BackgroundTo investigate the differences in postoperative patellofemoral pressures and patellar tracking during at least three years of follow-up in patients using three prostheses of different designs in total knee arthroplasty (TKA) without patellar resurfacing.Methods: Radiographic investigationsThe study included 401 patients who had a total of 480 knee prostheses implanted without patellar resurfacing. The prostheses used were Genesis II (external rotation design of femoral prosthesis), Triathlon (design with deep trochlear grooves), and Gemini MK II (deepening of trochlear groove and lateral condylar protrusion that closely follows the anatomical shape). The patients' patellar tracking was assessed by measuring patellar tilt and displacement during postoperative follow-up. Furthermore, postoperative knee function and pain were evaluated through range of motion, Knee Society scores (KSS), and Forgotten Joint Score (FJS) to compare the different groups.Finite element analysisConstructing a finite element model of the knee joint of a normal volunteer after total knee arthroplasty using different prostheses for nonpatellar replacement. The three models' von Mises stress distribution heat map, peak contact pressure, and patellar transverse displacement were compared at 30°, 60°, and 90°, respectively.Results: Radiographic investigationsA total of 456 knees of 384 patients were investigated at a 3-year follow-up after TKA without patellar resurfacing. There were no significant differences in patellar tracking between the three groups. Patients with all three prostheses demonstrated favorable clinical outcomes at 3 years postoperatively, with no statistically significant differences in knee scores (91.9 vs 92.3 vs 91.8) or range of motion (127.9° vs 128.5° vs 127.7°) between the groups. However, there was a significant difference between Genesis II and Gemini MK II in the Forgotten Joint Score (59.7 vs 62.4). Patients with persistent postoperative anterior knee pain were present in all three groups (16 vs 12 vs 10), but the incidence was not significantly different.Finite element analysisThe von Mises stress distribution heat map showed that during flexion, the patellofemoral stresses were mainly concentrated on the lateral side of the prosthesis side, and the contact site gradually shifted downward with increasing flexion angle. At the same time, the peak contact stress of the patellofemoral joint increased with the gradual increase in the flexion angle. Genesis II, with a wider and shallower trochlear groove, showed greater patellofemoral stresses and lateral patellar displacement after TKA without patellar resurfacing. The Gemini MK II with a deeper trochlear groove and slightly protruding lateral condyle is more in line with anatomical design, with smaller patellofemoral joint pressure and better patellar tracking.ConclusionsIn TKA without patellar resurfacing, a prosthesis with a deeper trochlear groove, a slightly higher lateral femoral condyle, and a more anatomically designed knee that better matches the patellar morphology should be a better choice.