Translate this page into:
Current outcomes of patellofemoral arthroplasty for isolated patellofemoral arthritis – A narrative review
∗Corresponding author: Rory Morrison. rorymorrison@nhs.net
-
Received: ,
Accepted: ,
This article was originally published by Reed Elsevier India Pvt. Ltd. and was migrated to Scientific Scholar after the change of Publisher.
Abstract
Abstract
Patellofemoral (PFJ) arthritis can be primary, or secondary to underlying trochlea dysplasia and patellofemoral malalignment. Although primary PFJ osteoarthritis affects an older patient population, just like tibiofemoral arthritis, it is common for younger patients to present with isolated PFJ arthritis secondary to an abnormal PFJ.
PFJ arthroplasty (PFJA) has many benefits including being less invasive, associated with lower blood loss, is more cost-effective, and leaves the kinematics of the tibiofemoral joint undisturbed. As a result, there are arguably better functional outcomes associated with PFJA, however the historical revision rate of this procedure is high.
Although registry outcome data associated with the first generation of PFJ implants shows a higher revision rate compared to TKA, the comparison of PFJ outcomes with TKA is not always age-matched and there is limited comparison on functional and patient-reported outcomes, something which is more important and relevant in a younger patient cohort. Improvements in implant design, instrumentation, surgical technique, and better patient selection has now resulted in outcomes which are comparable to that of TKA, and in some cases even better.
This narrative review outlines the current outcomes of PFJA including highlighting factors which need to be considered in optimising outcomes, as well as discussing advanced techniques of robotic assisted PFJA.
Keywords
Arthritis
Arthroplasty
Avon
Outcomes
Patellofemoral
PFJ
Robotic
Trochlea
1 Introduction
The patella is asymmetrically oval in shape with a distal apex and is the largest sesamoid bone. Its posterior surface comprises medial, lateral and odd facets which are covered by the thickest hyaline cartilage in the body, forming an articulation with the trochlea of the distal femur – the patellofemoral joint (PFJ). The femoral trochlea is bound by indistinct ridges, with the lateral ridge being most prominent. The patella begins to engage within the trochlea at 20–30° of flexion, and the contact point of the patella with the trochlea changes during the range of flexion. Typically, the anatomical congruency of the patella and the trochlea, along with surrounding soft-tissue structures such as the medial patellofemoral ligament (MPFL) stabilise the PFJ. The PFJ is subject to overall lower limb alignment and is influenced by the Q-angle, femoral and tibial torsion, as well as coronal plane angulation.
The patella acts as a fulcrum to increase the moment arm of the extensor mechanism, with patellectomy having been shown to reduce extension power by at least 15 %. The PFJ is subject to large compressive forces, and with certain activities such as deep squatting this may be up to 7x body weight.1 Activities which increase PFJ forces, or anatomical factors affecting normal PFJ congruency may therefore lead to the development of PFJ arthritis.
2 Patellofemoral arthritis
Isolated PFJ arthritis affects 10–24 % of those older than 40 years and is more common in females.2–5 The aetiology may be primary osteoarthritis, post-traumatic, or secondary to abnormal anatomy or biomechanics precipitating overloading of the PFJ cartilage. Possible causes of secondary arthritis include trochlea dysplasia, patella alta, abnormal patella tilt, lateralised extensor mechanism, soft-tissue instability or lower-limb torsional or angulation deformities.6 The higher incidence of females presenting with PFJ arthritis is thought to be due to associated underlying pathology, in particular dysplasia,2 and prevalence studies report young patients with PFJ arthritis are more likely to have preceding chronic knee pathology.4 Such patients with end-stage PFJ arthritis often present having frequently undergone previous surgical procedures including debridement, lateral release, lateral facetectomy, corrective osteotomies or trochleoplasty.7
Osteoarthritis of the knee has been shown to start in the PFJ, and in a longitudinal study of patients with a mean age of 55 presenting with knee pain, 16 % had evidence of isolated PFJ OA but no evidence of tibiofemoral arthritis at baseline.4 A systematic review of studies reporting MRI-defined PFJ arthritis in a symptomatic population report a 19 % incidence of isolated PFJ disease.8
Arthroplasty is an option for isolated end-stage PFJ arthritis which has not responded to conservative measures, with a choice between total knee replacement (TKR) and PFJ replacement. There are advantages to PFJ arthroplasty over TKR in that it is less invasive, preserves bone, is associated with less blood loss, is more cost-effective and minimises disruption to ligamentous structures of the knee, thus preserving more normal kinematics.9,10 However, PFJ arthroplasty has traditionally been associated with poor reported outcomes and high rates of early revision, and therefore there is scepticism for its use.11
3 Assessment of outcome
The requirement for revision surgery is often used as the metric to determine the success of arthroplasty. On first review of the National Joint Registry for England, Wales and Northern Ireland, headline figures for PFJ arthroplasty revision compared to TKR are alarming, and this is a similar finding across other national registries.11 For example, the Avon PFJ implant (Stryker, Mahwah, New Jersey) is the most used throughout England Wales and Northern Ireland and has a reported revision rate of 21 % at 15 years.12 The conclusion from the latest report is that the risk of revision of a PFJ replacement is at least 2.9 times higher for both males and females across all age groups at ten years, compared to a cemented TKR.12
However, there are important factors to consider when interpreting this data. The median age of patients undergoing PFJ arthroplasty is 58 (IQR 50–67) years compared to 70 (IQR 63–76) years for TKR. Younger patients are typically more active and place a higher demand on their arthroplasty and therefore the need for revision is higher.13 There is a reported three-fold higher risk of needing revision for young patients undergoing TKR compared to older patients, and revised TKRs have a much higher failure rate of 20 % at five years.14 Therefore when comparing revision rates between TKR and PFJ arthroplasty in similar aged patients, the difference in revision rate is less.
The most common cause for revision of PFJ arthroplasty is disease progression in the tibiofemoral joint, accounting for up to 50 % in some series.15–17 This should therefore not be seen as a failure of the PFJ implant, but instead represents a worsening of the patients’ condition. Patients with idiopathic PFJ OA are more likely to develop tibiofemoral OA, compared to those with secondary arthritis.18 Patient selection to identify those who may progress is the challenge, and pre-operative bone scans may help. Baker et al.16 report a retrospective review of 16 patients with pre-operative bone scans assessing for tibiofemoral arthritis and matched these to 16 patients with pre-operative MRI scans. No patient in the pre-operative bone scan group required revision whereas 31 % in the MRI group required revision at a mean follow-up of 52 months (30–105 months), and all were for OA progression.
PFJ arthroplasty may be used to delay TKR in younger patients with isolated PFJ arthritis. Due to the high incidence of re-revision of TKR in younger patients, then treating younger patients with a PFJ arthroplasty may be of benefit,19 accepting the requirement for revision if they subsequently go on to develop symptoms in the tibiofemoral joint. This strategy may mean patients are then at an age where their subsequent TKR lasts their lifetime. Encouragingly NJR data demonstrates that revised patellofemoral arthroplasties have the lowest risk of re-revision at 10 years compared to all other forms of arthroplasty (Fig. 1)12 which supports this approach.

There are differences in the reported reasons for revision between registry data and protocolled clinical studies, with a recent systematic review reporting progression of OA to be the most common reason for revision in studies, compared to pain and aseptic loosening in registry data.20 One possible reason for this difference may be surgeons have a higher tendency to scrutinize those patients entering a clinical study, or to record the specific reason for revision more accurately, compared to registry input. Furthermore, better results may be reported in clinical studies, where the participating surgeons may be PFJ enthusiasts, compared to results from registry data. On review of the NJR, most cases of PFJ arthroplasty are carried out by surgeons performing only 1–4 cases per year, with a mean of 3.7, compared to 52 cases for those performing TKR.7,12 In a reported series of 103 Avon PFJ replacements from a unit with surgeons performing high-volume PFJ arthroplasty, the revision rate was 9.7 % at up to 14 years21 compared to 21 % at 10 years using NJR data.12
Whilst need for revision, or post-operative complications, have been traditionally seen as the end-point for arthroplasty, Patient Reported Outcome Measures (PROMs) are increasingly used to evaluate success based on quality of life and functional outcome.22 Their use to judge success may therefore be more important in a younger more functionally-demanding cohort of patients such as those undergoing PFJ arthroplasty.
Recent systematic reviews have reported comparative outcomes between TKR and PFJ arthroplasty,9,15 and encouragingly some report better PROMs in the PFJ group.23,24 In a retrospective matched cohort review comparing PFJ arthroplasty versus TKR for isolated PFJ arthritis, those who underwent PFJ arthroplasty had better range of movement, Knee Society Scores and Forgotten Joint Scores at three years following surgery.3 Similar outcomes have been reported in a prospective randomised controlled trial comparing the Avon PFJ arthroplasty with TKR. At two years the PFJ group had better range of movement and PROMs, with a quicker return to baseline function and no difference in complications.25 Pogorzelski et al. report 94 % of patients with a mean age of 46 years returned to the same or higher level of sports following PFJ arthroplasty, and 74 % reported an improved ability to perform sports.26
4 Optimising outcomes
There are four key steps which can help to optimise post-operative outcomes. These are:1.Choosing the right patient2.Adopting an informed shared decision-making approach3.Using modern implant designs and instrumentation, including robotic replacement4.Paying careful attention to surgical detail
4.1 Patient selection and shared decision-making approach
Choosing the correct patient for surgery is vital to try to minimise poor post-operative outcomes. The ideal age of patient for PFJ arthroplasty is between 40 and 60.27 In the history, the patient's symptoms should be localised to the PFJ, and activities which stress the PFJ and worsen pain, such as stairs and squatting, are important to ask about. Previous PFJ pain, episodes of instability or a history suggestive of rotational or valgus deformities should be enquired about. Patients have often received treatment for PFJ symptoms previously, and these should be noted.7 It is important to note any history of inflammatory arthritis.
The BMI of the patient should be recorded, as whilst there are some conflicting reports, a high BMI is usually associated with poorer post-operative outcomes.10,18,24,28 Clinical examination should seek to evaluate discomfort and irritability of the PFJ, with no symptoms arising from the tibio-femoral joint. Evidence of fixed flexion, alignment or rotational abnormalities, patella alta and the presence of a lateralised or maltracking extensor mechanism should be examined. Core, hip and knee muscle strength should be evaluated.
A plain radiograph series of weightbearing AP, lateral, skyline and Rosenberg views should be obtained, to confirm presence of PFJ OA, as well as to exclude tibiofemoral arthritis. An MRI scan is useful if there is any uncertainty on x-rays due to its high sensitivity for detecting cartilage changes.29,30 A bone scan may also be utilised and may predict progression of osteoarthritis, due to activity of the underlying subchondral bone.31
If history, examination, and investigations confirm isolated PFJ OA which has failed non-operative treatment, then PFJ arthroplasty can be discussed with the patient. This is an important conversation to ensure the patient understands the rationale behind PFJ arthroplasty, as well as the reported higher revision rates in PFJ arthroplasty compared to TKR, and some of the possible reasons for this. The patient should be informed that PFJ arthroplasty is a joint-preserving option and may be used as part of a staged approach to management of their arthritis if they are young. The risk of disease progression in the tibiofemoral joint should be discussed, and that revision surgery to either add in a unicompartmental replacement or to revise the prosthesis to a TKR may be required in the future. Patients should be aware that this does not reflect failure of the PFJ arthroplasty. A shared decision-making approach is vital to ensure the patient is fully informed of the benefits and risks of PFJ arthroplasty.
4.2 Implant selection
The first isolated PFJ replacement was in 1955, where McKeever utilised a vitallium shell to replace the patella, leaving the native trochlea intact. Whilst early results were encouraging, patients' symptoms deteriorated within 5–7 years and this was thought to be due to not to addressing the trochlea. Lubinus and Blazina designed the first complete PFJ arthroplasty in 1979, with a polyethelene patella component and a metallic trochlea which had a short-anterior flange and a narrow trochlea groove. This, along with other first generation PFJ arthroplasties, utilised an ‘inlay’ approach, where the PFJ was inserted in the native trochlea to replace the articular cartilage. The subchondral bone remained untouched and any rotational malalignment was not corrected. First-generation PFJ arthroplasties had many limitations including rudimentary instruments, limited component size options and poorer surgical techniques, and along with poorer patient selection, was associated with poor outcomes, including patellar maltracking and clunking, instability, soft-tissue impingement. There was a consequential re-operative rate of up to 35 % at 5-years.
Second and third-generation PFJ arthroplasty designs have attempted to address these problems. New implants utilise an ‘onlay’ or trochlea cutting technique with better instrumentation which permits a greater control of implant positioning in the three planes to address underlying bony abnormalities. Newer trochlea implants have a larger more proximally-extending anterior flange, a radius of curvature similar to TKR, a thinner lateral margin, a wider trochlea groove and have been designed to prevent impingement on the anterior cruciate ligament and the femoro-tibial joint.
The patella component is comprised of polyethylene and may have an anatomic design or more commonly a dome design. Typically the patella design is governed by its compatibility with TKR components in case future revision is required.7
Current reviews report 10-year survivorship of 90 % using these newer designs of implants, with revision and complication rates more comparable to TKR.9,32,33
4.3 Surgical technique
The goals of PFJR are to have appropriately sized implants which do not ‘overstuff’ the PFJ, and which allow normal patella tracking throughout the range of movement with no tilt, subluxation or clunk. Modern instrumentation associated with second and third generation (onlay) implants.
4.3.1 Trochlea component
In the coronal plane, the implant should be in line with the anatomical axis of the femur, to aid in engagement of the patella as the knee moves from extension to flexion. In the sagittal plane, the component should be flush to the anterior cortex, or slightly flexed (Fig. 2). An extended trochlea component can lead to overstuffing, proximal anterior notching or possible impingement against the patella when the knee moves from flexion to extension. In the axial plane the implant should be either neutral or externally rotated. Optimum rotation is identified by a drop rod lining up with the tip of the medial malleolus.34

There should be a smooth transition from the implant to surrounding cartilage, and the implant should be adequately sized from medial to lateral to permit this. The component must not overhang and therefore under sizing rather than oversizing is preferable.
4.3.2 Patella component
The patella component should be sized to restore normal patella thickness (Fig. 3). In cases where there is no bone loss, then 2–3 mm should be added to the patella thickness measurement to compensate for lost articular cartilage. There should be at least 12 mm of patella bone left to avoid fracture when drilling the lug holes for the patella.35 Post-operative radiographs should be obtained to evaluate component placement (Fig. 4).


4.4 Robotic PFJ arthroplasty
There is increasing use of robotic-assisted arthroplasty and there are advantages for its use specifically in PFJ arthroplasty.36,37 Deciding on the optimal placement of the trochlea component for correct implant-cartilage transition can be difficult, especially if there is significant cartilage loss or if the trochlea has severe underlying dysplasia with loss of normal reference points. Pre-operative planning can help to ensure the correct trochlea implant position in all three planes, and this can plan can then be executed more reliably intra-operatively with the use of the robot to help prevent extreme implant positioning and to fine tune optimum position to achieve the correct cartilage-implant transition.
The long-term outcomes of robotic PFJ arthroplasty are awaited, however there is increasing evidence to suggest that good early outcomes can be achieved with correct implant cartilage transition.36 Robotic arthroplasty requires some specific techniques and these have been highlighted previously by the senior author.38 Future work to evaluate the influence of optimum component placement in three planes on clinical outcomes is of interest, and the use of robotics allows for precise referencing of the implant to help further this knowledge.
5 Conclusion
Whilst PFJ arthroplasty has traditionally been associated with higher revision rates compared to TKR, this is no longer the case and there are clear indications for its use. Correct patient selection with thorough examination and pre-operative work-up with imaging is important. The use of modern implant designs and instrumentation has improved survivorship of PFJ arthroplasty, and the most common reason for revision is progression of arthritis rather than failure of the implant. PFJ arthroplasty may be used as a staged approach to management of arthritis in younger patients, and a shared decision-making approach is vital. Future studies, particularly using robotics to optimise implant placement and allow accurate evaluation of this may help improve outcomes further.
Disclosures
Neither author declares any relevant disclosures or conflict of interest in relation to this narrative review.
Funding
No funding has been received for this narrative review.
CRediT authorship contribution statement
Rory Morrison: Resources, Writing – original draft, Writing – review & editing. Vipul Mandalia: Conceptualization, Resources, Writing – original draft, Writing – review & editing, Supervision.
References
- Biomechanics and pathomechanics of the patellofemoral joint. Int J Sports Phys Ther. 2016;11(6):820-830.
- [Google Scholar]
- Joint awareness after patellofemoral arthroplasty evaluated with the forgotten joint score: a comparison study. Orthop Surg. 2021;13(3):833-839.
- [Google Scholar]
- Incidence, prevalence, natural course and prognosis of patellofemoral osteoarthritis: the Cohort Hip and Cohort Knee study. Osteoarthritis Cartilage. 2017;25(5):647-653.
- [Google Scholar]
- The compartmental distribution of knee osteoarthritis – a systematic review and meta-analysis. Osteoarthritis Cartilage. 2021;29(4):445-455.
- [Google Scholar]
- Is tibiofemoral or patellofemoral alignment or trochlear morphology associated with patellofemoral osteoarthritis? A systematic review. Arthritis Care Res. 2016;68(10):1453-1470.
- [Google Scholar]
- The evolution and role of patellofemoral joint arthroplasty. Bone Jt Res. 2020;7(12):636-638.
- [Google Scholar]
- The prevalence of radiographic and MRI-defined patellofemoral osteoarthritis and structural pathology: a systematic review and meta-analysis. Br J Sports Med. 2017;51(16):1195-1208.
- [Google Scholar]
- Midterm results of modern patellofemoral arthroplasty versus total knee arthroplasty for isolated patellofemoral arthritis: systematic review and meta-analysis of comparative studies. Arch Orthop Trauma Surg. 2022;142(5):851-859.
- [Google Scholar]
- Advances in patellofemoral arthroplasty. Curr Rev Musculoskelet Med. 2018;11(2):221-230.
- [Google Scholar]
- Short-term revision risk of patellofemoral arthroplasty is high: an analysis from eight large arthroplasty registries. Clin Orthop Relat Res. 2020;478(6):1222-1231.
- [Google Scholar]
- Younger age increases the risk of early prosthesis failure following primary total knee replacement for osteoarthritis: a follow-up study of 32,019 total knee replacements in the Finnish Arthroplasty Register. Acta Orthop. 2010;81(4):413-419.
- [Google Scholar]
- Patellofemoral arthroplasty: current concepts. Journal of Clinical Orthopaedics and Trauma. 2018;9:24-28.
- [Google Scholar]
- Similar postoperative patient-reported outcome in both second generation patellofemoral arthroplasty and total knee arthroplasty for treatment of isolated patellofemoral osteoarthritis: a systematic review. Knee Surg Sports Traumatol Arthrosc. 2019;27(7):2226-2237.
- [Google Scholar]
- Isolated patellofemoral joint arthroplasty: can preoperative bone scans predict survivorship? J Arthroplasty. 2020;35(1):57-60.
- [Google Scholar]
- Why do patellofemoral arthroplasties fail today? A systematic review. Knee. 2017;24(1):2-8.
- [Google Scholar]
- Patellofemoral arthroplasty versus total knee arthroplasty in patients with isolated patellofemoral osteoarthritis. Am J Orthoped. 2010;39(10):487-491.
- [Google Scholar]
- The clinical outcome of patellofemoral arthroplasty vs total knee arthroplasty in patients younger than 55 years. J Arthroplasty. 2019;34(12):2914-2917.
- [Google Scholar]
- Failure modes of patellofemoral arthroplasty—registries vs. clinical studies: a systematic review. Acta Orthop. 2019;90(5):473-478.
- [Google Scholar]
- Mid-term survivorship and clinical outcomes of the Avon patellofemoral joint replacement. Knee. 2018;25(2):323-328.
- [Google Scholar]
- Patient-reported outcome measures in total joint arthroplasty: defining the optimal collection window. Arthroplast Today. 2020;6(1):62-67.
- [Google Scholar]
- Patient-related outcomes of patellofemoral arthroplasty: experience of a single center. Arthroplasty. 2021;3(1):3-8.
- [Google Scholar]
- Patellofemoral arthroplasty versus total knee arthroplasty for isolated patellofemoral osteoarthritis: a systematic review and meta-analysis. J Orthop Surg Res. 2021;16(1):1-12.
- [Google Scholar]
- The mark coventry award: patellofemoral arthroplasty results in better range of movement and early patient-reported outcomes than TKA. Clin Orthop Relat Res. 2018;476(1):87-100.
- [Google Scholar]
- Reliable improvements in participation in low-impact sports following implantation of a patellofemoral inlay arthroplasty at mid-term follow-up. Knee Surg Sports Traumatol Arthrosc. 2021 Oct;29(10):3392-3399.
- [Google Scholar]
- Patellofemoral joint arthroplasty: patient selection, surgical technique and outcomes. Orthop Traumatol. 2021;35(1):56-63.
- [Google Scholar]
- Patellofemoral arthroplasty: obesity linked to high risk of revision and progression of medial tibiofemoral osteoarthritis. Knee Surg Sports Traumatol Arthrosc. 2022;30(12):4115-4122.
- [Google Scholar]
- Accuracy of magnetic resonance imaging in assessing knee cartilage changes over time in patients with osteoarthritis: a systematic review. North Clin Istanbul. 2021;9(4):414-418.
- [Google Scholar]
- The diagnostic performance of MRI in osteoarthritis: a systematic review and meta-analysis. Osteoarthritis Cartilage. 2012;20(1):13-21.
- [Google Scholar]
- Prediction of the progression of joint space narrowing in osteoarthritis of the knee by bone scintigraphy. Ann Rheum Dis. 1993;52(8):557-563.
- [Google Scholar]
- Patellofemoral arthroplasty: current concepts and review of the literature. Joints. 2017;5(4):237-245.
- [Google Scholar]
- The short-term effectiveness and safety of second-generation patellofemoral arthroplasty and total knee arthroplasty on isolated patellofemoral osteoarthritis: a systematic review and meta-analysis. J Orthop Surg Res. 2021;16:1-10.
- [Google Scholar]
- The correct rotation of the femoral component in patellofemoral replacement: a laboratory assessment of a surgical technique. J Bone Jt Surg - Ser B. 2012;94 B(12):1637-1640.
- [Google Scholar]
- Effect of patellar thickness on early results of total knee replacement with patellar resurfacing. Knee Surgery. Sport Traumatol Arthrosc. 2014;22(12):3093-3099.
- [Google Scholar]
- Robotic-assisted patellofemoral replacement-correlation of preoperative planning with intraoperative implant position and early clinical experience: a minimum 2-year follow-up. J Knee Surg. 2022;35(7):731-738.
- [Google Scholar]
- Minimally invasive robotic-assisted patellofemoral arthroplasty. Arthrosc Tech. 2020;9(4):e425-e433.
- [Google Scholar]
- Robotic assisted patellofemoral joint replacement: surgical technique, tips and tricks. Indian J Orthop. 2022;56(12):2110-2118.
- [Google Scholar]
