Sign in to Peak Brain Path
Sign in to access your courses, books, and progress tracker. New here? Signing in creates your account automatically.
Want to explore courses first?
Browse courses and booksSign in to access your courses, books, and progress tracker. New here? Signing in creates your account automatically.
Want to explore courses first?
Browse courses and booksModule 8
Chapter 8 · 2.5 h · 8 quiz items · pass at 80%
This module is the abnormal-EEG core of IQCB Domain IV (EEG), 18% of the exam. A QEEG practitioner must recognize the epileptiform and pathological patterns that require referral and must not be quantified as if normal, and must avoid over-reading benign variants. The quiz confirms the learner can name an abnormal pattern by its defining criteria and state its clinical significance.
You bought a QEEG system to read brain maps, not to call seizures. So why does a certification didactic spend a chapter on spikes, triphasic waves, and burst-suppression?
Because the quantitative pipeline runs on the raw trace. Before any epoch reaches the fast Fourier transform, you look at the EEG. If you cannot recognize a focal slow-wave field, a run of frontal rhythmic delta, or an epileptiform discharge riding underneath the alpha, two things go wrong. You feed pathological data into a normative comparison that assumes a healthy waking adult, and you produce z-scores that describe an artifact of disease rather than a phenotype. And you miss the one finding that obligates you to stop, set the database aside, and route the recording to a physician.
This chapter teaches pattern recognition for two jobs: cleaning the record so the QEEG you build is honest, and knowing when a pattern is outside your scope and belongs with a board-certified electroencephalographer. Be clear with yourself about the boundary. A QEEG credential, at any level, does not license you to diagnose epilepsy, read a clinical EEG for seizure activity, or determine brain death. The standard descriptions below let you flag, describe, and refer. They do not let you interpret an epileptiform tracing as the clinical document of record. When in doubt, the recording goes to someone whose scope covers it.
The descriptive vocabulary used here follows the standard clinical-EEG references, principally Niedermeyer and da Silva's Electroencephalography for waveform morphology and the American Clinical Neurophysiology Society's standardized terminology for periodic and rhythmic patterns (Niedermeyer & Lopes da Silva, 2005; Hirsch et al., 2021). Where a specific prevalence, localizing value, or prognosis is stated, treat it as provisional until the citation gate clears it.
Every abnormality you will learn here sorts first along one axis. Is it focal, meaning it occupies a region and respects a boundary, or is it diffuse, meaning it covers both hemispheres more or less symmetrically?
Focal abnormalities point at tissue. A slow-wave field maximal at T5, dropping off toward the midline and toward the back of the head, says something is wrong under the left posterior temporal region. The EEG cannot tell you whether that something is a tumor, an old infarct, an abscess, or a resolving contusion, but it localizes the disturbance, and localization is the EEG's oldest and most reliable contribution.
Diffuse abnormalities point at the whole brain or at systems that drive the whole brain. Bilateral, symmetric slowing reflects a global process: a metabolic derangement, a toxin, a sedating drug load, a degenerative disease, or a falling level of consciousness. The pattern is not localizing because the cause is not local.
The two axes interact. A patient can have a diffuse encephalopathy and a focal lesion at the same time, and the focal finding can hide inside the diffuse slowing until you look for asymmetry rather than absolute frequency. Hold the distinction in mind as the organizing question for the rest of the chapter, because it determines both what the finding means and what you do next.
Slow activity that stays in one place, over one region, is the EEG signature of a regional disturbance of cortical and subcortical function. Two morphologies matter, and they carry different implications.
Polymorphic delta activity (PDA). Irregular, arrhythmic delta (under 4 Hz) that varies continuously in frequency, amplitude, and waveform, maximal over a region and persistent across the recording. Continuous, regional, polymorphic delta is the classic correlate of an underlying structural lesion involving the white matter beneath that region: tumor, infarct, abscess, contusion, or a focal area of cortical destruction (Niedermeyer & Lopes da Silva, 2005). The activity does not react to eye opening or alerting, which separates it from drowsiness-related slowing. When you see a delta field that ignores state changes and stays put, you are likely looking at a structural problem, and that recording needs a clinical read.
Focal rhythmic delta. More monomorphic, more sinusoidal, sometimes intermittent. Regional rhythmic delta can accompany deeper or less destructive pathology and is less specific for a fixed structural lesion than polymorphic delta (Niedermeyer & Lopes da Silva, 2005). One regional rhythmic pattern with a name of its own, temporal intermittent rhythmic delta activity, gets its own treatment below because of its epilepsy association.
The localizing value of focal slowing is real but coarse. The field maps to a region, not to a millimeter, and the EEG cannot read the pathology off the waveform. Your job is to recognize the field, describe its location and persistence, note its reactivity, and keep the involved channels out of any normative QEEG comparison you build, because regional delta will throw absolute power and asymmetry z-scores at exactly the sites it occupies.
On the quantitative side, focal slowing appears as a regional excess of absolute delta and theta power, often with a corresponding amplitude asymmetry toward the involved hemisphere and disrupted coherence between the affected region and its neighbors (Niedermeyer & Lopes da Silva, 2005). The map will light up. That QEEG picture is a description of the lesion's electrical footprint, not a phenotype to be matched against a database of healthy brains, and your report has to say so.
When the slowing is bilateral and roughly symmetric, the question shifts from "where" to "how bad and why." Diffuse slowing is the EEG's measure of global cerebral dysfunction, and it grades with the depth of that dysfunction (Niedermeyer & Lopes da Silva, 2005).
The progression runs in a recognizable direction. A healthy waking adult shows a well-organized posterior dominant rhythm in the alpha range that reacts to eye opening. As global function declines, the background slows and disorganizes: the posterior rhythm drops in frequency, theta intrudes into the background, then delta, and the reactivity to stimulation fades. Deep encephalopathy can reach a continuous, unreactive delta state, and beyond that lies burst-suppression and, at the limit, suppression of all cerebral activity. The faster the background at the time of recording, the better the cerebral function, as a rule (Niedermeyer & Lopes da Silva, 2005).
Three causal categories sit under the heading of diffuse slowing, and they overlap in their EEG appearance more than the textbook boxes suggest.
Metabolic encephalopathy. Derangements of the body's chemistry that the brain depends on: renal failure, hepatic failure, hypoglycemia, hyponatremia, hypoxia, hypercapnia. The shared EEG signature is diffuse slowing that often progresses to bilateral synchronous delta, sometimes frontally predominant, and the classic morphology of triphasic waves (covered below) appears across several of these states, most famously hepatic encephalopathy (Niedermeyer & Lopes da Silva, 2005).
Toxic encephalopathy. Drugs and poisons. The pattern depends on the agent. Many sedatives produce a paradoxical sequence: low doses drive prominent diffuse fast activity, particularly beta, before higher doses or accumulation tip the background into slowing (Niedermeyer & Lopes da Silva, 2005). Benzodiazepines and barbiturates are the cleanest examples of the beta-then-slowing continuum, which is why a sedating drug load is one of the first confounds you rule out before attributing diffuse slowing to disease.
Anoxic and post-anoxic states. Loss of oxygen delivery to the brain produces some of the gravest EEG patterns, including burst-suppression, periodic discharges, and the alpha-coma pattern, all discussed below. After cardiac arrest, the EEG carries prognostic weight, which is precisely why interpreting it sits with the treating neurologist and not with the QEEG technologist (Niedermeyer & Lopes da Silva, 2005).
The unifying clinical point: diffuse slowing tells you the brain is globally underperforming, it grades severity, and it does not name the cause by itself. A QEEG built on diffusely slow data will show a global excess of delta and theta and a deficit of alpha across all sites, which is a faithful description of a sick brain and a meaningless comparison against a healthy-adult norm. Recognize the state, describe it, and do not pretend the database output is a phenotype.
Here the scope boundary is sharpest. Epileptiform discharges are the interictal markers associated with a tendency toward seizures, and their identification, interpretation, and clinical correlation belong to a board-certified electroencephalographer. You learn the morphology so you can recognize a discharge in your record, stop, and refer. You do not learn it so you can tell a patient anything about epilepsy.
The morphologic definitions follow the standard EEG terminology (Niedermeyer & Lopes da Silva, 2005).
Spike. A transient, clearly distinguished from background, with a pointed peak and a duration of 20 to under 70 milliseconds. Spikes are followed by a slow wave and carry a surface-negative polarity at the cortical maximum (Niedermeyer & Lopes da Silva, 2005).
Sharp wave. The same kind of transient with a duration of 70 to 200 milliseconds. The line between spike and sharp wave is purely the duration. The clinical implication is similar (Niedermeyer & Lopes da Silva, 2005). The discipline of measuring duration is what separates a true sharp wave from the many normal sharp transients that are physiologic.
Spike-and-slow-wave complex. A spike followed by a slow wave, occurring singly or in runs. The repetition rate and the morphology of the complex carry syndromic information that, again, is read by the electroencephalographer.
Polyspike and polyspike-and-wave. Two or more spikes in succession, sometimes followed by a slow wave, associated particularly with the generalized epilepsies such as juvenile myoclonic epilepsy (Niedermeyer & Lopes da Silva, 2005).
Recognizing that a sharp transient is epileptiform, rather than one of the many benign sharp variants, rests on a set of standard features the field uses to separate real discharges from look-alikes: a di- or triphasic waveform that stands out from and interrupts the background, a duration and morphology unlike the ongoing rhythms, an after-going slow wave, a physiologic field that spreads across more than one electrode, and a polarity that makes anatomic sense (Niedermeyer & Lopes da Silva, 2005). A transient that meets these features is treated as epileptiform. A sharp blip confined to one electrode with no field and no after-going slow wave is not. You are not certifying these calls, but knowing the criteria is what lets you tell a probable discharge (refer) from a probable artifact or benign variant (do not over-read).
Two distinctions organize the epileptiform world.
Focal versus generalized. A focal discharge has a regional field, points at a region of cortex, and raises the question of a focal epilepsy or a structural focus. A generalized discharge appears bilaterally, synchronously, and symmetrically across the head, and is associated with the generalized epilepsies.
The named syndromic patterns you should be able to recognize by sight:
3 Hz spike-and-wave. Generalized, bilaterally synchronous spike-and-wave complexes repeating at about three per second, the electrographic signature classically associated with typical absence seizures in children, provoked by hyperventilation (Niedermeyer & Lopes da Silva, 2005). The clinical correlation, a brief behavioral arrest with the discharge, is part of what defines it, and that correlation is a clinical-EEG observation outside QEEG scope.
Hypsarrhythmia. A chaotic, high-amplitude, disorganized pattern of multifocal spikes and slow waves on a grossly abnormal background, the interictal EEG associated with infantile spasms in the first year or two of life (Niedermeyer & Lopes da Silva, 2005). It is one of the most disordered tracings in clinical EEG. A QEEG normative comparison on an infant with hypsarrhythmia is not a meaningful exercise. The recording is a referral, urgently.
Why a QEEG practitioner cares beyond referral: even a single unrecognized spike inside an epoch contaminates the spectral estimate at that site and at neighboring sites through volume conduction, inflating power in a way that has nothing to do with the resting rhythms you meant to quantify. Epileptiform transients are an absolute exclusion from any epoch you carry forward into the QEEG. There are software approaches that flag spikes and even quantify discharge frequency, but flagging is not interpreting, and the clinical meaning stays with the physician.
Intermittent rhythmic delta activity (IRDA) is a distinctive pattern: runs of relatively monomorphic, sinusoidal delta that come and go against a background that may be otherwise organized. It is named by location, and the location changes what it suggests.
FIRDA (frontal IRDA). Bilateral, frontally predominant rhythmic delta, in the 1.5 to 2.5 Hz range, in runs. In adults, FIRDA is a nonspecific marker. It accompanies diffuse encephalopathies (metabolic, toxic), increased intracranial pressure, and deep midline or posterior fossa lesions, and it is not, by itself, an epileptiform or localizing finding (Niedermeyer & Lopes da Silva, 2005). The classic teaching is that FIRDA points to a generalized disturbance or a deep structural process rather than to frontal cortex specifically.
OIRDA (occipital IRDA). The occipitally predominant counterpart, seen more in children, and associated in that population with generalized epilepsy, including absence epilepsy, more than the adult frontal pattern is (Niedermeyer & Lopes da Silva, 2005).
TIRDA (temporal IRDA). Rhythmic delta over the temporal region. TIRDA carries a stronger and more specific association: in adults it is linked with temporal lobe epilepsy and is treated as having localizing and epileptogenic significance approaching that of interictal temporal spikes (Niedermeyer & Lopes da Silva, 2005). Of the three rhythmic-delta patterns, TIRDA is the one most likely to mean focal epilepsy, and recognizing it as different from benign drowsy temporal slowing matters.
For your purposes, the practical sorting is: FIRDA, nonspecific, usually a sick or pressured brain; OIRDA, pediatric, generalized-epilepsy associated; TIRDA, focal temporal, epilepsy-associated and the one with the sharpest referral implication. All three are runs of rhythmic delta that will distort a power spectrum if they land in your analysis epochs, and all three describe states a normative database was not built to model.
A patient with a skull defect, after craniotomy, burr hole, or fracture, can show a striking focal pattern over the defect that the unprepared reader mistakes for pathology. This is breach rhythm.
The skull normally attenuates and smooths the EEG. Where the bone is missing, the underlying activity reaches the electrode less filtered, so the amplitude rises sharply over the defect, the higher frequencies are accentuated, and the waveforms look sharper and faster, sometimes with a spiky appearance (Niedermeyer & Lopes da Silva, 2005). The temptation is to call the sharp, high-amplitude focal activity epileptiform.
The defense against over-reading is the history and the topography. Breach rhythm sits exactly over a known skull defect, it is dominated by accentuated normal frequencies rather than by true spike-and-slow-wave complexes, and it makes sense given a surgical or traumatic history. The change is in the skull, not the brain. The corollary for QEEG is direct: a craniotomy site will produce a regional amplitude and high-frequency excess on the map that is a conduction artifact, not a cortical finding, and any z-scores over that region have to be read with the defect in mind, or the channels excluded.
Periodic patterns are waveforms or complexes that repeat at a regular interval. The terminology here was standardized to replace older, inconsistently used names, and the current ACNS framework describes these patterns by their laterality, type, and morphology (Hirsch et al., 2021). The two you should recognize:
LPDs (lateralized periodic discharges). Periodic, repetitive sharp or sharply-contoured complexes over one hemisphere or region, recurring at a roughly fixed interval, on the order of one per second. LPDs were long known as PLEDs (periodic lateralized epileptiform discharges), and the literature still uses both names (Hirsch et al., 2021). They are associated with acute or subacute focal destructive lesions, classically acute stroke, herpes encephalitis, and other rapidly evolving focal pathology, and they sit close to the seizure end of the spectrum, appearing in patients who have or have recently had focal seizures (Niedermeyer & Lopes da Silva, 2005). LPDs are a clinical emergency in their usual context and are unambiguously outside QEEG interpretive scope.
GPDs (generalized periodic discharges). The bilateral, synchronous, generalized counterpart. GPDs appear in severe diffuse processes, including anoxic injury, certain toxic-metabolic states, and the periodic complexes of Creutzfeldt-Jakob disease, which are taken up in the next chapter (Niedermeyer & Lopes da Silva, 2005; Hirsch et al., 2021).
Non-convulsive status epilepticus (NCSE). Ongoing seizure activity on the EEG without the obvious motor signs of a convulsion, in a patient who may simply appear confused, unresponsive, or "not themselves." NCSE is an under-recognized and serious cause of altered mental status, it can only be confirmed electrographically, and the EEG patterns that define it sit on a continuum with the periodic and rhythmic patterns above, which is part of what makes the diagnosis a specialist's (Hirsch et al., 2021). The reason NCSE belongs in a QEEG didactic at all is cautionary: a patient referred for a brain map because they are cognitively "off" could, rarely, be in non-convulsive status, and a quantitative summary is the wrong tool and the wrong scope for that question. If the raw EEG shows evolving rhythmic or periodic activity rather than resting rhythms, you stop and refer.
Burst-suppression is a pattern of deep cerebral depression: high-amplitude bursts of mixed activity, often with sharp or spike components, alternating with intervals of near-flat suppression in which cerebral activity drops to low voltage, conventionally under 10 microvolts (Niedermeyer & Lopes da Silva, 2005). The ratio of burst to suppression, and the length of the suppression intervals, index the depth of the state.
Two opposite situations produce it, and the difference is the whole clinical point.
It is induced, deliberately, by deep anesthesia or by high-dose sedatives given to suppress refractory seizures or to lower cerebral metabolic demand. In that setting, burst-suppression is a therapeutic target titrated against the EEG, and its depth is controlled by the drug.
It also appears, undesired, in severe anoxic or hypoxic brain injury, where it reflects grave global dysfunction and carries prognostic weight after cardiac arrest (Niedermeyer & Lopes da Silva, 2005). The same morphology means something benign and controllable in one context and something dire in another, and only the clinical context and the treating physician resolve which.
For the QEEG practitioner the lesson is brief: burst-suppression is not a resting state, it is not analyzable as a phenotype, and recognizing it tells you the patient is heavily sedated or critically injured. Either way the recording is not a candidate for normative QEEG, and the interpretation is not yours.
Two patterns at the severe end deserve naming, with a firm boundary around what you may do with them.
Electrocerebral inactivity (ECI). The absence of identifiable cerebral electrical activity above a defined low-voltage threshold, recorded under a strict technical protocol. ECI is part of the ancillary determination of brain death, and the recording that establishes it has demanding technical requirements: a full array of electrodes, raised electrode distances and amplifier sensitivity, documentation of integrity, exclusion of drug effect and hypothermia, and a specified recording duration, all defined by guideline (Niedermeyer & Lopes da Silva, 2005; Hirsch et al., 2021). This is named here so you recognize the term and the gravity. Performing or interpreting an ECI recording for brain-death determination is entirely outside any QEEG scope and is governed by separate certification and protocol.
Alpha coma. The EEG paradox where a comatose, unresponsive patient shows a widespread alpha-frequency pattern that superficially resembles the alpha of a relaxed, awake adult. The resemblance is only in the frequency. Alpha-coma alpha is diffuse rather than posterior, anteriorly distributed, monotonous, and crucially unreactive to stimulation, where normal waking alpha is posterior and blocks with eye opening (Niedermeyer & Lopes da Silva, 2005). It occurs after anoxic injury and with certain brainstem lesions and toxic states, and it carries a poor prognosis in the post-anoxic setting (Niedermeyer & Lopes da Silva, 2005). The teaching value for QEEG is the reactivity test: alpha that does not react is not the alpha of wakefulness, no matter what the frequency reads, and a quantitative pipeline that scores "normal alpha power" off such a record is being fooled by the spectrum.
Triphasic waves are a distinctive morphology: medium-to-high-amplitude waves with three phases, classically a small initial negative deflection, a dominant positive deflection, and a final negative deflection, occurring in bilateral, frontally-predominant runs, with an anterior-to-posterior or posterior-to-anterior time lag across the head (Niedermeyer & Lopes da Silva, 2005).
They are the textbook signature of metabolic encephalopathy, most associated with hepatic encephalopathy but seen across renal failure, other metabolic derangements, and some toxic states (Niedermeyer & Lopes da Silva, 2005). The morphology overlaps with periodic discharges and with some epileptiform patterns, and distinguishing triphasic waves of metabolic encephalopathy from generalized periodic discharges or from blunted spike-and-wave is a genuine interpretive challenge that the standardized terminology was partly designed to address (Hirsch et al., 2021), and that lands squarely with the electroencephalographer.
For QEEG, triphasic waves are one more reason a confused, encephalopathic patient is a poor candidate for a normative brain map: the runs distort the spectrum, the state is not a trait, and the morphology itself is a clinical-EEG call.
Everything above is described first as raw-trace morphology, because that is where you must catch it. But each pattern also leaves a quantitative footprint, and knowing the footprint serves two ends: it sometimes draws your eye back to a raw-trace abnormality you skimmed past, and it keeps you from reporting a disease signature as a phenotype.
Focal slowing surfaces as a regional excess of absolute delta and theta power, an amplitude asymmetry toward the lesioned side, and locally disrupted coherence between the involved region and its neighbors (Niedermeyer & Lopes da Silva, 2005). The map shows a hot region. The cause is structural. The z-scores describe the lesion, not the person's resting brain.
Diffuse slowing surfaces as a global excess of delta and theta with a global deficit of alpha and beta across all sites, the same pattern the dementia and encephalopathy literature describes as spectral slowing, with the posterior peak frequency dropping below its normal range (Niedermeyer & Lopes da Silva, 2005). A database built on healthy waking adults will read this as deviation everywhere, which is true and uninformative as a phenotype.
Epileptiform transients and periodic patterns are, from the spectral point of view, contaminants. A spike inflates power at its site through its broadband energy. Runs of periodic or rhythmic delta load the low-frequency bands. Breach rhythm inflates regional amplitude and high frequencies through the skull defect. None of these belong in an analysis epoch, and the discipline of artifact-free epoch selection, taught in the QEEG methodology chapters, is the same discipline that keeps them out.
The throughline: the QEEG is only as honest as the trace it is built from. Pattern recognition is the gatekeeper.
The mirror image of missing an abnormality is calling something abnormal that is not. Clinical EEG is full of benign variants with sharp or unusual morphology that catch the inexperienced eye, and naming the common ones protects you from manufacturing pathology out of normal physiology (Niedermeyer & Lopes da Silva, 2005).
A short list of the variants most often over-read, named here so you recognize them and leave them alone:
The unifying feature of most benign variants is their tie to drowsiness and light sleep, which is exactly the state your eyes-closed resting recordings drift toward. Misreading a wicket spike or a vertex sharp wave as an epileptiform discharge is a classic novice error, and the standard references catalog these variants precisely so they are not over-called (Niedermeyer & Lopes da Silva, 2005). The distinction is, once again, a clinical-EEG judgment, which is the recurring theme of this chapter.
Recognizing abnormal patterns is not a parallel credential in clinical EEG, and nothing here authorizes you to read a tracing for seizures or to tell a patient anything diagnostic. It is the gate on two of your real jobs. The first is producing an honest QEEG: the moment a focal delta field, a run of rhythmic delta, an epileptiform transient, a periodic pattern, or a breach rhythm appears in your record, those data are excluded from the epochs you carry into the spectral analysis, because a normative database built on healthy waking adults cannot model them and will return z-scores that describe disease or artifact rather than a phenotype. The second is the referral: when the raw EEG shows evolving rhythmic or periodic activity, deep depression, unreactive alpha, or any of the discharges above, you stop, describe what you see in neutral morphologic terms, and route the recording to a board-certified electroencephalographer whose scope covers what yours does not. On the exam, Domain IV will ask you to name the pattern, sort it focal versus diffuse, and state its general correlate. In the clinic, the same skill keeps your maps clean and your patients safe.