Pitfalls and limitations of Magnetic Resonance Imaging in bucket-handle tears of knee menisci.

PURPOSE: The aim of the present study was to assess the accuracy of magnetic resonance imaging (MRI) in the diagnosis of bucket-handle tears of the knee menisci, determining the sensitivity and specificity of the imaging for each of the signs typical of these tears.
MATERIALS AND METHODS: MR examinations of 495 patients suffering from knee traumas were assessed retrospectively. Forty-eight patients had arthroscopically-proven bucket-handle tears of the meniscus. MR examinations were performed using a 0.5-T superconducting magnet, with T1-weighted spin-echo (SE) and gradient-echo (GE) T2*-weighted sequences. Imaging findings used for the diagnosis were: a) double posterior cruciate ligament sign; b) flipped meniscus sign; c) presence of a displaced fragment of the meniscus in the intercondylar notch, visible in coronal and axial images; d) truncated triangular shape of the peripheral non-displaced portion of the meniscus, visible in coronal images.
RESULTS: A total of 43 out of 48 bucket-handle tears of the meniscus were correctly diagnosed at MR, thus overall MR accuracy was 98%. In 12 (28%) cases three signs were present simultaneously the double posterior cruciate ligament (sensitivity 28%, specificity 99%, accuracy 93%) + the displaced fragment in the intercondylar notch (sensitivity 69.8%, specificity 98.7%, accuracy 96.2%) + the truncated triangular shape of the peripheral portion of the meniscus (sensitivity 74.4%, specificity 98%, accuracy 96%). In 18 (42%) cases two signs were present together the displaced fragment of the meniscus + the truncated triangular shape of the peripheral portion of the meniscus. In 13 (30%) cases only one sign was presenting two cases the truncated triangular shape of the peripheral portion of the meniscus and in 11 cases the flipped meniscus sign (sensitivity 25.6%, specificity 93.4%, accuracy 87.5%). No statistically significant differences were found comparing the results for tears of the medial meniscus with those for the lateral meniscus.
CONCLUSIONS: MR imaging is highly accurate in diagnosing bucket-handle tears of the menisci due to its ability to identify a displaced fragment of the meniscus in the intercondylar notch or flipped over the anterior horn of the meniscus of origin. We speculate that bucket-handle tears not found by MR imaging are cases where the meniscus was displaced after MR examination.