The influence of femoral rollback on patellofemoral contact loads in total knee arthroplasty.

Increasing femoral rollback in flexion is thought to reduce patellofemoral contact load in total knee arthroplasty (TKA). The objectives of this study were to quantify the dependence of patellar load on rollback and to assess the effectiveness of posterior cruciate ligament (PCL)-retaining, PCL-sacrificing, and PCL-substituting TKA types in generating rollback. Nine cadaver knees were tested in simulated squatting. Six TKAs that were expected to produce varying amounts of femoral rollback were evaluated: PCL-retaining TKA, PCL-sacrificing TKA, a commercially available PCL-substituting TKA, and 3 modified PCL-substituting TKAs in which the anteroposterior position of the tibial post was varied. Kinematics, quadriceps loads, and patellofemoral contact loads were recorded. Significant differences in rollback were observed in the 30 degrees to 90 degrees flexion range. PCL-sacrificing TKAs generated the least rollback. PCL-retaining TKAs produced greater rollback but had the most variability. PCL-substituting TKAs produced the greatest and most reproducible rollback. Moving the tibial post posteriorly further increased rollback. Increased rollback correlated with reduced patellar load (-2.2%/mm). Reductions in patellar load of 17.6% were observed. Quadriceps loads were reduced by increasing rollback but to a smaller degree (-0.9%/mm). Rollback primarily affects patellar load rather than quadriceps load or efficiency.