10 Conditions Diagnosed With an EEG

An electroencephalogram records the brain’s electrical activity through sensors placed on the scalp, and clinicians rely on it to diagnose or help diagnose a wide range of neurological conditions. Epilepsy is the one most people associate with the test, but EEG findings also guide decisions in stroke care, coma prognosis, brain death determination, and several conditions that look nothing like seizures. Below are ten conditions where EEG plays a meaningful diagnostic role, along with what the test actually reveals in each case and where its limits lie.

Epilepsy

Epilepsy is the condition EEG was essentially built for, and it remains the single most common reason the test is ordered. Between seizures, the brain of a person with epilepsy often produces characteristic bursts of abnormal electrical activity called epileptiform discharges. These sharp waves and spikes, visible on the recording even when the person feels perfectly fine, help confirm a diagnosis that might otherwise rest entirely on a witness description of what the seizure looked like. Careful attention to the location and shape of those discharges lets neurologists distinguish focal epilepsies, which start in one brain region, from generalized epilepsies, which involve both hemispheres from the outset. The distinction matters because treatment and prognosis differ between the two.

A generalized pattern on EEG does not automatically mean the epilepsy is generalized. Secondary bilateral synchrony, where a focal discharge spreads rapidly to both sides of the brain and mimics a generalized pattern, can mislead clinicians who are not looking closely at where the discharge begins.

Childhood absence epilepsy offers a textbook example of how specific EEG can be: the hallmark is a roughly three-per-second generalized spike-and-wave discharge that appears during the brief staring spells these children experience.

Psychogenic Nonepileptic Seizures

Some people experience episodes that look exactly like epileptic seizures, with shaking, unresponsiveness, and falls, yet have no abnormal electrical discharge in the brain during the event. These are psychogenic nonepileptic seizures, and diagnosing them correctly prevents years of unnecessary antiseizure medication. Video EEG monitoring, where the patient is recorded on camera simultaneously with brain activity, is the gold standard for telling the two apart.

The challenge is that you need to capture an actual event on camera while the EEG is running. In one study of over 600 patients admitted to an epilepsy monitoring unit, about a quarter were ultimately diagnosed with psychogenic nonepileptic seizures, and of those, roughly 44% had a typical episode within the first eight hours of monitoring. The rate was higher in children, with about 55% having an event within that window.

⁴

For those whose episodes are less frequent, monitoring may need to continue for days. The payoff is substantial: a clear diagnosis allows patients to be redirected from epilepsy drugs to psychological treatment, which is the appropriate intervention.

Encephalopathy

Encephalopathy is a broad term for any diffuse brain dysfunction that impairs alertness or thinking, and EEG is one of the most sensitive tools for detecting it. The hallmark finding is a slowing of the brain’s normal background rhythms. In a healthy, awake adult, the dominant rhythm over the back of the head runs at about 8 to 13 cycles per second. When that rhythm drops to slower frequencies, or disappears altogether, it signals that the brain is not working properly. The degree of slowing tracks roughly with how confused the patient is.

⁵

This pattern shows up regardless of the cause, whether it is kidney failure, liver failure, a severe infection, or medication toxicity. Certain medications like lithium, baclofen, and methotrexate can produce encephalopathy with diffuse slowing and loss of normal sleep architecture on EEG.

⁶

One particular waveform deserves mention: triphasic waves, which are large, three-phased sharp deflections repeating once or twice per second. They have long been associated with hepatic encephalopathy, the brain dysfunction caused by liver failure, but they also appear in other metabolic derangements, infections, and even nonconvulsive seizures. That overlap means triphasic waves alone do not pin down a cause; they tell the clinician that something serious is affecting the brain diffusely.

⁷

Brain Death

Determining that a patient’s brain has irreversibly stopped functioning is among the highest-stakes decisions in medicine, and EEG is one of the ancillary tests used to support that determination. The standard is called electrocerebral inactivity: a 30-minute recording that shows no brain-generated electrical activity above two microvolts anywhere on the scalp.

⁸

Because this threshold is so tiny, technical standards are strict. Electrodes must be spaced at least 10 centimeters apart, sensitivity settings are pushed far higher than normal, and the technologist applies loud sounds and physical stimulation to make sure the brain does not respond to any outside input.

⁹

Critically, clinicians must first rule out sedative drugs, severe metabolic problems, and very low body temperature, all of which can temporarily suppress brain activity enough to mimic electrocerebral inactivity without the brain actually being dead.

¹⁰

EEG is just one part of the brain death evaluation; the clinical examination, including tests for brainstem reflexes and the ability to breathe independently, carries the primary weight. But when the clinical exam is difficult to complete, say because facial injuries prevent testing certain reflexes, EEG can provide corroborating evidence.

Coma After Cardiac Arrest

When someone survives a cardiac arrest but remains in a coma, one of the most pressing questions is whether the brain has a chance of recovering. EEG has become a central part of that prognostic assessment. A large multicenter study of 1,000 post-cardiac-arrest patients found that suppression, where background activity drops below 10 microvolts, was the most common pattern seen in those who did not recover, with roughly 76 to 83% of patients showing that pattern going on to have a poor outcome. Still, about one in five patients with suppression had a favorable outcome, so no single EEG pattern is an absolute death sentence.

¹¹

One pattern does come close: burst-suppression with identical bursts, where the brain produces periodic bursts of activity that look exactly the same each time, separated by stretches of silence. In one study, every single patient who displayed this pattern after cardiac arrest had a poor outcome, while only about a third of those with non-identical burst patterns did.

¹²

The timing of the EEG also matters. Predictive accuracy tends to improve over the first few days after the arrest, which is why many protocols call for serial recordings rather than a single snapshot.

Creutzfeldt-Jakob Disease

Creutzfeldt-Jakob disease is a rare and rapidly fatal brain disease caused by misfolded proteins called prions. EEG has a specific role here because the disease produces a distinctive pattern: periodic sharp wave complexes, which are brief, high-amplitude discharges repeating at regular intervals across both hemispheres. These complexes are considered both sensitive and specific for CJD.

¹³

The catch is timing. These discharges typically do not appear until about 8 to 12 weeks after symptoms begin, and in some cases even later.

¹⁴

A normal EEG early in the illness does not rule CJD out. MRI and cerebrospinal fluid testing are often more useful in the earliest stages, but once those periodic complexes appear, they are a powerful confirmation. They also tend to disappear in the very late stages of the disease, so the window for capturing them is finite.

Autoimmune Encephalitis

Autoimmune encephalitis occurs when the immune system mistakenly attacks the brain, and the most well-known form involves antibodies against the NMDA receptor. These patients can develop psychiatric symptoms, seizures, involuntary movements, and a decline in consciousness that can last weeks or months. EEG is abnormal in nearly all cases and has revealed a pattern that appears to be fairly unique to this condition: extreme delta brush.

The name comes from the waveform’s resemblance to a pattern normally seen only in premature infants. About 30% of adults with anti-NMDA receptor encephalitis display extreme delta brush, and its presence has been associated with longer and more severe hospitalizations.

¹⁵

Extreme delta brush is now considered a potential marker for this disease, which means spotting it on an EEG can prompt clinicians to order antibody testing they might not have considered otherwise.

¹⁶

In a condition where early treatment with immunotherapy dramatically improves outcomes, that kind of early diagnostic clue matters a great deal.

Stroke

EEG is not the first test most people associate with stroke, since brain imaging with CT or MRI is the initial go-to. But EEG is highly sensitive to the disruption in blood flow that a stroke causes, and changes can show up within minutes of onset. The most common finding is focal slowing: the region of the brain affected by the stroke generates abnormally slow electrical activity compared to the other side.

¹⁷

In one study of stroke patients, about half had abnormal EEGs, with focal slowing being the most frequent finding. Patients with anterior circulation strokes, which involve the front and middle portions of the brain’s blood supply, were roughly three and a half times more likely to show abnormalities on EEG compared to those with posterior circulation strokes. Higher stroke severity scores were independently associated with the presence of focal slowing.

¹⁸

Where EEG adds unique value in stroke care is in detecting seizures that can complicate recovery. Some post-stroke seizures are subtle or purely electrical, with no visible convulsions, and continuous EEG monitoring is the only way to catch them. There is also growing interest in quantitative EEG, which uses computer analysis to calculate ratios of slow to fast activity and detect changes in blood flow that correlate with the amount of brain tissue at risk.

¹⁹

Brain Tumors and Other Structural Lesions

Before modern brain imaging became widely available, EEG was one of the main ways to localize a brain tumor. Imaging has long since taken over that role, but EEG remains relevant because many patients with brain tumors develop seizures, and the EEG helps characterize those seizures and guide treatment. Both low-grade and high-grade brain tumors commonly produce focal slowing on EEG, particularly in the delta range, reflecting disruption to the normal electrical activity of nearby brain tissue.

²⁰

This type of slowing, called polymorphic delta activity, is a general sign that something is structurally wrong in the area producing it. Modeling studies have shown that even a localized patch of this slow activity can disrupt communication across distant brain regions.

²¹

EEG does not replace imaging for finding the tumor, but it does help answer the question of whether the tumor is causing seizures and, if so, where those seizures originate, which matters for surgical planning.

Developmental Epileptic Encephalopathies in Infants

In young infants, certain severe epilepsy syndromes produce EEG patterns so distinctive that they are part of the diagnostic criteria. Infantile epileptic spasms syndrome, formerly called West syndrome, is the most common example. The signature EEG pattern is hypsarrhythmia: a chaotic, high-amplitude background with multifocal sharp waves and spikes that looks dramatically different from anything seen in a healthy infant. In one comparative study, EEG scoring systems flagged features suggesting epileptic encephalopathy or hypsarrhythmia in about three-quarters of infants with spasms.

²²

EEG also has emerging value in predicting outcomes for these infants. Quantitative analysis of pre-treatment recordings has identified a measure, alpha-band event-related spectral power, that correlates with whether a child will become free of spasms and develop normally. Higher values before treatment began were linked to better outcomes across multiple measures.

²³

For families, this kind of prognostic information can guide decisions about how aggressively to pursue treatment and what level of developmental support to put in place early.

When EEG Misleads

For all its utility, EEG has a well-documented weak spot: overinterpretation. Normal brain activity includes a whole catalog of waveforms that look sharp and suspicious but are completely benign. These normal variants, which include patterns with names like wicket waves, small sharp spikes, and six-per-second spike-and-wave discharges, can fool even experienced readers into diagnosing epilepsy where none exists.

²⁴

In one EEG laboratory’s review, benign epileptiform variants appeared in about 12% of all recordings, with wicket waves and small sharp spikes being the most common offenders.

²⁵

One useful rule of thumb for EEG readers: genuine epileptic spikes are almost always followed by a slow wave, while most benign variants are not. If a sharp-looking waveform repeats with a stable shape and a single consistent rhythm throughout the recording, it is more likely a normal variant than a sign of epilepsy.

²⁶

Misdiagnosis in this direction is not trivial. Patients labeled with epilepsy on the basis of an over-read EEG may take anticonvulsant drugs for years, face driving restrictions, and carry a diagnosis that affects their employment and self-image, all unnecessarily. The takeaway for anyone handed an epilepsy diagnosis based on a single ambiguous EEG finding: a second opinion from a specialist in epilepsy is worth pursuing.

Investigational Uses on the Horizon

Quantitative EEG, where computers break the raw signal into numerical measures of power, coherence, and frequency ratios, has expanded the test’s reach into areas where conventional visual reading of the tracing is not enough. One active area of research is traumatic brain injury. Standard EEG often looks normal after a concussion, which is frustrating for patients who clearly feel that something is wrong. Quantitative analysis can detect subtle shifts, such as increased slow-wave power and decreased faster activity, that persist well beyond the initial injury.

²⁷

The enthusiasm here runs ahead of the evidence, though. A guideline from the American Clinical Neurophysiology Society found that current evidence does not support the routine clinical use of quantitative EEG for diagnosing mild traumatic brain injury, calling it an investigational tool based on the available literature.

²⁸

Machine learning models are being developed that may eventually change this, but for now, a concussion cannot be reliably diagnosed by EEG alone in a clinical setting.

Another area generating interest is ADHD. Research has shown that children with ADHD tend to have higher ratios of slow theta activity to faster beta activity in frontal brain regions compared to children without the condition. One pilot study found that this theta-to-beta ratio could meaningfully discriminate between children with ADHD and healthy controls, though the effect varied with age.

²⁹

The FDA cleared a quantitative EEG-based device as an aid to ADHD diagnosis several years ago, but the test supplements rather than replaces a comprehensive behavioral evaluation. The clinical applications of quantitative EEG continue to expand across neuropsychiatric conditions, stroke, and dementia, though the evidence base varies widely by condition.

³⁰

1
PubMed. Recognition of interictal and ictal discharges on EEG. Focal vs generalized epilepsy ↩︎
2
Epilepsy Research. EEG–fMRI study on the interictal and ictal generalized spike-wave discharges in patients with childhood absence epilepsy ↩︎
3
PubMed Central. Long-term video EEG monitoring for diagnosis of psychogenic nonepileptic seizures ↩︎
4
PubMed. Diagnostic Yield of 8-Hour Video-EEG in Detecting Psychogenic Non-Epileptic Seizures (PNES) ↩︎
5
MedLink Neurology. EEG in encephalopathic conditions ↩︎
6
Practical Neurology. Electroencephalography in encephalopathy and encephalitis ↩︎
7
Medical alphabet. Triphasic waves on electroencephalogram in patients with encephalopathy and their diagnosis significance. A review ↩︎
8
PubMed. EEG guidelines in the diagnosis of brain death ↩︎
9
Annals of Clinical Neurophysiology. Electroencephalography for the diagnosis of brain death ↩︎
10
PubMed. EEG guidelines in the diagnosis of brain death ↩︎
11
PubMed Central. Prognostic Significance of EEG Patterns in Post-Cardiac Arrest Coma: A Multicenter Retrospective Cohort Study of 1,000 Patients ↩︎
12
PubMed. Burst-suppression with identical bursts: a distinct EEG pattern with poor outcome in postanoxic coma ↩︎
13
PubMed. Disappearance of periodic sharp wave complexes in Creutzfeldt-Jakob disease ↩︎
14
PubMed. Early detection of periodic sharp wave complexes on EEG by independent component analysis in patients with Creutzfeldt-Jakob disease ↩︎
15
PubMed Central. Extreme delta brush A unique EEG pattern in adults with anti-NMDA receptor encephalitis ↩︎
16
PubMed Central. Using EEG and MEG to characterize extreme delta brush in a patient with anti-NMDA receptor encephalitis ↩︎
17
Clinical Neurophysiology Practice. Electroencephalography in stroke Care: From acute monitoring to Post-Stroke epilepsy ↩︎
18
PubMed Central. Qualitative electroencephalogram and its predictors in the diagnosis of stroke ↩︎
19
Clinical Neurophysiology Practice. Electroencephalography in stroke Care: From acute monitoring to Post-Stroke epilepsy ↩︎
20
American Epilepsy Society. EEG CHARACTERISTICS OF LOW-GRADE AND HIGH-GRADE GLIOMAS OF THE CENTRAL NERVOUS SYSTEM ↩︎
21
PubMed. Local polymorphic delta activity in cortical lesions causes global decreases in functional connectivity ↩︎
22
PubMed. EEG Scoring Systems in Developmental Epileptic Encephalopathies With and Without Epileptic Spasms: A Comparative Analysis of Reliability and Diagnostic Clarity ↩︎
23
PubMed. Alpha-ERSP as a prognostic biomarker in infantile epileptic spasms syndrome: Insights from quantitative EEG analysis ↩︎
24
PubMed. Normal variants and artifacts: Importance in EEG interpretation ↩︎
25
PubMed. Prevalence of benign epileptiform variants from an EEG laboratory in India and frequency of their misinterpretation ↩︎
26
PubMed Central. Overcoming traps and pitfalls leading to misinterpretation of normal EEG variants and variation of the background activity ↩︎
27
PubMed. Quantitative Electroencephalography Objectivity and Reliability in the Diagnosis and Management of Traumatic Brain Injury: A Systematic Review ↩︎
28
PubMed. Practice Guideline: Use of Quantitative EEG for the Diagnosis of Mild Traumatic Brain Injury: Report of the Guideline Committee of the American Clinical Neurophysiology Society ↩︎
29
PubMed. Quantitative electroencephalography in children with attention deficit hyperactivity disorder and healthy children: Behavioral and age correlates ↩︎
30
PubMed Central. The Role of Quantitative EEG in the Diagnosis of Neuropsychiatric Disorders ↩︎