The primary purpose of this study was to determine the preoperative predictors of gait biomechanics 6 months after unilateral total knee arthroplasty (TKA). There were 126 participants (age 64.4 ± 7.1 years, 75 females) who underwent instrumented biomechanical assessments while walking at a self-selected pace preoperatively, 10 weeks after (post-rehabilitation), and 6 months after unilateral TKA. Outcomes were peak knee extension moment (pKEM), knee angle excursion, and vertical ground reaction force (vGRF) ratio (surgical/contralateral). Potential clinical, demographic, and biomechanical predictors were tested univariately and considered a candidate for the final model if p < 0.15. Each multivariate model initially contained all candidates, and backward selection was used to determine the final model. Greater 6-month surgical limb pKEM was predicted (r² = 0.31) by greater preoperative pKEM (β = 0.44, p < 0.0001), better quadriceps activation (β = 0.23, p = 0.004), and male sex (β = −0.21, p = 0.009). Greater 6-month surgical knee excursion was predicted (r² = 0.34) by greater preoperative excursion (β = 0.39, p < 0.0001), male sex (β = −0.28, p = 0.0007), and preoperative quadriceps strength (β = 0.16, p = 0.047). Six-month vGRF ratio was predicted (r² = 0.16) by preoperative vGRF ratio (β = 0.37, p < 0.0001) and study treatment group (β = 0.18, p = 0.03). Preoperative biomechanical variables at post-rehabilitation were also the strongest predictors of 6-month biomechanics.

Statement of Clinical Significance:​ The strongest and most consistent predictor of gait biomechanics 6 months after TKA was the respective preoperative gait biomechanics variable, which may have important clinical implications for surgical decision making and prehabilitation/rehabilitation strategies. Biofeedback targeting vGRF predicted vGRF symmetry, but no other gait parameters, suggesting targeted interventions are needed. Improving quadriceps strength and activation may also facilitate knee biomechanics.