Lay Person's Summary:

Knee replacement surgery has been successfully carried out for many years. However, in time, some knee replacements will fail-usually either because of wear or loosening of the replacement parts. This may mean the patient has to undergo further surgery-this is known as revision total knee replacement (rTKR). This revision surgery is often more complex than the original operation and presents the operating surgeon with several technical challenges. Of particular concern, is that when the old knee replacement is removed, a large cavity can be left in the bone. The new knee replacement has to be placed into this, but it is essential that it is immediately stable and secure. The large cavity has to be somehow either filled in or bypassed to ensure the new knee replacement is secure enough for early weight-bearing and long term survival.

Different techniques have been used for many years to overcome this problem. Firstly, it may be possible to simply cement another knee replacement into the cavity- probably needing more cement than was previously used.

An alternative option to this is use a device called a "cone" which sits into the bony cavity created by removing the original knee replacement, and a new knee replacement can be cemented into this.

Bone grows onto the cone to ensure its stability. The new knee replacement can have either short stemmed components like the original Total Knee Replacement (TKR), or long stemmed components.

All of these types of rTKR are presently in use throughout the United Kingdom (UK). At this point in time, no-one knows which type is best. The research team are planning to run this study to see if it is possible to identify which type of rTKR gives the best outcome for patients. The research team will examine the results in different ways including questionnaires to measure how well the patients feel their knee is performing, specific tests to measure knee function, and by using x-rays and scans.

The investigators will identify patients suitable for inclusion in the study from their medical notes, x-rays and scans. If patients consent to be in the study, the research team will randomly allocate them to receive one of the 3 rTKR options (no cone and new knee replacement only, cone with short stemmed components or cone with long stemmed components) and monitor their progress for 5 years after the operation.

The investigators will identify patients suitable for inclusion in the study from their medical notes, x-rays and scans. If patients consent to be in the study, the investigators will randomly allocate them to receive one of the 3 rTKR options (no cone and new knee replacement only, cone with short stemmed components or cone with long stemmed components) and monitor their progress for 5 years after the operation.

A preliminary analysis of the data generated by the study will be carried out 2.5 years after recruitment of the 1st participant to assess compliance with the study protocol.

Methodology:

A randomised, prospective internal pilot study is to be set up to compare the results of different ways of approaching revision Total Knee Replacement (rTKR). The invsetigators will examine 3 differing ways of treating patients undergoing this procedure who present with a grade 2 Anderson Orthopaedic Research Institute (AORI) bony defect. One arm of the study will be randomised to receive a long or short stemmed cemented prosthesis (according to intra-operative findings and surgeon decision at that time), another to receive a metaphyseal cone into which a short stemmed prosthesis will be cemented (</=50mm) and a third arm to receive a metaphyseal cone with a long stemmed cemented prosthesis (>/=100 mm). The study will be carried out in the operating theatres, orthopaedic out-patients department and the physiotherapy department at the Royal Devon and Exeter (RD+E) Hospital.

Patients will be invited to take part in the study in one of 2 ways. Firstly, by their surgeon when they are initially listed for rTKR surgery at their out-patient consultation at the RD+E Hospital. Alternatively, if identified from the current surgical waiting list as suitable for inclusion. In this latter case, potential participants will be contacted by letter by their surgeon asking for permission for one of the research team to approach them to discuss the study. Patients who agree to enter the trial will be divided at random to enter the different treatment arms of the study.

Randomisation will be carried out using a computer generated random number assignment list. 51 patients will be initially included with allocation to either arm of the trial by drawing appropriately marked slips of paper from an opaque envelope. The operating surgeon will not be able to read the slips before they are drawn, nor will they perform the randomisation which will be carried out by one of the investigators not carrying out the surgery on the individual patient.

The research team will undertake a preliminary analysis of the study data 6 months after 51 participants have been recruited (i.e. 2.5 years after 1st patient recruited). This will be to see if the pilot study aims and objectives and study progression rules have been met. Based on this analysis, the aim is to recruit further participants to the number recommended in the preliminary analysis.

Participant Selection:

Patients will be under the care of the Exeter Knee Reconstruction Unit at the RD+E Hospital.

Sample Size:

The exact treatment effect of the use of cones in revision knee replacement is unknown. There are no prospective studies to use as a basis for calculating sample size. The choice of 51 participants initially with the potential for further recruitment after preliminary analysis of the 6 month post-surgery (i.e. 2.5 years after recruitment of the 1st participant) follow-up data is to allow an assessment of treatment size effect to allow the investigators to work out a definitive study sample size requirement. The investigators have taken statistical advice on this and are confident that this number will be sufficient to test the progression rules and to offer reasonable estimates of standard deviations of the intended outcome measures.

The research team will initially recruit 17 cases to each arm of the study (no cone used, cone with short stemmed knee replacement and cone with long stemmed knee replacement)

Pragmatically, (based on 2015 figures) the Exeter Knee Reconstruction Unit (EKRU) undertake approximately 62 rTKR procedure per year and in 2015, 27 of these did not meet the study inclusion criteria. The investigators believe this to be a representation of other years. Of the remaining 35 cases, if 20% are unwilling to participate this will leave 28 cases per annum. The study is therefore likely to recruit 56 potential participants in just under a 2 year period. The plan is to initially recruit 51 subjects (17 to the no cone arm and 17 each to the cone with short or long stem arms respectively). A preliminary analysis of the data will be carried out after 51 cases have been recruited, operated upon and have 6 months of follow up completed (i.e. 2.5 years after recruitment of 1st participant), to look for any significant trends. Based on these preliminary results of the internal pilot study the aim is to recruit further cases to reach a recruitment target that has been set based on the pilot study results. This figure would be calculated to allow the research team to identify a total number of participants that would give results with statistical significance in either the primary or secondary outcome measures.

Initial number of study participants is 51 patients but with potential for an increased number based on analysis of the pilot study.

Procedure:

Potential candidates for this study will be identified in one of 2 ways. Firstly, patients are listed for rTKR surgery whilst attending orthopaedic out-patient clinics at the RD+E Hospital. At the time of this consultation patients will be offered the opportunity to participate in the study. Potential participants will be given a copy of the information about the study and the consent form. This will allow the opportunity therefore, to discuss the study with family, friends or General Practitioner (GP) if necessary. Alternatively, patients will be identified from existing surgical waiting lists. In this case, potential recruits will be contacted in writing or by telephone asking for permission for one of the study team to approach them about the study. If permission is given, the patient will be sent a copy of the participant information sheet and a consent form so that they can decide if they would like to be involved. One of the research team- either the study coordinator or one of the research nurses- will then contact the patient and if they are willing to be involved, will obtain written consent for study participation. Consent is likely to be obtained at the pre-operative assessment clinic which usually occurs 2-4 weeks before surgery and with the patients consent, their GP will be informed of participation in the study.

3 copies of written consent will be obtained. One will be returned to the patient. One will be placed in the medical notes and one will be kept by the research team in the patient's case record file (CRF).

Patient Generated Clinical Outcomes:

All rTKR patients attend the hospital for routine pre-operative assessment-whether involved in the study or not. At this appointment, trial participants will complete the questionnaires related to the study- the Oxford Knee Score (OKS), the Knee Osteo-arthritis Outcome Score (KOOS) and the Euroquol Questionnaire (EQ5D). These 3 questionnaires are contained within the revision cones study Patient Recorded Outcome Measures (PROMS). Participants will also complete the Hospital Anxiety and Depression Score (HADS).

Initial Radiological Assessment:

All participants will undergo standardised long leg Antero-Posterior (AP) and lateral radiographs pre and post-operatively as part of routine care for rTKR patients. These will be reported upon by the surgeons and the results documented in the medical notes electronically via Clinical Document Manager (CDM).

In addition to routine care, all patients will undergo a Dual Emission X-ray Absorptiometry (DEXA) scan at the University of Exeter for pre-operative bone density evaluation. At the time of the initial DEXA scan patients will be asked to complete the following questionnaires: 1) Lower Extremity Functional Scale (LEFS) and 2) General health questionnaire (bone health female/male) to assess the impact this may have on bone density.

Pre-operative Functional Assessment:

Study participants will attend the physiotherapy department at the RD+E Hospital for a functional assessment. The Exeter Knee Reconstruction Unit (EKRU)Knee Arthroplasty Functional Assessment will be undertaken with analysis by the Nurture Functional Assessment Tool.

Randomisation:

The knee surgery team meet every Tuesday to discuss the operations planned for the next week. One of the research team (study coordinator or research nurse) will attend this meeting and open the envelope that assigns each patient to the appropriate arm of the study. (See methodology above). The operating surgeon and theatre staff will be informed. It is necessary to do this in advance of the day of the operation to ensure all instrumentation required for the surgery is ready for use, sterile, available and laid out in preparation for the surgery. The patient will be assigned a study number and their randomisation slip stored by the study team. The randomization schedule will be determined using an on-line random assignment sequence generator (www.graphpad.com/quickcalcs/randomize1.cfm) and will be kept and controlled by the study co-ordinator, and the surgeons will not have access to this schedule. A hand written note will be made by one of the research team in the orthopaedic section of the patient's medical notes after their surgery, confirming that they consented to the study and detailing which arm of the study they were allocated to.

Randomisation will be to one of 3 arms of the study. The patient will either receive:

1. A new TKR using the Triathlon Component System cemented into the bony cavity left after removal of the original TKR-stem length to be decided by surgeon during the operation.
2. A Triathlon Cone into which a new TKR will be cemented- stem length </=50mm.
3. A Triathlon Cone into which a new TKR will be cemented-stem length >/= 100mm

Surgery:

Operating techniques will be standard techniques as used by the 3 surgeons involved in the study. All 3 are experienced rTKR surgeons.

All stems will be cemented using a cement plug, suction catheter and pressurisation. All patients receive the Triathlon Total Stabilised (TS) Revision Implant with the TS polyethylene.

Both femoral and tibial components will be revised in all cases. A decision to revise the patella component will be taken intra-operatively by the operating surgeon. If the patient's patella was not resurfaced during the original operation, it is likely to be resurfaced during the revision procedure. If the patella was previously resurfaced it will be visually assessed per-operatively for component wear and manually assessed for loosening. It will be revised if found to be significantly loose or worn.

Patients will be discharged home allowed to fully weight-bear as comfort allows and with walking aids for use for a 6 week period if required.

Follow Up:

As per routine care, the patients will be reviewed at 3 months, 6 months, 1 year, 2 years and 5 years post-operatively. At each follow up appointment they will repeat the questionnaires above, (OKS, KOOS, HADS and EQ5D) and the weight bearing AP long leg and standard lateral radiographs.

Functional assessment will also be carried out at the 3 month, 6 month, 1 year, 2 year and 5 year follow up time frames.

All patients will undergo a DEXA scan to assess the bone density in the bone immediately adjacent either to the new knee replacement+/- cone post-operatively at 6 weeks, 3, 6, 12, 24 and 60 months. The DEXA scan will be performed at the University of Exeter.

All of the patients with a cone implanted will undergo a dual energy Computerised Tomography (CT) scan at 6 months after surgery to assess potential bony on-growth onto the cone.

The Primary outcome of the Pilot Study will be:

• An assessment of compliance with the study protocol. For this, the research team will measure compliance of study participants with the pre and post-operative attendance schedule.

Time Frame: 2.5 years after recruitment of 1st subject

Secondary Outcome Measures:

• An assessment of ease of recruitment to the study by comparing the potential number of recruits to those actually recruited
• Failure for any reason requiring further revision surgery: including but not exclusive to loosening of the components, infection, instability or pain.
• The assessment of early stability of the newly implanted prosthesis as measured radiographically on x-rays by subsidence
• Measurement of on-growth of bone onto the prosthesis via dual energy CT scan for patients in those arms of the study.
• Measurement of bone density in the bone surrounding the cone via DEXA scan
• Clinical assessment using the KOOS tool, the OKS, the HADS and the EQ5D questionnaires.
• Functional assessment using EKRU Knee Arthroplasty Functional Assessment and the Nurture assessment tool
• The development of post-operative infection

Data Collation:

To be undertaken by the clinician undertaking the specific procedure concerned or by the Chief Investigator of the study or by research nurses attached to the RD+E Hospital provided by the hospital Research and Development (R+D) department and the National Institute Health Research (NIHR).

The data will be stored on RD+E Hospital secure computers. Data will be entered into an Excel spread sheet for later transfer into a statistical package such as Statistical Package for the Social Sciences (SPSS) v15, StatsDirect or Stata for analysis. All patient identifiers will be removed. The study data will be kept for 5 years and destroyed thereafter by deleting it from the hard drive of the Hospital computers.

Data analysis:

As this is a pilot study, an estimate of effectiveness is not the primary aim of the analysis. However, as part of the secondary outcome measures, an analysis of the Minimally Clinically Important Difference results from the EQ5D (0.14 from the EQ5D index) between the 3 arms of the study will be performed to establish if there are differences identified that can allow the investigators to power a larger study. Regarding secondary outcomes, the research team will be interested in the central tendency and variation found in the outcome data. Thus descriptive analyses will be conducted giving means and standard deviations of outcome scores or median and interquartile ranges where appropriate, if the data are not normally distributed and cannot be easily transformed. This will apply to the various questionnaire-based outcome measures. Estimates of central tendency and variation will enable a sample size calculation to be conducted in order to power a larger definitive trial at a later date. The research team will arrange for independent statistical analysis of the results.

For the substantive study, there will be an analysis of covariance as there are 3 groups being compared with repeated measure Analysis Of Variance (ANOVA) as data are being collected at more than one time point (i.e. at follow-up).

Measures to avoid bias:

• The participants will be blinded as to which arm of the study they are allocated to. However, the participants allocated to one of the cone arms of the study may recognise this when they called for the post-operative CT scan. It is not ethical to CT scan all participants due to the radiation exposure when useful information from this examination can only be gleaned from the groups who have a cone implanted. Thus, patients with a cone device in situ may work out that they must have a cone device when they are having a CT scan. However, study participants will not be able to determine whether a short or long-stemmed TKR is in situ.
• The surgeon cannot be blinded to randomisation as he will be performing the surgery.
• The members of staff interpreting the results of the CT and DEXA scans cannot be blinded as to which group the patients are allocated to as it will be clear to see on the radiological images
• The physiotherapist carrying out the functional assessments will be blinded to study arm allocation
• No surgeon will report upon the x-rays of a patient upon whom they have operated
• The statistician analysing the results will be blinded to study arm allocation

Personnel:

• Chief Investigator- Andrew Toms
• Surgeons-Andrew Toms, Jonathan Phillips and Keith Eyres
• X-Ray interpretation: Andrew Toms, Jonathan Phillips and Keith Eyres. No surgeon will report on cases upon whom he has operated.
• Study correspondence and administration Patrick Hourigan
• Interpretation of CT scan results by Dr Rahul Anaspure -Consultant radiologist at the RD+E Hospital.
• Radiology staff employed at the RD+E Hospital as part of routine care to undertake imaging
• Functional outcome assessment team-Dean Chisling De Burgh -physiotherapy department RD+E Hospital.
• Statistical support provided by Dr Roy Powell
• Radiological assessment via DEXA scans by Dr Karen Knapp University of Exeter and Michael Knapp-radiographer
• Medical Physics Expert: Helen O'Brien at RD+E Hospital.
• Clerical staff attached to the EKRU.

Confidentiality:

All of the research staff will be trained as per the Guidelines for Good Clinical Practice (GCP) in research trained. All National Health Service (NHS) staff involved are bound by patient confidentiality requirements of the NHS.

Dr Knapp (who undertakes the DEXA scans at Exeter University) will be bound to confidentiality requirements as part of GCP adherence as will Michael Gundry.

The study participants GP will be informed that their patient is involved in the research- particularly as the patient may wish to discuss their involvement in the research with their GP.

All of the data associated with the study will be kept electronically on the hard drives of computers within the Exeter Knee Reconstruction Unit (EKRU). It will be stored on encrypted and password protected hospital computers. The data will only be accessible by members of the research team and by the R+D department at the RD+E Hospital who will require access to the data to ensure compliance with the protocol.

At the end of the study (see timescales below), all research data will be destroyed by being permanently deleted from the RD+E hospital computers. Copies will be archived by the study sponsor in line with standard research material archiving procedures. Radiographs and scans will remain on the secure hospital systems as per routine care.

Timescales:

62 rTKR operations were carried out at the RD+E in 2015. Of these, approximately 35 cases would have met the study inclusion criteria. If one considers that 20% of potential study participants will decline participation, it is estimated that 28 candidates could be recruited per year. Thus recruitment of 51 candidates is likely to take just under 2 years, provided adequate numbers of new cases are identified as needing revision and meet the entry criteria. However, there are at this time, approximately 15 patients awaiting surgery who already meet the entry criteria for this study. These patients have the potential to allow faster recruitment to the pilot study as their surgery is already planned as opposed to waiting for patients to be seen in clinics and for decisions to be taken regarding the need for surgery, and then patients being placed on surgical waiting lists.

A preliminary analysis of the data will be performed once the 51 patients are recruited and have completed 6 months post-operative follow-up (i.e. 2.5 years after recruitment of the 1st subject.

Complete follow up is for 5 years post entry into the study. It will take a further 1 year for data analysis, project write up and submission for publication post study end.

Therefore, if 51 participants are recruited over 2 years, followed up for 5 years and the data is analysed over 1 year, the total study length is 8 years.

If the preliminary analysis of the first 51 cases allows calculation of a sample size for a substantive study, the research team will then be able to give an estimate for study length for the substantive study.

Study data will be destroyed 5 years after completion of the study.

Dissemination of Results:

The participants will be informed in writing of the results if they so desire. The Research and Development department at the RD+E will be informed of the results.

The research team plan to present the results both nationally and internationally at knee surgery conferences (such as the British Orthopaedic Association and the European Knee Society) and to publish the results in a peer reviewed journal (such as the Bone and Joint Journal, the Journal of Bone and Joint Surgery-America, The Journal of Arthroplasty.)

Impact of Results:

As this is an internal pilot study, the impact of this study may be limited. The main aim is to answer the pilot study aims and objectives. However, if it is possible to identify a clear advantage of one study arm over the others, then it is likely that this technique will become standard rTKR procedure- with the aim of providing patients with the best clinical and functional results after their revision knee surgery. Similarly, if a clear difference is found in one study arm, it is hoped that presentation of the results to colleagues at other institutions will result in better long term results for their patients.

In the event that no technique proves superior to the others, then this will provide surgeons with the reassurance that no matter what their preferred technique, no patient is being disadvantaged.

If it is found that the cemented rTKR performs as well or better than the cone device plus knee replacement, and if this became the technique of choice for this operation, this could have significant cost savings for the NHS.

Conflicts of Interest:

Stryker UK have supported this study financially. However, it remains completely independent from them in study design, results and dissemination of findings. The 3 surgeons involved all undertake consultancy work for Stryker UK and will complete the necessary applications regarding this on the ethics application for this study.

Costs:

• The out-patient appointments, surgical costs, follow up costs and plain radiographs are all costs paid for by the NHS as part of routine care.
• Functional assessment, CT scan, DEXA scan. These costs will be covered by external funding provided by Stryker UK.
• Research support nurses are paid for by the RD+E Hospital.
• The study coordinator will provide the support for participant identification, recruitment, ensuring time frame adherence for follow up appointments, scans and functional assessment and will maintain the database of study data. Salary costs for this are being met by Stryker UK.
• Costs of statistician for analysis of results after the study will be met by the external funding provided by Stryker UK.

Patient Involvement:

The research team have consulted with patients on the feasibility of participating in the study at all and in particular about the demands of attending the appointments for functional assessment. The study was discussed at the EKRU patient research panel. (This panel is a group of patients who have undergone knee replacement surgery in the EKRU and who are therefore, well placed to comment on the practicalities of study involvement, as well as on the appropriateness and clarity of the patient information leaflet.) Attendance at all time frames is critical for completion of the secondary measure functional outcome and it was stressed by the patient representatives that the patients need to be fully informed of the commitment required should they agree to take part. Similarly, it was recommended that the extra commitment for CT and DEXA scan appointments be highlighted.

Bi-annual meetings of the patient research panel are held and if issues arise when running the study that require further patient involvement, they will be taken to the panel for consultation. If a meeting is not due for some time, the research team will contact a panel member personally for their advice.

When the research team consulted on the design and wording of the patient information leaflet and the consent form, various changes were recommended and implemented.

Risks of the Study:

All of the operative care is already part of routine practice in the RD+E Hospital. Thus the research team do not believe involvement in this study changes the risks from surgery that the patients already face as a part of routine care.

The study coordinator has discussed the study with the radiological protection officer (RPO) at the RD+E Hospital who has provided radiation dose calculations and a risk assessment based on plain x-ray as per routine care, and the pre-op, 6 week post-op and 3, 6, 12, 24 and 60 month post-op DEXA scans. The RPO will be our medical physics expert (MPE) on the ethics application.

All patients undergoing a knee replacement have 7 planar x-ray examinations at the same frequency of the study and will receive a radiation dose of 717.5µSv (milli Sieverts)- equivalent to 3.4months background radiation.

All patients involved in the study will receive 7 additional DEXA examinations of the Lumbar spine, dual hip, total body and both knees receiving a total dose of 740.5µSv which includes the additional 23µSv,i.e. the DEXA adds very little extra dose. The lifetime risk of developing a cancer resulting from this excess dose is 1 in 1,100,000 for a 50 year old patient, (this would fall to 1 in 1,600,000 for a 65 year old patient). For the ages of the expected cohort, this is considered to be a negligible risk (Health Protection Agency Centre for Radiation Chemical and Environmental Hazards (HPA-CRCE) 028, 2011 and National Radiation Protection Board (NRPB), 2001 and is equivalent to around 3.4 days of natural background radiation in the UK.

The patients who undergo 1 additional dual energy CT examination of the knee will receive a total dose of 890.5µSv, which includes 23 µSv for the DEXA studies and a dose of 150µSv for the CT. Their total additional dose is 173 µSv. The lifetime risk of developing a cancer from this excess dose is 1 in 150,000 for a 50 year old patient, (this would fall to 1 in 218,000 for a 65 year old patient). For the ages of the expected cohort, this is considered to be a very low risk (HPA-CRCE-028, 2011 and NRPB, 2001) and is equivalent to around 0.8 of a month of natural background radiation in the UK.

Monitoring Adverse Events/Stopping the Study:

A log of adverse events, serious adverse events, adverse device effects and serious adverse device effects will be kept by the study coordinator. Any such event events occurring in the immediate peri and post-operative events should be reported to the study coordinator by the consultant surgeon in charge of the case who will contact the study sponsor within the time frames as specified by them. Similarly any such adverse events may be identified in clinic and again, the surgeon concerned should report these back to the study coordinator to be logged and the sponsor will be informed.

It is possible that adverse events will occur that are dealt with either by the Accident and Emergency (A+E) department, the patient's GP or other hospitals. These may not be readily apparent to the research team. Therefore the team will specifically ask participants to inform their surgeon if any complications or adverse events occur and similarly ask the surgeons to specifically ask their patients for this information at each clinic attendance.

A meeting of the research team will be held on a 3 monthly basis to discuss progress of the study, any specific problems and to consider whether there are any justifications for stopping the study. As all 3 techniques being studied of performing rTKR are already part of routine practice, and it is not anticipated that an unexpected complication will be identified that would necessitation an early termination of the study.

Ethical Approval:

Ethical approval for this study has been provided by the National Research Ethics Committee and the Health Research Authority.

Project Management:

• Funding obtained by Mr Toms
• Identification of study participants-Mr Toms, Mr Phillips, Mr Eyres and the study coordinator
• Consent and randomisation for study participants will be carried out by the study coordinator or R+D department research nurses.
• Data collection- Clinical data will be collected by study coordinator and the R+D department research nurses.
• Dean Chisling De Burgh will collect the functional assessment data.
• Karen Knapp and Michael Gundry will collect the DEXA scan data.
• CT scans to be reported upon by Dr Anaspure
• Interpretation of plain radiographs will be by the 3 surgeons Mr Toms, Mr Phillips and Mr Eyres.
• Data entry- clerical staff attached to the EKRU.
• Statistical analysis-Dr Roy Powell.
• Study queries to be passed to the study coordinator- Telephone 01392-408562 or 07904-034538.
• 3 monthly study progress meetings will be co-ordinated by the study coordinator.
• Writing up of results by Mr Toms and Mr Phillips
• Results dissemination to patients by study coordinator
• Study write up for presentation and publication by Mr Toms and Mr Phillips

Inclusion Criteria:

• Patients undergoing first time revision knee replacement for aseptic loosening or wear of the components
• Patient has signed an ethics committee approved consent form
• rTKR to be performed at The Royal Devon and Exeter Hospital
• AORI class 2 defects of femur and/or tibia
• Skeletally mature male or female
• Patient is willing and able to comply with postoperative scheduled clinical and radiological evaluations and rehabilitation
• Patient must be suitable for a rTKR with the Triathlon TS system i.e. must not have gross collateral ligament laxity

Exclusion Criteria:

• Refusal to consent to participate in the study
• Patients known to have an infected joint replacement prior to revision surgery
• Patients identified with an unexpected infected joint replacement identified per-operatively
• AORI class 1 defects of tibia or femur where metaphyseal cone fixation is not indicated
• AORI class 3 defects of tibia and femur where distal femoral replacement or proximal tibial replacement is required.
• Cases of ligamentous instability where condylar knee revision is not indicated
• Patient is diagnosed with a systemic disease or metabolic disorder leading to progressive bone deterioration
• Patient has a neuromuscular or neurosensory deficiency
• Patients undergoing patella revision in isolation
• Pregnancy

Participants will be blinded to study arm allocation. The physiotherapists performing the functional assessments will be blinded to study arm allocation. T