RÉSUMÉ
Objective:To explore the diagnostic performance of patella-tilt angle and congruence angle in episodic patellar dislocation (EPD) and the quantitative measurements of the patellar and femoral axial parameters as well as their correlation with and contributions to the patellofemoral joint alignment with the knee extended.Methods:A case control study was conducted to analyze the radiological data of EPD patients (EPD group, n=106) and patients without patellar instability (control group, n=106) admitted to Sixth People′s Hospital Affiliated to Shanghai Jiao Tong University from January 2016 to December 2019. Each group consisted of 55 females and 51 males with the age range of 14-45 years [(23.0±5.3)years], showing 1∶1 match. Axial parameters included patellar parameters (patellar width, patellar thickness, Wiberg angle and Wiberg index), femoral parameters [trochlear sulcus angle, trochlear sulcus depth, trochlear sulcus height, trochlear width ratio (lateral/medial), trochlear height ratio (lateral/medial), lateral trochlear inclination and trochlear groove medialization], and patellofemoral joint parameters (patellar tilt angle and congruence angle). The receiver operating characteristic (ROC) curve of patellofemoral joint parameters for the prediction of EPD was analyzed. Univariate analysis was performed to determine the difference of those axial parameters between the two groups. Pearson correlation analysis was used to identify the correlation between those bony parameters and patellofemoral joint parameters. Stepwise regression model was further established to determine the influencing factors and corresponding contributions for patellofemoral joint parameters. Results:When the optimal cut-off values of patellar tilt angle and congruence angle were 17.2° and 25.5°, the area under the ROC curve (AUC) for predicting EPD was 0.91 (95% CI 0.87-0.95, P<0.01) and 0.92 (95% CI 0.87-0.95, P<0.01), and the Youden index was 0.745 (sensitivity=83.96%, specificity=90.57%) and 0.717 (sensitivity=81.13%, specificity=90.57%). Univariate analysis showed that Wiberg index, femoral parameters and patellofemoral joint parameters were significantly different between the two groups (all P<0.01). For all patients, Pearson correlation analysis showed that patellar tilt angle was moderately to strongly correlated with Wiberg index, trochlear sulcus angle, trochlear sulcus depth, trochlear width ratio (lateral/medial), trochlear height ratio (lateral/medial) and lateral trochlear inclination ( r=0.51, 0.41, -0.62, 0.43, -0.49, -0.65, all P<0.01) and that congruence angle has a moderate correlation with trochlear sulcus angle, trochlear sulcus depth, trochlear width ratio (lateral/medial) and lateral trochlear inclination ( r=0.43,-0.59,0.38,-0.51, all P<0.01). For all patients, Stepwise regression model analysis showed that lateral trochlear inclination, trochlear sulcus depth, trochlear sulcus angle, Wiberg index and trochlear height ratio (lateral/medial) could explain 60% of the variation of patellar tilt angle ( R 2=0.60, P<0.01) and that trochlear sulcus depth, lateral trochlear inclination, trochlear groove medialization, trochlear sulcus angle and Wiberg index could explain 44% of the variation of congruence angle ( R 2=0.44, P<0.01). Conclusions:The patellar tilt angle and congruence angle are reliable quantitative indicators representing patellofemoral axial alignment, with a good diagnostic performance for EPD. Variations in the patellar and femoral bony structures of EPD patients are related to the patellofemoral axial alignment, with the axial parameters differently contributing to the patellofemoral alignment.