Small deviations between planned and performed bone cuts using a CT-based robotic-arm-assisted total knee arthroplasty system

Purpose: Computed tomography (CT)-based robotic system for total knee arthroplasty (TKA) has shown improved accuracy compared to conventional. This study was designed to (1) confirm the accuracy of the robotic system in achieving the plan and (2) establish the alignment and positioning deviation between final components and planning, by measuring the discrepancy between final implant alignment and the corresponding planned cut. Methods: Ninety-six cementless robotic-arm assisted (RA) TKAs were assessed. Bone resections were performed using the haptically controlled robotic arm. Alignment in the coronal and sagittal plane and resection depth of the distal femoral and proximal tibial cuts were recorded with a navigation planar probe. After final components were impacted, the probe was positioned on each implant surface to determine its alignment and positioning. Results: The mean tibial resections and implanted tibial component's positioning were 0.4 mm (standard deviation, SD: 0.6) and 0.9 mm (SD: 0.8), respectively, higher than planned (p < 0.01). The tibial sagittal cut had 19/96 cases (19.8%) of ±1° outliers from plan. In 40/96 cases (41.7%), the tibial component was more prominent than planned of more than 1 mm. The mean femoral resections and impacted femoral component's positioning was 0.1 mm (SD: 0.8) and 0.2 mm (SD: 0.7), respectively, higher than planned. In 23/96 cases (24.0%), the femoral sagittal cut and femoral component coronal alignment deviated more than ±1° from plan. Conclusions: The computed tomography-based robotic-assisted TKA system showed good accuracy regarding bone preparation and component's positioning relative to the planning. Cementless tibial component impaction resulted in the most deviation from plan, with a large proportion of cases resulting in being more prominent than planned. Level of Evidence: Level III.