Rotational limb alignment changes following total knee arthroplasty

The aim of the present study was to assess the changes in rotational alignment introduced by total knee arthroplasty (TKA) and the reproducibility of pre- and postoperative CT measurements of rotational limb alignment. For this purpose we analyzed data from 196 consecutive cruciate-retaining, fixed bearing Columbus TKA procedures. Both pre- and postoperative scans torsion difference CT scans were available for measurements in 89 cases. Using these CT scans the neck-malleolar angle (NMA), the femoral posterior condylar angle (fPCA), the tibial posterior condylar axis (tPCA) and the tibial torsion angle (TTA) were independently assessed by three raters. CT scans were re-evaluated 8 weeks later by the most experienced rater for assessment of intraobserver agreement. Measurements of all angles were prone to high standard deviations reflecting interindividual variability. Mean fPCA changed from 1.3A degrees to 2.7A degrees internal rotation preoperatively to 0.1A degrees-1.9A degrees internal rotation postoperatively. Based on a relative external rotation of the tibial base plate as compared to the preoperative situation, we found a relative internal rotation of the postoperative NMA and tibial torsion of 3A degrees-5.4A degrees and 6A degrees-7.5A degrees, respectively. Intra- and interobserver agreement was strong for all angles assessed (ICCs 0.7-1.0) except for fPCA (ICC 0.2-0.6). However, mean absolute measurement differences for fPCA were clinically acceptable (1.2A degrees-2.6A degrees). Reproducibility of CT rotational limb alignment measurements was found to be clinically acceptable. Rotational alignment of the femoral and even more so of the tibial component will ultimately affect the rotational alignment of the entire limb-at least when fixed bearings are used. Diagnostic study, Level III.