Roentgen Stereophotogrammetric Analysis (RSA) can predict long-term outcome of prostheses by measuring migration over time. The Exeter femoral stem is a double-tapered highly polished implant and has been shown to subside within the cement mantle in 2 year RSA studies. It has a proven track record in terms of long-term survivorship and low revision rates. Several studies have demonstrated excellent clinical outcomes following its implantation but this is the first study to assess stem migration at 10 years, using RSA.

This is a single-centre study involving 20 patients (mean age: 63 years, SD=7) undergoing primary total hip replacement for degenerative osteoarthritis using the lateral (Hardinge) approach. RSA radiographs were taken with the patient bearing full weight post-operatively, at 3, 6, 12 months and at 2, 5 and 10 years follow-up. The three-dimensional migration of the Exeter femoral stem was determined.

The mean Oxford Hip Score at 10 years was 43.4 (SD=4.6) and there were no revisions. The stems subsided and rotated internally during a 10-year period. The mean migrations of the head and tip of the femoral stem in all three anatomic directions (antero-posterior, medio-lateral & supero-distal) were 0.69 mm posterior, 0.04 mm lateral and 1.67 mm distal for the head and 0.20 mm anterior, 0.02 mm lateral and 1.23 mm distal for the tip. The total migration at 10 years was 1.81 mm for the head and 1.25 mm for the tip.

The Exeter femoral stem exhibits migration which is a complex combination of translation and rotation in three dimensions. Comparing our 10 year with our previous 2 year migration results, the Exeter stems show continued, but slow distal migration and internal rotation. The subsidence continues to compress the cement and bone-cement interface which maintains secure fixation in the long term.

About ten years ago we introduced sophisticated instrumentation and an increased range of component sizes for the Oxford unicompartmental knee replacement (UKR) to facilitate a minimally invasive surgical (MIS) approach. The device is now routinely implanted through an incision from the medial pole of the patella to the tibial tuberosity. This has resulted in a more rapid recovery and an improved functional result. As the access to the knee is limited there is a concern that the long term results may be compromised. The aim of this study was to determine the 10 year survival.

A prospective follow up of all Phase 3 minimally invasive Oxford UKR implanted by two senior authors (DWM & CAFD) has been undertaken. So far 1015 UKRs have been implanted for anteromedial osteoarthritis. All patients received a cemented implant through a MIS approach and were followed up prospectively by an independent observer. The data was collected prospectively regarding pre-operative status, complications and clinical as well as functional outcome at predetermined intervals.

The average age of patients was 66.4 years (range: 33 – 88) with mean Oxford Knee Score 41 (SD: 7.9) at the time of last follow up, Knee Society Score (objective) of 84 (SD: 13) and Knee Society Score (functional) of 83 (SD: 21). At ten years the survival of this cohort is 96%. There were 22 revisions including 7 for progression of arthritis, 5 for infection, 5 for bearing dislocation, 4 for unexplained pain and one for rupture of ACL secondary to trauma.

We conclude that the Oxford Knee can be implanted reliably through a minimally invasive approach, giving excellent long term results.

Introduction: This study explores whether modern magnetic resonance imaging (MRI) with improved cartilage sequencing is able to show a more detailed view of antero-medial osteoarthritis of the knee (AMOA) than previously, so enabling a radiographic description of this common phenotype of disease. Modern MRI technology allows us to visualize in great detail the structures and cartilage within the knee, providing a better understanding of the pathoanatomy of AMOA. This description of the end stage of disease is useful as a baseline when investigating the progression of arthritis through the knee. Preoperative assessment of patients and selection of intervention is very important and preoperative imaging forms an integral part of this. This will also be useful in preoperative assessment and surgical management of patients.

Methods: 50 patients with a radiographic diagnosis of anteromedial osteoarthritis of the knee and had been listed for unicompartmental knee arthroplasty (UKA) had MRI as part of their pre-op workup. At operation all were deemed suitable for UKA using the current Oxford indications. The image sequences were coronal, axial and sagittal with a predetermined cartilage protocol. The state of the ACL, cartilage wear degree and location, presence and pattern of osteophytes, meniscal anatomy and subchondral high signal were assessed.

Results: All the ACLs were visualized and in continuity, however 40% showed intrasubstance high signal. 100% of medial compartments showed full thickness anteromedial loss with preservation of the posteromedial cartilage. When present, the meniscus was extruded in 75% of cases.

90% of lateral compartments were normal and none had full thickness cartilage loss. However 10% showed high signal in the tibial plateau. There was a highly reproducible pattern of osteophyte formation; 94% posteromedial and posterolateral aspect of medial femoral condyle; 90% medial tibial; 80% medial femoral and 84% lateral intercondylar notch.

Discussion: This study maps the pattern of anteromedial osteoarthritis using modern MRI techniques. This creates a baseline description of disease which is useful when investigating disease progression. This also has importance in determining preoperative indications (preservation of ACL and posteromedial cartilage); surgical technique (determine pattern of osteophytes requiring resection) and potentially important for long-term outcome (early lateral compartment changes).

Metal on Metal Hip Resurfacing Arthroplasty (MoMHRA) has gained popularity due to its perceived advantages of bone conservation and relative ease of revision to a conventional THR if it fails. Known MoMHRA-associated complications include femoral neck fracture, avascular necrosis/collapse of the femoral head/neck, aseptic loosening and soft tissue responses such as ALVAL and pseudotumours. This study’s aim was to assess the functional outcome of failed MoMHRA revised to THR and compare it with a matched cohort of primary THRs.

Method: We have revised 53 MoMHRA cases to THR; the reasons for revision were femoral neck fracture (Fracture Group, n=21), soft tissue reaction (Pseudotumour Group, n=16) and other causes (Other Group, n=16: loosening, AVN and infection). Average followup was 2.9 years. These MoMHRA revisions were compared with 103 matched controls from a primary THR cohort; matched for age, gender and length of followup. We compared, using the MannWhitney U test, operative time (OT, measured in minutes), and Oxford Hip Score (OHS) between the revised MoMHRA groups and their individual controls. We also compared, using the Kruskal Wallis test, UCLA Activity Score in the revised MoMHRA groups.

Results: There were no differences between the Fracture Group (mean OT 99.6, SD: 30.4; mean OHS 19.8, SD:9.2) and its controls (mean OT 95.9, SD: 31.8; mean OHS 17.3, SD: 7.5) nor between the Other Group (mean OT 129.4, SD: 36.7; mean OHS 22.2, SD: 9.4) and its controls (mean OT 104.4, SD: 39.2; mean OHS 20.3, SD: 10.1) in terms of OT and OHS. The Pseudotumour Group had significantly longer OT (mean 161.6, SD: 24.5, p< 0.001) and worse outcome (mean OHS 39.1, SD: 9.3, p< 0.001) than its controls (mean OT 113.1, SD: 51.7; mean OHS 20.0, SD: 9.2). In the Fracture Group, there were 3 infections requiring revisions. For the Pseudotumour Group, there were 3 recurrent dislocations, 1 femoral artery stenosis and 3 femoral nerve palsies. In the Other Group, there were 2 periprosthetic fractures. There was significant difference (p< 0.001) in UCLA scores between the MoMHRA groups. The Pseudotumour Group had the lowest mean UCLA score of 3.8 (SD: 1.89). The Fracture Group (mean: 7.0, S.D. 2.0) and the Other Group (mean: 6.7, S.D. 2.1) had similar UCLA scores.

Discussion: The results demonstrate that outcome after revision of MoMHRA is dependent upon the indication for revision. Patients revised for soft tissue reactions had significantly worse outcome. Patients with soft tissue reactions are more likely to experience complications and require further surgical intervention. The pseudotumour associated revisions were associated with a significantly prolonged OT. The overall complication rate for the study groups was quite high, with 11 (21%) revised MoMHRA cases experiencing a complication. The Pseudotumour Group had a higher complication rate (37%).

Patello-femoral instability (PFI) affects 40 individuals per 100,000 population and causes significant morbidity. The causes of patello-femoral instability are multi-factorial, and an isolated anatomical abnormality does not necessarily indicate instability. Patello-femoral subluxation ranges from 0% (stable patella tracking) to 100% (dislocation) and there is an established relationship between the amount of subluxation and anterior knee pain. Traditionally, magnetic resonance (MR) imaging and standard radiographs are used to guide the clinician towards a suitable corrective procedure for PFI. The multi-factorial nature of patello-femoral instability is not addressed with current imaging techniques. This study aims to address which anatomical variables assessed on MR images are most relevant to patello-femoral subluxation. This information will aid surgical decision making, particularly in selecting the most appropriate reconstructive surgery.

A retrospective analysis of MR studies of 60 patients with suspected patello-femoral instability was performed. All patients were graded for degree of subluxation using a dynamic MR scan.

The patient scans were assessed for the presence of a specific range of anatomical variables:

patella alta, (modified Insall-Salvatti)

The Wilk’s Lambda test for multi-variate analysis was used to establish whether any relationship was present between the degree of patello-femoral instability and bony or soft tissue anatomical variables. Non-parametric statistical tests were applied across the groups and within the groups to assess their relative significance.

The following variables showed a significant relationship with patellofemoral subluxation; distance of the VMO from the patella (< 0.001), TTD (< 0.001), patella engagement (0.001), sulcus angles (0.004) and patella alta (0.005).

This study agrees with previous work showing a significant correlation between subluxation and trochlea sulcus angle and TTD.

This is the first study to establish a significant correlation between patella engagement and radiological instability. The lower the percentage engagement of the patella in the trochlea, the greater the degree of patello-femoral instability. Patella engagement showed a more significant relationship with subluxation than patella alta.

We report a new method of predicting patello-femoral instability by measuring the overlap of the patella in the trochlea groove.

Introduction: Most unicompartmental knee replacements (UKRs) employ cement for fixation of the prosthetic components to the bone. The information in the literature about the relative merits of cemented and cementless UKR is contradictory, with some favouring cementless fixation while others favouring cemented fixation. There is concern about the radiolucency which frequently develops around the tibial component with cemented fixations. The exact cause of the occurrence of radiolucency is unknown but according to some, it may suggest suboptimal fixation.

Method: Following ethical approval, 62 patients with medial OA were randomised to receive either cemented (n=31) or cementless components (n=31). All patients underwent identical surgical procedure with either a cemented or cementless Oxford UKR. Patients were assessed clinically and radiologically. The x-rays were taken with an image intensifier (I.I.). The position of the I.I. was adjusted until it was perfectly aligned with the tibial bone-implant interface thereby allowing accurate assessment of presence and extent of the radiolucency.

Results: The patients in the two groups were well matched. There was no significant difference in the clinical scores between the two groups. The mean OKS for the cemented group was 40 (± 8.3) and 42 (± 4.6) for cementless group. Narrow radiolucent lines were seen at the bone-implant interfaces of 75% of the cemented tibial components; partial in 43% and complete in 32%. In the cementless implants, partial radiolucencies were seen in 7% and complete radiolucencies in none. The differences are statistically highly significant (p< 0.0001) and imply satisfactory bone ingrowth into the cementless implants.

Conclusions: The method of fixation influences the incidence of radiolucency. With identical designs, no patient with cementless components developed any complete radiolucency. The observation raises the question as to whether cementless rather than cemented components should be routinely used for UKR.

Purpose: The purpose of this study was to report the mid-term survival results of Oxford UKAs in patients of 50 years of age or less, using (1) revision surgery and (2) Oxford Knee Scores (OKS) as outcome measures.

Method: A literature review identified studies of Oxford mobile bearing UKAs containing individuals 1) 50 years old or less with 2) medial osteoarthritis and 3) 2 years or longer follow-up. Authors were approached to participate in a multi-centre survival analysis by submitting all their patients, 50 years of age or less, who received a medial UKA for osteoarthritis. Patients who had died, been lost to follow-up or who underwent revision were identified. OKS were established for all patients with surviving implants.

Results: Seven centres submitted 107 patients. The mean age was 47 years (range 32–50). The average follow-up was 4 years (range 1–25). Forty-seven patients had follow-up into their fifth year or longer. The cumulative 7-year survival using revision as the endpoint was 96% (CI 8). The mean post-operative OKS for surviving implants was 38 (CI 2) out of a possible 48.

Conclusion: While early survival rates and function are encouraging, long-term follow-up is required before concluding UKA is a viable treatment option in young patients with unicompartmental knee arthritis.

Introduction: We investigated the relationship between psychological distress and outcome after total hip replacement (THR) in the Exeter Primary Outcome Study (EPOS).

Materials & Methods: Data were collected from a number of centres across England between January 1999 and January 2002 for patients undergoing primary hip replacement using the cemented Exeter femoral component (Stryker). We recorded the Oxford Hip Score (OHS) for physical function and SF36 questionnaire for both physical and mental domain assessment annually for five years. We dichotomised the patients into the mentally distressed (MHS < 50) and the not mentally distressed (MHS ≥ 50) groups based on their pre-operative Mental Health Score (MHS) from the SF36 score.

Results: Complete data were available for 455 (407 not distressed and 48 distressed) patients. Pre-operative OHS and SF-36 score was significantly worse in the distressed group (both p< 0.001). Mean OHS improved from 43 to 20 at 1 year after surgery and remained the same thereafter in the non distressed group. In the mentally distressed group pre-operative mean OHS of 48 improved to 22 at 1 year after surgery. Maximum improvement in OHS occurred in the 1st yr after surgery. Mean MHS improved from 76 to 81 at 1 year after surgery and remained the same thereafter in the non distressed group. Mean MHS improved from 35 to 62 at 1 year after surgery, reaching 65 at 5 years after surgery in the mentally distressed group. The maximum improvement in MHS occurred in the 1st yr after surgery.

Conclusion: Pre-operative psychological distress did not compromise functional outcome after hip arthroplasty. There was a substantial improvement in mental distress in patients with mental distress prior to surgery. Both groups of patients experienced improvement in Oxford Hip Score, which was maximal by 1 year after surgery and was maintained over the 5 year follow up.

Purpose: This study was designed to establish the poly-ethylene wear rates in the Oxford medial unicompert-mental knee replacement.

Introduction: The Oxford meniscal bearing knee was introduced as a design to reduce polyethylene wear. There has been one previous retrieval study of the Oxford UKA, which reported very low wear rates in some specimens, but abnormal patterns of wear in others, including impingement. There has been no further investigation of these abnormal wear patterns.

Methods: Forty-seven bearings were retrieved from patients who had received a medial Oxford UKA for anteromedial osteoarthritis of the knee, none of which had previously been studied. Mean time to revision was 8.4 years (SD 4.1) and 20 had been implanted for over 10 years. The macroscopic pattern of polyethylene wear and the linear penetration (dial gauge measurement) was recorded for each bearing.

Results: The mean linear penetration rate (LPR) was 0.07mm/year. The patterns of wear fell into 4 categories, each with a different LPR; 1) No abnormal macroscopic appearance, n=16 (LPR = 0.01mm/year), 2) Abnormal macroscopic wear with extra-articular impingement, n=16 (LPR = 0.05mm/year), 3) Abnormal macroscopic wear with intra-articular impingement, n=6 (LPR = 0.10mm/year), 4) Abnormal macroscopic wear with impingement and signs of incongruous articulation, n=9 (LPR = 0.14mm/year). The differences in LPR were statistically significant (p< 0.05).

Conclusion: The results show that very low polyethylene wear rates are possible if the device functions normally. However if the bearing displays abnormal function (extra-articular, intra-articular impingement or incongruous articulation) wear rates increase significantly.

Introduction: Metal on metal Hip Resurfacing Arthroplasty (MoMHRA) has gained popularity due to its perceived advantages of bone conservation and relative ease of revision to a conventional THR if it fails. This retrospective study is aimed at assessing the functional outcome of failed MoMHRA revised to THR and comparing it with a matched cohort of primary THRs.

Method: Since 1999 we have revised 53 MoMHRA to THR. The reasons for revision were femoral neck fracture (Group A, n=21), pseudotumour (Group B, n=16) and other causes (Group C, n=16: loosening, avascular necrosis and infection). Average follow-up was 3 years months (1.2–7.3). These revisions were compared with 106 primary THRs which were age, gender and follow-up matched with the revision group in a ratio of 2:1.

Results: The mean Oxford Hip Score (OHS) was 20.1 (12–51) for group A, 39.1 (14– 56) for group B, 22.8 (12–39) for group C and 17.8 (12–45) for primary THR group. In group A, there were three infections requiring further revisions. In group B, there were three recurrent dislocations, three patients with femoral nerve palsy and one femoral artery stenosis. In group C, there were no complications. The differences in clinical and functional outcome between group B and the remaining groups as well as the difference in the outcome between group B and control group were statistically significant (p < 0.05).

Conclusions: THR for failed MoMHRA was associated with significantly more complications, operation time and need for blood transfusion for the pseudotumour group. In addition, the revisions secondary to pseudotumour also had significantly worse functional outcome when compared to other MoMHRA revisions or primary THR.

Introduction: Despite the satisfactory short-term implant survivorship of MoM hip resurfacing arthroplasty, symptomatic abnormal periprosthetic soft-tissue masses relating to the hip joint, ‘pseudotumours’, are being increasingly reported. These were found be locally destructive, requiring revision surgery in 75% of patients. Asymptomatic pseudotumours have not been previously investigated.

Methods: The aims were: (1) to investigate the prevalence of asymptomatic pseudotumours; and (2) to investigate their potential association with the level of metal ions. A total of 160 hips in 123 patients with a mean age 56 years (range 33–73) were evaluated at a mean follow-up of 61 months (range 13–88). Radiographs and OHS were assessed. Patients with a cystic or solid mass detected on the ultrasound/MRI had an aspiration or biopsy performed. Cobalt and chromium levels were analysed using Inductively-Coupled Plasma Spectrometer.

Results: Pseudotumours were found in 6 patients (5F: 1M). In 80% of bilateral cases, it was found in both sides. Histological examination showed extensive necrosis and diffuse lymphocyte infiltration. The presence of pseudotumour was associated with higher serum cobalt (9.2 μg/L vs. 1.9μg/L, p< 0.001) and chromium levels (12.0μg/L vs. 2.1μg/L, p< 0.001); higher hip aspirate cobalt (1182 μg/L vs. 86.2μg/L, p=0.003) and chromium levels (883μg/L vs. 114.8μg/L, p=0.006); and with inferior OHS (23 vs. 14 p=0.08).

Discussion: The prevalence of asymptomatic pseudotumour (5%) was higher than previously reported for the symptomatic pseudotumours (1%). There was a sixfold elevation of serum and a twelve-fold elevation of hip aspirate levels of cobalt and chromium in patients with pseudotumours. This suggests that pseudotumours may be a biological consequence of the large amount of metal debris generated in vivo. The association between pseudotumour and elevated metal ion levels might theoretically be explained by either systemic hypersensitivity responses to metal ions or local cytotoxic effects due to a high level of metal ions.

Introduction: This study explores whether modern magnetic resonance imaging (MRI) with improved cartilage sequencing is able to show a more detailed view of anteromedial osteoarthritis of the knee (AMOA). Preoperative assessment of patients and selection of intervention is very important and preoperative imaging forms an integral part of this. Modern MRI technology may allow us to visualize in great detail the structures and cartilage within the knee, providing a better understanding of the pathoanatomy of AMOA. This will be useful in preoperative assessment and surgical management of patients.

Methods: 50 patients with a radiographic diagnosis of anteromedial osteoarthritis of the knee and had been listed for unicompartmental knee arthroplasty (UKA) had MRI as part of their pre-op workup. At operation all were deemed suitable for UKA using the current Oxford indications. The image sequences were coronal, axial and sagittal with a predetermined cartilage protocol. The state of the ACL, cartilage wear location and pattern, presence of osteophytes and subchondral high signal were assessed.

Results: All the ACLs were visualized and in continuity, however 40% showed intrasubstance high signal.

100% of medial compartments showed full thickness anteromedial loss with preservation of the posteromedial cartilage. When present, the meniscus was extruded in 96% of cases.

Discussion: This study maps the pattern of anteromedial osteoarthritis using modern MRI techniques. This has importance in determining preoperative indications (preservation of ACL and posteromedial cartilage); surgical technique (determine pattern of osteophytes requiring resection) and potentially important for long-term outcome (early lateral compartment changes).

Introduction: Although resurfacing hip replacements are widely used there are few little independent outcome data to support this. The aim of this study was to report the 5 year clinical outcome and 7 year survival of an independent series.

Method: 610 Birmingham hip resurfacings were implanted in 532 patients with an average age of 51.8 years (range 16.5–81.6 years) and were followed for between 2 to 8 years; 120 of this series had minimum five year follow-up. Two patients were lost. There were 23 revisions, giving an overall survival of 95% (95% CI 85–99%) at seven years. Fractured neck of femur (n=13) was the most common reason for revision, followed by aseptic loosening (n=4). There were also 3 patients who had failures that were possibly related to metal debris. At a minimum of 5 year follow-up 93% had excellent or good outcome according to the Harris Hip Score. The mean Oxford Hip Score was 16.1 points (SD 7.7) and the mean UCLA activity score was 6.6 points (SD 1.9). There were no patients with definite evidence of radiographic loosening or greater than 10% of neck narrowing.

Discussion: The results demonstrate that with the Birmingham Hip Resurfacing, implanted using the extended posterior approach, the five year survival is similar or better to the reported survival rates for cemented and hybrid THR’s in young patients.

Conclusions: Considering these patients are young and active these results are good and support the use of resurfacing. However, further study is needed to address the early failures; particularly those related to fracture and metal debris.

Statement of purpose: The aim of this study is to evaluate different designs of unicompartmental knee replacement (UKR) by comparing the peak von Mises and contact stresses in polyethylene (PE) bearings over a step-up activity.

Summary of Methods: A validated finite element (FE) model was used in this study. Three UKR designs were modelled: a spherical femoral component with a spherical PE bearing (fully-congruent), a poly-radial femoral component with a concave PE bearing (semi-congruent), and a spherical femoral component with a flat bearing (non-congruent).

Kinematic data from in-vivo fluoroscopy measurements during a step-up activity was used to determine the relative tibial-femoral position as a function of knee flexion angle for each model. Medial and lateral force distribution was adapted from loads measured in-vivo with an instrumented implant during a step-up activity. The affect that varying the bearing thickness has on the stresses in the bearing was investigated. In addition, varus-valgus mal-alignment was investigated by rotating the femoral component through 10 degrees.

Summary of Results: Only the fully congruent bearing experienced peak von Mises and contact stresses below the PE lower fatigue limit (17MPa) for the step-up activity (fully congruent PE peak contact stress, 5MPa). The highest PE contact stresses were observed for the semi-congruent and non-congruent designs, which experienced approximately 3 times the PE lower fatigue limit. Peak PE von Mises stresses for the semi-congruent and non-congruent designs were similar, peaking at approximately 25MPa. Peak PE von Mises stresses were ameliorated with increased bearing thickness. Varus-valgus mal-alignment had little effect on the peak stresses in the three UKR designs.

Statement of Conclusions: Fully congruent articulating surfaces significantly reduce the peak contact stresses and von Mises stresses in the bearing. The FE model demonstrates that fully congruent bearings as thin as 2.5mm can be used without increasing the contact stresses significantly. Fully congruent designs can use thinner bearings and enable greater bone preservation.

Statement of purpose: Finite element (FE) models of bone can be used to evaluate new and modified knee replacements. Validation of FE models is seldom used, and the quantification of modelling parameters has a considerable effect on the results obtained. The aim of this study is to develop a FE model of a cadaveric tibia and validate it against a comprehensive set of experiments.

Summary of Methods: Seventeen tri-axial rosettes were attached to a cleaned, fresh frozen cadaveric human tibia and the tibia was subjected to 13 loading conditions. Deflection and strain data were used for comparison with the FE model. A geometric model was created on the basis of computed tomography (CT) scans. The CT data was used to map 600 orthotropic material properties to the tibia. All experiments were simulated on the FE model. Measured principal strains were compared to their corresponding FE values using regression analysis. The validated tibia model was reduced in size (75mm to the proximal) and then re-modelled to represent only the proximal tibia. This re-modelled tibia was validated against the reduced size FE model. Virtual surgery was performed on the validated proximal model to implant a UKR.

Summary of Results: For the whole tibia model, the regression line for all axial loads combined had a slope of 0.999, an intercept of −6.24 micro-strain, and an R2 value of 0.962. The root mean square error as a percentage was 5%. For the proximal tibia model, correlation coefficients of 0.989 and 0.976 were obtained for the maximum and minimum principal strains respectively.

Statement of Conclusions: An FE model of an implanted proximal tibia has been validated against experimental data. This model is able to accurately predict the deflection and stresses in a replaced knee joint to obtain clinically relevant information. This will provide a virtual model of unicompartmental arthroplasty, where variables such as fixation method and bearing mechanics can be assessed.

Introduction: The results of the mobile bearing Oxford unicompartmental knee replacement (UKR) in the lateral compartment have been disappointing with a five year survival of 82%. Therefore, it is recommended that mobile bearings should not be used for lateral UKR. This low survivorship is primarily due to high dislocation rate, all occurring in the first year. A detailed analysis of the causes of bearing dislocation confirmed the elevated lateral tibial joint line to be a contributory factor. A new surgical technique was therefore introduced in which care was taken neither to remove too much bone from the distal femur nor to over tighten the knee and therefore ensure that the tibial joint line was not elevated. Other modifications to the technique were also introduced including use of a domed tibial component.

Aim: The aim of this study is to compare the outcome of these iterations: the original series [series I], Series II with improved surgical technique and the domed tibial component [Series III].

Method: The primary assessment of outcome was bearing dislocation at one year. One year was chosen as all the dislocations in the first series occurred within a year. In series I, there were 53 knees, in series II 65 knees and in series III 60 knees, all with a minimum of one year follow up.

Results: In series I, there were 6 bearing dislocations (11%) and the average range of movement (ROM) was 110°. In the second series, there were 2 dislocations (3%) and the average ROM was 118°. In the third series, there were no primary dislocations and the average ROM was 125°.

Conclusions: The improved surgical technique and implant design has reduced the dislocation rate to an acceptable level so a mobile bearing can now be recommended for lateral UKR.

Introduction: Treatment options for the young active patient with isolated symptomatic medial compartment OA and pre-existing ACL deficiency are limited. Implant longevity and activity levels may preclude TKA, whilst HTO and unicompartmentasl knee arythroplasty (UKA) are unreliable due to ligamentous instability. UKAs tend to fail because of wear or tibial loosening resulting from eccentric loading. Combined UKA and ACL reconstruction may therefore be a solution.

Method: Fifteen patients with combined ACL reconstruction and Oxford UKA (ACLR group), were matched (age, gender and follow-up period) with 15 patients with Oxford UKA with intact ACL (ACLI group). Prospectively collected clinical and x-ray data from the last follow-up (minimum 3 years, range: 3 – 5) were compared. Ten patients from each group also underwent in-vivo kinematic assessment using a standardised protocol.

Results: At the last follow-up, the clinical outcome for the two groups were similar. One ACLR patient needed revision due to infection. Radiological assessment did not show any significant difference between relative component positions and none of the patients had pathological radiolucencies suggestive of component loosening. Kinematic assessment showed posterior placement of the femur on tibia in extension for the ACLR group, which corrected with further flexion.

Conclusions: The short-term clinical results of combined ACL reconstruction and UKA are excellent. Lack of pathological radiolucencies and near normal knee kinematics suggest that early tibial loosening due to eccentric loading is unlikely.

Aim: The aim of this study was to asses the accuracy of skyline radiographs in the assessment of the patellofemoral joint, when compared to open intraoperative assessment.

Methods: Eighty nine patients undergoing knee replacement surgery were included in the study. Skyline radiographs were obtained preoperatively. These radiographs were assessed and graded by an experienced musculoskeletal radiologist using the Altman and Ahlbäck classifications. The grades were calculated for both the medial and lateral facets of the PFJ. Intraoperative assessment of the Patellofemoral joint was undertaken at the time of surgery. The damage was graded using the modified Collins classification (0: Normal, 1: Superficial damage, 2: Partial thickness cartilage loss, 3: Focal Full thickness cartilage loss < 2cm2, 4: Extensive full thickness cartilage loss < 2cm2). Data was obtained for the Medial Facet, Lateral Facet and Trochlea.

Results: Spearman’s rank correlation coefficient between the radiographic and macroscopic changes within the lateral PFJ were poor with both the Altman 0.22 (p=0.0350) and Ahlbäck 0.24 (p=0.018). The correlation of the medial PFJ was slightly better with a coefficient for Altman 0.42 (P< 0.0001) and Ahlbäck 0.34 (P> 0.001).

Conclusion: In conclusion skyline radiographs provide a poor to moderate preoperative assessment of the degree of osteoarthritis within the patella-femoral joint. This has significant implications for establishing radiographic criteria for planning patella-femoral joint replacement.

Purpose: The aim of this study is to compare the long-term survival results of TKA in patients under the age of 60, using

revision surgery and

Method: From our knee database we identified a cohort of 60 total knee replacements that had been performed over 15 years previously. We identified those who had died, those who had been revised and established the Oxford Knee Score (OKS) for all those still surviving.

Results Using the following endpoint criteria the cumulative 15-year survival was (A) revision surgery alone = 78% (CI 12), (B) revision surgery or an OKS less than or equal to 24 (50% of total OKS) = 63% (CI 13), and (C) revision surgery or moderate pain = 48% (CI 14).

Conclusion The functional survival of TKA in patients under the age of 60 decreases in the second decade following implantation with a significant number of prostheses failing the patient due to knee pain

Introduction: The information in the literature about the relative merits of cemented and cementless unicompartmental knee replacement (UKR) is contradictory, with some favouring cementless fixation while others favouring cemented fixation. Cemented fixations give good survivorship but there is concern about the radiolucency which frequently develops around the tibial component. The exact cause of the occurrence of radiolucency is unknown but according to some, it may suggest suboptimal fixation.

Method: Sixty-two knees (31 in each group) were randomised to receive either cemented or cementless UKR components. The components were similar except that the cementless had a porous titanium and hydroxyappatite (HA) coating. Patients were prospectively assessed by an independent observer pre-operatively and annually thereafter. The clinical assessment included Oxford Knee Score, Knee Society Scores and Tegner activity score. Fluoroscopically aligned radiographs were assessed for thickness and extent of radiolucency under the tibial implant.

Results: At one year there were no differences in the clinical outcome between the groups and there were no loose components. No radiolucencies thicker than 1mm were seen. At one year none of the cementless tibias and 30% of the cemented tibias had complete radiolucencies. One out of 31 cementless (3%) and 12 out of 31 cemented (39%) had partial radiolucencies. This difference between these two groups was high significant (p< 0 0001).

Conclusions: This study clearly demonstrates that the incidence of radiolucency beneath the tibial component is influenced by component design and method of fixation. With identical designs of tibial component none of the cementless components developed complete radiolucences whereas 30% of the cemented components did. We conclude that HA achieves better bone integration than cement.