Patient-Specific Simulated Dynamics After Total Knee Arthroplasty Correlate With Patient-Reported Outcomes.

BACKGROUND: Component alignment variation following total knee arthroplasty (TKA) does not fully explain the instance of long-term postoperative pain. Joint dynamics following TKA vary with component alignment and patient-specific musculoskeletal anatomy. Computational simulations allow joint dynamics outcomes to be studied across populations. This study aims to determine if simulated postoperative TKA joint dynamics correlate with patient-reported outcomes. METHODS: Landmarking and 3D registration of implants was performed on 96 segmented postoperative computed tomography scans of TKAs. A cadaver rig-validated platform for generating patient-specific simulation of deep knee bend kinematics was run for each patient. Resultant dynamic outcomes were correlated with a 12-month postoperative Knee Injury and Osteoarthritis Outcome Score (KOOS). A Classification and Regression Tree (CART) was used for determining nonlinear relationships. RESULTS: Nonlinear relationships between the KOOS pain score and rollback and dynamic coronal alignment were found to be significant. Combining a dynamic coronal angular change from extension to full flexion between 0° and 4° varus (long leg axis) and measured rollback of no more than 6 mm without rollforward formed a "kinematic safe zone" of outcomes in which the postoperative KOOS score is 10.5 points higher (P = .013). CONCLUSION: The study showed statistically significant correlations between kinematic factors in a simulation of postoperative TKA and postoperative KOOS scores. The presence of a dynamic safe zone in the data suggests a potential optimal target for any given individual patient's joint dynamics and the opportunity to preoperatively determine a patient-specific alignment target to achieve those joint dynamics.

Patient Variation Limits Use of Fixed References for Femoral Rotation Component Alignment in Total Knee Arthroplasty.

BACKGROUND: Optimal rotational alignment of the femoral component is a common goal during total knee arthroplasty. The posterior condylar axis (PCA) is thought to be the most reproducible reference in surgery, while the transepicondylar axis (TEA) seems to better approximate the native kinematic flexion axis. This study sought to determine if rules based on patient gender or coronal alignment could allow reliable reproduction of the TEA from the PCA. METHODS: Three-dimensional models based on preoperative computed tomography were made representing a patient's arthritic knee joint. The landmarks were defined and angular relationships determined. RESULTS: The population group of 726 patients contained large anatomic variation. When applying the standard reference rule of 3° external rotation from the PCA, 36.9% of patients would have a rotational target greater than ±2° from their TEA. When applying the mean external rotation of the TEA from the PCA (1.85°) from this population, this proportion dropped to 26.0% of patients. The use of statistically significant gender and coronal alignment relationships to define the femoral rotation did not reduce the proportion of patients in ±2° error. CONCLUSION: This study shows that gender and coronal alignment relationships to the TEA to PCA angle are not clinically significant as a quarter of patients would still have a target for rotation greater than ±2° from the TEA using these relationships. Superior tools for orienting rotational cuts directly to the TEA in surgery or preoperative identification of relevant patient-specific angles might capture the proportion of patients for whom standard reference angles are not appropriate.