To analyze the postoperative functional recovery effect of posteromedial combined anterolateral dual portal technique in the treatment of complex tibial plateau fractures involving the posteromedial condyle, and explore the key factors affecting prognosis. A retrospective analysis of 85 patients with complex tibial plateau fractures involving the posterior medial condyle, all enrolled between January 2024 and January 2025, was conducted. Based on surgical approach, patients were categorised into a combined group (42 cases; combined posteromedial and anterolateral approach) and single-approach group (43 cases; anterolateral approach). The study compared postoperative 3-month joint function (HSS) scores, posterior tibial slope (PTS), medial proximal tibial angle (MPTA), knee and ankle range of motion (ROM) in flexion, complication rates, and key influencing factors identified through data mining between the two groups. At the three-month postoperative assessment, the combined approach group demonstrated significantly superior outcomes compared to the single-approach group. This was evidenced by higher scores across all domains of the HSS scale: joint stability (7.65 ± 1.54), flexion deformity (7.33 ± 1.19), muscle strength (7.58 ± 1.31), range of motion (13.32 ± 1.75), functional activity (16.44 ± 2.21), pain (21.13 ± 2.29), and total HSS score (73.36 ± 7.97) (all P < 0.05). Radiographic parameters, including MPTA (88.65 ± 3.62°), PTS (7.86 ± 1.22°), and knee flexion ROM (123.28 ± 14.39°), also showed significantly greater improvement in the combined group (P < 0.05). Furthermore, the postoperative complication rate was significantly lower in the combined group (2.38%) than in the single-approach group (16.28%) (P < 0.05). Through data mining, three predictive models—the C5.0 decision tree, Bayesian network, and C&R decision tree—were evaluated, each demonstrating an overall accuracy exceeding 90%. Among these, the Bayesian network achieved the highest performance and was therefore selected for detailed analysis. This model yielded an area under the curve (AUC) of 0.862 (95% confidence interval: 0.757–0.967). Its recall rate was 84.90%, precision 100.00%, accuracy 50.00%, and F1-score 62.93. Further analysis identified muscle strength, weight, and intraoperative bleeding volume as the three most influential predictors, with relative importance scores of 0.9786, 0.8739, and 0.7837, respectively. The combined posteromedial and anterolateral approach significantly improves anatomic reduction and functional recovery and reduces postoperative complications in complex tibial plateau fractures.
Liu et al. (Tue,) studied this question.