Abstract
Introduction
Robotically-assisted unicondylar knee arthroplasty (UKA) is intended to improve the precision with which the components are implanted, but the impact of alignment using this technique on subsequent polyethylene surface damage has not been determined. Therefore, we examined retrieved ultra-high-molecular-weight polyethylene UKA tibial inserts from patients who had either robotic-assisted UKA or UKA performed using conventional manual techniques and compared differences in polyethylene damage with differences in implant component alignment between the two groups. We aimed to answer the following questions: (1) Does robotic guidance improve UKA component position compared to manually implanted UKA? (2) Is polyethylene damage or edge loading less severe in patients who had robotically aligned UKA components? (3) Is polyethylene damage or edge loading less severe in patients with properly aligned UKA components?
Methods
We collected 13 medial compartment, non-conforming, fixed bearing, polyethylene tibial inserts that had been implanted using a passive robotic-arm system and 21 similarly designed medial inserts that had been manually implanted using a conventional surgical technique. Pre-revision radiographs were used to determine the coronal and sagittal alignment of the tibial components. Retrieval analysis of the tibial articular surfaces included damage mapping and 3D laser scanning to determine the extent of polyethylene damage and whether damage was consistent with edge loading of the surface by the opposing femoral component.
Results
Though the individual planar alignments did not differ between the two groups, overall 69% of the 13 robotically aligned components were well-positioned in both the coronal and sagittal planes, as opposed to only 18% of the manually aligned tibial components (Fig.1). Robotically aligned inserts had significantly less pitting, burnishing, and deformation than manually aligned inserts, and the maximum surface deviations (wear and deformation) were significantly smaller, though these differences could be explained by a longer length of implantation for the manually aligned inserts. Interestingly, no difference was found in the incidence of edge loading between the robotically aligned and manually aligned groups. When comparing polyethylene damage on the basis of alignment rather than surgical technique, neither the polyethylene damage nor surface deviation was different, aside from more burnishing and deformation in mal-positioned components and greater deviation in components mal-positioned in the sagittal plane.
Conclusions
Static radiographic alignment measurements were not useful in predicting the wear patterns that the tibial inserts experienced while implanted, suggesting that other factors, such as the patient's functional kinematics, influence the mechanical burden placed on the polyethylene articular surfaces (Fig. 2).
