Secondary Logo

Journal Logo


Do Fixed or Mobile Bearing Implants Have Better Survivorship in Medial Unicompartmental Knee Arthroplasty? A Study From the Australian Orthopaedic Association National Joint Replacement Registry

Kannan, Arun MS1; Lewis, Peter L. MBBS, FRACS(Orth), FAOrthA2,3; Dyer, Chelsea B.Math&Comp Sc4; Jiranek, William A. MD, FACS5; McMahon, Stephen FRACS(Orth), FAOrthA6

Author Information
Clinical Orthopaedics and Related Research: July 2021 - Volume 479 - Issue 7 - p 1548-1558
doi: 10.1097/CORR.0000000000001698



Unicompartmental knee arthroplasty (UKA) has seen a resurgence in popularity recently [24, 32] and constitutes 6% of annual knee replacement procedures performed in Australia [1]. Although the indications for UKA and its surgical techniques continue to evolve in an effort to improve the survivorship, it is unclear whether one bearing design philosophy is more durable than any other. There is support for two fundamentally different bearing designs, namely fixed bearing UKA, which allow a single articulation between the femoral component and the superior surface of the tibial component, and mobile bearing UKA, which has a similar femorotibial articulation but in contradistinction, also allows the polyethylene bearing to slide and rotate over a fixed metal tibial baseplate [2, 3].

Marmor [21], using a fixed bearing design, was one of the earliest to report successful use of UKA. Currently, fixed bearing UKA is a heterogeneous group consisting of three different tibial component designs [2]. These can be either UKA with a tibial component made entirely of polyethylene (all-polyethylene UKA), or those with a metal tibial baseplate (fixed metal-backed UKA), which are metal-backed tibial components with modular polyethylene inserts (fixed modular UKA), or metal-backed tibial components with molded polyethylene inserts attached (fixed molded metal-backed UKA). As thin polyethylene inserts used with metal tibial baseplates were implicated in wear, delamination, and deformation [8, 22, 26], all-polyethylene designs were initially preferred [13]. The all-polyethylene UKA has a theoretical advantage of allowing minimal tibial resection while maximizing polyethylene thickness. However, several studies have shown that all-polyethylene UKA has a higher revision rate than fixed, metal-backed UKA [9, 16, 18]. This has been attributed to the transmission of higher stresses to the proximal tibia with all-polyethylene tibial components resulting in loosening, subsidence, and tibial fracture [31, 33].

The mobile bearing UKA advocated by the Oxford group was introduced in 1982 [10] with the theoretical advantages of increased contact area, reduced contact stress, and reduced wear rates [11]. This design continues to be widely used [23, 27, 30]. Several studies comparing clinical outcomes and revision rates between fixed bearing and mobile bearing UKA have found no difference between the two designs [6, 7, 17, 25, 28, 29, 34]. Surprisingly, in the United Kingdom, widely considered to be the home of mobile bearing UKA, the proportion of fixed bearing UKAs compared with mobile bearing UKAs has steadily risen and accounted for nearly half of the UKAs implanted in 2018 [24].

Studies comparing fixed bearing and mobile bearing UKAs have been unable to detect a difference, possibly because of insufficient sample size [7, 25, 28, 34]. In a retrospective analysis by Whittaker et al. [34], the 5-year cumulative survival rate was 88% for mobile bearing UKA compared with 96% for fixed bearing UKA but without a difference between the groups, which suggested a higher point estimate but no actual difference, perhaps because of insufficient power. Emerson et al. [7] showed a higher revision risk for tibial component failure with fixed bearing UKA but similar overall revision rates compared with mobile bearing UKA. Systematic reviews and meta-analyses that have compared fixed and mobile bearing UKA have usually combined all-polyethylene and metal-backed UKA in the same group and then compared this combined group with mobile UKA. These studies may have been confounded by possible differences in survivorship between the different designs of fixed bearing UKA [6, 17, 29].

We therefore selected the four contemporary designs of UKA (mobile bearing, fixed modular, all-polyethylene, and fixed molded metal-backed) using data from the Australian Orthopaedic Association National Joint Replacement Registry (AOANJRR) to ask that when used for the treatment of medial compartment osteoarthritis: (1) How do the different designs of unicompartmental knees compare with survivorship as measured by the cumulative percentage revision (CPR)? (2) Is there a difference in revision rate between designs as a function of patient sex or age? (3) Do the reasons for revision differ, and what types of revision procedures are performed when these UKA are revised?

Patients and Methods

We used data from the AOANJRR (September 1999-December 2018). The AOANJRR commenced data collection on September 1, 1999, achieving complete national implementation by mid-2002. Since then, the registry has collected data on almost 100% of hip and knee arthroplasties performed in Australia [1]. These data are externally validated against patient-level data provided by all Australian state and territory health departments. A sequential, multilevel matching process is used to identify any missing data, which are subsequently obtained by follow-up with the relevant hospital. Each month, in addition to internal validation and data quality checks, all primary procedures are linked to any subsequent revision involving the same patient, joint, and side. Data are also matched biannually to the Australian National Death Index data to identify patients who have died.

There were 56,628 unicompartmental knees recorded during the study period (Supplementary Fig. 1; Supplemental Digital Content 1, Of these, 560 procedures were excluded due to a primary diagnosis other than osteoarthritis, and a further 5657 as these were lateral or patellofemoral UKA. An additional 31 procedures were excluded as the bearing type was unknown. There were 50,380 medial UKA procedures undertaken for osteoarthritis included in the analysis. The UKA procedures included 40% (20,208 of 50,380) mobile bearing UKA, 35% (17,822 of 50,380) fixed modular UKA, 23% (11,461 of 50,380) all-polyethylene UKA, and 2% (889 of 50,380) fixed molded metal-backed UKA (Table 1).

Fig. 1:
The CPR of primary medial unicompartmental knee replacement by bearing type (primary diagnosis of osteoarthritis). A color image accompanies the online version of this article.
Table 1. - Primary medial unicompartmental knee arthroplasty by bearing type, prosthesis combination, age, sex, and CPR (primary diagnosis osteoarthritis)
Bearing type Prosthesis combination Total Revised 5-year CPR (95% CI) 10-year CPR (95% CI) Female, % (n) Mean age in years ± SD
Mobile bearings BalanSys® Uni/BalanSys Uni Convex, Mathys Orthopaedics 53 11 7.5 (2.9-19) 25 (14-42) 64 (34) 60 ± 8
BalanSys Uni/BalanSys Uni Mobile, Mathys Orthopaedics 198 51 15 (11-21) 22 (17-28) 42 (84) 63 ± 9
Oxford®/Oxford, Zimmer Biomet 19,101 2481 7.9 (7.5-8.3) 14 (14-15) 48 (9095) 66 ± 10
Preservation®/Preservation® Mobile, DePuy Orthopaedics 348 119 19 (15-23) 27 (22-32) 41 (142) 61 ± 10
Uniglide™/Uniglide, Corin 484 95 12 (9.0-15) 18 (15-22) 44 (211) 64 ± 10
Other (2) 24 8 17 (6.9-40) 22 (9.7-45)
Total 20,208 2765
Fixed molded metal-backed Allegretto® Uni/Allegretto Uni, Zimmer Biomet 769 109 7.6 (5.9-9.7) 12 (10-15) 54 (416) 67 ± 10
Endo-Model® Sled/ Endo-Model Sled, Waldemar LINK GmbH Co 92 18 10 (5.3-18) 21 (13-32) 46 (42) 65 ± 10
Other (3) 28 1 0.0 (0.0-0.0) 5.9 (0.9-35)
Total 889 128
All polyethylene Allegretto Uni/ZUK, Zimmer Biomet 75 8 9.5 (4.7-19) 32 (24) 66 ± 9
Eius™/Eius, Stryker 132 40 18 (12-25) 23 (16-31) 63 (83) 60 ± 11
Endo-Model Sled/Endo-Model Sled, Waldemar LINK GmbH Co 736 87 6.6 (4.9-8.7) 12 (9.6-15) 43 (313) 70 ± 11
Freedom PKR/Active/Freedom PKR/Active, Samo 1337 348 14 (12-16) 27 (25-30) 40 (534) 64 ± 10
GRU™/GRU, Global Orthopedic Technology 1856 265 6.3 (5.3-7.5) 13 (12-15) 36 (676) 67 ± 10
Genesis™/Genesis, Smith & Nephew 1550 280 11 (9.7-13) 16 (14-18) 47 (725) 66 ± 10
Genesis/Genesis™ II, Smith & Nephew 70 7 1.4 (0.2-9.7) 9.8 (4.5-21) 44 (31) 69 ± 11
Genesis/Journey™ Uni All Poly, Smith & Nephew 60 14 14 (7.0-25) 24 (15-38) 45 (27) 64 ± 10
HLS Uni Evolution™/HLS Uni Evolution, Tornier 104 22 12 (6.9-20) 21 (15-31) 59 (61) 64 ± 9
Journey™ Uni/ Journey Uni All Poly, Smith & Nephew 277 23 9.0 (5.9-14) 42 (116) 64 ± 9
M/G®/M/G, Zimmer Biomet 387 70 9.2 (6.7-13) 14 (11-18) 51 (198) 67 ± 9
PFC® Sigma®/PFC Sigma, DePuy Synthes 104 23 7.8 (4.0-15) 15 (9.3-24) 54 (56) 62 ± 10
Preservation/Preservation® Fixed, DePuy Orthopaedics 1957 380 10 (8.9-12) 16 (15-18) 46 (903) 65 ± 10
Repicci II®/Repicci II, Zimmer Biomet 2111 493 7.1 (6.0-8.3) 18 (16-20) 39 (814) 66 ± 10
Uniglide/Uniglide, Corin 90 25 17 (11-27) 27 (19-38) 47 (42) 59 ± 12
Zimmer® Unicompartmental High Flex Knee (ZUK)/ZUK, Zimmer Biomet 532 45 6.7 (4.8-9.3) 9.8 (7.2-13) 55 (290) 67 ± 10
Other (6) 83 15 14 (7.7-23) 18 (11-28)
Total 11,461 2145
Fixed modular Allegretto Uni/ZUK, Zimmer Biomet 151 3 46 (70) 69 ± 8
BalanSys® Uni/BalanSys Uni Fixed, Mathys Orthopaedics 498 29 4.8 (3.1-7.6) 9.0 (6.0-13) 44 (218) 70 ± 10
GMK®-UNI/GMK-UNI, Medacta 148 27 24 (17-34) 51 (75) 61 ± 10
Genesis/Genesis, Smith & Nephew 143 31 9.3 (5.5-16) 25 (17-36) 41 (59) 65 ± 9
Genesis/Journey Uni (v1), Smith & Nephew 151 20 8.2 (4.7-14.0) 50 (75) 65 ± 10
Journey Uni/Journey Uni (v1), Smith & Nephew 51 6 5.9 (1.9-17) 51 (26) 65 ± 9
Journey Uni/Journey Uni (v2), Smith & Nephew 667 19 4.2 (2.6-6.7) 48 (320) 64 ± 9
M/G/M/G, Zimmer Biomet 1662 227 6.0 (4.9-7.3) 10 (8.9-12) 48 (804) 66±10
Preservation/Preservation Fixed, DePuy Orthopaedics 191 32 5.3 (2.9-9.5) 14 (9.5-20) 46 (88) 62 ± 9
Repicci II/Repicci II, Biomet 700 124 8.7 (6.8-11) 17 (14-21) 46 (320) 65 ± 10
Restoris® MCK/Restoris MCK, Stryker 2797 47 43 (1192) 66 ± 9
Sigma HP/Sigma HP, DePuy Orthopaedics 1074 39 4.9 (3.5-6.8) 42 (452) 64 ± 10
Triathlon® PKR/Triathlon PKR, Stryker 306 18 7.1 (4.3-12) 43 (132) 63 ± 9
Unix/Unix, Stryker 2231 244 6.9 (5.9-8.1) 12 (1-13) 46 (1017) 64 ± 10
ZUK/ZUK, Zimmer Biomet 6887 328 4.6 (4.1-5.2) 8.5 (7.5-9.6) 43 (2989) 65 ± 10
Other (11) 165 14 3.8 (1.6-8.9) 12 (6.8-21)
Total 17,822 1208
Only prostheses with more 50 procedures have been listed; CPR = cumulative percentage revision.

The study population had a mean age of 65 ± 9 years, and males comprised 55% (27,496 of 50,380) of the cohort. There were similar sex proportions and age distributions for each of the bearing groups (Table 1).

Primary and Secondary Study Outcomes

The primary outcome measure was CPR for the study period. We used hazard ratios, adjusted for sex and age, to compare revision rates between the groups. The secondary goals of the study were (1) to assess the difference in revision rates as a function of sex or age, (2) to compare the groups in terms of reasons for revision, and (3) to identify modes of revision. Males and females were analyzed separately to evaluate revision rates as a function of sex. We stratified the cohort into age groups of younger than 65 years (47% [23,930 of 50,380]) and 65 years and older (53% [26,450 of 50,380]) to compare revision rates in these age groups. For each major cause of revision, we compared the revision rates among the bearing groups. We considered the revision mode, and we studied whether there was a difference between the bearing groups in terms of revision to TKA, as opposed to a simpler revision such as an isolated polyethylene exchange.

Ethical Approval

Ethical approval was not sought for the present study.

Statistical Analysis

We used Kaplan-Meier estimates of survivorship to report the time to revision, with censoring at the time of death and closure of the dataset at the end of December 2018. The unadjusted CPR, with 95% confidence intervals, was calculated using unadjusted point wise Greenwood estimates. Sex- and age-adjusted hazard ratios were calculated from Cox proportional hazard models to compare the revision rate between groups. The assumption of proportional hazards was checked analytically for each model. If the interaction between the predictor and the log of time was statistically significant in the standard Cox model, then a time varying model was estimated. Time points were selected based on the greatest change in hazard, weighted by a function of events. Time points were iteratively chosen until the assumption of proportionality was met and HRs were calculated for each selected time. For the current study, if no period was specified, the HR was calculated over the entire follow-up. All tests were two-tailed at 5% levels of significance. Statistical analysis was performed using SAS software version 9.4 (SAS Institute Inc).


Comparison of Revision Rate Among Bearing Groups

At 15 years, fixed modular UKA had a CPR of 16% (95% CI 15% to 17%). In comparison, the CPR was 23% (95% CI 22% to 24%) for mobile bearing UKA, 26% (95% CI 24% to 27%) for all-polyethylene UKA, and 20% (95% CI 16% to 24%) for fixed molded metal-backed UKA. After adjustment for sex and age, mobile bearing UKA (HR 1.5 [95% CI 1.4 to 1.6]; p < 0.001) and fixed molded metal-backed UKA (HR 1.3 [95% CI 1.1 to 1.6]; p = 0.003) had higher revision rates than fixed modular UKA over the study period. When we compared fixed modular UKA with all-polyethylene UKA, there was no difference in the revision rate between 3 to 6 months and 9 months to 1 year. At all other points, fixed modular UKA had a lower revision rate than all-polyethylene UKA. The all-polyethylene group had the highest revision rate compared with the three other UKA groups (Fig. 1).

Differences in Revision Rate as a Function of Sex or Age

When we evaluated revision rates as a function of sex using HRs, we saw findings similar to the overall revision rates with the same comparisons in males, but in females, comparisons did not show a difference when we compared the fixed molded metal-backed and the fixed modular groups, and the difference between all-polyethylene and fixed modular groups was generally observed through the entire period (Fig. 2).

Fig. 2:
A-B (A) The CPR of primary medial unicompartmental knee replacement in males by bearing type (primary diagnosis osteoarthritis). (B) This graph shows the CPR of primary medial unicompartmental knee replacement in females by bearing type (primary diagnosis of osteoarthritis). A color image accompanies the online version of this article.

Fixed modular UKA, in comparison to all-polyethylene and mobile bearing UKA, had the lowest revision rates in both age groups (patients younger than 65 years and those 65 years or older). In patients younger than 65 years, fixed modular UKA also had a lower revision rate compared with fixed molded metal-backed UKA (Fig. 3). In patients 65 years or older, there was no difference in the comparison of fixed modular to fixed molded metal-backed UKA. In this age group, all-polyethylene UKA and mobile bearing UKA had similar revision rates (HR 1 [95% CI 0.9 to 1.1]; p = 0.53) (Fig. 4).

Fig. 3:
The CPR of primary medial unicompartmental knee replacement in patients aged younger than 65 years by bearing type (primary diagnosis of osteoarthritis). A color image accompanies the online version of this article.
Fig. 4:
The CPR of primary medial unicompartmental knee replacement in patients aged 65 years and older by bearing type (primary diagnosis of osteoarthritis). A color image accompanies the online version of this article.

Comparison of Revision Rates by Reasons for Revision and Mode of Revision

Component loosening and disease progression were the leading causes of revision and accounted for most revisions in all UKA groups (Table 2). Bearing dislocation was unique to the mobile bearing UKAs. Bearing dislocation led to the revision of 0.7% (140 of 20,208) of mobile bearing UKAs (percentage of primaries implanted) and this comprised 5.1% (140 of 2765) of revisions in the mobile bearing group (percentage of revisions). The sex- and age-adjusted HR for loosening with mobile bearing UKA was higher than fixed modular UKA (1.7 [95% CI 1.5 to 1.9; p < 0.001]) (Supplementary Fig. 2; Supplemental Digital Content 2, Similarly, the sex- and age-adjusted HR for progression of disease was higher for all-polyethylene UKA compared with fixed modular UKA (HR 1.4 [95% CI 1.3 to 1.6]; p < 0.001). The same reason for revision was higher for mobile bearing UKA compared with fixed modular UKA from 8 years onward (8 to 12 years: HR 1.4 [95% CI 1.2 to 1.7]; p < 0.001; 12 or more years: HR 1.9 [95% CI 1.5 to 2.3]; p < 0.001) (Supplementary Fig. 3; Supplemental Digital Content 3, The percentage of primaries revised for loosening was highest at 8.4% (965 of 11461) in the all-polyethylene UKA group (Table 2). Revisions for tibial insert wear accounted for the revision of 0.2% (37 of 20,208) of the mobile bearings, 0.2% (24 of 11,461) of all-polyethylene, 0.2% (33 of 17,822) of the fixed modular, and 0.7% (6 of 889) of the fixed molded metal-backed designs.

Table 2. - The 10 most common revision diagnoses for primary medial unicompartmental knee replacement by bearing type (performed for the primary diagnosis of osteoarthritis)
Mobile, primary (n = 20,208) Fixed molded metal-backed, primary (n = 889)
Number of revision procedures 13 (2716) 14 (128)
Progression of disease 4 (865) 4 (39)
Pain 1 (220) 2 (19)
Bearing dislocation 0.7 (140)
Infection 0.5 (111) 0.4 (4)
Lysis 0.2 (41) 0.1 (1)
Fracture 0.4 (74) 0.1 (1)
Instability 0.2 (41) 0.2 (2)
Wear tibial insert 0.2 (37) 0.7 (6)
Malalignment 0.2 (32) 0.1 (1)
Other 0.6 (154) 1.6 (14)
Fixed all-polyethylene, primary (n = 11,461) Fixed modular, primary (n = 17,822)
Number of revision procedures 19 (2126) 7 (1199)
Progression of disease 6 (684) 2 (402)
Pain 2 (168) 0.6 (109)
Bearing dislocation 0.0 (1)
Infection 0.5 (62) 0.4 (70)
Lysis 0.7 (82) 0.2 (39)
Fracture 0.3 (31) 0.2 (34)
Instability 0.1 (13) 0.1 (20)
Wear tibial insert 0.2 (24) 0.2 (33)
Malalignment 0.1 (13) 0.1 (18)
Other 0.7 (84) 0.4 (70)
Data are presented as % (n).

Most UKA revisions (87% [5368 of 6169]) were revisions to a TKA. Revision of a polyethylene insert alone was undertaken for 2% (320 of 20,208) of mobile bearing UKA and accounted for 12% (320 of 2716) of revisions in this group (Supplementary Table 1; Supplemental Digital Content 4, The sex- and age-adjusted HR for revision to a TKA over the study period for mobile bearing UKA was higher than that of fixed modular UKA (1.4 [95% CI 1.3 to 1.5]; p < 0.001) (Supplementary Fig. 4; Supplemental Digital Content 5,


Medial compartment UKA, with its promise of being a less-invasive alternative to TKA in appropriate patients, has continued to appeal to surgeons and patients. Medial UKA has four fundamentally different designs, namely, mobile bearing, fixed bearing with an all-polyethylene tibial component, and fixed bearing with metal-backed tibial component with either a modular polyethylene insert or as an integrated molded component. Although previous studies have shown that fixed bearing UKA with a metal-backed tibial component has better survivorship over fixed bearing UKA with an all-polyethylene tibial component [9, 16, 18], the relative merit of using mobile bearings remain unclear. We used data from the AOANJRR to frame a four-way comparison between these UKA designs and analyzed failure modes in each group. We found that fixed modular UKA had the lowest rate of revision compared with the other groups.


Our study has several limitations. Registry data does not record variables such as surgical expertise or preoperative severity of osteoarthritis, and it may be subject to selection bias not only in patients selected for UKA, but also in component use and in the decision to revise. These factors are thought likely to affect all groups in this comparison equally. Although clinical trials may be able to control for some of these variables, these studies are usually performed in centers with a specific expertise and when compared with registry data, may not be as reflective of community results. We have used revision rates for comparison between the groups. This may underestimate the true failure rate of UKA as some patients may be too unwell to or be unwilling to undergo a revision procedure. The threshold for revision of UKA may vary across centers and surgeons [12]; however, these factors should have comparable influence on all groups. The use of patient-recorded outcomes in addition to revision rates might have resulted in different conclusions. Also, the effect of radiographic alignment of UKA, which has been shown to have an influence on UKA survival [3, 14], BMI, and surgeon volume were not included in the current analysis. It is also notable that the mobile bearing group has fewer manufacturer designs and one manufacturer design accounted for 95% (19,101 of 20,208) of knees in this category. Additionally, over the study period, manufacturers have introduced changes in implants, such as peg design, method of fixation (cementless fixation), and polyethylene (highly crosslinked polyethylene) that could possibly affect the revision rates. Changes in implant design, independent of the bearing itself, that improve revision rates in UKA must be considered alongside the bearing design that our study attempts to identify.

Comparison of Revision Rate Among Bearing Groups

In this study, fixed modular UKA demonstrated the lowest revision rate. Prior studies comparing fixed bearing with mobile bearing UKA have not found a difference in revision rates [6, 7, 17, 25, 28, 29, 34]. One reason may be the inclusion of both fixed molded metal-backed UKA and all-polyethylene UKA in the same group [6, 17, 29]. In a retrospective analysis by Whittaker et al. [34], the 5-year cumulative survival rate was 88% for mobile bearing UKA compared with 96% for fixed bearing UKA, but they did not show a difference, likely due to inadequate statistical power. A strength of our analysis is the large sample size. In addition, the division of fixed bearings into all-polyethylene, fixed molded metal-backed, and fixed modular UKA, we believe, unmasked differences in the revision rates of these designs. Between the two commonly used bearing designs, namely, mobile bearings and fixed modular bearings, the latter was found to have a lower revision rate in our study.

Differences in Revision Rate as a Function of Sex or Age

The lower revision rate for fixed modular UKA compared with mobile bearing UKA and all-polyethylene UKA was seen across sex and age groups. The revision risk for all UKA groups was higher in patients younger than 65 years. The lower revision rate for UKA with increasing age has also been noted in the National Joint Registry for England, Wales, Northern Ireland, and the Isle of Man [24]. Sex differences are also consistent with the results noted in the United Kingdom where UKA had a higher risk of revision in females compared with age-equivalent males, except for patients younger than 55 years of age [24]. Our results suggest that fixed modular UKA outperforms mobile bearing UKA and polyethylene UKA in terms of revision rates across age and sex groups, prompting us to advocate its use for all patients undergoing medial UKA.

Comparison of Revision Rates by Reasons for Revision and Mode of Revision

The revision risk due to loosening was the lowest for fixed modular UKA followed by mobile bearing UKA and all-polyethylene UKA. The higher loosening rates in all-polyethylene UKA could be related to tibial component loosening because of higher proximal tibial stress transfer [31, 33]. The reason for the difference between mobile bearing and fixed modular UKA needs further investigation, particularly relating to wear and fixation differences between the bearing groups. Higher backside and volumetric wear have been seen with mobile bearing UKA compared with fixed modular UKA in in vitro and retrieval studies [5, 19, 20]. An in vitro study has shown that this difference is even greater with the introduction of highly crosslinked polyethylene for fixed bearing UKA [4]. The other common reason for revision was progression of disease in the lateral and patellofemoral compartments, which was also higher in the mobile bearing group compared with the fixed modular group. In our study, 5.2% of revisions in the mobile bearing UKA group were because of bearing dislocation. Bearing dislocation is a unique complication of mobile bearing UKA. To avoid this, there may be a tendency to use a thicker polyethylene component to tighten the joint. Overstuffing the medial compartment in mobile bearing UKA could lead to a valgus alignment and increase the lateral peak contact stress [15]. This is a possible cause for the higher revision rate for disease progression seen in mobile bearing UKA in this study. The higher risk of revision for disease progression in the lateral compartment for mobile bearing UKA compared with fixed bearing UKA has also been noted by Emerson et al. [7]. The similar and low incidence of revisions attributable directly to wear of the polyethylene insert between mobile bearing, all-polyethylene, and fixed modular groups (0.2% each) coupled with an over three times higher revision rate for bearing dislocation (0.7%) in the mobile bearing group in the registry leads us to question the clinical rationale for the use of mobile bearings in medial UKA.


Fixed-modular UKA had better survivorship than mobile bearing UKA, all-polyethylene, and fixed molded metal-backed UKA. The difference persisted across age and sex categories. The all-polyethylene UKA had the highest revision rates in both age categories except for patients aged 65 years and older where mobile bearing UKA showed no difference compared with all-polyethylene UKA. The revision risk for implant loosening and disease progression was higher for mobile bearing UKA compared with fixed modular UKA. Mobile bearing UKAs also carried a unique risk of early failure from bearing dislocation. The UKA, with its promise of being a less-invasive alternative to TKA in appropriate patients, has continued to evolve with time, and our study suggests that if a UKA is to be considered for the treatment of isolated medial compartment osteoarthritis, the fixed modular UKA has the best survivorship of the current UKA designs.


1. Argenson J-NA, Parratte S. The unicompartmental knee: design and technical considerations in minimizing wear. Clin Orthop Relat Res. 2006;452:137-142.
2. Australian Orthopaedic Association National Joint Replacement Registry. Hip, Knee & Shoulder Arthroplasty: 2019 Annual Report. Adelaide, South Australia, Australian Orthopaedic Association, 2019. Available at Accessed 17 October 2020.
3. Bonutti PM, Dethmers DA. Contemporary unicompartmental knee arthroplasty: fixed vs mobile bearing. J Arthroplasty. 2008;23:24-27.e22.
4. Brockett C, Jennings L, Fisher J. The wear of fixed and mobile bearing unicompartmental knee replacements. Proc Inst Mech Eng H. 2011;225:511-519.
5. Burton A, Williams S, Brockett CL, Fisher J. In vitro comparison of fixed-and mobile meniscal–bearing unicondylar knee arthroplasties: effect of design, kinematics, and condylar liftoff. J Arthroplasty. 2012;27:1452-1459.
6. Cheng T, Chen D, Zhu C, et al. Fixed- versus mobile-bearing unicondylar knee arthroplasty: are failure modes different? Knee Surg Sports Traumatol Arthrosc. 2013;21:2433-2441.
7. Emerson RH Jr, Hansborough T, Reitman RD, Rosenfeldt W, Higgins LL. Comparison of a mobile with a fixed-bearing unicompartmental knee implant. Clin Orthop Relat Res. 2002;404:62-70.
8. Engh GA, Dwyer KA, Hanes CK. Polyethylene wear of metal-backed tibial components in total and unicompartmental knee prostheses. J Bone Joint Sur Br. 1992;74:9-17.
9. Gladnick BP, Nam D, Khamaisy S, Paul S, Pearle AD. Onlay tibial implants appear to provide superior clinical results in robotic unicompartmental knee arthroplasty. HSS J. 2015;11:43-49.
10. Goodfellow J, Kershaw C, Benson M, O'Connor J. The Oxford Knee for unicompartmental osteoarthritis. The first 103 cases. J Bone Joint Sur Br. 1988;70:692-701.
11. Goodfellow JW, O'Connor J. Clinical results of the Oxford knee. Surface arthroplasty of the tibiofemoral joint with a meniscal bearing prosthesis. Clin Orthop Relat Res. 1986:21-42.
12. Goodfellow J, O’Connor J, Murray D. A critique of revision rate as an outcome measure: re-interpretation of knee joint registry data. J Bone Joint Sur Br. 2010;92:1628-1631.
13. Heck DA, Marmor L, Gibson A, Rougraff BT. Unicompartmental knee arthroplasty. A multicenter investigation with long-term follow-up evaluation. Clin Orthop Relat Res. 1993;286:154-159.
14. Hernigou P, Deschamps G. Alignment influences wear in the knee after medial unicompartmental arthroplasty. Clin Orthop Relat Res. 2004;423:161-165.
15. Heyse TJ, Slane J, Peersman G, Dworschak P, Fuchs-Winkelmann S, Scheys L. Balancing mobile-bearing unicondylar knee arthroplasty in vitro. Knee Surg Sports Traumatol Arthrosc. 2017;25:3733-3740.
16. Hutt J, Farhadnia P, Massé V, LaVigne M, Vendittoli P. A randomised trial of all-polyethylene and metal-backed tibial components in unicompartmental arthroplasty of the knee. Bone Joint J. 2015;97:786-792.
17. Ko Y-B, Gujarathi MR, Oh K-J. Outcome of unicompartmental knee arthroplasty: a systematic review of comparative studies between fixed and mobile bearings focusing on complications. Knee Surg Relat Research. 2015;27:141.
18. Koh IJ, Suhl KH, Kim MW, Kim MS, Choi KY, In Y. Use of all-polyethylene tibial components in unicompartmental knee arthroplasty increases the risk of early failure. J Knee Surg. 2017;30:807-815.
19. Kretzer JP, Jakubowitz E, Reinders J, et al. Wear analysis of unicondylar mobile bearing and fixed bearing knee systems: a knee simulator study. Acta Biomater. 2011;7:710-715.
20. Manson TT, Kelly NH, Lipman JD, Wright TM, Westrich GH. Unicondylar knee retrieval analysis. J Arthroplasty. 2010;25:108-111.
21. Marmor L. Results of single compartment arthroplasty with acrylic cement fixation. A minimum follow-up of two years. Clin Orthop Relat Res. 1977;122:181-188.
22. McCallum J 3rd, Scott RD. Duplication of medial erosion in unicompartmental knee arthroplasties. J Bone Joint Sur Br. 1995;77:726-728.
23. Murray D, Goodfellow J, O’Connor J. The Oxford medial unicompartmental arthroplasty: a ten-year survival study. J Bone Joint Sur Br. 1998;80:983-989.
24. National Joint Registry for England, Wales, Northern Ireland and the Isle of Man. NJR 16th Annual Report 2019. Hertfordshire, UK, NJR, 2019. Available at Accessed April 2, 2020.
25. Neufeld ME, Albers A, Greidanus NV, Garbuz DS, Masri BA. A comparison of mobile and fixed-bearing unicompartmental knee arthroplasty at a minimum 10-year follow-up. J Arthroplasty. 2018;33:1713-1718.
26. Palmer S, Morrison P, Ross A. Early catastrophic tibial component wear after unicompartmental knee arthroplasty. Clin Orthop Relat Res. 1998;350:143-148.
27. Pandit H, Hamilton T, Jenkins C, Mellon S, Dodd C, Murray D. The clinical outcome of minimally invasive Phase 3 Oxford unicompartmental knee arthroplasty: a 15-year follow-up of 1000 UKAs. Bone Joint J. 2015;97:1493-1499.
28. Parratte S, Pauly V, Aubaniac J-M, Argenson J-NA. No long-term difference between fixed and mobile medial unicompartmental arthroplasty. Clin Orthop Relat Res. 2012;470:61-68.
29. Peersman G, Stuyts B, Vandenlangenbergh T, Cartier P, Fennema P. Fixed-versus mobile-bearing UKA: a systematic review and meta-analysis. Knee Surg Sports Traumatol Arthrosc. 2015;23:3296-3305.
30. Price AJ, O'Connor JJ, Murray DW, Dodd CA, Goodfellow JW. A history of Oxford unicompartmental knee arthroplasty. Orthopedics. 2007;30:7.
31. Small SR, Berend ME, Ritter MA, Buckley CA, Rogge RD. Metal backing significantly decreases tibial strains in a medial unicompartmental knee arthroplasty model. J Arthroplasty. 2011;26:777-782.
32. Vakharia RM, Sodhi N, Cohen-Levy WB, Vakharia AM, Mont MA, Roche MW. Comparison of patient demographics and utilization trends of robotic-assisted and non-robotic-assisted unicompartmental knee arthroplasty. J Knee Surg. Published online October 22, 2019. DOI: 10.1055/s-0039-1698769.
33. Walker PS, Parakh DS, Chaudhary ME, Wei C-S. Comparison of interface stresses and strains for onlay and inlay unicompartmental tibial components. J Knee Surg. 2011;24:109-116.
34. Whittaker J-P, Naudie DD, McAuley JP, McCalden RW, MacDonald SJ, Bourne RB. Does bearing design influence midterm survivorship of unicompartmental arthroplasty? Clin Orthop Relat Res. 2010;468:73-81.

Supplemental Digital Content

© 2021 by the Association of Bone and Joint Surgeons