LAUSR

To investigate factors contributing to the de novo hyperextension after posterior cruciate ligament substituting (PS) total knee arthroplasty (TKA). Through a retrospective case–control study, de novo hyperextension patients were compared with patients without hyperextension after primary PS TKA. Eighty-five hyperextension patients were compared with 85 patients in a control group matched by age, sex, surgeon and implant. The clinical and radiographic parameters, including the mechanical axis (MA), joint line convergence angle (JLCA), posterior tibial slope angle (PTSA), posterior condylar offset (PCO), and the gamma angle, were evaluated preoperatively and immediate postoperatively. Comparisons between the two groups and logistic regression analysis were performed to investigate factors contributing to de novo hyperextension. Among the clinical factors, preoperative flexion contracture was less (5° ± 6° vs. 11° ± 6°, p < 0.001) and the range of motion was greater (125° ± 12° vs. 118° ± 5°, p = 0.041) in the de novo hyperextension group than in the control group. Among the radiographic parameters, preoperative and postoperative JLCA were greater (8.1° ± 4.4° vs. 6.1° ± 3.5°, p < 0.001, 1.0° ± 1.3° vs. 0.2° ± 0.8°, p = 0.002, respectively), postoperative PTSA was greater (3.7° ± 2.0° vs. 3.3° ± 1.6°, p < 0.001) and preoperative and postoperative PCO were less in the hyperextension group than in the control group (26.3 mm ± 3.3 mm vs. 29.1 mm ± 3.2 mm, p < 0.001, 26.4 mm ± 3.2 mm vs. 29.1 mm ± 3.0 mm, p < 0.001, respectively), whereas, change in PCO was greater in the hyperextension group (1.1 mm ± 7.7 mm vs. − 0.1 mm ± 3.3 mm, p < 0.001). In multivariate analysis, the degree of medial soft tissue release [odds ratio (OR) 2.83, p = 0.003], flexion contracture [OR 0.86, p = 0.028], postoperative JLCA [OR 2.45, p = 0.004], preoperative PCO and a change in PCO [OR 0.74, p = 0.002, OR 1.89, p = 0.001, respectively] were the factors associated with de novo hyperextension. An increased degree of medial soft tissue release, small preoperative flexion contracture, increased postoperative JLCA, decreased preoperative PCO and increased change in PCO were risk factors of de novo hyperextension. III.