Cemented femoral stems have shown decreased longevity compared to cementless implants in hip revision arthroplasty. The aim of this study was to evaluate the effect of an amphiphilic bonder on bone cement stability in a biomechanical femur expulsion test. A simplified hip simulator test setup with idealised femur stem specimens was carried out. The stems were implanted into bovine femurs (group 1: no bonder, n=10; group 2: bonder including glutaraldehyde, n=10; group 3: bonder without glutaraldehyde, n=10). A dynamic loading (maximum load: 800 N; minimum load: 100 N; frequency: 3 Hz; 105 cycles) was performed. Subsequently, the stem specimens were expulsed axially out of their implant beds and maximum load at failure was recorded. The static controls showed a mean maximum load to failure of 4123 N in group 1, 8357.5 N in group 2 and 5830.8 N in group 3. After dynamic loading, the specimens of group 2 reached the highest load to failure (8191.5 N), followed by group 3 (5649.5 N) and group 1 (3462 N), respectively. In group 2, we observed nine periprosthetic fractures at a load of 8400 N without signs of interface loosening. Application of an amphiphilic bonder led to a significant improvement of bonding stability, especially when glutaraldehyde was added to the bonder. This technique might offer an increased longevity of cemented femur revision stems in total hip replacement.