Background and purpose -- Given similar functional outcomes with mobile and fixed bearings, a difference in survivorship may favor either. This study investigated the risk of aseptic loosening for the most used subtypes of mobile-bearing rotating-platform knees, in Norway and Australia. Patients and methods -- Primary TKRs reported to the Norwegian and Australian joint registries, between 2003 and 2014, were analyzed with aseptic loosening as primary end-point and all revisions as secondary end-point. We hypothesized that no difference would be found in the rate of revision between rotating-platform and the most used fixed-bearing TKRs, or between keeled and non-keeled tibia. Kaplan-Meier estimates and curves, and Cox regression relative risk estimates adjusted for age, sex, and diagnosis were used for comparison. Results -- The rotating-platform TKRs had an increased risk of revision for aseptic loosening compared with the most used fixed-bearing knees, in Norway (RR = 6, 95% CI 4-8) and Australia (RR = 2.1, 95% CI 1.8-2.5). The risk of aseptic loosening as a reason for revision was highest in Norway compared with Australia (RR = 1.7, 95% CI 1.4-2.0). The keeled tibial component had the same risk of aseptic loosening as the non-keeled tibia (Australia). Fixation method and subtypes of the tibial components had no impact on the risk of aseptic loosening in these mobile-bearing knees. Interpretation -- The rotating-platform TKRs in this study appeared to have a higher risk of revision for aseptic loosening than the most used fixed-bearing TKRs. [ABSTRACT FROM AUTHOR]