Rationale: Post-anoxic tonic eyelid opening (PATEO) is a poorly understood phenomenon defined as non-periodic eyelid opening (EO) seen in comatose patients within 24-hours of cardiac arrest. In this study, we aim to explore potential mechanisms of this rare phenomenon using EEG and surface-EMG (sEMG) data.
Methods: We studied three post-cardiac arrest patients with vEEG and sEMG covering various bulbar muscles, sternocleidomastoid and biceps. We analyzed the duration of EEG bursts, onset of myoclonus and EO in relation to the bursts using video and sEMG data and total duration of EO. We also analyzed the sequence of myoclonus using myogenic activity in sEMG, EKG and frontal EEG channels to characterize the myoclonus as cortical or reticular in origin. Correlational analysis was made using Pearson correlation coefficient, and literature review was performed.
Results: All three patients had a generalized burst-suppression pattern on EEG within 24-hours of cardiac arrest with myoclonus and EO time-locked to the bursts. In all patients, EO occurred after the onset of EEG bursts with an average latency of ~200-400ms. The total EO duration was ~1600-7500ms. In patients 1 and 3, there was a strong positive correlation between EEG burst duration and EO duration (r=.91 and .81 respectively). After analyzing sEMG data, we concluded that the myoclonus and EO were of cortical origin in patient 2, reticular in patient 3 and cortico-reticular in patient 1. In patients 1 and 2, there was a negative correlation between EEG burst duration and myoclonus/EO latency. In patient 3, there was a weak positive correlation instead. Extensive literature search found 12 previous published reports (42 subjects) of tonic eye opening in the setting of cardiac arrest, two of which (two subjects) we excluded for pre-existing seizure disorder, or unclear semiology. Of the remaining 40 subjects, all showed burst-suppression on EEG with onset of eye opening corresponding to the burst pattern, and 24 (60%) also reported additional non-ocular myoclonic. A total of six EEGs were analyzed further, showing onset of EO was 350
+ 180ms after burst onset, with EO duration of 500-4,100 msec. All subjects with PATEO in the setting of burst suppression died.
Conclusions: PATEO is associated with generalized burst-suppression pattern on EEG. The EEG burst and EO durations have a strong positive correlation. We propose that PATEO and acute post-anoxic myoclonus can be cortical (descending volley) or reticular (ascending volley) or cortico-reticular in origin. Polygraphic analysis using sEMG electrodes covering various bulbar muscles can help localize the generator. Activation of the central caudal nucleus of the oculomotor complex in the midbrain is most likely responsible for the EO. Levator palpebrae superioris contains unique slow-twitch fibers distinct from other extra-ocular muscles, allowing it to maintain tonic eyelid contraction, accounting for the longer EO duration.
Funding: None