Robotic-assisted total knee arthroplasty does not increase procedure duration or adverse event incidence: a retrospective comparative cohort study in a secondary public hospital.

backgroundTotal knee arthroplasty (TKA) with various guidance methods for bone cuts and soft-tissue balancing yields excellent outcomes for end-stage knee osteoarthritis, offering durable pain relief and functional restoration. However, there is a growing interest in further enhancing intraoperative accuracy and consistency, with the aim of reducing complications and reducing recovery times through the introduction of robotic-assisted surgery (RAS). As with any new technology, concerns remain regarding the surgeon learning curve, operative efficiency, and safety profile of RAS compared to conventional techniques. This study evaluates the clinical utility of a RAS system (ROSA, Zimmer Biomet) in a public hospital setting, assessing its impact on procedure duration and adverse event incidence at 90 days postoperatively.

methodsA retrospective comparative cohort study was conducted at a secondary public hospital (Queen Elizabeth II Jubilee Hospital) in, Brisbane, Australia. Data was extracted from a departmental registry, electronic medical records, and intraoperative reports from September 2017 to February 2023. The study included 568 TKA cases: 173 instrumented or navigated TKAs performed before RAS introduction (Pre-RAS), 258 robotic-assisted TKAs after RAS adoption (RAS group), and 137 TKAs performed by other department surgeons who did not use RAS (non-RAS), serving as a benchmark for department-wide outcomes. The primary outcomes were procedure duration and adverse event incidence including surgical site infection (SSI), venous thromboembolism (VTE), knee stiffness, and all-cause readmission within 90 days postoperatively.

resultsAmong 67 recorded adverse events, no significant differences in total adverse event incidence were observed between the pre-RAS and RAS groups (12.1% vs. 11.6%, p = 0.78). A non-significant increase in superficial infections in males undergoing RAS-TKA was observed (p = 0.062). The pre-RAS group had significantly longer operative times than the RAS group (128 ± 21.6 min vs. 121.4 ± 19.5 min, p < 0.01). The non-RAS department group had shorter procedure durations than the pre-RAS group (118.3 ± 20.1 min, p < 0.01).

conclusionRAS-TKA was associated with shorter operative times and a comparable safety profile to non-robotic TKA in this cohort. Further investigation is warranted to determine whether these findings are maintained across different settings and over longer follow-up periods.