One hundred and forty hips scheduled for a total hip arthroplasty were randomized to receive a metal-on-polyethylene bearing (n=69 hips) or a ceramic-on-ceramic bearing (n=71 hips). Outcomes of interest included aseptic revision rate, rate of component loosening, rate of periarticular osteolysis, the Western Ontario and McMaster University (WOMAC) osteoarthritis scores, the incidence of ceramic fracture, and implant survivorship. The minimum follow-up time was 20 years. Results revealed significantly greater survivorship, lower aseptic revision rate, lower rate of periarticular osteolysis, and greater functional scores in the ceramic-on-ceramic group compared to the metal-on-polyethylene group (p<0.05 for all). No cases of component loosening were observed in the ceramic-on-ceramic group vs 3 (13%) cases in the metal-on-polyethylene group.