Both retrospective and blinded analyses of thin-section, high-resolution magnetic resonance (MR) images of the knee joint, produced using a solenoid surface coil, indicate that MR imaging is an effective technique for evaluating meniscal injuries. Images of 49 patients were evaluated, and the results were correlated with those of subsequent arthroscopy. A grading scale was developed to rate the index of suspicion of a meniscal tear based on the MR images. Overall, approximately 80% of menisci rated grade 4 (definite tear) or 3 (probable tear) were found to have corresponding tears at arthroscopy. In many other patients with a grade 4 or 3 meniscus in whom a corresponding tear was not found arthroscopically, meniscal tears at other sites or other abnormalities were correctly diagnosed using MR. A majority of the false-positive MR images involved the posterior horns of the menisci, the sites of most false-negative arthroscopic diagnoses. The predictive value of a negative MR image was almost 100%. Even in patients with moderate-to-large effusions, the menisci were accurately evaluated. The results imply that MR imaging is useful in the preoperative evaluation of suspected meniscal tears.

Traumatic injury to the knee remains a diagnostic and therapeutic challenge. Magnetic resonance imaging (MRI) has been applied to musculoskeletal pathoanatomy and has been shown to be an effective tool for definition and characterization of knee pathology. A systematic approach is taken to establish anatomical and pathoanatomical correlations, as well as the role of MRI in the management of knee injuries. Imaging was performed at the UCLA Medical Center using a permanent magnet system and a combination of solenoidal surface coils and thin-section, high-resolution scanning techniques. Images depict structural anatomical and spatial details of the knee that correlate well with corresponding cadaveric cryosections. To determine pathoanatomical correlations and the efficacy of MRI, 105 patients with preoperative diagnoses of meniscal tears, anterior and posterior cruciate ligament tears, tibial plateau fracture, and patella and quadriceps injuries were imaged. Results indicated that for the medial meniscus MRI demonstrated a 95.7% sensitivity, 81.8% specificity, 90% accuracy, 88.2% positive predictive value (PPV), and 93.1% negative predictive value (NPV). Imaging of the lateral meniscus demonstrated a 75% sensitivity, 95% specificity, 91% accuracy, 80% PPV, and 94% NPV. MRI of the ACL revealed 100% sensitivity, specificity, accuracy, positive and negative predictive values. MRI is a noninvasive tool which uses no ionizing radiation and can accurately define and characterize anatomy and pathoanatomy. This study indicates that MRI in conjunction with clinical evaluation can contribute to treatment decision-making processes and assist in preoperative planning. An algorithm demonstrating the potential clinical use of MRI is presented.

One hundred thirty patients with a diversity of knee joint injuries were evaluated with a high-resolution magnetic resonance (MR) imaging technique. The authors report the accuracy of this technique in the evaluation of 105 patients with suspected meniscal tears, 26 patients with suspected cruciate ligament tears, and eight patients with suspected patellar tendon injuries. Of those menisci rated as definitely or probably torn on MR imaging, 80% were found to be torn at subsequent arthroscopy. The predictive value of negative MR imaging results was 100%. MR imaging was 92% accurate in predicting the clinical outcome in patients with suspected meniscal tears who did not undergo surgery. MR permitted complete disruption of the patellar tendon to be differentiated from partial tears, ligamentous inflammation, and localized effusion of the infrapatellar bursa. Injuries to the anterior and posterior cruciate ligaments were identified on MR images, and the status of synthetic grafts of the anterior cruciate ligament was ascertained.

No abstract

Sixty patients with symptoms of chronic disease of the knee joint were evaluated with high-resolution, thin-section magnetic resonance (MR) imaging. MR imaging depicted a wide variety of knee joint abnormalities including osteochondritis dissecans, medullary infarcts, epiphyseal osteonecrosis, intraarticular osteochondral fragments, synovial cysts, joint effusions, intraarticular soft-tissue tumors, synovial disease, leukemic infiltration of bone marrow, Osgood-Schlatter disease, and nonossifying fibroma. In two cases MR imaging depicted bone infarcts not seen on both radionuclide bone scans and standard radiographs. The highly detailed depiction of the articular cartilage was of particular importance in predicting arthroscopic findings in cases of osteochondritis dissecans. In two cases, a soft-tissue mass (pigmented villonodular synovitis) and a large osteochondral fragment undetected at arthroscopy were accurately localized with MR imaging. The results indicate that MR imaging is capable of providing information that might otherwise require multiple, sometimes invasive diagnostic procedures.