Full-length weight-bearing radiographs and computed tomography show discrepancies in coronal alignment for robotic arm-assisted total knee arthroplasty

Assessment of lower limb coronal plane alignment is crucial in surgical planning of total knee arthroplasty (TKA) and is commonly performed with full-length weight-bearing radiographs (FLWBR). The aim of this study was to determine discrepancies in coronal limb alignment as assessed by conventional FLWBR versus non-weight-bearing computed tomography (CT) modalities in robotic arm-assisted TKA. We retrospectively analyzed 100 consecutive patients with osteoarthritic knees who underwent robotic arm-assisted TKA with preoperative FLWBR and CT assessments of knee alignment. The mechanical axes of the Mako system were established in accordance with the Mako TKA Surgical Guide. The following parameters were compared between the two imaging modalities: (1) mechanical hip–knee–ankle angle (mHKA), (2) medial proximal tibial angle (MPTA), (3) lateral distal femoral angle (LDFA), (4) arithmetic hip–knee–ankle angle (aHKA), (5) joint line obliquity (JLO), and (6) proportion of the Coronal Plane Alignment of the Knee (CPAK) types. Regression analyses were performed to identify potential factors associated with discrepancies in measurements of coronal alignment between these two imaging modalities. There were significant differences in terms of the mHKA between preoperative CT and FLWBR (CT vs. FLWBR, 7.0° vs. 8.5°; p < 0.001), which may be attributed to the extent of the weight-bearing condition. The CT measurements with the Mako system demonstrated a lower (more varus) MPTA than the FLWBR measurements (CT vs. FLWBR, 83.7° vs. 85.2°; p < 0.001). Furthermore, there were also significant differences between the CT and FLWBR measurements in terms of the aHKA (− 4.7° vs. − 3.7°; p = 0.028) and JLO (172.2° vs. 174.2°; p < 0.001). There were significant differences of proportion of CPAK types between the two imaging modalities; however, the most common category was type I in both imaging modalities. Multivariate analyses showed that greater preoperative mHKA and posterior tibial slope were associated with the discrepancy in the MPTA measurements between the two imaging modalities. There were discrepancies in coronal alignment between preoperative FLWBR and non-weight-bearing CT using the Mako system. When compared with conventional FLWBR, the Mako system tended to overestimate the magnitude of tibial varus deformity in the knee joint.