Objective To investigate the reliability of determining rotational positioning of femoral prosthesis using the coronal transverse tibial axis at 90° of knee flexion as the reference in total knee arthroplasty (TKA). Methods Fifty-five patients with knee osteoarthritis undergoing TKA between November 2020 and February 2026 who met the selection criteria were enrolled. There were 14 males and 41 females, aged 54-82 years, with a mean age of 66.4 years. Preoperative knee CT and full-length lower-extremity X-ray films were performed to measure the hip-knee-ankle angle (HKA) and posterior condylar angle (PCA). During surgery, after distal femoral osteotomy, the knee was maintained at 90° of flexion, and the tibial transverse axis (line a) perpendicular to the coronal projection of the tibial mechanical axis was marked on the osteotomy surface. Following tibial osteotomy and gap balancing, the external rotation line of the femoral component (line b) was determined. The final external rotation osteotomy angle was established based on the above references, and four-in-one positioning pins were fixed according to this angle. The line connecting the positioning pins served as the known external rotation reference line (line r). Accordingly, the external rotation angles of Line a and Line b relative to the posterior condylar line of the femur (∠A, ∠B) were calculated separately. The differences among PCA, ∠A, and ∠B were compared, and the correlation between ∠A and ∠B,∠A and HKA were analyzed. In addition, the distribution of ∠A in and around the “safe zone” (between PCA and ∠B) was counted. Results Preoperative HKA was (7.3±3.7)°, and PCA, ∠A, and ∠B were (3.1±1.4)°, (4.6±2.4)°, and (4.7±2.1)°, respectively. Compared with PCA, both ∠A and ∠B demonstrated greater external rotation, with significant differences (P<0.05); no significant difference was found between ∠A and ∠B (P>0.05). Correlation analysis revealed significant correlations between ∠A and ∠B, as well as between ∠A and HKA (r=0.674, P<0.001; r=0.350, P=0.009). In 45% (25/55) of patients, ∠A was between PCA and ∠B, which was within the “safe zone”. Conclusion The transverse tibial axis at 90° of knee flexion during TKA serves as a reliable reference for determining rotational positioning of femoral prosthesis. The severity of preoperative knee varus deformity correlates with the external rotation angle defined by this transverse tibial axis, and this angle shows high consistency with that derived from the gap balancing technique.
