Atlas of Neonatal Electroencephalography, 3rd Edition

Chapter 2

Techniques of Recording the Neonatal Electroencephalogram*


The general principles of recording the electroencephalogram (EEG) in older children and adults apply to the recording of the EEG in neonates, with some important additions and exceptions. Guidelines for the recording of the neonatal EEG have been established by the American Clinical Neurophysiology Society (ACNS, 1986) and the International Federation of Clinical Neurophysiology (De Weerd et al., 1999). In addition, a number of reports have detailed the technical aspects of neonatal EEG recording (Hanley, 1981; Kagawa, 1973;Mizrahi, 1986). These guidelines and reports were developed when only analog recordings were made. Technologic advances of digital recordings (Levy et al., 1998; Van Cott and Brenner, 1998) and bedside EEG-video monitoring (Kellaway, 1986; Mizrahi and Kellaway, 1987) have created the need for further delineation of their application in the neonate.


Critical to the recording of neonatal EEG is a well-trained staff of electroneurodiagnostic technologists (ENDTs) with expertise in the recording of newborn and young infants. Such technologists provide the expert interface between the patient and the interpreting clinical neurophysiologist by ensuring technical excellence, a clinical understanding of neonatal care, the detailed observation of normal and abnormal infant behaviors, a good working relationship with nursing staff, and an empathetic relationship with parents. The recording of a neonatal EEG is not just the recording of an EEG on a miniaturized adult or child—technologists require specialized training to produce clinically relevant records.

Clinical neurophysiologists with expertise in the interpretation of neonatal EEGs also are essential. The neurophysiologist also should have a thorough familiarity with the clinical problems that neonates may encounter to provide individualized and clinically relevant interpretations and clinical correlations.

It is essential to provide neonatal EEG services 24 hours/day, 7 days/week because the most frequent reasons for referral are suspicion of clinical seizures and acute alteration of mental status. For the EEG to be valuable in the assessment of such affected infants, it must be available around the clock. This requires the availability of specialized technologists and clinical neurophysiologists.

Biomedical engineering support staff and, more recently, computer technologists are often overlooked as essential members of the neonatal EEG team. These professionals ensure that instrumentation is well maintained, ready for use on an emergency basis, and quickly repaired if necessary.


The findings of the neonatal EEG are most valuable when considered in relation to an individual patient's history and clinical findings. To ensure maximal clinical relevance, the ENDT obtains basic information about each neonate to be recorded. This information is listed in Table 2-1 and includes standard demographic data, description of the recording environment, documentation of reason for referral, details of the medical history, a list and timing of medications, and the specifics of the infant's general medical condition. This information may be obtained from the infant's referring physician, parents, or hospital chart; thus the technologist must be trained to be familiar with the medical issues of neonates. To aid in data collection, well-designed data-collection forms tailored for use in each neurophysiology laboratory eventually become part of the infant's laboratory and hospital medical record.


TABLE 2-1. Data collection for recording of the neonatal EEG

1.    Patient demographic information

1.    Name

2.    Date of birth (chronologic age)

3.    Stated estimated gestational age at birth

4.    Birthweight

2.    Recording environment

1.    Location

1.    Hospital nursery

1.    Routine

2.    Special care units

2.    Laboratory

2.    Crib type

1.    Open bassinette

2.    Isolette

3.    Infant warmer

3.    Additional instrumentation

1.    Monitors

2.    Ventilator

3.    Phototherapy

4.    Extracorporeal membrane oxygenation

5.    Other

3.    Reason for referral

1.    Altered mental status

2.    Determination of conceptional age

3.    Identification of diffuse cerebral disturbance

4.    Identification of focal cerebral lesion

5.    Suspected seizures

1.    Description of suspected clinical event

2.    Timing and duration of events

6.    Other

4.    Medical history

1.    Prenatal

1.    Maternal history

2.    Pregnancy history

2.    Perinatal

1.    Route of delivery

2.    Apgar score

3.    Requirements for delivery room resuscitation

3.    Postnatal

1.    Hospital course since delivery

2.    History at home after initial discharge

5.    Medications

1.    Drug

1.    Sedation

2.    Antiepileptic drugs

3.    Paralytic agents

4.    Other

2.    Dose

3.    Timing

1.    Date

2.    Hours

3.    Routine schedule

4.    Route of administration

5.    Recipient

1.    Neonate

2.    Maternal (with potential neonatal effect)

6.    General medical condition

1.    Medical treatments

1.    Limb restraints

2.    Intravascular lines

3.    Intubation

4.    Gastric tube access

5.    Other

2.    Physical condition

1.    Body position

2.    Recent surgical wounds

3.    Healed scars

4.    Scalp swelling

5.    Ventriculoperitoneal shunt

3.    Degree of infant comfort

1.    Comfortable, without distress

2.    Irritable when handled, but consolable

3.    Inconsolable

4.    Apparent mental status

1.    Awake and alert

2.    Difficult to arouse

3.    Nonresponsive

5.    Feeding

1.    Type

2.    Route (oral, tube)

3.    Schedule

4.    Time of last feeding



The recording of the neonatal EEG may be routine in special care units where parents become familiar with the many medical procedures performed on their infants. However, it may be a unique experience for the infant and parents in the outpatient setting. In either circumstance, the ENDT should attempt to put the family at ease by explaining the procedure and answering any questions and concerns.

The clinical state of infants may vary: they may be content and comfortable, difficult to console, irritable, excessively sleepy or lethargic, or comatose. The technologist must consider the state of each infant and determine the best method to make the infant as comfortable as possible to obtain a complete recording. This may require having the infant fed, having diapers changed, adjusting room temperature and, often, just prolonging the recording until the infant becomes comforted. The purpose of these efforts is to facilitate a recording of spontaneous sleep cycles as well as an awake portion.


The International 10-20 System of Electrode Placement (Jasper, 1958) has been modified for recording neonates (Fig. 2-1) (Kellaway and Crawley, 1964). This is to accommodate the neonate's immature frontal lobes that do not extend as anteriorly relative to the skull compared with those in older children and adults. Typically, nine scalp positions are used (Fl, F2, C3, C4, Cz, T3, T4, O1, O2), although others may be added. In addition, electrodes are placed at Al and A2, and a ground electrode is placed either at mid-forehead or on a mastoid region. Because digital recordings are fundamentally referential, an additional reference electrode position may be needed (typically noncephalic), although some instruments provide a so-called “internal” reference. The Fl and F2 electrode positions are 20% of the inion-nasion distance above the nasion and 10% of the circumferential measurement from the midline. For neonates, placement of all of the standard electrodes of the International System would result in such close spacing on the infant's scalp that many electrodes would record overlapping, and thus redundant, electrical fields. The optimal number of electrodes has not been determined, but clinical experience in our laboratories has shown that the nine cephalic positions designated are sufficient to characterize the normal neonatal EEG and to detect, localize, and characterize major abnormalities.

FIG. 2-1. Electrode placement for the neonatal electroencephalogram designated by bolded circles.

After head measurement, the scalp is prepared with slight abrasion at each electrode site with a mild abrasive gel, using the soft end of a cotton applicator. This may cause slight erythema at the site, but this is a temporary effect. Conductive paste is used to secure the electrodes. A ball of cotton is placed over each electrode, and the electrode array is finally secured with paper tape. Paste can be used successfully for even lengthy recordings when applied and attended to by experienced ENDTs. Collodion is typically not used for routine neonatal EEG for a number


of reasons. The environment in which collodion is applied must be well ventilated because it is flammable and may be toxic to the lungs. This is a particular problem for infants in special care units who may be in confined areas such as isolettes that may concentrate fumes and for those infants with already compromised pulmonary function.


Polygraphic measures are integral to the recording of the EEG to assist in characterizing sleep states, eye movements, muscle contractions, cardiac rhythms, and respiratory patterns (ACNS, 1986; DeWeerd, 1999; Hanley, 1981; Kagawa, 1973; Mizrahi, 1990). The same procedures are followed for electrode application as described earlier for scalp electrodes. The basic polygraphic recording parameters are discussed later.


The electrooculogram (EOG) is recorded to detect and characterize eye movements. This assists in staging sleep and in the determination of the origin of some electrical potentials recorded in anterior cephalic electrodes that may have been generated by eye movement. For bipolar EOG recordings, one electrode is placed below and lateral to the outer canthus of one eye, and one, above and slightly lateral to the nasion (Fig. 2-2). This positioning will capture both horizontal and vertical eye movements.


The submental electromyogram (EMG) is recorded to assist in sleep staging and also to characterize some oral-lingual-pharyngeal muscle movements that also may contaminate the EEG. The EMG is recorded with electrodes placed bilaterally and symmetrically, immediately under the jaw. When movements recur in limbs, EMG may be recorded, to determine the precise relation of such movements to any change in the EEG. In addition, limb movements may be detected and characterized by triaxial accelerometry (Frost et al., 1978). This device will produce an analog signal in response to movements of the limb in any plane, compared with EMG, which identifies movement only in a specific direction according to muscle groups.


The electrocardiogram (EKG) is monitored to assess heart rate and rhythm. One electrode is placed over the midline chest and referenced to the right ear for a single-channel recording of the EKG. It also may be used to identify waveforms recorded from EEG electrodes that might have been generated by EKG.

FIG. 2-2. Electrode placement for recording electrooculogram in the neonate.


Respiration is recorded to assist in sleep staging, to characterize various types of apnea, and to assist in differentiating respiratory or body movements from the EEG. The methods used for the recording of respiration may be complex. The most complete method of characterizing respiration is by measurement of abdominal movement (typically with a strain gauge or pneumograph adapted to abdominal placement), thoracic movement (with a strain gauge, bipolar electrodes, or pneumograph adapted to thoracic placement), air flow at the nares or mouth (with a thermocouple device or other flow-measurement devices), end-expiratory CO2 and O2 saturation with a pulse oximeter.

Specialized Polygraphic Measures

A number of other parameters can be measured time-locked to EEG. These are most often used in research protocols or special clinical circumstances. The physiologic measure of greatest interest is systemic blood pressure. Intraarterial blood pressure can be measured from an indwelling catheter already placed for clinical


needs and, via a transducer blood pressure can be displayed numerically on a video screen or as an analog waveform on the recording.


Major strategies in the recording of the neonatal EEG are the documentation of wake/sleep cycles, the characterization of reactivity of the record to stimulation and identification of age-dependent graphoelements (see Chapter 4). These are best achieved in a sustained recording by using a single, bipolar montage with broad coverage over the scalp. In the recording of older children and adults, localization of focal abnormalities often requires the use of several montages; however, because of the range of abnormalities in neonates and the overriding need to characterize state changes over time, multiple montages are not used. A typical montage, with adequate coverage over the scalp, including the required channels with a Cz electrode placement, is given in Table 2-2. This recommendation has been developed for analog recordings. Digital recording of the neonatal EEG provides the opportunity to examine various waveforms with different montages after recordings are complete. New data obtained from these recordings may result in additional guidelines for neonatal recordings.

TABLE 2-2. Sample montage selection

Channel 1


Channel 2


Channel 3


Channel 4


Channel 5


Channel 6


Channel 7


Channel 8


Channel 9


Channel 10


Channel 11


Channel 12


Channel 13


Channel 14


Channel 15


Channel 16

RESP (pneumograph)

Channel 17

Time codea

a Instrumentation dependent, if time-synchronized video is used with analog/paper recordings.


Instrument settings used at the onset of recording are listed in Table 2-3. The filter settings and sensitivity of the EEG channels should be the same as those for EOG to allow accurate comparison of waveforms to differentiate cerebral activity from that of ocular origin. Paper speed is set at 30 mm/sec for analog recordings or 10 sec/screen or “page” for digital recordings. These paper-speed settings are used in many laboratories in the United States and are used in this atlas. However, several laboratories, particularly those with ties to the French school of recording, use a slow speed: 15 mm/sec or 20 sec/screen or “page.”


Over the past several years, considerable interest has been expressed in time-syn-chronized video-EEG monitoring of neonates (Boylan et al., 2002; Bye et al., 1997; Mizrahi and Kellaway, 1987). These studies used these techniques in the research setting. Only recently have other centers begun to record video with EEG routinely for clinical purposes in neonates. At the outset, it may seem as if the addition of video to the recording of neonatal EEG would be relatively easy, particularly with the introduction of new, light-weight, low-light, digital video cameras added to portable digital EEGs. However, many of the difficulties that were present during the development of these techniques still persist.

Nursery Environment

The most challenging aspect of neonatal EEG video monitoring is effective integration of recording efforts with the essential activities of the nursery. Instrumentation, camera mount, electrode cables, and junction box must all be placed not to interfere with the ongoing care of the neonate. A good working relationship between the ENDT and the nurse caring for the infant is essential for optimal recording. It also is important to keep the video image as free as possible from personnel and instruments that block the camera view. In addition, during paroxysmal clinical events, it is essential for personnel not to interfere with the recording of the entire seizure. The ambient temperature of the recording environment should be controlled as well as possible. It is important to keep the infant warm and covered. However, for infants with suspected seizures, it is essential that the infants remain uncovered, with all limbs in full view of the camera. These concerns must be addressed with nursing staff at the onset of the recording session at the bedside and with parents.


TABLE 2-3. Initial instrument settings

EEG channels



7 µV/mm



0.3 sec


Low-frequency filter

0.5 Hz


High-frequency filter

70 Hz


Notch filter (60 Hz, 50 Hz)



Display time


Paper speed

30 mm/sec


Screen display

10 sec/screen


Polygraphic channels







7 µV/mm

7 µV/mm

7 µV/mm

300 µV/mm


0.3 sec

0.03 sec

0.3 sec

0.3 sec

Low-frequency filter

0.5 Hz

5 Hz

0.5 Hz

0.5 Hz

High-frequency filter

70 Hz

70 Hz

70 Hz

70 Hz

Notch filter (60 Hz, 50 Hz)





EOG, electrooculogram; EMG, electromyogram; EKG, electrocardiogram.

a For a digital recording, a 7 µV signal should produce a vertical deflection equivalent to 1/30 of the horizontal distance spanned by 1 second.

b The use of time-constant or low-frequency filter is instrumentation dependent.

c EOG settings remain the same as EEG settings for comparison of simultaneously recorded waveforms.

d EMG, RESP (pneumograph), and EKG settings may be adjusted to optimize display.

Infant Positioning

Infants may be positioned by nursing or attending physician staff for clinical purposes. A number of constraints occur in maintaining a good video image: Limbs with intravascular line placements may be restrained; agitated infants may be swaddled; some infants may be intubated with limited range of head and neck movement; wound dressings may be present; or infants must remain covered because of temperature instability. These and other problems must be addressed to obtain the best video image for each study.

Camera Mount

The camera mount must be stable and must not “jitter” or bounce with movement of personnel around the EEG instrument. In addition, the camera must be mounted so that it is directly above the neonate to provide a full view of all limbs and head without the distortion that comes from a camera placed at the foot or to the side of the infant.


Lighting in neonatal special care areas can be suboptimal for video recording. Ambient light is dictated by clinical needs and day/night cycles imposed for optimal infant adaptation. Some cameras considered “low-light” still do not provide an adequate image. When additional light is used to enhance picture quality, additional problems may arise. Added light may provide added heat and cause the infant to perspire, enhancing sweat artifact. Added light also may provide stark shadows and obscure the image of limbs. These problems must be considered before beginning video recordings and must be corrected when encountered during monitoring.

Potential for Missed Events

A potential is always present for missing the video recording of important paroxysmal clinical events. The optimal video recordings are those time-linked with


EEG, but much can be determined from the video image alone. To maximize video-recording yield, the technologist must first arrange the video camera and begin video recording. This can start immediately after the EEG instrument is brought to the bedside or the infant is brought to the laboratory. Video recording continues through the gathering of historical data and preparation of the infant for EEG. It is terminated only after all electrodes are removed, the infant is cleaned, and the instrument is removed.


The most successful and clinically relevant neonatal EEG recordings are those in which objectives and strategies are identified before the beginning of each study. Advance planning is essential. This can be based on a basic algorithm adaptable to individual recording circumstances (Fig. 2-3). The basic tasks are to obtain historical data, determine the reason for referral, initiate technical recordings, examine the EEG in real time, observe the infant for clinical behaviors, record the infant in sleep and wakefulness, and attempt to provoke abnormal paroxysmal clinical events.

Close observation of the patient at all times is particularly important when recording a neonatal EEG. The record should be annotated when behavioral or autonomic changes occur or when other events happen that may affect the record. It is important that the technologist and clinical neurophysiologist know what behaviors suspected by the referring physician are thought to be clinical seizures. In addition, if the type of paroxysmal behavior for which the infant was referred occurs during the recording, the event should be noted and described. Other abnormal behaviors also should be noted on the record at their time of occurrence.

Some behaviors currently thought to be seizures can be elicited by tactile stimulation and suppressed by restraint or repositioning of the infant's limbs or trunk. Therefore, bedside maneuvers to elicit or suppress such clinical behaviors are essential to an adequate recording of the neonatal EEG. During the recording, infants suspected of having seizures should be subjected to proprioceptive or tactile stimuli such as gentle pinching or tickling of the skin or repositioning of the head, trunk, or extremities. When tonic posturing or motor automatisms occur, the technologist should determine whether they can be stopped by repositioning


the head, trunk, or limbs. If stimulation elicits abnormal behavior, it is important to determine the relation between the intensity of the stimulus and the intensity and irradiation of the response. Whether the intensity of the behavior or the degree of irradiation can be increased by stimulating at multiple sites or by repetitive stimulation of the same site also should be determined. Finally, attempts should be made to suppress abnormal movements with light restraint.

FIG. 2-3. Flow diagram of the sequence of recording neonatal electroencephalogram (EEG) with video. If video is not to be used, only those steps related specifically to video are eliminated, and the remainder are followed for EEG recording.

An important concern is the duration of the neonatal EEG recording. Two considerations apply: the time it may take for the infant to experience the full cycles of sleep and wakefulness, and a sufficient period for the infant to experience clinical or electrical events. The typical minimum duration of recording for a neonatal EEG is 1 hour. A recording may be longer in an infant who may be particularly irritable and may take more time to fall asleep. Conversely, circumstances for a critically ill infant may require that the recording time be shortened. However, even a critically ill infant who is comatose and who may have an EEG that is depressed, undifferentiated, and nonreactive to stimulation may require a recording of at least a full hour because seizure discharges of the depressed brain type may eventually occur. Thus the determination of recording duration is an interactive process, based on clinical circumstances and the unfolding EEG.