Article
Comparison of nucleus subthalamicus targeting by T2-, FLAIR- and SWI-3-tesla MRI confirmed by microelectrode recordings
Search Medline for
Authors
-
Witold H. Polanski
- Klinik und Poliklinik für Neurochirurgie, Medizinische Fakultät Carl Gustav Carus der Technischen Universität Dresden
-
Kay Engellandt
- Abteilung für Neuroradiologie, Medizinische Fakultät Carl Gustav Carus der Technischen Universität Dresden
-
Klaus D. Martin
- Klinik und Poliklinik für Neurochirurgie, Medizinische Fakultät Carl Gustav Carus der Technischen Universität Dresden
-
Lisa Klingelhoefer
- Klinik und Poliklinik für Neurologie, Medizinische Fakultät Carl Gustav Carus der Technischen Universität Dresden
-
Mareike Fauser
- Klinik und Poliklinik für Neurologie, Medizinische Fakultät Carl Gustav Carus der Technischen Universität Dresden
-
Alexander Storch
- Klinik und Poliklinik für Neurologie, Medizinische Fakultät Carl Gustav Carus der Technischen Universität Dresden
-
Gabriele Schackert
- Klinik und Poliklinik für Neurochirurgie, Medizinische Fakultät Carl Gustav Carus der Technischen Universität Dresden
-
Stephan B. Sobottka
- Klinik und Poliklinik für Neurochirurgie, Medizinische Fakultät Carl Gustav Carus der Technischen Universität Dresden
| Published: | May 13, 2014 |
|---|
Outline
Text
Objective: Successful deep brain stimulation (DBS) into the subthalamic nucleus (STN) for Parkinson’s disease is mostly dependent on an accurate position of the leads at the optimal target point. Recent studies showed, that the STN identification by MRI may differ from the physiological STN, which can be determined by intraoperative microelectrode recording (MER) and intraoperative neurological examination. We compared the accuracy for the identification of the STN in T2-weighted 3 tesla (3T)-MRI, fluid attenuated inversion recovery (FLAIR)-3T-MRI and susceptibility weighted image (SWI)-3T-MRI, which was additionally confirmed by intraoperative neurological examination and MER.
Method: 21 Patient underwent a bilateral STN-DBS in local anesthesia with intraoperative MER and neurological examination. 182 MER leads were implanted, consequently, 728 electrode contact positions in T2-weighted 3T-MRI, 552 electrode contact positions in FLAIR-3T-MRI and 490 electrode contact positions in SWI-3T-MRI were evaluated for a positive STN signal. After lead placement, preoperative MRI was merged with postoperative computer tomography scans (CT) to simulate the MER lead positions and to evaluate, whether they were located in the MRI-determinated STN. Medtronic lead configuration (model 3389) was used as reference.
Results: The highest sensitivity was measured for FLAIR-3T-MRI with 82.5%, while the highest specificity was observed for SWI-3T-MRI with 90.6%. The negative predictive value (npv) was nearly equal for SWI and FLAIR with 87.5% vs. 87.1% but the positive predictive value was higher in SWI-3T-MRI (86.0%) than in the other MRI sequences. The hit rates for STN targeting were markedly dependent on the electrode position. In every patient, at least 3 MER positions showed a positive STN signal. In central position, only 39.4% showed a positive MER signal. In contrast, in MRI, the hit rates were higher with 55.3% in T2-weighted MRI, 53.2% in FLAIR 3T MRI and 48.2% in SWI 3T MRI. The highest hit rate was measured in anterior and lateral electrode position.
Conclusions: The SWI-3T-MRI based STN localization shows the best accuracy compared with T2-weighted and FLAIR-3T-MRI, but appears to be insufficient for isolated determination of the optimal target. Further, the anatomical STN, determined by MRI, seems to differ from the physiological STN which was determined by MER. Therefore, we recommend using MER in addition to high resolution MRI to safely target the STN for DBS.
