Introduction: Knee osteoarthritis is the main indication for primary total knee arthroplasty (TKA). It is now accepted that cementless implantation of the femoral component provides equivalent results to cemented one, however, the optimal fixation method of the tibial component remains controversial. The purpose of this study was to compare the survivorship of cemented versus cementless tibial baseplate in primary total knee arthroplasty. Materials and methods: We carried out a retrospective, monocentric study, including 109 TKA (Zimmer® Natural Knee II ultra-congruent mobile-bearing) implanted between 2004 and 2010 for primary osteoarthritis, comparing 2 groups depending on tibial component fixation method, one cemented (n = 68) and one cementless (n = 41). Clinical (Knee Society Rating System (KSS), Hospital for Special Surgery (HSS) scores, range of motion) and radiodiological outcomes were assessed at last follow-up with a minimal follow-up of 5 years. Results: Mean follow-up were 8.14 [5.31–12.7] and 8.06 [5.22–12.02] years, respectively, in cemented and cementless groups. The tibial component survival rate was 100% [95CI: 91.4–100] in the cementless group and 97.1% [95CI: 89.78–99.42] in the cemented group (2 aseptic loosenings) (p = 0.27). Radiolucent lines were present in 31.7% (n = 13) of the cementless and 44.1% (n = 30) of the cemented baseplates (p = 0.2). The postoperative KSS knee score was higher in the cementless group (99 ± 3 vs. 97 ± 7.5; p = 0.02), but there was no significant difference in KSS function, global KSS and HSS scores. Mean range of flexion was 120 ± 10° in the cementless group and 122.5 ± 15° in the cemented group (p = 0.37). No significant differences were found on the radiographic data or on complications. Conclusion: In this study, the survival rate of the tibial component is not influenced by its fixation method at a mean follow-up of 8 years in osteoarthritis, which confirms the reliability of cementless fixation in primary TKA.