Intraspinal Tumors: The Value of Diffusion Tensor Imaging (DTI) and Tractography: A Preliminary Study

Aims: The aim of this prospective study was to evaluate the value of diffusion tensor imaging (DTI) in intraspinal cord tumors. Despite the advances made in microsurgical neurosurgery, operations of spinal cord tumors remain challenging. Diffusion-weighted imaging is a magnetic resonance (MR) technique that has seen vast application with many efforts undertaken to apply the technique in spinal cord pathologies.

Methods: DTI was performed in 31 patients with intraspinal lesions. Patients underwent MR imaging (MRI) on a 3.0T magnet (Verio, Siemens). A standardized MR imaging protocol for the cervical spine was used including T2 and T1-weighted as well as gadolinium-enhanced T1-weighted imaging. For DTI, a SE EPI sequence was used in two orientations with b-factors of 0 and 900 mm²/s. Three-dimensional tractography was calculated for all patients. Patients were classified into three groups according to the fiber course with respect to the lesion. Furthermore, the lesions were rated as resectable or nonresectable.

Results: Of the 31 patients (16 males, 15 females; mean age 47,7 ± 17 years, range, 19–74 years), 17 patients showed intramedullary tumors, 12 patients presented extramedullary intradural tumors, and 2 patients had epidural masses. Eight patients had ependymomas, six patients had meningiomas, three patients had cavernomas, one patient showed a ODG and another a LGG, and one patient had a GBM. One patient had a chordoma. One patient presented with an hemangioblastoma, three patients showed schwannoma, and one patient had no histology but a hemangioblastoma was suspected. In one patient, biopsy was inconclusive and imaging suspected an ependymoma. Two patients showed metastasis. One lesion was an epidural abscess with compression of the cord. One patient’s lesion was a multiple sclerosis plaque. Additionally, this patient presented with an arachnoidal cyst. The lesions could be classified into three types according to the fiber course. In type 1 (n = 14), fibers did not pass through the lesion. In type 2 (n = 8), some fibers crossed the lesion, but most of the lesion volume did not contain fibers. In type 3 (n = 10), the fibers were completely encased by tumor. Twenty tumors were considered resectable.

Conclusion: These preliminary data suggest that DTI of spinal cord tumors can be capable of showing the effect of tumors on the cord with more sensitivity than conventional MRI and thus having the potential of predicting the resectability. Further prospective studies are needed to confirm these results and effects on patient outcome.

Acknowledgment: We would like to thank Siemens Healthcare (Erlangen, Germany) for providing us with the work in progress DTI sequence package WIP #511 (author: Dr. Thorsten Feiweier).