PURPOSE: Coronal plane alignment of the knee (CPAK) classification is widely used in total knee arthroplasty (TKA) planning, but its three-dimensional (3D) relationship with sagittal and axial alignment remains unclear. This study aimed to investigate the association between the CPAK classification and 3D alignment parameters using EOS imaging. METHODS: This single-center retrospective cohort study included 1009 patients (511 non-KOA; 498 KOA). All participants underwent EOS full-body/lower-limb biplanar radiographs with 3D reconstruction. The CPAK classification and sagittal/axial parameters were measured. Generalized linear models (GLMs) were used to compare parameters across CPAK types while controlling for age, sex, and BMI. Effect sizes were reported as partial eta-squared (η²) with 95% confidence intervals obtained via bootstrap resampling. Associations between the arithmetic hip-knee-ankle angle (aHKA), joint line obliquity (JLO), and 3D parameters were assessed using partial Spearman correlation. RESULTS: Most 3D parameters showed no statistically significant differences across CPAK types after adjusting for age, sex, and BMI. In the non-KOA cohort, significant differences persisted for sagittal alignment (p 0.05). The aHKA and JLO were significantly correlated with some 3D parameters, although all correlation coefficients were weak (ρ < 0.30). CONCLUSION: The CPAK classification's weak correlation with most 3D parameters confirms its inherent limitations, accounting for limited sagittal and axial variance. Future phenotyping should integrate kinematic and soft tissue assessments for personalized TKA planning. LEVEL OF EVIDENCE: Level III, retrospective cohort study.
Wang et al. (Tue,) studied this question.