Abstract
Introduction
Preoperative templating of femoral and tibial components can assist in choosing the appropriate implant size prior to TKA. While weight bearing long limb roentograms have been shown to provide benefit to the surgeon in assessing alignment, disease state, and previous pathology or trauma, their accuracy in size prediction is continually debated due to scaling factors and rotated views. Further, they represent a static time point, accounting for boney anatomy only. A perceived benefit of robotic-assisted surgery is the ability to pre-operatively select component sizes with greater accuracy based on 3D information, however, to allow for flexibility in refining based on additional data only available at the time of surgery.
Methods
The purpose of this study was to determine the difference of pre-operative plans in size prediction of the tibia, femur, and polyethylene insert. Eighty four cases were enrolled at three centers as part of an Investigational Device Exemption to evaluate a robotic-assisted TKA. All patients had a CT scan as part of a pre-operative planning protocol. Scans were segmented and implant sizes predicted based on the patients boney morphology and an estimated 2mm cartilage presence. Additional information such as actual cartilage presence and soft tissue effects on balance and kinematics were recorded intra-operatively. Utilizing this additional information, surgical plans were fine tuned if necessary to achieve minimal insert thickness and balance. Data from the Preoperative CT plan sizing and final size were compared to determine the percentage of size and within one size accuracy.
Results
The pre-operative plan was able to determine the femoral and tibial components within one size for 100% of cases. Intra-operatively, surgeon upsized femoral 15 out of 85 (18%), downsized femoral 1 out of 85 (1%), baseplate 13 out of 85 (15%), and downsized baseplate 4 out of 85 (5%). Polyethylene exact size could be planned 93% of the time.
Discussion/Conclusion
Robotic-assisted pre-operative CT based planning was accurate over 70% of the time for the femur and tibial components, and over 90% with respect to the insert thickness Additionally, intraoperative information allowed for adjustments to provide patients with ideal coverage of articular surfaces and for joint balancing providing optimal individualized component placement. Further research is needed to determine the potential cost savings in hospital and OR inventory management.
