BACKGROUND AND PURPOSE: To assess the magnetic resonance (MR) imaging characteristics of spinal intramedullary ependymomas. METHODS: MR images were obtained from 60 patients with spinal intramedullary ependymomas. All patients were treated surgically, and pathological diagnoses were confirmed. MR images of ependymomas and associated cysts were correlated with surgical findings. RESULTS: Among 60 cases of intramedullary ependymomas, 39 were cervical and 11 thoracic, and 10 were located at the conus medullaris and filum terminale. The lesions measured 1 to 10 vertebral segments in length along the neuraxis, with a mean length of 3.7 segments. All tumors had slightly hyperintense signals on T2-weighted MR images. Clear tumor margins were seen in 77% of patients; 75% had uniform contrast enhancement. Rostral and caudal cysts were seen in 90% of all patients; all cysts were hypointense on T1-weighted images and hyperintense on T2-weighted images. The rostral cysts were 1 to 5 segments in length and caudal cysts 2 to 10 segments in length. CONCLUSION: Ependymomas occur most often at the upper cord rather than in the conus medullaris and filum terminale. Of the cervical and thoracic ependymomas, most were cellular or epithelial types. Papillary ependymomas occurred exclusively in the conus medullaris and filum terminale. Rostral and caudal cysts are frequently associated with intramedullary ependymomas. Clear tumor margins, more uniform enhancement, and central locations can help differentiate ependymomas from other intramedullary spinal cord tumors.
BACKGROUND AND PURPOSE: To assess the magnetic resonance (MR) imaging characteristics of spinal intramedullary ependymomas. METHODS: MR images were obtained from 60 patients with spinal intramedullary ependymomas. All patients were treated surgically, and pathological diagnoses were confirmed. MR images of ependymomas and associated cysts were correlated with surgical findings. RESULTS: Among 60 cases of intramedullary ependymomas, 39 were cervical and 11 thoracic, and 10 were located at the conus medullaris and filum terminale. The lesions measured 1 to 10 vertebral segments in length along the neuraxis, with a mean length of 3.7 segments. All tumors had slightly hyperintense signals on T2-weighted MR images. Clear tumor margins were seen in 77% of patients; 75% had uniform contrast enhancement. Rostral and caudal cysts were seen in 90% of all patients; all cysts were hypointense on T1-weighted images and hyperintense on T2-weighted images. The rostral cysts were 1 to 5 segments in length and caudal cysts 2 to 10 segments in length. CONCLUSION:Ependymomas occur most often at the upper cord rather than in the conus medullaris and filum terminale. Of the cervical and thoracic ependymomas, most were cellular or epithelial types. Papillary ependymomas occurred exclusively in the conus medullaris and filum terminale. Rostral and caudal cysts are frequently associated with intramedullary ependymomas. Clear tumor margins, more uniform enhancement, and central locations can help differentiate ependymomas from other intramedullary spinal cord tumors.
Authors: Tuncay Kaner; Mehdi Sasani; Tunc Oktenoglu; Bilgehan Solmaz; Ali Cetin Sarloglu; Ali Fahir Ozer Journal: J Korean Neurosurg Soc Date: 2010-02-28
Authors: Andrew T Parsa; Janet Lee; Ian F Parney; Philip Weinstein; Paul C McCormick; Christopher Ames Journal: J Neurooncol Date: 2004 Aug-Sep Impact factor: 4.130