IMAGING CHARACTERISTICS OF KNEE OSTEOARTHRITIS ON 3.0 TESLA MRI: A CROSS SECTIONAL STUDY IN VIETNAM

Tóm tắt

This study aim to investigate and characterize the imaging features of knee osteoarthritis (OA) using high-field 3.0 Tesla Magnetic Resonance Imaging (MRI) in a cohort of patients at Thong Nhat Hospital. A cross-sectional descriptive study was conducted on 156 patients with radiographically confirmed knee OA who underwent 3.0T MRI between November 2024 and May 2025. Pathologies of the menisci, cruciate ligaments, collateral ligaments, subchondral bone and joint space were systematically evaluated. Univariable and multivariable logistic regression analysis was employed to identify imaging features associated with OA. The study included 156 patients (mean age 55.8±10.6 years; 73 males). The majority of patients were classified radiographically as Kellgren-Lawrence (KL) grade I (n=99, 63.46%). Despite the prevalence of early-stage radiographic disease, 3.0T MRI revealed a high burden of soft-tissue pathology. The most common ligamentous injury was a partial tear of the anterior cruciate ligament (ACL) in 42 patients (27%), while medial meniscal tears were identified in 56 patients (35.9%). Mild joint effusion was a frequent finding, present in 107 patients (68.6%). In the multivariable model, age (Odds Ratio 1.03, 95% Confidence Interval [CI] 1.02–1.04), ACL injury (OR 1.14, 95% CI 1.02–1.28), posterior cruciate ligament (PCL) injury (OR 1.24, 95% CI 1.02–1.50), collateral ligament injury (OR 1.23, 95% CI 1.02–1.49) and lateral meniscal injury (OR 1.14, 95% CI 1.00–1.29) were identified as significant independent predictors of OA. High-field 3.0T MRI is a pivotal tool for the comprehensive assessment of knee OA, detecting a wide spectrum of radiographically occult soft tissue and bone marrow pathologies. These findings, particularly those related to ligamentous and meniscal integrity, are critical to understanding the disease's biomechanical underpinnings and may be present despite minimal radiographic changes.