Authors: Jiaju Yang, Xiaoke Li, Pengyu Liu, Xuanbo Liu, Liangliang Li, Min Zhang
Categories: Review, Bicompartmental knee arthroplasty, Knee osteoarthritis, Patellofemoral joint disease, Total knee arthroplasty, Unicompartmental knee arthroplasty
Source: Journal of Orthopaedic Surgery and Research
Authors: Jiaju Yang, Xiaoke Li, Pengyu Liu, Xuanbo Liu, Liangliang Li, Min Zhang
Patellofemoral joint (PFJ) diseases are chronic degenerative conditions that contribute to knee joint symptoms. Unicompartmental knee arthroplasty (UKA) is widely regarded as an effective treatment for knee osteoarthritis (KOA); however, its specific indications remain a subject of debate.
Patients with PFJ disease are expected to experience outcomes post-UKA comparable to those of patients without PFJ disease.
We conducted this meta-analysis following the guidelines outlined by the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA). A comprehensive search of PubMed, Embase, and Web of Science databases was conducted for studies examining the association between PFJ disease and UKA, including publications up to September 2024. Extracted data encompassed author, publication year, country, disease type, prosthesis type, sample size, mean patient age, gender distribution, follow-up duration, PFJ disease prevalence at surgery, diagnostic methods, and whether PFJ disease was considered a contraindication for UKA. To maintain objectivity, only studies in which PFJ diseases were visually identifiable were included in the meta-analysis. Statistical analyses were performed using Stata 15.0 and Review Manager 5.4.1. A random-effects meta-analysis was conducted to evaluate the Oxford Knee Score (OKS), Knee Society Score (KSS), flexion range of motion (ROM), Forgotten Joint Score (FJS), Tegner activity score, and prosthesis survival rate, with outcomes stratified by PFJ disease type (PFJ degeneration or patella cartilage injury). Mean differences, confidence intervals, and P values were calculated for comparisons between the PFJ disease and non-PFJ disease groups. The Methodological Index for Non-Randomized Studies (MINORS) criteria and the Newcastle–Ottawa Scale (NOS) were applied to evaluate the risk of bias. To address heterogeneity, sensitivity analyses were performed, and publication bias was assessed using funnel plots and Egger's test.
A total of 14,866 knees from 48 relevant studies were included in this systematic review. Methodological quality was assessed using the MINORS criteria, with case series scoring 11.0/16 and cohort studies scoring 18.2/24. PFJ degeneration emerged as the most studied condition, followed by patella cartilage injury. Clinical outcomes assessments indicated that medial PFJ degeneration, anterior knee pain, patella cartilage damage, and patella baja did not significantly impact UKA outcomes or prosthesis survival. However, severe lateral PFJ degeneration, lateral patellar subluxation, lateral trochlear osteophytes, and patellar bone marrow edema did influence results. Fifteen high-quality studies were included in the meta-analysis, involving 6080 patients-1338 in the PFJ disease group and 4,742 in the non-PFJ disease group. With an average NOS score of 7.2, the studies were generally of high quality. Meta-analysis results showed no significant differences between groups in final follow-up OKS, FJS, Tegner activity score, or prosthesis survival rate. However, the PFJ disease group had lower KSS and reduced flexion ROM compared to the non-PFJ disease group. Subgroup analysis further revealed that the PFJ degeneration group scored lower than the patella cartilage injury group on OKS, KSS, and flexion ROM following UKA.
In summary, PFJ disease was found to have limited impact on UKA outcomes; however, caution is recommended for cases involving severe lateral PFJ degeneration due to potential restrictions in postoperative knee function, particularly affecting flexion ROM in UKA patients.
The online version contains supplementary material available at 10.1186/s13018-024-05273-y.
Patellofemoral joint (PFJ) disease is a chronic degenerative condition affecting the PFJ, encompassing a range of disorders generally categorized as PFJ pain, instability, cartilage damage, and arthritis [1]. The prevalence of PFJ disease is influenced by both age and gender, affecting 21.8% of individuals over 40 and 34.7% of those over 60. Additionally, PFJ disease is more common in women, with a prevalence of 24%, compared to 11% in men [2, 3].
Most PFJ diseases are typically classified as patellofemoral osteoarthritis (PFA) based on clinical presentation and imaging findings [1]. These conditions are marked by progressive wear and tear, chafing, cartilage degradation, sclerosis, and osteocortical hyperplasia, leading to symptoms such as pain and restricted mobility in the anterior knee joint [4]. These conditions are marked by progressive wear and tear, chafing, cartilage degradation, sclerosis, and osteocortical hyperplasia, leading to symptoms such as pain and restricted mobility in the anterior knee joint [5]. PFA, more common in middle-aged individuals than osteoarthritis in the tibiofemoral compartment, is frequently overlooked due to its variable symptom presentation [6, 7]. Additionally, the tibiofemoral compartment is more prone to developing a range of KOA symptoms [8].
Knee-preserving surgical techniques primarily include high tibial osteotomy and unicompartmental knee arthroplasty (UKA) [9]. UKA specifically targets the affected knee compartment, offering several advantages over total knee arthroplasty (TKA) [10]. Notably, UKA is associated with reduced intraoperative trauma [11], minimized damage to extensor structures [12], a decreased need for osteotomy, reduced blood loss [13], and shorter operative times [14]. Early postoperative benefits include shorter hospital stays [15] and rapid recovery [16], while long-term advantages include enhanced knee mobility [17], better gait preservation [18], and intact proprioception [19]. Since its introduction into clinical practice, the number of UKA procedures and associated patient satisfaction have steadily increased, making it a widely adopted approach for managing KOA with positive outcomes across short, medium, and long-term follow-ups [20, 21].
Nevertheless, only a limited subset of patients meets the criteria for UKA, and there is variability among healthcare professionals regarding the specific indications for this procedure [22]. In 1989, Scott [23] proposed guidelines for UKA patient selection, identifying PFJ degeneration, anterior knee pain, and severe PFJ cartilage damage as contraindications. These guidelines were later reaffirmed by Stern [24], and have become widely accepted. However, with the growing popularity of UKA, increased surgical expertise, and advancements in techniques, these traditional contraindications have been criticized as overly restrictive, limiting patient eligibility for UKA [17]. Many studies, for example, have shown that PFJ disease frequently coexists with KOA in the medial or lateral compartments [25], persisting throughout the disease course in most patients [26]. This observation suggests that PFJ disease may not be an absolute contraindication, as doing so would significantly limit the eligible patient population for UKA [27].
A previous investigation provided a comprehensive analysis of PFJ disease's impact on mobile-bearing (MB) UKA, concluding that both lateral PFJ disease and lateral patellar subluxation can affect treatment efficacy [28]. However, this study was limited to MB UKA and only addressed four PFJ diseases, without considering the effects on fixed-bearing (FB) UKA. To guide clinicians, this analysis systematically evaluated the effects of various PFJ diseases-including PFJ degeneration, lateral patellar subluxation, bone marrow edema (BME) in the patella, anterior knee pain, PFJ cartilage injury, patellofemoral chondromalacia (PFCM), and patella baja on postoperative outcomes and prosthesis survival for both medial/lateral FB and MB UKA. To ensure result reliability and minimize subjective bias, studies that diagnosed PFJ diseases radiologically were excluded, retaining only those in which PFJ disorders were visually identified. The meta-analysis further stratified PFJ disease types, offering a comprehensive evaluation of their impact on functional outcomes and prosthesis longevity following UKA. Additionally, Bicompartmental Knee Arthroplasty (BKA) was briefly reviewed to provide clinical guidance on selecting between UKA, BKA, and TKA for patients of various age groups. In this study, broader inclusion criteria and comprehensive search strategies were employed to address existing gaps, enabling an in-depth investigation of PFJ diseases in relation to functional outcomes and prosthesis survival following UKA. The objective was to provide healthcare professionals with insights to improve the management of patients with these conditions. It was hypothesized that post-UKA outcomes for patients with PFJ disease would be comparable to those for patients without PFJ disease.
This systematic review and meta-analysis adhered to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines and was registered with the International Prospective Register of Systematic Reviews (Registration No. CRD42024503637). Both prospective and retrospective studies were included, focusing on patients who underwent UKA and were identified preoperatively or intraoperatively with PFJ disease.
Eligible studies for inclusion met the following (I) examined medial and/or lateral UKA as a treatment for patients with PFJ disease, with or without KOA; (II) provided functional outcome measures, including the Oxford Knee Score (OKS) [29], Knee Society Score (KSS) [30], Knee Injury and Osteoarthritis Outcome Score (KOOS) [31], Forgotten Joint Score (FJS) [32], Visual analog score (VAS), Tegner activity score, flexion Range of motion (ROM), and/or prosthesis survival rate before and after medial or lateral UKA; and (III) were human studies published in English through September 2024. Exclusion criteria (I) duplicate publications; (II) studies involving alternative treatments for PFJ disease, such as TKA or conservative approaches; and (III) animal studies, case reports, or reviews. Following study screening, two investigators (JJY and XKL) conducted a full-text review of the remaining studies, with disagreements resolved through discussion with a third reviewer (MZ).
To identify studies on the relationship between PFJ disease and UKA, systematic searches were conducted in PubMed, Embase, and Web of Science databases for relevant articles published up to September 2024. This meta-analysis followed the PICO (Patient, Intervention, Comparison, Outcome) framework, defined as the patient population consisted of individuals diagnosed with PFJ disease [33]; the intervention was UKA performed on patients with PFJ disease; the comparison was UKA in patients with PFJ disease versus UKA in those without PFJ disease; and the outcomes were assessed through clinical measures, including OKS, KSS, KOOS, FJS, VAS, Tegner activity score, ROM, and prosthesis survival rate, along with other relevant clinical outcomes. The search strategy employed keywords such as ((Unicompartmental Knee Arthroplasty) OR (Knee Replacement Arthroplasty) OR (Unicondylar Knee Arthroplasty) OR (Partial Knee Replacement)) AND ((Patellofemoral disease) OR (Patellofemoral chondral lesion) OR (Patellofemoral degenerative changes) OR (Patellofemoral joint)) (Tables S1–S3).
After compiling the identified studies and removing duplicates, the titles and abstracts of the remaining articles were independently reviewed for relevance by two investigators (JJY and XKL), with any disagreements resolved through discussion with a third investigator (MZ). Relevant articles were then thoroughly reviewed, and full texts were selected for inclusion. Study retrieval was managed using EndNote X9 software, while data of interest were extracted by two investigators (JJY and XKL) and organized in a Microsoft Excel table. Any discrepancies in data extraction were discussed and resolved among the reviewers, with the third reviewer (MZ) available for arbitration to ensure consistency across the search and data extraction process.
The extracted data included the author, publication year, country, disease type, prosthesis type, number of knees, mean patient age, gender, follow-up duration, PFJ disease prevalence at the time of surgery, diagnostic approach, and whether PFJ disease was considered a contraindication for UKA. Additionally, relevant clinical data were collected, including knee flexion and extension ROM, prosthesis survival rate, and knee function scores (such as OKS, KSS, KOOS, FJS, VAS, and Tegner activity score).
The adjusted Oxford Center for Evidence-Based Medicine 2011 Levels of Evidence criteria were applied to evaluate the evidence levels for all included studies [34], while the MINORS criteria were used to assess study quality [35]. The Newcastle–Ottawa Scale (NOS) [36] evaluated the methodological quality of cohort and registry-based studies in the meta-analysis, with scores of 7 or higher indicating high-quality research, scores between 4 and 6 denoting moderate quality, and scores below 3 classified as low quality. Study quality assessments were independently conducted by two investigators (JJY and XKL), with a third investigator (MZ) available to resolve any disagreements. Assessment of the quality of recommendations.
The quality of evidence was assessed using the Grading of Recommendations Assessment, Development, and Evaluation (GRADE) framework [25]. This evaluation considered study designs (randomized controlled trials or observational studies), limitations, result inconsistencies, evidence indirectness, imprecision, and publication bias. Additional factors, such as large effect sizes, confounding variables affecting treatment effects, dose–response relationships, and quality ratings for each outcome, were also reviewed. This assessment was independently conducted by two authors (JJY and XKL), with disagreements resolved by consensus or, if necessary, with the involvement of a third evaluator (MZ).
Descriptive statistics were performed using IBM SPSS software version 25, with mean ± standard deviation (SD) used for quantitative factors (e.g., age, follow-up period, and functional outcomes) extracted from the studies; alternatively, median or range values were used as needed. Statistical analyses were conducted using Stata 15.0 and Review Manager 5.4.1, and forest plots were created accordingly. For continuous outcomes, mean difference (MD) and 95% confidence intervals (CI) were used as effect sizes. When studies did not provide means and/or standard deviations, predefined criteria were applied to convert medians, ranges, or quartiles into SDs. Heterogeneity was assessed using the I^2^ statistic and P-value, with significant heterogeneity defined as P < 0.1 or I^2^ ≥ 50%, warranting the use of a random-effects model for meta-analysis. Insignificant heterogeneity (P > 0.1 or I^2^ ≤ 50%) led to the use of a fixed-effect model. Sensitivity analyses were conducted to identify potential sources of heterogeneity, and publication bias was evaluated with funnel plots and Egger’s test [37].
The initial search yielded 3,520 potentially relevant studies. After removing duplicates, 2,082 papers remained. A review of titles and abstracts led to the selection of 81 studies for full-text review. Of these, 33 studies were 15 were review articles, 5 lacked full-text availability, and 13 presented disparate outcomes. Consequently, 48 studies-comprising 21 cohort studies [12, 27, 38–56] and 27 registry-based studies [57–83] were included in the final analysis. Ultimately, 15 studies that visually identified PFJ disease were included in the meta-analysis, covering 6,080 patients who had undergone UKA (Fig. 1) [27, 38, 40, 42, 44, 47–49, 52, 55, 67, 72, 75, 82, 83].Fig. 1PRISMA flow diagram outlining the study selection process
Six studies recommended against medial UKA in patients with PFJ disease, and one study advised excluding PFJ disease from lateral UKA as well. The remaining 41 studies offered no evidence regarding PFJ disease as a contraindication for UKA. Overall, the included studies assessed 14,866 knees from patients across 14 countries, covering the years 2007 to 2024, with an average prevalence of PFJ disease at the time of surgery of 26.5% (Fig. S1). The patient cohort had an average of 53.7% women, a mean age of 65.1 years (range 26–93), and a mean follow-up duration of 5.0 years (range 0.1–19.6 years) (Table 1).Table 1Articles included in systematic review (n = 43)Source of study populationType of PFJ diseaseType of arthroplastyMedial or lateral UKAKnees at Follow-upSex (% women)Age, y^a^Follow-up, y^a^Prevalence (% of PFJ disease at surgery)Propose or oppose PFJ disease as a contraindication to UKADiagnostic methodsBeard [40]UKPatella cartilage injuryMobile bearing UKAMedial824NR66 (32–89)2 (1–7)13.7OpposeRadiology and visualBeard [41]UKPatella cartilage injuryMobile bearing UKAMedial10050.569 (53–89)254OpposeRadiologyBerend [64]USAPFJ degenerationMobile bearing UKAMedial3185861.6 (33.6–90.1)0.67 (0.12–2.3)NROpposeRadiologyNaal [77]CHPatella BajaFixed bearing UKAMedial8344.266.0 (46–84)24.8OpposeRadiologyKuipers [66]NLPFJ degenerationMobile bearing UKAMedial4376862.8 (39.3–84.6)2.6 (0.1–7.9)22.4OpposeRadiologySeyler [69]USAPFJ degenerationMobile bearing UKAMedial805872 (44–91)5 (2–14)48OpposeRadiologyBerend [65]USAPFJ degenerationMobile bearing UKAMedial63853.762.7 (33–90)NR37.9OpposeRadiologyPandit [47]UKPatella Cartilage InjuryMobile bearing UKAMedial10005266 (32–88)5.6 (1–10)15.8OpposeVisualMunk [46]DKLateral patella subluxationMobile bearing UKAMedial2604966 (9)118ProposeRadiology and visualKang [62]UKPFJ degenerationMobile bearing UKAMedial19542.965.5 (51–93)3.4 (2–7)64.1OpposeRadiologyXing [78]USAPFJ degenerationFixed bearing UKABoth178NR67 (36–90)2 (2–6.4)NROpposeRadiology and visualMofidi [68]USAPFJ degenerationFixed bearing UKAMedial14448.163 (11)NR30.6OpposeRadiologySong [74]KRPFCMMobile bearing UKAMedial6288.765.3 (49–78)NR66.1OpposeVisualLiddle [53]UKAnterior knee painMobile bearing UKAMedial4064564.8 (35–87)NR33OpposeRadiologySébilo [73]FRAnterior knee painBothBoth9446270 (26–93)5.2 (4.3)11.5OpposeRadiologyHurst [79]USAPFJ degenerationMobile bearing UKAMedial1000NRNR3.23.3OpposeRadiologyNeogi [76]KRPatella BajaMobile bearing UKAMedial13473.264.2 (51–74)6.2 (5.2–8.0)14.2OpposeRadiologySong [50]KRAnterior knee painMobile bearing UKAMedial1059764 (43–81)5.4 (3.1–10.2)54.3OpposeRadiology and visualWong [51]CNPFJ degenerationMobile bearing UKAMedial5172.965.7 (53–81)4.3 (1–7.7)22OpposeRadiologyThein [61]USAPFJ degenerationFixed bearing UKAMedial11347.363.9 (10.4)2 (1–4.2)27.5OpposeRadiologyKonan [45]UKAnterior knee painMobile bearing UKAMedial1004369 (41–82)10 (8–13)29OpposeRadiology and visualPongcharoen [49]THAIPFJ degenerationMobile bearing UKAMedial11486.865.3 (44–88)1.59 (1.0–2.48)22.8OpposeVisualAdams [39]USAPFCMFixed bearing UKAMedial834164.9 (49.1–88.0)2.17 (2.00–2.72)37.3OpposeVisualHamilton [27]UKPFJ degenerationMobile bearing UKAMedial10005266 (32–88)10.3 (5.3–16.6)15.8OpposeRadiology and visualYue [54]CNBME in the patellaMobile bearing UKAMedial14675.366 (46–86)2.9 (2.6–5.2)10.6ProposeRadiologyHamilton [44]UKPFJ degenerationMobile bearing UKAMedial8054866 (32–89)10 (5–17)20ProposeRadiology and visualLim [12]SGPFJ degenerationFixed bearing UKAMedial27958.462 (42–84)10.5 (2.1)15.6OpposeRadiologyHutt [70]UKPFJ degenerationMobile bearing UKAMedial18750.864.2 (49–84)3.6 (0.5–8)49.7OpposeRadiologyBerger [42]BEPFJ degenerationFixed bearing UKAMedial2404565 (10)3.426ProposeRadiology and visualBurger [58]USAPFJ degenerationFixed bearing UKAMedial63944.663.4 (41.1–86.8)4.3 (2.0–9.2)55.7OpposeRadiologyAbdulkarim [38]UKPFJ degenerationFixed bearing UKAMedial14753.767 (48–87)9.4 (5–10.7)46.9OpposeRadiology and visualBurger [57]USAPFJ degenerationFixed bearing UKALateral14057.963.4 (10.5)4.1 (2.0–8.5)52.9ProposeRadiologyDeckard [59]USAPFJ degenerationFixed bearing UKAMedial14343.464.1 (36.3–88.2)2 (0.9–6.9)NROpposeRadiology and visualFujita [60]JPPFJ degenerationMobile bearing UKALateral5490.772.92.8 (1–6.1)75.9OpposeRadiologyKennedy [75]UKPatella cartilage injuryMobile bearing UKALateral28562.564.7 (39–90)7 (1–13)16.1OpposeVisualWang [63]CNPFJ degenerationMobile bearing UKAMedial50064.861.1 (8.0)5.27 (1.59–11.60)79.2ProposeRadiologyCarlo [43]ITAPFJ degenerationFixed bearing UKAMedial1373554.3 (47–60)6.2 (5.2–7.5)46OpposeRadiologyJi [67]CNPatella cartilage InjuryFixed bearing UKAMedial33769.467.15.731.2OpposeVisualBull [72]UKPatella cartilage injuryFixed bearing UKAMedial23144.665.8 (41–89)3.1 (2–4)54OpposeRadiology and visualPlancher [48]USAPFJ degenerationFixed bearing UKAMedial15648.265.8 (39–89)9.28 (5–18.9)28.2OpposeVisualAndronic [71]CHPFJ degenerationFixed bearing UKAMedial4936.769 (47–82)5.17 (5–9)NROpposeRadiologySingh [52]USAPatella cartilage injuryMobile bearing UKAMedial31757.370.3NR36.4OpposeVisualCarender [80]USAPFJ degenerationMobile bearing UKAMedial2334760 (29–87)5.7 (2.0–13.1)9ProposeRadiologyRuderman [83]USALateral patella subluxationFixed bearing UKAMedial51742.263.2 (41.4–86.8)4.4 (2.0–9.0)20.4OpposeRadiologyGerow [56]USAPatella cartilage injuryBothMedial6735364 (58–69)111OpposeRadiology and visualMyatt [81]UKPatella cartilage injuryBothMedial11138.767 (47–91)5NROpposeRadiology and visualPlancher [82]USAPatella cartilage injuryFixed bearing UKALateral6168.965.310.9 (4–19.6)46OpposeRadiology and visualGaggiotti [55]ARPatella cartilage injuryFixed bearing UKABoth11072.765.8 (8.4)6.0 (3.1 to 7.3)80OpposeRadiology and visualPFJ, PF joint; Pt., Patients; SD, Standard deviation; y, Years; UKA, Unicompartemental arthroplasty; UK, United Kingdom; USA, United States of America; CH, Switzerland; NL, The Kingdom of the Netherlands; DK, Denmark; KR, Korea; FR, France; CN, China; THAI, Thailand; SG, Singapore; BE, Belgium; JP, Japan; ITA, Italy; AR, Argentina; PFCM, Patellofemoral chondromalacia; BME, Bone Marrow Edema; NR, Not reported^a^Data are expressed as mean or mean ± SD, or mean (range) unless otherwise indicated
The quality of the included studies was assessed using Review Manager 5.4.1 statistical software. Evaluation of the cohort studies and case series studies was performed using the MINORS criterion. The case series studies had a methodological quality score of 11.1 out of 16 points. Deficiencies were found in the case series studies regarding loss-to-follow-up rates and the prospective computation of sample sizes (Fig. S2A). While cohort studies scored 18.2 out of 24 points, these two factors were also scored relatively low (Fig. S2B). In contrast, a greater number of case-series studies had a loss-to-follow-up rate of over 5% and lacked prospective calculation of study. Notably, many case series had a follow-up loss rate exceeding 5% and lacked prospective sample size calculation (Table S4). The 15 cohort and registry-based studies included in the meta-analysis were evaluated with the NOS, yielding an average score of 7.2 points, suggesting relatively high quality across these studies (Table S5). The methodological quality is shown in Figs. S3 and S4.
The quality of GRADE evidence for functional outcomes is presented in the evidence certainty assessment (Additional file 2). The evidence quality for the Tegner activity score and prosthesis survival rate was downgraded to a very low level due to inconsistencies in study results. Evidence quality for all other outcomes was assessed as low, with unanimous agreement among the reviewers (JJY, XKL, and MZ).
Each study evaluated the correlations between various PFJ diseases and functional outcomes and/or prosthesis survival rates following UKA, along with prosthesis types and diagnostic methods used, as key outcome indicators (Table 1).
The most commonly studied PFJ disease was PFJ degeneration, followed by PFJ cartilage injury (Fig. 2A). After reviewing all included studies, it was observed that PFJ degeneration was typically diagnosed through preoperative imaging, whereas PFJ cartilage damage was identified visually by the surgeon during the procedure. Radiology was the primary diagnostic method, used in 52.1% of cases, followed by a combination of radiology and visual examination, and visual examination alone (Fig. 2C). Radiological diagnosis involved evaluating the preoperative knee via skyline radiographs, which showed signs of joint-space narrowing, subchondral sclerosis, and osteophytes for PFA. During UKA, visual inspection was applied to assess PFJ cartilage deterioration intraoperatively. MB devices were preferred in 54.2% of cases, followed by FB devices at 39.6% (Fig. 3A). Medial UKA was the most frequently performed arthroplasty type (85.4%), with lateral UKA representing the second most common choice at 8.3% (Fig. 3B).Fig. 2PFJ diseases covered in this study. A The type and number of studies that oppose or propose PFJ disease as a contraindication to UKA. Oppose vs Propose in PFJ degeneration (26 studies), PFJ cartilage injury (12 studies), anterior knee pain (4 studies), PFCM (2 studies), patella baja (2 studies), BME in the patella (1 study), and lateral subluxation of the patella (1 study). B Proportion of PFJ disease as a contradiction to UKA. Seven studies (14.6%) proposed that PFJ disease was a contraindication for UKA, in contrast, forty-one studies (85.4%) concluded that PFJ disease was not a contraindication for UKA. C Proportion of studies per diagnostic methods. The most used diagnostic method was radiology (52.1%), followed by radiology and visual (31.3%)Fig. 3PFJ diseases effects on UKA. A Proportion of studies per arthroplasty device. The most chosen type of arthroplasty was MB UKA (54.2%), followed by FB UKA (39.6%). B Proportion of studies per arthroplasty type. The most chosen type of arthroplasty was medial UKA (85.4%). C The distributions of clinical and functional No difference in green and Significant difference in red. The most used score was KSS (18 studies), followed by OKS (15 studies)
Severe lateral PFJ degeneration, lateral patellar subluxation, and BME in the patella were cited as reasons to avoid medial UKA in six studies (12.5%). One study (2.1%) proposed that PFJ disease was a contraindication for lateral UKA, as worse postoperative clinical and functional scores or prosthesis survival rates were observed, and significant differences were noted in patients who had a combination of PFJ diseases preoperatively or demonstrated that PFJ disease predicted poor postoperative clinical and functional scores or prosthesis survival. Significant differences in most outcomes after UKA were not found in these studies, regardless of the presence of PFJ disease. However, significantly worse KSS [49], VAS [67] and flexion ROM [49, 52], along with better OKS [40, 70], KSS [64] and extension ROM [38] after UKA were found in patients who had PFJ disease in several studies. In addition, apart from the study that observed a substantially higher number of modifications in patients who had PFJ degeneration, no significant variations in revision rates between the different groups were found in any other articles (Fig. 3C, Table S6) [38].
The impact of PFJ disease on the OKS following UKA was evaluated in eight studies [27, 40, 42, 44, 47, 67, 72, 75]. Overall, no significant association was observed between PFJ disease and OKS (MD = 0.30, 95% CI − 0.64 to 1.24, I^2^ = 67%, P < 0.01). Subgroup analysis revealed that PFJ degeneration significantly reduced postoperative OKS (MD = − 0.81, 95% CI − 1.81 to 0.18, I^2^ = 0%, P = 0.87), whereas patellar cartilage injury had no notable effect on OKS after UKA (Fig. 4A). Publication bias was assessed using a funnel plot, which showed studies distributed evenly on both sides of the regression line, with a P-value of 0.948, indicating no potential publication bias.Fig. 4Meta-analysis illustrating the impact of different types of PFJ diseases on the OKS (A), KSS (B), and flexion ROM (C)
The impact of PFJ disease on postoperative KSS following UKA was assessed in ten studies [27, 38, 40, 44, 47–49, 52, 55, 75]. Overall, PFJ disease significantly reduced KSS in patients undergoing medial UKA (MD = − 1.24, 95% CI − 2.27 to − 0.22, I^2^ = 0%, P = 0.77). Subgroup analysis indicated that PFJ degeneration significantly decreased postoperative KSS (MD = − 1.83, 95% CI − 3.13 to − 0.53, I^2^ = 0%, P = 0.95), while patellar cartilage injury had no significant effect on KSS after UKA (Fig. 4B). The funnel plot was symmetric, indicating no publication bias, and Egger’s test was not statistically significant (P = 0.227).
The impact of PFJ disease on postoperative flexion ROM following UKA was evaluated in six studies [38, 48, 49, 52, 55, 67]. Overall, PFJ disease was found to significantly reduce flexion ROM in UKA patients (MD = − 2.06, 95% CI − 3.99 to − 0.12, I^2^ = 78%, P < 0.01). Subgroup analysis indicated that PFJ degeneration significantly decreased postoperative flexion ROM (MD = − 3.46, 95% CI − 9.27 to 2.36, I^2^ = 86%, P < 0.01), as did patellar cartilage injury (MD = − 2.06, 95% CI − 3.99 to − 0.12, I^2^ = 78%, P < 0.01) (Fig. 4C). The funnel plot suggests a low risk of publication bias, with Egger’s test yielding a P-value of 0.583.
Three studies assessed the FJS [42, 67, 72], five evaluated the Tegner activity score [27, 44, 47, 75, 82], and five examined prosthesis survival rates [27, 42, 44, 47, 75]. No significant differences were found in these outcomes (FJS: MD = − 2.23, 95% CI − 5.94 to 1.48, I^2^ = 0%, P = 0.61; Tegner activity MD = 0.01, 95% CI − 0.12 to 0.14, I^2^ = 25%, P = 0.25; prosthesis survival MD = 1.24, 95% CI − 2.18 to 4.66, I^2^ = 0%, P = 0.75; Fig. 5). Publication bias was evaluated using funnel plots and Egger’s tests, with no indication of bias observed (P = 0.09; 0.556; 0.072).Fig. 5Meta-analysis illustrating the impact of different types of PFJ diseases on the FJS (A), Tegner activity score (B), and prosthesis survival rate (C)
The sensitivity analysis confirmed the stability and robustness of the overall results, even with the exclusion of individual studies, thereby ensuring the reliability of the findings. Funnel plots from the meta-analyses showed no significant asymmetry, and Egger’s regression test indicated a low likelihood of publication bias (Fig. S5a–g, Table S7).
Evidence certainty was rated as low for 19 outcomes and very low for 2 outcomes. The overall evidence quality was generally rated as low, mainly due to the observational design of the included studies, which impacted all outcomes. Additional factors, including study limitations, imprecise effect sizes, result inconsistencies, and potential publication bias, further contributed to this rating.
The primary findings of this systematic review indicated that severe lateral PFJ degeneration may be the most common cause of poor prognosis following UKA. This study systematically analyzed 48 studies to evaluate functional outcomes and prosthesis survival in patients with PFJ disease compared to those without (N-PFJ disease). To ensure objectivity, the meta-analysis was limited to 15 studies focusing on visually identifiable PFJ diseases. The quality of evidence for each outcome was assessed using the internationally recognized GRADE criteria. These outcomes should be interpreted alongside an evaluation of evidence quality to gauge confidence in the findings.
The most significant characteristic associated with PFJ disease was lateral PFJ degeneration, particularly in cases of severe degeneration. Preoperative Kellgren-Lawrence Grade IV lateral PFJ degeneration, or "groove-like changes," should be considered a contraindication for UKA [42, 44, 57, 63]. This recommendation is based on the rarity of concurrent lateral PFJ degeneration and medial KOA, as these conditions are generally driven by distinct lateral PFJ degeneration is commonly linked to knee valgus, while medial KOA is typically associated with knee varus [44]. Nevertheless, the impact of severe lateral PFJ degeneration on UKA outcomes could be managed. One study found no significant differences in functional scores between patients with severe lateral PFJ degeneration and those without, aside from a lower KSS in the degeneration group [49]. This suggests that preoperative PFJ status might not significantly affect postoperative functional scores or hip-knee-ankle angle changes [60]. Another study classified the extent of lateral PFJ degeneration, finding no differences in functional scores or prosthesis survival rates across groups, concluding that asymptomatic severe lateral PFJ degeneration did not influence UKA outcomes [48]. Consequently, these findings suggest that PFJ degeneration should not automatically disqualify patients from undergoing UKA.
Given the high prevalence of PFJ degeneration, it is argued that it should not be viewed as an absolute contraindication for UKA. Additionally, significant cartilage deterioration in the PFJ has been observed even in young individuals, worsening progressively with age; consequently, nearly all older adults exhibit severe PFJ degeneration [84]. Many patients also do not display clinical symptoms or radiological signs of PFJ degeneration [25]. Therefore, this degeneration may be considered a typical aspect of aging that, accompanied by minimal symptoms, appears to have no substantial impact on postoperative outcomes for UKA recipients.
Patellar instability is highly likely to occur, predominantly in the lateral direction, and is often referred to as lateral patellar subluxation [85, 86]. As a leading cause of knee pain, swelling, and perceived instability, lateral patellar subluxation may impact UKA outcomes. Studies indicate that lateral patellar subluxation elevates the risk of cartilage damage by altering stress distribution in the PFJ [87]. Notably, medial UKA is ineffective at correcting preoperative abnormal tracking or PFJ malalignment, potentially resulting in suboptimal outcomes for patients with concurrent lateral patellar subluxation [46]. Additionally, medial UKA can lead to overloading in the lateral compartment and exacerbate patellar subluxation, significantly raising the risk of complications [28]. The trochlea also plays a critical role. One study examined the effect of lateral trochlear osteophytes on UKA outcomes, using Cox proportional hazards models to show that lateral trochlear osteophytes increase the risk of UKA failure. Thus, the presence of lateral patellar subluxation and lateral trochlear osteophytes should be regarded as contraindications for UKA [42, 44, 80].
The presence of PFJ cartilage injury does not necessarily preclude the use of UKA, as research has shown that the extent and location of cartilage injury within the joint do not correlate with patient prognosis [88]. Additionally, PFJ cartilage injury was found to have minimal impact on early postoperative activity levels following UKA [52]. Neither the location nor the extent of preoperative PFJ cartilage injury influenced long-term outcomes or prosthesis survival rates after UKA [38, 40]. Patellofemoral chondromalacia (PFCM), marked by reduced cartilage density and softening of the patella cartilage, often results from patellar maltracking and microtrauma within the PFJ [89]. Historically, PFCM was considered a contraindication for UKA. However, our included studies suggest that preoperative PFCM may not significantly impact the operative outcomes of UKA [39, 74]. Patella baja, resulting from the shortening of the patellar tendon, is typically associated with symptoms such as PFJ pain [90] and muscle weakness [91]. While it has been shown to negatively affect postoperative outcomes after TKA [92], its impact on UKA remains debatable. One study used the Blackburne–Peel index [93] and the Insall–Salvati ratio to determine the patella height [94] to measure patella height, finding a weak negative correlation between preoperative patella baja and postoperative knee function scores or mobility. Consequently, patella baja was determined to have a relatively limited effect on UKA outcomes [77]. However, patella baja may increase the risk of impingement between the patella and femoral prosthesis [76].
Many KOA patients report anterior knee pain despite a lack of imaging evidence for PFJ disease, or vice versa, indicating a weak and independent association between these conditions. Anterior knee pain is not considered a contraindication for UKA, as postoperative functional improvements have been found to be similar in patients with and without anterior knee pain [50, 53, 73]. Patients were stratified into two groups based on their experience of anterior knee pain before UKA, and positive procedural outcomes were observed across both groups, with all patients showing complete relief from postoperative symptoms [45]. Additionally, patients were grouped based on the presence or absence of PFJ degeneration, revealing that those with PFJ degeneration had higher preoperative VAS scores. However, no statistically significant difference in anterior knee pain or VAS scores was found between the groups postoperatively [49].
The presence of BME in the patella may indicate underlying bone marrow damage, often presenting as knee pain, and is recognized as a primary reason for UKA revision [95, 96]. Consequently, a correlation between patellar BME and unfavorable postoperative outcomes or reduced survival following UKA may be inferred. Preoperative knee Magnetic Resonance Imaging scans were reviewed, and patients were classified based on the presence of patellar BME. Postoperative evaluations revealed that patients with patellar BME experienced worse VAS scores and reduced flexion ROM compared to other groups. Thus, it is recommended that TKA be considered for patients with PFJ degeneration combined with patellar BME [54].
UKA implants are categorized by design into MB and FB devices, each differing in bearing surface, design concept, and surgical technique [97]. In FB devices, the polyethylene cushion is fixed to the tibial prosthesis, aiding in the flexion and extension of the prosthesis. Conversely, in MB devices, the polyethylene cushion is removable, allowing the femoral prosthesis to rotate at a specific angle relative to the tibial prosthesis [98]. Both prostheses have unique advantages, limitations, and clinical applications, making them widely used in practice [99]. MB UKA has demonstrated better prosthesis-patella alignment, reduced impingement, and greater resilience to PFJ degeneration due to a more physiological knee motion. Patients undergoing MB UKA experienced stable long-term outcomes, with no observed correlation between preoperative medial PFJ degeneration severity and prosthesis longevity [27, 51, 69, 70]. In FB UKA, the femoral prosthesis position was generally more anterior relative to the intercondylar roof slope. This positioning difference, linked to biomechanics and design, as well as the use of larger prostheses, resulted in a design closely mimicking the femoral condyle’s natural form. However, this design was associated with increased patellar impingement, leading to pain and potentially reduced prosthesis survival rates [12, 44, 100]. Despite this, medial PFJ degeneration was not considered an absolute contraindication for FB UKA, as no correlation was observed between the extent of PFJ degeneration and patient recovery following FB UKA [12, 38, 78]. One study found no significant differences in knee-related scores between FB UKA patients with and without PFJ degeneration [42]. Furthermore, over a 10-year follow-up, patients with preoperative PFJ degeneration showed outcomes similar to those without degeneration, with improved flexion and extension recovery in the affected knee [38].
The potential impact of PFJ disease on UKA success is well recognized. UKA influences PFJ anatomy, relative positioning, and impingement rates. However, preoperative PFJ disease in KOA patients did not significantly affect postoperative outcomes due to improvements in PFJ conditions achieved through appropriate intraoperative management [38, 44, 50]. In UKA, the removal of damaged cartilage and restoration of joint space helps reestablish the lower extremity force line to its pre-arthritic state, reducing excessive pressure on the PFJ compartment. Additionally, patellar rim cauterization and concomitant denervation during UKA were performed to decrease PFJ impingement. UKA provided a new contact surface, allowing the patella to articulate with the prosthesis and removing damaged femoral condylar cartilage. UKA also altered the relative positioning of the femur and patella, with intermediate-term outcomes unaffected by certain preoperative conditions, such as abnormalities in patella tilt angle [101] and patella congruence angle [102]. Patients with preoperative patella or femoral excursion abnormalities experienced stress redistribution in the PFJ, leading to slowed degeneration, with a tendency for these angles to normalize postoperatively. Previous studies corroborated this, showing normalization in preoperative and postoperative patella congruence angles [61]. Early measurements of patella tilt angle and lateral PFJ displacement after UKA revealed slight increases, potentially due to proximal soft tissue tightening [60]. However, UKA may affect patellar tendon integrity, causing fibrosis, prolonged braking, and scar tissue contracture, all of which can contribute to reduced patellar height [103].
The impact of the femoral prosthesis on the patella, a significant factor contributing to anterior knee pain post-UKA. In MB UKA, meniscal pads were observed to move anteriorly and posteriorly during knee flexion and extension, reducing the risk of patellar impingement. However, patellar impingement is more common in FB UKA. Precise planning of the posterior femoral condyle osteotomy and careful selection of femoral prosthesis size are essential. Excessive osteotomy of the posterior femoral condyle can lead to prosthesis anteriority, significantly increasing the likelihood of impingement. This effect is more pronounced in lateral UKA than medial UKA, where patellar displacement during knee flexion further raises the risk of impingement [57, 58, 100].
The use of UKA has grown substantially in recent years, gaining wider recognition among both physicians and patients. This shift has been facilitated by the removal or relaxation of several procedural contraindications initially proposed by Scott [23]. Strict adherence to Scott's suggested contraindications [23] would limit suitable UKA candidates to only 10% of KOA patients [104], while broadening these criteria would raise this proportion to 47.6% [105]. PFJ diseases considered contraindications for UKA are primarily marked by abnormal anatomy. These anomalies may predispose patients to patellar maltracking and increase the risk of postoperative patellar impingement on the femoral prosthesis, potentially resulting in higher postoperative anterior knee pain and reduced mobility. Furthermore, most of these conditions are observed on the lateral side of the PFJ, likely due to the specific anatomical configuration of the patellar and trochlear lateral facets. The cartilage in this area is notably more susceptible to lesions arising from abnormal stress distribution. Consequently, it has been hypothesized that PFJ disease causing substantial changes in patellar morphology or position would negatively affect UKA postoperative outcomes. Examples include severe bone destruction, groove-like alterations in the lateral PFJ, lateral trochlear osteophytes, and lateral patellar subluxation. Conversely, it was found that less severe PFJ disease, such as mild to moderate degeneration without significant alterations, would have minimal impact on UKA outcomes. Notably, patella baja, despite potentially contributing to PFJ anatomical irregularities, does not contraindicate UKA. Additionally, potential adverse effects of patella baja on UKA may require longer observation to manifest, a consideration often overlooked due to limited relevant studies and inadequate follow-up durations. Another study reported that the incidence of PFJ cartilage lesions was unaffected by patella baja. Moreover, a majority of authors strongly contraindicated UKA in patients with severe bone destruction, groove-like alterations in the lateral PFJ, lateral patellar subluxation, lateral trochlear osteophytes, and patellar BME [42, 44, 46, 54, 57, 80]. These unfavorable results were likely due to the distinct anatomical and mechanical properties of the medial and lateral compartments. For example, it was found that medial UKA did not effectively address the biomechanical imbalance of the lateral PFJ, nor did it mitigate the risk of progressive arthritis when lateral PFJ degeneration was present [45]. Consequently, UKA is not recommended for managing these patients. Furthermore, a meta-analysis indicated that PFJ degeneration negatively affects OKS, KSS, and Flexion ROM following UKA. Notably, ROM is a recognized component of both OKS and KSS, suggesting that the primary impact of PFJ degeneration on postoperative knee function in UKA patients may be through a reduction in flexion ROM.
In applying UKA to treat patients with medial or lateral KOA combined with PFJ disease, a thorough preoperative assessment is essential due to the involvement of multiple knee compartments. TKA is recommended if the physician anticipates that the severity of the patient's condition may compromise prosthesis survival and functional outcomes following UKA. BKA involves replacing two of the three compartments while preserving the third [106], presenting a viable alternative to UKA or TKA [107]. Unlike UKA, BKA directly addresses PFJ pathology, thereby reducing its impact on operational outcomes and prosthesis longevity. Compared to TKA, BKA is associated with less trauma, fewer osteotomies [108], and lower rates of medical complications [109]. However, it is considered a complex procedure, contributing to longer operative times and lower long-term prosthesis survival rates [110], potentially due to the early limitations of PFJ prostheses [111]. BKA preserves the anterior and posterior cruciate ligaments, maintaining greater proprioception [112], knee stability, ROM [113], athletic capability [114], and natural knee kinematics [115], making it especially suitable for younger patients. Patients who undergo BKA exhibit improved short- and intermediate-term outcomes compared to those receiving TKA, with comparable long-term results [116]. For younger patients with severe PFJ degeneration, BKA is a prudent surgical option, as it eliminates the risk of unexplained postoperative pain and the need for further surgery due to osteoarthritis progression, thus avoiding TKA and enhancing surgical outcomes. In elderly patients, however, these three procedures should be selected based on the patient's symptoms and overall health.
This study provides reference on evaluating PFJ status for UKA surgical decisions, with severe degeneration possibly warranting alternative treatments such as TKA or BKA for improved outcomes. Evidence suggests that PFJ diseases-including lateral patellar subluxation, lateral trochlear osteophytes, and patellar BME-may influence the postoperative efficacy of UKA. The impact of PFJ disease on UKA outcomes may be underestimated due to short follow-up periods and the limited range of conditions assessed. The study shows that patients with PFJ degeneration experience greater extension deficits postoperatively, suggesting that biomechanical stress and altered knee mechanics may influence these outcomes. This underscores the need for tailored surgical techniques and preoperative assessments in UKA. Clinicians should focus on detailed preoperative planning and customized postoperative care to optimize surgical outcomes and patient recovery. Future research should prioritize larger, more diverse, and rigorously designed trials with extended follow-ups to accurately gauge this impact.
This study represents the second meta-analysis evaluating the impact of PFJ disease on UKA effectiveness, analyzing seven outcome metrics across fifteen studies. The inclusion of additional analyses elucidates variations in key outcomes. Furthermore, this study provides an examination of each aspect of patellofemoral joint disease. However, several limitations were (1) The number of studies included in Meta-analysis was relatively small, with variability in study quality. Additionally, the included studies encompassed patients with diverse types of PFJ disease, and disparities were observed in the diagnostic methods used preoperatively. Some studies lacked necessary clinical outcome data; (2) Both researchers conducted independent literature searches and resolved discrepancies collaboratively for final study inclusion. Despite these precautions, inherent biases and residual subjectivity could still affect the results; (3) The number of studies on each PFJ disease varied significantly, with only limited publications addressing specific PFJ conditions; (4) Only a small portion of studies focused on lateral UKA, while the majority examined medial UKA, restricting the ability to make conclusive comparisons between medial and lateral UKA outcomes and survival rates concerning PFJ disease; (5) Heterogeneity among study results likely reflects variations in clinicians' approaches to defining disease indicators. Cohort and case–control studies also typically reveal associations rather than causal relationships, which limits causal inference; (6) Variability in bias risk was observed among studies, with some showing high risk due to factors such as follow-up losses or prospective sample size estimations, potentially introducing performance and detection bias; (7) According to GRADE criteria, outcome confidence levels were low. Notable differences were observed in pooled outcomes for OKS, KSS, and ROM between groups; however, these estimates were marked by uncertainty; (8) The literature review’s comprehensiveness may be constrained by including only English-language publications.
In conclusion, while PFJ disease typically exerts a limited effect on UKA outcomes, caution is advised for cases of severe lateral PFJ degeneration, as it may restrict postoperative knee function. This study underscores the importance of PFJ evaluation in UKA procedures, particularly for managing patients with degenerative patterns that could compromise flexion ROM.
Below is the link to the electronic supplementary material.Additional file Figure S1. Publication years and study locations. Figure S2. Risk of bias graph. Figure S3. Risk of bias graph in Meta analysis. Figure S4. Risk of bias summary in Meta analysis. Figure S5a. Funnel plot of OKS, being stratified by PFJ disease types. Figure S5b. Funnel plot of KSS, being stratified by PFJ disease types. Figure S5c. Funnel plot of flexion ROM, being stratified by PFJ disease types. Figure S5d. Funnel plot of FJS, being stratified by PFJ disease types. Figure S5e. Funnel plot of Tegner activity score, being stratified by PFJ disease types. Figure S5f. Funnel plot of Prosthesis survival rate, being stratified by PFJ disease types. Table S1. Search strategy for PubMed. Table S2. Search strategy for Embase. Table S3. Search strategy for Web of Science. Table S4. Quality assessment of included studies using the Methodological Index for Non-Randomized Studies (MINORS) criteria. Table S5. Newcastle-Ottawa quality assessment of the included studies. Table S6. Overview of clinical and functional outcomes (scores) at last follow-up following UKA. Table S7. Egger’s test for the risk of bias. (PDF 34 KB)Additional file Review evidence certainty assessment (GRADE). (DOCX 661 KB)Additional file PRISMA 2020 checklist. (DOCX 33 KB)