16 Fluorescence-Guided Surgery, Intraoperative Imaging, and Brain Mapping (iMRI, DTI, and Cortical Mapping)

The basic idea for development of iMRI was the issue of intraoperative brain shift due to cerebrospinal fluid loss, tumor resection, and tissue edema leading to an increasing error of neuronavigation during surgery.1,2 The first iMRI systems were so-called open scanners or double donut scanners ( Fig. 16.1).³ Surgeons were operating in between the magnetic coils. Based on this configuration, it was only feasible to perform surgery using low magnetic field strength significantly below 1 T. Pre- and postoperative imaging, however, is routinely performed on a 1.5 or even 3-T magnet usually referred to as high-field MRIs. Thus, based on the low field strength, image quality was not comparable to standard diagnostic imaging. Scanning time is long and the number of available sequences is limited on a low-field iMRI. Nevertheless, early data from low-field iMRI-assisted surgeries showed a safe application and a significant benefit of extent of resection (EOR) based on the chance of an intraoperative scan especially for glioma patients.^4,5,6,7 Using a low-field iMRI as well, Senft et al performed a randomized controlled trial (RCT) assessing iMRI assisted versus conventional microsurgical resection in patients with glioblastoma (GBM) tumors eligible for a gross total resection (GTR). The authors showed a significantly increased rate of GTR in the iMRI group leading to an increased progression-free survival (PFS) for those patients.⁸ The data are monocentric and show a considerably lower case number as in the 5-aminolevulinic acid (5-ALA) trial by Stummer et al.⁹ However, Senft et al provide the first study with level-one evidence supporting the benefit of iMRI. Additionally, the importance of GTR as a crucial prognostic factor was emphasized. High-field iMRI systems require dedicated operating rooms (OR) with magnetic shielding and a 5-Gauss safety zone around the scanner ( Fig. 16.2). For an intraoperative scan, the patient is positioned in a dedicated head coil that allows for a sharp fixation of the patient’s head using three or more pins ( Fig. 16.3). Most available head holders are flexible. However, patient positioning is limited when using high-field iMRI. Certain standard positions like park bench or sitting/semi-sitting positions are not possible. For high-field intraoperative MRI, most standard radiological diagnostic MRI scanners are used inside the OR. Thus, in contrast to low-field iMRI due to the significantly higher magnetic field, the surgery has to be performed distant from the magnet and a scan during surgery requires a transfer of the patient into the scanner as shown in Fig. 16.4. Due to the increased image acquisition speed in the high-field MRI, the overall workflow is improved compared to low-field iMRI.¹⁰ Some centers even use the intraoperative MRI in a two-room solution for routine diagnostic imaging and intraoperative imaging for brain tumors ( Fig. 16.5).¹¹

Fig. 16.1 Magnetom open (0.2-T) installation: the surgeons are performing the case inside the isocenter of the scanner. (Reproduced with permission from Hlavac M, Konig R, Halatsch M, Wirtz CR. Intraoperative magnetic resonance imaging. Fifteen years’ experience in the neurosurgical hybrid operating suite. [in German] Unfallchirurg. 2012;115(2):121–124.)

Fig. 16.2 Hybrid operating room (Brainsuite by Brainlab, Feldkirchen, Germany) during an intracranial procedure at the University of Ulm, Campus Günzburg, Germany. The patient is fixed in a dedicated head holder. For an intraoperative scan, the OR table can be turned 90 degrees and the patient can be slid into the MRI bore.
(Reproduced with permission from Hlavac M, Konig R, Halatsch M, Wirtz CR. Intraoperative magnetic resonance imaging. Fifteen years’ experience in the neurosurgical hybrid operating suite. [in German] Unfallchirurg. 2012;115 (2):121–124.)

Fig. 16.3 Patient fixated in a two-part eight-channel head coil integrated in a dedicated holder for automatic intraoperative registration (Noras MRI Products, Höchberg, Germany). For an intraoperative scan, the upper part of the coil is mounted. During surgery, this part is removed.

The development of high-field iMRI (1.5 and 3.0 T) allows for comparable image quality and similar available sequences as in routine diagnostic MRI. Tumor depiction is similar to preoperative imaging. Fig. 16.6 shows a T1 MPRAGE sequence with contrast from an intraoperative scan with the skull open. Assessment of images is simplified, and better depiction of residual tumor is possible due to increased image resolution during glioma and pituitary surgery when compared to low-field imaging.¹⁰ The rate of GTR in low-grade gliomas was found to be increased in a retrospective multicenter series when a high-field iMRI was used instead of a low-field iMRI.¹² Further, advances in diagnostic imaging can usually be translated to intraoperative use.

16.1.2 Intraoperative Diffusion Tenser Imaging and Its Clinical Implications

During glioma surgery, high-field iMRI allows for intraoperative functional imaging sequences that can be of high surgical value. Especially relevant in this regard is the acquisition of diffusion tenser imaging (DTI) data.¹³ Based on this MRI sequence, a probabilistic or deterministic fiber tracking of functional relevant cerebral tracts like the corticospinal tract (CST), arcuate fascicle, or optic radiation can be performed preoperatively and integrated in the neuronavigation system.^14,15 Using high-field iMRI, these data can be updated intraoperatively. It has been shown that due to intraoperative brain shift, a significant change of the position of crucial tracts like the CST can occur.¹⁶ Fig. 16.7 shows an intraoperative example of shift of the CST that was verified by intraoperative subcortical mapping. iMRI-based intraoperative tractography shows a high sensitivity localizing the CST.¹⁷ However, intraoperative DTI imaging data can be disturbed by artifacts surgeons should be aware of. Especially in superficial locations or in resection cavities that cannot be filled with saline properly during the iMRI scan, the air to brain interface can distort DTI images significantly.¹⁸ Thus, a combined use of subcortical mapping of the CST and an intraoperative update of DTI images using iMRI is the method of choice to maximize safety for patients.

Fig. 16.4 Transfer of the patient from OR table to MRI table and insight into the MRI scanner (from left to right).

Fig. 16.5 Two-room solution of a 1.5-T installation at the Department of Neurosurgery, Bern, Switzerland.

Fig. 16.6 Screenshot of the navigation system (Iplan 3.0, Brainlab, Germany) demonstrating the intraoperative MRI scan (T1 MPRAGE with Gd-DTPA on a Simens Espree 1.5-T scanner; Erlangen, Germany) with open skull depicting residual faint contrast enhancement after microsurgical resection of the solid part of a glioblastoma. Residual enhancement was segmented in the navigation system (red object) and resected. It revealed tumor in all sections.

Fig. 16.7 Significant shift of the corticospinal tract (CST) during resection of a diffuse astrocytoma. Green shows the CST in preoperative fiber tracking and yellow shows the intraoperative data. A significant shift of the CST anteriorly toward the resection cavity was noticed, which could be replicated by subcortical mapping during surgery.

Additional MRI sequences beyond gadolinium-diethylenetriamine pentaacetic acid (Gd-DTPA) enhanced T1, T2, FLAIR, and diffusion-weighted imaging (DWI) are perfusion-weighted imaging (PWI) and MRI spectroscopy. Both methods are of significant value in diagnostic imaging but can also be used to detect residual tumor intraoperatively.19,20 Fig. 16.8 shows an example of intraoperative PWI compared to preoperative diagnostic imaging and amino acid PET (AA-PET) imaging. Even resting state connectivity networks were assessed intraoperatively in a first pilot study during general anesthesia, proving very interesting future perspective for functional imaging in patients with eloquent gliomas.21

Fig. 16.8 Comparison of preoperative perfusion-weighted imaging (cerebral blood volume sequence) (upper left) and preoperative methionine PET-CT (upper right) and intraoperative perfusion-weighted imaging (lower left) and intraoperative T1 gadolinium-diethylenetriamine pentaacetic acid enhanced images (lower right).

Concerning surgical workflow, an intraoperative MRI scan is time consuming, especially if additional diagnostic sequences beyond the routine are performed. Further, repetitive administration of Gd-DTPA leads to a nonspecific parenchymal enhancement. Surgical induced changes in T2 might further lead to a decrease of specificity after repetitive scans.22 Thus, while theoretically a GTR should be possible using iMRI as the gold standard of intraoperative imaging in all eligible cases, several practical constraints might hamper a GTR also when using iMRI. In the literature, EOR with iMRI in a patient in whom a GTR was feasible is reported to be around 97%.^8,23,24 Additionally to the interruption of surgical workflow, constructional requirements and high primary costs hamper the distribution of the technique. A central issue in the assessment of a method for accuracy of tumor detection is the definition of a lesion in the typical imaging techniques. In GBM, typically contrast-enhanced T1 imaging is used to define the borders of the solid tumor. Based on AA-PET and histopathological studies, it is well known that the tumor extends well beyond the borders of the contrast enhancement.^25,26,27,28 Thus, high-field iMRI-based resection is limited by the general interpretation of MRI data by surgeons and the respective conclusions and strategies derived from it.

16.1.5 Combined Use of 5-ALA and Intraoperative MRI

Other intraoperative imaging or visualization techniques like 5-ALA provide a different depiction of tumor tissue based on the technique itself. Due to the direct metabolism of 5-ALA to fluorescent protoporphyrin IX (PpIX), 5-ALA provides better detection of solid and invasive tumor compared to intraoperative Gd-DTPA enhancement alone during iMRI.²⁹ However, a single center retrospective analysis shows a potential advantage concerning rate of GTR of iMRI-assisted resection compared to 5-ALA-based surgery alone.²⁴ This might be due to the intraoperative update of navigation data and the excellent overview of the resection cavity using an intraoperative MRI. Further, potential satellite lesions can be detected with iMRI that might be missed using 5-ALA.

Only gold members can continue reading. Log In or Register to continue

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related

Related posts:

Intraoperative Fluorescence Guidance in Meningiomas 5-Aminolevulinic Acid and Recurrent High-Grade Gliomas 5-Aminolevulinic Acid in Low-Grade Gliomas Current Fluorescence-Guided Neurosurgery and Moving Forward Tozuleristide Fluorescence-Guided Surgery of Brain Tumors Indocyanine Green Videoangiography and Arteriovenous Malformations

Stay updated, free articles. Join our Telegram channel

Join