Coronal alignment behaviour across the knee range of motion is associated with constitutional alignment and soft‐tissue balance: The REAL HKA classification
Nikolaos Mylonakis, Eustathios Kenanidis, Alexander Maslaris, Peter Keyes Sculco, Eleftherios TsiridisAbstract
Purpose
To evaluate intraoperative, pre‐resection coronal alignment throughout the range of motion (ROM) in knees undergoing robotic‐assisted total knee arthroplasty (ra‐TKA), and to assess its relationship with the Coronal Plane Alignment of the Knee (CPAK) and the Robotic Evaluation of Articular Laxity (REAL) classifications.
Methods
Three hundred fifty‐two consecutive knees undergoing primary ra‐TKA using functional alignment were included in this single‐surgeon, platform and implant retrospective cohort study. Intraoperative coronal laxity was assessed before bone resections at 0°, 45° and 90° of flexion, under surgeon‐applied manual varus and valgus stress without a tensioning device. Laxity was recorded as the maximum angular deviation from the neutral coronal axis and shown as an excursion bar. At each flexion angle, a coronal phenotype was defined: neutral if the bar crossed the neutral axis, varus if entirely within negative hip–knee–ankle (HKA) values and valgus if entirely within positive. By combining the phenotypes at 0°, 45° and 90°, a composite ‘REAL HKA phenotype’ was derived. Associations between phenotype stability and the CPAK and REAL classifications were evaluated.
Results
Fourteen phenotypes were identified. Overall, 73.6% of knees demonstrated a stable phenotype. Phenotype stability varied significantly across CPAK groups ( χ 2 = 60.12, p < 0.001) and was lowest in valgus knees. Within each alignment group, apex direction influenced phenotype stability; Class IV had lower odds than I (odds ratio [OR] 0.28, 95% confidence interval [CI] 0.11–0.72) and Class V had lower odds than II (OR 0.23, 95% CI 0.09–0.60). REAL flexion grouping was also associated with stability ( χ 2 = 9.94, p = 0.007). Multivariable analysis confirmed independent associations of CPAK and REAL flexion grouping with phenotype stability ( χ 2 = 63.26, p < 0.001, Nagelkerke R 2 = 0.24).
Conclusions
Coronal behaviour across the ROM varies in knees undergoing ra‐TKA. Both constitutional alignment and soft‐tissue balance are associated with dynamic coronal phenotype stability.
Level of Evidence
Level IV, retrospective case series.