Abstract
BACKGROUND: Extramedullary systems are commonly used in knee arthroplasty, with the rod location being determined from the tibial torsion line during surgery. The traditional method for tibial torsion measurement is not in accordance with clinical practice. This study aimed to evaluate proximal and distal tibial torsion using 3-dimensional (3D) computed technology to establish a new evaluation method, as well as to investigate the association between tibial torsion and postoperative alignment deviation. METHODS: Fifty-five osteoarthritis tibias with >10°varus preoperatively were divided into valgus, neutral, and varus groups based on their postoperative alignment deviation. A new method based on clinical practice was built using a 3D tibial model. Proximal and distal tibial torsions were measured by both the new and traditional methods. In addition, tibial osteotomy that followed the intramedullary osteotomy system was simulated on the 3D model in the varus and valgus groups to investigate the association between tibial torsion and alignment deviation. RESULTS: Proximal tibial torsion was smaller and distal torsion was greater in the valgus group than the other two groups, according to the new method (p = 0.03 and p = 0.02, respectively). No significant difference was found when comparing these torsions by the traditional method (p = 0.782 and p = 0.753, respectively). In the valgus group, the postoperative alignment deviation improved after simulated osteotomy guided by the intramedullary system, while no significant improvement was found in the varus group. CONCLUSION: According to this new tibial-rotation evaluation method, valgus deviation in knee arthroplasty was identified as the main cause for knees in which the proximal tibial internal torsion is too small and the distal external torsion is too great. The use of an intramedullary system may help reduce this deviation. TRIAL REGISTRATION: Prospectively registered.