Article
nTMS-based DTI fiber tracking of language pathways in brain tumor patients: a feasibility study
Search Medline for
Authors
Published: | June 2, 2015 |
---|
Outline
Text
Objective: To describe and assess the feasibility of a new approach to DTI fiber tracking (DTI-FT) of language pathways based on navigated transcranial magnetic stimulation (nTMS) in patients affected by language-eloquent brain tumors.
Method: We collected neuroimaging data of 10 patients (6 m, 4 f, age 57 ± 14) suffering from left perisylvian brain tumors (9 malignant glioma, 1 hemangioma) that underwent language mapping using repetitive nTMS. Consequently, the following approach to DTI-FT of language pathways was applied:
- 1.
- standard: DTI-FT of arcuate fasciculus (AF), inferior longitudinal fasciculus (ILF), inferior fronto-occipital fasciculus (IFOF) and uncinate fasciculus (UF) was performed by using knowledge-based technique for defining two seeding regions of interest (ROI). In addition each fascicle was then converted into a 3D object to be used as 1st ROI for the nTMS-based DTI-FT (#2,3)
- 2.
- nTMS "all-spots": nTMS-based DTI-FT was performed by using all nTMS positive spots as 2nd ROI
- 3.
- nTMS "single-spot": each single nTMS positive spot was individually selected as 2nd ROI.
We compared the fractional anisotropy (FA) value and ratio (FA value/FA threshold) needed for the visualization of each fascicle by using the standard vs the nTMS-based all-spots technique. Also, we analyzed the ability of the three techniques in computing fibers connecting cortical nTMS spots to any of the main pathways.
Results: AF, ILF, IFOF and UF were computed in all patients by all three techniques. Mean FA values and ratios were higher when using the "all-spots" strategy compared to the standard technique. Difference was significant comparing FA ratio for AF, ILF and UF (p=0.006, p=0.002; p=0.04). We obtained a significantly higher number of cortical spots connected to the main fascicles comparing the "all-spots" to the standard method (AF p=0.0003, ILF p=0.0001, IFOF p=0.0001, UF p=0.0003, all p=0.02), and the "single-spot" to the "all-spots" strategy (AF p=0.02, ILF p=0.05, IFOF p=0.03, UF p=0.03, all p=0.009).
Conclusions: nTMS-based DTI-FT of language pathways allows for standardized computation of AF, ULF, IFOF and UF using higher FA values than standard DTI-FT, thus reducing the possibility of computing aberrant fibers. Especially the "single-spot" strategy enables reliable detection of connections between cortical language areas and subcortical language pathways.