American Clinical Neurophysiology Society's Standardized Critical Care EEG Terminology: 2021 Version.

Abstract

In the early 2000s, a subcommittee of the American Clinical Neurophysiology Society (ACNS) set out to “standardize terminology of periodic and rhythmic EEG patterns in the critically ill to aid in future research involving such patterns.” The initial proposed terminology was published in 2005.1 This was presented at many meetings on several continents, subjected to multiple rounds of testing of interrater reliability, underwent many revisions, and was then published as an ACNS guideline in 2013.2 Interrater agreement of the 2012 version (published in early 2013) was very good, with almost perfect agreement for seizures, main terms 1 and 2, the +S modifier, sharpness, absolute amplitude, frequency, and number of phases.3 Agreement was substantial for the +F and +R modifiers (66% and 67%) but was only moderate for triphasic morphology (58%) and fair for evolution (21%, likely at least partly because of the short EEG samples provided).3 The authors concluded that interrater agreement for most terms in the ACNS critical care EEG terminology was high and that these terms were suitable for multicenter research on the clinical significance of these critical care EEG patterns. With the help of infrastructure funding from the American Epilepsy Society and administrative and website support from the ACNS, a database that incorporated the ACNS terminology was developed for clinical and research purposes, tested during routine clinical care in multiple centers,4 and made available at no cost on the ACNS website (https://www.acns.org/research/critical-care-eeg-monitoring-research-consortium-ccemrc/ccemrc-public-database). This greatly enhanced the ability to complete multicenter investigations. After the establishment of the standardized terminology and free access to a database incorporating these terms, there have been many investigations into the clinical significance of rhythmic and periodic patterns (RPPs) in critically ill patients. Patterns such as lateralized rhythmic delta activity (LRDA) were found to be highly associated with acute seizures,5,6 equivalent to the association found with lateralized periodic discharges (LPDs) in one study.5 The association of all the main patterns in the nomenclature with seizures was defined in a multicenter cohort of almost 5,000 patients, with seizure rates highest for LPDs, intermediate for LRDA and generalized periodic discharges (GPDs), and lowest for generalized rhythmic delta activity (GRDA).6 This and other studies have shown that several of the modifiers within the nomenclature do indeed have clinically relevant meaning. For example, studies have shown that higher frequency (especially >1.5 Hz), higher prevalence, longer duration, and having a “plus” modifier are all associated with a higher chance of acute seizures.6,7 On the other hand, whether a pattern was spontaneous or “stimulus-induced” did not seem to have a significant effect on its association with seizures.6 In other investigations, the “triphasic morphology” modifier was investigated blindly with multiple expert reviewers, calling into question its relationship with metabolic encephalopathy and its lack of a relationship with seizures.8,9 For patients with refractory status epilepticus treated with anesthetic-induced coma, the presence of “highly epileptiform” bursts suggested that an attempted wean off of anesthetics at that time was much more likely to lead to seizure recurrence than if the bursts were not highly epileptiform.10 Even long-term outcome seemed to be associated with some modifiers, with a higher risk of later epilepsy found if LPDs were more prevalent, had longer duration, or had a “plus” modifier.7