Text

Objectives and questions: Navigation-assisted and robotic-assisted total knee arthroplasty (TKA) are increasingly used to improve implant positioning and potentially reduce early failure. Whether these technologies reduce early revision, particularly in cementless TKA, remains uncertain. This study evaluated early revision risk after primary TKA by surgical technique and assessed whether robotic assistance mitigates early failure within a cementless-only cohort.

Material and methods: We performed a retrospective cohort study of 5,512 primary TKAs recorded in an institutional arthroplasty database. Surgical technique was classified as conventional (n=4,116), navigation-assisted (n=1,129), or robotic-assisted (n=267). Fixation was categorized as cemented (n=3,863) or cementless (n=1,649), and implant design as cruciate-retaining (CR) or posterior-stabilized (PS). The primary endpoint was early revision within 1 year (time-to-revision ≤1.0 years). Due to rare events, multivariable penalized logistic regression (Firth method) estimated adjusted odds ratios (OR) with 95% confidence intervals, adjusting for age, sex, body mass index (BMI), surgery year, fixation, and design. A pre-specified cementless-only subanalysis (n=1,649) compared early revision between techniques within cementless fixation using the same modeling strategy.

Results: Overall, 50 early revisions (0.91%) occurred within 1 year. Early revision rates were 1.04% for conventional cases (43/4,116), 0.27% for navigation-assisted cases (3/1,129), and 1.50% for robotic-assisted cases (4/267). In the adjusted Firth model (n=5,343), navigation-assisted TKA demonstrated a trend toward reduced early revision compared with conventional technique (OR=0.36; 95% CI 0.12–1.14; p=0.083), while robotic-assisted TKA was not associated with a significant difference (OR=1.35; 95% CI 0.47–3.90; p=0.578). Cementless fixation was not associated with early revision (OR=0.79; 95% CI 0.38–1.63; p=0.521). In the cementless-only cohort (n=1,649), 16 early revisions (0.97%) occurred within 1 year. Early revision rates were 1.12% for cementless conventional cases (12/1,069), 0.28% for cementless navigation-assisted cases (1/362), and 1.38% for cementless robotic-assisted cases (3/218). In adjusted cementless-only Firth regression (n=1,579), navigation (OR=0.42; 95% CI 0.07–2.41; p=0.329) and robotics (OR=1.99; 95% CI 0.51–7.74; p=0.320) were not significantly associated with early revision.

Discussion and conclusions: Early revision within 1 year after primary TKA was rare (<1%). Navigation-assisted TKA demonstrated a consistent trend toward reduced early revision overall and within cementless fixation, whereas robotic-assisted TKA was not associated with reduced early failure in this cohort. Larger cohorts with longer follow-up are needed to more precisely define the impact of robotics on early cementless TKA outcomes.