Intraoperative magnetic resonance imaging for resection of intra-axial brain lesions: a decade of experience using low-field magnetic resonance imaging, Polestar N-10, 20, 30 systems.

Objective: The aim of this study was to determine the utility of an intraoperative magnetic resonance imaging (i-MRI) system (Polestar N-10, 20, 30) in achieving maximal resection of intra-axial brain lesions.

Methods: The subjects comprised 163 patients with intra-axial brain lesions who underwent resection at Sheba Medical Center using the Polestar from February 2000 through February 2012. Demographic and imaging data were obtained and analyzed retrospectively. The patients included 83 men (50.9%) and had a mean age of 43 years. High-grade gliomas were diagnosed in 72 patients, low-grade gliomas in 35, metastases in 22, and various pathologies (e.g., cavernous angiomas, juvenile pilocytic astrocytoma, pleomorphic xanthoastrocytoma, etc.) were diagnosed in 34. The majority of the lesions (84, 51.5%) were located in or near eloquent areas. Fifty-one patients had nonenhancing lesions.

Results: We intended to achieve complete resection in 110 of 163 cases, based on preoperative imaging. Complete resection was achieved in 90 of these 110 (81.8%) cases. Intraoperative MRI led to additional resection in 42.3% of the total cases and to complete resection in 43.3% of all the cases in which a complete resection was achieved. In 76.8% of these cases, 2 intraoperative scans were sufficient to achieve complete resection. Sex, age, intent of resection, recurrence, affected side, and radiologic characteristics did not differ significantly between cases in which intraoperative MRI led to further resection and cases in which it did not. Nonenhancing lesions of all types were 3 times more likely to require additional resection after obtaining intraoperative MRIs (P = .02).

Conclusions: The Polestar (N-10, 20, 30) proved useful for evaluating residual intra-axial brain lesions and achieving the maximal extent of resection in 42.3% of the total cases and in 43.3% of cases in which complete resection was achieved. Intraoperative MRI led to extended resection in 46.9% of patients for whom the initial intent was to perform an incomplete resection. Nonenhancement was the only independent variable predicting the usefulness of intraoperative MRI for additional lesion resection.