Impact of Intraoperative Magnetic Resonance Imaging and Other Factors on Surgical Outcomes for Newly Diagnosed Grade II Astrocytomas and Oligodendrogliomas: A Multicenter Study

Neurosurgery 88 (1) 2021: 63–73,

Few studies use large, multi-institutional patient cohorts to examine the role of intraoperative magnetic resonance imaging (iMRI) in the resection of grade II gliomas.

OBJECTIVE: To assess the impact of iMRI and other factors on overall survival (OS) and progression-free survival (PFS) for newly diagnosed grade II astrocytomas and oligodendrogliomas.

METHODS: Retrospective analyses of a multicenter database assessed the impact of patient-, treatment-, and tumor-related factors on OS and PFS.

RESULTS: A total of 232 resections (112 astrocytomas and 120 oligodendrogliomas) were analyzed. Oligodendrogliomas had longer OS (P < .001) and PFS (P = .01) than astrocytomas. Multivariate analyses demonstrated improved OS for gross total resection (GTR) vs subtotal resection (STR; P = .006, hazard ratio [HR]: .23) and near total resection (NTR; P = .02, HR: .64). GTR vs STR (P = .02, HR: .54), GTR vs NTR (P = .04, HR: .49), and iMRI use (P = .02, HR: .54) were associated with longer PFS. Frontal (P = .048, HR: 2.11) and occipital/parietal (P = .003, HR: 3.59) locations were associated with shorter PFS (vs temporal). Kaplan-Meier analyses showed longer OS with increasing extent of surgical resection (EOR) (P=.03) and 1p/19q gene deletions (P=.02). PFS improved with increasing EOR (P = .01), GTR vs NTR (P = .02), and resections above STR (P = .04). Factors influencing adjuvant treatment (35.3% of patients) included age (P=.002, odds ratio [OR]: 1.04) and EOR (P=.003,OR: .39) but not glioma subtype or location. Additional tumor resection after iMRI was performed in 105/159 (66%) iMRI cases, yielding GTR in 54.5% of these instances.

CONCLUSION: EOR is a major determinant of OS and PFS for patients with grade II astrocytomas and oligodendrogliomas. Intraoperative MRI may improve EOR and was associated with increased PFS.

Combining 5-Aminolevulinic Acid Fluorescence and Intraoperative Magnetic Resonance Imaging in Glioblastoma Surgery

Combining 5-Aminolevulinic Acid Fluorescence and Intraoperative Magnetic Resonance Imaging in Glioblastoma Surgery

Neurosurgery 78:475–483, 2016

Glioblastoma resection guided by 5-aminolevulinic acid (5-ALA) fluorescence and intraoperative magnetic resonance imaging (iMRI) may improve surgical results and prolong survival.

OBJECTIVE: To evaluate 5-ALA fluorescence combined with subsequent low-field iMRI for resection control in glioblastoma surgery.

METHODS: Fourteen patients with suspected glioblastoma suitable for complete resection of contrast-enhancing portions were enrolled. The surgery was carried out using 5-ALA–induced fluorescence and frameless navigation. Areas suspicious for tumor underwent biopsy. After complete resection of fluorescent tissue, low-field iMRI was performed. Areas suspicious for tumor remnant underwent biopsy under navigation guidance and were resected. The histological analysis was blinded.

RESULTS: In 13 of 14 cases, the diagnosis was glioblastoma multiforme. One lymphoma and 1 case without fluorescence were excluded. In 11 of 12 operations, residual contrast enhancement on iMRI was found after complete resection of 5-ALA fluorescent tissue. In 1 case, the iMRI enhancement was in an eloquent area and did not undergo a biopsy. The 28 biopsies of areas suspicious for tumor on iMRI in the remaining 10 cases showed tumor in 39.3%, infiltration zone in 25%, reactive central nervous system tissue in 32.1%, and normal brain in 3.6%. Ninety-three fluorescent and 24 non-fluorescent tissue samples collected before iMRI contained tumor in 95.7% and 87.5%, respectively.

CONCLUSION: 5-ALA fluorescence–guided resection may leave some glioblastoma tissue undetected. MRI might detect areas suspicious for tumor even after complete resection of all fluorescent tissue; however, due to the limited accuracy of iMRI in predicting tumor remnant (64.3%), resection of this tissue has to be considered with caution in eloquent regions.

Intraoperative MRI and DBS for Parkinson disease


J Neurosurg 124:62–69, 2016

The degree of clinical improvement achieved by deep brain stimulation (DBS) is largely dependent on the accuracy of lead placement. This study reports on the evaluation of intraoperative MRI (iMRI) for adjusting deviated electrodes to the accurate anatomical position during DBS surgery and acute intracranial changes.

MethodsTwo hundred and six DBS electrodes were implanted in the subthalamic nucleus (STN) in 110 patients with Parkinson disease. All patients underwent iMRI after implantation to define the accuracy of lead placement. Fifty-six DBS electrode positions in 35 patients deviated from the center of the STN, according to the result of the initial postplacement iMRI scans. Thus, we adjusted the electrode positions for placement in the center of the STN and verified this by means of second or third iMRI scans. Recording was performed in adjusted parameters in the x-, y-, and z-axes.

ResultsFifty-six (27%) of 206 DBS electrodes were adjusted as guided by iMRI. Electrode position was adjusted on the basis of iMRI 62 times. The sum of target coordinate adjustment was -0.5 mm in the x-axis, -4 mm in the y-axis, and 15.5 mm in the z-axis; the total of distance adjustment was 74.5 mm in the x-axis, 88 mm in the y-axis, and 42.5 mm in the z-axis. After adjustment with the help of iMRI, all electrodes were located in the center of the STN. Intraoperative MRI revealed 2 intraparenchymal hemorrhages in 2 patients, brain shift in all patients, and leads penetrating the lateral ventricle in 3 patients.

ConclusionsThe iMRI technique can guide surgeons as they adjust deviated electrodes to improve the accuracy of implanting the electrodes into the correct anatomical position. The iMRI technique can also immediately demonstrate acute changes such as hemorrhage and brain shift during DBS surgery.

Intraoperative MRI for transsphenoidal pituitary surgery

Intraoperative MRI for transsphenoidal pituitary surgery

J Neurosurg 120:346–356, 2014

Intraoperative MRI (iMRI) provides updated information for neuronavigational purposes and assessments on the status of resection during transsphenoidal surgery (TSS). The high-field technique additionally provides information about vascular structures at risk and precise information about extrasellar residual tumor, making it readily available during the procedure. The imaging, however, extends the duration of surgery. To evaluate the benefit of this technique, the authors conducted a retrospective study to compare postoperative outcome and residual tumor in patients who underwent conventional microsurgical TSS with and without iMRI.

Methods. A total of 143 patients were assessed. A cohort of 67 patients who had undergone surgery before introduction of iMRI was compared with 76 patients who had undergone surgery since iMRI became routine in TSS at the authors’ institution. Residual tumor, complications, hormone dependency, biochemical remission rates, and improvement of vision were assessed at 6-month follow-up. A volumetric evaluation of residual tumor was performed in cases of parasellar tumor extension.

Results. The majority of patients in both groups suffered from nonfunctioning pituitary adenomas. At the 6-month follow-up assessment, vision improved in 31% of patients who underwent iMRI-assisted surgery versus 23% in the conventional group. One instance of postoperative intrasellar bleeding was found in the conventional group. No major complications were found in the iMRI group. Minor complications were seen in 9% of patients in the iMRI group and in 5% of those in the conventional group. No differences between groups were found for hormone dependency and biochemical remission rates. Time of surgery was significantly lower in the conventional treatment group. Overall a residual tumor was found after surgery in 35% of the iMRI group, and 41% of the conventional surgery group harbored a residual tumor. Total resection was achieved as intended significantly more often in the iMRI group (91%) than in the conventional group (73%) (p < 0.034). Patients with a planned subtotal resection showed higher mean volumes of residual tumor in the conventional group. There was a significantly lower incidence of intrasellar tumor remnants in the iMRI group than in the conventional group. Progression-free survival after 30 months was higher according to Kaplan-Meier analysis with the use of iMRI, but a statistically significant difference could not be shown.

Conclusions. The use of high-field iMRI leads to a significantly higher rate of complete resection. In parasellar tumors a lower residual volume and a significantly lower rate of intrasellar tumor remnants were shown with the technique. So far, long-term follow-up is limited for iMRI. However, after 2 years Kaplan-Meier analyses show a distinctly higher progression-free survival in the iMRI group. No significant benefit of iMRI was found for biochemical remission rates and improvement of vision. Even though the surgical time was longer with the adjunct use of iMRI, it did not increase the complication rate significantly. The authors therefore recommend routine use of high-field iMRI for pituitary surgery, if this technique is available at the particular center.

Intraoperative Magnetic Resonance Ventriculography During Endoscopic Third Ventriculostomy

Usefulness of Intraoperative Magnetic Resonance Ventriculography During Endoscopic Third Ventriculostomy

Neurosurgery 73:730–738, 2013

Endoscopic third ventriculostomy (ETV) is the preferred method for the treatment of noncommunicating hydrocephalus. The different success rates of ETV indicate the difficulties in predicting the success of this procedure.

OBJECTIVE: To show the usefulness of intraoperative ventriculography performed by the low-field 0.15-T magnetic resonance imager Polestar N20 during ETV.

METHODS: The study was conducted in 11 patients with noncommunicating hydrocephalus caused by tumors or cysts of the third ventricle (n = 5), nontumoral stenosis of the sylvian aqueduct (n = 3), and fourth ventricle outlet obstruction (n = 3). Intraoperative magnetic resonance (iMR) ventriculography was performed before and after the ETV.

RESULTS: In each case, iMR-ventriculography was a safe procedure and determined the exact site of obstruction of cerebrospinal fluid flow. In all cases, iMR-ventriculography performed after ETV showed with the greatest accuracy the patency of the performed fenestrations, demonstrating in 9 patients good flow of the contrast from the third ventricle to the basal cisterns, restricted flow in 1 patient, and no flow in 1 patient. The results of ventriculography were consistent with the postoperative neurological status of operated-on patients. In 3 patients, the opinion of the surgeons about the patency of endoscopic fenestration, based on intraoperative observation of the third ventricle floor, was inconsistent with the results from iMR-ventriculography.

CONCLUSION: Low-field iMR-ventriculography is a safe procedure that can be successfully applied during ETV to determine the site of obstruction in hydrocephalus and the patency of performed ventricle fenestration.

Merging machines with microsurgery: clinical experience with neuroArm

Merging machines with microsurgery- clinical experience with neuroArm

J Neurosurg 118:521–529, 2013

It has been over a decade since the introduction of the da Vinci Surgical System into surgery. Since then, technology has been advancing at an exponential rate, and newer surgical robots are becoming increasingly sophisticated, which could greatly impact the performance of surgery. NeuroArm is one such robotic system.

Methods. Clinical integration of neuroArm, an MR-compatible image-guided robot, into surgical procedure has been developed over a prospective series of 35 cases with varying pathology.

Results. Only 1 adverse event was encountered in the first 35 neuroArm cases, with no patient injury. The adverse event was uncontrolled motion of the left neuroArm manipulator, which was corrected through a rigorous safety review procedure. Surgeons used a graded approach to introducing neuroArm into surgery, with routine dissection of the tumor-brain interface occurring over the last 15 cases. The use of neuroArm for routine dissection shows that robotic technology can be successfully integrated into microsurgery. Karnofsky performance status scores were significantly improved postoperatively and at 12-week follow-up.

Conclusions. Surgical robots have the potential to improve surgical precision and accuracy through motion scaling and tremor filters, although human surgeons currently possess superior speed and dexterity. Additionally, neuroArm’s workstation has positive implications for technology management and surgical education. NeuroArm is a step toward a future in which a variety of machines are merged with medicine.

Dual-room 1.5-T intraoperative magnetic resonance imaging suite with a movable magnet: implementation and preliminary experience

Neurosurg Rev (2012) 35:95–110. DOI 10.1007/s10143-011-0336-3

We hereby report our initial clinical experience of a dual-room intraoperative magnetic resonance imaging (iMRI) suite with a movable 1.5-T magnet for both neurosurgical and independent diagnostic uses. The findings from the first 45 patients who underwent scheduled neurosurgical procedures with iMRI in this suite (mean age, 41.3±12.0 years; intracranial tumors, 39 patients; cerebral vascular lesions, 5 patients; epilepsy surgery, 1 patient) were reported. The extent of resection depicted at intraoperative imaging, the surgical consequences of iMRI, and the clinical practicability of the suite were analyzed.

Fourteen resections with a trans-sphenoidal/transoral approach and 31 craniotomies were performed. Eighty-two iMRI examinations were performed in the operating room, while during the same period of time, 430 diagnostic scans were finished in the diagnostic room. In 22 (48.9%) of 45 patients, iMRI revealed accessible residual tumors leading to further resection. No iMRI-related adverse event occurred. Complete lesion removal was achieved in 36 (80%) of all 45 cases.

It is concluded that the dual-room 1.5-T iMRI suite can be successfully integrated into standard neurosurgical workflow. The layout of the dual-room suite can enable the maximum use of the system and save costs by sharing use of the 1.5-T magnet between neurosurgical and diagnostic use. Intraoperative MR imaging may provide valuable information that allows intraoperative modification of the surgical strategy.

Quantification of Glioma Removal by Intraoperative High-Field Magnetic Resonance Imaging: An Update

Neurosurgery 69:852–863, 2011 DOI: 10.1227/NEU.0b013e318225ea6b

The beneficial role of the extent of resection (EOR) in glioma surgery in correlation to increased survival remains controversial. However, common literature favors maximum EOR with preservation of neurological function, which is shown to be associated with a significantly improved outcome.

OBJECTIVE: In order to obtain a maximum EOR, it was examined whether high-field intraoperative magnetic resonance imaging (iMRI) combined with multimodal navigation contributes to a significantly improved EOR in glioma surgery.

METHODS: Two hundred ninety-three glioma patients underwent craniotomy and tumor resection with the aid of intraoperative 1.5 T MRI and integrated multimodal navigation. In cases of remnant tumor, an update of navigation was performed with intraoperative images. Tumor volume was quantified pre- and intraoperatively by segmentation of T2 abnormality in low-grade and contrast enhancement in high-grade gliomas.

RESULTS: In 25.9% of all cases examined, additional tumor mass was removed as a result of iMRI. This led to complete tumor resection in 20 cases, increasing the rate of grosstotal removal from 31.7% to 38.6%. In 56 patients, additional but incomplete resection was performed because of the close location to eloquent brain areas. Volumetric analysis showed a significantly (P , .01) reduced mean percentage of tumor volume following additional further resection after iMRI from 33.5% 6 25.1% to 14.7% 6 23.3% (World Health Organization [WHO] grade I, 32.8% 6 21.9% to 6.1% 6 18.8%; WHO grade II, 24.4% 6 25.1% to 10.8% 6 11.0%; WHO grade III, 35.1% 6 27.3% to 24.8% 6 26.3%; WHO grade IV, 34.2% 6 23.7% to 1.2% 6 16.2%).

CONCLUSION: MRI in conjunction with multimodal navigation and an intraoperative updating procedure enlarges tumor-volume reduction in glioma surgery significantly without higher postoperative morbidity.

Use of Movable High-Field-Strength Intraoperative Magnetic Resonance Imaging With Awake Craniotomies for Resection of Gliomas: Preliminary Experience

Neurosurgery 69:194–206, 2011 DOI: 10.1227/NEU.0b013e31821d0e4c

Awake craniotomy with electrocortical mapping and intraoperative magnetic resonance imaging (iMRI) are established techniques for maximizing tumor resection and preserving function, but there has been little experience combining these methodologies.

OBJECTIVE: To report our experience of combining awake craniotomy and iMRI with a 1.5-T movable iMRI for resection of gliomas in close proximity to eloquent cortex.

METHODS: Twelve patients (9 male and 3 female patients; age, 32-60 years; mean, 41 years) undergoing awake craniotomy and iMRI for glioma resections were identified from a prospective database. Assessments were made of how these 2 modalities were integrated and what impact this strategy had on safety, surgical decision making, workflow, operative time, extent of tumor resection, and outcome.

RESULTS: Twelve craniotomies were safely performed in an operating room equipped with a movable 1.5-T iMRI. The extent of resection was limited because of proximity to eloquent areas in 5 cases: language areas in 3 patients and motor areas in 2 patients. Additional tumor was identified and resected after iMRI in 6 patients. Average operating room time was 7.9 hours (range, 5.9-9.7 hours). Compared with preoperative neurological function, immediate postoperative function was stable/improved in 7 and worse in 5; after 30 days, it was stable/improved in 11 and worse in 1.

CONCLUSION: Awake craniotomy and iMRI with a movable high-field-strength device can be performed safely to maximize resection of tumors near eloquent language areas.

Stereotactic Brain Biopsy With a Low-Field Intraoperative Magnetic Resonance Imager

Neurosurgery 68[ONS Suppl 1]:ons217–ons224, 2011 DOI: 10.1227/NEU.0b013e31820826c2

Techniques for stereotactic brain biopsy have evolved in parallel with the imaging modalities used to visualize the brain.

OBJECTIVE: To describe our technique for performing stereotactic brain biopsy using a compact, low-field, intraoperative magnetic resonance imager (iMRI).

METHODS: Thirty-three patients underwent stereotactic brain biopsies with the PoleStar N-20 iMRI system (Medtronic Navigation, Louisville, Colorado). Preoperative iMRI scans were obtained for biopsy target identification and trajectory planning. A skull-mounted device (Navigus, Medtronic Navigation) was used to guide an MRI-compatible cannula to the target. An intraoperative image was acquired to confirm accurate cannula placement within the lesion. Serial images were obtained to track cannula movement and to rule out hemorrhage. Frozen sections were obtained in all but 1 patient with a brain abscess.

RESULTS: Diagnostic tissue was obtained in 32 of 33 patients. In all cases, imaging demonstrated cannula placement within the lesion. Histological diagnoses included 22 primary brain tumors and 10 nonneoplastic lesions. In 61% of the cases, initial trajectory was corrected on the basis of the intraoperative scans. In 1 patient, biopsy was nondiagnostic despite accurate cannula placement. No patient suffered a clinically or radiographically significant hemorrhage during or after surgery. There were no intraoperative complications.

CONCLUSION: Stereotactic biopsy with a low-field iMRI is an accurate way to obtain specimens with a high diagnostic yield. This accuracy, combined with the acceptable additional procedural time, may obviate the need for frozen section. The ability to correct biopsy cannula placement during surgery eliminates the chance of misdiagnosis because of faulty targeting, as well as the risks associated with inconclusive frozen sections and ‘‘blind’’ replacement of the cannula.