Over 300,000 primary total hip arthroplasty (THA) procedures were performed in 2012.¹ Many of these will be successful in achieving pain reduction but implant failure due to loosening, instability, and other factors does occur, making revision surgery necessary. Diagnostic tests such as radiography ², MRI ³, and serum metal ion levels ⁴ are available for orthopaedic surgeons to assess patients with THA but it is unclear which test, or combination of tests, is optimal for predicting implant failure. We previously developed an MRI evaluation protocol that is predictive of the pathologic adverse local tissue reactions (ALTRs) in patients with metal-on-metal ³ components but it is unclear how this technique relates to implant wear or corrosion. Therefore, the purpose of this study was to correlate indirect measures of ALTRs with direct, ex-vivo measures of implant wear and corrosion.Following local IRB approval with informed written consent, THA subjects undergoing revision surgery were enrolled. The types of THAs included: metal-on-metal (MOM), hip resurfacing (HRA), metal-on-poly (MOP), modular metal-on-poly (mMOP), ceramic-on-poly (COP), and ceramic-on-ceramic (COC). All subjects were scanned on 1.5T clinical MR scanners with an 8 channel phased array cardiac coil (GE Healthcare, Waukesha, WI). Three plane 2D-FSE images were acquired ⁵; coronal 3D MAVRIC-SL and MAVRIC-SL STIR images ⁶ were also acquired. MR images were evaluated for the presence and type of synovitis (fluid, solid, mixed), primary impression of synovium (normal, mildly abnormal, ALTR, infection, metallosis, polymeric, or old infection), synovial thickness and volume, presence and location of synovial decompression, and ALTR grade (none to severe) ⁷. Tissue samples were acquired during revision surgery and scored using Campbell’s aseptic lymphocyte-dominated vasculitis-associated lesion (ALVAL) score⁸, and the Natu⁹ and Fujishiro¹⁰ grading methods which evaluate the presence of histiocytes, particle type and load. The retrieved polyethylene implants were digitized with a 3D laser scanner to produce point clouds (NextEngine, Santa Monica CA, 12μm resolution) that were compared to CAD models of pristine implants and dimensional deviations, indicative of wear and/or deformation, were measured. The head and cup components of selected MOM implants also underwent contactless scanning (RedLux, Southampton, UK, 20 nm resolution). The point clouds were compared with best fit spheres fitted to the unworn portion of the surface to calculate linear wear and volumetric wear ¹¹. The head and neck tapers of modular designs were evaluated for corrosion: none, mild, moderate, or severe ¹². Statistical Analysis: Spearman rank correlation coefficients, Wilcoxon rank-sum tests, Fisher’s exact tests, and exact Cochran-Armitage trend tests were used to evaluate the associations between continuous or ordinal variables, categorical and continuous variables, categorical variables, and categorical and ordinal variables respectively, across all implant types. Statistical significance was set at p<0.05 (SAS, V9.3, Cary, NC). 117 THAs patients (118 hips) have been enrolled to date, a subset of which have had wear evaluation completed: MOM (n=5), HRA (n=3), mMOP (n=18), MOP (n=17), COP (n=7), and COC (n=2). For MR evaluation, significant correlations were found between synovial thickness with visual damage at the femoral stem head trunion (ρ =0.48, p=0.049), ALTR grade with visual damage of the femoral head taper (ρ=0.35, p=0.036), femoral stem in modular designs (ρ=0.59, p=0.006), and femoral head volumetric wear (ρ=0.87, p=0.001, Fig. 1). Greater volumetric wear and visual damage of the femoral stem was found in subjects with ALTR on MR images, p=0.034 and p=0.017, respectively. The visual damage score differed by impression of the synovium (p=0.017). Volumetric wear was associated with greater prevalence of histiocytes (r=0.89, p=0.041), and a greater particle density (femoral head: ρ=0.87, p=0.001, acetabular cup: ρ=0.89, p=0.041, Fig. 2). The presence of edge loading generated greater levels of histiocytes, particles present, and metal particles present (p<0.05). ALVAL scores tended to be higher with increasing wear metrics but the relation was not significant. The correlations of synovial thickness, a known predictor of ALVAL and soft tissue damage ³, with visual damage, and ALTR grade with volumetric wear, indicates that MRI detects physical wear debris from the implant. Greater wear was correlated with histologic evaluation; however, the lack of additional correlations with wear measurements and damage evaluation may be attributed to tissue sampling strategy. Further patient enrollment will aid in increasing the statistical power for correlations of implants. THAs of different bearing materials produce wear debris, the presence of which is detectable using noninvasive MR imaging. Pre-revision MRI allows for accurate definition of the magnitude and quality of the synovial response and permits targeted biopsy.