Robotic-assisted simultaneous bilateral total knee arthroplasty was associated with a more favorable perioperative profile than a historical conventional cohort.

PURPOSE: To compare perioperative, clinical, and functional outcomes between robotic-assisted and conventional simultaneous bilateral total knee arthroplasty in a historical comparative cohort. METHODS: We conducted a retrospective historical comparative cohort study of patients undergoing simultaneous bilateral total knee arthroplasty at a single high-volume arthroplasty centre between March 2013 and July 2025. Patients were stratified into a historical conventional cohort or a computed tomography-based robotic-arm-assisted cohort. Outcomes included post-operative haematocrit, blood transfusion, tourniquet ischaemia time, length of stay, post-operative complications, and Forgotten Joint Score-12 values. RESULTS: Sixty-two patients were included: 41 in the robotic-assisted group and 21 in the conventional group. Baseline preoperative haematocrit was higher in the robotic cohort. Post-operative haematocrit was higher after robotic-assisted surgery (31.74 ± 4.81% vs. 26.68 ± 4.41%, p < 0.001), whereas haematocrit drop did not differ significantly (p = 0.295). The robotic-assisted group had a lower transfusion requirement (14.6% vs. 76.2%, p < 0.001). After multivariable adjustment, conventional surgery remained associated with increased odds of post-operative transfusion (adjusted odds ratio 20.78, 95% confidence interval 3.19-135.24). Robotic assistance was also associated with shorter tourniquet ischaemia time and hospital stay. Post-operative complications occurred only in the conventional cohort, although the difference was not statistically significant. Forgotten Joint Score-12 values were available for 34 of 62 patients (54.8%) at a median follow-up of 47.9 months and did not differ significantly between groups (81.06 ± 25.76 vs. 74.83 ± 23.86, p = 0.486). CONCLUSIONS: Robotic-assisted simultaneous bilateral total knee arthroplasty was associated with a more favourable perioperative profile than a historical conventional cohort, particularly regarding transfusion requirement, post-operative haematocrit, tourniquet ischaemia time, and hospital stay. These findings should be interpreted cautiously because of temporal confounding, incomplete functional follow-up, and limited sample size. LEVEL OF EVIDENCE: Level III.