Trochlear Morphological Parameters Based on Measurements at the Trochlear Groove Entry Plane Demonstrate Excellent Predictive Efficacy for Identifying the High‐Grade J‐Sign

ABSTRACT Objective The correction of high‐grade J‐sign is critical for improving clinical outcomes in patients with trochlear dysplasia, but insufficient attention has been given to anatomical abnormalities at the trochlear groove entry plane, the established gold standard for describing trochlear morphology. We aim to compare the diagnostic accuracy and reliability of axial computed tomography (CT) measurements at the trochlear groove entry plane versus the Roman arch plane in predicting the J‐sign among patients with patellar instability. Methods This retrospective study enrolled 120 knees diagnosed with recurrent patellar dislocation. Based on patellar tracking assessments, the J‐sign was categorized into a low‐grade group (J‐sign negative and Grade I) and a high‐grade group (J‐sign Grades II and III). Preoperative measurements of trochlear depth, sulcus angle (SA), and lateral trochlear inclination (LTI) were obtained at both the trochlear groove entry plane and Roman arch plane using computed tomography (CT). Receiver operating characteristic (ROC) curve analysis was utilized to evaluate the predictive accuracy of the three parameters measured at both anatomical planes for J‐sign severity. Results Parameters measured at the trochlear groove entry plane demonstrated superior diagnostic accuracy on ROC curves compared with those obtained at the Roman arch plane. In both planes, the trochlear depth and LTI of the high‐grade group were significantly lower than that of the low‐grade group, while the SA was significantly higher than that of the low‐grade group (all p < 0.01). At the trochlear groove entry plane, the area under the curve (AUC) values for distinguishing low‐grade and high‐grade J‐sign were as follows: AUC Trochlear depth = 0.858, AUC SA = 0.912, and AUC LTI = 0.803. For SA, sensitivity was 0.87, specificity 0.87, and the optimal cutoff value 149.8°. In contrast, measurements at the Roman arch plane exhibited lower discriminative performance with AUC Trochlear depth = 0.714, AUC SA = 0.740, and AUC LTI = 0.694. Specifically, SA demonstrated reduced sensitivity of 0.67, comparable specificity of 0.77, and a lower cutoff threshold of 140.2°. Conclusions When evaluating knee parameters in patients with recurrent patellar instability, axial CT measurements at the trochlear groove entry plane demonstrate superior predictive efficacy for the J‐sign compared to the Roman arch plane. Specifically, CT‐based measurement of the SA enables enhanced differentiation between low‐grade and high‐grade J‐sign subgroups, providing critical preoperative insights of patellar maltracking.