High-yield points
- SEEG is now the main pediatric invasive technique: 94% of invasive cases in an international survey of 61 programs, vs. 3% grids alone. A third of invasive cases are MRI-negative .
- SEEG vs. grids in children (non-randomized pooled studies):
- Grid complications: CSF leak about 12%, hemorrhage about 11% and infection about 11%.
- SEEG hemorrhage: about 3%.
- Seizure freedom: 67% after SEEG vs. 52% after grids .
- SEEG complications overall: about 1% each for hemorrhage and infection; 0.3% mortality .
- Robotic vs. frame-based SEEG: no significant difference in accuracy or safety, but faster (about 6.6 minutes saved per electrode) .
- Age is not a barrier: SEEG is accurate (about 2–3 mm) and safe even under 3 years, with a skull about 2 mm thick .
- Grids still have a role for contiguous convexity mapping. Generous initial coverage was associated with fewer reoperations .
Intracranial EEG (iEEG) records seizures from electrodes placed on or in the brain. It is used when noninvasive evaluation cannot define:
- the epileptogenic zone;
- its relationship to eloquent cortex.
There are two techniques:
- subdural grids and strips (SDG): contiguous surface coverage through a craniotomy;
- stereo-EEG (SEEG): depth electrodes through twist-drill holes, sampling deep and bilateral structures.
In children, SEEG has largely replaced grids .
Indications
- Current practice (ILAE survey, 61 programs in 29 countries, 2,427 children operated on in 2023) :
- Invasive evaluation: 21.1% of children, more often in North America (33.7%) than Europe (18.0%).
- MRI: 32.3% of invasive cases were MRI-negative.
- Indications:
- seizure-onset localization, 88.1%;
- motor or sensory mapping, 17.2%;
- language mapping, 14.2%.
- Technique:
- SEEG in 93.6% of invasive cases;
- subdural electrodes alone in only 3.2%;
- combined approaches in 3.2%.
- Treatment through the electrodes: SEEG-guided radiofrequency thermocoagulation in 40.9% of SEEG cases.
- No definitive surgery: 16.2% of invasively evaluated children did not go on to resection, disconnection or ablation.
- Typical scenarios (see caveats):
- MRI-negative focal epilepsy;
- discordant noninvasive data;
- multiple lesions (for example tuberous sclerosis);
- suspected bilateral, deep or insular/opercular onset;
- seizure onset near eloquent cortex that needs mapping.
- Choosing a technique:
- Favors SEEG:
- deep structures (hippocampus, insula, cingulate, sulcal depths);
- bilateral or multilobar hypotheses;
- prior surgery;
- avoiding a craniotomy.
- Favors grids: a convexity hypothesis needing contiguous functional mapping.
- Combining them: grids can follow SEEG to map motor cortex before resection (see caveats).
- Who will benefit (the 5-SENSE score, 79 children):
- What it is: five noninvasive variables (focal MRI lesion, regional ictal onset, no bilateral independent spikes, localizing semiology, neuropsychology) that predict whether SEEG will find a focal epileptogenic zone.
- Performance: sensitivity 70.7% and specificity 71.4%.
- Caution: the score was developed in adults; in children its accuracy was moderate, and a low score doesn't rule out a focal onset .
Technique
- Subdural grids and strips:
- Procedure: a craniotomy; grids placed on the cortex and strips slid under the dura; leads tunneled out.
- Second operation: removal and resection.
- Coverage: in the St. Louis series of 112 children, aggressive initial coverage was not associated with more transfusion or complications, and was associated with fewer reoperations for extra electrodes .
- SEEG planning:
- Planning: depth electrodes planned on MRI with vascular imaging (CTA, MRA/MRV or catheter angiography), along avascular trajectories, entering through 2–2.5 mm twist-drill holes with anchor bolts (see caveats).
- Pediatric series (25 children, 342 electrodes):
- Monitoring: a mean of 8 days.
- Localization: SEEG localized the hypothesized onset in 92%.
- Discharge: most went home the day after explant .
- Frame-based vs. robotic placement (meta-analysis of 8 retrospective cohorts, 758 patients with drug-resistant epilepsy; not pediatric-specific) :
- Accuracy: no significant difference in entry, target, radial or depth error.
- Speed: robotic placement was about 33 minutes faster overall and about 6.6 minutes faster per electrode.
- Safety: no significant difference in complications.
- Robotic SEEG across ages (88 children, 675 electrodes) :
- Error: target point 1.93 mm and entry point 1.30 mm, with no difference between children aged 3 and younger, 3–6, and over 6.
- Complications: 6.06%, with no severe or long-term deficit.
- Children under 3 (21 children, 443 electrodes) :
- Head fixation: a Mayfield headframe with an infinity support system in 65%.
- Skull: thickness at electrode sites was a mean of 2.0 mm (range 1.5–3.2).
- Error: entry point 1.9 mm and target point 2.8 mm.
- Complications: only one, a bolt-site CSF leak.
- After SEEG:
- Removal and treatment together: electrodes can be removed in the same session as LITT or open resection.
- Outcomes (74 children, non-randomized): Engel I–II in 74% after either. LITT was used for deep targets such as the insula, and tended toward shorter stays (3 vs. 5 days, not significant) .
- Radiofrequency thermocoagulation through the electrodes is a further option .
Risks and complications
- SEEG vs. grids in children (meta-analysis, 697 grid and 277 SEEG patients) :
- With grids:
- CSF leak 11.9%, the commonest grid complication (only one leak with SEEG);
- hemorrhage 10.7% (vs. 2.9% with SEEG);
- infection 10.8%.
- Seizure freedom at last follow-up: 66.5% after SEEG vs. 52.1% after grids.
- The St. Louis grid series (122 implantations) :
- reoperation for more electrodes, 5.7%;
- wound infection, 2.4%;
- CSF leak, 1.6%;
- subdural hematoma, symptomatic pneumocephalus, bone-flap osteomyelitis and strip fracture, 0.8% each;
- transient deficits, 3.3%;
- no permanent deficit or death.
- SEEG complications overall (meta-analysis, not pediatric-specific) :
- hemorrhage, 1.0%;
- infection, 0.8%;
- all surgical complications, 1.3%;
- mortality, 0.3% (5 deaths).
- Resource use (9 SEEG and 9 grid patients with similar characteristics, one center) :
- Shorter with SEEG: operating time 288 vs. 356 minutes and ICU stay 1.0 vs. 2.1 days.
- Longer with SEEG: SEEG patients had two hospitalizations (implant and explant, then resection), and a longer time to treatment.
- Grid-specific risks to plan for (see caveats):
- cerebral edema and mass effect under large grids;
- CSF leak at lead exits;
- infection rising with monitoring duration;
- bone-flap problems.
- SEEG-specific risks:
- intracerebral or subdural hemorrhage from vessels along the trajectory (see caveats);
- electrode deflection into the subdural space ;
- superficial scalp infection ;
- bolt-site CSF leak ;
- misplacement and breakage (see caveats).
Related topics
- Epilepsy Surgery Overview: where invasive monitoring fits.
- Brain Mapping: cortical stimulation mapping.
- Extratemporal Epilepsy Surgery: the main use of SEEG.
- Responsive Neurostimulation: SEEG-guided lead placement.
- Temporal Lobe Epilepsy Surgery: when invasive monitoring is needed.
- Matching the Anesthetic to the Operation: Effects on Neuromonitoring: anesthetic effects on monitoring.
- Responsive Neurostimulation (RNS) Placement: step-by-step procedure.
References
- Ramantani G, Feucht M, Cserpan D, et al. Pediatric epilepsy surgery: global survey of invasive explorations. Epilepsia 2026;67(9):4470–4482. doi:10.1002/epi.70325. PMID 42274288. verified
- Sacino MF, Huang SS, Schreiber J, et al. Is the use of stereotactic electroencephalography safe and effective in children? A meta-analysis of the use of stereotactic electroencephalography in comparison to subdural grids for invasive epilepsy monitoring in pediatric subjects. Neurosurgery 2019;84(6):1190–1200. doi:10.1093/neuros/nyy466. PMID 30351382. verified
- Johnston JM Jr, Mangano FT, Ojemann JG, et al. Complications of invasive subdural electrode monitoring at St. Louis Children's Hospital, 1994–2005. J Neurosurg 2006;105(5 Suppl):343–347. doi:10.3171/ped.2006.105.5.343. PMID 17328255. verified
- Mullin JP, Shriver M, Alomar S, et al. Is SEEG safe? A systematic review and meta-analysis of stereo-electroencephalography-related complications. Epilepsia 2016;57(3):386–401. doi:10.1111/epi.13298. PMID 26899389. verified
- Goldstein HE, Youngerman BE, Shao B, et al. Safety and efficacy of stereoelectroencephalography in pediatric focal epilepsy: a single-center experience. J Neurosurg Pediatr 2018;22(4):444–452. doi:10.3171/2018.5.PEDS1856. PMID 30028270. verified
- Abbas A, Sabet H, El Refaei K, et al. Robot-assisted versus frame-based stereoelectroencephalography (sEEG) electrode implantation in drug-resistant epilepsy: a meta-analysis of accuracy, efficiency, and safety. Acta Neurochir (Wien) 2026;168(1). doi:10.1007/s00701-026-06787-6. PMID 41721878. verified
- Lu R, Wang M, Zhang Y, et al. Safety, accuracy, and efficacy of robot-assisted stereo electroencephalography in children of different ages. Neurosurgery 2024;95(1):137–145. doi:10.1227/neu.0000000000002853. PMID 38299855. verified
- Uchitel J, Fariyike OA, Rockwood SJ, et al. Safety, feasibility, and technique of stereo-electroencephalography in children younger than 3 years. J Neurosurg Pediatr 2026;37(5):440–449. doi:10.3171/2025.9.PEDS25194. PMID 41650398. verified
- Salehi A, Yang PH, Smyth MD. Single-center cost comparison analysis of stereoelectroencephalography with subdural grid and strip implantation. J Neurosurg Pediatr 2022;29(5):568–574. doi:10.3171/2022.1.PEDS21523. PMID 35180694. verified
- Singh AS, Seri S, Lo WB, et al. Can the 5-SENSE score guide stereo-electroencephalography in children? First validation in a purely pediatric cohort. J Neurosurg Pediatr 2026;38(2):128–133. doi:10.3171/2025.12.PEDS25446. PMID 42139735. verified
- Slingerland AL, Chua MMJ, Bolton J, et al. Stereoelectroencephalography followed by combined electrode removal and MRI-guided laser interstitial thermal therapy or open resection: a single-center series in pediatric patients with medically refractory epilepsy. J Neurosurg Pediatr 2023;31(3):206–211. doi:10.3171/2022.11.PEDS22262. PMID 36681974. verified
Verification (2026-11-17):
- Each reference's authors, title, journal, year, volume, pages, DOI and PMID were checked on PubMed. Numbers were checked against the PubMed abstracts.
- Reference 6 is a 2026 article without page or article numbers in PubMed.
Caveats:
- The following are standard practice, not drawn from the cited abstracts:
- the list of typical indications;
- the technique-selection factors;
- SEEG planning details (vascular imaging, avascular trajectories, 2–2.5 mm twist-drill holes);
- the grid-specific and SEEG-specific risk lists beyond the cited series (including trajectory hemorrhage, misplacement and breakage).
- Reference 5 reports using grids after SEEG in 2 patients, to map motor function near the seizure onset; the broader "combining" guidance is general practice.
- Reference 4 is not pediatric-specific (its abstract does not give ages). Reference 2 compares separate study populations, not randomized groups, so the outcome differences may reflect selection.
- The figure is an original schematic.