A Clinical Decision-Making Algorithm for Posterolateral Corner Injuries of the Knee: Development and Internal Validation.

Abstract

Posterolateral corner (PLC) injuries of the knee remain a complex and heterogeneous clinical entity, with substantial variability in surgical management and no widely accepted criteria to guide treatment selection. The purpose of this study was to develop and internally validate a clinically applicable decision-making algorithm for PLC injuries and to evaluate the outcomes of a minimally invasive reconstruction strategy in a consecutive cohort of patients. A retrospective cohort of 120 patients undergoing surgical treatment for PLC injuries, including both isolated PLC reconstructions and combined ligament procedures (ACL and/or PCL), was analyzed. Baseline demographic, clinical, radiographic, and intraoperative variables were collected. The primary outcome was defined as a successful clinical outcome at final follow-up, including absence of revision surgery, restoration of clinical stability, and functional improvement. Predictors of outcome were identified using multivariable logistic regression. Internal validation was performed using bootstrap resampling (1000 iterations). Model performance was assessed using discrimination (area under the curve [AUC]) and calibration analysis. A simplified clinical score and decision-making algorithm were derived from the final model. At a mean follow-up of 24.8 ± 6.3 months, 92 patients (76.7%) achieved a successful clinical outcome. Independent preoperative predictors of unsuccessful outcome included varus laxity >4 mm on stress radiographs (OR 3.12; p = 0.005), a positive dial test at 90° of knee flexion (OR 2.67; p = 0.018), and combined ligamentous injury involving the posterior cruciate ligament (OR 2.94; p = 0.010). The final model demonstrated good discrimination (AUC 0.81; 95% CI 0.73-0.89), with a bootstrap-corrected AUC of 0.78 and good calibration (slope 0.94). The derived clinical algorithm, based on a weighted score including varus laxity >4 mm, dial test positivity at 90°, and PCL-associated injury, stratified patients into low-, intermediate-, and high-risk categories, supporting tailored selection of minimally invasive versus anatomic reconstruction strategies. Each variable was assigned points according to its regression coefficient, allowing classification into three groups based on cumulative score: low risk (0-1 points), intermediate risk (2-3 points), and high risk (≥4 points). A clinically applicable and internally validated algorithm can effectively guide surgical decision-making in PLC injuries of the knee. By integrating key clinical and radiographic variables, this model enables safe selection of minimally invasive reconstruction in appropriately selected patients while preserving the indication for anatomic techniques in complex instability patterns. Level III, retrospective cohort study.