Introduction

Total knee arthroplasty (TKA) represents a safe and reliable procedure to reduce pain and functional limitation caused by end-stage osteoarthritis (OA) [1]. The incidence of primary TKA is 450/100,000 and annual rates of surgical procedures are widely increasing worldwide [2,3,4].

Despite the continuous studies to develop new prosthetic designs with advanced kinematic concepts, unsatisfactory results are still reported in 20% of patients undergone TKA [5]. The importance of a reliable prosthetic design together with surgical and medical strategies has been emphasized to improve the functional outcome and achieve better clinical results in TKA [6].

Native kinematic of the medial compartment of the knee is a “ball-and-socket” mechanism, with medial femoral condyle constrained in a pivot motion, while the lateral femur is free to translate posteriorly through a complete arc of flexion [7].

The medial pivot design was introduced in 1994 miming the physiological knee kinematic, to ensure greater efficacy of extensor mechanism in full range of motion (ROM) [8, 9]. This philosophy is based on femoral component with single- or multi-radius curve and a tibial insert with a highly congruent medial compartment and flat lateral compartment. The anteroposterior stability is ensured by a raised anterior lip of polyethylene with minimum risk of condylar lift off [9, 10].

Several studies showed promising mid-term results of medial pivot TKA [11], but the long-term survivorship and clinical outcome have not been extensively investigated with high level of evidence.

The purpose of this systematic review is to analyze survivorship and clinical and radiographic outcomes of medial pivot design TKA.

Material and methods

Literature search and inclusion criteria

A systematic review of the literature has been performed, following Cochrane Handbook of Systematic Reviews of Interventions [12] and Preferred Reporting Items for a Systematic Reviews and Meta-Analyses (PRISMA) [13] for study selection (Fig. 1).

Fig. 1
figure 1

PRISMA flow diagram

Full size image

A systematic search from January 1st, 1990, until October 1st, 2020, was performed in the following databases: the Cochrane Central Register of Controlled Trials (CENTRAL), MEDLINE/PubMed, Embase, Scopus, the Science Citation Index Expanded from Web of Science, ScienceDirect, CINAHL, and LILACS. The research was conducted using the following keywords alone and in all the various combinations: “Medial pivot knee”, “TKA design”, “medial congruent knee”, and “medial stabilized knee”.

Clinical studies reporting objective and patient-reported outcome of medial pivot design TKA were considered eligible for inclusion. There was no quality restriction for study inclusion. Case report, technical notes, editorial commentaries, ex vivo, biomechanical, pre-clinical, and clinical studies without adequate quantitative or qualitative data were excluded. Studies that did not report clear clinical-functional data or survivorship of primary medial pivot TKA were excluded from this research.

Two reviewers independently screened each title and abstract collected from the primary electronic search. In case of relevant title and abstract, the full-text version was obtained.

All references of each study were screened to find any additional relevant paper potentially missed with the first review process. The two reviewers independently followed the same checklist to screen all studies and evaluate the eligibility criteria. Disagreements were resolved through discussion with third reviewer.

The primary endpoints of this analysis were the survivorship and clinical outcome of medial pivot design TKA and revision rate. Secondary endpoints were radiological alignment and complications.

Appraisal of studies’ quality and risk of bias

The level of evidence of included studies were evaluated through the adjusted Oxford Centre For Evidence-Based Medicine 2011 Levels of Evidence [14]. The quality of the studies was defined using the Grading of Recommendations Assessment, Development, and Evaluation (GRADE) system [15], rating quality of evidence in systematic reviews. After the evidence is collected and summarized, the GRADE system provides explicit criteria for rating the quality of evidence that include study design, risk of bias, imprecision, inconsistency, indirectness, and magnitude of effect.

The risk of bias was assessed with “Risk Of Bias” (Rob 2.0) for Randomized Trial (RCT) and “Risk of bias in non-randomized studies” (ROBINS-I) [16] to evaluate non-RCT studies.

Relevant conflict of interest having potential impact on study design and results were considered and reported.

Data extraction and analysis

A stepwise analysis on study design, aim of study, level of evidence, journal, number of procedures included in the study, prosthetic implant used, indication to TKA, mean age, body mass index (BMI), follow-up, and patellar resurfacing. Disparities in data extraction were discussed and resolved by consensus meeting between the authors. When studies reported mixed cohorts of patients, data extraction was selectively focused on procedures involving medial pivot implants.

Radiological data reported as pre- and post-operative femoral tibial angle (FTA), implant alignment with alpha angle (α), femoral beta angle (β), sagittal femoral gamma angle (γ), and sagittal tibial delta angle (δ) were collected. The femoral and tibial radiolucent lines (RLL) were also noted, and divided into major and progressive and minor and non-progressive, according to the knee society total knee arthroplasty roentgenographic evaluation and scoring system [17]. Continuous variables were expressed as weighted means and weighted standard deviation. Mean survivorship, clinical-functional outcomes, and radiographic angles were calculated as weighted means. All studies were assessed for revision for any reason. To evaluate survival rate of implants included in studies with different follow-up times, revisions per 100 components years (CY), which is a well-established method in orthopedic literature [18], were calculated. Studies missing data on the number of revisions performed were excluded from this analysis.

Results

The database research identified a total of 3655 studies. After initial screening, 191 studies were retrieved for full assessment. A total of 34 studies were included in the systematic review. Details are reported within Table 1. Two studies were level of evidence I [10, 19], 8 studies level II [9, 20,21,22,23,24,25,26], and 24 level IV [8, 11, 27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48].

Table 1 General characteristics of included studies
Full size table

The overall quality of the included study was very low according to GRADE system. Twenty-two studies were rated as very low quality [8, 11, 27, 28, 30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45, 47, 48], 6 low quality [22,23,24, 26, 29, 46], 4 moderate [9, 10, 21, 25], and 2 high quality [19, 20] (Table 1). The risk of bias was considered high in 1 research [11], serious in 19 [8, 27, 30,31,32,33,34,35,36, 38,39,40,41,42,43,44,45, 47, 48], and moderate in 4 [28, 29, 37, 46]; 9 [9, 19,20,21,22,23,24,25,26] had some concerns and one RCT [10] had low risk of bias.

A total of 3058 patients (3377 medial pivot design TKA) were finally included in this systematic review.

The mean age at surgery was 69.9 ± 4.4 years and the mean BMI was 29.9 ± 1.0 kg/m2. The mean follow-up was 85.8 months (range, 12–182). Patellar replacement was described in 23 studies [9,10,11, 20, 21, 23, 24, 26, 28, 30, 33, 35,36,37,38,39,40, 42,43,44,45,46], and it was performed in 880 (44.0%) cases on a total of 2000 procedures.

Survivorship and revision rate

The revisions per 100 CY were 0.19, which correspond to a revision rate of 1.9% after 10 years (Table 2). The causes of revision were: 16 cases of periprosthetic joint infection (PJI), 9 cases of aseptic loosening, 8 cases of periprosthetic fractures, 6 cases of persistent pain, 3 component failures (at least one TKA component), and 2 instabilities. Seven of the included studies did not report appropriate data on revisions rate, and then, these studies were not included in the calculation of overall revisions per 100 CY.

Table 2 Revisions per 100 observed components years of the included studies
Full size table

Functional outcomes

The pre-operative mean ROM was 103.4 ± 1.5 (CI 95% 103.3–103.5°). Mean pre-operative KSS and KSS for function were 38.7 ± 1.7 (CI 95% 38.6–38.9) points and 45.8 ± 4.4 (CI 95% 45.7–45.9) points, respectively, and mean pre-operative OKS was 38.6 ± 8.6 (CI 95% 38.2–39.0) points. Considering post-operative results, mean ROM was 117.3 ± 0.4° (CI 95% 117.29–117.31), clinical and functional KSS were 85.9 ± 1.1 (CI 95% 85.88–85.92), and 84.7 ± 3.6 (CI 95% 84.6–84.8). WOMAC and KOOS values were 39.1 ± 7.6 (CI 95% 38.8–39.4) and 84.9 ± 2.2 (CI 95% 84.6–85.2), OKS was 28.2 ± 5.7 (CI 95% 27.9–28.5), and FJS mean value was 68.5 ± 1.0 (CI 95% 68.4–68.6).

Mean ROM of RCTs improved from 112.2 ± 7.2° (CI 95%111.5–112.9) preoperatively to 115.9 ± 1.6° (CI 95% 115.7–116.1) at final follow-up, while clinical and functional KSS varied from 34.0 ± 5.1 (CI 95% 33.3–34.7) and 44.7 ± 0.2 (CI 95% 44.6–44.8) before surgery to 87.7 ± 1.0 (CI 95% 87.6–87.8) and 78.1 ± 2.6 (CI 95% 77.8–78.4). OKS improved from 25.9 ± 0.9 (CI 95% 25.7–26.1) to 35.9 ± 1.5 (CI 95% 35.7–36.1).

Detailed functional outcomes are reported within Table 3.

Table 3 Clinical Outcomes of included studies
Full size table

Radiographic outcomes

Eight studies [19, 23, 27, 35, 39, 41, 43, 48] reported a pre-operative varus deformity (699 TKA [47.4%], mean FTA value: 186.9 ± 0.4° [CI 95% 186.8–187.0]), and other 6 [9, 21, 29, 36, 38, 44] reported a valgus FTA (774 TKA [52.6%], mean FTA 174.6 ± 0.3° [CI 95% 174.5–174.7]). The overall pre-operative FTA was 180.5 ± 0.8°. Mean post-operative FTA after 6.9 years was 177.1 ± 0.1° (CI 95% 177.0–177.2). Post-operative alfa and beta knee angles mean values after 8,4 years were 95.6 ± 0.4° (CI 95% 95.5–95.7) and 89.1 ± 0.2° (CI 95% 89.0–89.2), respectively, while the gamma and delta angles and their mean values were 2.3 ± 0.7° (CI 95% 2.2–2.4) after 8.4 years and 86.7 ± 0.4° (CI 95% 86.6–86.8) after 8.7 years.

Minor (< 2 mm) and non-progressive femoral radiolucent lines were found in 146 (7.5%) knees on the femoral side and in 175 (9.0%) cases on the tibial side. Major or progressive femoral RLL were reported in 11 (0.5%) cases. Two studies [8, 35] reported 12 (0.6%) RLL minor case without mentioning the exact localization.

Post-operative outliers were 116 (24.3%) cases (mechanical axis alignment ± 3 degrees). Regarding RCTs, the FTA varied from 188.1 ± 5.7° (CI 95% 187.4–188.8) after the procedure to 175.9 ± 2.3° (CI 95% 175.2–176.6) after 9.1 years.

Post-operative alfa and beta angles were 96.6 ± 2.2° (CI 95% 96.3–96.9) and 88.6 ± 0.1° (CI 95% 88.5–88.7), while gamma and delta values were 2.8 ± 0.4° (CI 95% 2.7–2.9) and 86.5 ± 0.5° (CI 95% 86.4–86.6). In RCT, studies were not reported any case of RLL or AL. Details of radiological measurements are reported within Table 4.

Table 4 Radiographic outcomes of included studies
Full size table

Complications

The main complications were 30 (1.0%) cases of stiffness, 25 (0.9%) cases of deep vein thrombosis, 17 (0.6%) PJIs, 17 (0.6%) superficial wound infections, 16 (0.6%) cases of persistent pain, 16 (0.6%) retarded wound healings, and 11 (0.4%) periprosthetic fractures.

Other reported complications were 7 (0.2%) cases of pulmonary thromboembolism, 6 (0.2%) AL, 6 (0.2%) peroneal neurapraxias, 4 (0.1%) cases of persistent knee swelling, 3 (0.1%) cases of patellar fractures, 3 (0.1%) cases of knee instability, 1 regional pain syndrome, and 1 patellar tendon rupture. In RCT studies, 14 (3%) cases of stiffness and 4 PJI were reported. Of these, three patients needed reoperation.

Discussion

The aim of this systematic review was to summarize the literature evidence on survivorship and clinical–radiological outcomes of the medial pivoting design TKA. To the best of our knowledge, this is the first systematic review of the literature with a detailed ROM report, patient-reported and objective outcome measures, radiological outcomes, and complications of patients who underwent medial pivot TKA.

Although knee replacement is one of the most performed surgical procedures worldwide, some concerns are related to the relatively high percentage of unsatisfactory outcomes [49, 50]. Native knee kinematics is complex and consists of a constrained pivoting medial compartment and a lateral femoral condyle which can slide posteriorly at high grades of flexion [51, 52]. Medial pivoting designs have been proposed to mimic native knee kinematic and potentially improve clinical outcomes of classic PS TKA designs.

The most important finding of this research is the excellent overall survivorship of medial pivoting design TKA. In fact, the revisions per 100 CY were 0.19, corresponding to a revision rate of 1.9% after 10 years (Fig. 2). Only 51 revision TKA procedures were reported in the literature included in the current review. However, despite the overall excellent survivorship of these implants, survival analysis showed some outliers, as represented in Fig. 2 [8, 36, 38]. In particular, the mean survivorship reported in the retrospective study by Dehl et al. [38] was 93.0% at the final 9.5 year follow-up which is lower than the median value of the overall population studied. However, it should be considered that the small sample size of this study could have overestimated the revision rate, which main causes were not related to the implant design, such as arthrofibrosis and infections. Moreover, it should be highlighted that median values are not significantly affected by the presence of outliers. Some values points reported in the scatterplot (Fig. 2) are outside the 95% CI, showing the quite large dispersion of values around the line representing the projected median; then, any further conclusion based on the data presented in this review should be weighted considering this evidence. Nevertheless, it must be reminded that the 95% CI is a tool to assess the method to esteem values; then, real values should not be expected to be included into the interval.

Fig. 2
figure 2

Scatterplot representing the relationship between survivorship and follow-up in each study included in the review. Solid line, linear median value of survivorship at different follow-up times. Dotted line, 95% CI of median survivorship calculated with the Wilcoxon t test

Full size image

Furthermore, medial pivot TKA showed good-to-excellent results in term of objective and patient-reported outcomes measures (Table 3). The RCTs included in this review demonstrated good-to-excellent clinical results of the medial pivot TKA, with strong improvement of KSS (from 34.0 to 87.7) and slight increase of ROM (from 112° to 115°). However, we point out that on a total of 9 RCTs included, only three studies demonstrated that the medial pivoting TKA had a better clinical results when compared to other knee implants (posterior stabilized and cruciate retaining) [9, 24, 25]. French et al. [25] reported significantly better FJS and quality of life subscale of the KOOS and KOOS-12 in a subgroup of patients who had undergone medial pivot TKA compared to cruciate retaining TKA. However, other patient-reported outcomes and ROM were comparable between the groups. Gill et al. [24] found better results in the group treated with medial pivot implants demonstrating better KSS, ROM, and FJS. On the contrary, Kim et al. [23] observed higher complication rates and worse knee scores, ROM, and patient satisfaction in the medial pivot design group than in the cruciate retaining group. Jeremic et al. [48] reported higher 1 year performance of kinematically medial pivot TKA compared to mechanically aligned ones.

A potential conflict of interest was disclosed in ten studies [9, 10, 21, 25, 28, 38, 42, 44, 45, 48] where authors reported receipt of payment, either direct or indirect, institutional support, and association with a biomedical field entity related to the medial pivot TKA, raising some concerns and precluding the reliable interpretation of clinical results.

Hossain et al. [9] found better ROM in the medial pivot TKA than in the posterior stabilized design (114.9° vs 100.1°). Moreover, physical component scores of SF-36 and Total Knee Function Questionnaire were better in the medial conforming ball-and-socket group. However, no differences were found in the American Knee Society, WOMAC, and Oxford Knee scores. Benjamin et al. [10] performed a gait analysis comparing the medial pivot TKA with the single radius PS implant. They found no significant differences in cadence, walking speed, stride length and stance time, peak stride, mid-support, and push-off forces.

This study has several limitations. First, there is a low level of evidence among included studies, since 22 papers were classified as level of evidence IV. Moreover, only one research was considered at low risk of bias precluding strong conclusions on the results of the included studies. No quality restriction was applied to obtain the largest population of medial pivot TKA. There are possible selection biases deriving from different diagnosis and high heterogeneity in TKA indications that include population (i.e., comorbidities, age, and pre-operative level of activity).

Heterogeneous RCTs studies have been included where the medial pivoting design was compared to the conventional posterior stabilized and cruciate retaining TKA or compared to other models of the medial pivot TKA. All reported outcome measures were highly heterogeneous resulting in a difficult systematic analysis. To reduce bias, the largest number of procedures available in the literature were included and variables gathered from RCTs were analyzed separately.

Conflict of interests were disclosed in several included studies, and this aspect can overestimate the medial pivot TKA outcomes. Only six studies [23, 33, 34, 38, 42, 45] reported survivorship and complications over 10 years raising some concerns of long-term failure risk of the medial pivot TKA.

Conclusions

Medial pivoting design TKAs provided high survivorship of implants, with a revision rate of 1.9% after 10 years. Moreover, good-to-excellent results were obtained in term of objective and patient-reported outcomes measures. Radiological evaluation of studies showed excellent post-operative correction of axial deformities. Three RCTs demonstrated better functional outcomes of medial pivoting designs when compared to the conventional TKA. Only one RCT showed worse results in the medial pivoting group, whereas the remaining RCTs demonstrated non-significant differences between groups. However, several limitations and biases affect this review and further high-quality studies are needed to clarify the role of medial pivoting implants in TKA.