Electroencephalography for the Sleep Specialist

Sudhansu Chokroverty, Eli S. Neiman and Sushanth Bhat

Method of Electroencephalographic Recording

Electrical signals are transmitted by the electrodes and conducting gel through electrode wires, which connect the electrodes and the jack box of the polysomnography (PSG) equipment. Different types of electrodes are available, but for recording EEG the best electrodes to use are gold cup electrodes with holes in the center and the silver–silver chloride electrodes. Silver–silver chloride electrodes need repeated chloriding for proper maintenance. Electrode paste or conducting gel is used to secure the electrodes. Positive and negative charges are generated between the scalp and recording electrode as a result of ionic dissociation. The electrode-electrolyte interface is the most critical link in the PSG machine, because most artifacts originate at this site; careful preparation is therefore very important. The impedance in a pair of electrodes should be measured by an impedance meter and should not be greater than 5000 ohms. High impedance impairs the ability of the electrical signal to reach the amplifier and interferes with the capacity of the amplifier to eliminate environmental noise, thus increasing artifacts. All electrode wires terminate in a pin, called a jack, that is plugged into the electrode box. All the wires from the head are gathered into a ponytail in back of the head and then secured by tying them together. A wrap around the scalp secures the electrodes and the wires. The arrangement is therefore as follows: electrodes from the surface of the scalp are connected by electrode wires into the jacks, which are numbered, and then the wires from the jack box are connected through a shielded conductor cable to the electrode montage selector containing rows of switches in pairs corresponding to the inputs of the amplifier.

The placement of the electrodes is determined by the 10-20 electrode placement system that is recommended by the International Federation of Societies for EEG and Clinical Neurophysiology and was published by Jasper in 1958. The 10-20 system is based on definable anatomical landmarks (Fig. 3.1). The system consists of letters denoting the parts of the brain underneath the area of the scalp and numbers denoting specific locations. The odd numbers refer to the left side of the head, and the even numbers refer to the right side. The technician should be familiar with the measurement technique for placement of the electrodes according to the 10-20 international system. Important landmarks for measuring include the inion, the nasion, and the right and left preauricular points. The distance from the nasion to inion along the midline through the vertex should be measured. FPz in the midline is 10% above the nasion of the total distance between the inion and nasion. The electrodes marked FP1 and FP2 are located laterally 10% above the nasion of the total distance between the inion and nasion measured along the temporal regions through the preauricular points. Oz denotes an electrode placed at a distance of 10% above the inion of the total distance between the nasion and inion. T3 and T4 electrodes are placed in a location 10% above the preauricular points of the total distance between the two preauricular points. The rest of the electrodes are located at a distance of 20% measured from inion to nasion anteroposteriorly or laterally through the ears as well as transversely between the ears. The montage refers to an arrangement of two electrodes (derivations). Both bipolar montage (connection of the electrodes between two relatively active sites over the scalp) and referential montage (connection of the electrodes between an active and relatively inactive site, e.g., M1, M2, Cz, Pz; see Fig. 2.1) are recommended. The nomenclature was recently changed to rename T3, T4, T5, and T6 to T7, T8, P7, and P8, respectively, in a modified 10-20 electrode placement system.

The 2007 American Academy of Sleep Medicine manual recommends three channels; this results in significant limitations (see Chapter 1). Most importantly, such a limited montage would result in the inability to capture focal epileptiform activity arising from the temporal lobes, the most common location for such discharges, as well as missing focal slowing from the hemisphere not being recorded. In our own laboratory we use between four and eight EEG channels, including temporal leads from both hemispheres (see Table 1.2); this increases the yield of capturing focal or diffuse slow waves or epileptiform activities over recording with two to four channels. Many patients are referred to the sleep laboratory with a diagnosis of possible nocturnal seizures. In these cases an extended EEG montage (Fig. 2.1; see Table 1.3), covering the bilateral temporal and parasagittal regions and including both bipolar and referential channels, is recommended. A full complement of electrodes and special electrode placements (e.g., Tl and T2 electrodes) should be used. Simultaneous video recording (video-PSG study) for correlation of EEG activities with the actual behavior of the patient is crucial. In computerized PSG recordings (digital PSG recordings), which are currently performed in most laboratories, it is easy to change the recording speed from the standard 10 mm/sec of the usual sleep recording to 30 mm/sec of the standard EEG recording for proper identification of the epileptiform discharges.

Normal Waking and Sleep Electroencephalographic Rhythms in Adults

The dominant rhythm in adults during wakefulness is the alpha rhythm, consisting of 8 to 13 Hz activity distributed synchronously and symmetrically over the parietooccipital regions (see Fig. 2.1). The frequency of the rhythm between the two hemispheres should not vary by more than 1 Hz, and the amplitude should not vary by more than 50%. This alpha rhythm is best seen during quiet wakefulness with eyes closed and is significantly attenuated by eye opening or mental concentration. A small percentage of normal adults show no alpha rhythm, and their EEG is characterized by a dominant rhythm of low-amplitude EEG in the beta frequencies (greater than 13 Hz) with amplitudes varying between 10 and 25 μV (Fig. 2.2). In most adults the beta rhythms are seen predominately in the frontal and central regions intermixed with the posterior alpha rhythms.

There are characteristic changes in the background EEG rhythms as an individual progresses through the three stages of non–rapid eye movement (NREM) sleep and the tonic and phasic stages of rapid eye movement (REM) sleep. The various characteristics during different sleep stages are described in detail in Chapter 3.

EEG in older adult subjects shows a progressive slowing of the alpha frequency during wakefulness and diminution of alpha blocking and photic driving responses. Focal temporal slow waves, particularly in the left temporal region, often called benign temporal delta transients (Fig. 2.3) of older adults, are seen in many apparently normal older adults, and are sometimes associated with sharp transients. Transient bursts of anteriorly dominant rhythmical delta waves may also be seen in some older adults in the early stage of sleep. Other changes in older adults consist of sleep fragmentation with frequent awakenings, including early morning awakening and multiple sleep-stage shifts. Another important finding in the sleep EEG of older adults is the reduction in amplitude of the slow waves. Because of this, many slow waves do not meet American Academy of Sleep Medicine scoring guidelines; the percentage of slow-wave sleep is therefore often reduced in these subjects.

Abnormal Electroencephalographic Patterns

Abnormal EEG patterns consist of focal or diffuse slow waves and epileptiform discharges. The importance of multiple-channel EEG recordings (see Table 1.3) is to document this focal (Fig. 2.4) or diffuse (Figs. 2.5 and 2.6) slowing and epileptiform discharges (see later).

Recognition of Epileptiform Patterns in the Polysomnographic Tracing

Epilepsy is a clinical diagnosis; therefore history must be obtained from patients, relatives, witnesses, or medical personnel if a patient is referred to the sleep laboratory with a suspected diagnosis of nocturnal seizures. The history must include descriptions of ictal, preictal, postictal, and interictal phenomena, family and drug histories, as well as history of any significant medical or surgical illnesses that might be responsible for triggering the seizures. Physical examination must be conducted to find any evidence of neurological or other medical disorders before PSG recording in the laboratory.

Electroencephalographic Signs of Epilepsy

EEG is the single most important diagnostic laboratory test for patients with suspected seizure disorders. Certain characteristic EEG waveforms correlate with a high percentage of patients with clinical seizures and therefore can be considered of potentially epileptogenic significance. These epileptiform patterns consist of spikes, sharp waves, spike and waves, sharp and slow-wave complexes, as well as evolving pattern of rhythmical focal activities, particularly in neonatal seizures. In addition, a pattern that correlates highly with complex partial seizure is temporal intermittent rhythmical delta activity (TIRDA) (Fig. 2.7). Another pattern that is considered a marker of the seizure-onset zone is interictal scalp high-frequency oscillations (HFOs) consisting of gamma frequency activity (30-80 Hz), ripples (80-250 Hz), and fast ripples (250-1000 Hz), recorded noninvasively using amplifiers with appropriate filter settings (Fig. 2.8). Therefore recording HFOs may help map epileptogenic zones.

Recognition of a spike or a sharp wave depends on the presence of certain characteristics. A spike is defined as a waveform that suddenly appears out of the background rhythm with a brief duration of 20 to 70 milliseconds showing a field of distribution and is often followed by an after-going slow wave (Fig. 2.9, A). The characteristic feature of a true epileptiform spike or sharp wave is a biphasic or triphasic appearance with a sharp ascending limb followed by a slow descending limb in contrast to augmented background rhythm of sharp contour, which is generally monophasic and has uniform ascending and descending limbs (Fig. 2.9, B). The amplitude of a true epileptiform spike or sharp wave is at least 30% higher than the background activity, and often there is a disturbance in the background rhythm in the neighboring region of the spikes or the sharp waves. Generally spikes or sharp waves are surface negative. A sharp wave fulfills all the criteria described for a spike except that the duration is 70 to 200 milliseconds. Spike and waves and sharp- and slow-wave complexes may be transient or may repeat for several seconds or longer, and in some patients these repeat in a rhythmical manner, such as may be seen in patients with absence seizure showing 3-Hz spike and wave complexes in the EEG (Fig. 2.10).

An EEG showing all the characteristic features of sharp waves or spikes and accompanied by the behavioral correlates simultaneously during the recording in the laboratory confirms the diagnosis of epilepsy. However, it is rare to observe the occurrence of a clinical seizure during an EEG recording (Fig. 2.11). Therefore a definitive statement about the diagnosis of epilepsy in a particular patient cannot be made even in the presence of the characteristic EEG signs of epilepsy. In patients suspected of seizure disorder it is advantageous to include activation procedures such as sleep, hyperventilation, and photic stimulation during routine EEG recording in the daytime to bring out the ictal or interictal epileptiform patterns. Special basal temporal electrodes (T1, T2, and sphenoidal electrodes) should be used in addition to the routine electrode placement in patients suspected of partial complex seizure. As described earlier, in such patients the PSG study must include multiple EEG channels covering the parasagittal and temporal regions bilaterally.

Nonepileptiform Patterns Mimicking Epileptiform Discharges

Several nonepiletpiform EEG patterns may mimic epileptiform discharges but are not of epileptogenic significance. It is crucial that the polysomnographer be familiar with such normal variants to avoid an erroneous report of epileptiform activity, which may lead to unnecessary further testing and unwarranted medication administration. Some commonly seen normal variants are described in the following sections.

Epileptiform Patterns in Sleep

Patients with focal or generalized seizures may sometimes present only at night. On the one hand, such nocturnal seizures may be mistaken for other motor disorders during sleep such as parasomnias. On the other hand, patients with a variety of sleep-related movement disorders (severe sleep apnea, periodic limb movements of sleep, hypnic jerks, confusional arousals, night terrors) are misdiagnosed as having nocturnal seizures. It is therefore important for the sleep specialist to be familiar with the clinical and EEG patterns seen in primary generalized seizure (e.g., absence spells or generalized tonic-clonic seizures) and partial complex seizures of temporal or extratemporal (frontal) origin.

Patients with nocturnal frontal lobe epilepsy are often referred to the laboratory for PSG recording (see Chapter 11). The characteristic EEG pattern in the primary generalized epilepsy of the absence type is the presence of symmetrical and synchronous frontally dominant 3-Hz spike and wave discharges (see Fig. 2.10) seen either as an ictal or interictal pattern. Patients with tonic-clonic seizure may show a sudden burst of spike and slow waves beginning at a rate of 4 to 6 Hz and then gradually slowing down and stopping before the postictal period. These discharges are seen in a bilaterally symmetrical and synchronous fashion. Postictal slow waves are followed by gradual recovery to the preictal normal background rhythm after several hours or sometimes after a day or two. Interictal EEG may show generalized synchronous and symmetrical 4- to 6-Hz spike and slow-wave discharges or multiple spike and waves.

Certain types of seizures are characteristically observed during sleep, including benign epilepsy of childhood with centrotemporal spikes (Fig. 2.29), tonic seizures, juvenile myoclonic epilepsy (Fig. 2.30), continuous spike and waves during slow-wave sleep (Fig. 2.31), nocturnal frontal lobe epilepsy, and a subtype of partial complex seizure of temporal lobe origin. Some patients with generalized tonic-clonic and partial complex seizure may also have predominantly nocturnal seizures. It is well known that sleep deprivation can lower seizure threshold; indeed, EEGs performed when patients are sleep deprived increase the yield of identifying epileptiform activity.

It is important to differentiate nocturnal seizures from motor and behavioral parasomnias and other movement disorders persisting during sleep. Tonic seizures are typically activated by sleep, occur frequently during NREM sleep, and are never seen during REM sleep. A typical EEG shows interictal abnormality as slow spike and waves intermixed during sleep with trains of fast spikes. Benign focal epilepsy of childhood with centrotemporal spikes is characterized by the presence of centrotemporal or rolandic spikes or sharp waves (see Fig. 2.29). Seizures generally resolve by the age of 15 to 20 years without any neurological sequelae. The EEG in patients with juvenile myoclonic epilepsy (see Fig. 2.30) typically shows symmetrical and synchronous multiple spike and wave discharges. The EEG in the entity of continuous spike and waves during NREM sleep consists of generalized 2- to 2.5-Hz spike and wave discharges occupying 85% of NREM sleep that are suppressed during REM sleep (see Fig. 2.31). This entity is generally seen in children and is associated with developmental regression; it may be mistaken for autism and illustrates the importance of capturing sleep even on daytime EEGs. Nocturnal frontal lobe seizures are discussed in Chapter 11.

Diagnostic Approach to Patients With Suspected Nocturnal Seizures

In a patient suspected of having nocturnal seizures, a routine daytime EEG recording with hyperventilation, sleep, and photic stimulation is the initial recommended diagnostic step. If no epileptiform discharges are noted, then the EEG should be repeated after partial or total sleep deprivation. After three of four negative EEGs in these particular patients, an overnight video-PSG study should be obtained for electroclinical correlation. An appropriately devised seizure montage with full complement of electrodes (see earlier) or special electrode placements (e.g., T1 and T2 electrodes) should be used. During the interpretation of the PSG study, the montage can be reformatted to analyze any abnormal behavior or suspected clinical seizures at 10 mm/sec (the conventional EEG speed) instead of the usual 30 mm/sec (the conventional PSG speed); this makes it easier to recognize abnormal EEG patterns.

Bibliography