THE EFFECT OF USING FIXED OR VARIABLE DISTAL FEMORAL RESECTION ANGLES ON POSTOPERATIVE LIMB ALIGNMENT IN TOTAL KNEE ARTHROPLASTY.

Abstract

In Total Knee Arthroplasty (TKA) restoring the mechanical alignment of the knee joint is essential. This can be improved by considering the individual variability in the angle between the mechanical and anatomical axes of the femur (FMA angle). However with the traditional instrumentation and the use of the most common fixed distal femoral resection angle of 6° we assume little or no variation in the FMA angles in different patients. In a previous study we showed that the FMA angle had a wide distribution and that there was a good correlation between the FMA angle and the pre-operative lower limb alignment in the coronal plane. Our hypothesis was that improved post operative limb alignment would be achieved with traditional instrumentation by individual measurement of the FMA angles pre-operatively and adjusting the distal femoral resections accordingly. In the study we compared the post-operative coronal limb alignment for a cohort of patients with a variable distal femoral resection angle to the previous cohort of fixed distal femoral resection angle.

The study consisted of 103 patients undergoing 103 consecutive primary TKAs between October 2008 and March 2009. All patients had pre- and post-operative Hip-Knee-Ankle digital radiographs and had TKAs performed using a variable distal femoral cut angle. The FMA angle and the mechanical femoro-tibial (MFT) angles were measured in all cases. Inter-observer variation was measured by second observer readings. We compared our results with the group of 158 consecutive patients undergoing 174 primary TKAs operated between January and October 2007 using fixed distal femoral resection angle.

Patient demographics of the two cohorts (age, gender, BMI) were similar.

The pre-operative coronal deformity for the variable cohort was less than the fixed, mean 3.7° varus (SD 5.8°) compared to 4.7° varus (SD 7.9°). The FMA angles for the variable cohort ranged from 4° to 8°, (the fixed cohort from 2° to 9°). The variable valgus resection angles cohort showed a correlation between FMA and pre-operative MFT angles as had previously been shown in the fixed cohort (r = −0.499 and r = −0.346 respectively). Post op alignment showed that accuracy within ±5° increased from 86% (fixed resection angle group) to 96% (variable resection group). When using the more commonly quoted accuracy of within ±3°, this changed from 67% (fixed resection angle group) to 85% (variable resection group). These improvements were statistically significant (chi-squared 0.025 and 0.002, respectively). To further evaluate the effect of using variable angles we analysed the improvement of each of the different groups of deformity identified in the previous study (> 8° varus, 8° varus to 1° valgus, > 2° valgus). The range was reduced in both the extreme varus and valgus groups with the variable angles. The most significant improvement was found in the valgus group with the median reducing from 3° to 2° and range from 14° to 8°.

It seems logical to use a variable distal femoral resection angle based on the patient’s individual anatomy. By doing so, our results show significant improvement of postoperative limb alignment compared to traditional method of using fixed distal femoral resection angle. In units where preoperative long leg film radiographs are available, measuring the FMA angle and setting the distal femoral resection angle guide accordingly improves the postoperative limb alignment. However, where long leg radiographs are not available, changing the distal femoral resection angle according to the pre-operative varus-valgus deformity is likely to improve the post operative limb alignment. (e.g. 4°–5° distal femoral resection angle for preoperative valgus, 6° for preoperative mild/moderate varus, and 7°–8° for preoperative severe varus).Computer navigation, however, enables us not only to use customised distal femoral cut for each patients, but it also provides many other useful information such as dynamical limb alignment through motion, component rotation, soft tissue balancing.