Amplitude-integrated EEG (aEEG)
A bandpass-filtered, rectified, time-compressed EEG envelope, the workhorse of the NICU and PICU bedside for the long view.
1. Bedside vignettes: why this matters
Vignette A. Term newborn, day 2 of cooling for HIE
A 39-week neonate with a sentinel uterine rupture, Apgars 1/3/5, cord pH 6.83. Therapeutic hypothermia (target core 33.5 °C) started at 2 h of life. At 24 h on the Olympic CFM you see a discontinuous trace: upper margin 20–25 µV, lower margin dipping to 3–4 µV every 8–10 seconds. By 48 h the lower margin has lifted to 7 µV, the trace is now continuous normal voltage, and a faint sleep-wake cycle is appearing. The clinical question: is the imaging at 72 h going to find diffuse injury, or is this child recovering?
Vignette B. School-age status epilepticus in the PICU
An 8-year-old presents in convulsive status, treated with lorazepam, levetiracetam, midazolam infusion. Clinical motor activity stops; you put two leads on (C3-P3, C4-P4) for an aEEG strip while waiting for the cEEG technologist. The aEEG envelope shows rhythmic broadening of the upper margin every 90 seconds: NCSE. The midazolam infusion is up-titrated. When the cEEG arrives 90 minutes later it confirms ongoing electrographic seizures arising from the right central region. The aEEG bought time.
Vignette C. The "lush" envelope that is not reassurance
A 6-month-old infant post-cardiac-arrest after a near-drowning. At 36 hours the aEEG looks continuous, voltages 15–35 µV, no obvious burst-suppression. The team is reassured. The neurologist asks for raw cEEG: it shows alpha-coma rhythm, a low-amplitude monomorphic alpha frequency unresponsive to noise, eye opening, or pain, classic of severe brainstem and thalamic dysfunction after global hypoxia. The aEEG envelope cannot read frequency content; in alpha-coma it looks deceptively normal. Outcome at 6 months: profound neurodisability.
2. What aEEG is, and what it is not
aEEG is not a separate measurement. It is a display transform of the raw EEG signal, designed so that hours of activity collapse onto a single trend line that a bedside nurse or PICU fellow can read without an electroencephalographer in the room.
The pipeline, in four steps:
- Bandpass filter, 2–15 Hz. This is the most consequential step. It cuts the low-frequency artefact (sweat, eye movement, slow drift) and the high-frequency contamination (EMG, mains noise) while preserving the frequencies where most clinically relevant background activity lives. It also throws away delta-dominant patterns; very slow rhythms (< 2 Hz) are systematically under-represented.
- Full-wave rectify. Negative deflections are flipped positive; the absolute amplitude of the signal becomes the input to the next stage.
- Smooth and envelope-detect. A short moving average smooths the rectified signal; the upper margin of the envelope is the peak-to-peak amplitude per epoch (~15-second bin), the lower margin is the floor.
- Plot on a semi-logarithmic y-axis, time-compressed. The y-axis runs ~0–100 µV with a linear segment up to ~10 µV and a logarithmic segment above. The x-axis is compressed: a single screen shows 6 hours (one channel) or 12–24 hours (multi-channel).
What aEEG does well
- Long view: compresses hours of activity into a glance.
- Pattern recognition: five canonical patterns are taught in under an hour.
- Bedside-friendly: a two-lead montage that a nurse can place; output requires no FFT or montage knowledge.
- Trend prognostication: time to return of sleep-wake cycling in cooled HIE is one of the strongest non-imaging prognostic markers in the first 72 h.
- Seizure burden quantification: rhythmic upper-margin elevations, even brief, flag NCSE and high-burden electrographic seizures.
What aEEG cannot do
- Focal seizures with limited spatial spread can be invisible: the two-lead montage covers a small fraction of cortex.
- Frequency content is collapsed into amplitude only: alpha-coma, theta-coma, periodic patterns are unreadable on the envelope alone.
- Artefact (movement, ECG, ventilator, high-frequency oscillator) elevates the lower margin and mimics CNV; an envelope that "looks fine" can be misleading without raw EEG cross-check.
- Adult cortex generates lower-amplitude background than neonatal cortex; aEEG thresholds for CNV/DNV are derived from neonatal data and do not transfer directly to adults.
aEEG is a trend tool, not a montage. Treat it like a continuous BP cuff: useful for hours of drift, hopeless for the single beat-to-beat decision. When something looks abnormal on aEEG, open the raw EEG window. When it looks fine, periodically open the raw EEG window anyway.
The five aEEG patterns (Hellström-Westas classification) were defined for term neonates. In preterm infants ≤ 32 weeks GA the background is naturally discontinuous, the lower margin sits near 2–3 µV, and the trace tracks gestational-age maturation more than acute injury. A trace that would mean "severely abnormal" in a term newborn is normal at 28 weeks. The Burdjalov score, not the Hellström-Westas classification, is the right tool for preterm aEEG interpretation.
3. Anatomy: where the leads go
The standard aEEG montage is a single bipolar channel: P3-P4 in neonates (parietal-parietal), or two bipolar channels (C3-P3 and C4-P4) on multi-channel CFM devices. The two-lead choice is deliberate: most NICU staff can place two leads reliably in under 5 minutes, and the parietal montage straddles watershed territories where global injury shows up earliest.
| Lead | 10–20 location | Why this site |
|---|---|---|
| P3-P4 (neonate single channel) | Bilateral parietal | Watershed; captures global cortical activity; minimal frontal artefact |
| C3-P3 / C4-P4 (dual channel) | Centro-parietal each side | Adds hemispheric symmetry; surfaces left-vs-right asymmetry |
| F3-T3 / F4-T4 (adult ICU) | Frontotemporal | Used in adult ICU sedation monitors; less prognostic data in pediatrics |
Reference electrodes (Cz or AFz) and the ground electrode complete the array. Leads are typically subdermal needles (preterm) or hydrogel patches (term and older). Impedance < 5 kΩ is the bedside target; > 20 kΩ produces unreadable artefact.
Asymmetry on a two-lead aEEG can be a real focal injury or an artefact of one bad electrode. Before calling left-vs-right asymmetry pathological, check impedances and the raw trace.
4. The signal: anatomy of the envelope
The envelope has three readable features.
- Upper margin: the peak amplitude in each ~15-second epoch. Continuous well-developed background has an upper margin > 10 µV in the term newborn and > 25 µV in the older child.
- Lower margin: the trough amplitude. A lower margin > 5 µV in the term newborn is the threshold for "continuous"; below 5 µV is "discontinuous".
- Bandwidth: the vertical distance between upper and lower margins. Narrow bandwidth (< 5 µV) with both margins low signals CLV or FT; very wide bandwidth (> 30 µV) with periodic interruptions signals BS.
Sleep-wake cycling (SWC) appears as a slow, regular undulation of the lower margin every 20–60 minutes: a more discontinuous "quiet sleep" phase alternates with a more continuous "active sleep" phase. SWC normally appears by 30 weeks GA in preterm infants and is robust by term. Its disappearance is an early marker of acute injury; its reappearance after HIE injury is the strongest envelope-based predictor of favourable outcome.
A SWC index > 0.3 is a clean cycle; flat (< 0.1) means no cycling.
5. The numbers to record: the aEEG six-pack
For every patient, on every shift, log this six-pack into the chart:
| Variable | Symbol | What it tells you |
|---|---|---|
| Upper margin (mean over 1 h) | UM | Background voltage; falls in injury, sedation, post-ictal suppression |
| Lower margin (mean over 1 h) | LM | Continuity; the key threshold for CNV vs DNV |
| Bandwidth (UM − LM) | BW | Narrow + low = CLV/FT; wide + periodic = BS |
| Pattern classification | CNV/DNV/BS/CLV/FT | Hellström-Westas grade |
| Sleep-wake cycling | SWC present / absent | Strongest favourable marker post-HIE |
| Electrographic seizure count and burden | n/hr, %/hr | NCS / NCSE flag; quantify if cEEG available |
Document the time since insult with every reading: a BS pattern at 6 hours is much less prognostic than the same pattern at 48 hours after rewarming. Time-stamp every change.
6. What is normal? Age- and GA-banded reference
Normative aEEG bandwidth and continuity are dominated by post-menstrual age in the neonatal period and by state (wake / sleep / sedation) thereafter.
| Age / GA | Expected pattern | Lower margin (µV) | Sleep-wake cycling |
|---|---|---|---|
| 24–28 weeks GA | Highly discontinuous; long inter-burst intervals (IBI 20–60 s) | 2–3 | Absent or rudimentary |
| 28–32 weeks GA | Discontinuous; IBI 5–20 s | 3–5 | Emerging |
| 32–36 weeks GA | Mixed CNV/DNV; IBI 2–10 s | 4–7 | Established |
| Term (37–42 weeks) | CNV with clear SWC | > 5 | Present |
| 1–6 months | CNV; SWC mature | 7–12 | Present |
| 6 months – 5 years | CNV; voltages peak | 10–25 | Present |
| 5–18 years | CNV with age-typical alpha background | 10–20 | Present |
| Adult | CNV with alpha background | 10–20 | Present |
Sedation, propofol or pentobarbital coma, and natural deep sleep all push the trace toward discontinuous or burst-suppression. A neonate at 24 h of cooling with DNV on a midazolam-fentanyl infusion is not the same physiology as the same DNV off all sedation: sedation accounts for at least one-grade downgrade of pattern.
In the preterm population the Burdjalov maturational score combines continuity, sleep-wake cycling, amplitude, and bandwidth into a 0–13 ordinal score that tracks gestational-age maturation. A score below the GA-expected band suggests injury beyond what GA explains.
7. What is abnormal? Pattern library
The five Hellström-Westas patterns plus the seizure overlay form the canonical pattern set. The same envelope features carry different prognostic weight at different time-points and on or off sedation.
| Pattern | Defining features | Where you see it | Implication |
|---|---|---|---|
| Continuous normal voltage (CNV) | UM > 10, LM > 5 µV, no periodic interruption | Healthy term newborn; recovering HIE post-rewarming | Favourable; pair with SWC for prognostic punch |
| Discontinuous normal voltage (DNV) | UM > 10, LM 3–5 µV, periodic dips | Moderate HIE; preterm; sedated child | Borderline; trend over 24 h is the question |
| Burst-suppression (BS) | Clean baseline LM < 5 µV with periodic bursts UM > 25 µV | Severe HIE; deep barbiturate coma; status epilepticus suppression target | Severe injury or pharmacologic; time-point and sedation status critical |
| Continuous low voltage (CLV) | UM < 10 µV, LM < 5 µV, no periodicity | Severe HIE; profound diffuse injury | Poor; concordant with poor MRI |
| Flat trace (FT) | UM < 5 µV, LM < 5 µV | Electrocerebral inactivity | Devastating; brain-death evaluation context |
| Electrographic seizure overlay | Rhythmic broadening of UM with abrupt onset/offset, often periodic | NCS / NCSE in any pattern | Treat; quantify burden; cEEG confirmation |
| Sleep-wake cycling (SWC) | Slow undulation of LM every 20–60 min | Healthy term and older; recovering HIE | Strongest favourable marker |
| Loss of SWC | Flat LM trend | Acute injury; sedation; deep coma | Sensitive marker of clinical change |
Decision tree: reading the envelope in cooled HIE
flowchart TD
Trace[aEEG trace 24-48 h on TH] --> Sedation{Heavily sedated?}
Sedation -->|Yes| Caveat[Pattern downgraded; recheck 12 h post-wean]
Sedation -->|No| Pattern{Pattern}
Pattern -->|CNV with SWC| Favorable[Favourable; pair with MRI]
Pattern -->|DNV no SWC| Repeat[Repeat at 48-72 h]
Pattern -->|BS, CLV, FT| Concern[Concerning; raw EEG, MRI, family discussion]
Pattern -->|Rhythmic UM elevations| Seizure[NCS / NCSE; treat; cEEG]
The single most prognostic question on the aEEG round is whether sleep-wake cycling has returned by 48–72 hours of cooling. Children whose SWC returns by 72 h have markedly better neurodevelopmental outcomes at 18 months than those who do not, even when the static pattern at 24 h was the same. Trend beats snapshot.
8. Try it: interactive widget
9. Management: how aEEG drives the bedside decision
aEEG itself does not titrate a drug; it changes the decisions around drugs, imaging, and prognostic conversations.
9.1 During therapeutic hypothermia for HIE
The 72-hour cooling window is the canonical aEEG-driven workflow.
- Place leads within the first hour of cooling (3-hour upper limit). Earlier placement gives a full evolutionary curve.
- Document the pattern at 6, 24, 48, 72 hours and at any clinical change.
- Annotate sedation events: fentanyl boluses, midazolam infusion changes, paralysis (rare in cooling) all interpret the envelope.
- Open the raw EEG at every shift change and at every concerning aEEG change.
- At 72 h (rewarming complete) the pattern and SWC presence are the single most predictive non-imaging variables; combine with MRI at 4–7 days of life and the clinical exam.
- Family discussions based on aEEG patterns at 24 h are premature; wait until 48–72 h.
9.2 Seizure detection and burden quantification
- Suspicion threshold for NCSE: any rhythmic broadening of the upper margin, even brief, in a sedated or post-status patient.
- Confirm on raw EEG: aEEG sensitivity for individual seizures is 25–80% (operator and seizure-character dependent); specificity is high. Treat the rhythmic envelope, document the cEEG confirmation.
- Burden quantification: seizure-burden > 1 minute per hour or > 13 minutes total in any rolling hour is the threshold that drives most pediatric escalation protocols.
9.3 Sedation titration in the PICU (adjunct, not primary)
aEEG can support sedation depth assessment in paralysed ICU patients (where clinical sedation scores fail), but BIS is the more validated tool in that role. Use aEEG to detect break-through seizures during sedation and to monitor for excessive suppression that may delay neuroprognostication later.
Decision support, not a clinical protocol. Every threshold here is age-, sedation-, and centre-dependent. Pair with raw cEEG, clinical exam, and imaging; defer to your unit's neuro-prognostication pathway.
10. Clinical contexts: aEEG across the spectrum
10.1 Neonatal HIE and therapeutic hypothermia
The canonical use. aEEG is part of the routine cooling bundle in every NICU that runs the protocol. The 6-h pattern was diagnostic in the pre-TH era; under hypothermia, the diagnostic and prognostic time-points have shifted to 24–48 hours (during cooling) and 48–72 hours (post-rewarming). The combination of persistent BS/CLV/FT at 48 h and failure of SWC return by 72 h is the strongest non-MRI predictor of severe disability.
10.2 Neonatal seizures (term and preterm)
Neonatal seizures are predominantly electrographic-only after the first 24 h of life, particularly under sedation. aEEG with raw-EEG review is the screening modality. The ILAE 2017 framework recommends cEEG as the gold standard but explicitly endorses aEEG-plus-raw-trace as the bedside surrogate when cEEG is not immediately available.
10.3 Pediatric cardiac arrest and post-arrest
aEEG is part of the post-arrest neuromonitoring bundle in PICU practice. A return of CNV with SWC by 24–48 h post-ROSC is associated with favourable outcome; persistent CLV/FT or BS after sedation washout predicts poor outcome. The AHA 2021 pediatric resuscitation guidelines list cEEG (with aEEG screening) as the bedside electrophysiologic standard.
10.4 Severe TBI
In paralysed PICU TBI patients, clinical seizure detection is impossible. aEEG with raw cEEG cross-check detects NCS / NCSE that can otherwise present only as unexplained autoregulatory failure or refractory ICP. The 2019 BTF pediatric TBI guidelines and the ACNS critical-care EEG terminology guidelines support continuous EEG monitoring in severe TBI; aEEG is the bedside screening layer.
10.5 Aneurysmal SAH and DCI
Adult and emerging pediatric data show that qEEG-derived alpha-delta ratio is the most sensitive EEG marker of pre-clinical DCI in SAH; aEEG alone is too coarse to track DCI. Use aEEG to screen for NCS and to monitor sedation depth in coiled / clipped patients; rely on qEEG for the DCI question.
10.6 Pediatric ECMO
ECMO patients are paralysed, sedated, often hypothermic, and frequently anti-coagulated; clinical seizure detection is unreliable and stroke incidence is 5–10%. aEEG with raw-trace cross-check is the bedside screen for NCS / NCSE and the trend tool for acute neurologic deterioration during cannulation, cannula manipulation, or pump events.
10.7 Bacterial meningitis and acute encephalitis
In severe acute encephalitis (HSV, autoimmune, post-infectious), NCS / NCSE is common and clinically undetectable in the obtunded child. aEEG-screened EEG monitoring in the first 48–72 h is now standard. A burst-suppression pattern after status epilepticus treatment is the target for many sedation protocols; aEEG tracks the achievement and duration.
10.8 Refractory status epilepticus
aEEG is the bedside trend tool for continuous infusion treatment of refractory SE: midazolam, pentobarbital, ketamine. The target (electrographic seizure suppression, burst-suppression, or isoelectric, depending on protocol) is set and titrated on the envelope, then confirmed on cEEG. ESETT and ECLIPSE provide the second-line evidence; the third-line continuous-infusion endpoints are protocol-driven.
10.9 Brain-death determination
aEEG is not a substitute for the multi-channel EEG required by the World Brain Death Project framework, but it can prompt the ancillary testing pathway when an envelope becomes persistently isoelectric in a clinical brain-death candidate. The formal test requires full montage, defined sensitivity, ECG monitoring, and trained interpretation.
11. Multimodal integration: aEEG in the MMM/MNM stack
| Pair with… | What you gain | Worked scenario |
|---|---|---|
| cEEG | Confirm focal seizures, frequency content, alpha-coma | Any aEEG abnormality in a salvageable patient |
| NIRS / rSO2 | Combine cortical electrophysiology with cortical oxygenation; the HIE prognostic bundle | HIE monitoring bundle |
| NPI / pupillometry | Cortical (aEEG) plus brainstem (NPI) view; the post-arrest 72-h prognostic call | Post-arrest prognostic bundle |
| MRI 4–7 d | Imaging anchor for the envelope-based prognosis in HIE | HIE monitoring bundle |
| SSEP at 72 h | The most specific bilateral-absent-N20 sign in post-arrest comatose patients | Brain death and ancillary testing |
| ICP / PRx | Detect NCSE-driven CBF surges as autoregulatory failure on PRx | EEG / TCD non-convulsive seizure pair |
| BIS | Adjunct in paralysed PICU patients on continuous infusion sedation | Status epilepticus suppression targets |
12. Setup and technique
12.1 Equipment
- Bedside aEEG device: Olympic Brainz Monitor, Natus CFM, Nicolet BRM, Massimo CFM. All output the canonical envelope; visual styling differs.
- Electrodes: subdermal needles (preterm), hydrogel patches or cup electrodes with conductive gel (term and older). Three to five electrodes per side (active, reference, ground).
- Conductive medium and skin prep: alcohol clean, light skin abrasion (NuPrep), then patch. Avoid acetone in neonates.
- Cable shielding: critical in the high-EMI NICU/PICU environment; isolated input modules reduce 50/60-Hz mains contamination.
- Optional: simultaneous raw cEEG video integration for offline review.
12.2 Placement: the 5-minute bedside protocol
- Skin prep: clean the scalp at the two parietal sites (P3, P4 in the 10–20 system) with alcohol, gentle abrasion.
- Apply electrodes: hydrogel patches or cup electrodes with conductive gel, ensuring full contact and securing with light tape.
- Add reference and ground: Cz or AFz for reference; Fpz for ground.
- Test impedances: < 5 kΩ ideal; < 10 kΩ acceptable; > 20 kΩ unreadable, re-prep and re-apply.
- Confirm signal: visualise the raw trace for 30 seconds; you should see physiologic EEG activity, not a flat line (electrode off) or 60-Hz mains contamination.
- Start aEEG recording: 6 or 12-hour display window; annotate the start time and the patient's sedation state.
12.3 Reading routine
- Glance at every shift change: pattern, SWC, any rhythmic broadening.
- Open raw EEG at every concerning aEEG event: rhythmic upper-margin broadening (seizure?), abrupt voltage drop (sedation? rebleed? cannulation event?).
- Annotate the envelope with clinical events: sedation boluses, paralysis, cooling/rewarming, suction, intubation, cannulation.
- Re-impedance check every 8–12 hours: drying gel and patient sweat raise impedances and can produce artefactual "abnormalities".
12.4 Common artefact patterns and the fix
| Artefact | Envelope appearance | Cause | Fix |
|---|---|---|---|
| ECG | Periodic narrow blips on upper margin at heart rate | Cardiac activity bleeding into scalp electrode | Re-prep electrode; check impedance |
| High-frequency oscillator (HFOV) | Continuous high upper margin, narrow bandwidth | Mechanical vibration | Filter setting; flag on chart |
| Sweat / drift | Slow rise of lower margin | Patient warmer, sweat under electrode | Re-prep |
| Movement | Spike of upper margin, irregular | Patient agitation, suctioning, hand-on-head | Annotate; ignore short bursts |
| Mains 50/60 Hz | Continuous wide bandwidth | Poor shielding | Check ground, move away from infusion pumps |
| Lead-off | Flat trace, sudden | Electrode disconnected | Check impedance, re-apply |
12.5 Documentation in the chart
Every shift the bedside nurse documents the pattern, lower margin, SWC present/absent, seizure burden, and sedation state. The neurology team adds the cEEG interpretation. The combined entry is what informs the family discussion and the prognostic call.
13. Pitfalls
- Sedation downgrades the pattern. A child on midazolam and fentanyl looks one Hellström-Westas grade worse than off. Do not call BS at 24 h on full sedation; wait until sedation hold or interpret cautiously.
- Alpha-coma is invisible on the envelope. Periodically open the raw EEG.
- Focal seizures may not move the envelope. A two-lead aEEG covers a small fraction of cortex. Add lateralising leads or full cEEG for any focal clinical suspicion.
- Hypothermia and rewarming change the baseline. Voltages dip with cooling and recover with rewarming; do not over-call deterioration around the rewarming window.
- The 6-hour cutoff is pre-TH data. Use 24–48 h time-points for cooled neonates; do not over-weight a 6-h pattern.
- Asymmetry between two leads may be a bad electrode. Always check impedance before calling lateralised injury.
- Burst-suppression as a sedation target (for refractory SE) looks identical to BS as a marker of severe injury. Annotation of the clinical context is essential.
- High-frequency oscillator contaminates the trace with mechanical vibration. Flag the ventilation mode on the chart.
- A "normal-looking" envelope in a comatose patient with absent brainstem reflexes is alpha-coma or theta-coma until proven otherwise. Open the raw EEG.
14. Combine with…
- Continuous full-montage EEG: the gold standard; aEEG is the bedside trend layer.
- Quantitative EEG: adds spectral and asymmetry analysis on top of cEEG.
- NIRS: rSO2 plus aEEG for the HIE prognostic bundle.
- Pupillometry: NPI plus aEEG for the post-arrest 72-h call.
- SSEP: N20 plus aEEG plus NPI for the post-arrest prognostic triangulation.
- Foundations: pediatric physiology: why aEEG thresholds are age-dependent.
- Integration: HIE monitoring bundle: how aEEG fits the bundle.
- Integration: refractory status epilepticus: aEEG as the bedside trend in third-line therapy.
15. Evidence summary
| Topic | Source | Grade |
|---|---|---|
| Original CFM description | foundational | |
| Pre-TH aEEG prognosis (term HIE) | B | |
| Sleep-wake cycling return as prognostic marker | B | |
| Modern review of aEEG technique | review | |
| Neonatal seizure detection / ILAE framework | expert | |
| ACNS critical-care EEG terminology | expert | |
| Pediatric NCSE / cEEG | B | |
| Post-cardiac-arrest aEEG/cEEG | expert | |
| THAPCA-OH trial (outcomes context) | A | |
| Pediatric severe TBI (BTF 4th ed.) | expert | |
| Status epilepticus second-line evidence | A | |
| Status definition (ILAE) | expert | |
| Neonatal cEEG (sansevere) | review | |
| HIE NICHD cooling trial | A | |
| Pediatric MMM consensus | expert | |
| ECMO neuro outcomes | C | |
| Brain death determination | expert |
16. Recent literature (2022–2025)
- Sansevere 2023 (neonatal cEEG review): contemporary practice patterns for neonatal cEEG with aEEG as the screening layer in NICUs without 24/7 EEG technologist coverage.
- Naim 2023 (pediatric brain injury post-cardiac arrest): review of multimodal post-arrest neuromonitoring with aEEG/cEEG as the cortical electrophysiology arm.
- Tasker 2023 (pediatric neurocritical care review): positions aEEG as a tier-1 bedside modality in resource-stratified pediatric neurocritical care.
- Pediatric MMM consensus (Figaji 2025): formalises continuous EEG (with aEEG screening) in the recommended tier of pediatric multimodal monitoring.
- Topjian 2021 AHA pediatric: continuous EEG screening (aEEG-supported) in post-arrest pediatric care now part of the algorithm.
- Quantitative EEG / DCI: in SAH, qEEG (alpha-delta ratio) has eclipsed aEEG for DCI detection; aEEG remains the bedside screening layer for NCS / NCSE during the same admission.
17. Self-check
References
- Toet MC, Hellström-Westas L, Groenendaal F, Eken P, de Vries LS. Amplitude integrated EEG 3 and 6 hours after birth in full term neonates with hypoxic-ischaemic encephalopathy. Archives of Disease in Childhood. Fetal and Neonatal Edition 1999;81(1):F19–23. doi:10.1136/fn.81.1.f19 link
- Hellström-Westas L, Rosén I, de Vries LS, Greisen G. Amplitude-integrated EEG: classification and interpretation in preterm and term infants. NeoReviews 2006;7(2):e76–e87.
- Herman ST, Abend NS, Bleck TP, et al.. Consensus statement on continuous EEG in critically ill adults and children, part I and II (ACNS). Journal of Clinical Neurophysiology 2015;32(2-3):87-105 / 96-108.
- Abend NS, Arndt DH, Carpenter JL, et al.. Electrographic seizures in pediatric ICU patients: cohort study of risk factors and mortality. Neurology 2013;81(4):383–391. doi:10.1212/WNL.0b013e31829c5cfe link
- Topjian AA, Scholefield BR, Pinto NP, et al.. Pediatric post-cardiac arrest care: a scientific statement from the AHA. Circulation 2021;144(13):e194-e233.
- Naim MY, Friess SH, Sutton RM, et al.. Multimodal neuromonitoring in pediatric post-cardiac-arrest care. Pediatric Critical Care Medicine 2023.
- Hellström-Westas L, Rosén I, de Vries LS, Greisen G. Atlas of Amplitude-Integrated EEGs in the Newborn. Informa Healthcare 2008.
- Toet MC, Hellström-Westas L, Groenendaal F, Eken P, de Vries LS. Amplitude integrated EEG 3 and 6 hours after birth in full term neonates with hypoxic-ischaemic encephalopathy. Arch Dis Child Fetal Neonatal Ed 1999;81(1):F19–F23.
- al Naqeeb N, Edwards AD, Cowan FM, Azzopardi D. Assessment of neonatal encephalopathy by amplitude-integrated electroencephalography. Pediatrics 1999;103(6 Pt 1):1263–1271. doi:10.1542/peds.103.6.1263 link
- Shankaran S, Laptook AR, Ehrenkranz RA, et al.. Whole-body hypothermia for neonates with hypoxic-ischemic encephalopathy. NEJM 2005;353(15):1574-1584.
- Pressler RM, Cilio MR, Mizrahi EM, et al.. The ILAE classification of seizures and the epilepsies: modification for seizures in the neonate (ILAE Task Force). Epilepsia 2021;62(3):615-628.
- Sansevere AJ, Kapur K, Peters JM, et al.. Continuous EEG in the neonatal ICU: utility in seizure detection and neuroprognostication. Pediatric Neurology 2023.
- Moler FW, Silverstein FS, Holubkov R, et al.. Therapeutic hypothermia after out-of-hospital cardiac arrest in children (THAPCA-OH). NEJM 2015;372(20):1898-1908.
- Kochanek PM, Tasker RC, Carney N, et al.. Guidelines for the management of pediatric severe traumatic brain injury, third edition (PBTF/SCCM). Pediatric Critical Care Medicine 2019;20(3S):S1-S82.
- Tasker RC. Cerebrovascular reactivity in pediatric severe traumatic brain injury: a review. Pediatric Critical Care Medicine 2023.
- Hoh BL, Ko NU, Amin-Hanjani S, et al.. Guideline for the management of patients with aneurysmal subarachnoid hemorrhage: a guideline from the American Heart Association/American Stroke Association. Stroke 2023;54(7):e314-e370.
- Rass V, Helbok R. How to diagnose delayed cerebral ischaemia and symptomatic vasospasm and prevent cerebral infarction in patients with subarachnoid haemorrhage. Current Opinion in Critical Care 2021;27(2):103-114.
- Sandsmark DK, Foreman B, Claassen J. Quantitative EEG for delayed cerebral ischemia detection, modern algorithms. Critical Care 2024.
- Lorusso R, Taccone FS, Belliato M, et al.. Brain monitoring in adult and pediatric ECMO patients: the importance of early and late assessments. Minerva Anestesiologica 2017;83(10):1061-1074.
- Cho SM, Ziai W, Geocadin R, et al.. Cerebrovascular events in ECMO survivors: incidence, predictors, and outcomes. Critical Care Medicine 2024.
- Tunkel AR, Hartman BJ, Kaplan SL, et al.. Practice guidelines for the management of bacterial meningitis (IDSA). Clinical Infectious Diseases 2004;39(9):1267–1284.
- Tunkel AR, Glaser CA, Bloch KC, et al.. The management of encephalitis: clinical practice guidelines by the Infectious Diseases Society of America. Clinical Infectious Diseases 2008;47(3):303-327.
- Glauser T, Shinnar S, Gloss D, et al.. Evidence-based guideline: treatment of convulsive status epilepticus in children and adults. Epilepsy Currents 2016;16(1):48-61.
- Kapur J, Elm J, Chamberlain JM, et al.. Randomized trial of three anticonvulsant medications for status epilepticus (ESETT). NEJM 2019;381(22):2103-2113.
- Trinka E, Cock H, Hesdorffer D, et al.. A definition and classification of status epilepticus, Report of the ILAE Task Force. Epilepsia 2015;56(10):1515–1523.
- Greer DM, Shemie SD, Lewis A, et al.. Determination of brain death/death by neurologic criteria: the World Brain Death Project. JAMA 2020;324(11):1078-1097.
- Nakagawa TA, Ashwal S, Mathur M, et al.. Guidelines for the determination of brain death in infants and children: an update of the 1987 task force recommendations. Critical Care Medicine 2011;39(9):2139-2155.
- Figaji AA, Tasker RC, Bell MJ, Kochanek PM. Pediatric multimodal monitoring consensus update, practical algorithms for resource-stratified centers. Intensive Care Medicine, Paediatric and Neonatal 2025.
- Helbok R, Tasker RC, Kochanek PM, Bell MJ. Pediatric multimodal monitoring: where are we and where do we go?. Pediatric Critical Care Medicine 2024.
- Tasker RC, LaRovere KL, Riviello JJ, et al.. Pediatric multimodal neuromonitoring: international Delphi consensus. Pediatric Critical Care Medicine 2023.
- Maynard DE, Prior PF, Scott DF. Device for continuous monitoring of cerebral activity in resuscitated patients. BMJ 1969;4(5682):545–546.