Influence of 1.5-Tesla intraoperative MR imaging on surgical decision making.

To determine the frequency that high-field magnetic resonance (MR) imaging sequences influenced surgical decision making during intraoperative MR-guided surgery. From January 1997 to February 2001, 346 MR-guided procedures were performed using a 1.5-Tesla MR system (NT-ACS, Philips Medical Systems). This system can perform functional MR imaging (fMRI), diffusion weighted imaging (DWI), MR spectroscopy (MRS), MR angiography (MRA), and MR venography (MRV) in addition to T1-weighted, T2-weighted, and turbo FLAIR (fluid-attenuated inversion recovery) imaging. FMRI was used to determine areas of brain activation for language, motor function, and memory. DWI was utilized after tumor resection to exclude cerebral ischemia or infarction. MRS was obtained to identify areas of elevated choline that were suspected to correlate with tumor presence. MRA and MRV localized vascular structures adjacent to tumors prior to resection. The intraoperative procedures performed included 140 brain biopsies of which 82 utilized a trajectory guide and prospective stereotaxy. MRS was used in 42 biopsies (30%), of which 29 had turbo spectroscopic imaging (TSI) and 21 had single voxel spectroscopy (SVS). In all biopsy cases, diagnostic tissue was obtained. There were 103 tumor resections of which 18 (17%) had MRS. Functional MRI was used in 17 cases; 3 biopsies (2%) and 14 planned resections (14%). Speech function was localized in 3 cases, memory function in 3, and motor function in 11. In one case where the motor function of the tongue was intimately involved with a low-grade glioma, resection was not attempted. DWI was used in less than 10% of tumor resections. MRA and MRV were performed in 3 (3%) and 2 (2%) of tumor resections, respectively. The imaging capabilities (i.e., fMRI, DWI, MRA, MRV) associated with high-field intraoperative MR influenced surgical decision making primarily for tumor resections. MRS influenced target selection during brain biopsy.