Introduction: Currently there are no accepted international guidelines for the correct placement of reconstruction prostheses in the axial plane of the femur after en bloc resection. The most accepted method is based on the linea aspera as an intraoperative landmark, indicating posterior. This study was conducted to address the reliability of the linea aspera as a landmark for rotational alignment. Material and methods: 50 CT angiographies of the right limb were used for this purpose. These 2D images were reconstructed into 3D models using proprietary software (materialize NV, Leuven, Belgium). The posterior condylar line was used as a reference axis. The orientation of the linea aspera was described as the angle between the perpendicular line to the PCL, through the center of the diaphysis, and the lateral (a) and medial labium (beta). Results: The linear mixed model shows that the alpha- and beta-angles are significantly associated with the distance from the joint line (p<0.001) and vary significantly between subjects (p<0.001). The alpha-angle has the lowest variance and approximates more closely true posterior, while the median beta-angle never overlaps true posterior. Discussion: When a surgeon would blindly rely on the linea aspera as a posterior landmark roughly 78% of the femoral implants would exceed the accepted +/- 3 degrees deviation around the surgical transepicondylar axis (sTEA) as defined in total knee replacement. The linea aspera is not a reliable landmark for axial rotation of femoral implants. The position is highly dependent on the osteotomy height and in addition differs between individual patients. Preoperative assessment of the linea aspera is advocated in order to reduce the risk of malrotation. As the height of the osteotomy cannot always be determined correctly preoperatively, a table was designed as a guideline for how much a deviation from the planned resection height will affect the rotation of the implant. (c) 2020 IPEM. Published by Elsevier Ltd. All rights reserved.