Background. Pseudotumours have been associated with metal-on-metal (MoM) hip replacements. We define it as a solid mass which may have cystic components that is neither neoplastic nor infectious in aetiology. The cause of a pseudotumour is not fully understood but could be due to excessive wear, metal hypersensitivity or due to an as-yet unknown cause. Aim. We present the retrieval analysis of early failure MoM hips revised for pain, loosening or a symptomatic mass. Tissues and implants were examined for the possible causes of failure and pseudotumour formation. Corrosion as a potential new cause for pseudotumour formation will be presented. Methods. A group of 16 MoM hip replacements were collected for retrieval analysis. Six of which had a pseudotumour. An Artificial Hip Profiler (Redlux Ltd) was used to measure wear. Edge loading was determined using the 3D wear data. Tissues were histologically evaluated using a 10-point ALVAL scoring system, which strongly suggests hypersensitivity (1). Cases were assigned to one of three categories: high wear (rates >5m/yr), hypersensitivity or corrosion. Results. Of the 6 pseudotumours, 3 had edge loading, 2 had high ALVAL scores and one had corrosion at the head taper junction. The high wear group, (3 cases) all demonstrated edge loading. Histology revealed more metal wear particles and macrophages, with a low ALVAL score in these cases. Two cases were found to have hypersensitivity with a high ALVAL score, more lymphocytes with less visible wear debris and macrophages. The wear rates were also lower, <4m/yr. One total hip replacement had corrosion at the junction between the head and adaptor taper for the stem. The bearings had low wear and the tissue had extensive necrosis as well as products of corrosion enclosed in fibrin (2). There was no suggestion of hypersensitivity (low ALVAL score). Conclusion. Pseudotumours can be caused by high wear, but if this is ruled out, a hypersensitivity or corrosion product reaction should be considered

INTRODUCTION. Femoral neck narrowing (NN) following Metal-on-Metal Hip Resurfacing Arthroplasty (MoMHRA) is a well-recognised clinical phenomenon. The incidence of resurfaced hips with NN > 10% is reported to be up to 27%. Its pathogenesis is thought to be multi-factorial secondary to stress shielding, impingement, osteolysis secondary to wear/ion particles and as a result of reduced vascularity and pressure effect on cancellous bone secondary to the presence of a soft-tissue mass around the resurfaced hip. Recognised risk factors for its development include: female gender and the presence of a pseudotumour. Serum Chromium (Cr) and Cobalt (Co) are recognised surrogate markers of in-vivo wear of MoMHRA. The aims of this study were to establish whether NN is associated with increased wear. METHODS. A cohort of 214 patients with unilateral MoMHRA (139M: 75F) was included in this study. Primary osteoarthritis was the diagnosis leading to surgery for the majority of patients (208). The average age at surgery was 54.1 years old (13–73). Six different implants were used; BHR (116), Conserve plus (92), Recap (2), ASR (2), Adept (1) and Cormet (1). The average femoral component size was 49.2mm (range: 38–59). The average follow up was 4.3 years (range: 2–10). Patients were subdivided into 3 groups as per implant size. Small size component group had implants <45mm, average size group had components 45–51mm and large component size group had components >51mm. All patients had Prosthesis-Junction-Ratio (PJR) measured from postoperative (PJRpost) and at latest follow up (PJRfollow) radiographs. Measurements were made using the method described by Lilikakis1. Metal ion levels (Cr/Co) were measured at last follow-up for all patients. Cr level >5.1g/ml and Co levels >4.4 g/ml were considered high2 and patients with such levels formed the high ion group. RESULTS. For the whole cohort, mean NN was 3.2% (range: 0–32%, SD:4.3). Females (4.7%, range: 0–32%, SD: 5.8) had significantly greater NN than males (2.4%, range: 0–23%, SD: 2.9) (p=0.001). Patients in the large component group had less NN (2.1%, SD: 2.3) in comparison to the average (3.2%, SD: 3.7%) (p=0.046) and the small (5.9%, SD: 7.9) (p=0.007) component size groups. There was no difference in NN between small and average sized components (p=0.1). Patients in the high ion group (25 patients) had significantly greater NN than patients with low ions (10.1%, range: 1%–32%, SD: 8.3 Vs 2.3%, range: 0–11, SD 2.3) (p<0.0001). NN >10% was seen in 16 hips (7.5%). The odds ratio of having high ions if NN >10% was 125 (p<0.001). DISCUSSION. The findings of this study showed that greater NN occurs in females and patients with small components and in patients with high ions. Furthermore, hips with NN >10% should be investigated further as they are more likely to have increased wear