Abstract
Introduction
Computed tomography (CT) can be utilized to design patient specific instruments (PSI) for total knee arthroplasty (TKA). The PSI preoperative plans predict bone resection, anterior-posterior implant position, implant rotation and implant size. The purpose of this study was to compare preoperatively predicted implant sizes (tibia and femur) to the actual implanted sizes. Data were compiled from two surgeons, one in the United Kingdom (Surgeon 1, cruciate retaining) and one in the United States (Surgeon 2, posterior stabilizing). Both used the same primary TKA implant systems (Sigma® and Attune®; DePuySynthes®, Warsaw, Indiana). This is the largest comparison of CT-based PSI size accuracy between two implant systems.
Methods
An international cohort of 396 CT-based PSI-TKA preoperative plans (TruMatch®)were compared to postoperative implant records. Data were retrospectively analyzed for Sigma®(n=351) and Attune® (n=45), both as separate cohorts and as a combined cohort (Sigma® + Attune®). Three analyses were performed: Tibia and femur plan accuracy, major size changes (femoral size change or tibial size change resulting in a femoral size change) and minor size changes (tibial size change not impacting femoral size). Inter-rater reliability analyses using ICC (intra-class correlation) and the Kappa statistic were performed to determine reliability and agreement among the groups.
Combined TKA implant data (Sigma® + Attune®) for surgeons 1 and 2 were compared for accuracy between users utilizing different implant designs, cruciate retaining (CR) versus posterior stabilized (PS).
Results
In the combined system analysis (Sigma® + Attune®) femoral implant prediction was 97.0% accurate and combined tibial implant size accuracy was 79.5%. There were no significant differences between the systems for tibial or femoral accuracy. See Table 1. There were 12 major size changes, 11 downsized femoral implants and 1 upsized femoral implant (all femoral changes were with Sigma® system). There were 81 minor size changes. Per Kappa, the plans were in excellent agreement with the femoral implant size and had substantial agreement with tibial implant size (p<.01). See Table 2.
Comparing size accuracy between Surgeon 1 and Surgeon 2, Surgeon 1 had significantly greater tibial accuracy (p<.01), while femoral accuracy showed no significant difference (p=0.49). See Table 3.
Discussion
In this combined data set of two surgeons, we report high implant sizing accuracy overall. This accuracy was noted across implant systems (Sigma® and Attune®) and across surgeons (1 and 2) utilizing different implant designs (CR vs. PS) using TruMatch® PSI. In all cohorts, the femur was more accurately predicted than the tibia. Accurate size reconstitution and reconstruction of the femur is critical for maintenance of posterior condylar offset, avoidance of anterior compartment overstuffing and avoidance of anterior femoral cortical notching. This study demonstrates the reproducibility of CT-based PSI TKA across different implant systems (Sigma® and Attune®), implant designs (CR and PS) and different surgeons. The ability to accurately predict implant size can also contribute system efficiencies: improved implant inventory management, development of size-focused instrumentation sets and potentially reduced workload for sterile processing departments.
