Skip to content
How It Works
Explore
Ask a Question Resources Pricing
Who It's For
Patients
Decode results, prep for appointments, weigh options.
Caregivers & advocates
Carry the research load for someone you love.
Providers & clinicians
Resources for patients in your practice.
Log In Ask a Question
PubMed This is a summary of 18 peer-reviewed journal articles Updated
Pediatric Neurology

What Is the Spike-Wave Index (SWI) on a Child's EEG?

At a Glance

The Spike-Wave Index (SWI) estimates how much of a child's non-REM sleep EEG contains abnormal spikes. A high percentage may indicate SWAS or ESES, but no single cutoff predicts development; doctors interpret it with the EEG method, sleep captured, and the child's skills, behavior, and seizures.

In this answer

3 sections

01

How is the SWI Calculated?

02

What Do the Percentages Mean?

03

Does a High SWI Mean My Child Will Lose Skills?

Seeing a high percentage on your child’s EEG report can be frightening, but you are not expected to interpret it alone. The Spike-Wave Index (SWI) is a calculation of the percentage of your child’s non-rapid eye movement (NREM) sleep that contains abnormal electrical discharges, or “spikes” [1]. For children with Self-limited epilepsy with centrotemporal spikes (SeLECTS, formerly Rolandic epilepsy), doctors use the SWI to monitor for a condition historically called ESES (Electrical Status Epilepticus during Sleep) [2]. Today, this is often referred to as SWAS (Spike-Wave Activation in Sleep) or, if accompanied by developmental regression, DEE-SWAS (Developmental and/or Epileptic Encephalopathy with Spike-Wave Activation in Sleep) [3].

How is the SWI Calculated?

When a child falls asleep, they cycle through lighter and deeper stages of NREM sleep. During NREM sleep, it is common for children with Rolandic epilepsy to experience an increase in electrical spikes [4].

To calculate the SWI, neurologists often break the NREM portion of the EEG into one-second blocks (epochs). They count how many blocks contain at least one spike, and divide that by the total number of NREM blocks analyzed [1].

However, methods vary. Some laboratories measure the exact duration of the spikes rather than counting blocks, or they only sample the first few minutes of sleep [5][6]. Furthermore, an SWI calculated from a 20-minute daytime nap may differ significantly from one calculated during a full overnight video-EEG, as naps may not capture enough deep slow-wave sleep [7]. Because of this, the SWI is highly dependent on the scoring method and the sleep period analyzed [8].

What Do the Percentages Mean?

There is no single universally accepted SWI cutoff that applies to every child [9]. Instead, neurologists use certain percentages as historical or study-specific reference points to gauge the “burden” of spikes during sleep:

  • The 50% Marker (SWAS): Many modern studies define Spike-Wave Activation in Sleep (SWAS) or “atypical ESES” as an SWI of 50% or higher [10][11]. This is generally considered a warning level where the risk for cognitive or behavioral challenges increases [4].
  • The 85% Marker (Typical ESES): Historically, an SWI of 85% or greater was the strict criteria for diagnosing “typical ESES” [2]. Other studies have suggested that an SWI over 60% is highly suggestive of this heavy spiking pattern [2].

It is important to know: A high SWI describes an EEG pattern; it does not mean your child is having a continuous physical convulsion, nor is it automatically a medical emergency.

Does a High SWI Mean My Child Will Lose Skills?

A high SWI is a risk factor, but it does not dictate a child’s future.

When a high SWI occurs alongside cognitive regression (the loss of previously learned skills), or new struggles with language, attention, impulsivity, or school performance, a neurologist may diagnose DEE-SWAS [4][12][13].

However, an elevated SWI does not guarantee regression. Some children have a high SWI without experiencing major drops in overall intelligence [14][15]. At the same time, a child might maintain their overall IQ but develop specific difficulties with language, executive function, or behavior [16].

Because the number alone is not a perfect predictor, your pediatric epileptologist will look at the whole picture rather than treating the SWI in isolation. They will consider:

  • Is the SWI persistently high across multiple EEGs? [17]
  • How are the spikes distributed on the EEG? While bilateral spikes (on both sides of the brain) are a known risk factor, significant unilateral (one-sided) activity is also carefully evaluated [17][18].
  • How is the child functioning at home and school? [18]

If your child has a high SWI, the most important next step is communicating with your care team about any changes in language, learning, or behavior. Do not change medications based on the EEG number alone.

Bottom Line: No single SWI number predicts a child’s future or definitively diagnoses DEE-SWAS. Interpreting the SWI requires understanding the EEG method, ensuring adequate NREM sleep was captured, and closely monitoring the child’s real-world developmental progress [3].

Common questions in this guide

What does the SWI percentage on my child's EEG measure?
The SWI estimates the percentage of non-REM sleep on the EEG that contains abnormal electrical discharges called spikes. It describes the amount of spike activity recorded, not the number of physical seizures your child had.
Why can two EEG reports show different SWI percentages?
SWI can change depending on whether the laboratory counts one-second segments with a spike, measures spike duration, or samples only part of sleep. A short nap may also capture less deep sleep than an overnight video-EEG, so the results are not always directly comparable.
Does a 50% or 85% SWI prove that my child has ESES?
No. An SWI of 50% or higher is often used as a reference for SWAS, while 85% was a historical criterion for typical ESES, but there is no universal cutoff. The interpretation depends on the scoring method, amount of sleep recorded, EEG pattern, and your child's developmental picture.
Does a high SWI mean my child is having a seizure continuously?
No. A high SWI reflects frequent abnormal electrical discharges during non-REM sleep and does not mean that your child is having a continuous physical convulsion or that the result is automatically an emergency.
Can a high SWI cause my child to lose skills?
A high SWI can be associated with developmental, language, attention, learning, or behavior difficulties, but it does not guarantee regression. The neurologist will consider changes over time, repeated EEGs, spike distribution, seizures, and your child's functioning at home and school.
How are ESES, SWAS, and DEE-SWAS related to Rolandic epilepsy?
Rolandic epilepsy is the former name for SeLECTS, a childhood epilepsy in which spikes may increase during non-REM sleep. ESES is the older term for a heavy sleep-activated spike pattern, SWAS is the newer term often used for that pattern, and DEE-SWAS refers to the pattern when developmental regression or broader developmental and epileptic encephalopathy is present.
What should I do if my child's EEG shows a high SWI?
Ask the care team how the percentage was calculated, how much non-REM sleep was recorded, and whether the EEG was a nap or an overnight study. Tell the team about changes in language, learning, attention, behavior, or seizures, and do not change medication based on the EEG number alone.

Questions to Ask Your Doctor

Curated prompts to bring to your next appointment.

  1. 1.What exact scoring method was used to calculate my child's SWI, and how much NREM sleep was analyzed?
  2. 2.Was this SWI captured during a short nap or a full overnight video-EEG, and did it capture enough slow-wave sleep?
  3. 3.Does this result represent Spike-Wave Activation in Sleep (SWAS), or is it simply an expected increase in spikes for Rolandic epilepsy?
  4. 4.Given the EEG pattern and distribution of the spikes (unilateral vs. bilateral), how frequently should we repeat the video-EEG?
  5. 5.Do you recommend a formal neuropsychological evaluation to establish a baseline for my child's learning, language, and executive functioning?

Questions For You

Tap a prompt to share your answer — we'll use it plus this page's context to start a tailored conversation.

Related questions

Are Follow-Up EEGs Necessary for Rolandic Epilepsy?At What Age Does Rolandic Epilepsy Usually Go Away?Can Seizure Medication Make Rolandic Epilepsy or ESES Worse?Can Rolandic Epilepsy Cause Full-Body Seizures in Children?Does ADHD Go Away After Rolandic Epilepsy Resolves?Does Watchful Waiting in Rolandic Epilepsy Damage the Brain?Can EEG Spikes Cause Learning Problems in SeLECTS?Is Ospolot (Sulthiame) Used for Rolandic Epilepsy in the US?Is Rolandic Epilepsy Genetic? Heredity and Sibling RiskWhy Does SeLECTS Cause Drooling and Facial Twitching?Why Do Rolandic Epilepsy Seizures Happen During Sleep?Why Is a Brain MRI Usually Normal in Rolandic Epilepsy?

References

References (18)
  1. 1

    Self-limited epilepsy with centrotemporal spikes (SeLECTS) with spike-and-wave activation in sleep (SWAS): clinical characteristics and electroencephalogram outcomes at the 2-year follow-up.

    Wu SN, Dai JM, Ye XG, et al.

    Seizure 2026; (139()):84-90 doi:10.1016/j.seizure.2026.05.005.

    PMID: 42107919
  2. 2

    [A comparative study of three systems for quantifying the spike and wave index in patients with continuous spikes and waves during slow sleep].

    Azcona G, Gurtubay IG, Mosquera A, et al.

    Revista de neurologia 2017; (65(10)):439-446.

    PMID: 29130467
  3. 3

    Rebuilding the Tower of Babel: The Current Landscape and Emerging Opportunities in DEE-SWAS.

    Stowe RC, Kelley SA, Bhatia S, et al.

    Epilepsy currents 2025; (25(6)):369-377 doi:10.1177/15357597251344191.

    PMID: 40502808
  4. 4

    Evolution into spike-and-wave activation in sleep in patients with self-limited focal epilepsies.

    İriş M, Yaşgüçlükal MA, Yalçınkaya C, Demirbilek V

    Seizure 2026; (135()):34-38 doi:10.1016/j.seizure.2026.01.006.

    PMID: 41570778
  5. 5

    Relationship between brain activity, cognitive function, and sleep spiking activation in new-onset self-limited epilepsy with centrotemporal spikes.

    Li Y, Li Y, Sun J, et al.

    Frontiers in neurology 2022; (13()):956838 doi:10.3389/fneur.2022.956838.

    PMID: 36438972
  6. 6

    The Spike-Wave Index of the First 100 Seconds of Sleep Can Be a Reliable Scoring Method for Electrographic Status Epilepticus in Sleep.

    Kulkarni N, Albert DVF, Klamer B, et al.

    Journal of clinical neurophysiology : official publication of the American Electroencephalographic Society 2023; (40(6)):547-552 doi:10.1097/WNP.0000000000000918.

    PMID: 35025840
  7. 7

    [Epileptiform activity index for the evaluation of treatment efficacy in patients with epileptic encephalopathy].

    Kozhokaru AB

    Zhurnal nevrologii i psikhiatrii imeni S.S. Korsakova 2019; (119(10)):121-126 doi:10.17116/jnevro2019119101121.

    PMID: 31793553
  8. 8

    Semiautomated Spike Detection Software Persyst 13 Is Noninferior to Human Readers When Calculating the Spike-Wave Index in Electrical Status Epilepticus in Sleep.

    Joshi CN, Chapman KE, Bear JJ, et al.

    Journal of clinical neurophysiology : official publication of the American Electroencephalographic Society 2018; (35(5)):370-374 doi:10.1097/WNP.0000000000000493.

    PMID: 29933261
  9. 9

    EEG features in Encephalopathy related to Status Epilepticus during slow Sleep.

    Gardella E, Cantalupo G, Larsson PG, et al.

    Epileptic disorders : international epilepsy journal with videotape 2019; (21(S1)):22-30 doi:10.1684/epd.2019.1054.

    PMID: 31162114
  10. 10

    Spike-Wave Index Assessment and Electro-Clinical Correlation in Patients with Encephalopathy Associated with Epileptic State During Slow Sleep (ESES / CSWS); Single-Center Experience.

    Öztoprak Ü, Yayici Köken Ö, Aksoy E, Yüksel D

    Epilepsy research 2021; (170()):106549 doi:10.1016/j.eplepsyres.2021.106549.

    PMID: 33450525
  11. 11

    Perinatal thalamic injury: MRI predictors of electrical status epilepticus in sleep and long-term neurodevelopment.

    van den Munckhof B, Zwart AF, Weeke LC, et al.

    NeuroImage. Clinical 2020; (26()):102227 doi:10.1016/j.nicl.2020.102227.

    PMID: 32182576
  12. 12

    Cognition in patients with benign epilepsy with centrotemporal spikes: A study with long-term VEEG and RS-fMRI.

    Yan X, Yu Q, Gao Y, et al.

    Epilepsy & behavior : E&B 2017; (76()):120-125 doi:10.1016/j.yebeh.2017.06.033.

    PMID: 28882722
  13. 13

    The influence of EEG-detected nocturnal centrotemporal discharges on the expression of core symptoms of ADHD in children with benign childhood epilepsy with centrotemporal spikes (BCECTS): A prospective study in a tertiary referral center.

    Danhofer P, Pejčochová J, Dušek L, et al.

    Epilepsy & behavior : E&B 2018; (79()):75-81 doi:10.1016/j.yebeh.2017.11.007.

    PMID: 29253678
  14. 14

    Influence of epileptic activity during sleep on cognitive performance in benign childhood epilepsy with centrotemporal spikes.

    Nissenkorn A, Pappo A, Feldmann Y, et al.

    European journal of paediatric neurology : EJPN : official journal of the European Paediatric Neurology Society 2017; (21(6)):858-863 doi:10.1016/j.ejpn.2017.07.001.

    PMID: 28734769
  15. 15

    The Prevalence and Risk Factors of Electrical Status Epilepticus During Slow-Wave Sleep in Self-Limited Epilepsy With Centrotemporal Spikes.

    Lu G, Cheng Y, Wang Y, et al.

    Clinical EEG and neuroscience 2024; (55(2)):265-271 doi:10.1177/15500594231182758.

    PMID: 37331959
  16. 16

    Language-related brain areas in childhood epilepsy with centrotemporal spikes studied with MEG.

    Xu F, Xu Y, Wang Y, et al.

    Clinical neurophysiology : official journal of the International Federation of Clinical Neurophysiology 2023; (152()):11-21 doi:10.1016/j.clinph.2023.05.005.

    PMID: 37257319
  17. 17

    Predicting progression from SeLECTS with SWAS to EE-SWAS: risk factor identification and model development.

    Hu Q, Luo Y, Deng Y, et al.

    Frontiers in human neuroscience 2025; (19()):1641421 doi:10.3389/fnhum.2025.1641421.

    PMID: 41625547
  18. 18

    Language dysfunction is associated with age of onset of benign epilepsy with centrotemporal spikes in children.

    Ma Y, Chen G, Wang Y, Xu K

    European neurology 2015; (73(3-4)):179-83 doi:10.1159/000371417.

    PMID: 25614147

This page is for informational purposes only and does not constitute medical advice about your child's SWI. A pediatric neurologist or epileptologist should interpret the EEG with your child's development, seizures, and clinical history.

Get notified when new evidence is published on Self-limited epilepsy with centrotemporal spikes.

We monitor PubMed for new peer-reviewed studies on this topic and email a short summary when something meaningful changes.