Background and purpose - Backside wear of the polyethylene insert in total knee arthroplasty (TKA) can produce clinically significant levels of polyethylene debris, which can lead to loosening of the tibial component. Loosening due to polyethylene debris could theoretically be reduced in tibial components of monoblock polyethylene design, as there is no backside wear. We investigated the effect of 2 different tibial component designs, monoblock and modular polyethylene, on migration of the tibial component in uncemented TKA. Patients and methods - In this randomized study, 53 patients (mean age 61 years), 32 in the monoblock group and 33 in the modular group, were followed for 2 years. Radiostereometric analysis (RSA) was done postoperatively after weight bearing and after 3, 6, 12, and 24 months. The primary endpoint of the study was comparison of the tibial component migration (expressed as maximum total point motion (MTPM)) of the 2 different implant designs. Results - We did not find any statistically significant difference in MTPM between the groups at 3 months (p = 0.2) or at 6 months (p = 0.1), but at 12 and 24 months of follow-up there was a significant difference in MTPM of 0.36 mm (p = 0.02) and 0.42 mm (p = 0.02) between groups, with the highest amount of migration (1.0 mm) in the modular group. The difference in continuous migration (MTPM from 12 and 24 months) between the groups was 0.096 mm (p = 0.5), and when comparing MTPM from 3-24 months, the difference between the groups was 0.23 mm (p = 0.07). Interpretation - In both study groups, we found the early migration pattern expected, with a relatively high initial amount of migration from operation to 3 months of follow-up, followed by stabilization of the implant with little migration thereafter. However, the modular implants had a statistically significantly higher degree of migration compared to the monoblock. We believe that the greater stiffness of the modular implants was the main reason for the difference in migration, but an initial creep in the polyethylene metal-back locking mechanism of the modular group could also be a possible explanation for the observed difference in migration between the 2 study groups.