Paper 105: Changes in Patellar Alignment After Total Knee Arthroplasty

Abstract

Although the results of total knee arthroplasty continue to improve, problems related to the patellofemoral joint remain significant. This study examined the factors affecting patellar alignment after total knee arthroplasty and subsequent changes during a postoperative period of 5.3 years. Fifty-five knees in 39 patients were retrospectively evaluated by radiography for postoperative patellar alignment. This study included four men and 35 women with a mean follow-up time of 5.3 years (range: 2.5 to 8.4). All procedures utilized identical surgical technique; through a midline skin incision and medial parapatellar arthrotomy, the patella was fully everted. We adjusted the rotation of the femoral component until it was parallel to the surgical epicondylar axis. Using the posterior femoral condyles as references, we inserted the femoral components with 3° to 5° of external rotation to the posterior condylar axis according to preoperative CT scans. Tibial components were set using the medial third of the tibial tuberosity as a reference. All patellae were resurfaced with three-peg offset-dome all-polyethylene patellar component. Patellar alignment was assessed by measuring the lateral tilt and lateral displacement in skyline views taken with the knees at approximately 60° of flexion. Patellar alignment was evaluated one month after surgery, and at final follow-up. Changes in patellar lateral tilt angle and lateral displacement in the postoperative period were calculated using data obtained one month after surgery and at final follow-up. We did not observe severe malalignment of the femorotibial joint, patellar bone fracture, prosthesis failure, or any other clinical complications after TKA in any cases. The mean patellar lateral tilt angle was -0.7° (range: −13.7 to 7.0°) one month after surgery and −0.4° (range: −11.4 to 11.6°) at final follow-up. The mean patellar lateral displacement was 1.1mm (−2.7 to 6.6mm) one month after surgery and 0.9mm (−3.7 to 5.8mm) at final follow-up. We observed a negative correlation between the patellar resection angle and lateral tilt one month after surgery (p<0.0001) and between the patellar resection angle and patellar lateral displacement one month after surgery (p=0.0003). Lateral retinacular release was required for thirteen knees. In these cases, the resection angle was significantly smaller than that seen in knees without lateral retinacular release (p=0.0012). These results demonstrate that a smaller resection angle correlated with an increased frequency of lateral retinacular release. The change in patellar lateral tilt angle after 5.3 years was 0.3° (range: −7.2 to 7.0°), while the change in patellar lateral displacement was −0.3 mm (range: −5.2 to 3.7 mm). These changes did not significantly correlate with patellar alignment at any time point, resection angle, resection thickness, lateral retinacular release, or clinical results. The postoperative femorotibial alignment, however, exhibited a statistically significant correlation with the changes in lateral tilt angle (p=0.020). A more varus postoperative alignment correlated with increased lateral tilt of the patella in the postoperative period. We observed a satisfactory patellar alignment after a postoperative period of 5.3 years. The patellar resection angle had a significant effect on postoperative patellar alignment. Our findings suggest that leaving a thicker patellar remnant on the medial side can decrease postoperative lateral tilt and the resulting need for lateral retinacular release. Postoperative changes in patellar alignment were very small; none of the knees demonstrated complications of the patellofemoral joints. In this study, we observed the tendency that a postoperative varus alignment leads to patellar lateral tilt. Surgeons should pay scrupulous attention to femorotibial alignment during total knee arthroplasty to decrease patellofemoral complications and femorotibial problems.