Introduction. Model-based radiostereometric analysis (MBRSA) allows the in vivo measurement of implant loosening (i.e. migration) from a host bone by acquiring a pair of biplanar radiographs of the patient's implant over time. Focusing on total knee replacement patients, the accuracy of MBRSA in calculating tibial baseplate migration depends on the accuracy in registering a 3D model onto the biplanar radiographs; thus, the shape of the baseplate and its orientation relative to the imaging planes is pertinent. Conventionally, the baseplate coordinate system is aligned with the laboratory coordinate system, however, this reference orientation is unnecessary and may hide unique baseplate features resulting in less accurate registration (Figure 1). Therefore, the primary objective of this study was to determine the optimal baseplate orientation for improving accuracy during MBRSA, and an acceptable range of orientations for clinical use. A second objective was to demonstrate that a custom knee positioning guide repeatably oriented the baseplate within the acceptable range of orientations. Materials and Methods. A tibia phantom consisting of a baseplate rigidly fixed to a sawbone was placed in 24 orientations (combination of six rotations about X (i.e. knee flexion) and four rotations about Z (i.e. hip abduction)) with three pairs of radiographs acquired at each orientation. The radiographs were processed in MBRSA software, and the mean maximum total point motion (MTPM), an indicator of bias error during model registration, was plotted as a function of the two rotations to determine the optimal orientation and a range of acceptable orientations (Figure 2). A custom knee positioning guide was manufactured with the goal of orienting the baseplate close to the optimal orientation and within the acceptable range of orientations (Figure 3). Ten independent pairs of biplanar radiographs were acquired by repeatedly placing a knee model in the knee positioning guide, and the images were processed in MBRSA software to determine the baseplate orientation. Results and Discussion. Results showed an 85% decrease in bias error between the reference orientation (i.e. no rotation) and the optimal orientation (10° rotation about X and 5° rotation about Z). An acceptable range of orientations from 5° − 20° rotation about an axis perpendicular to the sagittal imaging plane and from 5° − 15° rotation about an axis perpendicular to the coronal imaging plane was defined as these orientations decreased the bias error by more than 50%. Additionally, the custom knee positioning guide controlled the mean orientation ± one standard deviation within the acceptable range of orientations. Conclusions. The accuracy of MBRSA is significantly improved if the tibial baseplate is placed in the range of acceptable orientations as opposed to the conventional reference orientation. A custom knee positioning guide can be used during a clinical study to repeatably position the patient's knee within the range of acceptable orientations. For any figures or tables, please contact the authors directly

The purpose of this study was to examine the influence of weight-bearing on the measurement of in vivo wear of total knee replacements using model-based RSA at 1 and 2 years following surgery. Model-based RSA radiographs were collected for 106 patients who underwent primary TKR at a single institution. Supine RSA radiographs were obtained post-operatively and at 6-, 12-, and 24-months. Standing (weight-bearing) RSA radiographs were obtained at 12-months (n=45) and 24-months (n=48). All patients received the same knee design with a fixed, conventional PE insert of either a cruciate retaining or posterior stabilized design. Ethics approval for this study was obtained. In order to assess in vivo wear, a highly accurate 3-dimensional virtual model of each in vivo TKA was developed. Coordinate data from RSA radiographs (mbRSA v3.41, RSACore) were applied to digital implant models to reconstruct each patient's replaced knee joint in a virtual environment (Geomagic Studio, 3D Systems). Wear was assessed volumetrically (digital model overlap) on medial and lateral condyles separately, across each follow-up. Annual rate of wear was calculated for each patient as the slope of the linear best fit between wear and time-point. The influence of weight-bearing was assessed as the difference in annual wear rate between standing and supine exams. Age, BMI, and Oxford-12 knee improvement were measured against wear rates to determine correlations. Weight bearing wear measurement was most consistent and prevalent in the medial condyle with 35% negative wear rates for the lateral condyle. For the medial condyle, standing exams revealed higher mean wear rates at 1 and 2 years, supine, 16.3 mm3/yr (SD: 27.8) and 11.2 mm3/yr (SD: 18.5) versus standing, 51.3 mm3/yr (SD: 55.9) and 32.7 mm3/yr (SD: 31.7). The addition of weight-bearing increased the measured volume of wear for 78% of patients at 1 year (Avg: 32.4 mm3/yr) and 71% of patients at 2 years (Avg: 48.9 mm3/yr). There were no significant (95% CI) correlations between patient demographics and wear rates. Volumetric, weight-bearing wear measurement of TKR using model-based RSA determined an average of 33 mm3/yr at 2 years post-surgery for a modern, non-cross-linked polyethylene bearing. This value is comparable to wear rates obtained from retrieved TKRs. Weight-bearing exams produced better wear data with fewer negative wear rates and reduced variance. Limitations of this study include: supine patient imaging performed at post-op, no knee flexion performed, unknown patient activity level, and inability to distinguish wear from plastic creep or deformation under load. Strengths of this study include: large sample size of a single TKR system, linear regression of wear measurements and no requirement for implanted RSA beads with this method. Based on these results, in vivo volumetric wear of total knee replacement polyethylene can be reliably measured using model-based RSA and weight-bearing examinations in the short- to mid–term. Further work is needed to validate the accuracy of the measurements in vivo

Purpose. The purpose of this study was to examine the influence of weight-bearing on the measurement of in vivo wear of total knee replacements using model-based RSA at 1 and 2 years following surgery. Methods. Model-based RSA radiographs were collected for 106 patients who underwent primary TKR at a single institution. Supine RSA radiographs were obtained post-operatively and at 6-, 12-, and 24-months. Standing (weight-bearing) RSA radiographs were obtained at 12-months (n=45) and 24-months (n=48). All patients received the same knee design with a fixed, conventional PE insert of either a cruciate retaining or posterior stabilized design. Ethics approval for this study was obtained. In order to assess in vivo wear, a highly accurate 3-dimensional virtual model of each in vivoTKA was developed. Coordinate data from RSA radiographs (mbRSA v3.41, RSACore) were applied to digital implant models to reconstruct each patient's replaced knee joint in a virtual environment (Geomagic Studio, 3D Systems). Wear was assessed volumetrically (digital model overlap) on medial and lateral condyles separately, across each follow-up. Annual rate of wear was calculated for each patient as the slope of the linear best fit between wear and time-point. The influence of weight-bearing was assessed as the difference in annual wear rate between standing and supine exams. Age, BMI, and Oxford-12 knee improvement were measured against wear rates to determine correlations. Results. Weight bearing wear measurement was most consistent and prevalent in the medial condyle with 0–4% of calculated wear rates being negative compared to 29–39% negative wear rates for the lateral condyle. For the medial condyle, standing exams revealed higher mean wear rates at 1 and 2 years; supine, 16.3 mm. 3. /yr (SD: 27.8) and 11.2 mm. 3. /yr (SD: 18.5) versus standing, 51.3 mm. 3. /yr (SD: 55.9) and 32.7 mm. 3. /yr (SD: 31.7). The addition of weight-bearing increased the measured volume of wear for 78% of patients at 1 year (Avg: 32.4 mm. 3. /yr) and 71% of patients at 2 years (Avg: 48.9 mm. 3. /yr). There were no significant (95% CI) correlations between patient demographics and wear rates. Discussion and Conclusion. This study demonstrated TKA wear to occur at a rate of approximately 10 mm. 3. /year and 39 mm. 3. /year in patients imaged supine versus standing, respectively, averaged over 2 years of clinical follow-up. In an effort to eliminate the effect of PE creep and deformation, wear was also calculated between 12 and 24 months as 9.3 mm. 3. (standing examinations), This value is comparable to wear rates obtained from retrieved TKRs. Weight-bearing exams produced better wear data with fewer negative wear rates and reduced variance. Limitations of this study include: supine patient imaging performed at post-op, no knee flexion performed, and unknown patient activity level. Strengths of this study include: large sample size of a single TKR system, linear regression of wear measurements and no requirement for implanted RSA beads with this method. Based on these results, in vivo volumetric wear of total knee replacement polyethylene can be reliably measured using model-based RSA and weight-bearing examinations in the short- to mid–term. For any figures or tables, please contact authors directly

This study used model-based radiostereometric analysis (MBRSA) to compare migration of a recently introduced cementless hip stem to an established hip stem of similar design. Novel design features of the newer hip stem included a greater thickness of hydroxyapatite coating and a blended compaction extraction femoral broach. Fifty-seven patients requiring primary total hip arthroplasty (THA) were enrolled at a single centre. Patients were randomized to receive either an Avenir collarless stem and Trilogy IT cup (ZimmerBiomet) or a Corail collarless stem and Pinnacle cup (DePuy Synthes) via a posterior or lateral approach. Both stems are broach-only femoral bone preparation. RSA beads (Halifax Biomedical) were inserted into the proximal femur during surgery. Patients underwent supine RSA imaging a 6 weeks (baseline), 6, 12, and 24 months following surgery. The primary study outcome was total subsidence of the hip stem from baseline to 24 months as well as progression of subsidence between 12 and 24 months. These values were compared against published migration thresholds for well-performing hip stems (0.5mm). The detection limit, or precision, of MBRSA was calculated based on duplicate examinations taken at baseline. Patient reported outcome measures were collected throughout the study and included the Oxford-12 Hip Score (OHS), EuroQoL EQ-5D-5L, Hip Osteoarthritis Score (HOOS) as well as visual analogue scales (VAS) for thigh pain and satisfaction. Analysis comprised of paired and unpaired t-tests with significance set at p≤0.05. Forty-eight patients (30 males) were included for analysis; 7 patients received a non-study hip stem intra-operatively, 1 patient suffered a traumatic dislocation within three weeks of surgery, and 1 patient died within 12 months post-surgery. RSA data was obtained for 45 patients as three patients did not receive RSA beads intra-operatively. Our patient cohort had a mean age of 65.9 years (±;7.2) at the time of surgery and body mass index of 30.5 kg/m2 (±;5.2). No statistical difference in total stem migration was found between the Avenir and Corail stems at 12 months (p=0.045, 95%CI: −0.046 to 0.088) and 24 months (p=0.936, 95% CI: −0.098 to 0.090). Progression of subsidence from 12-24 months was 0.011mm and 0.034mm for the Avenir and Corail groups which were not statistically different (p=0.163, 95%CI: −0.100 to 0.008) between groups and significantly less than the 0.5mm threshold (pNo statistically significant differences existed between study groups for any pre-operative function scores (p>0.05). All patients showed significant functional improvement from pre- to post-surgery and no outcome measures were different between study groups with exception of EQ-5D-5L health visual analogue scale at 12 months which showed marginally superior (p=0.036) scores in the Avenir group. This study was not powered to detect differences in clinical outcomes. This study has demonstrated no statistical difference in subsidence or patient-reported outcomes between the Corail hip stem and the more recently introduced Avenir hip stem. This result is predictable as both stems are of a triple-tapered design, are coated with hydroxyapatite, and utilize a broach-only bone preparation technique. Both stem designs demonstrate migration below 0.5mm suggesting both are low-risk for aseptic loosening in the long-term

Introduction. The most common method for accurate kinematic analysis of the knee arthroplasty uses bi-planar fluoroscopy and model-based RSA. The main challenge is to have access to reverse-engineered CAD models of the implant components, if not provided by the company, making this method impractical for a clinical study involving many types or sizes of implants. An alternative could be to reconstruct the 3D primitive features of the implant, such as cylindrical pegs, flat surfaces and circular boundaries, based on their 2D projections. This method was applied by Kaptein et al. (2006) for hip implants. However, despite its broad potential, it has not yet been applied for studying TKA kinematics. This study develops a methodology for feature-based RSA of TKA and investigates the range of accuracies in comparison to model-based RSA. Methods. Joint-3D software was developed in the MATLAB programming language to segment and fit elementary 2D features such as circles, lines, and ellipses to the edges of the parts on the radiographs (Figure 1). The software has the capability to reconstruct the 3D location and orientation of the components based on their 2D projections. To test the accuracy of the system a standard primary knee replacement system (Zimmer NexGen) was implanted on bone replica models, and positioned at 0° to 120° flexion at 30° intervals, simulating a lunge activity. For each pose, a multi-planar radiography system developed in our lab (Amiri et al., 2011) was used to take a sagittal and a 15° distally rotated radiograph (Figure 2a). Figure 1 shows the features C, L, and E segmented on the tibia and femur. The 3D reconstruction is performed based on a number of functions: Functions ‘f’ and ‘g’ reconstruct a 3D point or line based on their 2D projections. Function ‘h’ finds the plane containing the 3D circular edge based on its two projection ellipses. Function ‘i’ finds the 3D location of a line based on one projection line, and a known 3D vector normal to the solution 3D line. Based on these, the coordinate systems of the components were reconstructed (Figure 2b):. Femur_Origin=f(C1A,C1B);. Femur_Anteroposterior=g(L1A, L1B);. Femur_Proximodistal=g(L2A,L2B);. Femur_Mediolateral=i(L,C1A–C1B),{L=L1: if flexion<45°; L=L2: if flexion>45°};. E_3D=h(E1A,E1B);. Tibia_Origin=f(E1A_Centre,E1B_Centre);. Tibia_Anteroposterior=g(L3A,L3B);. Tibia_Mediolateral=cross(E_3D, Tibia_Anteroposterior);. Tibia_Proximodistal=cross(Tibia_Anteroposterior, Tibia_Mediolateral). To determine the errors, model-based RSA measures were used as the reference using the reverse-engineered models of the components in JointTrack software (University of Florida). Results. The overall accuracies in terms of bias (the mean error) and precision (standard deviation of the errors) are shown in Figure 3. The bias was within 0.5–1 mm and 0.9–1.2°, and the calculated precision was in the range of 0.4–0.6 mm and 0.7–1.0°. The overall accuracy was 0.8±0.6 mm and 1±0.7°. Discussion. The very good accuracies obtained show the practicality of the methodology. The methodology can be easily worked out for any type of implant based on the primitive geometric features at the bone-implant interface. This method can be extremely useful in a large clinical study by eliminating the need for having the 3D models of many types and sizes of the implant available

Purpose. The purpose of this study was to assess the biomechanical stability of the a total ankle arthroplasty system using longitudinal migration (LM) and inducible displacement (ID) measures. This study is the first study of its kind to assess total ankle arthroplasty (TAA) implant micromotion using model-based radiostereometric analysis (MBRSA). Method. Twenty patients underwent TAA that implanted the Mobility(TM) (DePuy, Warsaw IN). The mean (SD) age was 60.4 (12.5) and BMI was 29.1 (2.8) kg/m. 2. One surgeon performed all surgeries. All patients included in this study had given informed consent. Capital Health Research Ethics Board had approved this study. Uniplanar medial-lateral RSA X-ray exams were taken postop (double exam), at six wk, three mth, six mth, one yr and two yr followup times using a supine, unloaded position. Standing medial-lateral exams were taken at three mth, six mth, one yr and two yr followup intervals. LM and ID micromotions were assessed using Model-based RSA 3.2 software (Medis specials, Leiden, The Netherlands). Implant micromotions (x, y, z, Rx, Ry, Rz, MTPM) were determined and assessed for each subject using model-based pose estimation, and the implant-based coordinate system. The Elementary Geometric Shapes module from the Model-based RSA 3.2 software was used to assess the micromotion of the tibial component spherical tip due to implant symmetry. Results. The median (range) maximum total point motion (MTPM) for the implants at 2 year followup were 1.23 mm (0.39–1.95 mm) for the talar implant and 0.96 mm (0.17–2.28 mm) for the spherical tip of the tibia implant. Generally for each subject and implant component, the slopes of the migration curves decreased over time. The talar and tibial implants mean LM showed initial subsidence in the y-direction (migration into the bone) followed by stabilization patterns at one year followup. The median (range) of two year MTPM ID for the talar component was 0.39 (0.27–1.06) mm. At the one year and two year followup times the ID were almost all below the detection limit of 0.85 mm. The highest measured displacement for any one talar component at either of these times was 1.06 mm. Hence, the implant was displaced at least 0.21 mm under loading. The median (range) of one year and two year MTPM ID for the tibial component spherical tip was 0.08 (0.03–0.19) mm. The tibial component spherical tip demonstrates no ID in terms of MTPM greater than the 0.22 mm detection limit. Conclusion. The implant subsides directly into the bone in the line of primary loading during standing or walking. For most of the patients the two year LM for the Mobility(TM) demonstrates a typical subsidence-stabilization behaviour seen in many RSA studies of orthopaedic implants. Based on the results of this study the Mobility(TM) components show no measurable ID. This is the first study of its kind internationally for total ankle arthroplasty and offers novel insight into the need for prosthetic design change

Background. Migration analysis after total joint arthroplasty are performed using EBRA analysis (Krismer et al., 1997) or - more accurate but also much more cost-intensive and time-consuming – via radiostereometric analysis (RSA). For the latter, additional radiographs from two inclined perspectives are needed in regular intervals in order to define the position of the implant relative to tantalum bone markers which have been implanted during surgery of the artificial joint (Fig. 1). Modern analysis software promises a migration precision along the stem axis of a hip implant of less than 100 μm (Witvoet-Brahm et al., 2007). However, as the analysis is performed semi-automatically, the results are still dependent on the subjective evaluation of the X-rays by the observer. Thus, the present phantom study aims at evaluating the inter- and intra-observer reliability, the repeatability as well as the precision and gives insight into the potential and limits of the RSA method. Materials and Methods. Considering published models, an RSA phantom model has been developed which allows a continuous and exact positioning of the prostheses in all six degrees of freedom (Fig. 2). The position sensitivities of the translative and rotative positioning components are 1 μm and 5 to 24, respectively. The roentgen setup and Model-Based RSA software (3.3, Medis specials bv, Leiden, Netherlands) was evaluated using the SL-PLUS® standard hip stem (size 7, Smith & Nephew, Baar, Switzerland). The inter-observer (10 repetitions) and intra-observer (3 observers) reliability have been considered. Additionally, the influences of the model repositioning and inclination as well as the precision after migration and rotation along the stem axis are investigated. Results and Discussion. Precision along the stem axis was determined to 161 μm (± 230 μm), in the lateral plane 100 μm (± 85 μm) and maximal rotations to 0.524° (± 1.268°). High reproducibility (intra-observer reliability) is reported with relevant influences of the inclination of the implant on the radiograph, in particular for the first clinical scene which serves as a reference. Deviations after translations along the stem axis are 0.37 ± 1.92% and −3.28 ± 6.62% after rotations. In conclusion, the precision given by the software producer of less than 100 μm could not be verified. Beside the limitations from the software, potential sources of errors are the subjective analysis by the observer, a small number of bone markers and the positioning of the implant (patient) during X-ray examination. Though, Model-Based RSA largely outmatches the EBRA approach in terms of measuring implant migration. However, standardization of the X-rays and RSA analysis is recommended

Introduction. Surgeons performing a total knee replacement (TKR) have two available techniques available to help them achieve the proper bone resections and ligament tension – gap balancing (GB) and measured resection (MR). GB relies on balancing ligaments prior to bony resections whereas bony resections are made based on anatomical landmarks in MR. Many studies have been done to compare the joint kinematics between the two techniques, however the results have been varied. These studies were not done with anatomically designed prostheses. The Journey II (Smith & Nephew, Memphis, TN) is one such design which attempts to mimic the normal knee joint structure to return more natural kinematics to the joint, with emphasis on eliminating both paradoxical anterior motion and reduced posterior femoral rollback. Given the design differences between anatomical and non-anatomical prostheses, it is important to investigate whether one technique provides superior kinematics when an anatomical design is used. We hypothesize that there will be no difference between the two techniques. Methods. A total of 56 individuals were recruited to receive a Journey II prosthesis and randomized evenly to groups where the GB technique or MR technique is used. For all patients in the study, a series of radiostereometric analysis (RSA) images were acquired at 3-months post-operatively at different knee flexion angles, ranging in 20° increments from 0° to 120°. Model-based RSA software (RSACore, Leiden, Netherlands) was used to obtain the 3D positions and orientations of the femoral and tibial implant components, which were in turn used to obtain kinematic measures (contact locations and magnitude of excursion) for each condyle. Results. Preliminary results for the anterior-posterior (AP) contact locations from 33 patients (18 GB, 15 MR) are displayed in Figure 1. There were no significant differences in medial and lateral contact locations between the GB and MR groups for all angles of flexion. However, the pattern of medial contact for the MR technique displays more paradoxical anterior motion at mid-flexion (40°–60°) than the GB group. There were no significant differences in magnitude of excursion between groups on both medial (mean difference=1.96 mm, p=0.16) and lateral (mean difference=0.21 mm, p=0.79) condyles, indicating that posterior femoral rollback is similar between groups. Conclusions. Early results suggest that the MR technique is associated with slightly more abnormal kinematics than the GB technique when an anatomical prosthesis design is used for TKR. The GB technique may be more appropriate than MR technique for implanting anatomically designed knee replacements. For any figures or tables, please contact authors directly

Introduction. Despite improvements in the survivorship of total knee replacements (TKR) over the years, patient satisfaction following TKR has not improved, with approximately 20% of patients recording dissatisfaction with their new knee joint. It is unclear why many patients feel this way, but it may relate in part to implant designs that do not provide a “natural” feeling knee. Implant manufacturers continue to introduce new concepts for implant design, which are essential for reaching the goal of a “normal” knee after TKR surgery. The Journey II TKR (Smith & Nephew) was developed with this goal in mind. Its anatomical design attempts to mimic the normal knee joint structure to return more natural kinematics to the joint, with emphasis on eliminating both paradoxical anterior motion and reduced posterior femoral rollback. Our objective is to examine patients receiving the Journey II TKR to measure the knee joint contact kinematics of the Journey II TKR compared to a non-anatomically designed implant by the same manufacturer. We hypothesize that the Journey II TKR will have more natural contact kinematics that differ from the non-anatomically designed implant. Methods. A total of 28 individuals were recruited to receive a Journey II TKR, matching an existing prior cohort with a non-anatomical design from the same manufacturer (Legion TKR, Smith & Nephew). For both groups, a series of radiostereometric analysis (RSA) images were acquired at different knee flexion angles, ranging in 20° increments from 0° to 120°. Model-based RSA software (RSACore, Leiden, Netherlands) was used to obtain the 3D positions and orientations of the femoral and tibial implant components, which were in turn used to obtain kinematic measures (contact locations and magnitude of excursion) for each condyle. Results from the Journey II TKR group at 3 months post-operation were compared to the 2-year post-operative measurements from the Legion TKR group. Results. Preliminary results for the anterior-posterior (AP) contact locations from 38 patients (15 Journey II TKR, 23 Legion TKR) are displayed in Figure 1. The Journey II TKR group showed more anterior contact medially and laterally relative to the Legion TKR group at many angles, however the overall pattern from 0°–120° was similar. There was no significant difference between groups with respect to magnitude of excursion on both medial (mean difference=1.29 mm, p=0.19) and lateral (mean difference=0.69 mm, p=0.60) condyles. Conclusions. Early results suggest that paradoxical anterior motion and reduced posterior femoral rollback are present in the anatomically designed Journey II TKR, suggesting that it does not provide substantially different knee kinematics compared to the non-anatomically designed Legion TKR. For any figures or tables, please contact authors directly

Background. Surgeons generally perform total knee replacement using either a gap balancing or measured resection approach. In gap balancing, ligamentous releases are performed first to create an equal joint space before any bony resections are performed. In measured resection, bony resections are performed first to match anatomical landmarks, and soft tissue releases are subsequently performed to balance the joint space. Previous studies have found a greater rate of coronal instability and femoral component lift-off using the measured resection technique, but it is unknown how potential differences in loading translate into component stability and fixation. Methods. Patients were randomly assigned at the time of referral to a surgeon performing either the gap balancing or measured resection technique (n = 12 knees per group). Both groups received an identical cemented, posterior-stabilized implant. At the time of surgery, marker beads were inserted in the bone around the implants to enable radiostereometeric analysis (RSA) imaging. Patients underwent supine RSA exams at 0–2 weeks, 6 weeks, 3 months, 6 months, and 12 months. Migration of the tibial and femoral components including maximum total point motion (MTPM) was calculated using model-based RSA software. Knee Society Scores were also recorded for each group. Results. At 12 months follow-up, there were no revisions or adverse events. There were no differences in translation or rotation between the measured resection and gap balancing groups at 12 months, including for MTPM of the tibial component (mean 0.67 mm vs. 0.69 mm, p = 0.77, Fig. 1) and the femoral component (mean 0.71 mm vs. 0.51 mm, p = 0.25, Fig. 2). At 6 weeks, tibial components had greater (p = 0.01) anterior tilt in the measured resection group (0.08 deg) while the gap balancing group had greater posterior tilt (0.14 deg), but there were no differences from 3 months onwards (Fig. 3). Patients in both groups improved in Knee Society scores from pre- to post-operatively, with no difference in score between the groups at pre-operation (p = 0.56) or post-operation (p = 0.54). Discussion. Implants in both the gap balancing and measured resection groups were well fixed after 12 months, with no differences in translations or rotations between the two groups as of the latest time points. Both surgical techniques result in adequate fixation for total knee replacement. Future work will include measuring the contact location and possible condylar lift-off with flexion within this cohort. For any figures or tables, please contact authors directly (see Info & Metrics tab above).

Introduction. The number of revisions of total knee replacements (TKR) increases annually. Because of reduced bone stock, stable fixation of the implant is important. The femoral and tibial components are usually cemented whereas stems can be placed either cemented or press-fit (hybrid construct). To assess the stability of revision TKR with either cemented or hybrid places implants a randomized controlled trial (RCT) was executed, by using radiostereometric analysis (RSA). The short-term results of this RCT showed no differences between the two groups in stability and clinical outcomes. Although there were no clinical or radiological signs of loosening, both groups showed implants micromotion > 1 mm or degree. These findings might indicate the possibility of loosening later in time; therefore, the current study investigated the stability of cemented versus hybrid-placed revision TKR 6.5 years after surgery. Additionally, clinical results were evaluated. Methods. Of the 32 patients in the original RCT, 23 (12 cement, 11 press-fit) were available for mid-term follow-up measures. RSA images taken at baseline, 6 weeks, 3, 6, 12 and 24 months postoperatively were used from the previous study. New RSA images were taken at median 6.5 years (range 5.4–7.3) postoperatively. Stability of the femoral and tibial implants was assessed by using model-based RSA software (RSAcore, Leiden, The Netherlands) to determine micromotion. Clinical results were evaluated using the Knee Society Score (KSS), the Knee injury and Osteoarthritis Outcome Score (KOOS), active flexion, and VAS pain and satisfaction. Stability and clinical outcome were compared between the two groups using independent t-tests or Mann-Whitney U tests when applicable. Results. The median total translation at 6.5 years was 0.37 (0.13–1.96) mm and median total rotation 0.62 (0.11 – 2.81)° for the femoral component. For the tibia component the median total translation was 0.41 (0.10 – 1.04) mm and the median total rotation 0.61 (0.09 – 1.99)°. There were no differences in total translation and total rotation of the femoral and tibial component between the two groups. Additionally, none of the clinical scores differed between the groups. Interestingly, in the group with cemented stems five tibia implants showed > 1 mm or degree migration compared to zero in the hybrid group (p=0.02; Figure 1). Conclusion. There was no difference in stability and clinical outcome between fully cemented and hybrid-placed revision TKR 6.5 years postoperatively. Until now micromotion >1 mm or degree in the tibial components of the cemented group has not yet resulted in re- revisions. The patients will be followed to examine the consequences of these amounts of micromotion in this type of implant in the long-term. Figure 1. Scatter plot of total translation (x-axis) and total rotation (y-axis) for the tibia component at 6.5 years follow-up for fully cemented and hybrid-placed revision TKA

Introduction. Measuring the step off during total knee replacement (TKR) is a newly developed operative strategy (“spacer technique”; Heesterbeek et al, KSSTA 2014;22(3):650–9) to determine the optimal contact point (CP) of the femur with the tibia postoperative and to balance the posterior cruciate ligament (PCL) in cruciate-retaining TKR. Engineers have calculated the ideal step off for every size of the TKR, for which the tibiofemoral contact point in 90° will be at the designed position. With this study we determined the postoperative CP in CR-TKA and investigated whether (adverse) clinical outcome was correlated with the CP. Methods. 23 patients presenting with non-inflammatory osteoarthritis, a good functioning PCL, and indication for surgery with a PCL-retaining TKR were selected. Intraoperative PCL balancing was performed with the spacer technique. At 3 months postoperative, a pair of mediolateral radiographs was made using a set-up used for radiostereometric analysis (RSA). The patient was positioned standing with the operated leg in 90 degrees, 50% weight-bearing, knee flexion on a 30 cm-step. Model-based RSA software (RSAcore) was used to determine the 3D positions of the femur and tibia component, that were exported to custom-written software for determining the CP. The CP was defined as the point with the smallest distance between both the medial and lateral femur condyles and tibia plateau. It is expressed as the ratio of the anterior-posterior CP distance and the maximum anterior-posterior tibia plateau size, with 0 being anterior, 1 being posterior. Patients with reduced flexion capacity at follow-up, leading to manipulation under anaesthesia and/or scopic releases, were categorized as COMP, the other patients as no-COMP. CP was compared between these groups. Results. Preliminary data show that the mean medial CP of the total group was 0.51 (sd 0.05), mean lateral CP was 0.61 (sd 0.03) (p<0.001). Six out of 23 patients had flexion-related complications and for this reason further patient inclusion was stopped. The medial CP of the COMP-group (n=6) was at 0.54 (sd 0.01), which was significantly more posterior than the medial CP of the no-COMP group (n=17) (0.50 (sd 0.05)) (p=0.004). (Figure 1) The lateral CP was similar for both groups (p=0.76). Discussion. The medial CP relates to clinical outcome; patients with reduced flexion capacity had a more posterior CP. This might be an indication for a too tightly balanced PCL, but we need to investigate this further. None of the patients had a medial CP at the theoretically optimal position

Articulation of the polyethylene (PE) insert between the metal femoral and tibial components in total knee replacements (TKR) results in wear of the insert which can necessitate revision surgery. Continuous PE advancements have improved wear resistance and durability increasing implant longevity. Keeping up with these material advancements, this study utilises model-based radiostereometric analysis (mbRSA) as a tool to measure in vivo short-term linear PE wear to thus predict long-term wear of the insert. Radiographic data was collected from the QEII Health Sciences Centre in Halifax, NS. Data consisted of follow-up RSA examinations at post-operative, six-, 12-, and 24-month time periods for 72 patients who received a TKR. Implanted in all patients were Stryker Triathlon TKRs with a fixed, conventional PE bearing of either a cruciate retaining or posterior stabilised design. Computer-aided design (CAD) implant models were either provided by the manufacturer or obtained from 3D scanned retrieved implants. Tibial and femoral CAD models were used in mbRSA to capture pose data in the form of Cartesian coordinates at all follow-ups for each patient. Coordinate data was manually entered into a 3D modeling software (Geomagic Studio) to position the implant components in virtual space as presented in the RSA examinations. PE wear was measured over successive follow-ups as the linear change in joint space, defined as the shortest distance between the tibial baseplate and femoral component, independently for medial and lateral sides. A linear best-fit was applied to each patient's wear data; the slope of this line determined the annual wear rate per individual patient. Wear rates were averaged to provide a mean rate of in vivo wear for the Triathlon PE bearing. Mean linear wear per annum across all 72 patients was 0.088mm/yr (SD: 0.271 mm/yr) for the medial condyle and 0.032 mm/yr (SD: 0.230 mm/yr) for the lateral condyle. Cumulative linear wear at the 2-year follow-up interval was 0.207mm (SD: 0.565mm) and 0.068mm (SD: 0.484mm) for the medial and lateral condyles, respectively. Linear PE wear measurements using mbRSA and Geomagic Studio resulted in 0.056mm/yr additional wear on the medial condyle than the lateral condyle. Large standard deviations for yearly wear rates and cumulative measurements demonstrate this method does not yet exhibit the accuracy needed to provide short-term in vivo wear measurement. Inter-patient variability from RSA examinations is likely a source of error when dealing with such small units of measure. Further analysis on patient age and body mass index may eliminate some variability in the data to improve accuracy. Despite high standard deviations, the results from this research are in proximity to previously reported linear wear measurements 0.052mm/yr and 0.054mm/yr. Linear wear analysis will continue upon completion of >100 patients, in addition to volumetric PE wear over the entire articulating surface

Introduction. Debate over appropriate alignment in total knee arthroplasty has become a topical subject as technology allows planned alignments that differ from a neutral mechanical axis. These surgical techniques employ patient-specific cutting blocks derived from 3D reconstructions of pre-operative imaging, commonly MRI or CT. The patient-specific OtisMed system uses a detailed MRI scan of the knee for 3D reconstruction to estimate the kinematic axis, dictating the cutting planes in the custom-fit cutting blocks machined for each patient [1, 2]. The purpose of this study was to evaluate the correlation between post-operative limb alignment and implant migration in subjects receiving shape match derived kinematic alignment. Methods. In a randomized controlled trial comparing patient-specific cutting blocks to navigated surgery, seventeen subjects in the patient specific group had complete 1 year data. They received cruciate retaining cemented total knee replacements (Triathlon, Stryker) using patient-specific cutting blocks (OtisMed custom-fit blocks, Stryker). Intra-operatively, 6–8 tantalum markers (1 mm diameter) were inserted in the proximal tibia. Radiostereometric analysis (RSA) [3, 4] exams were performed with subjects supine on post-operative day 1 and at 6 week, 3, 6, and 12 month follow-ups with dual overhead tubes (Rad 92, Varian Medical Systems, Inc., Palo Alto, CA, USA), digital detectors (CXDI-55C, Canon Inc., Tokyo, Japan), and a uniplanar calibration box (Halifax Biomedical Inc., Mabou, NS, Canada). RSA exams were analyzed in Model-based RSA (Version 3.32, RSAcore, Leiden, The Netherlands. Post-operative limb alignment was evaluated from weight-bearing long-leg films. Results. Post-operative limb alignments ranged from 5 degrees of varus to 5 degrees of valgus. Comparing implant migration to post-operative alignment did not demonstrate a relationship between deviation from neutral mechanical alignment and increased migration (Pearson correlation coefficient = 0.25, P = 0.33) (Figure 1). Conclusions. Previous studies have suggested that alignment of greater than 3 degrees from neutral may have adverse effects on implant survivorship [5], but this early data does not suggest increased migration with non-neutral alignment. Continued evaluation with RSA to 2 years will be performed to monitor these subjects over the longer term

Background. Patient specific instrumentation (PSI) for total knee replacement (TKR) has demonstrated mixed success in simplifying the operation, reducing its costs, and improving limb alignment. Evaluation of PSI with tools such as radiostereometric analysis (RSA) has been limited, especially for cut-through style guides providing mechanical alignment. The primary goal of the present study was to compare implant migration following TKR using conventional and PSI surgical techniques, with secondary goals to examine whether the use of PSI reduces operative time, instrumentation, and surgical waste. Methods. The study was designed as a prospective, randomized controlled trial of 50 patients, with 25 patients each in the PSI and conventional groups, powered for the RSA analysis. Patients in the PSI group received an MRI and standing 3-foot x-rays to construct patient-specific cut-through surgical guides for the femur and tibia with a mechanical alignment. All patients received the same posterior-stabilized implant, with marker beads inserted in the bone around the implants to enable RSA imaging. Intraoperative variables such as time, number of instrumentation trays used, and mass of surgical waste were recorded. Patients underwent supine RSA exams at multiple time points (2&6 weeks, 3&6 months and yearly) with 6 months data currently available. Migration of the tibial and femoral components was calculated using model-based RSA software. WOMAC, SF-12, EQ5D, and UCLA outcome measures were recorded pre-operatively and post-operatively. Results. There were no demographic differences between groups. One patient in the PSI group was revised for infection, and three patients required manipulation, with no revisions or manipulations in the conventional group. There was no difference in maximum total point motion between groups for the tibia (mean 0.50 vs. 0.50 mm, p = 0.98) or femur (mean 0.46 vs. 0.48 mm, p = 0.87). The PSI group displayed greater tibial posterior tilt (p = 0.048, Fig. 1) and greater femoral anterior tilt (p = 0.01) and valgus rotation (p = 0.04, Fig. 2) than the conventional group, but there were no other differences in migrations. The PSI group required less instrument trays than the conventional group (mean 4.8 vs. 8.1 trays, p < 0.0001), but procedure time was equivalent (mean 79 vs. 74 min, p = 0.06). The PSI group produced less recyclable waste (mean 0.3 vs. 1.4 kg, p < 0.001), but total waste (Fig. 3) was equivalent between groups (mean 10.1 vs. 10.6 kg, p = 0.32). At 6 months there was no difference between groups for SF-12, WOMAC, EQ5D, or UCLA scores. Discussion. At early RSA follow-up, the two groups were broadly similar in implant fixation except for small rotational changes in the tibial and femoral components. The PSI group provided minimal or no advantage over the conventional group for operative time, instrumentation used, or surgical waste produced. The observed increase in manipulations in the PSI group is concerning, and requires additional investigation. Further radiographic and economic analysis is underway to determine if there is any benefit to the use of PSI for TKR during the perioperative and early follow-up period. For any figures or tables, please contact authors directly (see Info & Metrics tab above).

Purpose. To determine if minimally invasive surgery (MIS) for primary hip replacement surgery increases the risk of long term aseptic loosening as predicted by implant micromotion measured with radiostereometric analysis (RSA). Method. Ninety patients undergoing primary THA for osteoarthritis (exclusion criteria: post-traumatic arthritis, rheumatoid arthritis, hip dysplasia, previous hip infection) were randomized to undergo THR surgery utilizing the standard direct lateral approach (n=45; 24 male; age=58 yrs; BMI=27) or MIS via a one-incision direct lateral approach using specific instrumentation (n=45; 23 male; age=55; BMI=29). Uncemented acetabular and femoral (ProfemurZ) components were used with ceramic on ceramic bearings. The femur was marked with 9 tantalum beads placed in the greater trochanter, lesser trochanter, and femoral shaft distal to the tip of the prosthesis. Post-operative care was be standardized according to the care maps at our institution. Primary outcome measure was femoral stem MTPM (maximum total point motion) measured using Model-based RSA. Stereo supine X-rays were taken before weight bearing and 3, 6, and 12 months postoperatively. At the same time intervals Harris Hip Score, Oxford-12, WOMAC, and SF36 questionnaires were administered. Rates of infection, dislocation and revision were recorded. Results. Nine patients were lost to follow-up (4 due to missing post-op exams; 5 did not have enough beads placed during surgery). Ten patients were revised (2 due to failure of the ceramic femoral head; 8 due to long (38.5 mm) neck fractures at 17–30 months postop). There were no differences between groups for all outcome measures. Mean MTPM at 24 months was 3.0mm (SD=2.0mm) for the MIS group and 2.9mm (SD=2.1mm) for the standard group. Conclusion. No difference between groups at one year indicates MIS for uncemented primary THR through a direct lateral approach does not appear to negatively affect stability of the femoral stem. However, there was a high rate of femoral neck fracture in the study cohort that was not related to micromotion or treatment group that indicates a problem with the design of the titanium alloy modular femoral neck

Purpose. The purpose of this study was to assess the clinical outcomes over two years for total ankle arthroplasty (TAA) using Short Form-36, Foot Function Index and Ankle Osteoarthritis Scores, and to compare these with radiostereometric analysis longitudinal migration and inducible displacement results. Method. Twenty patients undergoing TAA implanting the Mobility Total Ankle System (DePuy, Warsaw IN) were assessed at 3mth, 6mth, 1yr and 2yr followup periods by model-based radiostereometric analysis, MBRSA 3.2 (Medis specials, Leiden, The Netherlands), for longitudinal migration (LM) and inducible displacement (ID). The same subjects completed clinical outcome questionnaires at these followup periods for Short Form-36 (SF-36; Physical Component Scores (PCS) and Mental Component Scores (MCS)), Foot Function Index (FFI) and Ankle Osteoarthritis Scores (AOS). Descriptive statistics and Pearson correlations (alpha = 0.05) were calculated using Minitab 15 (Minitab Inc., State College PA). Results. For the PCS of SF-36, FFI and AOS the scores were significantly different at 2 year followup when compared to preoperative values; p = 0.005, 0.0002 and 0.0003 respectively. The PCS on average increased with respect to pre-operative by 10 points (SD = +/−13), while the MCS on average did not change with respect to pre-operative (SD = +/− 12). The FFI on average decreased by 25 points with respect to pre-operative (SD = +/− 18) and AOS on average decreased by 23 points with respect to pre-operative (SD = +/− 21). There were several correlations for the 2 yr results: AOS to FFI of r = 0.92 (p = 0.000); AOS to PCS of r = −0.67 (p = 0.005); AOS to MCS of r = −0.51 (p = 0.046); AOS to talar component ID of r = 0.70 (p = 0.004); AOS to the talar component LM of r = 0.62 (p = 0.046). PCS related better than MCS to both AOS and FFI. The LM of the talar component and tibial component were not significantly correlated, r = 0.18 (p = 0.62). The ID of the talar component and tibial component were not significantly correlated, r = 0.48 (p = 0.07). The latter result may be too underpowered to determine a significant difference; due to the small sample size. Conclusion. The outcome scores of AOS, FFI, SF-36 (PCS) and SF-36 (MCS) were correlated to each other. The strongest outcome score relationships were AOS to FFI, followed by AOS to SF-36 (PCS). The correlation of AOS to the talar component LM and ID suggests that the implant performance may be related to the stability of the talar component