BACKGROUND: Robotic-assisted total knee arthroplasty (RATKA) has been shown to improve the accuracy of component alignment and to potentially reduce the need for soft-tissue releases. To date, however, the potential benefits of expedited recovery, improved functional outcomes, and improved longevity of RATKA remain unproven. The purpose of this study was to compare functional outcomes and rates of soft-tissue releases between RATKA, conventional instrumentation (CONV), and accelerometer-based navigation (ABN) in primary TKAs. METHODS: A retrospective study of 2,338 consecutive TKAs performed by two surgeons was performed. There were 1,216 TKAs performed with CONV and the goal of neutral mechanical alignment. There were 724 TKAs performed with ABN and restricted kinematic alignment goals. There were 398 RATKAs performed with a functional alignment philosophy. Radiographs were reviewed for all knees. We compared Patient-Reported Outcomes Measurement Information System (PROMIS) scores (two weeks, six weeks, and one year) and rates of soft-tissue releases between ABN and RATKA cohorts. Chi-square tests were used to compare rates of releases between cohorts. Generalized estimating equations were used to evaluate outcomes over time. RESULTS: Overall, the rate of soft-tissue releases was 47.9% in CONV, 74.4% in the ABN group, and 29.9% in the RATKA group. The RATKAs required significantly fewer medial releases in varus knees than CONV (19.9 versus 46.3%, P <
0.002) and ABN TKAs (19.9 versus 68.8%, P <
0.002). In valgus knees, RATKAs required significantly fewer lateral releases than CONV (33.7 versus 61.6%, P <
0.002) and ABN TKAs (33.7 versus 46.6%, P <
0.002). The RATKA had a greater reduction in pain scores (mean 4.5 more points) than the ABN cohort from preoperative to six weeks (P = 0.038) with no difference in pain scores beyond six weeks. There were no differences in PROMIS scores at 1-year follow-up. CONCLUSION: In this series, RATKA with a functional alignment goal performed by adjusting component placement and bony cuts to balance the knee resulted in fewer soft-tissue releases when compared to conventional instrumented TKA and navigated TKA. The RATKAs also demonstrated a larger reduction in short-term pain scores when compared to ABN TKAs, but no differences in patient-reported outcome scores at one year.