Cementless versus cemented fixation in robotic-assisted total knee arthroplasty: a systematic review and meta-analysis.

Robotic assistance can improve the reproducibility of bone preparation in total knee arthroplasty (TKA), but whether this translates into different outcomes for cementless and cemented fixation is uncertain. PubMed, Embase.com, Web of Science Core Collection and CENTRAL were searched from inception to 29 July 2026 for comparative studies of cementless versus cemented primary robotic-assisted TKA. The review was prospectively registered in PROSPERO (CRD420261446055). Random-effects models used restricted maximum likelihood estimation and Hartung-Knapp adjustment. Risk of bias and certainty were assessed using ROBINS-I and GRADE, respectively. Ten studies represented by 11 reports were included. In three-study syntheses, the pooled risk ratio (RR) was 0.63 for mid-term all-cause revision (95% confidence interval [CI] 0.09 to 4.53; I²=0%) and 1.03 for manipulation under anaesthesia (95% CI 0.09 to 11.74; I²=31.6%). The pooled estimate suggested a potential reduction in operative time in selected settings (mean difference [MD] -12.91 min, 95% CI -36.38 to 10.55; I²=94.4%), but it was imprecise and highly heterogeneous. The length-of-stay estimate was imprecise (MD -0.24 days, 95% CI -3.67 to 3.19; I²=0%). Adjusted registry and institutional estimates were directionally mixed and were kept separate from crude risks. Certainty was very low for every pooled outcome. Available observational evidence is insufficient to favour either fixation strategy for survival or stiffness outcomes after robotic-assisted TKA. Any potential reduction in operative time appears setting dependent and remains uncertain. PROSPERO registration: CRD420261446055.