What is the Best Way to Identify Femoral and Tibial Anatomical Landmarks for Component Alignment? A Cadaver-Based...

Femoral and tibial component malalignment in total knee arthroplasty has been associated with postoperative instability, patient dissatisfaction, and early component failure. Precise placement and rotational alignment are therefore critical to achieving optimal clinical results. Femoral and tibial component positioning relies on identification of bony landmarks relative to native anatomy. Previous studies have demonstrated that manual identification of these landmarks is inconsistent; however, few studies have directly compared manual probe identification with computed tomography (CT) measurements. The purpose of this study was to compare variability in femoral and tibial landmark identification using manual probing versus CT-based identification. Femoral and tibial bony landmarks were identified on six cadaver knees by two evaluator groups, including the anterior trochlear groove, intercondylar notch, medial and lateral posterior femoral condyles, medial and lateral epicondyles, tibial tubercle, medial and lateral tibial plateau, and medial and lateral malleoli. The first group consisted of three fellowship-trained arthroplasty surgeons who identified landmarks intraoperatively using a manual probe, whereas the second consisted of three evaluators who identified landmarks digitally on corresponding preoperative CT scans. Each landmark was identified three times with rotation between specimens for randomization, and data were collected using a robotic system. The three-dimensional landmark locations were converted into a distance value, and variation between landmarks was analyzed. A two-sample student t-test was utilized to evaluate statistical significance at p < 0.05. The CT-based identification demonstrated decreased variation than manual probe techniques across all landmarks. Statistically significant reductions in variability were observed for the anterior trochlear groove (p < 0.001), medial posterior femoral condyle (p = 0.010), lateral posterior femoral condyle (p = 0.021), medial epicondyle (p = 0.006), lateral tibial plateau (p = 0.029), and lateral malleolus (p = 0.028). Identification using CT-based techniques of femoral and tibial landmarks demonstrated less variability than manual probing, which may lead to more consistent implant component placement. Further studies are needed to determine the influence of landmark variation on component positioning.