This study compared the midterm results between press-fit and cemented implantation of a highly congruent, all-polyethylene patellar component. We followed prospectively 172 implants (cemented, n = 133; press-fit, n = 39). Average follow up was 6 years (range, 5-8 years). Patellofemoral complications occurred in 4 cemented patellae (2.3%). Two of these patellae required revision. Cemented implants had a significantly higher incidence of patellar maltracking (30% vs 8%; P= .005). No significant differences in the overall Knee Society scores (mean, 165; SD, 27) or any of its components relevant to patellofemoral function were detected between fixation methods. A retrieved specimen showed an intervening fibrous membrane at the implant-bone interface. The potential for macrophage-mediated osteolysis at this site is unknown. No other adverse outcome was associated with press-fit implantation. These results suggest that at midterm follow-up, press-fit implantation of this all-polyethylene patellar component may improve tracking and represents a viable alternative to cement fixation.