This is an old revision of the document!


Signs and Symptoms of Seizures detectable Pediatric Eye Exam

Distinguishing Benign Non-Epileptic Ocular Paroxysmal Events from Seizures in Infants

Preserved consciousness, normal horizontal eye movements, absence of postictal features, and a normal ictal EEG are the key clinical discriminators that separate benign non-epileptic ocular paroxysmal events from seizures — but because clinical features overlap substantially in infants, polygraphic video-EEG remains the criterion standard when doubt persists (Falsaperla et al., Pediatr Neurol 2021; Lőrincz-Molnár et al., Dev Med Child Neurol 2026; Kartal, Pediatr Emerg Care 2019). The single most reliable clinical maneuver is capturing the event (ideally on parental video) and assessing responsiveness, the character of the eye movements, triggers, and the state (sleep vs. wake) during the episode (Falsaperla et al., Pediatr Neurol 2021; Lagorio et al., Neurol Clin Pract 2022).

Distinguishing features that separate PTU from seizures (e.g., partial seizures or atypical absences, for which these children are frequently referred) (Bonnet et al., Mov Disord 2010):

Each has preserved consciousness and a normal EEG: shuddering attacks/benign myoclonus of early infancy, benign paroxysmal torticollis, Sandifer syndrome (dystonic posturing during/after feeding with GERD), non-epileptic head drops, and opsoclonus (which must prompt evaluation for neuroblastoma) (Beinvogl et al., Pediatrics 2016; Lagorio et al., Neurol Clin Pract 2022; Paolicchi, Epilepsia 2002).

One important non-benign mimic is Glut1 deficiency syndrome, in which paroxysmal eye-head movements often begin before 6 months and precede seizures; these gaze-saccade movements are distinguished from opsoclonus by a clear intermovement fixation interval and a same-direction head movement, and recognition matters because the condition is treatable (Pearson et al., Neurology 2017).

Polygraphic video-EEG/EMG demonstrating paroxysmal motor activity (e.g., shuddering) with a completely normal concurrent EEG confirms a non-epileptic event (Caraballo et al., Epilepsia 2009).