Purpose Coronal tibiofemoral subluxation (CTFS) is considered a controversial and potential contraindication to unicompartmental knee arthroplasty (UKA) but is less discussed. The study aims to observe the CTFS in a cohort of patients before and after mobile-bearing UKA and to investigate the relationship between preoperative variables (preoperative CTFS and preoperative CTFS under valgus stress) and postoperative CTFS after mobile-bearing UKA. Methods The study retrospectively analyzed 181 patients (224 knees) undergoing mobile-bearing UKA from September 1 2019 to December 31 2021. By using hip-to-ankle anterior-posterior (AP) standing radiographs and valgus stress force radiographs, preoperative CTFS, preoperative CTFS under valgus stress, and postoperative CTFS were measured. CTFS was defined as the distance between the tangent line to the outermost joint edge of the lateral condyle of the femur and the tangent line of the lateral tibial plateau. All patients were divided into two groups based on postoperative CTFS, group A (postoperative CTFS <= 5 mm) and group B (postoperative CTFS > 5 mm). The Student's t-test, one-way ANOVA together with Tukey's post hoc test, the chi-square test, the Fisher's exact test, Pearson correlation analysis, simple and multiple linear regression, and univariate and multiple logistic regression were used in the analyses. Results The means +/- standard deviations (SD) of preoperative CTFS, preoperative CTFS under valgus stress, and postoperative CTFS were 4.96 +/- 1.82 mm, 3.06 +/- 1.37 mm, and 3.19 +/- 1.27 mm. The difference between preoperative CTFS and postoperative CTFS was statistically significant (p < 0.001). The preoperative CTFS (6.35 +/- 1.34 mm) in Group B (n = 22) was significantly higher than that (4.81 +/- 1.82 mm) in Group A (n = 202) (p < 0.001), so was the variable-preoperative CTFS under valgus stress (5.41 +/- 1.00 mm (Group B) > 2.80 +/- 1.14 mm (Group A), p < 0.001). In Pearson correlation analysis, there was a correlation between preoperative CTFS and postoperative CTFS (r = 0.493, p < 0.001), while the correlation between preoperative CTFS under valgus stress and postoperative CTFS was stronger (r = 0.842, p < 0.001). In multiple linear regression analysis, preoperative CTFS under valgus stress (beta = 0.798, 95% confidence interval (CI) = 0.714-0.883, p < 0.001) was significantly correlated with postoperative CTFS. In multiple logistic regression analysis, preoperative CTFS under valgus stress (OR = 12.412, 95% CI = 4.757-32.384, and p < 0.001) was expressed as the risk factor of postoperative CTFS (> 5 mm). Conclusion Preoperative CTFS can be improved significantly after mobile-bearing UKA. In addition, postoperative CTFS is correlated with preoperative CTFS under valgus stress and a higher preoperative CTFS under valgus stress will increase the risk of higher postoperative CTFS (> 5 mm).