Whether the kinematics of cadaveric knees recreate those of the patient's knees after total knee arthroplasty (TKA) remains unknown. This study compared in vivo and in vitro fluoroscopic kinematics of knees after TKA during knee bending using the same implant design. Patients who had undergone cruciate-retaining and cruciate-substituting total knee arthroplasty (CR-TKA and CS-TKA) did squatting motions. Fresh-frozen human cadaveric knees were tested under dynamic closed-chain knee extension in an Oxford knee rig. Fluoroscopic kinematics were measured in vivo and in vitro using a 2-dimensional to 3-dimensional registration technique. The axial rotation angle and anterior-posterior translation of medial and lateral contact points of the femoral implant relative to the tibial implant was evaluated in each flexion angle. No notable differences in the axial rotation angles were found between in vivo and in vitro measurements in both CR-TKA and CS-TKA. In early flexion, the medial and lateral contact points in cadaver knees were located more posterior than those in in vivo after CR-TKA. From early flexion to high flexion, the medial and lateral contact points in cadaver knees were located more posterior than those in in vivo after CS-TKA. The axial rotation angle of cadaveric knees in the Oxford knee rig was similar to that measured in vivo after TKA. However, the anterior-posterior location of the femoral implant in cadaver knee was more posterior than that in in vivo knees after TKA.
Read full abstract