Can an EEG detect past seizures?

An electroencephalogram (EEG) is a test that detects electrical activity in the brain using small metal discs (electrodes) attached to the scalp. EEG is commonly used to diagnose and monitor epilepsy and seizures. It can record brain wave patterns during seizures and between seizures. But can an EEG detect evidence of seizures that have occurred in the past, even if one is not occurring during the test?

Table of Contents

How Does EEG Work?

EEG records the summation of synchronized electrical activity from thousands or millions of neurons that have similar spatial orientation. The electrodes detect fluctuations resulting from ionic current flows within the neurons of the brain. EEG measures voltage fluctuations resulting from ionic current within the neurons of the brain.

In clinical contexts, EEG refers to the recording of the brain’s spontaneous electrical activity over a period of time, as recorded from multiple electrodes placed on the scalp. Diagnostic applications generally focus on the spectral content of EEG, that is, the type of neural oscillations (popularly called “brain waves”) that can be observed in EEG signals.

EEG Wave Patterns

EEG activity demonstrates oscillations at a variety of frequencies. Several of these oscillations have characteristic frequency ranges, spatial distributions and are associated with different states of brain functioning. These oscillations represent synchronized activity over a network of neurons. The neuronal networks underlying some of these oscillations are understood (e.g., the thalamocortical resonance underlying alpha waves), while many others are not fully understood.

The main frequency bands are:

Delta waves (0.5–3 Hz)

Theta waves (4–7 Hz)
Alpha waves (8–12 Hz)
Mu waves (8–13 Hz)

Beta waves (12–30 Hz)
Gamma waves (25–100 Hz)

EEG in Epilepsy

EEG is a critical tool for evaluating, monitoring, and managing patients with epilepsy. It plays a role in:

Diagnosing epilepsy – EEG can help determine whether episodes of unusual movement, sensation, behavior or loss of awareness are due to epilepsy.
Classifying seizures – EEG patterns during and between seizures help classify seizure type.
Identifying epilepsy syndrome – Specific EEG patterns can help identify certain epilepsy syndromes that have characteristic features.

Locating seizure focus – EEG can identify the region of the brain where seizures originate.
Monitoring anti-seizure treatment – EEG provides feedback on the effectiveness of medications or other treatments for controlling seizures.
Assessing surgery candidacy – Candidates for epilepsy surgery have prolonged EEG monitoring to precisely characterize their seizures.

EEGs are also used to investigate epilepsy in infants and children. However, EEG patterns evolve across development, so pediatric EEG interpretation requires specialized expertise.

Interictal Epileptiform Discharges

In between seizures, the EEG of a person with epilepsy may reveal intermittent abnormal spikes called interictal epileptiform discharges (IEDs). These reflect the epileptogenicity of the brain tissue generating the abnormal discharge. IEDs have a characteristic appearance:

They are transient, distinct waves or complexes that stand out from the background EEG activity.

They have a pointed shape with a narrow base and duration of <80 msec.
Their polarity generally differs from the background rhythm.

IEDs can arise independently or in a repetitive pattern. They may be generalized or focal. Certain IED locations, morphology and distribution correlate with particular epilepsy syndromes. IEDs can fluctuate in frequency over time. Their presence can help establish an epilepsy diagnosis and seizure focus even if clinical seizures were not captured on EEG.

Postictal EEG

The postictal EEG represents the state of the brain in the period immediately following a clinical or electrographic seizure. Characteristic postictal EEG patterns can provide insight into the region and structures involved in seizure activity.

In the immediate 10-30 seconds after a seizure, the postictal EEG usually shows diffuse slowing and attenuation of brain wave activity. This represents depressed brain function following the intense, hypersynchronous discharges of the seizure.

Other postictal patterns that may emerge later include:

Regional slowing – reflecting impaired function in the seizure onset zone.
Spike wave activity – indicating residual hyperexcitability in the epileptogenic region.
Electrodecremental events – brief bursts of slowing/attenuation, suggesting dysfunction of deeper brain structures.

Electrocerebral inactivity – complete lack of discernible brain activity for a period of time.

Postictal EEG allows the affected brain regions and networks to be mapped. This helps elucidate seizure semiology as well as predict future seizure risk and Postictal EEG patterns also provide information relevant to classification of seizure type.

Seizure Detection from EEG

EEG is the gold standard test for assessing seizures as they occur. However, EEG has limitations in capturing seizures. EEG monitoring may go on for days before a clinical seizure happens to be recorded. Or there may be electrographic seizures seen on EEG that have no clinical manifestation.

Can EEG show evidence that a seizure happened in the recent past, even if it was not recorded at the time? This is an area of active research. There are changes in cerebral activity and metabolism following a seizure that may leave detectable “fingerprints” in the EEG.

Some ways past seizures might be identifiable on later EEG include:

Periodic discharges – spike-and-wave or sharp-and-wave patterns repeating at regular intervals, reflecting postictal neuronal instability.

Amplitude suppression – reduced EEG amplitude, indicating depressed brain activity following a seizure.
Cerebral blood flow changes – altered perfusion and metabolism linked to recent seizures may modulate EEG signal.
Network connectivity – seizures may cause transient changes in functional connectivity between brain regions that are measurable on EEG.

Researchers are also using quantitative EEG analysis and machine learning approaches to try to detect signatures of recent seizures. However, no EEG patterns have been validated as definitive biomarkers of past seizures. Ongoing studies aim to assess the sensitivity and specificity of proposed EEG biosignatures.

Prolonged EEG Monitoring

Because EEG provides only a brief sample of brain activity, it often fails to capture spontaneous seizures. Prolonged EEG monitoring over days or weeks increases the likelihood of recording seizures and maximizes diagnostic yield.

There are several approaches to prolonged EEG monitoring:

Routine EEG – 20-40 minutes of recording as an outpatient.
Ambulatory EEG – 1 to 4 days of portable EEG recording at home or work.
Inpatient video-EEG monitoring – Continuous EEG monitoring for 1-2 weeks in an epilepsy monitoring unit (EMU).

Among patients admitted to the EMU, over 80% will show epileptiform abnormalities on their first routine EEG. About 50-70% of EMU admissions will record spontaneous clinical seizures, while up to 90% may reveal subclinical electrographic seizures. Thus, inpatient video-EEG monitoring substantially increases diagnostic yield compared to standard EEG.

Prolonged recordings are more likely to capture patterns potentially suggestive of seizures in the recent past, such as periodic discharges or cerebral hypoactivity. But there remains a lack of definitive validation of EEG biomarkers that confirm past clinical seizures.

EEG Limitations in Detecting Past Seizures

While EEG is excellent at capturing real-time ictal activity, there are limitations to using it as a detector of previous seizures:

EEG changes reflecting seizures may be very subtle/transient.
Artifacts or normal variations can resemble postictal patterns.
EEG findings lack specificity – changes could have other causes.

Unknown time course – unsure how long postictal EEG changes persist.
Research is early – proposed biomarkers need more validation.
No way to determine precise timing of presumed earlier seizure.

In particular, the potential EEG indicators of past seizures described above also occur in a variety of other conditions unrelated to seizures. For example periodic discharges may be seen with encephalopathy of various causes, including metabolic disorders. And amplitude suppression or cerebral hypoactivity can result from sedative medications, among other etiologies.

So while EEG shows promise for indirect seizure detection, the evidence base remains limited. EEG still functions best for directly capturing real-time ictal activity as it occurs. Wider validation is needed before EEG biomarkers of prior seizures can be applied in clinical practice.

Other Tools for Detecting Past Seizures

Given the limitations of EEG, what other tools may potentially serve as markers of previous seizure activity? Some options include:

fMRI – Changes in blood flow and metabolism from recent seizures visible on functional MRI.
MRI – Signs of recent neuronal injury may point to earlier unobserved seizures.
Implanted electrodes – Recordings from implanted cortical or depth electrodes can help identify earlier subclinical seizures.

Biomarkers – Chemical biomarkers indicative of recent seizures may be measurable in blood or other body fluids.
Smartwatches – Wearable devices with seizure detection algorithms could log information about timing of suspected seizures.

However, most of these approaches currently lack sufficient evidence. Implanted electrodes are invasive and not routinely feasible. fMRI and MRI changes are nonspecific. Biomarker and wearable tech research remains in early stages.

As of now, these modalities serve only as adjuncts to EEG monitoring. But future research may enhance their utility for post hoc seizure detection.

Conclusion

In summary, EEG is the gold standard test for evaluating clinical and subclinical seizures as they occur. But its ability to determine if a seizure happened in the recent past is more limited. While there are EEG patterns that may suggest a prior seizure, these lack specificity.

Prolonged EEG monitoring can increase detection of interictal and ictal epileptiform discharges. But the evidence base for EEG biomarkers that confirm earlier seizures remains sparse. EEG still functions best for real-time seizure recording rather than retroactive detection.

Ongoing research seeks to validate EEG signatures and other diagnostic tools that could serve as markers of antecedent seizures. But for now, history taking and clinical correlation remain central for assessing prior suspected seizure activity that was not captured on EEG. In the future, EEG or other modalities may provide additional confirmatory data about previous unobserved seizures.