[Stereotactic mapping for radiosurgical treatment of vestibular schwannomas].

RATIONALE As an exclusively image-guided surgery method, radiosurgery requires special attention in the choice of imaging modalities and acquisition parameters must be set with extreme care. METHODS Quality control for resolution and accuracy of computed tomography (CT) scanners must be performed. Magnetic resonance imaging (MRI) distortions should be limited through magnetic field homogeneity adjustment (shimming) and acquisition parameters optimization. These inaccuracies should then be quantified through systematic combination of MRI and CT in the radiosurgery planning system. MRI pulse sequences selection criteria are defined by their ability to delineate tumor contrast enhancement and to image cranial nerves and vessels relative arrangement in the cistern and canal. Topography of the petrous structures, such as cochlea, vestibulum and facial nerve canal should be visible. Exact definition of real extension of the lesion at the end of the canal may require specific technical solutions. These technical requirements must be balanced depending on the lesion Volume staging (Koos), the treatment history (microsurgery), the clinical condition (hearing quality), the pathological context (NF2) or the age of the patient. RESULTS T1-weighted Volumetric MRI pulse sequences (3D-T1) show a contrast enhanced signal that is useful for both the pons interface delineation in Koos III cases, and the canal ending in Ohata A and B. On the other hand, 3D-T1 introduce inaccuracies from magnetic susceptibility distortions and partial Volume effects. High resolution CISS T2-weighted Volumetric pulse sequences (3D-T2) give superior stereotaxic definition attributable to their better resolution (half a millimeter) minimizing partial Volume effects and to their lower magnetic susceptibility minimizing distortions. 3D-T2 allows direct nerve visualization. Moreover, this pulse sequence with contrast injection, show improved distinction between the pons and the nerves due to signal differences within the schwannomas. Fat saturation pulse sequences are of interest in post-microsurgery conditions. CONCLUSIONS Radiology phase quality is critical and its complexity requires a high commitment to obtain satisfactory clinical results. Solelt the 3D-T1 MRI modality seems to us not to comply to minimum security criteria.