The Lateral Femoral Condyle is not Hypoplastic Relative to the Medial Condyle in 6829 Magnetic Resonance Images Irrespective of Gender, Age, or Extent of Arthritis

Understanding the relationship between the radii of the medial/lateral femoral condyles (MFC/LFC respectively) is important for restoring kinematics in knee arthroplasty. The objective of this study is to use a large cohort of patient magnetic resonance Images (MRIs) to investigate whether asymmetry exists between the radii of the medial/lateral femoral condyles. Patients recruited into the Osteoarthritis Initiative (OAI) with knee MRIs were included. Using a validated machine learning algorithm, the radii of each condyle was calculated. The study sample was split into cohorts depending on medial and lateral compartment wear patterns of each knee in addition to their KL classification. The radii of each condyle in each cohort were compared using paired t-tests. Finally, a multivariable regression was run to evaluate factors that could impact differences between medial/lateral condylar size. 6,829 MRIs were included in this study of which 89% were classified as varus knees. The average best fit radius of the MFC was significantly smaller than that of the LFC (15.3mm vs. 16.8mm, p<0.001) in general and in patient knees with medial wear with KL 0, 1, 2, 3 and 4 and patient knees with lateral wear with KL 0, 2, 3, and 4. After adjusting for age, BMI, and Gender, patients with lateral wear patterns and patients with increasing KL score had a statistically significant larger lateral condylar size. ML was effectively used to automate the measurement of femoral condyle size suggesting that the LCF has a slightly larger radius than the MFC and is not hypoplastic.