Ambulatory EEG

1437

Ambulatory EEG

William O. Tatum, IV

Ambulatory electroencephalogram (AEEG) is a diagnostic extension of standard EEG. It is a well established technique that is best suited for assessing paroxysmal neurological events in patients with epilepsy or in those suspected to have seizures. AEEG uses prolonged overnight EEG monitoring that takes place in the patient’s habitual environment. Like other techniques recording EEG, AEEG is primarily used to support the clinician’s suspicion of an epilepsy diagnosis. Psychogenic nonepileptic attacks represent a “conversion disorder with seizures”. The diagnosis is based upon clinical features that may resemble convulsive movements of people with epilepsy. Syncope is the most common physiological nonepileptic event encountered that mimics epileptic seizures. Syncope manifests as episodes or “spells” with loss of consciousness. AEEG identifies, classifies seizure type, and quantifies seizure frequency and duration. Artifact is present in essentially every EEG and may be over-interpreted to be mistaken for an electrographic seizure.

ambulatory electroencephalogram, artifact, epilepsy diagnosis, epileptic seizures, psychogenic nonepileptic attacks, syncope

Artifacts, Electroencephalography, Epilepsy, Seizures, Syncope

Ambulatory EEG (AEEG) is a diagnostic extension of standard EEG. It is a well- established technique that is best suited for assessing paroxysmal neurological events in patients with epilepsy or in those suspected to have seizures. The International League Against Epilepsy (ILAE) recommends long-term EEG monitoring where there is uncertainty as to the diagnosis of epilepsy. Further, it has established utility in patients with epilepsy to classify an epilepsy syndrome and to quantify seizures manifesting a diurnal or circadian seizure pattern. AEEG uses prolonged overnight EEG monitoring that takes place in the patient’s habitual environment. Because it is performed outside of the hospital setting, capture of the habitual events is more likely to occur. Like other techniques recording EEG, AEEG is primarily used to support the clinician’s suspicion of an epilepsy diagnosis. The duration of AEEG typically lasts long enough to capture targeted events (up to 96 hours) or sample interictal EEG (usually 24–48 hours).

AEEG is utilized primarily as a diagnostic tool capable of detecting epileptiform activity in people suspected to have epilepsy. However, AEEG may also indirectly impact treatment. Prolonged EEG monitoring may characterize patients with focal seizures and determine the cumulative spike and seizure burden in patients with generalized epilepsies. This information can significantly alter anti-seizure medication (ASM) or prompt evaluation for nonmedical therapies. When epileptiform discharges or seizures are recorded, localization and the spatial distribution as well as the presence of self-awareness (signified by push-button activation) are identified by push-button activation.

144TABLE 7.1. Advantages and disadvantages of AEEG compared with standard EEG

Advantages	Disadvantages
Seizures suspect but standard EEG is nondiagnostic	Inappropriate for seizure emergencies
Differentiates epileptic from nonepileptic attacks	May not be diagnostic in nonepileptic attacks without video and EEG change
Prolonged interictal EEG recording	Ancillary testing unavailable (i.e., SPECT)
Records deeper stages of sleep and sleep transitions (awakening)	Unable to taper anti-seizure medication like VEM due to safety reasons
Performed in home environment with typical “triggers” (no “hospital effect”)	Semiology may be unavailable when AEEG is recorded without video
Classifies seizures for appropriate anti-seizure medication management	Nursing and support staff not on-site to perform response testing
Quantifies epileptiform activity	Usually not performed in “real-time”
Identifies seizures without self-awareness (i.e., focal and generalized)	Delayed results due to time required to process EEG that has been recorded

Note: AEEG, ambulatory EEG; VEM, video-EEG monitoring.

To evaluate patients with drug-resistant focal epilepsy, inpatient video-EEG monitoring (VEM) is the gold standard. Rarely, AEEG has been utilized as an adjunct in patients with nondiagnostic inpatient VEM to localize epileptiform activity as an outpatient to avoid the “hospital effect” and characterize selected patients for epilepsy surgery. The benefits of AEEG compared with inpatient VEM include EEG recording in the patient’s typical environment, easy accessibility, greater convenience for patients, and lower cost. Like VEM, AEEG samples deeper stages of sleep and circadian rhythms. There are disadvantages of AEEG compared with VEM and these are listed in Table 7.1. These features have resulted in greater utility for selected patients especially when event capture is desirable and VEM is not accessible or feasible.

Most modern AEEG systems are high-quality, lightweight devices (about 1–2 pounds) using complex portable computerized multiplexed microprocessors to record computer-assisted ambulatory EEG (CAA-EEG). Systems vary and can range from a minimum of 16 channels to 32 channels, though usually 21 electrodes are applied to the scalp and covered by a cap or gauze dressing. These wraps may be customized to make them relatively unintrusive to limits social isolation during AEEG recording (Figure 7.1). The electrodes are connected to a head-mounted preamplifier that digitizes and multiplexes the EEG. The preamplifier connects to a recording device worn around the waist or in a sling over the shoulder. The patient and/or caregiver are instructed to use a push-button activation device to mark an event on the AEEG. Written comments are also provided in a daily written diary. AEEG may be operated using a battery-powered device when portability is desirable or remain connected to an AC-powered home computer for maintaining electrical power support. High-resolution color and infrared video cameras may be included with CAA-EEG monitoring. However, AEEG is commonly performed without video recording in developing countries. Many AEEG systems use 16 bit analogue-digital convertors with sampling rates of 256 to 512 Hz including event recording, spike and seizure software detection, Fourier analysis for quantitative EEG, and artifact reduction algorithms with 145additional DC channels available for ambulatory sleep monitoring. Other non-EEG signals, including ECG, EMG, pulse oximetry, and sometimes carbon dioxide thermistors, permit multimodal ambulatory monitoring for application in a variety of neurological conditions. Wireless networks commonly store up to 30 GB of data over 24 to 72 hours with newer systems using Cloud storage. By using the Cloud for storage and retrieval, secure personal health information meets HIPAA compliance, backs up and shares data with other sources, and can apply software analytics to the AEEG recording. As a result, this limits IT support and maintenance as well as reduces cost of performing AEEG. Fast data transfer can be accomplished using either a DSL or T1 dedicated line to facilitate remote review and even allow for real-time AEEG monitoring.

**FIGURE 7.1.** Ambulatory EEG (customized) headwrap demonstrating portability while shopping at Beall’s department store.

*Source:* From Tatum WO, ed. Ambulatory EEG Monitoring. Demos Medical Publishers, LLC. New York, New York. 2017:42.

146DIAGNOSIS

**FIGURE 7.2.** Burst of left temporal sharply contoured rhythmic mid-temporal theta of drowsiness on ambulatory EEG (AEEG; circle) mimicking an abnormality in a patient with recurrent nonepileptic events.

*Source:* From Tatum WO, ed. Ambulatory EEG Monitoring. Demos Medical Publishers. New York, New York. 2017:63.

Compared with other forms of EEG such as VEM, AEEG has advantages and disadvantages. The incidence of unprovoked seizures is estimated to be 5% in the general population with about 1% of people who develop epilepsy. Standard EEG is the foundational neurodiagnostic test performed in children and adults to assess the risk of epilepsy. Standard EEG is a routine neurodiagnostic test performed in adults as well as in children following a first unprovoked seizure. The definition of epilepsy now includes a single seizure and abnormal epileptiform EEG where > 60% likelihood of recurrence is anticipated (Figure 7.2).

**FIGURE 7.3.** In this case, generalized polyspike-and-wave complexes were present during sleep in a patient with a single generalized tonic–clonic (GTC) and myoclonus to support long-term treatment of juvenile myoclonic epilepsy (JME). The yellow represents special clipped epochs from a 24-hour ambulatory EEG (AEEG). Clips (in yellow) are acquired to condense AEEG for rapid review and to limit memory usage during archiving and storage.

The sensitivity of AEEG has been noted to be 2.23 times greater than standard EEG in the diagnosis of suspected epilepsy (Figure 7.3). AEEG may be useful in the diagnosis of clinically suspected epilepsy that has remained unsupported in up to 30% of patients with epilepsy that are unsubstantiated by standard EEG.

148PSYCHOGENIC NONEPILEPTIC ATTACKS

**FIGURE 7.4.** This 17-year-old female was evaluated for recurrent “petit mals.” She had a single generalized tonic–clonic (GTC) seizure after excessive doses of tramadol were taken for fibromyalgia pain at 15. Ambulatory EEG (AEEG) was without ictal changes during her events with loss of responsiveness. She was diagnosed with psychogenic nonepileptic attacks (PNEA) and had no further symptoms, did not receive anti-seizure medications (ASM), and successfully discontinued tramadol. Most episodes of PNEA are “convulsive.”

The differential diagnosis of epilepsy is extensive. Multiple studies have evaluated the diagnostic utility of AEEG to separate epileptic seizures and nonepileptic events. Paroxysmal neurological events may mimic epileptic seizures because the semiology shares common clinical presentations with a wide array of disorders. Nonepileptic attacks are particularly amenable to AEEG when the diagnosis is in question following a normal standard EEG. Psychogenic nonepileptic attacks (PNEA) represent a “conversion disorder with seizures.” The diagnosis is based upon clinical features that may resemble convulsive movements of people with epilepsy. On average, AEEG records a typical event in approximately one-half of patients. Depending upon the study, about one-third or more have epileptic seizures, up to 60% have PNEA, 10% or less have physiologic events, <5% have both epileptic seizures and PNEA recorded, and <1% have causative ECG changes. Generalized rhythmic tremor is the most common semiology for people with PNEA (Figure 7.4). However, a limp collapse with unresponsiveness or “pseudo-syncope” may occur in 20% to 30% of patients admitted for VEM.

149SYNCOPE

**FIGURE 7.5.** A 26-year-old female had a 1-year history of recurrent spells of “passing out.” Psychogenic nonepileptic attacks (PNEA) was suspected. Following ambulatory EEG (AEEG) with pauses in the ECG she was admitted for video-EEG monitoring (VEM), where bradycardia/asystole was identified. A pacemaker was implanted and no further “spells” were encountered.

Syncope is the most common physiological nonepileptic event encountered that mimics epileptic seizures. Syncope manifests as episodes or “spells” with loss of consciousness. Syncope is commonly associated with convulsive movements manifest as either multifocal myoclonus or tonic posturing that may falsely lead to an epilepsy diagnosis. EEG recordings during an episode of syncope (Figure 7.5). AEEG recordings during an episode of syncope reveal a pattern of increasing theta to delta slowing with subsequent progressive voltage suppression to an isoelectric background. With recovery, the sequence reverses with increasing voltage, delta, theta and ultimately return to baseline. The ECG may demonstrate arrhythmia and differentiate syncope from seizures. Other seizure mimics include limb-shaking transient ischemic attacks, acute states of confusion and delirium, migraine with aura, hyperkinetic movement disorders and paroxysmal dystonia, dyskinesia, and tremor, as well as sleep disorders (Figure 7.6).