After-hours EEG is increasingly used in hospital patients. Although the detection of seizures and interictal epileptiform discharges has been shown to be higher with prolonged EEG (pEEG) than routine-duration EEG, the relative value for particular indications can inform utilization.
The Mayo EEG Report System was queried for after-hours emergent routine EEG (ErEEG) and pEEG performed between January 2015 and June 2015. Total 296 after-hours ErEEG were identified, of which 140 converted into pEEG were included in the study for direct comparison of two modalities. Indications were categorized as: mental status changes, recent seizures rule out continued nonconvulsive seizures, spells, and prognosis after anoxic brain injury. Categorical data were analyzed using the McNemar and Fisher exact tests; a P value of 0.05 was considered significant.
Prolonged EEG was superior to ErEEG for detection of interictal epileptiform discharges (61 vs. 48/140, P = 0.004) and seizures (29 vs. 17/140, P = 0.012). Seizure detection was greater for pEEG than ErEEG for the indication of evaluating for subclinical seizures after recent observed clinical seizures (14/41 [34.1%] versus 4/41 [9.8%], P = 0.002). There was no significant difference between modalities for seizure detection in patients undergoing evaluation of spells or mental status changes without previous observed seizures. Detection of seizures on pEEG was higher for recordings greater than 24 hours (8/46, 17%, P = 0.014) and 48 hours (19/26, 73%, P < 0.001) than recordings less than 24 hours (2/68, 3%). Seizure detection was higher with pEEG in comatose patients than ErEEG (17/51, 33% vs. 12/89, 13%; P = 0.009).
Increased value was demonstrated for pEEG over ErEEG in patients undergoing evaluation after observed recent clinical seizures and for coma. No significant difference was found between ErEEG and pEEG for seizure detection in the relatively small subgroups of patients with mental status changes other than coma and without preceding seizure or spells.