J Neurol Surg A Cent Eur Neurosurg 2020; 81(01): 001-009
DOI: 10.1055/s-0039-1691822
Original Article
Georg Thieme Verlag KG Stuttgart · New York

Diffusion Tensor Imaging for Surgical Planning in Patients with Cervical Spondylotic Myelopathy

Karsten Schöller*
1  Department of Neurosurgery, Justus-Liebig-University Gießen, Gießen, Germany
2  Department of Neurosurgery, University Hospital, Ludwig-Maximilians-University Munich, Munich, Germany
Christian Brem
3  Institute of Neuroradiology, University Hospital, Ludwig-Maximilians-University Munich, Munich, Germany
Jürgen Lutz
4  Institute of Neuroradiology, Ingolstadt Hospital, Ingolstadt, Germany
Stefan Zausinger
2  Department of Neurosurgery, University Hospital, Ludwig-Maximilians-University Munich, Munich, Germany
› Author Affiliations
Further Information

Publication History

22 November 2018

04 March 2019

Publication Date:
10 June 2019 (online)


Background and Study Aims/Objective Despite its invasiveness, computed tomography myelography (CTM) is still considered an important supplement to conventional magnetic resonance imaging (MRI) for preoperative evaluation of multilevel cervical spondylotic myelopathy (CSM). We analyzed if diffusion tensor imaging (DTI) could be a less invasive alternative for this purpose.

Material and Methods In 20 patients with CSM and an indication for decompression of at least one level, CTM was performed preoperatively to determine the extent of spinal canal/cerebrospinal fluid (CSF) space and cord compression (Naganawa score) for a decision on the number of levels to be decompressed. Fractional anisotropy (FA) and apparent diffusion coefficient (ADC) were correlated with these parameters and with MRI-based increased signal intensity (ISI). Receiver operating characteristic analysis was performed to determine the sensitivity to discriminate levels requiring decompression surgery. European Myelopathy Score (EMS) and neck/radicular visual analog scale (VAS-N/R) were used for clinical evaluation.

Results According to preoperative CTM, 20 levels of maximum and 16 levels of relevant additional stenosis were defined and decompressed. Preoperative FA and particularly ADC showed a significant correlation with the CTM Naganawa score but also with the ISI grade. Furthermore, both FA and ADC facilitated a good discrimination between stenotic and nonstenotic levels with cutoff values < 0.49 for FA and > 1.15 × 10−9 m2/s for ADC. FA and especially ADC revealed a considerably higher sensitivity (79% and 82%, respectively) in discriminating levels requiring decompression surgery compared with ISI (55%). EMS and VAS-N/R were significantly improved at 14 months compared with preoperative values.

Conclusion DTI parameters are highly sensitive at distinguishing surgical from nonsurgical levels in CSM patients and might therefore represent a less invasive alternative to CTM for surgical planning.

Previous Presentations

* These two authors contributed equally to the article and should be considered co-first authors.