Introduction. Primary mechanical stability is important with uncemented THR because early migration is reduced, leading to more rapid osseointegration between the implant and bone. Such primary mechanical stability is provided by the design features of the device. The aim of this study was to compare the migration patterns of two uncemented hip stems, the Furlong Active and the Furlong HAC stem; the study was designed as a randomised control trial. The implants were the Furlong HAC, which is an established implant with good long term results, and the Furlong Active, which is a modified version of the Furlong HAC designed to minimise stress concentrations between the implant and bone, and thus to improve fixation. Materials and methods. The migration of 43 uncemented femoral components for total hip replacement was measured in a randomised control trial using Roentgen Stereophotogrammetric Analysis (RSA) over two years. Twenty-three Furlong HAC and twenty Furlong Active stems were implanted into 43 patients. RSA examinations were carried out post-operatively, and at six months, 12 months and 24 months post-operatively. The patients stood in-front of a purpose made calibration frame which contained accurately positioned radio-opaque markers. From the obtained images, the 3-D positions of the prosthesis and the host bone were reconstructed. Geometrical algorithms were used to identify the components of the implant. These algorithms allowed the femoral component to be studied without the need to attach markers to the prosthesis. The migration was calculated relative to the femoral coordinate system representing the anterior-posterior (A-P), medial-lateral (M-L) and proximal-distal (P-D) directions respectively. Distal migration was termed subsidence. Results. Both stems subsided significantly during the first six months following surgery but almost all stems did not progressively subside thereafter. The Furlong Active stem experienced approximately three times the amount of subsidence of the Furlong HAC stem; this difference was significant (p = 0.02). There was one subsidence outlier (four standard deviations from the mean) for the Furlong Active stem between one and two years post-operatively. Both the stems migrated laterally and rotated into valgus. Lateral migration was greater for the Furlong Active stem; at 12 and 24 months there was a significant migration of the Furlong Active head laterally of 0.51 mm (p = 0.012) and 0.58 mm (p = 0.013) respectively. There was no significant difference in clinical scores between the implants at any RSA examination post-operatively. Discussion. The initial fixation of the Furlong Active stem was not as good as the established stem making it less likely to integrate effectively with the bone. In this study, the theoretical design of a hip replacement to minimise the stress concentration between the implant and bone and thus improve fixation actually resulted in worse implant fixation. Stems designed theoretically to improve fixation may not achieve this. Therefore we recommend that new devices should be tested using Roentgen Stereophotogrammetric Analysis. Acknowledgments This work was funded by the Furlong Charitable Research Foundation

Studies using roentgen stereophotogrammetric analysis (RSA) have shown that the femoral components of cemented total hip replacements (THR) migrate distally relative to the bone, but it is not clear whether this occurs at the cement-implant or the cement-bone interface or within the cement mantle. Our aim was to determine where this migration occurred, since this has important implications for the way in which implants function and fail. Using RSA we compared for two years the migration of the tip of the stem with that of the cement restrictor for two different designs of THR, the Exeter and Charnley Elite. We have assumed that if the cement restrictor migrates, then at least part of the cement mantle also migrates. Our results have shown that the Exeter migrates distally three times faster than the Charnley Elite and at different interfaces. With the Exeter migration was at the cement-implant interface whereas with the Charnley Elite there was migration at both the cement-bone and the cement-implant interfaces

Summary. In this study we validate that weight-bearing images are needed for accurate polyethylene liner wear measurement in total knee prostheses by measuring the difference in minimum joint space width between weight-bearing and non-weight-bearing RSA views. Introduction. Recent studies show that Model-based Roentgen Stereophotogrammetric Analysis is superior to the conventional in vivo measurements of polyethylene liner wear in total knee prostheses. Although it is generally postulated that weight-bearing (standing) views are required to detect liner wear, most RSA images are acquired in non-weight-bearing (supine) view for practical reasons. Therefore, it would be of interest to know if supine views would be sufficient for measuring TKA liner wear, defined as a change in minimum joint space width (mJSW). As a difference in mJSW between weight-bearing and non-weight-bearing RSA images has never been validated, the aim of this study is to compare the outcome of in vivo measurements of mJSW in total knee prosthesis when conducted with weight-bearing and non-weight-bearing RSA views. Method. We selected the first 16 patients with a Triathlon total knee prosthesis from a clinical study for which subsequently a weight-bearing eand a non-weight-bearing RSA image pair were acquired at one year follow up. For both images the mJSW and the corresponding contact point locations in terms of mediolateral (ML) and anterioposterior (AP) coordinates were measured. In addition, the ML stability was scored to assess knee laxity. The size of the mJSW difference between non-weight-bearing and weight-bearing views was determined and a regression analysis was conducted to investigate the effect of knee laxity to this difference, while correcting for differences between the contact point locations. Results. On average, the measured mJSW was 0.22 mm larger in the non-weight-bearing views (T-test, p < 0.05). The standard deviation of the difference was 0.22 mm. The regression analysis showed that a difference in the ML position of the contact location was an important covariate (beta = 0.255±0.054, Wald 95% CI). 13 patients had a high ML stability (< 5 deg) and three had a medium stability (5–10 deg). The regression analysis showed that on average patients with medium ML stability had a 0.17 mm larger difference in mJSW than patients with a high ML stability. Conclusion. The study shows that the mJSW is larger in non-weight-bearing views. The differences found are clinically relevant, as wear rates in practice are as small as 0.1 mm per year. Hereby it is validated that weight-bearing RSA images are more capable of detecting the mJSW in total knee prostheses. The size of the difference in mJSW between the views seems to be in relation with the knee laxity