Abstract

The purpose of the present study was to implement a unique low-field open magnetic resonance scanner for perioperative imaging in neurosurgery.

A paradigm was created for joint intraoperative/interventional MRI, including premises, surgical practice and an operational model. The feasibility of the paradigm was tested in clinical work. The joint use of the facilities between the Departments of Neurosurgery and Diagnostic Radiology was found to enhance the economic rationale and provide for perioperative imaging. It was also found to be organizationally viable in the long run.

Intraoperative MRI was implemented and studied in connection with neuronavigation and other intraoperative instruments, tools and imaging modalities. The unique shut down possibility of the magnet enabled staged operating-imaging practice, use of non-MRI-compatible instruments and devices, multimodal imaging with navigation, and avoidance of safety risks associated with operating in magnetic fringe fields.

Two dynamic contrast enhanced MR imaging sequences, which used undersampled projection reconstruction, were implemented in the low-field scanner. The applicability of these imaging sequences to follow contrast enhancement of meningiomas was studied in laboratory experiments and in two patient cases. The laboratory experiments showed a nearly linear response in signal intensity to the concentration of gadopentetate dimeglumine in purified water up to 1.25 mM. The patient cases showed results consistent with an earlier study performed at high-field strength.

The potential of low-field MRI study including dynamic contrast enhanced imaging to predict surgical and histopathologic characteristics of meningiomas was studied in a series of 21 patients. Dynamic contrast enhanced imaging could be used to evaluate microvessel densities of meningiomas. Surgical bleeding, blood loss during operation, progesterone receptor expression and collagen content were statistically best correlated to the relative intensity of meningioma on FLAIR images. Tissue hardness correlated best with relative intensity on T2-weighted images.