Distal Femoral Cortical Irregularity at Knee MRI: Increased Prevalence in Youth Competitive Alpine Skiers

Background Tumor-like cortical irregularities at the posterior distal femur are common incidental findings in adolescents, but the origin of these irregularities is debated. Purpose To compare the prevalence of distal femoral cortical irregularities (DFCIs) at different tendon attachment sites in youth competitive alpine skiers with that in young adults. Materials and Methods In this secondary analysis of a prospective trial, unenhanced 3-T knee MRI scans obtained in youth competitive alpine skiers were compared with images in control participants of the same age from 2014 to 2019 (Cantonal Ethics Committee Zurich registry number: KEK-ZH-2017-01395) for presence of DFCIs at the femoral attachment of the medial head of the gastrocnemius muscle (MHG) and/or lateral head of the gastrocnemius muscle (LHG) and adductor magnus tendon by two radiologists. DFCI size and tendon attachment position were measured. Tendon attachment position and associated MRI findings (meniscus, cartilage, bone marrow edema, joint effusion, ligaments, tendons) were examined for an association with DFCI. Pearson χ$^{2}$, Student t test, logistic regression, and κ statistics were applied. Results Unilateral knee MRI scans obtained in 105 skiers (mean age, 14.8 years ± 0.6 [standard deviation]; 66 boys) and in 105 control participants (mean age, 14.6 years ± 0.5; 59 boys) were evaluated. DFCIs were found in 61 of 105 skiers (58%; 95% confidence interval [CI]: 48.5%, 67.2%) compared with 28 of 105 control participants (27%; 95% CI: 18.9%, 35.7%) (P < .001). Two skiers had more than one DFCI. Distribution of DFCIs for skiers and control participants was 60 of 63 (95.2%) and 26 of 28 (92.8%) at the MHG, three of 63 (4.8%) and one of 28 (3.6%) at the LHG, and zero of 63 (0%) and one of 28 (3.6%) at the adductor magnus attachment site, respectively. Interreader agreement was almost perfect (κ = 0.87; 95% CI: 0.80, 0.93). The mean size of MHG-related DFCIs in skiers (3.7 mm) was not different compared to the size of those in control participants (3.4 mm) (P = .32), nor was a difference found for the MHG tendon attachment position in knees with DFCI (63.9 mm vs 63.0 mm, P = .83) or without DFCI (63.6 mm vs 62.8 mm, P = .86). Regarding associated MRI findings, increased signal intensity of the MHG tendon showed a significant association with MHG-related DFCI in both groups (P = .01 for both). Conclusion A distal femoral cortical irregularity at the attachment sites of tendons was a frequent incidental finding on knee MRI scans, with an increased prevalence in youth competitive alpine skiers. © RSNA, 2020 Online supplemental material is available for this article