The impact of increasing femoral width on anatomic risk factors for patellar instability in paediatric patients: a retrospective MRI case-control study

Abstract Background Chronological age does not reliably reflect skeletal development in children and adolescents. Anatomical risk factors for non-traumatic lateral patellar dislocation (LPD) may therefore be better interpreted relative to knee size. This study aimed to evaluate the relationship between femoral width and MRI-based anatomical risk factors for PFI in skeletally immature patients with and without a history of patellofemoral instability. Methods In this retrospective MRI case–control study, 630 knee MRIs of skeletally immature patients were analysed, including 315 patients with first-time, non-traumatic lateral patella dislocation and 315 age- and sex-matched controls without a history of PFI. Each group comprised 146 males and 169 females, with a mean age of 14.5 years. MRI-based measurements included parameters of trochlear dysplasia, patellar tilt, patellar lateralization, and patellar height. Bicondylar femoral width was measured to normalize anatomical parameters to knee size. Independent t-tests compared group differences, and linear regression analyses assessed associations between femoral width and patellofemoral parameters. Results Compared with controls, patients with PFI showed significantly reduced trochlear depth (3.6 ± 0.09 mm vs. 4.5 ± 0.09 mm), lower lateral trochlear inclination (14.4° ± 0.29 vs. 20.8° ± 0.32), greater trochlear sulcus angle (151.0° ± 0.58 vs. 136.5° ± 0.50), lower Laurin angle (− 0.7° ± 0.55 vs. 2.5° ± 0.57), lower Fulkerson angle (2.2° ± 0.55 vs. 8.6° ± 0.57), higher patellar inclination angle (TILT; 21.2° ± 0.53 vs. 18.5° ± 0.55), greater TTTG distance (13.3 ± 0.27 mm vs. 12.0 ± 0.27 mm), and greater TTPCL distance (23.1 ± 0.33 mm vs. 19.4 ± 0.34 mm), as well as a higher Caton–Deschamps Index (1.2 ± 0.01 vs. 1.1 ± 0.01) (all p < 0.05). The Patellotrochlear Index did not differ significantly between groups (47.1 ± 1.09 vs. 47.5 ± 1.02; p = 0.764). Linear regression analysis demonstrated significant associations between increasing femoral width and several parameters, including trochlear depth, patellar inclination, Fulkerson angle, TTPCL distance, Patellotrochlear Index, and Caton–Deschamps Index in at least one cohort ( p < 0.05). Conclusion Anatomical risk factors for non-traumatic LPD differ significantly between skeletally immature patients with and without a history of patellar dislocation. Several patellofemoral parameters vary with femoral width, suggesting that knee size should be considered when interpreting patellofemoral morphology during skeletal development.