Quantitative EEG (qEEG) transforms the visual EEG signal into mathematical features — power in frequency bands, asymmetry indices, coherence between regions, spectral edge frequencies — that can be tracked numerically over time. In clinical practice, qEEG is most useful in continuous EEG monitoring in the ICU, where it compresses 24 hours of EEG into a single screen, allows rapid detection of seizures and trend changes, and provides quantitative endpoints for treatment response. qEEG should never replace visual EEG review for diagnosis, but it adds an analytic layer that human visual review cannot match. This page covers the major qEEG techniques, their ICU and outpatient applications, and the pitfalls that limit interpretation.

Spectral Analysis (Fast Fourier Transform)

  • EEG is decomposed into its frequency components using Fourier analysis.
  • Power in each frequency band quantified: delta (0.5–4 Hz), theta (4–8 Hz), alpha (8–13 Hz), beta (13–30 Hz), gamma (>30 Hz).
  • Displayed as power spectra, density spectral array (DSA), or compressed spectral array (CSA).
  • Tracks the dominant frequency over time — a sensitive index of cortical activity.

Quantitative Measures

Spectral Edge Frequency (SEF)

  • The frequency below which a specified percentage (e.g., 95%) of EEG power lies.
  • SEF 95 is a common index of cortical activity.
  • Decreases in slow encephalopathies, anesthesia, coma.

Median Frequency / Peak Frequency

  • Median frequency: the frequency that divides EEG power 50/50.
  • Peak frequency: the single frequency with the most power.
  • Track over time for treatment response.

Power in Specific Bands

  • Absolute power: μV² in each band.
  • Relative power: percent of total power in each band.
  • Useful for tracking treatment effect (e.g., burst-suppression ratio during anesthetic infusion).

Asymmetry Index

  • Quantifies left-right or anterior-posterior amplitude/power differences.
  • Can detect subtle focal slowing or seizure activity.
  • Continuous monitoring of asymmetry indices alerts to lateralized changes.

Coherence

  • Statistical measure of synchronization between two channels at a given frequency.
  • High coherence = related signals; low coherence = independent signals.
  • Used to characterize network behavior, especially in epilepsy and consciousness research.

Burst-Suppression Ratio

  • Percentage of an epoch in suppression vs activity.
  • Standard endpoint for refractory status epilepticus treatment (target 50–80%).
  • Quantifiable, reproducible, and continuous.

Compressed Spectral Array (CSA / DSA)

  • Spectrogram display: time on x-axis, frequency on y-axis, color = power.
  • Compresses hours of EEG into a single visual.
  • Allows rapid detection of:
    • Seizure activity (sudden frequency change, bursts).
    • Periodic discharges (rhythmic pattern in spectrogram).
    • Background changes (overall slowing or speeding).
    • Trend over time.
  • Standard in modern ICU continuous EEG.

Trends Useful in ICU EEG

Alpha-Delta Ratio

  • Ratio of alpha power to delta power.
  • Sensitive to ischemic injury — decreases hours before clinical decline.
  • Useful for monitoring patients at risk of delayed cerebral ischemia after subarachnoid hemorrhage.

Suppression Ratio

  • Percentage of epoch with very low EEG activity.
  • Targeted endpoint for thiopental coma in status epilepticus.
  • Continuous tracking for ICU staff.

Cerebral Function Monitor (CFM)

  • Older device for trending EEG over hours.
  • Uses amplitude-integrated EEG (aEEG).
  • Now largely replaced by modern qEEG software.

Amplitude-Integrated EEG (aEEG)

  • Single-channel rectified, smoothed, time-compressed signal.
  • Standard in neonatal ICU for brain function monitoring.
  • Captures background activity, sleep-wake cycling, and seizure activity in compressed display.
  • Sensitivity for neonatal seizures: ~50–80% depending on operator experience.

Spectral Analysis in Epilepsy

  • Quantitative search for spike-wave discharges via specific frequency patterns.
  • Some software (e.g., Persyst seizure detection) provides automated marking.
  • Not a replacement for human review; complements visual interpretation.
  • Source localization (LORETA, sLORETA) uses spectral analysis combined with anatomic head models.

QEEG in Specific Clinical Settings

ICU Continuous EEG (cEEG)

  • Compressed spectral array displays 24 hours on one screen.
  • Real-time alerts for: seizure activity, rhythmic patterns, decline in background.
  • Allows physician/nurse oversight without continuous EEG technician review.
  • Standard of care in modern neuroICU monitoring.

Subarachnoid Hemorrhage Monitoring

  • Alpha-delta ratio tracks risk of delayed cerebral ischemia.
  • Decrease in alpha-delta ratio precedes clinical decline by hours.
  • Triggers earlier intervention.

Anesthesia and Sedation Monitoring

  • Bispectral Index (BIS): processed EEG number reflecting anesthetic depth.
  • SedLine and similar: spectrogram-based depth-of-anesthesia monitor.
  • Endpoint for procedural sedation, anesthesia depth in OR, and ICU sedation.

Post-Cardiac Arrest Prognostication

  • Background EEG features (suppression, periodic discharges, burst-suppression) carry prognostic weight.
  • Spectral features (low SEF, decreased alpha-delta ratio) corroborate visual findings.
  • Used in combined model with SSEP and clinical exam.

Status Epilepticus

  • Visual identification of seizures supplemented by qEEG alerts.
  • Continuous EEG during anesthetic-infusion treatment.
  • Burst-suppression ratio guides titration.

Outpatient and Research qEEG

Dementia

  • Quantitative EEG shows slowing of dominant frequency, decreased alpha, increased theta and delta in Alzheimer disease.
  • Sensitive to disease progression but not specific for diagnosis.
  • Research applications; not standard clinical practice.

Concussion / mTBI

  • Decreased alpha, increased theta and delta in early phase.
  • Resolution over weeks.
  • Some commercial qEEG systems marketed for concussion evaluation; clinical utility debated.

ADHD and Mood Disorders

  • Various qEEG profiles described.
  • Clinical utility for diagnosis: minimal evidence.
  • Not standard practice; should not be used alone for diagnosis.

Limitations of qEEG

  • Artifact contaminates spectral analysis (muscle, ECG, electrode pop, line noise).
  • Subtle epileptiform activity (sharp waves, spikes) may not be detected by spectral methods alone.
  • Specific patterns (e.g., morphology of IEDs) are lost in spectral averaging.
  • “Brain mapping” with topographic spectral maps can be misleading without raw EEG correlation.
  • Commercial qEEG systems for non-ICU clinical use have limited evidence base.
  • Cannot replace clinical context — a slow PDR in metabolic encephalopathy may have the same spectral signature as that in dementia, but the clinical differential is entirely different.

Quality Standards for Clinical qEEG

  • Visual review of raw EEG essential before interpreting any quantitative measure.
  • Artifact rejection or epoch exclusion required.
  • Standardized protocols for filter settings, sampling rate, recording duration.
  • Awareness of pharmacologic effects that affect spectral analysis.
  • Clinical correlation always required.

Modern ICU Continuous EEG Workflow

  1. Patient placed on continuous EEG monitor with full 10-20 array.
  2. Real-time qEEG display shows compressed spectral array, alpha-delta ratio, seizure detection alerts.
  3. Bedside RN or physician reviews the qEEG display.
  4. EEG technician reviews raw EEG periodically (e.g., every 8 hours).
  5. Neurologist reviews raw EEG once daily and after any alert.
  6. Treatment adjustments based on combined raw and qEEG findings.

🔍 Did You Know?

The alpha-delta ratio (ADR) on quantitative EEG has emerged as one of the most clinically valuable qEEG markers for predicting delayed cerebral ischemia (DCI) after subarachnoid hemorrhage. Studies have shown that a sustained decline in ADR of more than 10% precedes clinical signs of DCI by 12–48 hours — providing a window for therapeutic intervention before neurologic decline becomes irreversible. In a typical ICU workflow, the patient with SAH is placed on continuous EEG with real-time ADR display; nurses and physicians monitor the trend; a downward shift triggers earlier ultrasound, repeat imaging, or treatment adjustment. This is a beautiful example of how qEEG transforms a complex visual signal into a single trended number that bedside staff can monitor continuously without specialized training. The same approach is being extended to other conditions: ADR trends in traumatic brain injury, asymmetry indices for stroke evolution, suppression ratios in status epilepticus management. The lesson is that qEEG’s greatest value is in continuous monitoring, not single-point diagnosis: it lets us see trends that would be invisible to intermittent visual review. For practicing neurophysiologists, the practical implication is that modern neuroICU competence increasingly requires fluency in qEEG dashboard interpretation, not just visual EEG reading.

Pitfalls and Pearls

  • Spectral analysis: decomposes EEG into frequency bands; quantifies power in delta, theta, alpha, beta, gamma.
  • Compressed spectral array (CSA/DSA): 24 hours on one screen; standard for ICU EEG.
  • Spectral edge frequency (SEF): continuous index of cortical activity.
  • Alpha-delta ratio: ICU monitoring for delayed cerebral ischemia post-SAH.
  • Burst-suppression ratio: endpoint for thiopental treatment of status epilepticus.
  • aEEG: standard in neonatal ICU; single-channel time-compressed display.
  • BIS / SedLine: anesthesia depth monitors; processed EEG-based.
  • qEEG complements but does not replace visual EEG review.
  • Artifact contaminates spectral analysis: reject artifact epochs before interpreting.
  • Visual confirmation required for any qEEG-detected pattern before treatment changes.
  • Subtle epileptiform activity: may not be detected by spectral methods alone.
  • Outpatient qEEG for psychiatric/cognitive disorders: limited clinical evidence.
  • ICU workflow: continuous EEG with qEEG dashboard + intermittent visual review.
  • Asymmetry indices: detect lateralized changes that visual review might miss in compressed time.
  • Coherence: research applications; not standard clinical.
  • Clinical context essential: same spectral pattern can mean different things in different patients.

References

  1. Foreman B, Claassen J. Quantitative EEG for the detection of brain ischemia. Crit Care. 2012;16(2):216.
  2. Hirsch LJ, Brenner RP. Atlas of EEG in Critical Care. Wiley-Blackwell; 2010.
  3. Claassen J, Hirsch LJ, Kreiter KT, et al. Quantitative continuous EEG for detecting delayed cerebral ischemia in patients with poor-grade subarachnoid hemorrhage. Clin Neurophysiol. 2004;115(12):2699-2710.
  4. Hirsch LJ, Fong MWK, Leitinger M, et al. American Clinical Neurophysiology Society’s Standardized Critical Care EEG Terminology: 2021 Version. J Clin Neurophysiol. 2021;38(1):1-29.
  5. Toet MC, Lemmers PMA. Brain monitoring in neonates. Early Hum Dev. 2009;85(2):77-84.
  6. Nuwer M. Assessment of digital EEG, quantitative EEG, and EEG brain mapping: report of the American Academy of Neurology and the American Clinical Neurophysiology Society. Neurology. 1997;49(1):277-292.