Intraoperative MR Imaging: Making an Impact on Outcomes for Patients with Brain Tumors
======================================================================================

* Mitchel S. Berger

There is little question that the extent of tumor removal significantly impacts outcome for the majority of patients with brain tumors, especially gliomas. Notwithstanding, it is still a considerable feat to achieve a complete (100%) or radical (> 95%) radiographic tumor resection despite the advances that have been made in the surgeon's armamentarium. Two articles in this issue of the *AJNR* bring us closer to achieving the desired surgical result, thanks to advances with intraoperative MR (iMR) imaging.

The study by Schneider et al (page 89), describes the utility of a 0.5-T vertical “doughnut” configuration magnet; the surgeon stands within a 58-cm space between the magnets. The main advantage of this system is that the operating table is stationary, and does not require movement into the magnet. Likewise, the magnet does not have to move into the physical space of the patient. As with any iMR imaging system, requirements for a successful procedure include MR-compatible surgical and anesthetic equipment, proper room shielding, and suitable head coils. This iMR imaging system also uses the technique of interactive, image-guided surgery to allow surgical navigation during the procedure with the Flashpoint Position Encoder and the MR Track Pointer. The disadvantages of this intraoperative imaging system include a small space to work in, a relatively long image acquisition time, and a magnetic field strength that makes intraoperative functional or metabolic imaging difficult, at best.

Despite the potential limitations, Schneider et al achieved complete, or nearly complete, resection of low-grade gliomas in 11 of 12 patients by use of the feedback they received as intraoperative images were obtained. Although this is a spectacular result, one large caveat remains; namely, the inability to avoid surgical morbidity with anatomic images alone. In other words, the use of direct physiological stimulation mapping of functional (eg, motor, sensory, language) pathways cannot be replaced with iMR imaging. iMR imaging must be used in conjunction with the fundamental principles of functional brain mapping in order to achieve radical resections with the least morbidity. Assessing intraoperative complications, such as swelling or bleeding, is also critical in achieving the best outcome possible for patients with brain tumors.

Although surgeons recognize the value iMR imaging adds to the surgical procedure, a potentially dangerous problem exists with spurious contrast enhancement leaking into the resection margin. This has made it difficult to assess adequately the true extent of resection in lesions that preoperatively enhance with contrast agents. In the study by Knauth et al (page 99), the authors report the novel use of monocrystalline iron oxide nanoparticles, or MIONs, to bypass this problem. MIONs are stored within glioma cells for longer periods than they circulate in the blood. This creates a window of opportunity to avoid surgically induced leakage of contrast enhancement. The result of this finding, as elegantly described by the investigators, is that intracellular storage of MIONs may yield an excellent means by which a tumor can be enhanced on preoperative imaging and at the time of surgical resection without causing false leakage during intraoperative imaging. The latter could result in unnecessary resection of tissue that does not contain tumor. Thus, MIONs may be the ideal contrast agent for high-grade gliomas. As the authors readily admit, a limitation to imaging low-grade gliomas may exist, and it is not known whether these tumors will be able to undergo endocytosis of MION particles. Because a disrupted blood-brain barrier is not essential to MION-induced enhancement of tumors, this strategy could solve the problem of spurious intraoperative MR signals simulating residual tumor caused by edema or microhemorrhage within the resection margin.

It certainly appears that iMR imaging is here to stay and will make a significant impact on patient outcome. Nonetheless, a number of issues must be resolved, including the type of magnet system (ie, low vs high field strength), ease of imaging during a complicated operative procedure, and the need to generate intraoperative functional and metabolic data. Neurosurgeons must realize that input from neuroradiologists will be as important intraoperatively for interpreting these findings as it is extraoperatively. One thing is certain—excitement and widespread optimism have been created by knowing that iMR imaging is now a reality.

*   Copyright © American Society of Neuroradiology