Haptic robotic-arm-assisted total knee arthroplasty (RATKA) seeks to leverage three-dimensional planning, intraoperative assessment of ligament laxity, and guided bone preparation to establish and achieve patient-specific targets for implant position. We sought to compare (1) operative details, (2) knee alignment, (3) recovery of knee function, and (4) complications during adoption of this technique to our experience with manual TKA. We compared 120 RATKAs performed between December 2016 and July 2018 to 120 consecutive manual TKAs performed between May 2015 and January 2017. Operative details, lengths of stay (LOS), and discharge dispositions were collected. Tibiofemoral angles, Knee Society Scores (KSS), and ranges of motion were assessed until 3 months postoperatively. Manipulations under anesthesia, complications, and reoperations were tabulated. Mean operative times were 22 minutes longer in RATKA (p < 0.001) for this early cohort, but decreased by 27 minutes (p < 0.001) from the first 25 RATKA cases to the last 25 RATKA cases. Less articular constraint was used to achieve stability in RATKA (93 vs. 55% cruciate-retaining, p < 0.001; 3 vs. 35% posterior stabilized (PS), p < 0.001; and 4 vs. 10% varus-valgus constrained, p_ = _0.127). RATKA had lower LOS (2.7 vs. 3.4 days, p < 0.001). Discharge dispositions, tibiofemoral angles, KSS, and knee flexion angles did not differ, but manipulations were less common in RATKAs (4 vs. 17%, p = 0.013). We observed less use of constraint, shorter LOS, and fewer manipulations under anesthesia in RATKA, with no increase in complications. Operative times were longer, particularly early in the learning curve, but improved with experience. All measured patient-centered outcomes were equivalent or favored the newer technique, suggesting that RATKA with patient-specific alignment targets does not compromise initial quality. Observed differences may relate to improved ligament balance or diminished need for ligament release.
Thieme. All rights reserved.