Introduction. Failure of metal on metal (MOM) total hip arthroplasty (THA) and resurfacing arthroplasty (HRA) due to development of adverse local tissue reaction (ALTRs) is a significant problem. The prevalence of ALTRs in asymptomatic MOM arthroplasty patients is highly variable. The purpose of this prospective, longitudinal study was to: 1) determine MRI ALTR prevalence in patients with HRA; 2) determine if patients with HRA have a greater rate of MRI ALTRs compared to control patients with ceramic on poly (COP) THA; and 3) evaluate changes in patient reported outcomes between these implant designs. Methods. Following IRB approval with informed consent, self-reported asymptomatic primary COP and HRA patients greater than one year post arthroplasty were evaluated with 4 annual (TP1–4) MRIs using a standardized protocol and serum ion level testing. Morphologic and susceptibility reduced images were acquired for each hip and evaluated for synovial thickness, volume, capsule dehiscence and the presence of ALTR. Patient reported outcomes were evaluated by Hip Disability and Osteoarthritis Outcome Scores (HOOS). Analyses were performed to detect differences of synovial thickness and volume, and HOOS subgroups between and within bearing surfaces at each time point and over time, and to compare the time to and the risk of developing MRI ALTR. Analyses were adjusted for age, gender, and length of implantation. Results. 86 hips were evaluated at the initial time point (TP1): HRA, n=41, age=58±8 y.o., 34/7 (M/F); COP, n=45, age=65±9 y.o., 22/23 (M/F). The mean synovial volume of the subjects was similar at TP1, and the synovial volume in HRA subjects increased significantly from TP1 to TP4 (mean volume difference=9 cm3, p=0.031) while COP subjects displayed a minimal increase from TP1 to TP4 (mean volume difference =0.3 cm3, p=0.9). The synovial volume in HRA subjects tended to be larger and with greater variability than COP subjects at TP4 (HRA=15±58 cm3; COP=6 ±10 cm3, p=0.3). By TP4, MRI ALTRs developed in 13/41 HRA subjects (mean time to ALTR=1.5 yr) compared to 2/25 COP subjects (mean time to ALTR=2.0 yr), p=0.001. While risk of MRI ALTRs in HRA subjects was 9.9 times the risk in COP subjects (p=0.07), 68% of HRA subjects did not have an MRI ALTR present. Significant differences of [Co] and [Cr] in HRA subjects with MRI ALTR present were not detected at 2/4 time points and 1/4 time points, respectively, and there were no significant differences in ion levels in the COP group. The HOOS sub-scores of Pain and Activity of Daily Living scores of HRA subjects were similar to COP subjects, and HRA subjects had better Sporting Ability than COP subjects at all time points. HRA subjects had higher Symptoms at TP1 (p=0.016) but both cohorts had high scores (HRA=92.3/100; COP=87.7/100). Initially dehiscent posterior capsules of HRA subjects (9%) tended not to resolve (10/11, 91%), while dehiscent COP subjects (15%) had greater resolution (6/22, 27%). The presence of osteolysis was limited (HRA: 6/41; COP: 3/45; p=0.101). Conclusion. A greater synovial volume and higher rate of MRI ALTRs was found in HRA subjects, while the patient symptomatology via HOOS remained slightly better in HRA subjects than COP subjects, suggesting a poor association between MRI ALTRs and symptoms. The presence of posterior decompression of the posterior capsule into the bursa alone was not necessarily indicative of an ALTR, and is likely a normal postoperative finding. The inconsistent findings of the serum ion level testing further supports MRI as a non-invasive imaging modality capable of assessing peri-prosthetic soft tissue complications, and should be considered as part of the routine patient follow up to allow early detection and monitoring of ALTRs
