Figure 130.21 Hypnagogic hypersynchrony. This less common but well recognized normal variant of drowsiness is seen in children aged 3
months to 13 years. It consists of paroxysmal bursts (3–5 Hz) of high-voltage (as high as 350 mV) rhythmic waves. Note the normal alpha
rhythm in the first two-thirds of the sample, which precedes drowsiness.
Figure 130.22 Stage 2: sleep spindles. This segment depicts typical short bursts (1–3 seconds) of sinusoidal 12–16 Hz central activity that
waxes and wanes.
THE NORMAL EEG 1005
By strict sleep staging criteria on polysomnography,
REM sleep is defined by (1) rapid eye movements, (2) muscle
atonia, and (3) EEG ‘‘desynchronization’’ (compared to
slow-wave sleep). Thus two of the three defining character-
istics are not cerebral waves and theoretically require
monitoring of eye movements (electro-oculogram (EOG))
and muscle tone (electromyogram (EMG)). Fortunately,
muscle activity and eye movements can be evaluated
on EEG; thus REM sleep is usually not difficult to
identify (Figure 130.25). In addition to the three features
already named, ‘‘sawtooth’’ waves also are seen in REM
sleep.
� EEG Desynchronization. The EEG background activ-
ity changes from that seen in slow-wave sleep (stage 3
or 4) to faster and lower voltage activity (theta and
beta), resembling wakefulness.
� Rapid Eye Movements. These are saccadic,
predominantly horizontal, and occur in repetitive
bursts.
� Despite the lack of a dedicated EMG channel, the
muscle atonia that characterizes REM sleep is usually
apparent as a general sense of ‘‘quiet’’ muscle activity
compared to wakefulness.
� Sawtooth waves are a special type of central theta
activity that has a notched morphology, resembling
the blade of a saw, and usually occurs close to rapid
eye movements (i.e., phasic REM). They are only
rarely identifiable on EEG.
The duration of REM sleep increases progressively with
each cycle and tends to predominate late in the sleep period
into early morning.
Figure 130.23 Stage 2: K-complexes. This sample shows the typical long-duration (�1 second) diphasic waveform. Note the widespread
distribution on this double banana montage, but on a transverse montage one would typically see a maximum (phase reversal) at the midline
(Cz, Fz).
1006 INTRODUCTION TO SLEEP ELECTROENCEPHALOGRAPHY
Less Common Patterns and Normal Variants
Just like human anatomy, sizes and shapes vary somewhat
among individuals, so too do brain waves. There is a wide
range of variability, and it is important to read EEG ‘‘con-
servatively’’ and avoid overreading normal variants [9].
There are many normal variants that have been well
described as benign and have no association with epilepsy.
These include small sharp spikes (also called benign epilep-
tiform transients of sleep, Figure 130.26), wicket spikes
(Figure 130.27), 14 and 6 Hz positive spikes, phantom
spike waves, rhythmic midtemporal theta of drowsiness
(also called psychomotor variant), and subclinical rhythmic
epileptiform discharges of adults. These can be overread as
abnormal. Lambda waves (Figure 130.28) are occipital
sharp transients that resemble POSTS but occur in wake-
fulness when subjects (especially children) scan the envir-
onment. Despite these, the most commonly overread
patterns are ‘‘nameless’’ fluctuations of background
activity [9].
THE ABNORMAL EEG
Like most neurophysiologic tests, EEG is a test of cerebral
function, and as such is for the most part nonspecific as to
etiology. Although earlier investigators have attempted to
identify the reliability of EEG in differentiating types of
lesions, this has clearly become a senseless and futile exer-
Figure 130.24 Slow-wave sleep. This is the same sample at an EEG time-base (left) and at a PSG time-base (right). There is widespread
delta (2 Hz) activity.
THE ABNORMAL EEG 1007
cise in the modern era of neuroimaging [10]. The exercise
to describe EEG abnormalities by pathology (stroke,
abscess, tumor, etc.), which was common in old EEG
texts [11], is clearly obsolete and will not be followed
here.
There are many different ways to classify EEG abnorm-
alities. This chapter will use a very practical classification
developed at the Cleveland Clinic Foundation [12] and used
at many centers. The outline of the classification is shown
in Table 130.1.
Epileptiform Abnormalities
Types of Epileptiform Abnormalities EEG is useful in
epilepsy because it is the only test that gives direct evidence
for epileptogenicity. Most commonly, clinicians have to
rely on interictal abnormalities. Interictal epileptiform
abnormalities include spikes, sharp waves, and spike–
wave complexes. In addition to type and morphology, loca-
tion is very important and is typically either focal or gener-
alized. Generalized epileptiform discharges (i.e., spikes,
sharp waves, polyspikes, spike–wave complexes), as seen
in the primary generalized epilepsies, are usually maximal
in the frontal regions, with typical ‘‘phase reversals’’ at the
F3 and F4 electrodes or less commonly at F7 and F8.
� Spikes and sharp waves are sharp transients that have a
very strong association with epilepsy. The two are
distinguished only by their duration (spikes <70 ms,
sharp waves 70–200 ms), but they have no differences
in terms of clinical significance. Several characteris-
tics help distinguish these from benign epileptiform
variants but this distinction can at times be difficult.
Helpful features that indicate pathologic discharges
include high amplitude, which makes them ‘‘stand
out’’ from ongoing background activity, and after-
going slow waves, which indicate ‘‘disruption’’ of
background activity (Figures 130.29 and 130.30).
The terms ‘‘spikes’’ and ‘‘sharp waves’’ should be
reserved for these abnormalities thought to be patholo-
gical and indicative of epileptogenicity. If one wants to
use purely descriptive (uncommitted) terms, then
‘‘generic’’ phrases should be used, such as ‘‘sharp
transients’’ or ‘‘sharply contoured waveforms.’’
Figure 130.25 REM sleep. This segment shows rapid eye movements (3rd and 10th seconds). Note also the sawtooth waves: notched theta
(�5 Hz) transients in the central regions.
1008 INTRODUCTION TO SLEEP ELECTROENCEPHALOGRAPHY
Figure 130.26 Small sharp spikes (SSS) or benign epileptiform transients of sleep (BETS). Typical brief (<50 ms) and low-amplitude (<50
mV) sharp transients in the temporal region, typically in stage 1 or 2 sleep.
Figure 130.27 Wicket spikes. These are typically sharp transients seen in the temporal regions during wakefulness or stage 1 sleep. They
have a symmetric upgoing and downgoing phase, typically arise from an ongoing background activity, and do not disrupt the background
(i.e., no after-going slow wave). They are basically fluctuations of background activity.
THE ABNORMAL EEG 1009
TABLE 130.1 Cleveland Clinic EEG Classification
Slow activity
Background slow (BS)
Intermittent slow (IS)a
Intermittent rhythmic slow (IRS)a
Continuous slow (CS)a
Epileptiform patternsa
Spikes
Sharp waves
Benign epileptiform discharges of childhood (BEDC)
Spike–wave complexes (SWCs)
Slow spike–wave complexes (SSWCs)
3 Hz spike–wave complexes (3 Hz SWCs)
Polyspikes
Hypsarrhythmia
Photoparoxysmal response (PPR)
EEG seizure pattern
Artifact obscured EEG
Special patterns
Asymmetry
Excessive fast
Sleep-onset REM
Periodic pattern
Triphasic waves
PLEDsa
Background suppressiona
Special patterns used only in coma
Alpha coma
Spindle coma
Beta coma
Theta coma
Delta coma
ECI
a Abnormalities that require localization:
� Generalized
� Lateralized
� Multiregional
� Regional
� Multifocal
� Focal
Source: From [12].
Figure 130.28 Lambda waves. This sample shows the typical occipital sharp transients that resemble POSTS but occur in wakefulness
when subjects (especially children) scan the environment.
1010 INTRODUCTION TO SLEEP ELECTROENCEPHALOGRAPHY
� Polyspikes are usually generalized (rarely focal),
although focal spikes can at times have a multiphasic
‘‘polyspike-like’’ morphology. Polyspikes are multi-
ple repetitive spikes occurring at about 20 Hz and
are typically seen in the primary generalized epilepsy
with myoclonic seizures (Figure 130.31).
� Spike–wave complexes (SWCs) are the repetitive
occurrence of a spike followed by a slow-wave. Since
any significant spike or sharp wave usually is followed
by a slow wave (see earlier), a run of 3 seconds is
required to classify a record as SWC, as opposed to
the categories already mentioned (spike or sharp
wave). SWCs can be divided further into two more
specific types, as follows:
3 Hz SWC. This pattern is characterized by a
frequency of 2.5–4 Hz and a very monomorphic
(‘‘perfectly regular’’) morphology (Figure 130.32).
It occurs in very discrete bursts, and between
bursts the EEG is normal.
Slow SWC. This pattern is not only slower (<2.5
Hz) but also more irregular (less monomorphic)
than the 3 Hz SWC. Bursts are less discrete than
the 3 Hz SWC, and between bursts other abnorm-
alities are seen in symptomatic/cryptogenic epi-
lepsies of the Lennox–Gastaut type.
Hypsarrhythmia. This is characterized by continuous
(during wakefulness) high-amplitude >200 mV
(microvolts) generalized and polymorphic slow-
ing with no organized background, and multifocal
spikes (Figure 130.33). During NREM sleep, the
pattern becomes discontinuous and fragmented,
resembling a burst-suppression or pseudoperiodic
pattern, while it tends to disappear in REM sleep.
� Electrographic Seizures (Ictal Patterns)
Figure 130.29 Typical left anterior temporal sharp wave. The maximum (negativity) is at F7 and T1 (phase reversal). Note the after-going
slow wave. Classification: sharp wave, regional, left frontotemporal.
THE ABNORMAL EEG 1011
Focal seizures are discharges characterized by rhyth-
micity and evolution (‘‘buildup’’) in frequency,
amplitude, and distribution. The discharge can
consist of rhythmic theta or delta activity, or
repetitive spikes or sharp waves, but the most
characteristic features of electrographic focal sei-
zures are rhythmicity and evolution (Figure 130.34).
Generalized seizures are also characterized by rhyth-
mic discharges that evolve. They can be variable
and there are rhythmic or periodic abnormalities
that are not ictal (see later under status epilepticus).
Spike–wave complexes are both an interictal and an
ictal pattern. The distinction is arbitrary and based
on duration and the presence of detectable clinical
symptoms (i.e., alteration of awareness).
Electrodecrement consists of abrupt attenua-
tion (‘‘flattening’’) of background activity, often
preceded by a high-amplitude transient, which
can be sharply contoured or broad, or generalized
paroxysmal fast activity (GPFA). This typically is
associated with infantile spasms and tonic or ato-
nic seizures (Figure 130.35).
Figure 130.30 This sample shows a left frontal spike in a patient with left frontal lobe epilepsy. The left panel is a double banana, the right
is a referential montage showing the same sample. Note that there are two different spikes: the first is maximum at F7, the second at Fp1.
Classification: spike, regional left frontal.
1012 INTRODUCTION TO SLEEP ELECTROENCEPHALOGRAPHY
Epileptiform discharges have to be differentiated from
normal patterns and normal variants that can look like
epileptiform, and this can at times be difficult. The overin-
terpretation of EEG is a common and underreported pro-
blem [9, 13, 14] and a major cause for the misdiagnosis
of epilepsy. If read conservatively (being careful not to
overinterpret normals), the specificity of EEG for epilepsy
is very high, that is, greater than 90%. It is generally
accepted that <3% of the general population has interictal
epileptiform abnormalities, with a slightly higher percen-
tage in children [15] (largely because of benign focal
epilepsies in that age group, see later) Because of the rela-
tively frequent misinterpretation (overreading) of benign
EEG recordings, it is unfortunately true that ‘‘routine inter-
ictal EEG is one of the most abused investigations in clin-
ical medicine and is unquestionably responsible for great
human suffering’’ [16]. One of the important remedies to
this pitfall is that when in doubt, one should err on the
side of underreading. Clinical experience strongly supports
the view that less harm will be done by underinterpreting an
abnormality than by overinterpreting a normal pattern.
Electrographic Status Epilepticus Status epilepticus (SE)
is typically an obvious clinical diagnosis, but in some situa-
tions an EEG is required to diagnose ‘‘nonobvious’’ SE.
Electrographically, SE can take the form of either repetitive
discrete seizures or, more commonly, a continued pattern of
rhythmic or periodic discharges such as generalized peri-
odic epileptiform discharges (GPEDs) [17, 18]. Unfortu-
nately this pattern (Figure 130.36) is not specific for
electrographic SE and can be seen in severe metabolic
encephalopathies such as anoxic, uremic, or hepatic distur-
bances [19, 20, 21]. Thus the final answer as to whether it
does or does not represent SE in a given patient often will
depend on the clinical circumstances and the response to
treatment [21].
Figure 130.31 Polyspike. This is a typical generalized polyspike in a patient with an idiopathic (‘‘primary’’) generalized epilepsy. Note that
the discharge is followed by about 2 seconds of delta activity. This after-going slow activity indicates substantial disruption of the background
and is a significant argument to consider this discharge pathological and not a normal variant. Classification: polyspike, generalized.
THE ABNORMAL EEG 1013
Figure 130.32 3 Hz spike–wave complexes. This shows typical 3 Hz spike–wave complexes in a patient with an idiopathic (‘‘primary’’)
generalized epilepsy. This patient likely has absence seizures. If tested with a clicker, a discharge of this duration (�4.5 seconds) is likely
associated with a brief impairment of awareness. Classification: 3 Hz spike–wave complexes, generalized.
Figure 130.33 Hyparrhythmia. These recordings are from two different patients. Note on both the lack of any reactive background, the very
high-amplitude polymorphic delta activity, and the multifocal spikes. Classification: hypsarrhythmia.
1014 INTRODUCTION TO SLEEP ELECTROENCEPHALOGRAPHY
Figure 130.34 Focal seizure. Left temporal seizure (singular), showing the typical rhythmic activity that evolves (build up). Classification:
EEG seizure, regional left temporal.
Figure 130.35 Generalized paroxysmal fast activity followed by attenuation (electrodecrement) in a patient with a symptomatic general-
ized epilepsy of the Lennox–Gastaut type. This could be an interictal (asymptomatic) discharge but could also be an ictal pattern associated
with a tonic or atonic seizure. Classification: EEG seizure, generalized.
THE ABNORMAL EEG 1015
EEG-Video Monitoring and the Differential Diagnosis ofEpilepsy Prolonged EEG-video monitoring is indicated
when seizures do not respond to medications [4]. About
20–30% of patients referred for refractory seizures do not
have epilepsy but have psychogenic nonepileptic seizures
(PNESs) instead [3, 13]. A small percentage have syncope
[14, 22]. Occasionally other paroxysmal conditions can be
misdiagnosed as epilepsy, but PNESs are by far the most
common condition, followed by syncope. Unfortunately,
the current average delay in the diagnosis of PNESs is
over 7–9 years and 80% of patients with PNESs have
received antiepileptic drugs [23]. This means that EEG-
video monitoring is probably underutilized and the ‘‘estab-
lished’’ diagnosis of seizures is not verified often enough.
Once the diagnosis of PNESs is suspected clinically [7],
it is usually easily confirmed by EEG-video monitoring,
which can even be performed as an outpatient. In the
hands of experienced epileptologists, the combined electro-
clinical analysis of both the clinical semiology of the
‘‘ictus’’ and the ictal EEG findings allows a definitive diag-
nosis in nearly all cases. The second reason why medica-
tions may fail is that 30% of epilepsies are intractable
and require nonpharmacologic treatments [4].
Encephalopathic Patterns
These patterns are associated with diffuse (generalized)
brain dysfunction, that is, diffuse encephalopathies. Again,
in general, they are completely nonspecific in terms of
etiology. In order of increasing severity, they include:
� Background Slowing (Classification as ‘‘Background
slow’’). There is a posterior dominant background
that is reactive, but the frequency is too slow for age
(<8 Hz after the age of 8) (Figure 130.37).
� Intermittent Generalized Slowing (Classification as
‘‘Intermittent Slow, Generalized’’). There are intermit-
tent bursts of generalized slowing, in the theta or
(more commonly) delta range (Figure 130.38), which
are not attributable to normal circumstances such as
hyperventilation or sleep. A subtype of intermittent
generalized slowing is intermittent rhythmic slowing,
which is usually frontally predominant and thus often
referred as ‘‘FIRDA.’’
� Continuous Generalized Slowing (Classification as
‘‘Continuous Slow, Generalized’’). Here the slowing
is continuous or nearly continuous (>80% of the
Figure 130.36 Periodic pattern, periodic complexes, or generalized periodic epileptiform discharges (GPEDs). This pattern is equally com-
patible with some severe metabolic encephalopathies (e.g., anoxic, uremic, or hepatic) or nonconvulsive status epilepticus. Only the clinical
situation can distinguish among these possibilities, and when in doubt patients with this pattern should be treated for status epilepticus and
monitored for clinical improvement. Classification: periodic pattern, generalized.
1016 INTRODUCTION TO SLEEP ELECTROENCEPHALOGRAPHY
Figure 130.37 Background slowing. Note that there is a posterior dominant background, which is even normally reactive (attenuation with
eye opening in the last 2 seconds). However, the frequency is only 6 Hz, which is too slow (for a subject older than 3 years) and thus is
evidence for a mild diffuse encephalopathy. Classification: background slow.
Figure 130.38 Intermittent generalized slowing. Note the brief (�4 sec) burst of polymorphic delta (�2 Hz) activity. When this is
intermittent or reactive, it is indicative of a mild to moderate diffuse encephalopathy. Classification: intermittent slow, generalized.
THE ABNORMAL EEG 1017
record) and is unreactive. Unreactive means that
it does not change with external stimulation, and
that there are no state changes such as evidence
for drowsiness, sleep, or alerting responses (Fig-
ure 130.39).
� Periodic patterns are generalized discharges that occur
with periodicity (i.e., at regular intervals) and are often
referred to as generalized periodic epileptiform dis-
charges (GPEDs) (Figure 130.35). The discharges
are often sharply contoured, and the periodicity is
most often 1–3 seconds. The etiology is diverse and
includes nonconvulsive status epilepticus [17, 18] as
well as severe metabolic encephalopathies such as
anoxic, uremic, or hepatic disturbances [19, 20, 21].
Another cause of generalized periodic pattern is
Creutzfeldt–Jakob disease.
Burst-suppression (Figure 130.40) is a subtype of
periodic patterns where the activity between
complexes is suppressed (i.e., <10 mV). This is
indicative of an extremely severe degree of ence-
phalopathy, which in fact immediately precedes
eletrocerebral inactivity. It is typically caused by
either drugs (e.g., anesthetic agents), in which
case it is reversible, or anoxia.
Triphasic waves (Figure 130.41) can also be viewed
as a subtype of periodic patterns. These are high-
amplitude surface-positive sharp transients pre-
ceded and followed by lower amplitude negative
components (thus ‘‘triphasic’’) and occur semi-
periodically at 1–3 seconds. They are typically
seen in metabolic encephalopathies, especially
hepatic or renal.
� Background suppression is the absence of any cerebral
activity greater than 10 mV, including with attempts
at activating/stimulating the patient (unreactive).
This is often used for severely ‘‘flat’’ EEG recordings
that do not meet the criteria for electrocerebral
inactivity (see later).
� Electrocerebral inactivity (ECI) (Figure 130.42) is the
EEG pattern of brain death. The criteria are precisely
defined [24], but it should be remembered that, at least
in the United States, brain death is a clinical diagnosis
and EEG is a supportive test (i.e., not required) for
the diagnosis. The main criteria for ECI include no
Figure 130.39 Continuous generalized slowing. Note the polymorphic delta (�2 Hz) activity. When this is continuous (>80% of the
record) and unreactive, it is indicative of a severe diffuse encephalopathy. Classification: continuous slow, generalized.
1018 INTRODUCTION TO SLEEP ELECTROENCEPHALOGRAPHY
Figure 130.41 Triphasic waves. Note the quasiperiodicity, triphasic morphology with a prominent frontal (anterior) positivity, and the
(mild) anteroposterior lag shown by the vertical bold line. Classification: triphasic waves.
Figure 130.40 Burst-suppression. This sample depicts the typical periods of suppression (activity < 10 mV), which in this example last 3–5
seconds, interrupted by ‘‘bursts’’ of activity. Classification: burst-suppression.
THE ABNORMAL EEG 1019
activity greater than 2 mV with double interelectrode
distances, complete lack of reactivity, and exclusion
of confounding factors such as sedative drugs and
hypothermia.
Patterns of Focal Abnormalities
Slow Activity Abnormal slow activity is by far the most
common EEG manifestation of focal brain dysfunction.
Like generalized slow activity, focal slow activity can be
either intermittent (classification as ‘‘Intermittent slow,
regional ___’’) or continuous (classification as ‘‘Continuous
slow, regional___’’). Continuous focal slowing is typically
in the delta range and polymorphic and is also known as
polymorphic delta activity (Figure 130.43). This is a very
strong finding, highly associated with the presence of a
structural lesion (but completely nonspecific as to the nat-
ure of the lesion). Continuous focal slow activity is the only
nonepileptiform abnormality that can unequivocally be
interpreted as abnormal as an isolated finding. By contrast,
intermittent focal slowing (Figure 130.44) is a weak and
nonspecific finding, which can even be normal (e.g., ‘‘tem-
poral slowing of the elderly’’).
As outlined above, focal slowing is completely non-
specific as to etiology, and in the era of imaging the EEG
has no role in diagnosing the nature of a lesion. Focal slow-
ing is the most common abnormality associated with focal
lesions of any type, including (but not limited to) neoplastic,
vascular, subdural collections, traumatic, and infectious [10].
Amplitude Asymmetry In this discussion, the term asym-
metry refers to an asymmetry of amplitude and refers to
normal rhythms. By contrast, a focal frequency asymmetry
would be classified as focal slow.
Destructive lesions can attenuate normal rhythms. How-
ever, normal rhythms are never perfectly symmetric in
amplitude, so when to consider asymmetries significant is
not always clear. A good rule of thumb is that, with very
few exceptions, significant focal asymmetries will be asso-
ciated with slowing. In general, as with other types of focal
EEG abnormalities such as slowing, asymmetry is nonspe-
cific as to etiology.
Although asymmetry in amplitude is usually indicative
of dysfunction on the side of depressed amplitude, one
very common exception to this rule is the so-called breach
rhythm [25] (Figure 130.45). This is caused by a skull
Figure 130.42 Electrocerebral inactivity. There is no cerebral activity greater than 2 mV. Because this is recorded at a sensitivity of 2 mV/mm,
there is prominent EKG and 60 Hz artifact. Classification: electrocerebral inactivity.
1020 INTRODUCTION TO SLEEP ELECTROENCEPHALOGRAPHY
Figure 130.44 Intermittent lateralized slowing in the left hemisphere. Note the brief (�2 second) burst of delta (�2 Hz) activity over the left
hemisphere. This type of focal slowing, unlike that showed in Figure 130.42, is indicative of mild dysfunction and is a ‘‘weak’’ (nonspecific)
abnormality. Classification: intermittent slow, lateralized left hemisphere.
Figure 130.43 Continuous lateralized slowing in the right hemisphere. Note the clear polymorphic delta (�3 Hz) activity over the right
hemisphere. Such focal slow activity, when persistent (>80% of the record), is strongly associated with a structural lesion. Classification:
continuous slow, lateralized right hemisphere.
THE ABNORMAL EEG 1021
Figure 130.45 Breach rhythm. These are two different examples in two patients who are status post craniotomy. Left sample: Note the
increase in amplitude very focally at C3, of what is likely a mu rhythm. Classification: asymmetry, increased mu, left central. Right sample:
Here the increase is in fast (beta) activity at C4. Classification: asymmetry, increased beta, right central. The increased amplitude should not
be mistaken for sharp waves.
Figure 130.46 PLEDs—periodic lateralized epileptiform discharges. These discharges occur periodically, here with a periodicity of 1–1.5
seconds, and are lateralized (here in the right hemisphere). Note that the discharges look to be of low amplitude but in reality are �100 mV
(see scale). Classification: PLEDs, lateralized right hemisphere.
1022 INTRODUCTION TO SLEEP ELECTROENCEPHALOGRAPHY
defect, which attenuates the high-frequency filter function
of the intact skull. As a result, faster frequencies (alpha,
spindles, beta) are of higher amplitude on the side of the
defect. Since morphology is often sharply contoured, deter-
mining the epileptogenicity of these discharges can be
extremely difficult, and in this situation one should prob-
ably err on the conservative side by not interpreting them
as epileptiform. Finally, it should be kept in mind that
amplitude asymmetries are best interpreted on referential
montages, since amplitude is highly dependent on interelec-
trode distances.
PLEDs Periodic lateralized epileptiform discharges
(PLEDs) (Figure 130.46) are a special type of focal
abnormality. As implied by their name, they are periodic,
lateralized, and often epileptiform in morphology. Periodi-
city is the most characteristic feature, and the one that sets
PLEDs apart from other focal abnormalities. Periodicity
refers to a relatively constant interval between two dis-
charges and varies between 0.5 and 3 seconds, most often
around 1 second. The epileptiform morphology of the dis-
charges is not invariable, as PLEDs are often closer to slow
waves than to sharp waves in morphology.
PLEDs are caused by acute destructive focal lesions and
are a transitory phenomenon: they tend to disappear in
weeks, even if the causal lesion persists. With time, periodi-
city and duration increase, and the record takes on a less
specific focal slow appearance that is more likely to persist.
By far the most common etiology is an acute cerebrovascu-
lar event, followed by focal encephalitis such as herpes
[26–28]. Although the periodic patterns of Creutzfeldt–
Jakob disease are usually generalized and bisynchronous,
occasionally, especially early in the course, they may be
unilateral or markedly asymmetric and thus take on the
appearance of PLEDs [29].
As implied by their name, PLEDs have a high associa-
tion with clinical seizures, and on average about 80%
of patients with PLEDs will have clinical seizures. In
clinical practice, PLEDs are usually managed as interictal
discharges (spikes or sharp waves). They indicate a
high risk of focal seizures but are usually not treated with
antiepileptic drugs unless there is clinical evidence for sei-
zures, although there is some controversy about this [10].
REFERENCES
1. Ajmone-Masan C, Zivin LS. Factors related to the occurrence
of typical paroxysmal abnormalities in the EEG records of
epileptic patients. Epilepsia 11:361–381(1970).
2. Tatum WO IV, Winters L, Gieron M, Passaro EA, Benbadis S,
Ferreira J, Liporace J. Outpatient seizure identification: results
of 502 patients using computer-assisted ambulatory EEG.
J Clin Neurophysiol 18:14–19(2001).
3. Benbadis SR, O’Neill E, Tatum WO, Heriaud L. Outcome of
prolonged EEG-video monitoring at a typical referral epilepsy
center. Epilepsia 45:1150–1153(2004).
4. Benbadis SR, Tatum WO IV, Vale FL. When drugs don’t work:
An algorithmic approach to medically intractable epilepsy.
Neurology 55:1780–1784(2000).
5. Anonymous. Patient referral to specialty epilepsy care. Epilep-
sia 31:S10–S11(1990).
6. Benbadis SR, Wyllie E, Bingaman W. Intracranial EEG and
localization studies. In: Wyllie E (Ed), The Treatment of Epi-
lepsy: Principles and Practice, 3rd ed. Lippincott, Williams &
Wilkins, Philadelphia, 2001, pp 1067–1075.
7. Benbadis SR. Psychogenic non-epileptic seizures. In: Wyllie E
(Ed), The Treatment of Epilepsy: Principles and Practice. 4th
ed. Lippincott, Williams & Wilkins, Philadelphia, in press.
8. Lesser RP, Luders H, Dinner DS, Morris H. An introduction
to the basic concepts of polarity and localization. J Clin
Neurophysiol 2:45–61(1985).
9. Benbadis SR, Tatum WO. Over-intepretation of EEGs
and misdiagnosis of epilepsy. J Clin Neurophysiol 20:42–
44(2003).
10. Benbadis SR. Focal disturbances of brain function. In: Levin
KH, Luders HO (Eds), Comprehensive Clinical Neurophysiol-
ogy. Saunders, Philadelphia, 2000, pp 457–467.
11. Niedermeyer E, Lopes da Silva F. Electroencephalography:
Basic Principles, Clinical Applications, and Related Fields,
3rd ed. Williams & Williams, Baltimore, 1993.
12. Luders H, Noachtar S (Eds.) Atlas and Classification of
Electroencephalography. Saunders, Philadelphia, 2000.
13. Smith D, Defalla BA, Chadwick DW. The misdiagnosis of epi-
lepsy and the management of refractory epilepsy in a specialist
clinic. QJM 92:15–23(1999).
14. Eiris-Punal J, Rodriguez-Nunez A, Fernandez-Martinez N,
Fuster M, Castro-Gago M, Martinon JM. Usefulness of the
head-upright tilt test for distinguishing syncope and epilepsy
in children. Epilepsia 42:709–713(2001).
15. Cavazzuti GB, Capella L, Nalin A. Longitudinal study of epi-
leptiform EEG patterns in normal children. Epilepsia 21:43–
55(1980).
16. Chadwick D. Diagnosis of epilepsy. Lancet 336:291–295(1990).
17. Towne AR, Waterhouse EJ, Boggs JG, Garnett LK, Brown AJ,
Smith JR Jr, DeLorenzo, RJ. Prevalence of nonconvulsive
status epilepticus in comatose patients. Neurology 54:340–
345(2000).
18. Treiman DM. Electroclinical features of status epilepticus.
J Clin Neurophysiol 12:343–362(1995).
19. Brenner RP, Schaul N. Periodic EEG patterns: classification,
clinical correlation, and pathophysiology. J Clin Neurophysiol
7:249–267(1990).
20. Husain AM, Mebust KA, Radtke RA. Generalized periodic
epileptiform discharges: etiologies, relationship to status epi-
lepticus, and prognosis. J Clin Neurophysiol 16:51–58(1999).
21. Benbadis SR, Tatum, WO IV. Prevalence of nonconvulsive
status epilepticus in comatose patients (letter). Neurology
55:1421–1423(2000).
REFERENCES 1023
22. Zaidi A, Clough P, Cooper P, Scheepers B, Fitzpatrick
AP. Misdiagnosis of epilepsy: many seizure-like attacks
have a cardiovascular cause. J Am Coll Cardiol 36:181–
184(2000).
23. Benbadis SR. How many patients with pseudoseizures receive
antiepileptic drugs prior to diagnosis? Eur Neurol 41:114–
115(1999).
24. American Electroencephalographic Society. Guideline three:
minimum technical standards for EEG recording in suspected
cerebral death. J Clin Neurophysiol 11:10–13(1994).
25. Cobb WA, Guillof R, Cast J. Breach rhythm: the EEG related
to skull defects. Electroencephalogr Clin Neurophysiol
47:251–271(1979).
26. Schaul N, Green L, Peyster R, Gotman J. Structural determi-
nants of elctroencephalographic findings in acute hemispheric
lesions. Ann Neurol 20:703–711(1985).
27. Schwartz MS, Prior PF, Scott DF. The occurrence and evolu-
tion in the EEG of a lateralized periodic phenomenon. Brain
96:613–622(1973).
28. Snodgrass SM, Tsuburaya K, Ajmone-Marsan C. Clinical signif-
icance of periodic lateralized epileptiform discharges: relation-
ship to status epilepticus. J Clin Neurophysiol 6:159–172(1989).
29. Heye N, Cervos-Navarro J. Focal involvement and lateraliza-
tion in Creutzfeldt–Jakob disease: correlation of clinical,
electroencephalographic and neuropathological findings. Eur
Neurol 32:289–292(1992).
1024 INTRODUCTION TO SLEEP ELECTROENCEPHALOGRAPHY