Demirjian Thomas, Crues John, Powers Christopher M
Division of Biokinesiology and Physical Therapy, University of Southern California, Los Angeles, California, USA.
RadNet MSK Imaging, Los Angeles, California, USA.
J Orthop Res. 2025 Mar;43(3):567-575. doi: 10.1002/jor.26014. Epub 2024 Nov 18.
Individuals who undergo anterior cruciate ligament reconstruction are at elevated risk for developing early-onset patellofemoral joint osteoarthritis. Our objective was to use K-means clustering to ascertain whether individuals at risk for patellofemoral joint osteoarthritis could be identified as determined by the presence of multiple co-existing anatomical and patella alignment risk factors. Forty participants (20 after anterior cruciate ligament reconstruction, 20 healthy controls) underwent magnetic resonance imaging assessment of the patellofemoral joint. Measures of hypothesized risk factors for patellofemoral joint osteoarthritis were obtained including patella alignment (lateral patella displacement and tilt), trochlear morphology (sulcus angle, lateral inclination angle), patella height (Insall-Salvati ratio and patella articulating overlap), and patellofemoral joint contact area. K-means clustering (k = 2) was used to ascertain whether a high-risk group could be identified. Following clustering, two distinct groups were detected. Participants assigned to cluster 1 exhibited features consistent with patellofemoral joint osteoarthritis including greater lateral patellar displacement and tilt, flatter trochlear grooves and lower lateral trochlear inclination, less patella articulating overlap, and reduced contact area. The proportion of females after anterior cruciate ligament reconstruction assigned to cluster 1 was 75% (N = 15) compared to 25% of healthy females (N = 5). K-means clustering was capable of characterizing individuals at elevated risk for patellofemoral joint osteoarthritis based on the presence of multiple co-existing anatomical and patella alignment risk factors. The fact that a significant percentage of females were assigned to the high-risk cluster supports the clinical observation that these individuals may be at higher risk of early-onset patellofemoral joint osteoarthritis.
接受前交叉韧带重建的个体发生早发性髌股关节骨关节炎的风险较高。我们的目的是使用K均值聚类法来确定是否可以根据多种并存的解剖学和髌骨对线风险因素来识别髌股关节骨关节炎的高危个体。40名参与者(20名前交叉韧带重建术后患者,20名健康对照者)接受了髌股关节的磁共振成像评估。获取了髌股关节骨关节炎假定风险因素的测量值,包括髌骨对线(髌骨外侧移位和倾斜)、滑车形态(沟角、外侧倾斜角)、髌骨高度(Insall-Salvati比率和髌骨关节重叠)以及髌股关节接触面积。使用K均值聚类法(k = 2)来确定是否可以识别出高危组。聚类后,检测到两个不同的组。分配到聚类1的参与者表现出与髌股关节骨关节炎一致的特征,包括更大的髌骨外侧移位和倾斜、更平坦的滑车沟和更低的外侧滑车倾斜度、更少的髌骨关节重叠以及更小的接触面积。前交叉韧带重建术后分配到聚类1的女性比例为75%(N = 15),而健康女性为25%(N = 5)。基于多种并存的解剖学和髌骨对线风险因素,K均值聚类法能够对髌股关节骨关节炎高危个体进行特征描述。相当比例的女性被分配到高危聚类这一事实支持了临床观察结果,即这些个体可能有更高的早发性髌股关节骨关节炎风险。