The purpose of the study is to measure and compare the wear rates of retrieved PEI after failed TKA with respect to various causes for revision surgery, implant size, surgical and patient factors as well as to identify a characteristic linear wear distribution pattern of the articular surface in most common causes of TKA failure. Consequently, we want to contribute to a better understanding of wear mechanisms and identify a possible additional factor for increased PEI wear in TKA.

INTRODUCTION / BACKGROUND The most common indications for revision (or causes for dissatisfaction) in total knee arthroplasty (TKA) include prosthetic joint infection (PJI), malposition of the implant, wear of the tibial polyethylene insert (PEI) and consequent induced osteolysis and aseptic loosening, mechanical failure, instability and stiffness of the prosthetic joint. Increased wear of PEI is the result of a multifactorial problem that involves factors from the implant, the surgeon, and the patient. There is also a large number of patients with failed knee arthroplasty where the exact cause for failure is not known. In these cases, probably, a single cause alone would not lead to a problem, but a combination of causes could generate severe dissatisfaction finally leading to revision without clearly defined diagnosis. The appropriate diagnosis in these knees might be extremely difficult to clarify.

Analyses on retrieved PEI allows for a better understanding of TKA failure mechanisms and the detection of factors that cause increased wear rates. There is a lack of research on homogeneous PEI samples, which are analyzed using modern methods and information on the 3D wear distribution of the articular surface in different clinical causes for TKA failure. To our knowledge, it has not been investigated, whether PEI size and the femur to PEI size ratio (within the manufacturer's guaranteed compatibility) affect the wear rate.

PURPOSE AND HYPOTHESIS The purpose of the study is to measure and compare the wear rates of retrieved PEI after failed TKA with respect to various causes for revision surgery, implant size, surgical and patient factors as well as to identify a characteristic linear wear distribution pattern of the articular surface in most common causes of TKA failure. Consequently, we want to contribute to a better understanding of wear mechanisms and identify a possible additional factor for increased PEI wear in TKA.

We expect increased polyethylene wear rates of prosthesis failed due to aseptic causes compared to those failed due to PJI and different linear wear distribution patterns in different failure causes. We expect higher wear rates in smaller PEI and in PEI with a lower femoral to PEI component size ratio (within the framework of the compatibility guaranteed by the manufacturer).

MATERIALS AND METHODS The study will include approximately 60 retrieved Sigma® Cruciate-Retaining Curved GVF i2 from DePuy Synthes (Johnson & Johnson, Warsaw, IN 46582, USA) PEIs, that were removed during the revision operation between 1.1.2013 (start of Institutional storage of removed knee endoprosthesis components) and 31.12.2021. The listed Sigma® PEI represents the largest share of inserted PEV in primary total knee arthroplasty in Valdoltra Orthopedic Hospital (VOH) in the last decade (from 1.1.2012 to 31.12.2021) and the highest number of single types retrieved PEIs in this period and represents the only reason why we selected this particular implant.

All inclusion criteria are:

primary TKA performed in VOH, cement TKA, tibial component Sigma® Modular CoCr Tibia Tray Component (Non-Porous); tibial PEI Sigma® Cruciate-Retaining Curved GVF i2; femoral component Sigma® CR Femoral Component (Non-Porous); revision operation with PEI removal performed in VOH and PEI storage.

Exclusion criteria: endoprosthesis fracture, missing patient, prosthesis type or course of primary and revision surgery data.

Clinical and demographic data of patients will be obtained from the Valdoltra Arthroplasty Registry (VAR) and Institutional information system database. Radiological diagnostic studies will be reviewed in the Institutional program IMPAX. Following the first collection, all data will be anonymized and managed under the hospital's internal registry index.

The approval of the Institutional review board of Valdoltra Orthopaedic Hospital (issue number 1/2022) was obtained to carry out the study.

Wear measurements and calculation of wear rates Measurements will be performed on the articular and the entire surface of the retrieved PEI. The latter measurement will also incorporate polyethylene deformation or creep. First, the PEIs will be scanned with an EviXscan 3D Heavy Duty Quadro 3D optical scanner (EVIXSCAN 3D, Bielsko-Biala, Poland) mounted on a tripod stand. For point cloud processing we will use the measurement software EviXscan 3D Suite 2.7 (EVIXSCAN 3D, Bielsko-Biala, Poland). With the described systems, we will obtain an STL file of each retrieved PEI. Geomatic Control X software (3D Systems - USA / Canada) will be used to post-process the measurements and then comparison of the scanned PEI with its reference CAD model of appropriate dimensions. The specified accuracy of the optical scanner is from 0.013 mm onwards.

We will calculate the linear wear of the articular surface, which is defined as the maximum deviation in mm from the reference PEI measured at any point of this surface. Volumetric wear of the articular and entire surface in mm3 will also be obtained by subtracting the measurement of retrieved PEI from the reference values of its CAD model. The linear and volumetric wear rate will be calculated by dividing the wear by the component time in situ (time from primary TKA to revision operation) and given as mm/year and mm3 /year, respectively.

CAD models of the selected PEIs will be provided by DePuy Synthes (DePuy International Ltd, registered no. 3319712, registered office at St. Anthony's Road, Leeds, LS11 8DT, UK) as support for the study.

Determination of characteristic wear patterns of the articular surface For each PEI we will obtain an articular surface linear penetration wear map. We will try to calculate and show the average linear wear distribution on this surface for each cause of TKA failure and identify a possible characteristic pattern. For this purpose, PEI size will be normalized. Side of TKA will also be normalized in order to describe medial vs lateral surface characteristics.

Identification of causes for TKA failure

The hospital code list allows the operating surgeon to select only one cause of TKA failure and consequent revision surgery. We will check for the presence of any possible additional causes for failure. For this matter, we will review the perioperative and follow up records, as well as available radiological and laboratory studies. Every single case will be classified into one or more of the following groups of common causes of TKA failure:

1. PJI according to the current European Bone and Joint Infection Society (EBJIS) definition
2. Osteolysis without component loosening (radiological assessment according to the Knee Society Roentgenographic Evaluation and Scoring System
3. Clinical loosening of the component (intraoperative evaluation)
4. Joint malalignment or component malposition
5. Instability (positive history and clinical examination)
6. Other

PJI will be also compared to all the other groups together as septic to aseptic causes, respectively.

STATISTICS AND RESULTS PRESENTATION IBM SPSS Statistics (IBM Corporation, Armonk, NY, USA) will be used for statistical analysis. The normal distribution of the wear rates of the group will be checked with the Kolmogorov-Smirnov test. In the case of normal distributions, results will be given as the arithmetic mean and standard deviation (SD). Otherwise, the median with interquartile range (IQR) will be given. The value of p <0.05 will be considered as statistically significant. In addition to the p value, we will also give a 95% confidence interval (CI), which will be used to recognize the differences between the groups and the study hypotheses.

The results will be presented in appropriate tabular and graphical representations.

NOVELTY AND CONTRBUTION TO SCIENCE We believe that we will be one of first to use CAD models for reference values to calculate wear on retrieved PEIs. Will be the first to present volumetric wear rates per cause of TKA failure. The wear of the entire PEI is far more telling than linear, wear alone and the result also incorporates the creep of polyethylene, not only the wear. We will first demonstrate the wear distribution map pattern per cause of TKA failure.

If hypotheses will be confirmed, it could be easier to declare the main cause of prosthesis failure in diagnostically demanding cases by measured PEI wear rate and its linear wear distribution pattern. The findings would influence the diagnostic algorithm in failed TKA. Identification of the articular surface linear wear distribution according to different causes of TKA failure is a new method, that represents an upgrade of current quantitative PEI measurements. Additionally, if a characteristic distribution pattern could be identified preoperatively by imaging investigations, a revision operation could be better planned and treatment targeted to the identified cause. This work represents the basis for further research.

Inclusion Criteria:

• primary TKA performed in Valdoltra Orthopedic Hospital (VOH), cement TKA, tibial component Sigma® Modular CoCr Tibia Tray Component (Non-Porous); tibial PEI Sigma® Cruciate-Retaining Curved GVF i2; femoral component Sigma® CR Femoral Component (Non-Porous); revision operation with PE insert removal performed in VOH and PE insert storage.

Exclusion Criteria:

• endoprosthesis fracture, missing patient, prosthesis type or course of primary and revision surgery data.

Sigma® Cruciate-Retaining Curved GVF i2 from DePuy Synthes (Johnson & Johnson, Warsaw, IN 46582, USA) PEIs, that were removed during the revision operation between 1.1.2013 (start of Institutional storage of removed knee endoprosthesis components) and 31.12.2021. The listed Sigma® PEI represents the largest share of inserted PEV in primary total knee arthroplasty in Valdoltra Orthopedic Hospital (VOH) in the last decade (from 1.1.2012 to 31.12.2021) and the highest number of single types retrieved PEIs in this period and represents the only reason why we selected this particular implant

No intend to share individual participant data (IPD).