Detection of meniscal tears and marrow lesions using coronal MRI

The usefulness of coronal images in addition to sagittal images for detection of additional meniscal injuries or bone lesions has been questioned. Some authors believe meniscal tears are rarely seen only in the coronal plane. We performed a retrospective review of coronal and sagittal MR images of the knee to determine whether coronal imaging resulted in the detection of any additional meniscal tears or bone lesions when compared with sagittal MRI of the knee alone.

Two musculoskeletal radiologists retrospectively reviewed 200 consecutive sets of MR images of the knee by consensus. Both observers recorded their retrospective findings on sagittal proton density images, sagittal and coronal T2-weighted images, and coronal T1-weighted images. Findings recorded were those of meniscal tears, bone marrow abnormalities, and bone lesions.

On these 200 consecutive sets of MR images of the knee, 114 meniscal tears were shown. Ninety-three meniscal tears were shown on sagittal proton density images only. One hundred fourteen meniscal tears were seen on sagittal proton density and coronal T1- and T2-weighted images. Use of coronal images resulted in the confident detection of 21 additional meniscal tears not well seen on sagittal proton density images alone. Twelve of the 21 additional meniscal tears were seen on coronal T1- and T2-weighted images. Nine of the 21 additional meniscal tears were seen only on coronal T1-weighted images. Most of these additional meniscal tears were in the body of the meniscus. These 21 additional meniscal tears were confirmed arthroscopically. Eight marrow lesions were well characterized only on coronal T1-weighted images. Five were osteochondral lesions, one was an intraosseous hemangioma, and two were microfractures.

Coronal MR images of the knee allowed better detection and characterization of some meniscal tears than sagittal images alone. Radial meniscal tears, bucket-handle tears, and horizontal tears in the body of the meniscus may be difficult to characterize in the sagittal plane alone. Use of coronal T1-weighted images, rather than coronal T2-weighted or sagittal proton density images alone, allows accurate characterization of some additional marrow lesions. The addition of a T1-weighted sequence in the coronal plane adds only 1 min 30 sec to the scanning time.