HICU - Overview of standardized critical care terminology.pdf

PART III: EEG INTERPRETATION 16 Overview of Standardized Critical Care EEG Terminology Jessica W. Templer and Elizabeth E. Gerard...

PART III: EEG INTERPRETATION 16 Overview of Standardized Critical Care EEG Terminology Jessica W. Templer and Elizabeth E. Gerard IN THIS CHAPTER Concepts and organization of the American Clinical Neurophysiology Society (ACNS) standardized critical care EEG terminology Description of the main terms and modifiers Interrater reliability KEY POINTS Critical care EEG terminology was developed by the American Clinical Neurophysiology Society (ACNS) to standardize the description of rhyth- mic, periodic, and fluctuating patterns commonly encountered in critically ill patients. The core nomenclature of the ACNS terminology includes two discrete main terms: Main term 1 describes the location of each pattern as generalized (G), lateralized (L), bilateral independent (BI), or multifocal (Mf). Main term 2 describes the morphology of each pattern as periodic discharges (PDs), rhythmic delta activity (RDA), or spike-wave (SW). Additional modifiers include descriptions of frequency, amplitude, number of phases, prevalence, duration, and polarity. Among trained users, interrater reliability is almost perfect (kappa = 0.8–1.0) for main terms 1 and 2 and the majority of the modifiers. Recent studies have shown that this nomenclature can be used for multi- center research and may help identify which patients are most at risk for seizures. Copyright Springer Publishing Company. All Rights Reserved. From: Handbook of ICU EEG Monitoring, Second Edition 141 DOI: 10.1891/9780826168627.0016 142 Part III EEG Interpretation I. BACKGROUND A. History In 2005, the American Clinical Neurophysiology Society (ACNS) Subcommittee on Research Terminology for Continuous EEG Monitoring first proposed standard- ized terminology for EEG patterns frequently encountered in critically ill patients (1). The ACNS terminology has undergone several modifications since the original pub- lication and a revised version was published in 2013 (2,3) with changes made on the basis of solicited feedback and studies investigating interrater reliability (4,5). A recent study evaluated the interrater reliability of the most recent version of the terminology and found interrater agreement (IRA) for the main terms to be almost perfect among trained users, while agreement on the modifiers was more variable, though still substantial, or almost perfect, for the major modifiers (6). Although the terminology was originally developed for research purposes, it is now commonly used for clinical purposes at many academic medical centers utilizing continuous EEG (cEEG). B. Rationale for the development of standardized terminology cEEG monitoring is an important tool for assessing neurologic function, and is becoming the standard of care in many ICUs. There had been no consensus on the EEG terms used to describe the many peri- odic and rhythmic patterns frequently seen in critically ill patients, and which pat- terns are associated with neuronal injury and may require treatment. Many of the terms commonly used to describe EEG patterns in critically ill patients carry clinical connotations that have not been validated, including the term gener- alized periodic epileptiform discharges (GPEDs) which implies that the pattern is epileptiform, and the term triphasic waves which traditionally has been associated with a toxic or metabolic etiology. One of the main goals of the ACNS terminology is to create an objective EEG clas- sification scheme using descriptive terms that are not biased regarding the clinical significance of a given pattern. For example, under the ACNS terminology, the terms GPEDs (generalized periodic epileptiform discharges) and PLEDs (periodic lateralized epileptiform discharges) have been replaced by the terms GPDs (generalized periodic dis- charges) and LPDs (lateralized periodic discharges). The ACNS committee recognized that research regarding the clinical significance of these EEG patterns requires standardized terminology to allow for systematic, multicenter clinical trials, and for comparison of results between centers. Recently, a database has been created to allow for multicenter trials utilizing the ACNS terminology which has thus far compiled data from over 5,000 patients, and demonstrated that specific patterns and modifiers can aide in predicting which patients are at higher risk of seizures on continuous EEG (7). C. Concepts and organization The terminology is designed to describe patterns of uncertain significance and spe- cifically excludes patterns that most electroencephalographers consider definite electrographic seizures.  Chapter 16 Overview of Standardized Critical Care EEG Terminology 143 Therefore, the terminology excludes: Generalized spike-wave (SW)/sharp-waves greater than or equal to 3 per sec- ond Evolving discharges that reach greater than 4 per second (whether focal or gen- eralized) However, the terminology does include some patterns that meet criteria for elec- trographic seizures based on evolution in the field, morphology, or frequency, as long as the maximum frequency is less than 4 Hz. Generalized SW discharges at frequencies less than 3 Hz are also included. The terminology includes any rhythmic or periodic pattern that continues for at least six cycles (e.g., 1 per second for 6 seconds, or 3 per second for 2 seconds). All patterns described by the terminology are assigned two main terms. Several modifiers also can be applied as appropriate. II. BASICS (FIGURE 16.1) A. Main terms Main term 1: The location of a given pattern. Each pattern should be assigned only one of the following terms: Generalized (G)—any bilateral, symmetric, and synchronous pattern, even if the field is restricted. The pattern may be frontally, midline, or occipitally predomi- nant, but cannot be asymmetric. Additional localizing information: Frontally predominant: amplitude in anterior derivations 50% or less than posterior derivations Occipitally predominant: amplitude in posterior derivations 50% or less than anterior derivations Midline predominant: amplitude in midline derivations 50% or less than parasagittal derivations on an average or noncephalic referential recording Generalized, not otherwise specified Lateralized (L)—unilateral hemispheric or focal patterns. This also includes pat- terns seen synchronously over both hemispheres but clearly more prominent on one side (bilateral asymmetric). The lobe most involved should be specified (i.e., F, P, T, O, or hemispheric if unable to better localize). Patterns only seen over one hemisphere should be specified as “lateral- ized, unilateral.” Patterns that are synchronous and seen bilaterally but clearly more prom- inent over one hemisphere are termed “lateralized, bilateral, asymmetric.” Bilateral-independent (BI)—asynchronous hemispheric or focal patterns occur- ring independently over both hemispheres Specify symmetric versus asymmetric Specify lobes most involved in both hemispheres (i.e., F, P, T, O, or hemi- spheric if unable to better localize) Multifocal (Mf)—patterns occurring in at least three discrete brain regions and involving both hemispheres Specify symmetric versus asymmetric 144 Part III EEG Interpretation PLUS MODIFIERS TERM #1 TERM #2 (ADD ONLY IF PRESENT WITH PATTERN AND NOT IN BACKGROUND) Generalized (G) Periodic Discharges +F Superimposed FAST activity; use with PDs or RDA only (PDs) +R Superimposed RHYTHMIC activity; use with PDs only Lateralized (L) + FR Use for PDs if both subtypes apply Rhythmic Delta +F Superimposed FAST activity; use with PDs or RDA only Bilateral Activity Independent (BI) (RDA) +S Superimposed SHARP waves or Spikes; use with RDA only + FS Use for RDA if both subtypes apply Multifocal (Mf) Spike Wave No + modifiers (SW) MODIFIERS: Prevalence Duration Frequency Number of Sharpness (% of record) (cycles/sec) phases · Rare (