Patterns of Uncertain Diagnostic Significance



Patterns of Uncertain Diagnostic Significance

A number of waveforms and patterns have not been reported to occur in normal infants, do not fit the criteria of normality, and are not associated with well-defined abnormal clinical correlates. These patterns are of questionable or uncertain diagnostic significance.


Currently, little understanding exists of the significance of focal sharp waves in premature and term infants. Various definitions of “normal” spikes and sharp waves have been proposed. For example, sharp waves occurring two or three times in an hour-long recording may have no significance in the newborn. To be considered normal, sharp waves occurring more frequently than this must be truly random and without persistent focality (Monod et al., 1972) and may arise from any scalp location (Engel, 1975; Hagne, 1972; Schulte, 1970). However, no stringent criteria have been developed to separate, unambiguously, abnormal from normal sharp-wave activity in the newborn.

Sharp waves or sharp-and-slow-wave complexes are commonly recorded from the midtemporal region during sleep in apparently normal premature and term infants. The features our group uses to differentiate between normal and abnormal temporal sharp waves are as follows:

Amplitude and Duration

In general, as the amplitude and duration of sharp waves increase, so does the likelihood that they are abnormal. For example, in a premature or term infant, random midtemporal sharp waves having a voltage of less than 75 µV, a duration of less than 100 msec, and occurring only rarely during sleep, would be considered normal (Figure 5.1). Conversely, midtemporal sharp waves with a voltage exceeding 150 µV and a duration greater than 150 msec would be considered abnormal. Unfortunately, numerous temporal sharp waves fall in a “gray zone” by these criteria. The most controversial is that of amplitude, since some sharp waves of greater than 100 µV may be considered normal if they are normal by all other parameters. However, other features, discussed in this section, often aid the neurophysiologist in determining the probability of normality or abnormality.

Incidence and Persistence

Temporal sharp waves that are considered normal occur randomly. They usually occur bilaterally but asynchronously and may be symmetrical or asymmetrical (Figure 5.2). Generally, these sharp waves appear infrequently; thus, temporal sharp waves occurring in quick succession, in series or long runs, and occurring on only one side are likely to be considered abnormal (Figures 5.3–5.5).

Morphology and Complexity of Waveforms

Temporal sharp waves classified as normal are usually mono- or diphasic. The morphology of abnormal temporal sharp waves is more variable. Although they are sometimes mono- or diphasic, abnormal sharp waves more often are polyphasic and followed by a high voltage slow wave (Figures 5.6–5.9).


The initial component of a normal temporal sharp wave is surface negative. Temporal sharp waves with an initial or prominent surface-positive component are abnormal (Figure 5.9).

Features Associated With Changes in Wake/Sleep States

In the premature infant, normal temporal sharp waves may occur in any state. In the term infant, however, normal temporal sharp waves are more common during transitional sleep. If temporal sharp waves appear in the waking record of term infants, they probably are abnormal, regardless of other characteristics (Figure 5.10).

Midtemporal sharp waves satisfying the above criteria of normality occur in some healthy-term newborns during sleep. These sharp waves disappear rapidly during the first month of life and are seen in less than 5% of healthy newborns after 6 weeks post-term. Even temporal sharp waves considered clearly abnormal often do not persist beyond this age.


Midline Frontal Rhythmic Activity

Bursts of rhythmic 50 to 200 µV, 5 to 9 Hz activity may occur in the midline frontal regions of neurologically normal infants between 38 and 42 weeks conceptional age (CA) (Hayakawa et al., 1987). The waves may have a sharp appearance and may occur in brief bursts usually lasting less than 1.5 seconds. They may occur in transitional and nonrapid eye movement (NREM) sleep, but typically not in REM sleep. They have been reported to occur in 55% of neurologically normal term infants, only rarely occur in infants with abnormal outcomes (4% of abnormal infants), and are not present in infants with moderate to severely depressed background activity (Hayakawa et al., 1987).

Midline Central Rhythmic Activity

Rhythmic activity similar to the frontal midline activity described earlier may also be present in the midline central in an otherwise normal electroencephalography (EEG) (Hrachovy et al., 1990) (Figures 5.11–5.13). This activity may also accompany other EEG features that are considered abnormal, particularly central sharp waves, in infants who have experienced various central nervous system (CNS) insults.

Occipital Rhythmic Theta Activity

In the term infant’s EEG, an uncommon finding is the occurrence of brief runs of rhythmic 4 to 6 Hz, 50 to 100 µV activity in the occipital regions with a duration of less than 1 second. This activity appears in wakefulness or sleep and usually occurs asynchronously on the two sides (Figure 5.14). In some infants, this activity may be unilateral (Figure 5.15). Occipital rhythmic theta activity can occur as an isolated feature in an otherwise normal EEG, but usually it accompanies other abnormalities such as sharp waves or an abnormal background.


Transient unilateral attenuation of background EEG activity occurring during NREM sleep is an unusual finding reported to occur in about 3% to 4% of all newborns (Challamel et al., 1984; O’Brien et al., 1987) (Figures 5.16 and 5.17). The attenuation episodes usually last about 1 minute and begin and end abruptly. The background EEG activity is normal in many infants. In our experience, these episodes usually occur only once during a recording session and usually within minutes of the time the infant first enters NREM sleep.

The significance of this finding, which occurs in apparently normal as well as in abnormal infants, has not been determined. Such episodes of unilateral attenuation of background activity should not be confused with the more common finding of marked generalized and/or lateralized voltage attenuation seen in the EEGs of infants who have experienced a variety of CNS insults. The latter attenuation episodes are short in duration (2–15 sec), occur repeatedly during the recording session, and may be present in all wake/sleep states. The isolated transient unilateral attenuation episodes seen in NREM sleep, and the generalized and/or lateralized attenuation episodes associated with various CNS insults, must also be distinguished from the generalized attenuation of background EEG activity that is seen when transient arousal occurs from NREM sleep.


Only gold members can continue reading. Log In or Register to continue

Mar 8, 2018 | Posted by in PEDIATRICS | Comments Off on Patterns of Uncertain Diagnostic Significance
Premium Wordpress Themes by UFO Themes