Background Methods to reduce the revision rate of total knee prostheses because of wear-related issues are important to examine, particularly because younger patients have a disproportionately high risk of revision.Questions/purposes(1) Are long-term Knee Society knee and function scores better in younger patients with an oxidized zirconium (OxZr) TKA compared with those with a cobalt-chrome (CoCr) TKA? (2) Are there differences in radiographic signs of loosening or CT findings of osteolysis between OxZr TKAs and CoCr TKAs? (3) Are there fewer polyethylene wear particles in the OxZr TKA than CoCr TKA? (4) Do the groups differ in terms of survivorship free from revision surgery at 13 years?MethodsFrom April 2003 to January 2007, we enrolled 110 patients younger than 55 years of age in this randomized, double-blind, prospective trial. Each patient served as his or her own control and each received an OxZr femoral component in one knee and a CoCr femoral component in the other. The minimum followup was 10 years (mean, 13 years; range, 10-14 years); two died and nine were lost to followup before that time, leaving 99 patients (198 knees) for analysis. There were 28 men and 71 women with a mean age of 53 ± 6 years (range, 40-55 years). We obtained Knee Society knee scores for each knee, but Knee Society function scores, WOMAC scores, and UCLA activity scores were obtained for each patient preoperatively and at each followup. Additionally, we performed radiographic examination preoperatively and at each followup. At a minimum of 10 years (mean, 13 years) followup, we obtained CT scans in all patients. Polyethylene wear particles in the synovial fluid were analyzed at the final followup using thermogravimetric methods and scanning electron microscopy. Survivorship was ascertained using the Kaplan-Meier calculator. A sample size calculation determined that to detect a difference in the Knee Society knee score of 5 points, assuming a SD of 5 points, with an α = 0.05 and β = 0.80, a total of 90 patients would be needed in each group.ResultsAt the most recent followup, the mean Knee Society knee scores (92 versus 93 points; p = 0.857), function scores (85 versus 85 points; p = 1.000), WOMAC scores (23 versus 23 points; p = 1.000), UCLA activity scores (6.5 versus 6.5 points; p = 1.000), and range of knee motion (125° versus 127°; p = 0.365) were not different between the two groups. There was no radiographic evidence of loosening and no osteolysis visible on CT scan in either group. The weight of polyethylene wear particles produced at the bearing surface was 0.046 ± 0.010 g in 1 g of synovial fluid in patients with an OxZr femoral component and 0.0448 ± 0.0108 g in patients with a CoCr femoral component (p = 0.583). Kaplan-Meier survivorship free from revision was 97% for the OxZr group (95% confidence interval [CI], 93-100) and 98% for the CoCr group (95% CI, 93-100) at 13 years after surgery (p = 0.918).ConclusionsGiven the absence of demonstrated superiority of either the CoCr implant or the OxZr implant, we recommend that surgeons and healthcare systems can reasonably choose the less expensive device for routine use, unless there is some compelling reason in an individual patient to choose one over the other (such as severe, documented metal sensitivity).Level of EvidenceLevel I, therapeutic study.