Computed Tomography Analysis of Postsurgery FemoralComponent Rotation Based on a Force Sensing DeviceMethod versus Hypothetical Rotational Alignment Based onAnatomical Landmark Methods: A Pilot Study

Rotation of the femoral component is an important aspect of knee arthroplasty, due to its effects on postsurgery knee kinematics andassociated functional outcomes. It is still debated which method for establishing rotational alignment is preferable in orthopedicsurgery. We compared force sensing based femoral component rotation with traditional anatomic landmark methods to investigatewhich method is more accurate in terms of alignment to the true transepicondylar axis. Thirty-one patients underwent computer-navigated total knee arthroplasty for osteoarthritis with femoral rotation established via a force sensor. During surgery, threealternative hypothetical femoral rotational alignments were assessed, based on transepicondylar axis, anterior-posterior axis, or theutilization of a posterior condyles referencing jig. Postoperative computed tomography scans were obtained to investigate rotationcharacteristics. Significant differences in rotation characteristics were found between rotation according to DKB and other methods(𝑃<0.05). Soft tissue balancing resulted in smaller deviation from anatomical epicondylar axis than any other method. 77% ofoperated knees were within a range of±3∘of rotation. Only between 48% and 52% of knees would have been rotated appropriatelyusing the other methods. The current results indicate that force sensors may be valuable for establishing correct femoral rotation