Purpose
Revision constrained-condylar total knee arthroplasty (CCK-TKA) is often used to provide additional mechanical constraint after failure of a primary TKA. However, it is unknown how much this translates to a reliance on soft-tissue support. The aim of this study was therefore to compare the laxity of a native knee to the CCK-TKA implanted state and quantify how medial soft-tissues stabilise the knee following CCK-TKA.

Methods
Ten intact cadaveric knees were tested in a robotic system at 0°, 30°, 60° and 90° flexion with ± 90 N anterior–posterior force, ± 8 Nm varus-valgus and ± 5 Nm internal–external torques. A fixed-bearing CCK-TKA was implanted and the laxity tests were repeated with the soft tissues intact and after sequential cutting. The deep and superficial medial collateral ligaments (dMCL, sMCL) and posteromedial capsule (PMC) were sequentially transected and the percentage contributions of each structure to restraining the applied loads were calculated.

Results
Implanting a CCK-TKA did not alter anterior–posterior laxity from that of the original native knee, but it significantly decreased internal–external and varus-valgus rotational laxity (p < 0.05). Post CCK-TKA, the sMCL restrained 34% of the tibial displacing load in anterior drawer, 16% in internal rotation, 17% in external rotation and 53% in valgus, across the flexion angles tested. The dMCL restrained 11% of the valgus rotation moment.

Conclusions
With a fully-competent sMCL in-vitro, a fixed-bearing CCK-TKA knee provided more rotational constraint than the native knee. The robotic test data showed that both the soft-tissues and the semi-constrained implant restrained rotational knee laxity. Therefore, in clinical practice, a fixed-bearing CCK-TKA knee could be indicated for use in a knee with lax, less-competent medial soft tissues.