ra0015">Thirty-one patients who underwent magnetic resonance imaging of a TKA between April 2008 and November 2015 were retrospectively reviewed. A single surgeon performed the TKA using a posterior condylar angle of 3° (PCA group) or surgical transepicondylar axis (TEA group) to determine femoral component rotation. The hip-knee-ankle angle and the tibial plateau-tibial shaft angle (TPTSA) were measured on full-length x-rays, and the rotation of the femoral component was measured as compared to the TEA on magnetic resonance imaging (negative values indicate internal rotation).
ra0020">The median measured rotational deviation of the femoral component from the TEA was significantly higher in the PCA group than in the TEA group (−3.8 ± 2.9° and −1.4 ± 1.9°, respectively) (P = .02). When knees with preoperative varus and neutral alignment were evaluated, the median measured rotational deviation of the femoral component was significantly higher in the fixed PCL group than in the TEA group (−3.4 ± 3.3° and −0.61 ± 1.3°, respectively) (P = .04). Linear regression used to evaluate the relationship between the TPTSA and femoral component rotational deviation from the TEA revealed similar near zero slopes (P = .90); however, the Y intercepts in the TEA group were significantly higher than the PCA group (−2.8 ± 0.7 and −5.5 ± 1.1, respectively) (P = .007).
ra0025">The use of the surgical TEA as an intraoperative rotational reference is more reliable than the PCA in valgus, varus, and neutrally aligned knees independent from the magnitude of the TPTSA.