The effect of bone cement viscosity and cement mantle thickness on the migration of the Exeter total hip prosthesis was studied in a prospective, randomized, double-blind clinical Roentgen Stereophotogrammetric Analysis study. Forty-one cemented total hip arthroplasty in 39 patients were included and randomized into a low/medium Simplex P cement group and a high-viscosity Simplex AF cement group. At time of stem introduction, 5 minutes after mixing, the Simplex AF was more viscous than Simplex P. No statistical difference existed between the 2 cement groups, for neither translation nor rotation migration data. Subsidence of the stem at 2-year follow-up was 1.1 +/- 0.56 mm for Simplex AF cement and 1.5 +/- 1.00 mm for Simplex P cement. The mean rotation of the acetabular components about the sagittal axis was 1.7 degrees +/- 3.8 degrees in the Simplex AF group and 0.7 degrees +/- 2.1 degrees for the Simplex P group. No effect of cement mantle thickness on migration of neither the acetabular cups nor the femoral stems was found. Although there were no differences in migration data for the cups and the stems, 2 acetabular cups in the Simplex AF group (almost 10%) were revised because of mechanical loosening. Because of these findings, we suggest caution before using this new high-viscosity bone cement for fixation of acetabular components.