Intracranial Monitoring

Published on 26/03/2015 by admin

Filed under Neurosurgery

Last modified 26/03/2015

Print this page

This article have been viewed 1464 times

CHAPTER 60 Intracranial Monitoring

Kenneth P. Vives, Andy J. Redmond, Dennis D. Spencer

Rationale

Intracranial monitoring is one of an increasing number of tools available to neurosurgeons for the investigation of brain physiology and pathophysiology. The investigative process for neurological diseases always begins with the patient’s history and physical examination to provide clues about the current state of brain function and dysfunction. For periodic disorders such as epilepsy, examination of the patient during the ictus and immediately after can provide further information about the areas of the brain involved. Neuroimaging with modalities such as computed tomography and magnetic resonance imaging (MRI) can further elucidate anatomic details of the pathology and its relationship to surrounding brain structures. Taken one step further, functional MRI can, in some patients, visualize areas of brain related to specific functions. Although these techniques are promising, the number of specific brain functions that can be mapped is limited. In many cases in which long-standing pathology exists and plasticity has allowed the relocation of function—perhaps in a more diffuse pattern—this type of imaging may not reveal the precise areas involved. Furthermore, the tasks that patients have to perform may be demanding, especially while confined in the MRI gantry. Such problems may be magnified in children and the neurologically impaired. Other imaging modalities, such as single-photon emission computed tomography (SPECT), can identify areas with abnormal blood flow, and positron emission tomography (PET) and magnetic resonance spectroscopy can identify areas of abnormal metabolism. These studies are somewhat lacking in anatomic resolution but can be enhanced by coregistration with anatomic MRI.

Scalp electroencephalograms (EEGs) may provide clues to regional localization of electrophysiologic disturbances; however, the anatomic resolution provided by such studies is inadequate to delineate the relationship of involved and uninvolved structures and allow safe operative intervention. The thick layers of the skin and skull protect the fragile structures of the central nervous system, but they the hinder electrophysiologic localization within the brain.

Thus, despite the technologic advances in functional and metabolic imaging and more sophisticated analysis of interictal abnormal electrophysiology with dipole modeling or magnetoencephalography, there are only two ways that the surgeon can be assured of localizing the epileptogenic substrate. The first is indirect and involves the use of MRI to identify anatomic abnormalities that are known to be highly epileptogenic, such as hippocampal atrophy in the setting of medial temporal lobe epilepsy or low-grade tumors. The second is the use of electrodes to directly record from suspected areas of the brain. The goal of such studies is twofold: to elucidate areas of brain pathology (i.e., epileptogenicity) and to localize brain functions for assessment of the potential risks and benefits of further intervention. Although the remainder of this chapter focuses primarily on intracranial monitoring for epilepsy, use of these devices for extraoperative brain mapping is also addressed.

History

The contribution of Otto von Guericke and later Ewald von Kleist and Pieter van Musschenbroek of Leyden in the late 1600s and early 1700s provided scientists with the means to generate and discharge static electricity. Although abundant hypotheses existed regarding the possible role of electricity in nerve and muscle conduction, Luigi Galvani was among the earliest scientists to demonstrate the role of, as he termed it, “animal electricity” in his experiments with frog leg preparations. Using electrostatic machines and static electricity generated from storms, he demonstrated contraction of muscle in response to the discharge of electricity and published these results in 1791. Galvani believed that these contractions resulted from the discharge of electricity from within the preparation; fortunately, not all agreed. Volta argued that intrinsic electricity was responsible, having been conducted into the tissue, possibly through the nerves. In the early 1800s, Hans Christian Oersted and J. S. C. Schweigger developed devices to measure small amounts of electricity (galvanometers). The development of such sensitive instruments allowed Richard Caton to record “feeble currents of the brain” directly from the cerebral cortex of animals in 1875.¹ Fifty-four years later, Hans Berger is credited with being the first to describe the human EEG. His initial measurements were performed on patients with skull defects or trephine holes and later on intact patients. In 1931, he recorded spike wave activity from the brain of a person with epilepsy. Similarly, in 1935, Frederic and Erna Gibbs recorded comparable spike wave patterns with a frequency of 3 Hz from the scalp of a woman suffering from petit mal seizures. In 1929, Sachs, Schwartz, and Kerr first recorded such activity from the surface of the human brain. Victor Horsely had been using direct cortical stimulation to guide resections for epilepsy, but it was Wilder Penfield and Herbert H. Jasper who began recording abnormal electrical activity directly from the surface of the brain at the time of such surgery. The first use of stereotactic depth electrodes for the treatment of intractable seizures dates to 1950, when E. A. Spiegel and H. T. Wycis recorded from and subsequently lesioned the lateral thalamus in an attempt to relieve seizures. These and other early studies emphasized the use of interictal recordings to guide resections. Talairach and Bancaud realized the limited ability of interictal activity to delineate the areas of paroxysmal ictal discharge,² and under this influence the American neurosurgeon Paul H. Crandall began chronic monitoring for the recording of spontaneous seizures in 1973.

Indications

Epilepsy

Patients being evaluated for intractable epilepsy undergo a fairly standard work-up, with some variability from center to center. Tertiary care centers, where patients with intractable seizures are referred for surgical evaluation, initially screen patients for appropriateness. In general, patients with focal epilepsy that is manifested as a consistent seizure type are likely to have an anatomic substrate for the symptomatic seizures. A second group of patients likely to benefit from surgery consists of those with multiple seizure types but with one type that is more frequent or more disabling. Last are patients with suspected diffuse or bihemispheric seizures that cause drop attacks on generalization. These patients may be candidates for monitoring before consideration of corpus callosotomy.

At Yale, we believe that the ictal onset best defines the volume of tissue that, when resected, will render the patient seizure free. The primary goal of intracranial monitoring of epilepsy patients for resection is to define this volume. It is worthwhile to note that not all institutions use these data in the same way. Some believe that interictal activity should play a major role in defining the resection and may use intraoperative interictal recordings to define the limits of resection.

In a typical phased epilepsy surgery evaluation, patients undergo detailed history taking and physical examination, followed by MRI and an outpatient EEG. This is usually followed by an inpatient continuous audiovisual EEG to further document interictal patterns on the EEG and to capture detailed semiologic (semiology, or the physical and experiential manifestation of a seizure) data, along with the ictal patterns on the EEG. During this time, patients may also undergo interictal and ictal SPECT and interictal PET. Neuropsychological examination may take place at this time as well. At the end of this period of monitoring, all the data are examined in detail to assess whether a specific area is responsible for the seizures. A decision tree is presented in Figures 60-1 and 60-2. Each part of the preoperative data set has its own characteristics that must be weighed in making the final decision with regard to concordance. Some data, by their very nature, point to only broad areas of brain dysfunction, whereas others may be very specific. Other pieces of data, such as an ictal scalp EEG, may be more regional at best and confer more weight to an MRI-detected structural abnormality, or the ictal EEG may indicate multifocality or generalization or be characterized only by muscle artifact.

FIGURE 60-1 Part 1 of a decision-making tree for the evaluation of patients with intractable epilepsy.

FIGURE 60-2 Part 2 of a decision-making tree for the evaluation of patients with intractable epilepsy.

Concordance of these data with an MRI abnormality is necessary to proceed to surgery without further monitoring. Several scenarios are likely. An initial determination of whether the patient has a lesion evident on MRI can be made. Patients with lesions can then be divided on the basis of whether their other data are concordant or nonconcordant. In general, patients with abnormal MRI findings and concordant data are good candidates for resection. In those with normal MRI findings, intracranial monitoring is necessary to delineate the areas of brain to be resected. The study may be designed as a focused study to cover the areas of brain identified by the initial data set as being potentially epileptogenic. In general, if the patient has one predominant seizure type, the ictal EEG may point to a particular area for invasive study. Some patients may have multiple areas of ictal onset but a common spread pattern, thus generating a single seizure type. Other patients may have a single area of ictal generation with a variable spread pattern that is manifested as multiple seizure types. Seizure onset in particular areas of the brain may be prone to rapid spread, thereby making the ictal EEG appear bilateral or diffuse.

These and many other factors need to be considered before deciding that a patient is or is not a candidate for further investigation. For example, a patient may have an ictal EEG that specifies one hemisphere, semiology on an audiovisual EEG that is consistently lobar (i.e., hemifield flashing lights from one occipital lobe), SPECT scans confirming regionally abnormal blood flow and metabolism, and neuropsychological testing that is abnormal but nonspecific. An invasive study should be designed to cover the posterior aspect of the hemisphere in question.

The other set of patients consists of those with evidence of a tumor, vascular malformation, or hippocampal sclerosis on MRI. Typically, these patients, in whom the other data are concordant, proceed to surgery without intracranial monitoring. Exceptions are individuals with evidence of brain developmental disorders on neuroimaging. These patients may have more diffuse brain abnormalities, and at our center, the threshold for invasive monitoring in these instances is low. In many cases in which a lesion is present and the other data are concordant, further study is warranted to ensure that the lesion is or is not responsible for the seizures. In other cases, patients may have multiple lesions, but all the data point to one particular lesion as being responsible for the seizures. Unless there are pressing reasons to address the other lesions (e.g., a patient with multiple cavernous malformations, one of which is larger and has evidence of hemorrhage, but the data point to a smaller lesion without evidence of hemorrhage), many of these patients can proceed directly to surgery.

A special case involves patients with lesions and evidence of hippocampal sclerosis. There is speculation in the literature about the relationship of these entities.³^–⁵ It is possible that an extrahippocampal lesion may cause seizures that spread through the hippocampus. Over time, this spread pattern could damage the hippocampus through excitotoxicity and render it an independent source of seizures. At our institution, in general, patients with hippocampal atrophy and dysfunction (temporal lobe–specific poor memory) in whom the other data are concordant for the temporal lobe undergo combined resection of the lesion and the hippocampus. The more medial the lesion, the more likely we are to assume true dual pathology and resect both. Patients without evidence of hippocampal dysfunction on neuropsychological studies and the intracarotid amobarbital (Amytal) procedure (IAP) undergo lesionectomy without hippocampectomy, regardless of the volume of the hippocampus, although a small proportion of these patients may need further resection. In the context of these cases, the IAP (Wada test) is also specifically used to ensure the ability of the contralateral hemisphere to encode memory when hippocampal resection is planned. Poor memory performance of the contralateral hemisphere on the IAP would largely be a contraindication to resection and would necessitate consideration of nonresective strategies for treatment of the patient’s seizure disorder, such as investigational responsive electrical neurostimulation or multiple subpial transections, when applicable.

Brain Mapping

A separate indication for intracranial monitoring of patients without epilepsy is for extraoperative brain mapping (see Fig. 60-8). The goal is typically to map the areas of appreciable function in relation to a lesion (1) to provide data about the risks associated with surgical resection so that the clinician and patient can make a decision about the appropriateness of surgery and (2) to allow surgical planning to minimize those risks. In some instances, intraoperative monitoring may be the most efficient way to address these issues. Mapping of simple motor modalities can be accomplished at surgery through direct cortical stimulation. Similarly, sensory mapping can be performed via somatosensory evoked potentials. Intraoperative language mapping requires an awake patient who can participate at the time of surgery and demonstrate reversible deficits during cortical stimulation. Testing of higher cognitive function may be difficult during surgery because of the complexity of the tasks that must be performed. Beyond these straightforward tests, particular patients and particular types of mapping may make extraoperative mapping more desirable.

Some patients, such as children and adults who are unable to cooperate, may be unable to participate in an awake craniotomy. The claustrophobic nature of the drapes, even if they are transparent, can be overwhelming. No matter how good the anesthesia, the discomfort from craniotomy may also be a hindrance. Even preoperative language testing in children may be difficult because the patient is required to focus on the tests for a substantial period; the same can be said of neurologically impaired adults. Other brain modalities may be difficult to map intraoperatively. For example, a professional in the field of personal relations who has a lesion in the posterior fusiform gyrus near the facial recognition area may not desire resection if substantial impairment could result. Mapping this area requires multiple visual stimuli that would be difficult to accomplish intraoperatively.

Hardware

A variety of electrodes are available for use in electrophysiologic monitoring, and current research is focused on developing wireless implantable systems. A combination of depth electrodes and subdural strip or grid electrodes are routinely used in most situations (Fig. 60-3). In certain situations in which the plane between the dura and brain surface is scarred and safe dissection is impossible, epidural peg electrodes can be used. Alternatively, a subdural electrode can be placed in the epidural space, but unless the dura has been denervated, these electrodes cannot be used for stimulation-based mapping. Some centers supplement scalp studies or intracranial studies with foramen ovale or sphenoidal electrodes. The construction, surgical technique, and use of each are discussed here.

FIGURE 60-3 Intracranial monitoring hardware. The arrow points to a depth electrode.

Depth Electrodes

Depth electrodes are typically constructed with one or more contacts in a thin shaft with a blunt tip to minimize traumatic brain and vascular injury. Both rigid and flexible types exist. For insertion, the flexible electrodes are made rigid through a stylet that is fixed adjacent to the electrode or placed down the center and removed after insertion. An added advantage is that these electrodes can be tunneled subcutaneously, thereby reducing the risk for infection. The contacts themselves are usually made of platinum or platinum-iridium construction and spaced from 3 to 10 mm apart. For multicontact electrodes, individual insulated wires run up the inner cannula of the hollow electrode and lead to contacts that can be connected to a ribbon cable. Electrodes with multiple contacts can be used to record along the longitudinal dimension of a structure, such as the hippocampus, or they may allow simultaneous recording of a deep structure and the cortical surface through which the electrode was inserted (Figs. 60-4 and 60-5).

FIGURE 60-4 Reconstructions of electrode positions for a typical dominant frontotemporal study superimposed on preoperative magnetic resonance images. A, Anterior medial temporal depth electrode. B, Posterior medial temporal depth electrode. C, Electrode entrance sites (arrows) in the context of the rest of the study from a lateral view. D, Inferior view.

FIGURE 60-5 Coronal and sagittal reconstructions of an orbitofrontal depth electrode.

Subdural Electrodes—Strips and Grids

Subdural electrodes are typically constructed from a flexible material that conforms to the surface anatomy of the brain. Usually, they have multiple contacts in one or more columns to allow simultaneous recording from different areas of the cortex. In general, two types are used: one that consists of a thin shaft or reed with multiple contacts (similar to depth electrodes) and one in which multiple disks are embedded in a thin sheet of Silastic. We find the second type to be more useful because of the enhanced ability to direct the electrodes to specific sites from a remote bur hole. The larger contact surface and the Silastic coating make cortical stimulation possible without dural pain. The wires from a single strip then exit together in a hollow Silastic tube. Newer constructs have multiple strips passing through a large Silastic tube that lead to contacts that can be connected to a ribbon cable. The contacts are constructed of stainless steel or platinum.

Most grids are similarly constructed, but they have columns of multiple contacts for a wider area of coverage (see Fig. 60-4). These grids terminate in multiple thin, hollow tubes with contacts for connecting to ribbon cables. One special type is an L-shaped grid for recording from the medial surface of the hemisphere (Fig. 60-6). Once again, the contacts may be made of platinum or stainless steel. The platinum construction allows safe postoperative performance of MRI.⁶

FIGURE 60-6 Reconstruction of two interhemispheric “L”-shaped grid placements for monitoring of the supplementary motor area and cingulate. The interhemispheric fissure is entered where the veins permit, and both strips are placed, one directed anteriorly and the other posteriorly. Frequently, both sides can be accessed from a unilateral approach by making an opening in the falx between the superior and inferior sagittal sinuses.

Epidural Peg Electrodes

Recording of electrical activity from electrodes placed in the epidural space traces its history to Penfield and Jasper, who used epidural ball electrodes in a few difficult cases.⁷ In comparison to scalp EEGs, epidural recordings have the advantage of providing an improved signal-to-noise ratio. The effect is created by reducing volume conduction and amplitude attenuation and by eliminating the myogenic and kinesigenic artifacts inherent with scalp EEGs.⁸ Epidural recording has also been considered an alternative to subdural and intraparenchymal monitoring because of its semi-invasive nature and the presumed advantage of fewer infectious and hemorrhagic complications.

Epidural electrode designs have included ball electrodes,⁷ screws,⁸ and pegs.⁹ Epidural screws and peg electrodes are widely applied in some epilepsy surgery centers. The screw is usually made of titanium and has a shaft to prevent overpenetration. Screw length varies to allow stable placement and to accommodate the varying thickness of the scalp and calvaria. The screw head is hexagonal to permit easy placement and removal of the electrodes with a wrench. Right-angled EEG monitoring leads can be placed in the screw head.⁸ Epidural peg electrodes are composed of mushroom-shaped Silastic elastomer, and the stalk tapers from a diameter of 4.7 mm to a diameter of 0.5 mm. At the base of the stalk, either stainless steel or platinum tips are used to conduct the electrical current. The tips are continuous with the Teflon-coated steel wire that is tunneled through the peg and exits through the cap. Strip arrays of epidural electrodes have also been described.¹⁰

Foramen Ovale and Sphenoidal Electrodes

Foramen ovale electrodes were first developed in 1985 as a semi-invasive EEG alternative to intracerebral depth electrodes for the evaluation of mesial temporal lobe epilepsy.¹¹ These electrodes generally consist of helical, wound, Teflon-coated silver wires that end in multicontact poles.¹² The construct is mounted on a thin stainless steel wire. The mechanical properties conferred to this construct allow appropriate flexibility to avoid puncturing the pia-arachnoid layer. The external diameter permits easy passage through a specially constructed 18-gauge introducer cannula.¹¹

Buy Membership for Neurosurgery Category to continue reading. Learn more here