Cemented total hip arthroplasty yields reliable results in short to medium term studies, but aseptic loosening remains a problem in long-term follow up, especially in young and active patients. Aseptic loosening has been related to wear, and in order to minimize wear various alternatives to the traditional metal on polyethylene have been proposed. Both ceramic on polyethylene (COP) and metal on metal (MOM) have been shown to produce less wear than metal on polyethylene (MOP). In order to study the effect of the bearing, we have utilized identical stems and cups while comparing the different bearings. Methods and material. 396 hips were randomized to MOP, COP or MOM using a cemented triple tapered polished stem (MS-30; Sulzer Orthopedics) with a cemented polyethylene cup (Weber; Sulzer, Orthopedics) and a 28 millimeter head. For the MOP and COP articulations, a all-polyethylene cup was used with a Protasul™ metal head or a Sulox™ alumina head, whereas a polyethylene cup with metal insert was used for the MOM articulations (Weber Polyethylene Cup with Metasul™ Insert) with a Metasul™ metal head. Harris Hip Score (HHS) and radiological evaluation was performed after two, five and seven years. Ethical approval was obtained. Results. HHS was available for 338 hips after seven years. The HHS in the MOP group (116 hips) was 93.7 (SD 9.0), 93.5 in the COP group (112 hips) (SD 8.8), and 91.0 (SD 13.4) in the MOM group (110 hips). Radiographic evaluation was available for 335 hips. Radiolucencies around the stem larger than one millimeter were found in five of 115 MOP hips, seven of 111 COP hips and in seven of 110 MOM hips. Periacetabular radiolucencies identified as larger than one millimeter were found in none of the 116 MOP hips, five of the 112 COP hips and in 19 of the 110 MOM hips. Ten revisions were performed. In the MOP group there were three revisions (infection, dislocation, pain); one in the COM group (infection), and six in the MOM group (three infections, two aseptic loosening and one septic loosening). Discussion. Clinically, the three bearing types perform equally well at seven years, and there are few signs of impending failure for the stem. However, there are an alarmingly high proportion of cups showing signs of loosening. Two hips have been revised in the MOM group for aseptic loosening as opposed to none in the two other groups, which adds to the concern. The MOM articulation used in this study should be monitored closely for signs of aseptic loosening

Introduction. Radiolucencies beneath the tibial component are well recognized in knee arthroplasty; the aetiology and significance are poorly understood. Non-progressive narrow radiolucencies with a sclerotic margin are thought not to be indicative of loosening. Factors which decrease the incidence of radiolucencies include cementless fixation and the use of pulse lavage. Leg/component alignment or BMI do not influence radiolucency. We are not aware of any studies that have looked at the effect of load type on radiolucency. The Oxford domed lateral tibial component was introduced to decrease the bearing dislocation rate that was unacceptably high with the flat tibial tray. However, the introduction of the domed tibial component alters the forces transmitted through the implant-cement-bone interface. As the Oxford UKR uses a fully congruent mobile bearing, the forces transmitted through the interface with a flat tray are compressive, except for the effect of friction. However, with the domed tibial component shear forces are introduced. The aim of this study was to assess the prevalence of radiolucency beneath the previous flat design and the new domed tibial tray. Patients and methods. A consecutive series of 248 cemented lateral UKRs (1999–2009) at a single institution were assessed. The first 55 were with a flat tibia and the subsequent 193 with a domed component. One year post-op radiographs were assessed, by two observers, for the presence (full or partial) and distribution of radiolucency. The distribution and thickness of each radiolucency. Cases were excluded for missing or poorly aligned radiographs. Results. In the flat group there were 42 patients (17 male, 25 female) with 42 joints. In the domed group there were 139 patients (44 male, 95 female) with 146 joints. There was a statistical significant difference between groups for the presence and extent of radiolucency (p<0.001). In the flat group, 6 (14%) had a full radiolucency and 18 (43%) had a partial radiolucency. 14 of the 18 partial radiolucencies involved the far lateral zones (1&2) and 3 partial radiolucencies involved the keel. In the domed group 6, (4.1%) had a full radiolucency and 13 (8.9%) had a partial radiolucency. All radiolucencies observed were less than 1 mm thick. Intra-class correlation for inter-observer agreement for assessment of radiolucency was 0.679. Discussion. The introduction the domed lateral tibial component results in a reduction in prevalence of radiolucency compared to a flat component. This suggests, as the mode of fixation is identical, that the forces across the tibial-cement-bone interface have a significant effect on the development of the tissue at that interface. In particular it suggests that compressive forces alone tend to cause soft tissue to be formed at the interface, whereas shear force causes bone