Diagnostic Accuracy of Ultrasonography for Knee Pathologies: A Prospective Comparative Study With MRI

Background Knee joint pathologies are a common cause of orthopedic consultation, with many requiring imaging. Magnetic resonance imaging (MRI) is considered the non-invasive imaging reference standard with superior soft tissue resolution. However, it is expensive and time-consuming. High-resolution ultrasonography (HRUS), on the other hand, provides high soft tissue resolution for superficial structures, is cheaper, and is readily available with dynamic capabilities. Methods This prospective study was carried out on 50 patients referred from orthopedics to the radiology department of Government Medical College and Hospital, Jammu, with a request for MRI of the symptomatic knee joint. All patients underwent USG first, followed by MRI with standard knee protocol on a 1.5 T Siemens MRI (Munich, Germany). MRI was used as the reference standard. Ultrasonography (USG) findings were compared with MRI findings, and diagnostic indices, including sensitivity, specificity, accuracy, positive predictive value, and negative predictive value, were calculated. Results The anterior cruciate ligament and the medial meniscus were the most commonly damaged structures. HRUS showed low sensitivity (46.1% and 40%, respectively) but high specificity (95.8% and 97.5%, respectively) for medial and lateral meniscus injuries. Diagnostic performance of USG was excellent for superficial structures, including collateral ligaments, tendons, synovitis, collections, and popliteal cysts, with specificity approaching 100%. Joint effusion detection showed 94.7% sensitivity and 100% specificity. However, the cruciate ligaments were poorly visualized on USG. Conclusion Although MRI remains the gold standard for thorough evaluation of knee pathologies, USG has shown strong specificity and good diagnostic performance for superficial soft tissue structures. Therefore, USG may serve as a valuable screening tool, particularly in resource-limited settings.