Ebook Practical guide for clinical neurophysiologic testing EEG (2/E): Part 2

(BQ) Part 2 book Practical guide for clinical neurophysiologic testing EEG has contents: Activation procedures, EEG of premature and full term infants, artifact recognition and technical pitfalls, benign EEG patterns,... and other contents.

Activation Procedures

THORU YAMADA and ELIZABETH MENG

Activation procedures include various sensory and pharmacological stimulations

to alter the physiological state They are usually aimed at eliciting or enhancingabnormal activity, especially epileptiform activity The most commonly usedsensory stimulation is photic stimulation Others include tactile or electricalstimuli for somatosensory stimulation and music or sounds for auditorystimulation Pharmacological activation includes pentylenetetrazol to induce aseizure or benzodiazepine to attenuate one The most routine activationprocedures in any EEG laboratory are hyperventilation (HV), photic stimulation(PS), and sleep

Hyperventilation

This procedure consists of deep and regular breathing at a rate of about 3 to 4 per

10 seconds for a period of 3 to 5 minutes In young children, HV can besuccessfully performed by asking the child to blow on a pinwheel Acharacteristic HV response consists of bilaterally diffuse and synchronous slow-wave bursts, initially with theta frequency and then progressing to deltafrequency This is called “HV buildup” (Figs 9-1A and B and 9-2A and B) Theamplitude may reach as high as 500 μV Theta–delta buildup by HV is usuallyanterior dominant in adolescents or adults but may be posterior dominant inchildren These occur in a serial semirhythmic fashion with fluctuatingamplitude (Video 9-1) The effect is most prominent in children between theages 8 and 12 years and progressively decreases toward adulthood (compare

Figs 9-1B and 9-2B); a clear HV response is seen in about 70% of children, but

in adults, it may be less than 10%.1 HV effects, however, vary considerably fromone individual to another Physiologically, HV reduces the carbon dioxideconcentration (PCO2), which causes vasoconstriction and reduction of cerebralblood flow The reduction of PCO2 (hypocapnia) is likely the major factor in

producing HV buildup.1 HV buildup is enhanced by a blood sugar level below

80 mg/100 mL.2 Therefore, HV buildup may be more prominent when thepatient is hungry or his/her last meal was some time ago In subjects who show alow-voltage and poorly defined alpha rhythm, HV may bring out a better definedalpha rhythm

EEG during hyperventilation (B) in a 38-year-old man Note the

generalized bursts of 4- to 6-Hz theta waves This is unusuallyprominent HV response in this age of patient The frequency of bursts

is faster than that seen in children (Fig 9-1B)

Some delta bursts induced by HV may include “spiky” or spike-likedischarges (small notched spikes preceding or mixed with theta–delta activity)especially in children Unequivocal spikes or clear focal or lateralizing (focal)changes elicited by HV are considered to be abnormal After cessation of HV,the patient may complain of numbness or tingling in the fingers and lips,transient blurring of vision, or ringing in the ears Some may even show changes

of consciousness or awareness These symptoms are self-limiting and are notnecessarily associated with EEG changes or related to the degree of buildup.Likewise, after cessation of HV, the slow waves disappear quickly and the EEGreturns to the pre-HV state within 30 seconds In some subjects, though, theeffect may continue for a minute or longer One should be cautious ininterpreting a long-lasting post-HV effect, since some subjects may continue to

hyperventilate even after being told to stop The technologist should observecarefully to make sure that the patient did indeed stop HV If the delta burstsappear longer than 1 minute in the post-HV period, they are not likely related to

the HV effect (An exception to this is seen in moyamoya disease.)3,4

HV is a well-known activation procedure for inducing absence seizures HVactivates more than 80% of untreated children with absence seizures.5 It isimportant for the technologist to document clinical signs associated with anabsence seizure With a sudden onset of rhythmic (monomorphic) 3-Hz spike–wave bursts (Fig 9-3; Video 9-2, also see Video 10-6), the patient usually stops

HV and often stares into space, sometimes with eyelid or facial muscle twitches

If 3-Hz spike–wave bursts last longer than 5 seconds, the technologist is usuallyable to observe a clinical change by examining the patient’s level ofconsciousness An astute technologist will quickly ask the patient to rememberwords presented during the event and ask the patient after the event if he or shecan recall the presented words If the patient’s communication or consciousness

is impaired, the patient will not be able to recall the word spoken during theepisode A more accurate assessment may be made by testing reaction time; thepatient is instructed to press a button (which makes a mark on the EEGrecording) in response to an auditory signal given by the technologist during theevent of 3-Hz spike–wave bursts With cessation of the spike–wave bursts, thepatient usually resumes HV without being prompted by the technologist There

is no postictal confusion or impaired consciousness

FIGURE 9-3 | Typical 3-Hz spike–wave bursts, characteristic for

absence seizures, induced by hyperventilation in an 8-year-old girl.Apparently, a clinical seizure was associated with the event involvingstaring and blinking as noted by a technologist

HV may also activate focal or other types of generalized seizures, orprecipitate interictal epileptiform activity, though the incidence of suchactivation is far less (~5%) than that for absence seizures Much more vigorousand prolonged HV is usually required to elicit partial seizures.6

HV may accentuate focal slowing, which is sometimes useful for verifyinguncertain or subtle focal features observed in the resting EEG One unique HV

effect has been observed in moyamoya disease in which the delta bursts reappear

3 to 5 minutes after cessation of HV, called the “re-buildup” HV effect.3,4

CONTRAINDICATIONS OF HYPERVENTILATION

The American Clinical Neurophysiology Society (formerly American EEGSociety)7 recommended that HV should not be performed in certain clinicalsettings

Included contraindications are acute stroke, recent intracranial hemorrhage,large-vessel stenosis, recent TIA, moyamoya disease, severe cardiopulmonarydisorders, and sickle cell disease or trait All these conditions are related tocerebrovascular problems

Photic stimulation is a routine activation procedure performed in most EEGlaboratories This is done primarily to elicit a photoparoxysmal response for thediagnosis of photosensitive epilepsy Photic stimulation provides otherphysiological responses but of less diagnostic value

Photic stimulation commonly uses a stroboscope to deliver a repetitive, diffuse,flashing light of brief duration (10 to 30 ms) The repetitive rate is usually 1 to

30 Hz The strobe light is placed directly in front of the patient’s eyes at adistance of about 30 cm The patient’s eyes are usually closed during delivery ofphotic stimulation, but some laboratories prefer to have the eyes open at first andclosed during the middle of photic stimulation (because this may enhance theincidence of the photoparoxysmal response)

The duration of each set of repetitive stimuli is usually 10 seconds, followed

by 10 seconds of resting time before delivering the next set of stimuli Thestimulus rates vary according to laboratory protocols and technologistpreference It is typical to deliver a series of six to eight different frequencies.Some laboratories include crescendo type and decrescendo type of stimulus rateswith a progressively increasing stimulus rate from 1 to 30 Hz and then aprogressively decreasing stimulus rate from 30 to 1 Hz (see Fig 7-26C)

The photic driving responses elicited by 5- to 30-Hz stimuli consist of occipitaldominant rhythmic waves with a one-to-one frequency relationship with eachflash (Fig 9-4A) Within one set of flashes, the driving response frequency maychange from one-to-one to a harmonic or subharmonic pattern (Fig 9-4B) Theresponses are usually most prominent at the flash frequency closest to thefrequency of the individual’s alpha rhythm In a young child, the drivingresponse may be elicited at theta frequency flashes The amplitude of the photicdriving response is generally higher in children and better visualized in adultswith low-voltage background activity The patients who have large lambdawaves and/or POSTS tend to have large photic driving responses (see “LambdaWaves,” Chapter 7; see also Fig 7-26A–C) A photic driving response is absent

in a blind person, but absence of a photic driving response per se is not anabnormal finding It should be noted that the photic response with a slowfrequency (<3 Hz) is not a driving response but rather an evoked potentialelicited by each flash (Fig 9-5) In some subjects, there are diffuse sharpdischarges at the onset or offset of photic stimulation, called the on-response andoff-response, respectively (see Fig 9-4B)

FIGURE 9-4 | Typical photic driving response with 14-Hz photic

stimuli Note the rhythmic activity with photic stimulation The

driving responses fade toward the end of the stimulation (A) In some

individuals, the frequency of the occipital response becomes half

(subharmonic) of the frequency of photic stimuli (B) In this

individual, the driving response started with the same frequency as thestimulus frequency at first but became half the frequency in the middle

of photic stimulation Note the “on-response” by broad sharp transientoccurring approximately 150 to 200 ms after the onset of photic

stimulation (indicated by vertical line).

FIGURE 9-5 | Photic responses at slow-frequency rates In some

individuals, there may be a distinct response at slow-frequency (1- to3-Hz) stimulation Note the small sharp and wave complex in theoccipital electrodes, time locked to, but slightly following, each flash

Examples are shown in the rectangular box This is not a driving

response but rather a photic-induced evoked potential

This was formerly referred to as the “photoconvulsive response,” but the use of this term is now discouraged Photoparoxysmal responses (PPRs) typically

consist of high-amplitude generalized irregular spike–wave or polyspike–wavebursts, with either bianterior or biposterior dominance (Figs 9-6A and B and 9-7A and B; Videos 9-3; see also Video 10-7; Figs 10-6B, 10-18B, and 10-21B).The most effective frequency is 15 to 20 Hz.8 The incidence of PPR is highestbetween ages 6 and 15 and decreases with age.9,10 PPRs correlate highly with adiagnosis of epilepsy; approximately 70% to 80% of patients with PPRs haveepilepsy.11,12 The seizure correlation is especially high in PPRs with spike–wavebursts that persist well beyond termination (>200 ms) of the flash stimulus(sustained PPR) (see Figs 9-6A and 9-7B), as compared to the PPRs in whichthe spike–wave bursts cease at or before the termination of the flash stimulus13(self-limited PPR) (Fig 9-6B) The frequency of PPRs is typically independent

of the photic stimulus rates Some of the atypical photic driving responses mayconsist of spike–wave-like discharges which are time locked and sustained in aone-to-one relationship with the stimulus rate These should not be considered aPPR (Fig 9-8)

FIGURE 9-6 | Example of photoparoxysmal responses In (A),

generalized irregular polyspike–wave bursts started immediately afterthe onset of 11-Hz photic stimulation Spike–waves continued despitequick cessation of photic stimulation (unlimited photoparoxysmal

response) In (B), generalized irregular spike–wave bursts started in

the middle of photic stimulation and ceased despite continuation ofphotic stimulation (self-limited photoparoxysmal response) A

sustained photoparoxysmal response (A) is more epileptogenic and

more likely associated with a history of seizure than the self-limited

photoparoxysmal response (B).

FIGURE 9-7 | Photoparoxysmal response with occipital onset in an

11-year-old girl The photoparoxysmal response started in the occipitalregions with frequency-dependent repetitive sharp discharges andbecame frequency-independent assuming spike–wave discharges

toward the end of photic stimulation (A) In crescendo photic

stimulation, spike discharges started in the occipital electrodes ataround 12-Hz photic stimuli and then evolved to generalized spike–wave bursts, which continued well after the cessation of photic

stimulation (B).

FIGURE 9-8 | Prominent photic responses in a 69-year-old woman.

Note the repetitive spike–wave discharges consistently time lockedwith each flash This is not a photoparoxysmal response

The seizure type most often associated with PPR is generalized tonic–clonic (>80%) Juvenile myoclonic epilepsy (JME) has an incidence of PPR greater

than 1/3.14 Absence and myoclonic seizures represent less than 10% of all

PPRs.15 Partial seizures, especially temporal lobe seizures, associated with PPRsare extremely rare, if any, but may occur with occipital lobe seizures (see Fig.10-6A and B).16 If PPR is observed in a patient with complex partial seizure, thislikely suggests that this patient has both types of seizures: primary generalizedand complex seizures

Some PPRs are more readily activated by a pattern such as dots or stripes.Overall, pattern stimulation is more effective for eliciting PPRs than a diffusestrobe light.17 The color of photic stimulation also affects the degree ofphotosensitivity A red color is more effective for eliciting PPRs than whiteflashes, and blue tinted sunglasses have been recommended for preventingseizure due to photosensitive epilepsy.18 Technologists should be able to quickly

identify the onset of a photoparoxysmal response (epileptiform activity) so thatthe stimulus may be stopped before provoking a clinical generalized tonic–clonicseizure Once the technologist recognizes the PPRs, however, the samefrequency of photic stimulation should be repeated to verify that the evokedepileptiform activity is indeed induced by photic stimulation and is not anincidental occurrence during photic stimulation In this situation, thetechnologist must be extremely alert so as to stop photic stimulation immediatelyupon the onset of a photoparoxysmal response (see Fig 9-6A) The technologistshould also be able to differentiate a photoparoxysmal response fromphysiological variants of photic stimulation (see Fig 9-8) or from a

photomyogenic response (see “Photomyogenic Response”), which is not

considered to be abnormal

One remarkable incidence occurred in December 1997 in Japan;approximately 700 children throughout Japan had seizures almostsimultaneously while watching a television cartoon program called pocketmonster, or “Pokemon.”19 This was apparently caused by alternating red/blueframes flickering at 12 Hz on the TV screen A photosensitive seizure may also

be triggered by playing video games.20

The animal model of photosensitive epilepsy has been studied extensively in

the photosensitive baboon, Papio papio.21

The photomyogenic response (PMR) (formerly referred to as the

“photomyoclonic response” [the term is now discouraged]) consists of EMG

artifacts time locked with the flash frequency (Fig 9-9A) These musclepotentials most often arise from frontal and orbicularis oculi muscles Visiblemuscle twitches, time locked with the stimulus, may appear in the eyelids orface In some occasions, muscle contractions progressively increase (Fig 9-9B),involving larger muscle groups, spreading to the neck or upper body as thestimulus continues This may appear clinically to be a generalized clonic seizure

In addition to the time-locked characteristics, stopping the stimulus willimmediately stop the response in PMR It is important for the technologist tonote if there are any muscle twitches (often eyelid twitches) associated with thePMRs

artifact time locked with the frequency of photic stimulation In (A),

EMG artifacts started abruptly with the onset of photic stimulation andstopped abruptly with the cessation of photic stimulation, whereas in

(B), EMG gradually increased but abruptly stopped with the cessation

of photic stimulation

PMR may be enhanced with alcohol22 or in a barbiturate withdrawal state.23However, PMR is essentially a nonspecific finding and should not be considered

an abnormal response

Hz) may bring out delta activity, called “delta driving” (this is not an evoked

potential),24 enhancing the preexisting underlying slowing (Fig 9-10) In someindividuals, the photic stimulation elicits small spike–waves, which are timelocked to the stimulus (see Fig 9-8) This is an “exaggerated” evoked potentialand should not be confused with a PPR A relatively specific diagnostic pattern

of photic stimulation is high-amplitude spikes in the occipital region, timelocked with a slow stimulus rate in a young child This is a characteristic photic

response for a patient with ceroid lipofuscinosis (Batten’s disease) (Fig

9-11).25,26 In some individuals, a prominent electroretinogram (ERG), which

ordinarily is recorded by a corneal electrode, may be seen at the Fp1 and Fp2electrodes (see Fig 15-14A and B) The ERG consists of two peaks, sharplycontoured A and rounded B waves These waves should also not be confusedwith spike–wave discharges

FIGURE 9-10 | “Delta driving” in a 65-year-old man Waking

background activity in this patient consisted of a mixture of theta and

delta activities as shown by oval circles (A) The 3-Hz photic

stimulation introduced rhythmic 3-Hz delta activity in the occipital

electrodes as indicated by vertical lines (B).

with a diagnosis of Batten’s disease The time-locked spikes with slowphotic stimulation are characteristic for this diagnosis as shown in

rectangular box.

Sleep is an essential tool for activating both generalized and focal interictal

epileptiform discharges (IEDs) The technologist should always try to obtain a

sleep record in patients with suspected seizures If an awake-only EEG shows noIEDs or only questionable IEDs in suspected seizure patients, a repeat EEG with

a sleep record should be justified It is preferable to record an EEG after sleepdeprivation so that a sleep record is achieved without sedation If sedation isrequired, most EEG laboratories prefer chloral hydrate because the medication isrelatively safe, does not increase beta activity, and does not attenuate IEDs.Benzodiazepines or barbiturates are not ideal for obtaining sleep recordingsbecause they tend to produce excessive beta activity and may attenuate the IED

In about one third of patients with partial complex seizures, no IEDs occur inthe awake state and appear only in sleep.27 Focal IEDs tend to be moregeneralized or may become multifocal in sleep Since most sleep activationoccurs in stage I or II sleep, 20 to 30 minutes of stage I and II sleep recordingwill suffice to activate most IEDs.28 In REM sleep, like during the awake EEG,IEDs are decreased or abolished

Some epileptic syndromes have dramatically increased IEDs in sleep

Examples are continuous spike–wave pattern during slow-wave sleep (CSWS),29

Landau–Kleffner syndrome (Fig 9-12A and B),30 and benign epileptiform

central midtemporal spikes (BECTS)31 (Fig 9-13A and B) (see “BenignEpilepsy of Childhood with Central Midtemporal Spikes,” Chapter 10, forfurther details).31

in a 9-year-old boy with diagnosis of Landau–Kleffner syndrome.Note the sporadic sharp-wave discharges from the left hemisphere

(indicated by asterisk in A) in the awake state and more or less

continuous generalized spike–wave bursts becoming electrographic

status epilepticus in sleep (B).

to asleep (B), characteristic for Rolandic spikes in a 10-year-old boy.

In wakefulness, there were sporadic sharp–spike discharges from the

right central and midtemporal regions (shown by asterisk in A) In

sleep, there was a dramatic increase of sharp–spike discharges fromthe left and right central midtemporal regions independently

Sleep deprivation is an effective method for activating IEDs.32 Sleepdeprivation not only promotes sleep but also tends to enhance IEDs In somepatients, IEDs may appear in the awake state only after sleep deprivation.33

Pharmacological activation, therefore, mainly focuses upon the withdrawal

of antiepileptic drugs Due to the risk of seizure recurrence, including statusepilepticus, drug withdrawal has to be done in an inpatient setting and underclose observation in an epilepsy monitoring unit Drug withdrawal may unmask(but not create) a seizure focus that was previously not evident or not in the samelocation of the habitual seizures.34 Also, drug withdrawal may provokegeneralized tonic–clonic convulsions in patients who previously did not havegeneralized seizures Although it is unlikely that drug withdrawal may activate

an entirely new focus, it is important to verify that the recorded seizures areclinically the same as the habitual seizures This is especially important forlocalizing a seizure focus when planning epilepsy surgery

There are rare epileptic occurrences, which are triggered by specific

stimulations, called “reflex epilepsy” and their stimulations include startle,35,36

reading,36,37 listening to music (musicogenic),38 and tasks of decision-making.36Startle epilepsy can be triggered by a sudden, unexpected noise or touch TheEEG pattern may be vertex-dominant spikes or spike/waves Reading epilepsymay be triggered by either silent reading or reading out loud The EEG pattern isusually a generalized spike/wave or focal parietal spikes

In musicogenic epilepsy, music of any type may be epileptogenic, but insome patients, only a specific tone or piece of music triggers a seizure Theremay be additional emotional factors that potentiate seizures The EEG may showanterior temporal spikes

Decision-making–induced epilepsy may be triggered by, for example, chessplaying, solving mathematical problems, or other forms of decision-making TheEEG pattern is usually generalized spike–wave bursts

Other extremely rare activations include eating,39 writing,40 body or limbmovement,41 teeth brushing,42 and tapping.43 If the patient claims that a certainmaneuver or stimulus tends to induce the “spell,” it is important to try toreproduce the same maneuver or stimulus during EEG recording One suchexample is vasovagal syncope (blackout spell), which is often triggered bystanding up or coughing or a Valsalva maneuver Although this is not a seizure,EEG usually shows dramatic diffuse delta slow waves associated with the spell(see Video 10-14) All these activating procedures should be performed with theattendance of a neurologist/electroencephalographer and with simultaneousvideo EEG recording

Takahashi T, Tsukahara Y Pocket Monster incident and low luminance visual stimuli: Special

reference to deep red flicker stimulation Acta Paediatr Jpn 1998;40:631–637.

Harding GF, Fylan F Two visual mechanisms of photosensitivity Epilepsia 1999;40:1446–1451 Killam KF, Killam EK, Naquet R An animal model of light sensitive epilepsy Electroencephalogr

Clin Neurophysiol 1967;22:497–513.

Victor M, Brausch C The role of abstinence in the genesis of alcoholic epilepsy Epilepsia

1967;8:1–20.

Wikler A, Essig CF Withdrawal seizure following chronic intoxication with barbiturates and other

sedative drugs In: Niedermeyer E, ed Modern Problems of Pharmacopsychiatry—Epilepsy Vol 4.

Basel, Switzerland: Karger, 1970:170–184.

Kooi, K Electrographic signs of cerebral disorder In: Fundamentals of Electroencephalography New

Holmes GL, Blair S, Eisenberg E, et al Tooth-brushing induced epilepsy Epilepsia (New York)

1982;23:657–661.

Negrin P, DeMarco P Partial focal spike evoked by tactile somatotopic stimulation in sixty

non-epileptic children: The nocturnal sleep and clinical and EEG evolution Electroencephalogr Clin

Neurophysiol 1977;43:312–316.

EEG and Epilepsy

THORU YAMADA and ELIZABETH MENG

Since the introduction of CT (computerized axial tomography) in the early1970s, neuroimaging diagnostic tests such as MRI (magnetic resonance image),fMRI (functional MRI), PET (position emission tomography), and SPECT(single photon emission computerized tomography) have made remarkableadvancements in revealing the anatomical as well as functional disturbances ofbrain lesions Despite these significant contributions to the diagnosis of variousneurological disorders, EEG continues to play a pivotal role in the diagnosis andmanagement of patients with epilepsy

In evaluating the EEG of a patient with possible seizures, we may see

interictal epileptiform discharges (IEDs) and/or nonspecific paroxysmal

discharges, with or without focal or diffuse slowing IEDs, represented by spike

or spike-wave discharges, are the most sensitive and specific markers for thediagnosis of seizures Other paroxysmal discharges represented sharp, alpha,theta or delta pattern is less specific for diagnosis of epilepsy In a routine EEG(a recording of ~30 minutes), the chance of recording a clinical seizure (ictal)event is rather rare, unless the patient is having frequent seizures or is in statusepilepticus Thus, we often rely primarily on IEDs for the diagnosis of epilepsy.The likelihood of detecting IEDs varies depending on seizure type, age, andseizure frequency An EEG that includes sleep or that is recorded after sleepdeprivation increases the yield of IEDs Generally, greater seizure frequency isassociated with a higher yield of IEDs.1 IEDs are also recorded more often inchildren than in adults Detection of IEDs differs depending on the origin of theepileptiform activity: if a relatively small area of the cortex is involved as theepileptogenic zone, IEDs may not be detected by scalp electrodes Also,epileptiform activity arising from deep brain structures such as the medialtemporal lobe, subfrontal lobe, or interhemispheric medial cortex may not bereadily recorded using scalp electrodes

The specificity of IEDs is determined by the incidence of IEDs in the normalpopulations (false positive) compared with that in patients with epilepsy IEDsare found in 1.9% to 3.5% of healthy children2,3 and 0.5% of healthy adults.4Specificity also varies depending on the type of IEDs: only about 40% of

patients with benign Rolandic spikes of childhood or BECTS (benign epilepsy of

childhood with central midtemporal spikes) and 50% of patients with childhood epilepsy with occipital paroxysms (benign occipital spikes of childhood) have a