BACKGROUND The purpose of this study was to determine the effects of different cementing techniques on the rate of early femoral loosening of high-flexion total knee arthroplasties (TKAs). METHODS A total of 734 knees from 486 patients treated with high-flexion design TKA between July 2001 and July 2010 were divided into two groups based on the cementing technique used. For 403 knees (group N), cement was applied onto the distal and anterior cut surfaces of the femur and the posterior flanges of the femoral component without pressurization. For 331 knees (group P), cement was applied onto distal and anterior femoral cut surfaces with digital pressurization and whole cement surfaces of the femoral component. Two groups were subjected to clinical and radiological evaluation with a minimum five year follow-up period. Cox proportional hazards model with revision surgery of the prosthesis or radiological loosening as an endpoint was used to evaluate the effect of the cementing technique and other covariates. RESULTS The pressurizing and bi-surface cementing technique resulted in significant reduction in femoral radiological loosening incidence compared to that without pressurization (0.3% vs. 2.5%, P=0.015) and revision rate for aseptic causes (0.9% vs. 3.2%, P=0.032). Cox proportional hazard regression analysis revealed a significant difference in component survival rate between the two groups if femoral radiological loosening was considered as failure (hazard ratio, 4.229, 95% confidence interval (CI): 1.256-14.243, P=0.020). CONCLUSIONS Pressurizing and bi-surface cementation can reduce the occurrence rate of early loosening around the femoral component in high-flexion TKAs.