MAPPING FUNCTIONAL LOCALISATION ONTO STRUCTURAL ANATOMY

CLINICAL BACKGROUND

The use of MEG to do brain mapping is an established tool of clinical neuroscience with FDA clearance and related CPT coding. Brain mapping in general terms is defined as functional localization mapped on structural anatomy.

Presurgical mapping or intraoperative mapping of areas of functional cortex using invasive electrodes may be necessary when brain lesions to be resected are located near areas of primary cortex involved in motor, sensory, or language function. Intraoperative cortical mapping is considered the gold standard for locating functional regions, there are several limitations associated with it.

The cortical mapping procedure is invasive, a surgical procedure is involved with a craniotomy, and since this type of mapping often occurs just before the planned surgical procedure, little time is allowed to discuss options.

This has led to the use of MEG as a noninvasive preoperative assessment in a clinical situation that allows for high accuracy in mapping motor and somatosensory cortical functions within the brain.

The use of MEG or other noninvasive techniques are expected to provide guidance in surgical planning and intraoperative guidance with the least number of surgical/neurologically based deficits. Cortex that shows a reproducible relation to any given motor or sensory function is termed eloquent cortex. Removal of eloquent cortex during the course of epilepsy or other lesional surgeries could result in the loss of function. Jeff Lewine in the Journal of Neurosurgery presented a method of incorporating noninvasive functional brain mapping data into a stereotactic MR imaging dataset used for resection of brain lesions near eloquent cortex. (1) To correlate methods of preoperative and intraoperative mapping localization directly, Lewine developed techniques of importing magnetic source (MS) imaging and fMRI imaging data into a planning workstation that proved useful in assessing accuracy of various functional brain mapping techniques.

In recent years, the technique of preoperative mapping of functionally important brain areas has made enormous advances. In addition to functional magnetic resonance (fMRI) imaging and positron emission tomography (PET), magnetoencephalography (MEG) has proved to be a valuable tool for the localization of intracranial neuroelectrical sources. MEG offers the possibility of localizing electrical sources with high temporal and spatial accuracy. Magnetic source imaging (MSI) technique allows for MEG results to be overlaid onto MRI images. In addition, with the advent of image guided frameless stereotaxy; there was not only the possibility of using anatomical information, but also of integrating functional data which could be used to identify eloquent brain areas.

The use of MEG/MSI and other techniques makes it possible to identify cental sulcus and motor-sensory cortices accurately. fMRI has often had difficulties in clinical practice such as movement artifacts and close quarters for the patient where MEG does not have these problems.

MEG allows temporal resolution in milliseconds that is not possible with other functional imaging modalities. This type of resolution is critical not only in multifocal epilepsy evaluation but also in mapping of complex cognitive functions such as language. AC Papanicolaou of the University of Texas in Houston has demonstrated the concordance between MSI and direct cortical stimulation for mapping receptive language cortex.

Breir in 1999 in a prospectively blinded study at the University of Texas in Houston explored a comparison of MEG/MSI and the Wada procedure in determining language domination. He found in a small study of adult candidates (26 in number) for epilepsy surgery complete agreement of the MEG/MSI with the Wada procedure. (2)

Oliver Ganslandt in the Journal of Neurosurgery illustrated cases in which his group was able to use MEG to obtain data on the primary sensory or motor cortex which was in agreement with intraoperative recordings. In addition, Ganslandt was able to allow neurosurgeons to choose different treatment strategies in patients with tumor invasion of the motor cortex revealed by MEG. (3) According to Ganslandt, the main advantage of MEG/MSI imaging localization of the sensorimotor cortex is that it allows preoperative assessment of the relation of a lesion to the motor cortex and is quite helpful in guided the type and degree of resection needed to avoid compromise of brain function.

JB Vieth of the University of Erlangen found that electrophysiological signs of deteriorated function of brain tissue altered by surrounding brain tumor tissue can be focal slow and fast activity shown by representative MEG activity. In 2000, Vieth studied 44 patients with brain tumors located near speech related cortical areas. He found that in 43 patients (97.7%) the sensory speech area and in 38 patients (86.4%) the motor speech area could be localized utilizing MEG. (4)

K Kamada highlights in the Journal of the Neurological Sciences that multichannel MEG reflects intracellular electric current flow in the brain and thus provides direct information on neural activity. Kamada used MEG for functional brain mapping of the primary cortex by measuring evoked magnetic fields in addition to having studied increased abnormal spontaneous magnetic activities in the areas to find functionally abnormal lesions surviving in pathological states. (5) In the spontaneous MEG analysis, Kamada observed increased pathological activities and clearly demonstrated the ability of MEG to localize the sources of the abnormal brain activities.

Jyrki Mäkelä of Helsinki University Central Hospital in the Journal of Human Brain Mapping, illustrated that MEG studies could encourage operations on eloquent brain tumors when functional regions are pushed aside but remain unaffected by tumor tissue. Mäkelä sought to improve visualization of functional brain anatomy within the sensory-motor strip in neurosurgical patients with the use somatosensory evoked responses, comparing the sources of the MEG signals with data from intraoperative cortical sub-dural stimulation/recording. In this study, the central sulcus was identified using MRI/anatomy landmarks. The study aim was to locate the sensory-motor strip in a group of brain tumor patients as reliably as possible using a combination of functional methods. (6)

In this study, source locations obtained from MEG data were shown on 3D surface renditions of the individual brains including veins.

The conclusion of the study was that preoperative visualization of functional anatomy within the sensorymotor strip assisted in the design of operative procedures, orientated the neurosurgeon during the tumor resection, and helped to prevent damage during surgery involving eloquent brain regions.

PRESURGICAL TUMOR MAPPING CLINICAL CASE

Clinical History

The patient is a 52 year old female with a oligoastrocytoma Grade II located in the left parietal region. The patient had a sensory-motor clinical finding affecting her right limbs.

Methods

The MEG recordings were performed in a magnetically shielded room utilizing a whole head SQUID system with a 122 sensor array. Somatosensory evoked potentials were elicted by electrical stimulation of the median nerve at the wrist and the posterior tibial nerve at the ankle, face, and auditory regions.

MRI imaging was performed using a 1.5T Siemens Vision system with gadolinium enhanced images used for visualizing the venous system and allowing tumor enhancement. In addition, MR angiography (MRA) was used to additionally view the venous system.

All of the obtained data was analyzed by a semiautomatic system that offered a comprehensive set of 2D and 3D processing operation.

Current sources of evoked responses (ER)

Methods

The MEG recordings were performed in a magnetically shielded room utilizing a whole head SQUID system with a 122 sensor array. Somatosensory evoked potentials were elicted by electrical stimulation of the median nerve at the wrist and the posterior tibial nerve at the ankle, face, and auditory regions.

MRI imaging was performed using a 1.5T Siemens Vision system with gadolinium enhanced images used for visualizing the venous system and allowing tumor enhancement. In addition, MR angiography (MRA) was used to additionally view the venous system.

All of the obtained data was analyzed by a semiautomatic system that offered a comprehensive set of 2D and 3D processing operation.

Current sources of evoked responses (ER)

Figure 1. Lower Left 3D surface rendering including cortical veins with sources of SER’s and AER’s. On Right: Post-operative surface rendering with sources of evoked responses

Figure 2. Brain surface during surgery.	Figure 3. Enlarged 3D surface rendering.

Post-operative result

Left: Post-operative surface rendering with sources of evoked responses.
Right: Sagital and coronal sectioning of post-operative MRI.

MEG RESULTS:

Summary:

Cortical stimulation of the area predicted by somatosensory evoked potentials sources produced paraestesia and sensations in the right hand and corner of the mouth. Cortical evoked potentials using a subdural grid inverted in polarity along the course of the estimated central sulcus. The corresponding hand motor fMRI was off by a sulcus.

A Grade II oligoastrocytoma was removed during an awake craniotomy microsurgically with no permanent postoperative speech problems or limb weakness. The postoperative MRI showed a total tumor removal with the source of the MEG activity now clearly separated. The postoperative MRI clearly showed that the slow growing tumor was distorting cortical anatomy.

An overall conclusion of this case is a clinical example of a brain tumor where sources of pathological focal activities and simultaneously analyzed metabolic changes showed functionally abnormal lesions vs. normal cortex.

Conclusion

Clinical applications of MEG have shown utility such that magnetic spike sources correlate with conventional noninvasive and invasive EEG findings in patients who have undergone epilepsy surgery. Such studies have concluded that MEG was useful in localizing the epileptogenic zone in patients with refractory seizures.

In addition, another aspect of the clinical utility of MEG has been illustrated in the use of the technology to reliably deal with the functional mapping of tumors near the eloquent cortical areas to avoid creating new neurological deficits postoperatively.

Acknowledgements

The author would like to thank Jeffrey Lewine PhD, Michael Funke MD, PhD, and Professor Jyrki Mäkelä for their contributions to this manuscript.

The author is a neurophysiological and MEG consultant to Elekta Inc.

References

1. McDonald JD, Chong BW, Lewine JD et al. Integration of preoperative and intraoperative functional brain mapping in a stereotactic environment for lesions near eloquent cortex J Neurosurgery 1999 90, 591-598.
2. Breier JI, Simos PG, Zouridakis G. et al. Language dominance determined by magnetic source imaging-a comparison with the Wada procedure Neurology 1999 53, 938-45.
3. Ganslandt O, Fahlbusch R, et al. Functional neuronavigation with magnetoencephalography: outcome in 50 patients with lesions around the motor cortex Journal of Neurosurgery 1999 91, 73-79
4. Vieth JB, Kober H, Ganslandt O, et al. The Clinical Use of MEG Activity associated with Brain Lesions Biomedical Engineering 1999 44 Suppl. 2 61-69.
5. Kamada K, Moller M, Saguer M, et al. A combined study of tumor related brain lesions using MEG and proton MR spectroscopic imaging Journal of the Neurological Sciences 2001 186 13-21.
6. Mäkelä J, Kirveskari E, et al. Three Dimensional Integration of brain anatomy and function to facilitate intraoperative navigation around the sensorimotor strip Journal of Human Brain Mapping 2001 12: 180-192.