Abstract
Introduction and purpose: Magnetic resonance imaging is frequently used in the diagnosis of knee conditions. We carried out a retrospective study to assess the correlation that exists between MRI and arthroscopic findings during diagnosis of meniscal and cartilage lesions. Arthroscopy is considered the ‘gold standard.’
Materials and methods: Between January 2006 and February 2007 we studied 72 patients with clinical symptoms compatible with pathological meniscal conditions. We carried out one MRI and subsequently performed arthroscopy. Mean age of the patients was 49 years (range: 10–77), 38 were men (53%) and 34 women (47%). We assessed sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of MRI using 4 lesion locations: medial meniscus, lateral meniscus, patellofemoral cartilage, tibiofemoral cartilage (both compartments together).
Results: Diagnostic correlation was seen to be: 85.9% for the medial meniscus (sensitivity 96%, specificity 62%, PPV 86%, NPV 87%); for the lateral meniscus 81.25% (sensitivity 87%, specificity 95%, PPV 81%, NPV 96%); for the patellofemoral cartilage 68.2% (sensitivity 68%, specificity 93%, PPV 93%, NPV 65%) and for the tibiofemoral cartilage 36.8% (sensitivity 63%, specificity 98%, PPV 87.5%, NPV 81%).
Conclusions: Greater correlation is seen in the case of meniscal lesions than in the case of chondropathies. With reference to meniscal lesions MRI shows greater sensitivity in lesions of the medial meniscus and greater specificity in lesions of the lateral meniscus. A normal MRI rules out, with a high degree of certainty, any meniscal injury, but if the MRI shows abnormalities, this does not mean that there is actually a lesion. On the other hand, an MRI that does not show cartilage lesions has a high probability of error; there may well be grade I or II chondropathies, especially those that are incipient and difficult to detect. If the MRI shows cartilage abnormalities, the lesion is sure to exist, although MRI staging does not usually coincide with arthroscopic findings, especially in the case of incipient lesions.
The abstracts were prepared by E. Carlos Rodríguez-Merchán, Editor-in-Chief of the Spanish Journal of Orthopaedic Surgery and Traumatology (Revista de Ortopedia y Traumatología). Correspondence should be addressed to him at: Sociedad Española de Cirugía Ortopédica y Traumatología, calle Fernández de los Ríos 108, 28015-Madrid, Spain