Real-time intraoperative motion-following robotic assistance improves efficiency and accuracy in total knee arthroplasty: a retrospective comparative study.

Background: Conventional robotic-assisted total knee arthroplasty (RA-TKA) relies on rigid limb fixation to suppress intra-operative motion, adding complexity and potential inefficiency. A novel motion-following control system dynamically compensates for limb movement, allowing real-time adjustment of the tool-bone relationship without immobilization. This study evaluated whether motion-following improves efficiency and osteotomy accuracy while preserving alignment and early function.

Methods: Sixty consecutive primary RA-TKA cases performed with the SkyWalker robotic platform (MicroPort, Shanghai, China) between September 2022 and August 2024 were retrospectively reviewed. Thirty procedures used conventional rigid fixation (control group) and thirty employed motion-following tracking (motion-Following group). Primary endpoints were operative time and resection thickness error, measured intraoperatively with a caliper. Secondary outcomes included coronal alignment assessed by HKA (hip-knee-ankle angle), CFCA (coronal femoral component angle), and CTCA (coronal tibial component angle), as well as functional recovery assessed by WOMAC (Western Ontario and McMaster Universities Osteoarthritis Index) at 6 months. Values are expressed as mean ± standard deviation unless otherwise specified.

Results: Mean operative time was shorter with motion-following (118.8 ± 9.3 min) than with conventional fixation (133.9 ± 11.9 min; p < 0.001). Mean resection-thickness error was lower with motion-following (0.53 mm vs 0.82 mm), with 93.9% versus 68.3% of cuts within ≤ 1 mm. At the plane level, motion-following achieved smaller errors on all six surfaces, with four planes: DF-M (distal femur medial), distal femur lateral (DF-L), posterior femur medial (PF-M), and tibial plateau lateral (TP-L) reaching statistical significance (p < 0.05). Post-operative coronal alignment closely reproduced the pre-operative plan in both groups, with mean deviations of approximately 1° across all parameters and no statistically significant between-group differences. WOMAC scores improved substantially in both groups, with no significant between-group difference (ΔWOMAC 32.8 ± 8.5 vs 30.1 ± 7.9; p = 0.21).

Conclusions: Motion-following robotic control streamlines TKA by eliminating rigid fixation, improving workflow efficiency, and slightly enhancing osteotomy precision without compromising alignment or recovery. This dynamic, real-time tracking approach refines execution of the surgical plan and may represent a meaningful evolution toward more efficient, surgeon-friendly robotic arthroplasty.