PURPOSE

MRI evaluation of torn lateral meniscus was compared with arthroscopy. This study calculates the sensitivity, specificity, and accuracy of MRI in determining the presence or absence of discoid lateral meniscus (DLM) for different tear types.

METHODS

MR imaging of 156 knees with arthroscopically confirmed lateral meniscus tears was analysed. There were 78 knees (70 patients) in non-DLM group and 78 knees (74 patients) in DLM group on arthroscopy as the reference standard. The presence of DLM on MRI was determined by an orthopaedic surgeon and a radiologist, who were blinded to the arthroscopic findings. The presence of discoid meniscus on MRI was determined by coronal and sagittal measurements, considering the tear pattern of lateral meniscus. The tear pattern was categorized into six types based on arthroscopic findings: horizontal, longitudinal, radial, combined radial, degenerative, and complex tear. The sensitivity, specificity, and accuracy of MRI were calculated for each type of lateral meniscus tear. In addition, we analysed the reason for non-detection of discoid meniscus on preoperative MRI.

RESULT

The sensitivity for determining the presence of discoid meniscus was 58% for radial tear, 57% for combined radial tear, and 65% for longitudinal tear, whereas the specificity was 100% for all tear groups. In the presence of radial or longitudinal tear, the accuracy of MRI was significantly lower than having no radial and longitudinal tear (p < 0.001). The presence of discoid meniscus was not recognized on MRI because of large radial tear (12 knees), deformed bucket-handle tear (6 knees), and inverted flap tear (3 knees).

CONCLUSIONS

MRI was not successful in determining the presence or absence of DLM in radial tear, combined radial tear, and longitudinal tear. When there are large radial tear, deformed bucket-handle tear, and inverted flap tear in lateral meniscus, it is recommended to consider the possibility of DLM. This information can help to make accurate diagnosis of DLM, which allows appropriate surgical planning and facilitates patient's information on poor prognosis of DLM.

LEVEL OF EVIDENCE

Level I.