Artikel
Identification of different resting state networks under anesthesia in neurosurgical patients operated under intraoperative MRI control
Suche in Medline nach
Autoren
-
Hussam Metwali
- INI Hannover, Neurochirurgie, Hannover, Deutschland
-
Mathjis Raemaekers
- Brain Center Rudolf Magnus, University Medical Center Utrecht, Utrecht, Niederlande
-
Nic Ramsey
- Brain Center Rudolf Magnus, University Medical Center Utrecht, Utrecht, Niederlande
-
Amir Samii
- INI Hannover, Neurochirurgie, Hannover, Deutschland
| Veröffentlicht: | 18. Juni 2018 |
|---|
Gliederung
Text
Objective: The resting state functional magnetic resonance imaging (rs-fMRI) is a method for evaluation of the functional connectivity of the brain and it was recently introduced in neurosurgical practice. In this study, we are presenting the possibility of reproducing different resting state networks in rs-fMRI performed during neurosurgical procedures.
Methods: Forty patients were operated in our institution for cerebral tumor resection under 1.5 Tesla intraoperative MRI control (Magnetom Espree Siemens AG Medical Solution, software NUMARIS/4 version syngo MR B17). A rs-fMRI EPI sequence (TR 2200 msec, TE 30 msec, voxel size 3x3x3mm, number of slices 30, interleaved slice order, number of volumes 178) together with high-resolution anatomical imagining (T1) were performed before the intervention and repeated at the end of surgery. The images were preprocessed (realignment, slice time correction, co-registration, normalization, segmentation, and smoothing). We performed single subject analysis (ROI-based) and group-based analysis (ICA) using CONN toolbox and GIFT toolbox, respectively.
Results: The motor network was present in 37 patients. The visual network was present in 35 patients. The language network was present in 30 patients. The connectivity (z-score) of each network showed however significant variation between patients which was not correlated to the preoperative clinical status. Nevertheless, the connectivity of the network stayed stable until the end of the procedure. The sensory network was present in 35 patients. In the ICA group analysis, the default mode network showed a disconnection between the frontal cortex and the parietal cortex. Salience network showed decreased connectivity to the cingulate gyrus. We could not detect the attention and executive networks.
Conclusion: Identification of different networks is possible under anesthesia and can be used for operative planning and monitoring of the neurological functions. The variability of the reproducibility of the resting networks between patients requires further investigations.
