conventional PROMs. Killen et al. found no difference in KSS between RP TKA and FB TKA after 12 years [24]. Theirs is the longest study to date comparing the two prostheses and included 132 patients contributing to 140 surgeries. Unfortunately, given the longer timeframe, there was a significant loss to follow-up (40%), with 20% of patients passing away and an additional 20% withdrawing from the study. The remaining 76 patients completed the KSS PROM. There was no statistically significant difference in the study at the two-year interval between RP TKA and FB TKA. Kalisvaart et al. conducted a randomised control trial (RCT) comparing 76 patients with RP TKA to 152 with FB TKA and found no difference at two or five years in KSS [25]. Bailey et al. conducted an RCT of 352 patients, of whom 163 underwent RP TKA, and 176 underwent FB TKA [26]. The researchers found no difference in OKS postoperatively at two years. Meta-analyses on this topic have found no differences between groups RP TKA and FB TKA. Most recently, a meta-analysis by Chen et al., looking at clinical and functional outcomes between MB TKA and FB TKA at a minimum 10-year follow-up, did not find any differences in KSS [27]. Wang et al. conducted a meta-analysis on 17 studies with a five-year minimum follow-up and did find some improvements in KSS in the MB TKA group; however, they noted significant heterogeneity between studies, including a lack of baseline KSS and small sample sizes [23]. 

We found statistically significant differences in RP TKA and FB TKA in several PROMs at two years when adjusted for BMI. However, these findings should be interpreted cautiously, as this analysis was exploratory and the study was not designed to evaluate BMI as a true effect modifier across predefined BMI subgroups. In our analysis, all differences favoured RP TKA. However, the magnitude of these differences was modest, and it remains uncertain whether they represent clinically meaningful improvements from a patient perspective. Given the increasing prevalence of obesity among patients undergoing TKA, understanding the potential influence of BMI on functional outcomes remains clinically relevant. A higher proportion of patients who are classified as obese (BMI of over 30 kg/m2) are undergoing TKA, yet outcomes in this population remain ambiguous [28]. Boyce et al. conducted a systematic review of studies comparing outcomes in morbidly obese and non-obese patients [29]. They found that a higher BMI was associated with a higher complication rate for TKA but reported similar improvements in functional scores. O’Neill et al. conducted an analysis at a single centre on 2,180 patients undergoing primary TKA and found no differences in functional scores between normal weight and overweight groups at six months postoperatively [30]. Their subgroup analysis, however, did suggest worsening functional scores with increasing BMI. Whilst a large study, this was a retrospective study with a limited duration of follow-up. No study to our knowledge has compared outcomes between RP TKA and FB TKA in elevated BMI populations. Our results suggest functional differences between RP and FB TKA may exist at higher BMI; however, we did not perform analysis on sub-groups of different BMI, and this may be the focus of a future project.

Survival rates were excellent in both groups at the two-year follow-up. Revision of either femoral or tibial components remained low in both RP TKA and FB TKA groups at two years. There was a statistically significant difference between FB TKA and RP TKA for the total number of revisions; however, no difference was found when comparing revisions involving the removal of a femoral or tibial implant. Only two patients underwent revision for loosening as the primary indication for revision, and both were in the RP TKA cohort. These findings are similar to other studies investigating survival rates of RP TKA. Meftah et al. conducted a long-term follow-up of over 10 years in 117 patients with RP TKA and found the 10-year survival was 97.7% when revision for any reason was an endpoint [31]. Our study was not able to address the issue of long-term wear, which is the index theoretical advantage of the RP TKA, as the scope of observation was too limited to confidently provide any meaningful conclusions. 

The current study involved a modern implant, the ATTUNE Knee System, which has shown encouraging results for its use in TKA at two years, with no signs of increased revision rates when compared to a well-established implant, the P.F.C. Sigma [32]. This is further supported by findings in the Australian and American National Joint Registries [2,33].

Our study is the largest data set, to our knowledge, comparing RP TKA and FB using prospectively collected data. Our analysis is also the first to compare RP TKA and FB TKA using the new PKIP PROM and adjust for BMI between these two groups. Additionally, the international pool of patients increases the generalisability of our results.

Limitations

There are limitations in our study. In addition to the short timeframe, we did not use any objective tools in our analysis, such as radiographic data, gait analyses, or range of motion. Our KSS PROM dataset was reduced as there was a different version of the KSS PROM used across centres, and analysis was conducted only on the updated KSS questionnaire. This did not affect other PROMs. 

Another limitation of our study is the lack of randomisation in the recruitment process, which may contribute to selection bias and residual confounding. In particular, differences in recruitment across participating centres and surgeons may have contributed to baseline differences between RP TKA and FB TKA cohorts prior to statistical adjustment. Although propensity score adjustment was used to account for measured confounders, unmeasured factors may still have influenced outcomes.

Additionally, the BMI-adjusted analysis should be interpreted cautiously. While differences favouring RP TKA were observed when adjusting for BMI alone, this analysis was exploratory, and the study was not designed to evaluate BMI as a true effect modifier across predefined BMI subgroups. Further investigation in studies specifically designed to examine BMI-related outcomes would be required to confirm these findings.

Conclusions

The study findings demonstrate similar improvements in the short-term in both RP TKA and FP TKA across all PROMs. Patients reported similar scores in satisfaction, confidence, mobility, and stability in the PKIP PROM for both RP and FB groups. At two years when adjusting for BMI using the 0.05 significance threshold, the following PROMs favoured RP TKA: PKIP Overall, PKIP Confidence, PKIP Stability, PKIP Modified Activities, Oxford Knee Score, Knee Society Objective Knee Score, and Knee Society Functional Activity Score, warranting further investigation.

We did not identify any disadvantage in short-term patient-reported functional outcomes associated with RP TKA at two years. Our findings indicate that, from a patient-reported performance perspective, RP TKA performs comparably to FB TKA in the early postoperative period. Long-term follow-up data from RCTs are needed to establish with certainty any difference between RP and FB designs. 

Disclosures

Human subjects: Informed consent for treatment and open access publication was obtained or waived by all participants in this study. Western Institutional Review Board issued approval WIRB PRO NUM 20121213/20111599.

Animal subjects: All authors have confirmed that this study did not involve animal subjects or tissue.

Conflicts of interest: In compliance with the ICMJE uniform disclosure form, all authors declare the following:

Payment/services info: All authors have declared that no financial support was received from any organization for the submitted work.

Financial relationships: Piers Yates declare(s) royalties from Matortho, Corin. Piers Yates declare(s) Paid presentations from Depuy Synthes. Piers Yates declare(s) personal fees from Matortho, Carinal, Depuy Synthes. Piers Yates declare(s) stock/stock options from 360. Piers Yates declare(s) Research Support from Depuy Synthes, Corin. Piers Yates declare(s) Medical/Orthopaedic Publications Editorial Governing Board from Hip International Journal. Thierry Bernard declare(s) employment from Depuy Synthes.

Other relationships: All authors have declared that there are no other relationships or activities that could appear to have influenced the submitted work.

Author Contributions

Concept and design:  Naseer Mohammed Abdul, Piers Yates

Acquisition, analysis, or interpretation of data:  Naseer Mohammed Abdul, Piers Yates, Thierry Bernard

Drafting of the manuscript:  Naseer Mohammed Abdul

Critical review of the manuscript for important intellectual content:  Naseer Mohammed Abdul, Piers Yates, Thierry Bernard

References

  • 1.Lewis PL, Gill DR, McAuliffe MJ, et al. 2024 Annual Report: Hip, Knee and Shoulder Arthroplasty. Adelaide, Australia: Australian Orthopaedic Association National Joint Replacement Registry; 2024. Hip, Knee and Shoulder Arthroplasty: 2024 Annual Report, Australian Orthopaedic Association National Joint Replacement Registry. AOA: Adelaide, South Australia. 2024. [Google Scholar]
  • 2.Hip, Knee & Shoulder Arthroplasty: Annual Report 2021. Adelaide, Australia: Australian Orthopaedic Association National Joint Replacement Registry; 2021. Australian Orthopaedic Association National Joint Replacement Registry Annual Report 2021, https://aoanjrr.sahmri.com/annual-reports-2021. [Google Scholar]
  • 3.The prevalence and predictors of patient dissatisfaction 5-years following primary total knee arthroplasty. Ayers DC, Yousef M, Zheng H, Yang W, Franklin PD. J Arthroplasty. 2022;37:0–8. doi: 10.1016/j.arth.2022.02.077. [DOI] [PubMed] [Google Scholar]
  • 4.Fixed bearings versus rotating platforms in total knee arthroplasty. White PB, Ranawat AS, Ranawat CS. J Knee Surg. 2015;28:358–362. doi: 10.1055/s-0035-1550338. [DOI] [PubMed] [Google Scholar]
  • 5.Migration and clinical outcome of mobile-bearing versus fixed-bearing single-radius total knee arthroplasty. Van Hamersveld KT, Marang-Van De Mheen PJ, Van Der Heide HJ, Van Der Linden-Van Der Zwaag HM, Valstar ER, Nelissen RG. Acta Orthop. 2018;89:190–196. doi: 10.1080/17453674.2018.1429108. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Wear-simulation analysis of rotating-platform mobile-bearing knees. Fisher J, McEwen H, Tipper J, Jennings L, Farrar R, Stone M, Ingham E. https://www.proquest.com/docview/203915028?sourcetype=Scholarly%20Journals. Orthopedics. 2006;29:36–41. [PubMed] [Google Scholar]
  • 7.No difference in terms of radiostereometric analysis between fixed- and mobile-bearing total knee arthroplasty: a randomized, single-blind, controlled trial. Schotanus MG, Pilot P, Kaptein BL, Draijer WF, Tilman PB, Vos R, Kort NP. Knee Surg Sports Traumatol Arthrosc. 2017;25:2978–2985. doi: 10.1007/s00167-016-4138-6. [DOI] [PubMed] [Google Scholar]
  • 8.A randomized controlled trial of fixed- versus mobile-bearing total knee arthroplasty: a follow-up at a mean of ten years. Abdel MP, Tibbo ME, Stuart MJ, Trousdale RT, Hanssen AD, Pagnano MW. Bone Joint J. 2018;100-B:925–929. doi: 10.1302/0301-620X.100B7.BJJ-2017-1473.R1. [DOI] [PubMed] [Google Scholar]
  • 9.Comparison of high-flexion fixed-bearing and high-flexion mobile-bearing total knee arthroplasties-a prospective randomized study. Kim YH, Park JW, Kim JS. J Arthroplasty. 2018;33:130–135. doi: 10.1016/j.arth.2017.07.025. [DOI] [PubMed] [Google Scholar]
  • 10.Mobile bearing total knee arthroplasty is associated with an increased risk of mechanical loosening: a follow-up analysis of the American Joint Replacement Registry. Hegde V, Pelt CE, De A, Zaniletti I, Kagan R. J Arthroplasty. 2025;40:2900–2909. doi: 10.1016/j.arth.2025.04.091. [DOI] [PubMed] [Google Scholar]
  • 11.Reliability and validity of the EQ-5D-5L compared to the EQ-5D-3L in patients with osteoarthritis referred for hip and knee replacement. Conner-Spady BL, Marshall DA, Bohm E, Dunbar MJ, Loucks L, Al Khudairy A, Noseworthy TW. Qual Life Res. 2015;24:1775–1784. doi: 10.1007/s11136-014-0910-6. [DOI] [PubMed] [Google Scholar]
  • 12.The Oxford Knee Score; problems and pitfalls. Whitehouse SL, Blom AW, Taylor AH, Pattison GT, Bannister GC. Knee. 2005;12:287–291. doi: 10.1016/j.knee.2004.11.005. [DOI] [PubMed] [Google Scholar]
  • 13.Knee Injury and Osteoarthritis Outcome Score (KOOS)--development of a self-administered outcome measure. Roos EM, Roos HP, Lohmander LS, Ekdahl C, Beynnon BD. http://10.1016/0168-8510(90)90421-9. J Orthop Sports Phys Ther. 1998;28:88–96. doi: 10.2519/jospt.1998.28.2.88. [DOI] [PubMed] [Google Scholar]
  • 14.The reliability of the American Knee Society Score. Liow RY, Walker K, Wajid MA, Bedi G, Lennox CM. Acta Orthop Scand. 2000;71:603–608. doi: 10.1080/000164700317362244. [DOI] [PubMed] [Google Scholar]
  • 15.What makes patients aware of their artificial knee joint? Loth FL, Liebensteiner MC, Giesinger JM, Giesinger K, Bliem HR, Holzner B. BMC Musculoskelet Disord. 2018;19:5. doi: 10.1186/s12891-017-1923-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Questionnaire on the perceptions of patients about total hip replacement. Dawson J, Fitzpatrick R, Carr A, Murray D. https://pubmed.ncbi.nlm.nih.gov/8666621/ J Bone Joint Surg Br. 1996;78:185–190. [PubMed] [Google Scholar]
  • 17.Psychometric evaluation of the Patient's Knee Implant Performance questionnaire. Coles T, Williams V, Dwyer K, Mordin M. Value Health. 2018;21:1305–1312. doi: 10.1016/j.jval.2018.05.006. [DOI] [PubMed] [Google Scholar]
  • 18.ClinicalTrails.gov: 10003 PRO Current Products (10003) [ May; 2026 ]. 2019. https://clinicaltrials.gov/study/NCT01497730 https://clinicaltrials.gov/study/NCT01497730
  • 19.ClinicalTrails.gov: Prospective, multiconfiguration study to assess functional performance of primary total knee arthroplasty system (10004) [ May; 2026 ]. 2019. https://clinicaltrials.gov/study/NCT01746524 https://clinicaltrials.gov/study/NCT01746524
  • 20.A comparison of EQ-5D-3L population norms in Queensland, Australia, estimated using utility value sets from Australia, the UK and USA. Clemens S, Begum N, Harper C, Whitty JA, Scuffham PA. Qual Life Res. 2014;23:2375–2381. doi: 10.1007/s11136-014-0676-x. [DOI] [PubMed] [Google Scholar]
  • 21.The new Knee Society Knee Scoring System. Scuderi GR, Bourne RB, Noble PC, Benjamin JB, Lonner JH, Scott WN. Clin Orthop Relat Res. 2012;470:3–19. doi: 10.1007/s11999-011-2135-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Psychometric evaluation of the Patient's Knee Implant Performance questionnaire. Value Health. 2024;27:820. doi: 10.1016/j.jval.2024.05.001. [DOI] [PubMed] [Google Scholar]
  • 23.Patient-reported outcome measures used in patients undergoing total knee arthroplasty. Wang Y, Yin M, Zhu S, Chen X, Zhou H, Qian W. Bone Joint Res. 2021;10:203–217. doi: 10.1302/2046-3758.103.BJR-2020-0268.R1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Minimum twelve-year follow-up of fixed- vs mobile-bearing total knee arthroplasty: double blinded randomized trial. Killen CJ, Murphy MP, Hopkinson WJ, Harrington MA, Adams WH, Rees HW. J Clin Orthop Trauma. 2020;11:154–159. doi: 10.1016/j.jcot.2019.03.019. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Randomized clinical trial of rotating-platform and fixed-bearing total knee arthroplasty: no clinically detectable differences at five years. Kalisvaart MM, Pagnano MW, Trousdale RT, Stuart MJ, Hanssen AD. J Bone Joint Surg Am. 2012;94:481–489. doi: 10.2106/JBJS.K.00315. [DOI] [PubMed] [Google Scholar]
  • 26.No clinical difference between fixed- and mobile-bearing cruciate-retaining total knee arthroplasty: a prospective randomized study. Bailey O, Ferguson K, Crawfurd E, et al. Knee Surg Sports Traumatol Arthrosc. 2015;23:1653–1659. doi: 10.1007/s00167-014-2877-9. [DOI] [PubMed] [Google Scholar]
  • 27.Mobile bearing versus fixed bearing for total knee arthroplasty: meta-analysis of randomized controlled trials at minimum 10-year follow-up. Chen P, Huang L, Zhang D, Zhang X, Ma Y, Wang Q. J Knee Surg. 2022;35:135–144. doi: 10.1055/s-0040-1713356. [DOI] [PubMed] [Google Scholar]
  • 28.Do TKAs in patients with higher BMI take longer, and is the difference associated with surgeon volume? A large-database study from a national arthroplasty registry. Quayle J, Klasan A, Frampton C, Young SW. Clin Orthop Relat Res. 2022;480:714–721. doi: 10.1097/CORR.0000000000002047. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.The outcomes of total knee arthroplasty in morbidly obese patients: a systematic review of the literature. Boyce L, Prasad A, Barrett M, Dawson-Bowling S, Millington S, Hanna SA, Achan P. Arch Orthop Trauma Surg. 2019;139:553–560. doi: 10.1007/s00402-019-03127-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Effect of body mass index on functional outcome in primary total knee arthroplasty - a single institution analysis of 2180 primary total knee replacements. O'Neill SC, Butler JS, Daly A, Lui DF, Kenny P. World J Orthop. 2016;7:664–669. doi: 10.5312/wjo.v7.i10.664. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Ten-year follow-up of a rotating-platform, posterior-stabilized total knee arthroplasty. Meftah M, Ranawat AS, Ranawat CS. J Bone Joint Surg Am. 2012;94:426–432. doi: 10.2106/JBJS.K.00152. [DOI] [PubMed] [Google Scholar]
  • 32.Early experience with the ATTUNE Total Knee Replacement System. Giaretta S, Berti M, Micheloni GM, Ceccato A, Marangoni F, Momoli A. Acta Biomed. 2019;90:98–103. doi: 10.23750/abm.v90i12-S.8997. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.The Eighth Annual Report of the AJRR on Hip and Knee Arthroplasty: Annual Report 2021. Rosemont, IL: American Academy of Orthopaedic Surgeons (AAOS); 2021. Surgeons AAoO. American Joint Replacement Registry (AJRR): 2021 Annual Report. [Google Scholar]

Articles from Cureus are provided here courtesy of Cureus Inc.

RESOURCES