The importance of bony impingement in restricting flexion after total knee arthroplasty: computer simulation model with clinical correlation

We constructed patient-specific models from computed tomography data after total knee arthroplasty to predict knee flexion based on implant-bone impingement. The maximum flexion before impingement between the femur and the tibial insert was computed using a musculoskeletal modeling program (KneeSIM; LifeModeler, Inc, San Clemente, California) during a weight-bearing deep knee bend. Postoperative flexion was measured in a clinical cohort of 21 knees (low-flex group: 6 knees with <100° of flexion and high-flex group: 15 size-matched knees with >125° of flexion at 2 years). Average predicted flexion angles were within 2° of clinical measurements for the high-flex group. In the low-flex group, 4 cases had impingement involving the bone cut at the posterior condyle, and the average predicted knee flexion was 102° compared with 93° measured clinically. These results indicate that the level of the distal femoral resection should be carefully planned and that exposed bone proximal to the tips of the posterior condyles of the femoral component should be removed if there is risk of impingement.