Table 3

Summary of evidence that supports the signs used to distinguish between psychogenic non-epileptic seizures (PNES) and epileptic seizures (ES)

Sign that favour PNESEvidence from primary studiesSensitivity (%) for PNESSpecificity (%) for PNES
Long durationGood––
Fluctuating courseGood

  • 69 (events)

  • 47–88 (patients)

  • 96

  • 96–100

Asynchronous movementsGood (frontal-lobe partial seizures excluded)

  • 44–96 (events)

  • 9–56 (patients)

  • 93–96

  • 93–100

Pelvic thrustingGood (frontal-lobe partial seizures excluded)

  • 1–31 (events)

  • 7.4–44 (patients)

  • 96–100

  • 92–100

Side-to-side head or body movementGood (convulsive events only)

  • 25–63 (events)

  • 15–36 (patients)

  • 96–100

  • 92–100

Closed eyesGood

  • 34–88 (events)

  • 52–96 (patients)

  • 74–100

  • 97

Ictal cryingGood

  • 13–14 (events)

  • 3.7–37 (patients)

  • 100

  • 100

Memory recallGood

  • 63 (events)

  • 77–88 (patients)

  • 96

  • 90

Signs that favour ESEvidence from primary studiesSensitivity for ESSpecificity for ES
Occurrence from sleepGood

  • 31–59 (events)

  • –

  • 100

  • –

Postictal confusionGood

  • 61–100 (events)

  • 67 (patients)

  • 88

  • 84

Stertorous breathingGood (convulsive events only)

  • 61–91 (events)

  • –

  • 100

  • –

Other signsEvidence from primary studies
Gradual onsetInsufficient
Non-stereotyped eventsInsufficient
Flailing or thrashing movementsInsufficient
Opisthotonus, ‘arc en cercle’Insufficient
Tongue bitingInsufficient
Urinary incontinenceInsufficient

  • The sensitivity and specificity values were calculated from the frequencies of clinical signs in PNES and ES. We were not able to obtain the CIs in most cases.