Video-EEG monitoring (VEM) includes a variety of techniques used simultaneously to record electrical and behavioral characteristics of paroxysmal disturbances in cerebral function over extended periods. , It has been called “long-term monitoring” in the past; however, the increasing availability of the recording technology has resulted in it being used in a larger variety of clinical circumstances and for a greater range for durations. Clinically valuable results may require recording durations shorter than an hour or longer than a month. Overall, VEM is useful for investigating events that are difficult to record during routine electroencephalography (EEG) or involve key behavioral changes that are to be assessed in the context of the ongoing EEG. It is employed for four purposes: (1) to distinguish between epileptic seizures and other intermittent behaviors; (2) to characterize the electroclinical features of habitual ictal events in order to diagnose seizure type and, when possible, a specific epileptic syndrome; (3) to determine the frequency and temporal pattern of ictal events to identify precipitating factors and assess the effectiveness of therapeutic interventions; and (4) to localize the site of seizure origin in patients with medication-resistant seizures as part of the evaluation for surgical treatment.
VEM is designed for the diagnosis of intermittent abnormalities that include behavioral change. Clinical situations wherein behavior is unchanging, but neurologic monitoring is needed, such as for a critically ill patient with neurologic disease, typically do not require video recording. Continuous EEG monitoring alone often is sufficient. However, if the critically ill patient is manifesting episodic behaviors of unknown neurologic significance, then simultaneously recording video and EEG provides insight into the underlying cause for the episodes. Simultaneous recording allows a direct comparison of behavior and electrocerebral activity. As such, VEM’s usefulness includes recording for prolonged periods to capture an infrequent and unpredictable behavioral change, as well as recording for brief periods when the behavior is predictable or frequent. For brief recordings, an outpatient setting may be possible. ,
VEM became readily available for long-term monitoring and as a clinical diagnostic tool in the 1960s as a result of two independent technologic developments. First, the use of stereotactically implanted chronic depth electrodes provided a means of easily recording ictal EEG discharges during complex partial and secondarily generalized epileptic seizures without contamination by muscle artifact. Second, EEG telemetry was devised by scientists working with the National Aeronautics and Space Administration (NASA) in order to record electrophysiologic changes occurring in animals (and later, humans) put into earth’s orbit. Combining telemetry technology with depth electrodes chronically implanted into brains of epileptic patients allowed artifact-free ictal EEGs to be recorded during spontaneous seizures occurring unpredictably over prolonged periods. Since then, many advances in VEM have occurred, and a variety of approaches are currently available. No specific approach is generally accepted for standard use; rather, each of the various technologies and methods has relative strengths and weaknesses that allow specific equipment and configurations to be chosen, depending on the needs and limitations of individual clinical facilities.
The American Clinical Neurophysiology Society and the International Federation of Clinical Neurophysiology have issued “Guidelines for Long-Term Monitoring for Epilepsy,” which reviews available technologies and methods, and makes recommendations concerning indications for their use. The technology for VEM continues to advance. Nevertheless, the suggestions for a standard terminology for the most common subcategories of VEM, shown in Table 6-1 , are still appropriate.
|
This chapter is concerned with the indications for VEM, the available monitoring equipment and methods of procedure, the interpretation of data, and recommendations for specific diagnostic purposes. The equipment and procedures discussed here are appropriate for VEM of adults and children; more details on adaptations necessary for application to infants and neonates can be found elsewhere.
Indications
VEM can be expensive and labor-intensive, but it is cost-effective in many circumstances. Its use should be limited to diagnostic problems that cannot be resolved easily in the routine EEG laboratory.
Differential Diagnosis
VEM is often used as the ultimate test in the differential diagnosis between epilepsy and other disorders associated with intermittent or paroxysmal disturbances that resemble epileptic seizures. A definitive diagnosis is made easily when habitual events are shown to consist of clinical behaviors characteristic of epilepsy and are associated with well-defined ictal EEG discharges, or when other etiologies can be clearly demonstrated (e.g., cardiac arrhythmias or sleep disturbances). Often, however, the events in question occur without obvious EEG or other electrophysiologic changes. In this case, a diagnosis usually is reached with reasonable confidence based on features of the ictal behavior in association with other clinical and laboratory information.
Focal seizures without amnesia or alteration of consciousness (traditionally called simple partial seizures) usually have no EEG correlates that can be recorded with extracranial electrodes. Consequently, seizures without impaired consciousness are diagnosed most often by characteristic behavioral features and, at times, elevated serum prolactin levels, rather than by the occurrence of ictal EEG discharges. Myoclonic jerks may also have no EEG correlates but usually can be diagnosed on the basis of characteristic motor signs. Many other nonepileptic disorders can be recognized readily by clinical examination during the habitual event, by review of video recordings, or both. In most of these situations, however, the results of VEM merely confirm the clinical impression derived from historical and other information, and are not diagnostic in themselves.
The most difficult, and most important, differential diagnosis for which VEM is used is the distinction between epileptic seizures and psychogenic nonepileptic seizures. Because, by definition, psychogenic nonepileptic seizures have no abnormal EEG correlates, this diagnosis is usually one of exclusion. Certain features may distinguish them from epileptic seizures. These include: gradual onset, eye closure, waxing and waning motor activity, uncoordinated nonsynchronous thrashing or undulation of the limbs, quivering, pelvic thrusting, side-to-side head movements, opisthotonic posturing, weeping, screaming and talking throughout the ictal episode, prolongation for many minutes or even hours, abrupt termination without postictal confusion, evidence of some recall during the ictal event, and features that are not stereotyped but differ from one episode to another. Any of these symptoms, however, can be epileptic phenomena, and differential diagnosis between nonepileptic psychogenic seizures and complex partial seizures is particularly difficult. Furthermore, nonepileptic psychogenic seizures may be associated with autonomic changes (e.g., pupillary dilatation, depressed corneal reflexes, Babinski responses, cardiorespiratory changes, and urinary and fecal incontinence) as well as self-injury induced by falling or biting the lips and tongue. Consequently, a definitive diagnosis cannot be made solely on the basis of ictal clinical features observed during VEM; other positive evidence (e.g., secondary gain) is necessary.
A conclusion that an ictal event captured by VEM is nonepileptic and psychogenic does not, in itself, rule out the existence of an epileptic condition, because some patients with epilepsy also have nonepileptic psychogenic seizures. When the two phenomena coexist, it should be possible to obtain a history of more than one seizure type and to use VEM to record examples of each type to determine which are nonepileptic and which are epileptic. Based on this information, the patient and family can be instructed to record these events separately to determine the effects of antiepileptic and psychiatric interventions on each independently.
Nonepileptic psychogenic seizures, which are involuntary and as disabling as epileptic events, need to be distinguished from malingering, which is voluntary simulation, and from factitious disorder, which is self-induction of epileptic attacks for the purpose of gaining patient status. This differential diagnosis sometimes can be accomplished by data obtained during VEM, but usually it depends on historical information.
Characterization and Classification
Patients with known epileptic seizures that do not respond to antiepileptic medication may be undergoing treatment for the wrong type of epilepsy. In this situation, VEM can provide crucial information that characterizes the epileptic events so that the physician can make a seizure diagnosis and, when possible, a diagnosis of a specific epilepsy syndrome. Seizure type usually determines the most appropriate antiepileptic drugs and whether a patient might be a candidate for surgical intervention. A specific epilepsy syndrome is often associated with a known prognosis, which is helpful information for the patient and physician.
VEM is particularly useful for distinguishing focal from generalized seizures on the basis of characteristic ictal EEG features. It is important to distinguish the different epileptic causes of brief loss of consciousness, which can be a typical absence seizure, an atypical absence seizure, or a focal seizure characterized by impaired consciousness only. It may be impossible to distinguish between atypical absences resulting from bilateral or diffuse brain damage and focal seizures originating primarily from frontal lobe lesions with secondary bilateral synchrony.
VEM is also useful for determining the degree of alteration of consciousness during focal seizures, particularly when trained personnel are available to examine the patient during the ictal event. Although this information usually has no direct therapeutic relevance, because all focal seizures are treated with the same medications, a definitive diagnosis can at times have medicolegal implications (e.g., a patient without alteration of consciousness might be allowed to drive), and it is important for counseling patients regarding activities of daily living. Documentation of the degree of disability during and after the seizure can be important in the decision about whether surgical treatment is warranted. More detailed tests (e.g., reaction time tasks) during ictal events can be used to identify subtle disturbances of function for the same purposes.
Determination of Frequency and Temporal Pattern of Seizures
When patients are known to have epileptic seizures of a specific type, but it is unclear how often these seizures are occurring, VEM can be used to determine the frequency of ictal events. Patients with seizures that involve relatively brief lapses of consciousness may not be aware of each ictal event; therefore they depend on observers to know whether therapeutic interventions have resulted in benefit. In these situations, VEM is a more accurate way of documenting seizure frequency before and after treatment. Unexplained deterioration in mental function can be caused by unrecognized brief daytime seizures or more severe nocturnal events. Knowledge about the occurrence and frequency of such ictal events is important for medicolegal reasons, for counseling patients regarding the activities of daily living, and for deciding whether surgical intervention is warranted.
VEM can reveal when seizures are most likely to occur. At times, this information is useful for identifying specific precipitating factors that might be avoided. Combining VEM with serum drug level assessments can help to determine whether seizures occur because serum levels are subtherapeutic at specific times of the day, and it can aid in suggesting more effective drug dosing schedules.
Localization of the Epileptogenic Region
The most common use of VEM is for localization of a discrete epileptogenic region in patients with drug-resistant epilepsy who are candidates for surgical therapy. Consideration for surgical therapy is essentially the only situation in which detailed information about localization is of clinical value. This localization includes not only identification of an area that can be removed when localized resection is contemplated, but also demonstration that no such well-defined epileptogenic region exists in patients who are candidates for nonlocalized therapeutic surgical procedures such as hemispherectomy and corpus callosum section.
VEM is capable of revealing clinical signs and symptoms of habitual ictal events that have localizing value (see Appendix 6-1 ); however, these clinical behaviors may result from propagation to distant cortical areas and can never be considered definitive evidence of an epileptogenic region. Consequently, identification of the area to be surgically resected usually requires clear demonstration of the site of electrographic ictal onset. Reliable localization of an epileptogenic region can often be determined with scalp EEG recordings, in association with a variety of other confirmatory tests; but at times VEM with intracerebral, epidural, or subdural recordings is necessary. A variety of electrode types are used for this purpose, including depth, strip, grid, and foramen ovale electrodes. The performance of intracranial VEM requires specialized technical and clinical expertise to place electrodes, to guarantee patient safety in the recording unit, and to interpret the EEG data obtained.
In large part as a result of technical advances in VEM, approximately twice as many patients underwent surgical treatment for medically refractory epilepsy in 1991 as did so in 1985. For some surgical procedures (e.g., standard anterior temporal lobectomy), 70 to 90 percent of patients with medically refractory focal seizures with dyscognitive features can expect to become seizure-free, whereas almost all of the remainder experience worthwhile improvement. The results of VEM techniques are also used increasingly to guide extratemporal cortical resections with beneficial results, and patients with secondary generalized epilepsies who would not have been considered surgical candidates only a few years ago are now benefiting from large multilobar resections and, to a much lesser extent, corpus callosum sections.
Technical considerations
Equipment, methods of procedure, and typical system configurations used for VEM are considered in this section, which is adapted, with permission, from guidelines published elsewhere.
Equipment for EEG Recording
Electrodes
Disk electrodes are used for scalp EEG recordings; needles and electrode caps are not recommended. Electrodes should have a hole in the top to permit periodic re-gelling, and should be applied with collodion and gauze.
Sphenoidal electrodes sometimes are used to record from the mesial or anterior aspects of the temporal lobe in the region of the foramen ovale, but they are used rarely now because true temporal electrodes (T1/T2) usually suffice. They are constructed of fine, flexible, braided stainless steel wire, insulated except at the tip, and are inserted through a needle guide, as illustrated in Figure 6-1 . Sphenoidal electrodes may be left in place from days to several weeks. Nasoethmoidal, supraoptic, and auditory canal electrodes have also been used but are difficult to place; nasopharyngeal electrodes cannot lateralize and their use is discouraged. Some investigators prefer to use a variety of basal electrodes (e.g., sphenoidal plus T1/T2, and other placements on the face and ear) to define the fields of basal epileptiform transients better.
Recordings from the surface of the brain are performed with strip electrodes or grid electrodes, which can be placed epidurally or subdurally. Strip electrodes are inserted through burr holes, whereas grid electrodes require a craniotomy for placement. Electrode strips consist of a row of stainless steel or magnetic resonance imaging (MRI)–compatible platinum disks embedded in Silastic, or a bundle of fine wires with recording contacts at the tips. Electrode grids consist of 4 to 64 small platinum or stainless steel disks arranged in two to eight rows and embedded in soft Silastic. The disks in strip and grid electrodes typically are spaced so that there is 1 cm from disk center to disk center. Strip and grid electrodes are preferred in patients with focal seizures whose epileptogenic region is likely to be in the lateral neocortex. Strips are easier to insert than grids and can be used bilaterally. Grids usually are used only unilaterally because bilateral craniotomy is rarely justified, but their extensive coverage allows not only accurate topographic mapping of interictal and ictal epileptic events but also detailed functional mapping of normal essential cortex.
A variety of rigid and flexible depth electrodes are used for intracerebral recording. Most are multicontact and are constructed of either stainless steel or metals compatible with MRI, such as platinum and nickel-chromium alloy. They are inserted stereotactically by several techniques that enter the skull from the side, back, or top of the head. Depth electrodes are best suited for recording from structures deep within the brain (e.g., the hippocampus and amygdala), orbital frontal cortex, and cortex in the interhemispheric fissure (e.g., supplementary motor area and anterior cingulate). Consequently, depth electrode evaluations usually are preferred in patients with focal seizures of limbic origin. Depth electrodes also are useful when the evaluation requires bilateral coverage or coverage over a large region, although electrode strips also may be used in these situations, depending on the experience and preference of the surgeon.
Intermediately invasive electrodes are used at some centers. Foramen ovale electrodes are constructed of flexible stainless steel or MRI-compatible metals and contain one to four recording contacts. They are placed in the ambient cistern through a needle inserted into the foramen ovale and record from hippocampal gyrus in a manner similar to the most mesial contacts of subtemporally inserted strip or grid electrodes. Although foramen ovale electrodes cannot record directly from hippocampus and amygdala, as do depth electrodes, and do not record as broad a field as strip and grid electrodes, they do have a definite advantage over sphenoidal and other extracranial basal electrodes. Epidural peg electrodes are inserted through twist drill holes in the skull; they can record from selected areas of the lateral cortical surface as an alternative to strip or grid electrodes.
Foramen ovale electrodes may be used in association with grid electrodes for more comprehensive recordings of mesial temporal structures, including those on the contralateral side. Also, strip electrodes and epidural pegs may be used contralateral to grid electrodes as sentinel electrodes to identify the occurrence of distant ictal discharge, even though the spatial distribution of these epileptic events cannot be mapped. Similarly, strip or epidural peg electrodes can be used in association with depth electrodes to provide additional information about ictal discharges at the cortical surface.
Amplifiers
Amplifiers used for VEM should have the following performance specifications: low-frequency response of at least 0.5 Hz; high-frequency response of at least 70 Hz; noise level less than 1 μV RMS; input impedance of at least 1 Mohm; common mode rejection of at least 60 dB; and dynamic range of at least 40 dB. If a preamplifier is used, preamplification input impedance must be greater than 100 Mohm.
Amplifiers should be able to capture a minimum of 24 channels or, more preferably, 32 or more channels at a minimum of 200 samples per second. The analog-to-digital converter should have a minimum of 12-bit resolution with the ability to discriminate the EEG at 0.5 μV steps or less. To prevent aliasing artifacts, a high-frequency antialiasing filter with a minimum rolloff of 12 dB/octave must be used before digitization. The maximum cutoff frequency for the antialiasing filter is determined by the sampling rate. For example, at a sampling rate of 200 Hz, the amplifier must have an antialiasing high-frequency filter no greater than 70 Hz. For higher sampling equipment, proportionately higher-frequency cutoffs can be used. For recording purposes, the low-frequency filter should be set at 0.16 Hz or less. The use of notch filters for acquisition is discouraged. Because VEM systems allow data to be modified when reviewed, frequency filters and gain of the recording system should be set initially to obtain maximum information rather than clean tracings.
Amplifiers may be mounted on the head, carried on the body, or positioned remote from the patient. The closer the amplifier is to the signal source, the shorter are the electrode leads and the less artifact that results from movement or interference. The amplifier is carried most commonly on the body because of size and weight. In scenarios where the patient has violent or continuous movement, the amplifier may be mounted on the head to minimize artifact. Small, lightweight amplifiers and preamplifiers are commercially available for this purpose. Movement and interference artifact is worst with remote amplifiers, and this arrangement is least satisfactory for recording seizures.
Transmission
Standard cable used for routine EEG is the simplest and most widely available means of transmission, but it greatly impairs patient mobility and is associated with the greatest amount of artifact. Telemetry refers to a system in which the EEG signal is amplified close to the patient, multiplexed, and then transmitted to a remote recording device, where it is decoded. With cable telemetry, the multiplexed signal is transmitted over a lightweight cable that consists of a single wire that is long enough to allow the patient to move around the room. Cable telemetry is inexpensive, is associated with low interference, and is the most common form of EEG telemetry used; however, patient mobility is relatively limited. More mobility is possible with radio or infrared telemetry, in which signals are transmitted without a wire, but the range of these devices is still restricted, and there is increased interference with the EEG signal from outside sources. To overcome this patient restriction, some recently developed systems can store the multiplexed signal directly onto flash memory within the amplifier/headbox and retransmit it to the system at a later time. This allows the patient to be unhooked from the system for moderate periods and to move about freely without loss of data. Although this does not offer the same degree of freedom as radio telemetry, it has the advantage of allowing patients limited periods in which they are not restricted by cable lengths without subjecting the data to radio frequency interference artifacts.
Ambulatory recording for longer periods is a specific form of VEM in which the patient wears the recording device and no remote transmission is necessary. However, because of limitations of storage, the number of channels are limited, and often these devices are event recorders and do not provide continuous long-term records.
Amplification close to the body and multiplexing of the resultant strong signal accounts for the major advantage of telemetry. The low-voltage signal travels only a short distance and so is less prone to movement artifact and interference from outside sources. Recordings of epileptic seizures, which usually are associated with considerable movement, are therefore relatively artifact-free compared with ictal recordings obtained in a routine EEG laboratory. These advances have made feasible VEM with scalp electrodes. Multiplexers typically combine 16 to 64 channels into a single channel of information, which then is demultiplexed at some later point. If the multiplexed signal is recorded and stored, the recording apparatus must have a higher-frequency response than if the signal is demultiplexed before recording and storage.
Recording, Storage, Retrieval, and Review
Many methods of EEG recording and storage are currently available. Current systems use analog-to-digital conversion methods, and the data are stored on a computer hard drive, server, or compact disk (CD)/digital video disk (DVD) media. Digital video allows data to be modified easily on review, and it is amenable to computer reduction and analysis. The storage capacity of the system depends on the sampling rate and the number of channels acquired. For 32 channels of EEG sampled at 200 Hz, a CD can hold approximately 20 hours of continuous data, whereas DVD media can store approximately 90 hours. Higher sampling rates and greater numbers of channels decrease this amount proportionately. A minimum recording capability of 30 GB or 24 hours of VEM is the guideline of the American Clinical Neurophysiology Society (formerly the American Electroencephalographic Society).
Systems that permit selected EEG storage employ a built-in time delay so that the data stored include EEG activity recorded both before and after the device is triggered. Computer-recognized interictal or ictal events are used to activate the system, so only the events of interest are stored. Because computerized detection programs are not completely accurate, false-positive detections and failure to detect genuine events occur. However, this approach greatly reduces the amount of data that needs to be stored. Storage can also be activated by a pushbutton that the patient or an observer uses to identify ictal events. Such an event-recording approach requires that the patient or an observer be able to recognize when a seizure is occurring, and seizures that occur without warning or subclinically may be missed. A portable system is now available that allows computerized detection of ictal and interictal events for use with an ambulatory recorder.
Digital video includes the ability to zoom in on the image after recording and the ease of EEG and video time-locking on review, but detailed zooming requires higher image resolution and therefore larger video files. Thus there is a trade-off between the quality of the image and the amount of video that can be stored. To overcome the difficulty with storage, most systems offer the ability to edit and delete unwanted video online before permanent storage, keeping only the clinically necessary video segments.
The most common method for review is to display both the EEG and the video on a high-resolution monitor. This has several advantages, including the ability to reformat the EEG montage, filter settings, and gain. In addition, the digital video image can be time-locked to the EEG via a cursor to correlate clinical behavior easily with the EEG. In order to display the EEG waveform accurately, the display must have certain minimal characteristics. The display should have a minimum scaling ability for each channel such that 1 second of EEG occupies approximately 30 mm, with a resolution of 120 data points per second. Vertical scaling depends on the number of channels displayed, but a minimum of four pixels per vertical millimeter is required for accurate reproduction of the waveforms. The system should also offer the ability to mark the EEG digitally for technologist comments, pushbutton events, and automatic detections.
The major advance in this area of VEM is the capability to record digitally onto disk or flash memory so that the continuous attention of an EEG technologist is not necessary. This has made 24-hour monitoring less expensive. The added refinements of computer-detected and patient-triggered selective identification and/or storage have reduced the amount of data that need to be reviewed. Techniques for rapid review of EEGs have also facilitated data analysis, particularly for ambulatory recording, in which the addition of an audio channel has enhanced recognition of meaningful events. The development of digital recording systems makes it possible to reproduce data on a high-resolution computer monitor in any montage, gain, or filter setting desired. Movable time markers, spike maps, and other programs for analysis permit facilitated interpretation, and networking between workstations at multiple locations is feasible.
Equipment for Monitoring Clinical Behavior
Methods of monitoring behavior include self-reporting, observer-reporting, video and/or audio recording, and detailed monitoring of specific physical or cognitive functions. Self-reporting requires the patient to make notes in a daily diary or log, indicating the occurrence of the specific events in question. This self-reporting can be accompanied by the use of a pushbutton that marks the occurrence of an event on the EEG tracing. Self-reporting is the common method of recording clinical behavior during ambulatory monitoring. Observer-reporting uses the same methods but someone else, such as a parent, maintains a log and activates the event marker. For inpatient VEM, specially trained nurses or technologists can be made available to examine the patient during a seizure in order to elucidate the clinical behavior and to elicit mental and neurologic deficits. A useful approach to the clinical examination during spontaneous epileptic seizures is shown in Table 6-2 . Video and/or audio recording ideally is used in association with self- and observer-reporting, and is the usual practice for inpatient VEM. This approach requires one or more video cameras (which are continuously focused on the patient) and a mechanism for synchronizing the video-recorded behavior with simultaneously recorded EEG activity. In selected situations it may also be helpful to use polygraphic instrumentation, continuous reaction time monitoring, or other automated techniques to identify specific physiologic or cognitive disturbances during ictal events. These data usually are recorded along with EEG activity.
A. ICTAL PHASE
|
B. POSTICTAL PHASE
|
Related posts:
The Emergence of Electrophysiology as an Aid to Neurology
Intraoperative Electroencephalographic Monitoring During Carotid Endarterectomy and Cardiac Surgery
Invasive Clinical Neurophysiology in Epilepsy and Movement Disorders
Somatosensory Evoked Potentials
Neuromuscular Ultrasound as a Complement to the Electrodiagnostic Evaluation
Brainstem Auditory Evoked Potentials: Methodology, Interpretation, and Clinical Application
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
