NeuroRad TractsNeuroradiology education
Search→Residents→Fellowship→Explore Research↗Anatomy Lab↗Neuroradiology Oral Boards→
Brain

MRI in Epilepsy: Find the Substrate, Earn the Cure

Fellowship-level neuroradiology teaching by E. Brooke Schrickel, MD. Open this topic in the interactive reading room →

Watch · concise explainerVisual summary

Core clinical idea

A 'normal' routine brain MRI in epilepsy is usually a failure of technique and search pattern, not a normal brain. Findings are subtle and finding the substrate decides whether the patient gets a curative resection — so use a dedicated protocol, run a fixed search (HIPPO SAGE), and know the normal variants cold to neither miss the lesion nor invent one.

Bottom line

Epilepsy MR is a hunt: dedicated protocol, HIPPO SAGE every time, find the substrate, rule out dual pathology, and don't call a variant a lesion.

Core workstation questions

  • Is this on a dedicated seizure protocol, and do I need 3T / 3D volumetric sequences for a subtle lesion?
  • Is the head rotated (check the IACs) and faking hippocampal asymmetry?
  • Is the hippocampus atrophic with increased T2/FLAIR signal and loss of internal architecture?
  • Is there a second lesion — dual pathology — that changes the resection?
  • Is this cortical thickening / transmantle band a focal cortical dysplasia, and is the true epileptogenic zone wider than the visible abnormality?
  • Is this CSF-signal finding just a normal variant (hippocampal sulcus cyst, choroidal fissure cyst)?

What changes reporting / management

  • Read on a dedicated seizure protocol; escalate to 3T/3D for developmental or subtle lesions — dedicated protocol plus a trained reader roughly doubles yield over routine brain MRI.
  • Suggested phrasing for MTS: hippocampal volume loss with increased T2/FLAIR signal and loss of internal architecture, consistent with mesial temporal sclerosis.
  • State when the epileptogenic zone (e.g., in cortical dysplasia) is likely more extensive than the visible MR abnormality — it affects resection margins.
  • If truly negative, report MR-negative and recommend a dedicated/3T protocol rather than implying a normal brain.

Practical traps

  • Calling a small hippocampus without first excluding head rotation, which causes pseudo-asymmetry.
  • Stopping after one finding and missing dual pathology (MTS coexisting with a developmental anomaly); leaving either unresected worsens outcome.
  • Mistaking a normal variant (hippocampal sulcus residual cyst or choroidal fissure cyst, both CSF-isointense on all sequences in characteristic locations) for a lesion.
  • Confusing focal cortical dysplasia with a low-grade tumor (tumor more often expands/has mass effect) or with heterotopia (periventricular gray matter, not cortical thickening).

Teaching pearls

  • An MR-negative epilepsy scan often means wrong protocol or wrong search pattern, not a normal brain.
  • Check head rotation before you call a small hippocampus.
  • Find one lesion, keep looking — dual pathology changes the operation.
  • In cortical dysplasia, the epileptogenic zone is bigger than the blob.
  • CSF-isointense on every sequence in a known spot = variant, not pathology.

Teaching visuals

Mesial temporal sclerosis — imaging signsOblique-coronal section through the hippocampal body · original schematicCoronal section at the hippocampal levelimage left = NORMAL side · image right = MTS sidemidbrainNORMALplump volume · dark on FLAIRdigitations preservedMTSsmall & bright on FLAIRinternal architecture lostfornix ↓mammillary body ↓normal hippocampussclerotic (MTS) — atrophic & FLAIR-brightPrimary signs of MTSHippocampal volume loss (atrophy)smaller than the contralateral sideIncreased T2 / FLAIR signalbright where it should be darkLoss of internal architectureeffaced digitations / interdigitationsSecondary signsIpsilateral fornix atrophyIpsilateral mammillary body atrophyDilated ipsilateral temporal hornAtrophy of ipsilateral collateral white matterSearch patternUse a dedicated epilepsy protocol.Thin oblique-coronal T2 / FLAIR slices,angled PERPENDICULAR to the long axisof the hippocampus.ALWAYS compare side-to-side.Subtle MTS is a relative judgment,not an absolute one.
Mesial Temporal Sclerosis — Imaging Signs & Epilepsy-Protocol Search Pattern

Source lectures

  • MR Imaging of Epilepsy

Educational material for radiology residents and neuroradiology fellows. Nothing here drives individual patient care, and it contains no patient data.