Article
Three-dimensional reconstruction of cranial nerve nuclei and coregistration with MRI
Search Medline for
Authors
-
J. Rachinger
- Klinik für Neurochirurgie, Universitätsklinikum Halle
-
A. Beißwenger
- Klinik für Neurochirurgie, Universitätsklinikum Halle
-
D. Stoevesandt
- Klinik für Diagnostische Radiologie, Universitätsklinikum Halle
-
E. Peschke
- Institut für Anatomie und Zellbiologie, Martin-Luther-Universität Halle-Wittenberg
-
C. Strauss
- Klinik für Neurochirurgie, Universitätsklinikum Halle
-
S. Rampp
- Klinik für Neurochirurgie, Universitätsklinikum Halle
| Published: | June 4, 2012 |
|---|
Outline
Text
Objective: Intraoperative identification of motor areas of the rhomboid fossa is routinely performed by direct electrical stimulation, but is not possible for all cranial nerve nuclei at risk. Furthermore, there is no method for visualizing, the nuclei in relation to the surgical approach and the lesion to be operated upon. To facilitate intraoperative orientation and provide safe access to the brainstem, a cadaver study with 3D-reconstruction of relevant cranial nerve nuclei and co-registration with MRI were performed.
Methods: We used three human brainstems, which were cut into serial slices of 50 µm thickness. Staining was done with cresyl violet. All slices were digitized with a high-resolution scanner and imported into a dedicated graphics workstation. Cranial nerve nuclei were segmented slice-by-slice using the workstation and a microscope to validate correct identification of cell nuclei. Cranial nerve nuclei and the volume of the brainstem were three-dimensionally reconstructed. Shrinkage due to fixation and preparation of the specimens was evaluated by CT- and MRI studies at different stages of the process as well as by calliper measurements. Brainstem volume with cranial nerve nuclei was co-registered to a MRI-dataset.
Results: The dorsal motor nucleus of the vagus, the nuclei of the cranial nerves VI, VII and XII and the motor nucleus of the trigeminal could be clearly identified, reconstructed three-dimensionally and measured on both sides. Definite identification of N. ambiguus turned out to be more difficult. Shrinkage analysis allowed for an adequate correction of the size of the reconstructed brainstem model. Co-registration with MRI-datasets was possible and provided an individual 3D-model of the brainstem containing relevant cranial nerve nuclei.
Conclusions: Individualized, three-dimensional modelling of the brainstem and clinically relevant nuclei combined with frameless stereotaxy has the potential for becoming a useful tool in brainstem surgery by providing substantial spatial information in addition to intraoperative, electrophysiological findings.
