Magnetic Resonance Imaging-Arthroscopy Correlation in Traumatic Knee Injuries: A Retrospective Diagnostic Accuracy Study.

Abstract

Introduction: Magnetic resonance imaging (MRI) is widely used for the evaluation of traumatic knee injuries because it provides a detailed, non-invasive assessment of intra-articular structures. However, discrepancies between MRI and arthroscopic findings may occur, particularly in subtle or complex lesions. This study evaluated the diagnostic performance of pre-operative MRI by correlating MRI findings with arthroscopic findings in patients undergoing therapeutic knee arthroscopy after trauma.

Materials and methods: This single-center retrospective diagnostic accuracy study included 35 patients aged 15-60 years with traumatic knee injuries who underwent clinical evaluation, pre-operative MRI, and therapeutic knee arthroscopy. MRI findings were compared with arthroscopic findings as the reference standard. Sensitivity, specificity, positive predictive value, negative predictive value (NPV), diagnostic accuracy, and Cohen's kappa coefficient were calculated.

Results: Arthroscopy identified 31 anterior cruciate ligament (ACL) injuries, three posterior cruciate ligament (PCL) injuries, 25 medial meniscal injuries, 14 lateral meniscal injuries, and 10 osteochondral defects. MRI demonstrated sensitivities ranging from 76.0% to 100.0%, with specificity of 100.0% for all evaluated lesions. Diagnostic accuracy ranged from 82.9% for medial meniscal injuries to 100.0% for PCL injuries and osteochondral defects. Agreement with arthroscopy was substantial for ACL injuries (κ = 0.768) and medial meniscal injuries (κ = 0.644), and almost perfect for PCL injuries (κ = 1.000), lateral meniscal injuries (κ = 0.940), and osteochondral defects (κ = 1.000).

Conclusion: MRI demonstrated high diagnostic performance and substantial to almost-perfect agreement with arthroscopic findings for the evaluated traumatic intra-articular knee lesions. Medial meniscal injuries showed the lowest sensitivity and NPV, indicating that a negative MRI finding may not completely exclude clinically relevant meniscal pathology when clinical suspicion remains high.