Background: The optimal soft tissue release technique for severe varus total knee arthroplasty (TKA) remains controversial. The technique of tibial plateau reduction was suggested for severe varus deformity during TKA. The purpose of this study was to evaluate the clinical and radiological outcomes of the technique of tibial plateau reduction. Methods: Fifty-one knees (39 patients) with a mechanical femoro-tibial angle (MFTA) greater than 15 degrees treated with TKA was presented. The clinical and radiographic data were gathered prospectively. The patients were divided into two groups according to the preoperative angle: Group A comprised patients with a preoperative MFTA less than or equal to 20 degrees, while Group B included patients with a preoperative MFTA greater than 20 degrees. Results: Among the 51 knees, 46 knees (90.2%) had femoro-tibial component size mismatch. The KSS-KS improved from preoperative 39.9 +/- 11.3 points (range:18-68 points) to 92.3 +/- 5.2 points (range: 83-100 points) at the final follow up (t = 30.141, P < 0.001); the KSS-FS improved from preoperative 48.0 +/- 11.5 points (range: 31-71 points) to 87.2 +/- 8.7 points (range: 63-100 points) at the final follow up (t = 19.413, P < 0.001). The mean MFTA was corrected from 21.0 +/- 4.7 degrees preoperatively to 2.9 +/- 1.8 degrees at the latest follow up (P < 0.001). The preoperative MFTA was less than or equal to 20 degrees in 27 knees (Group A), and greater than 20 degrees in 24 knees (Group B). The mean postoperative MFTA was 1.9 +/- 0.9 degrees in Group A, and 4.1 +/- 1.7 degrees in Group B (P < 0.001). The neutral limb alignment was reached in 26 knees (96.3%) in group A, which was significantly higher than that in group B (29.7%) (P = 0.019). Conclusions: The technique of tibial plateau reduction provided satisfactory clinical and radiological outcomes without instability for severe varus TKA. However, surgeons should remain aware that the greater the preoperative varus deformity, the greater the postoperative residual varus deformity, and that there is a high incidence of femoro-tibial component size mismatch using the technique. (c) 2024 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies.