Introduction

Unicompartmental knee arthroplasty (UKA) is a treatment option for patients with end-stage arthritis limited to a single knee compartment (medial or lateral). The advantages of UKA include preservation of the cruciate ligaments, reduced loss of bone stock, and fewer complications [5, 14, 23] compared to total knee arthroplasty (TKA). In addition, UKA grants superior knee kinematics [12, 17] and improved return to activities [34] and sports [35], reduced length of stay [35], better patient-reported outcomes [35] and superior cost-effectiveness [3, 6]. Survival of UKA is more contentious, historically reported to be lower than that of TKA [7, 9, 14, 25, 27] but more recently suggested to be higher [15, 34], particularly in high-volume centers [24].

The ideal indications and patient age for UKA remain a matter of debate. The original selection criteria, as described by Kozinn and Scott [21], are: age over 60 years, weight under 82 kg, low functional demand, preoperative flexion arc greater than 90°, flexion contracture under 5°, and minimal varus or valgus deformity (< 15°). These indications have previously been challenged [29], with one particular area of investigation being younger patients [20, 22]. Furthermore, while expansion of indications have been documented and established for mobile-bearing UKA implants [19, 26], similar considerations for fixed-bearing prostheses have been limited despite increased use [22, 28].

There is currently limited published data available on long-term survival, functional scores and patient satisfaction amongst younger patients receiving fixed-bearing medial UKA [18, 31, 32]. The purpose of this study was therefore to evaluate outcomes of this procedure in patients aged ≤ 60 years and establish whether results are favorable despite non-conventional age criteria. Our hypothesis is that equivalent outcomes can be achieved to comparable cohorts receiving TKA.

Materials and methods

The authors retrospectively reviewed the records of all consecutive medial UKAs performed in patients aged 60 or under, during the period November 2000 to July 2007, under the care of one senior surgeon. The indications for surgery were severe, bone-on-bone medial compartment osteoarthritis (IKDC Grade 4) [16] revealed by radiographic and clinical diagnosis, no significant lateral compartment osteoarthritis, varus deformity < 15° and flexion contracture < 10°. Patients with asymptomatic patellofemoral osteoarthritis and/or deficient but symptomatically stable anterior cruciate ligament deficiency were not excluded.

Out of a total of 312 patients (352 knees) receiving UKAs, 71 patients (91 knees) were aged under 60 and, therefore, eligible for inclusion, comprising 41 women (53 knees) and 30 men (38 knees), aged 54.6 ± 4.2 years at index surgery. All patients received the same fixed-bearing, cemented M/G Unicompartmental Knee System (Zimmer, Warsaw, IN) Patients were operated in the supine position, with a tourniquet and a single-side support. A paramedial incision was completed, extending from the superomedial corner of the patella towards the tibial tuberosity. A minimal medial parapatellar arthrotomy, with subsequent limited medial release and partial resection of the fat pad, facilitated adequate visualization and access. A 2–3 mm tension gauge was used to ensure the knee was left with 2°–3° residual deformity in all patients rather than aiming for mechanical alignment, to reduce the likelihood of progressive wear within the adjacent compartment. Final alignment was checked with long rods from hip to ankle, ensuring that the weight bearing axis remained through the medial compartment. Patients were encouraged to return to non-contact sports but asked to avoid heavy repetitive impact activities (including jogging/running).

All patients were contacted by mail and phone to update their records, noting any complications or revisions. If patients were deceased, their general practitioner or next of kin was contacted to confirm the date and cause of death, and whether any of their knee components had been revised. All patients were assessed using a questionnaire including the Oxford Knee Scores (OKS, 0 = worst; 48 = best) and a subjective evaluation of overall patient satisfaction on a 5-point scale. All patients provided informed consent for the use of their data for research and publications, and thus the institutional review board at the Harrogate and District NHS Foundation Trust deemed that formal approval was not required for this study.

Statistical analysis

Descriptive statistics were used to summarize the data. Implant survival was calculated using the Kaplan–Meier (KM) method, with revision of any component for any reason as endpoint. Uni- and multi-variable linear regression analyses were performed to determine associations of OKS and 3 variables (age, BMI and sex). The multivariable regression models were deemed sufficiently powered, considering the sample size recommendations of Austin and Steyerberg [2] of 10 Subject Per Variable (SPV). Statistical analyses were performed using R version 3.5.2 (R Foundation for Statistical Computing, Vienna, Austria). P values <0.05 were considered statistically significant.

Results

Of the initial cohort of 91 knees, 10 were revised, 6 were deceased, and 1 was lost to follow-up. This left a final cohort of 74 knees (45 female) aged 54.3 ± 4.3 years with BMI of 28.9 ± 3.0 at index surgery (Table 1).

Table 1 Post-operative outcomes for final cohort (74 knees)
Full size table

Survival

Using revision of any component as endpoint, the present series had a KM survival of 92.9% (CI 84.8–96.7%) at 10 years, and 87.8% (CI 78.4–93.2%) at 15 years (Fig. 1). Of the ten knees revised, six received primary TKA, one required a stemmed revision TKA, and three were re-operated at different centers and details of revision surgery were not available. The causes of revision included tibial component loosening (n = 3), femoral component loosening (n = 1), combined tibial and femoral component loosening (n = 1), tibial fracture (n = 1) and spread of osteoarthritis (n = 1). Six of the knees were revised within the first ten years while four were revised between ten and fifteen years.

Fig. 1
figure 1

15-years Kaplan–Meier survival curve of fixed-bearing medial UKA with revision of any component as endpoint

Full size image

Complications

The only postoperative complication observed which did not require revision was a non-fatal case of deep vein thrombosis. There were no cases of deep infection recorded.

Patient-reported outcomes

At final follow-up of 15.0 ± 1.3 years, OKS (available for all 74 knees) was 38.4 ± 8.4. Univariable analysis revealed that OKS was better for men (β = 5.7, p = 0.004) and worsened for patients with higher BMI (β = − 0.7, p = 0.028). Multivariable analysis confirmed that OKS was better for men (β = 5.7, p = 0.003) and improved with age (β = 0.6, p = 0.011) (Table 2). Overall patients were pleased or very pleased with 72 of the knees (97%).

Table 2 Uni- and multi-variable regression analysis to identify factors associated with post-operative OKS
Full size table

Discussion

The most important finding of this study is that fixed-bearing medial UKA yields favorable results in the treatment of single compartment osteoarthritis of the knee in patients ≤ 60 years when performed by adequately trained and experienced practitioners. The present series had a 10-year survival of 93% and a complication rate of 1%. For the final cohort of 74 knees, the mean OKS was 38 and satisfaction was favorable in 97%. Multivariable analysis revealed that there was a positive association between OKS and age as well as male gender, with the positive association with BMI observed in univariable analysis being an indirect one, due to a correlation between BMI and age in the cohort.

The findings of this study compare favorably to previously published literature on UKA. A recent review of 26 studies on 42,791 knees reported a mean 10-year survival of 91.3% for medial UKA [15]. Amongst those reporting on younger cohorts, Schai et al. [31] reviewed a consecutive series of 28 patients aged ≤ 60 years (25 medial and 3 lateral) using the fixed-bearing PFC (DePuy) UKA with 2- to 6-year follow-up and demonstrated a survival of 93%. Pennington et al. [30] published a series of 41 patients (46 knees, ≤ 60 years) undergoing the M/G system UKA (Zimmer) with a mean follow-up of 11 years, with 11-year survival 91%. Lee et al. [22] compared two propensity-score-matched cohorts (n = 71) receiving medial UKA (one of 55 years or younger and one older than 55 years) and found no difference in 10-year survival, reporting 96% for both cohorts, although 13% of patients were lost to follow-up in this cohort. With respect to the results of revisions amongst the present study, it is worth noting that all 10 cases of revision reported favorable results at final follow-up.

Although the success of an implant is often assessed with revision as endpoint, the incidence of medical complications has been shown to be reduced in UKA (as compared to TKA) [1, 5, 14, 23]. Liddle et al. [23] performed an analysis of matched groups of UKA vs TKA from the National Joint Registry for England, Wales, Northern Ireland and the Isle of Man (NJR), which demonstrated in patients undergoing TKA a twofold increase in the risk of venous thromboembolism, myocardial infarction or deep infection, a threefold increase in the risk of stroke, and a fourfold increase in the risk of death within the first 30 days [23].

Patient outcomes reported in the present study at a mean follow-up of 15 years (range 11–18) are also comparable to those reported in the literature. Hamilton et al. [13] reported a mean OKS of 40 at a mean follow-up of 10 years amongst 1000 patients receiving mobile-bearing medial UKA (with the 96 highly active patients in the cohort reporting a significantly higher OKS of 46). In a review of 15 studies assessing a total 3417 mobile-bearing UKAs, Mohammad et al. [26] reported a mean OKS of 40 at 10-year follow-up. In terms of patient satisfaction, a study of 1,703 TKAs by Bourne et al. [4] found that 81% were satisfied or very satisfied 1 year post-operatively, while Felts et al. [11] reported 94% satisfaction for fixed-bearing medial UKA at a mean follow-up of 11 years.

More specifically, the present findings corroborate the most recent literature comparing UKA with TKA in younger patients. The NJR [28] reports for younger patients (< 55 years) receiving the most popular TKA (cemented, unconstrained, fixed; n = 35,125) implant survival rates of 92% and 93% for males and females respectively. It is difficult to meaningfully compare the long-term (> 10 year) functional scores from this study with registry data which are at 1-year follow-up. However, in their review of the literature comparing the results of UKA and TKA in patients ≤ 65 years, Kleeblad et al. [20] systematically assessed 43 studies with 2224 UKAs in patients of mean age 54.7 and 4737 TKAs in patients of mean age 51.7. They found that while 10-year extrapolated survival was significantly better for TKA than UKA (95% vs 90%), range of motion and activity scores were significantly better in UKA. Amongst these younger patients, those receiving UKA (at mean follow-up of 7.4 years) reported a mean OKS of 41 and 94% reported good-to-excellent satisfaction; amongst those receiving TKA (at mean follow-up of 6.7 years), mean OKS was 36 and 90% good-to-excellent satisfaction. The authors concluded that “good-to-excellent outcomes are achievable with medial UKA and TKA in the young and often more active patient population”. It is worth noting that postoperative satisfaction depends largely on patient expectation and preoperative scores [10] though there are no established guidelines or thresholds to enable predictions [8].

There is increasing understanding of the factors which can lead to improved results in UKA. One of these is the volume of UKA surgeries performed by a surgeon. In a review of 37,121 UKAs and 422,149 TKAs recorded in the NJR, Liddle et al. [24] reported better 8-year survival for surgeons performing > 30 operations per year (91.2%) compared to those performing < 10 (87.2%). The same effect was less observable in surgeons performing the same volumes of TKA (96.2% vs. 95.8%). The senior author of the present study is a high-volume UKA surgeon performing > 150 UKA per year with a total experience of > 2000 cases over the past 19 years, with approximately 1700 total knee arthroplasties over the course of the study period.

Ideal selection criteria for UKA have often been subject to debate, and this cohort represents a group of patients historically not recognized to be “strongly indicated” for a UKA procedure. In 1989, Kozinn and Scott [21] described relative contra indications of age < 60 years, weight > 82 kg and extremely active patients or manual laborers. This was derived from an unpublished series of around 100 patients within which the majority of failures from aseptic loosening occurred in these groups. Although UKA has shown to achieve benefits in terms of both functional outcomes [33] and patient mortality or morbidity [23], it is still an underutilized treatment option, partly due to patient and disease-specific strict exclusion criteria.

The limitations of this study include its retrospective design, lack of propensity-matched cohorts receiving alternative treatment (high tibial osteotomy, TKA), limited availability of pre-operative OKS (32 out of 86 knees only) and lack of pre- and post-operative range of motion, comorbidities, radiographic outcomes and return to activity data. It should be noted that surgeon experience and volume could have influenced patient selection and reduced the number of complex severe cases. Its main strengths are the long follow-up, particularly considering the scarcity of similar cohorts reported in the literature, cohort size and low loss to follow-up.

Conclusion

Fixed-bearing medial UKA yields favorable results in the treatment of single compartment osteoarthritis of the knee in patients ≤ 60 years. The present study demonstrates low complication rates, good-to-excellent long-term patient outcomes, and satisfactory implant survival for this age group. The clinical relevance is that UKA can be considered a safe and less invasive alternative than TKA for this demanding patient group.