Epilepsy Mimics & DDx
Epilepsy Mimics & Differential Diagnosis
What Do You Need to Know?
- 20–30% of "drug-resistant epilepsy" referrals do not have epilepsy — misdiagnosis is one of the most common errors in neurology
- PNES is the most important mimic: mean diagnostic delay 7–10 years; ictal eye closure (96% sensitivity, 98% specificity) is the single most reliable distinguishing sign
- ECG is MANDATORY for every first seizure presentation — cardiac channelopathies (Long QT, CPVT, Brugada) cause convulsive syncope and carry risk of sudden death
- Convulsive syncope occurs in up to 12% of syncope episodes and does NOT indicate epilepsy
- Dual diagnosis: 10–50% of PNES patients also have true epilepsy — never assume all events are the same type
- Gold standard: video-EEG monitoring with capture of a habitual event and normal ictal EEG definitively excludes epilepsy for that event type
PNES (Psychogenic Nonepileptic Seizures) — The Most Important Mimic
Epidemiology & Impact
- 20–30% of patients referred to epilepsy monitoring units with "drug-resistant epilepsy" have PNES
- Mean diagnostic delay: 7–10 years of unnecessary ASM exposure
- Also termed functional seizures or dissociative seizures (FND spectrum)
- Consequences: iatrogenic ASM toxicity, ICU admissions for "status," psychosocial burden
Semiologic Features
| Feature | Sensitivity | Specificity | Notes |
|---|---|---|---|
| Ictal eye closure | 96% | 98% | MOST RELIABLE SIGN; epileptic seizures present with eyes open |
| Waxing/waning course | 94% | 100% | Fluctuating intensity with pauses; seizures evolve but do not wax/wane |
| Asynchronous limb movements | 84% | 100% | Out-of-phase alternating; caveat: rarely in frontal lobe seizures |
| Duration >2 minutes | 65% | 93% | GTCS typically 1–2 min; PNES often 5–30 min |
| Side-to-side head movement | 63% | 100% | Lateral head shaking during event |
| Pelvic thrusting | 24% | 97% | Low sensitivity but highly specific; rarely in frontal lobe epilepsy |
| Ictal crying/weeping | Low | Very high | Virtually pathognomonic when present |
Dual Diagnosis
- 10–50% of PNES patients also have coexisting epilepsy
- Each event type must be independently identified on video-EEG
- Never assume all events in one patient are the same type
Diagnosis
- Gold standard: video-EEG capturing habitual event with normal ictal EEG
- Staged approach (ILAE): documented → clinically established → confirmed (EEG-video confirmed)
- Prolactin: elevated after GTCS (not after PNES or absence)
- Must draw within 10–20 minutes of the event
- Limited sensitivity; NOT reliable as sole diagnostic tool
- Not elevated after frontal lobe seizures
Treatment
- Psychotherapy is the treatment — NOT ASMs
- CBT has the best evidence (CODES trial: CBT-informed therapy reduced seizure frequency)
- Empathetic communication of the diagnosis is itself therapeutic
- Gradual ASM taper if no concurrent epilepsy — tapering alone reduces event frequency
- Avoid iatrogenic harm: no IV benzodiazepines for prolonged PNES events
💎 Board Pearl
- Ictal eye closure = PNES until proven otherwise. Epileptic seizures almost always present with eyes OPEN. This single sign has 96% sensitivity and 98% specificity — the highest-yield feature on boards
Syncope vs. Seizure
Comparison Table
| Feature | Syncope | Epileptic Seizure (GTCS) |
|---|---|---|
| Trigger | Positional/situational (standing, pain, heat) | Usually spontaneous; sleep deprivation, alcohol |
| Prodrome | Lightheadedness, tunnel vision, warmth, nausea | Focal aura (déjà vu, epigastric rising) or none |
| Position | Usually upright | Any position (including supine or from sleep) |
| Skin color | Pale, diaphoretic | Cyanotic (respiratory compromise) |
| Motor features | Brief myoclonus (<15 sec), irregular, multifocal | Sustained tonic → rhythmic clonic (1–2 min) |
| Duration of LOC | Seconds (<30 sec) | Minutes (1–3 min + postictal) |
| Recovery | Rapid (<1 min), oriented quickly | Prolonged postictal confusion (5–30+ min) |
| Tongue bite | Tip of tongue (if any) | Lateral tongue laceration (specific) |
| Incontinence | Uncommon | Common in GTCS |
Convulsive Syncope
- Occurs in up to 12% of syncope episodes
- Mechanism: transient cerebral hypoperfusion → cortical disinhibition → subcortical motor release
- Brief (<15 sec) tonic stiffening or irregular myoclonic jerks
- Does NOT indicate epilepsy; does NOT require ASMs
- No prolonged postictal confusion
💎 Board Pearl
- Lateral tongue laceration = seizure. Tip tongue bite = syncope (if any). This is the highest-yield tongue-bite distinction for boards
Cardiac Mimics — Critical for Boards
ECG Is Mandatory
- ECG must be obtained for EVERY first seizure presentation
- Cardiac channelopathies cause convulsive syncope indistinguishable from GTCS by observation alone
- Family history of unexplained sudden death (<40 years) = critical red flag
Key Cardiac Channelopathies
| Condition | Gene | Presentation Clue | Diagnosis |
|---|---|---|---|
| Long QT (LQT1) | KCNQ1 | "Seizure" during swimming = LQT1 until proven otherwise | Prolonged QTc on resting ECG |
| Long QT (LQT2) | KCNH2 | Syncope with auditory startle | Prolonged QTc on resting ECG |
| Long QT (LQT3) | SCN5A | Events during sleep/rest | Prolonged QTc on resting ECG |
| CPVT | RYR2 (AD), CASQ2 (AR) | 30% misdiagnosed as epilepsy; resting ECG is NORMAL | Exercise stress test: bidirectional VT (pathognomonic) |
| Brugada | SCN5A | Nocturnal "seizures"; young/SE Asian males | Coved ST elevation V1–V3; Na channel blocker provocation |
SUDEP & Cardiac Overlap
- 10–15% of SUDEP autopsy cases have cardiac channel gene variants
- Overlap between epilepsy and cardiac channelopathy genes (e.g., SCN5A, SCN1A)
- Some "SUDEP" cases may be primary cardiac arrhythmia deaths
💎 Board Pearl
- CPVT has the highest misdiagnosis rate of all cardiac channelopathies because the resting ECG is completely normal. Exercise stress test is required. Exercise-triggered "seizures" + normal resting ECG = think CPVT before epilepsy
Paroxysmal Dyskinesias
| Type | Gene | Trigger | Duration | Treatment |
|---|---|---|---|---|
| PKD | PRRT2 | Sudden movement (standing, starting to walk) | <1 min (seconds) | Dramatic response to low-dose CBZ |
| PNKD | MR-1/PNKD | Alcohol, caffeine, stress (NOT movement) | 10 min–hours | Clonazepam; CBZ NOT effective |
| PED | SLC2A1 (GLUT1) | Sustained exercise (10–15 min continuous) | 5–30 min | Ketogenic diet |
PRRT2: One Gene, Three Phenotypes
- BFIS: seizure clusters in infancy, resolve by age 2
- PKD: movement-triggered dyskinesias in childhood/adolescence
- ICCA syndrome: BFIS in infancy followed by PKD in adolescence (same individual)
- All three respond to low-dose carbamazepine; all have excellent prognosis
Key Principle
- Paroxysmal dyskinesias are NOT epileptic — consciousness preserved; EEG normal during events
- PKD is one of the most gratifying diagnoses: dramatic response to minimal treatment
💎 Board Pearl
- PRRT2 = one gene, three phenotypes (BFIS + PKD + ICCA). All respond to low-dose CBZ. Exercise-induced dystonia + low CSF glucose = think GLUT1 (SLC2A1); treat with ketogenic diet
Benign Paroxysmal Events of Infancy
| Condition | Key Features | Distinguishing Clue |
|---|---|---|
| Shuddering attacks | Brief tremor-like shivering; head tremor, shoulder elevation | Normal development; resolves by age 2; family history of essential tremor |
| Jitteriness (neonatal) | Rhythmic tremulous movements; stimulus-sensitive; symmetric | Stops with passive flexion (seizures do NOT); higher frequency/lower amplitude |
| Sandifer syndrome | Dystonic posturing with GERD; arching, head rotation during feeds | Related to feeding; resolves with antireflux treatment (PPI) |
| Benign tonic upgaze | Sustained upward gaze; compensatory head retroflexion | Consciousness preserved; horizontal movements normal; resolves by age 2–4 |
| Spasmus nutans | Triad: head bobbing + nystagmus + torticollis | MRI MANDATORY to exclude chiasmal/hypothalamic glioma |
| Benign neonatal sleep myoclonus | Myoclonic jerks ONLY in sleep; may be dramatic | STOPS on awakening; WORSENED by benzos; resolves by 3–6 months |
| Non-epileptic head drops | Brief head drops mimicking epileptic spasms | Normal EEG (no hypsarrhythmia); self-gratification behavior |
💎 Board Pearl
- Benign neonatal sleep myoclonus: sleep-only, stops on waking, WORSENED by benzos — the paradoxical benzodiazepine worsening is a classic board question
- Jitteriness vs. seizure: jitteriness is stimulus-sensitive and suppressible by passive flexion; seizures are NOT suppressible — key bedside maneuver
Sleep Events vs. Nocturnal Seizures
| Condition | Key Features | Distinguishing Clue |
|---|---|---|
| SHE (Sleep-related Hypermotor Epilepsy) | Brief (<2 min); hypermotor; stereotyped; multiple per night; NREM | Any time of night; frequent (5–10+/night); genes: CHRNA4, CHRNB2, CHRNA2 |
| NREM parasomnias | Sleep terrors, sleepwalking, confusional arousals; longer; N3 | First third of night; variable/non-stereotyped; difficult to arouse |
| REM behavior disorder | Dream enactment; violent movements; vivid dream recall | Latter third of night; older adults; synucleinopathy association |
| Hypnagogic jerks | Single brief myoclonic jerk at sleep onset; falling sensation | Sleep-wake transition ONLY; single jerk; universal (60–70%) |
SHE vs. Parasomnias — Quick Comparison
- SHE: brief, stereotyped, frequent, any time of night, rapid return to sleep
- Parasomnias: longer, variable, infrequent, first third of night, hard to arouse
- Frontal seizures may have no scalp EEG correlate — video-EEG with extended sleep montage often required
Other Mimics
Hyperekplexia (Startle Disease)
- Gene: GLRA1 (glycine receptor α1-subunit); NOT epileptic
- Exaggerated startle to unexpected stimuli (especially nose tap); neonatal hypertonia
- Vigevano maneuver (forced head/limb flexion) aborts attacks; treatment: clonazepam
Tics & Stereotypies
- Suppressible (with effort); premonitory urge unique to tics
- Stereotypies: interruptible by distraction; no postictal state; normal EEG
Migraine with Aura vs. Seizure
- Migraine aura: gradual march over 20–30 minutes (cortical spreading depolarization)
- Seizure aura: rapid spread over seconds
- Key: minutes to develop = migraine; seconds = seizure
TIA vs. Seizure
- TIA: negative symptoms (weakness, numbness, vision loss)
- Seizure: positive symptoms (jerking, tingling, flashing lights)
- Exception: limb-shaking TIA (hemodynamic insufficiency) mimics focal motor seizures
Daydreaming vs. Absence Seizure
- Daydreaming: interruptible; longer duration; no automatisms
- Absence: NOT interruptible; abrupt onset/offset; eyelid flutter; 3 Hz spike-wave
- Hyperventilation provokes absences but NOT daydreaming — useful bedside test
Diagnostic Approach to Paroxysmal Events
Systematic Algorithm
- Step 1 — Detailed history: patient AND witnesses; triggers, timing, position, motor features, awareness, recovery
- Step 2 — Home video: smartphone recording; reviewed by specialist achieves ~89% PPV
- Step 3 — ECG for ALL: mandatory 12-lead; measure QTc; exclude cardiac channelopathies
- Step 4 — Routine EEG: sensitivity 50–60% on first study; normal EEG does NOT exclude epilepsy
- Step 5 — Sleep-deprived EEG: increases yield; serial studies reach 80–90%
- Step 6 — Prolonged video-EEG: gold standard; habitual event + normal EEG = excludes epilepsy
- Step 7 — MRI brain: epilepsy protocol if epilepsy suspected
- Consider: ambulatory EEG, exercise stress test, tilt-table test, PSG
Red Flags for Non-Epileptic Events
- Failure to respond to 2+ appropriately chosen ASMs
- Prolonged duration (>5 min) without postictal state
- Eye closure during convulsive event; waxing/waning motor activity
- Preserved responsiveness during bilateral motor event
Red Flags for Cardiac Cause
- Event during exercise, swimming, or emotional stress
- Abrupt LOC WITHOUT prodrome; rapid full recovery
- Family history of sudden cardiac death at young age (<40)
Board Pearls
💎 Board Pearl
- Ictal eye closure = PNES (96%/98%); epileptic seizures present with eyes OPEN — the single most reliable sign for distinguishing PNES from epileptic seizures
- "Seizure" during swimming = Long QT type 1 (KCNQ1) until proven otherwise — ECG with QTc measurement is the critical first test
- CPVT: 30% epilepsy misdiagnosis rate; resting ECG is NORMAL — exercise stress test showing bidirectional VT is diagnostic; the normal resting ECG is why it gets missed
- Benign neonatal sleep myoclonus: sleep-only, stops on waking, WORSENED by benzos — the paradoxical benzodiazepine worsening is a classic board question
- Lateral tongue bite = seizure; tip tongue bite = syncope — only objectively confirmed lateral laceration is specific; self-reported biting occurs equally in PNES and epilepsy
- Drug-resistant epilepsy failing 2 ASMs = mandatory diagnostic reassessment — 20–30% of these patients do not have epilepsy at all
- 10–15% of SUDEP autopsy cases have cardiac channel gene variants — overlap between epilepsy and cardiac channelopathy genetics
Clinical Pearls
Clinical Pearl
Prolactin for seizure diagnosis is a trap. While prolactin elevation after GTCS (not after PNES or absence) seems useful, it must be drawn within 10–20 minutes, is not elevated after frontal lobe seizures, and has limited sensitivity. A normal prolactin does NOT exclude epileptic seizures. Video-EEG remains the gold standard.
Clinical Pearl
SHE vs. parasomnias — the key is stereotypy. If the family describes the same bizarre nocturnal behavior happening identically multiple times per night, think SHE. If events are variable, prolonged, and happen only in the first hours of sleep, think parasomnia. Frontal seizures may have NO scalp EEG correlate.
Clinical Pearl
Convulsive syncope is not epilepsy. Up to 12% of syncope episodes include brief (<15 sec) tonic or myoclonic movements from cerebral hypoperfusion. These do NOT indicate epilepsy, do NOT require ASMs, and should not lead to an epilepsy diagnosis. Key: positional trigger, pallor, rapid recovery without postictal confusion.
References
- LaFrance WC Jr, Baker GA, Duncan R, et al. Minimum requirements for the diagnosis of psychogenic nonepileptic seizures: a staged approach. Epilepsia 2013;54(11):2005–2018.
- Goldstein LH, Robinson EJ, Mellers JDC, et al. Cognitive behavioural therapy for adults with dissociative seizures (CODES): a pragmatic, multicentre, randomised controlled trial. Lancet Psychiatry 2020;7(6):491–505.
- Chung SS, Gerber P, Kirlin KA. Ictal eye closure is a reliable indicator for psychogenic nonepileptic seizures. Neurology 2006;66(11):1730–1731.
- Lempert T, Bauer M, Schmidt D. Syncope: a videometric analysis of 56 episodes of transient cerebral hypoxia. Ann Neurol 1994;36(2):233–237.
- Schwartz PJ, Ackerman MJ, Antzelevitch C, et al. Inherited cardiac arrhythmias. Nat Rev Dis Primers 2020;6(1):58.
- Leenhardt A, Lucet V, Denjoy I, et al. Catecholaminergic polymorphic ventricular tachycardia in children. Circulation 1995;91(5):1512–1519.
- Tinuper P, Bisulli F, Cross JH, et al. Definition and diagnostic criteria of sleep-related hypermotor epilepsy. Neurology 2016;86(19):1834–1842.
- Bruno MK, Hallett M, Gwinn-Hardy K, et al. Clinical evaluation of idiopathic paroxysmal kinesigenic dyskinesia: new diagnostic criteria. Neurology 2004;63(12):2280–2287.
- Chen WJ, Lin Y, Xiong ZQ, et al. Exome sequencing identifies truncating mutations in PRRT2 that cause paroxysmal kinesigenic dyskinesia. Nat Genet 2011;43(12):1252–1255.
- Derry CP, Duncan JS, Berkovic SF. Paroxysmal motor disorders of sleep: the clinical spectrum and differentiation from epilepsy. Epilepsia 2006;47(11):1775–1791.
- Hallett M, Aybek S, Dworetzky BA, et al. Functional neurological disorder: new subtypes and shared mechanisms. Lancet Neurol 2022;21(6):537–550.
- Ertan D, Aybek S, LaFrance WC Jr, et al. Functional (psychogenic non-epileptic/dissociative) seizures: updated review. J Neurol Neurosurg Psychiatry 2022;93(10):1029–1037.
- Uldall P, Alving J, Hansen LK, et al. The misdiagnosis of epilepsy in children admitted to a tertiary epilepsy centre with paroxysmal events. Arch Dis Child 2006;91(3):219–221.
- Kotagal P, Costa M, Wyllie E, et al. Paroxysmal nonepileptic events in children and adolescents. Pediatrics 2002;110(4):e46.
- Continuum (Minneap Minn). Epilepsy. Volume 31, Number 1, February 2025.