Risk of seizure clusters and status epilepticus following rapid and ultra-rapid medication discontinuation during intracranial EEG monitoring

• More than half of all patients whose anti-seizures medications were withdrawn either rapidly (i.e., tapered) or ultra-rapidly (i.e., held) at the start of invasive EEG (iEEG) developed seizure clusters • Among iEEG patients who had seizure complications, two-thirds resolved spontaneously • A longer duration of epilepsy was associated with higher odds of developing seizure complications during medication withdrawal for iEEG • Patients in the rapid withdrawal group had shorter latencies to first seizure, but there was no significant difference between the overall duration of the iEEG in the two groups • Patients in the rapid withdrawal group had higher odds of having status epilepticus relative to patients in the ultra-rapid withdrawal group, even after adjusting for potential confounders Anti-seizure medications (ASMs) are discontinued in the course of intracranial EEG (iEEG) monitoring for presurgical evaluation. The ASM withdrawal facilitates an emergence of seizures but may also precipitate seizure clusters (SC) and status epilepticus (SE). The aim of this study was to compare the rates of SC and SE during the ultra-rapid withdrawal (URW) and rapid withdrawal (RW) of ASMs during iEEG. We performed a retrospective observational study of all consecutive patients with drug resistant epilepsy who completed iEEG at our comprehensive epilepsy center from 2012-2018. SC was defined as three or more seizures in 24 h with a return to baseline between the events. SE was defined as ≥ 5 min of clinical seizure or ≥ 10 min of ictal electrographic activity or series of seizures with no return to the neurological baseline between the events. Of 107 patients who completed iEEG with intracranial grid or strip electrodes, 46 (43%) were male. Median age at the time of iEEG was 35.4 years (interquartile range [IQR], 26.4 - 44.9). Ninety patients (84.1%) had all AEDs held on admission, while 16 patients (15%) underwent a rapid taper. The median time to first seizure was 15.1 (8.2 - 22.6) h. Sixty-two patients (57.9%) developed SC, while 10 (9.4%) developed SE. Twenty-six patients (36.1%) with these complications required intravenous lorazepam or other rescue ASMs, while the remaining patients had spontaneous resolution of seizures; intubations were not required. While there were differences in the proportions in patients who experienced SC, SE, or neither in the URW and RW groups, these differences were not significant at the 0.05 alpha level. Ultra-rapid and rapid ASM withdrawal are accompanied by SC and SE the majority of which terminate spontaneously. These data support the use of either approach of the medication taper for seizure provocation in iEEG.