Surgeon accuracy in identifying anatomic landmarks for femoral component positioning in tka

Summary There is tremendous variability amongst surgeons9 ability to reference anatomic landmarks. This may suggest the necessity of other objective methods in determining femoral alignment and rotation. Introduction Despite the durability of total knee arthroplasty, there is much room for improvement with regards to functional outcome and patient satisfaction. One important factor contributing to poor outcomes after TKA is malrotation of the femoral component. It has been postulated that this is due to failure of surgeons to correctly reference bony landmarks, principally the femoral epicondyles, however, this is unproven. The purpose of this study was to evaluate the accuracy of joint surgeons and trainees in identifying anatomic landmarks for positioning the femoral component and to determine the effect of prior training and experience. Methods 23 surgeons (17 attending surgeons, 6 trainees) participated in this study. Using custom-made computer software, each surgeon interactively defined the epicondylar axis (EA), the anterior-posterior axis (AP) of the distal cut (Whiteside9s Line) on 3D computer models of 10 normal femora reconstructed from CT scans. Each surgeon then aligned a standard distal cutting guide on the resected distal surface of each femoral model. A standardized procedure was employed to determine the true location of the epicondyles, the direction of Whiteside9s Line and the orientation of the cutting guide. Each participant was surveyed to ascertain their extent of formal training in joint arthroplasty, their annual volume of TKA cases, and whether they routinely aligned their TKAs using Whiteside9s and the transepicondylar axis. The difference between the ideal and surgeon-selected parameters were calculated and correlated with data describing each surgeon9s training and experience. Results Landmark selection and guide placement was highly variable between surgeons. Overall, surgeons placed Whiteside9s line in 1.83°± 7.01° of internal rotation vs. the calculated axes. Additionally, surgeons placed the transepicondylar axis in 1.40°± 3.72° of internal rotation vs. the calculated axes. On average, the guide was placed in 1.44°± 2.59° of additional internal rotation in comparison to the selected transepicondylar axis. Surgeons who routinely utilized the transepicondylar axis intraoperatively placed the guide significantly closer to the selected transepicondylar axis than those who did not (0.74°± 1.28° vs. 1.85°± 3.05°, p=.0024). Surprisingly, fellowship training, years of training, and volume of cases per year had no statistical effect the outcome of placement. Conclusion This study suggests that there is tremendous variability amongst surgeons9 ability to accurately reference the femoral epicondyles, Whiteside9s line, and the transepicondylar axis. Our results also indicate that surgeons are not able to identify Whiteside9s line with sufficient reliability for it to be a dependable indicator of correct component alignment in TKA. Our data also support the use of other methods to reliably determine correct rotational alignment of the femoral component in total knee arthroplasty.