Improved accuracy of implant placement with an imageless handheld robotic system compared to conventional instrumentation in patients undergoing total knee arthroplasty: a prospective randomized controlled trial using CT-based assessment of radiological outcomes

PURPOSE: Image-free handheld robotic-assisted total knee arthroplasty (RATKA) has shown to achieve desired limb alignment compared to conventional jig-based instrumented total knee arthroplasty (CTKA). The aim of this prospective randomized controlled trial (RCT) was to evaluate the accuracy of a semi-autonomous imageless handheld RATKA compared to CTKA in order to achieve the perioperative planned target alignment of the knee postoperatively. METHODS: Fifty-two patients with knee osteoarthritis were randomized in 1:1 ratio to undergo unilateral CTKA or an imageless handheld RATKA. A full-length lower limb CT-scan was obtained pre- and 6-week postoperative. The primary outcomes were radiologic measurements of achieved target hip-knee-ankle axis (HKA-axis) and implant component position including varus and external rotation and flexion of the femur component, and posterior tibial slope. The proportion of outliers in above radiographic outcomes, defined as?>?3° deviation in postoperative CT measurements as compared to perioperative planned target, were also noted. Knee phenotypes were compared with use of the Coronal Plane Alignment of the Knee (CPAK) classification. RESULTS: Baseline conditions were comparable between both groups. The overall proportion and percentage of outliers (n?=?38, 24.4% vs n?=?9, 5.8%) was statistically significant (p?<?0.001) in favor of RATKA. The achieved varus-valgus of the femoral component (varus 1.3°?±?1.7° vs valgus -?0.1°?±?1.9°, p?<?0.05) with statistically significant less outliers (0% vs 88.5%, p?<?0.01), the achieved HKA-axis (varus 0.4°?±?2.1° vs valgus -?1.2°?±?2.1°, p?<?0.05) and the posterior tibial slope (1.4°?±?1.1° vs 3.2°?±?1.8°, p?<?0.05) were more accurate with RATKA. The most common postoperative CPAK categories were type II (50% CTKA vs 61.5% RATKA), type I (3.8% CTKA vs 23.1% RATKA) and type V (26.9% CTKA vs 15.4% RATKA). CPAK classification III was only found in CTKA (19.2%). Type VI, VII, VIII, and IX were rare in both populations. CONCLUSIONS: The present trial demonstrates that an imageless handheld RATKA system can be used to accurately perioperatively plan the desired individual component implant positions with less alignment outliers whilst aiming for a constitutional alignment. LEVEL OF EVIDENCE: I.