Standardized Critical Care EEG Terminology PDF

L. J. Hirsch, et al. Standardized Critical Care EEG Terminology FIG. 20. Periodic Discharges (PDs)....

L. J. Hirsch, et al. Standardized Critical Care EEG Terminology FIG. 20. Periodic Discharges (PDs). 1. Repetition of a waveform with relatively uniform morphology and duration, 2. with a clearly discernable interdischarge interval between consecutive waveforms, and 3. recurrence of the waveform at nearly regular intervals: having a cycle length (i.e., period) varying by ,50% from one cycle to the next in the majority (.50%) of cycle pairs. A pattern can qualify as rhythmic or periodic if and only if it continues for at least 6 cycles (e.g. 1 Hz for 6 seconds, or 3 Hz for 2 seconds). 2. Main Term 2: PDs, RDA or SW (i.e., the period) of the rhythmic pattern should vary by a. Periodic Discharges (PDs): ,50% from the duration of the subsequent cycle for most  Periodic: Repetition of a waveform with relatively uniform (.50%) cycle pairs to qualify as rhythmic. An example of a morphology and duration with a clearly discernible inter- rhythmic pattern would be a sinusoidal waveform, although discharge interval between consecutive waveforms and there are other examples; a pattern can be sharp at the top recurrence of the waveform at nearly regular intervals and/or the bottom of the waveform and still be rhythmic (but (Fig. 20). “Nearly regular intervals” is defined as having a would no longer be sinusoidal). Irregular or polymorphic cycle length (i.e., period) varying by ,50% from one cycle delta should not be reported as RDA. to the next in most (.50%) cycle pairs.  RDA: Rhythmic activity 0.5 to #4.0 Hz.  Discharges: Waveforms lasting ,0.5 seconds, regardless of c. Spike-and-wave or Sharp-and-wave (SW): Spike, polyspike, number of phases, or waveforms $ 0.5 seconds with no or sharp wave consistently followed by a slow wave in a more than 3 phases. This is as opposed to Bursts, defined as regularly repeating and alternating pattern (spike-wave- waveforms lasting $ 0.5 seconds and having at least 4 spike-wave-spike-wave), with a consistent relationship phases. Discharges and bursts must clearly stand out from between the spike (or polyspike or sharp wave) component the background activity. and the slow wave for at least six consecutive cycles and with no interval between one spike-wave complex and the next b. Rhythmic Delta Activity (RDA): (see Supp EEG 14, Supplemental Digital Content 1, http://  Rhythmic: Repetition of a waveform with relatively uniform links.lww.com/JCNP/A134) (Fig. 22) (if there is an interval, morphology and duration and without an interval between this would qualify as PDs, where each discharge is a spike- consecutive waveforms (Fig. 21). The duration of one cycle and-wave). FIG. 21. Rhythmic Delta Activity (RDA). 1. Repetition of a waveform with relatively uniform morphology and duration and 2. without an interval between consecutive waveforms. 3. The duration of one cycle (i.e., the period) of the rhythmic pattern should vary by ,50% from the duration of the subsequent cycle for the majority (.50%) of cycle pairs to qualify as rhythmic. A pattern can qualify as rhythmic or periodic if and only if it continues for at least 6 cycles (e.g. 1 Hz for 6 seconds, or 3 Hz for 2 seconds). 12 Journal of Clinical Neurophysiology Volume 38, Number 1, January 2021 clinicalneurophys.com Copyright © by the American Clinical Neurophysiology Society. Unauthorized reproduction of this article is prohibited.  Standardized Critical Care EEG Terminology L. J. Hirsch, et al. FIG. 22. “Spike-and-wave” or “Sharp- and-wave” (SW). Spike-and-wave or Sharp-and-wave (SW): Polyspike, spike, or sharp wave consistently followed by a slow wave in a regularly repeating and alternating pattern (spike-wave-spike- wave-spike-wave), with a consistent relationship between the spike (or polyspike or sharp wave) component and the slow wave for at least 6 cycles; and with no interval between one spike-wave complex and the next (if there is an interval, this would qualify as PDs, where each discharge is a spike-and-wave). NOTE: A pattern can qualify as rhythmic or periodic if and only if it cutoffs listed below for each term. Also record the longest continues for at least six consecutive cycles (e.g., 1 Hz for 6 seconds, continuous duration. or 3 Hz for 2 seconds). i. Very long: $ 1 hour. NOTE: If a pattern qualifies as both PDs and RDA simulta- ii. Long: 10 to 59 minutes. neously, it should be coded as PDs1R rather than RDA1S (see iii. Intermediate duration: 1 to 9.9 minutes. “plus” modifier below). iv. Brief: 10 to 59 seconds. v. Very brief: ,10 seconds. c. Frequency ¼ rate per second: Specify typical rate and range 3. Main Modifiers (Most of the following section can for all patterns (e.g., LPDs with typical frequency of 1 Hz and be applied to any EEG phenomenon) range of 0.5–2 Hz). Record typical, minimum, and maximum frequency using a. Prevalence: Specify percent of record or epoch that includes the following categories: ,0.5, 0.5, 1, 1.5, 2, and 2.5; and if very the pattern. This should be based on the proportion of an epoch brief duration: 3, 3.5, and 4 Hz. that includes or is within the pattern. The time between widely NOTE: For PDs and SW, typical frequencies.2.5 Hz can spaced PDs counts as part of the pattern duration. For example, only be applied to RPPs ,10 second duration (“very brief” by 2-Hz PDs present for 1 minute every 10 minutes is 10% definition); if PDs or SW have a typical frequency.2.5 Hz and prevalence, and 0.25-Hz PDs present for 1 minute every are $ 10 seconds, these would qualify as electrographic seizures 10 minutes is also 10% prevalence. When categorizing or (criterion A) and should be referred to as such rather than as PDs using qualitative terms, follow the cutoffs listed below for each or SW. term. Suggested clinical terms are given as well. If two or NOTE: No RPP in this terminology can have a typical more patterns are present, record the presence and prevalence frequency of.4 Hz; if an RPP is.4 Hz and $ 0.5 seconds, it of each one (e.g., w20% GRDA, 20% GPDs, and 30% would always meet the criteria for either BIRDs (if ,10 seconds) BIPDs). or electrographic seizure (if $ 10 seconds) (see definitions i. Continuous: $ 90% of record/epoch. below). If ,0.5 seconds, this would not qualify as any RPP ii. Abundant: 50% to 89% of record/epoch. but might qualify as a polyspike. iii. Frequent: 10% to 49% of record/epoch. iv. Occasional: 1% to 9% of record/epoch. d. Number of phases ¼ 1 1 number of baseline crossings of the v. Rare: ,1% of record/epoch. typical discharge as assessed in longitudinal bipolar and in the channel in which it is most readily appreciated. A phase is that b. Duration: Specify typical duration of pattern if not continuous. part of the signal that is on one side of (above or below) the When categorizing or using qualitative terms, follow the imaginary baseline (Fig. 23). The start and end points do not FIG. 23. The Number of Phases. Number of Phases ¼ 1 1 number of baseline crossings of the typical discharge. In this case there are a total of 2 baseline crossings, therefore the number of phases is 1 1 2 ¼ 3 phases. A phase is the part of the signal above or below the imaginary baseline. In this case, phase 1 (pink) is above, phase 2 (blue) is below, and phase 3 (yellow) is above again. clinicalneurophys.com Journal of Clinical Neurophysiology Volume 38, Number 1, January 2021 13 Copyright © by the American Clinical Neurophysiology Society. Unauthorized reproduction of this article is prohibited.  L. J. Hirsch, et al. Standardized Critical Care EEG Terminology count as baseline crossings. Applies to PDs and the entire iii.Spontaneous: never clearly induced, exacerbated, improved, spike-and-wave or sharp-and-wave complex of SW (including or terminated by stimulation. the slow wave). This does not apply to RDA. Categorize as 1, iv. Unknown: includes unclear or untested. 2, 3 or.3. e. Sharpness: Specify for both the dominant phase (phase with NOTE: Specify type of stimulus (auditory; light tactile; patient greatest voltage) and the sharpest phase if different. For both care and other nonnoxious stimulations; or noxious: suction, sternal phases, describe the typical discharge. Applies only to PDs and rub, nailbed pressure, nostril tickle, trapezius squeeze, or other). the spike/sharp component of SW, not RDA. Categorize as NOTE: The term “SIRPIDs” refers to stimulus-induced (or i. Spiky: duration of that component, measured at the EEG stimulus-exacerbated) rhythmic, periodic, or ictal-appearing baseline, is ,70 ms discharges and is a term that includes all SI- patterns together ii. Sharp: duration of that component is 70 to 200 ms (SI-RDA, SI-PDs, SI-SW, SI-IIC, SI-BIRDs, or SI-seizures). In iii.Sharply contoured: used for waveforms that have a sharp general, one should refer to the specific “SI–" pattern rather morphology (steep slope to one side of the wave and/or than using the general term “SIRPIDs,” especially for a given pointy or apiculate at inflection point[s]) but are too long in patient. duration to qualify as a sharp wave. iv. Blunt: having smooth or sinusoidal morphology. h. Evolution: Evolving, Fluctuating, or Static: terms refer to f. Voltage (amplitude) [of PDs, SW or RDA; not background changes in frequency, location, or morphology. EEG, which is in section A8, page 7, above]: i. Evolving: At least two unequivocal, sequential changes in i. Absolute: Typical voltage measured in standard longitu- frequency, morphology, or location defined as follows: Evolu- dinal bipolar 10-20 recording in the channel in which the tion in frequency is defined as at least two consecutive changes pattern is most readily appreciated. For PDs, this refers to in the same direction by at least 0.5 Hz, e.g., from 2 to 2.5 to 3 the highest voltage component. For SW, this refers to the Hz, or from 3 to 2 to 1.5 Hz (Fig. 24); Evolution in morphology spike/sharp wave. Voltage should be measured from peak to is defined as at least two consecutive changes to a novel trough (not peak to baseline). Specify for RDA as well. morphology (Fig. 25); Evolution in location is defined as Categorize as sequentially spreading into or sequentially out of at least two a. Very low: ,20 mV different standard 10-20 electrode locations (Fig. 26). The two b. Low: 20 to 49 mV consecutive changes must be in the same category (frequency, c. Medium: 50 to 149 mV morphology or location) to qualify. d. High: $ 150 mV  To qualify as evolution in frequency, a single frequency ii. Relative: For PDs only (PDs require two voltages, absolute must persist for at least three cycles (e.g., 1 Hz for 3 seconds, and relative). Typical ratio of voltage of the highest voltage or 3 Hz for 1 second). Thus, the following pattern would component of the typical discharge to the voltage of the typical qualify as evolving: 3 Hz for $ 1 second, then 2 Hz for background between discharges, measured in the same channel $ 1.5 seconds (the first change), then 1.5 Hz for $ 2 and montage as absolute voltage. Categorize as #2 or.2. seconds (the 2nd change) (see Supp EEG 16, Supplemental Digital Content 1, http://links.lww.com/JCNP/A134).  To qualify as evolution in morphology, each different g. Stimulus-Induced (SI-) or Stimulus-Terminated (ST-): SI- morphology or each morphology plus its transitional forms versus ST- versus spontaneous: Categorize as must last at least three cycles. Thus, the following example i. Stimulus-Induced (SI-): reproducibly brought about or would qualify: Spiky 4-phase PDs for three cycles then sharp exacerbated by an alerting stimulus, with or without 2–3 phase PDs for three cycles then blunt diphasic PDs for clinical alerting, when a patient is in their less-stimulated three cycles. state; an SI- pattern may also be seen spontaneously at  To qualify as evolution in location, the pattern must spread other times (due to spontaneous alerting or arousal) (see into or out of two standard 10-20 electrode locations and the Supp EEG 15, Supplemental Digital Content 1, http:// involvement of each additional electrode must be present for links.lww.com/JCNP/A134). Even if most instances of at least three cycles, e.g., 1-Hz LPDs only at T7, the pattern are spontaneous, it can still qualify as “SI-” if spreading to include F7 for 3 seconds then F7, T7, and it can be reproducibly brought about by an alerting P7 for 3 seconds. stimulus.  The criteria for evolution must be reached without the ii. Stimulus-Terminated (ST-): reproducibly terminated or attenuated evolving feature (frequency, morphology, or location) by an alerting stimulus, with or without clinical alerting, when a remaining unchanged for five or more patient is in their less-stimulated state; an ST- pattern may also continuous minutes. Thus, the following pattern would self-terminate at other times. Even if most instances of the pattern not qualify as evolving: 3 Hz for 1 minute, then 2 Hz for resolve or attenuate spontaneously, it can still qualify as “ST-” if it 7 minutes, then 1.5 Hz for 2 minutes. can be reproducibly terminated or attenuated by an alerting  Evolution in voltage (amplitude) alone does not qualify as stimulus. evolving and does not qualify as a different morphology. 14 Journal of Clinical Neurophysiology Volume 38, Number 1, January 2021 clinicalneurophys.com Copyright © by the American Clinical Neurophysiology Society. Unauthorized reproduction of this article is prohibited.  Standardized Critical Care EEG Terminology L. J. Hirsch, et al. FIG. 24. Evolution of frequency. At least 2 unequivocal, sequential changes in frequency; defined as at least 2 consecutive changes in the same direction by at least 0.5 Hz. To qualify as present, a single frequency must persist for at least 3 cycles. The criteria for evolution must be reached without the evolving feature (frequency) remaining unchanged for 5 or more continuous minutes. FIG. 25. Evolution of morphology. At least 2 consecutive changes to a novel morphology. Each different morphology or each morphology plus its transitional forms must last at least 3 cycles. FIG. 26. Evolution of location. Defined as sequentially spreading into or sequentially out of at least two different standard 10–20 electrode locations. To qualify as present, a single location must persist for at least 3 cycles. clinicalneurophys.com Journal of Clinical Neurophysiology Volume 38, Number 1, January 2021 15 Copyright © by the American Clinical Neurophysiology Society. Unauthorized reproduction of this article is prohibited.  L. J. Hirsch, et al. Standardized Critical Care EEG Terminology FIG. 27. Fluctuating frequency. $ 3 changes, not more than one minute apart, in frequency (by at least 0.5 Hz), but not qualifying as evolving. This includes patterns fluctuating from 1 to 1.5 to 1 to 1.5 Hz. To qualify as present, a single frequency must persist at least 3 cycles (e.g. 1 Hz for 3 seconds, or 3 Hz for 1 seconds). FIG. 28. Fluctuating morphology. $ 3 changes, not more than one minute apart, in morphology, but not qualifying as evolving. This includes patterns alternating between 2 morphologies repeatedly. To qualify as present, a single morphology must persist at least 3 cycles. FIG. 29. Fluctuating location. $ 3 changes, not more than one minute apart, in location (by at least 1 standard inter- electrode distance), but not qualifying as evolving. This includes patterns spreading in and out of a single electrode repeatedly. To qualify as present, a single location must persist at least 3 cycles. 16 Journal of Clinical Neurophysiology Volume 38, Number 1, January 2021 clinicalneurophys.com Copyright © by the American Clinical Neurophysiology Society. Unauthorized reproduction of this article is prohibited.  Standardized Critical Care EEG Terminology L. J. Hirsch, et al. FIG. 30. Lateralized Periodic Discharges PLUS fast activity (LPDs1F). Code as 1F if the fast activity is part of the RDA or PDs pattern and not simply part of the background activity. fast activity cycling with the periodic discharge. NOTE: Evolution of an RPP is now limited to patterns that then 1.5 Hz for 30 seconds, then 2 Hz for 5 minutes. The are #4 Hz AND ,10 seconds duration. Any.4-Hz RPP with changes are too far apart (.1 minute). evolution lasting ,10 seconds would qualify as definite BIRDs (see Section E, page 24). Any RPP with evolution lasting $ 10  The following would qualify as fluctuating: 2 Hz for 10 seconds meets criterion B of an electrographic seizure and should seconds, then 2.5 Hz for 30 seconds, then 2 Hz for 5 be coded as such. seconds, then 2.5 Hz for 5 seconds. ii. Fluctuating: $ 3 changes, not more than 1 minute apart, in iii. Static: Not qualifying as evolving or fluctuating. frequency (by at least 0.5 Hz) (Fig. 27), morphology (Fig. NOTE: Change in voltage (amplitude) alone would not 28), or location (by at least 1 standard interelectrode distance) qualify as evolving or fluctuating. (Fig. 29), but not qualifying as evolving (see Supp EEG 17, NOTE: If evolving or fluctuating in frequency, a minimum Supplemental Digital Content 1, http://links.lww.com/JCNP/ and maximum frequency should be specified under the “fre- A134). This includes patterns fluctuating from 1 to 1.5 to 1 to quency” modifier above. For nongeneralized patterns, specify 1.5 Hz; alternating between two morphologies repeatedly; or degree of spread (none, unilateral, or bilateral). spreading in and out of a single additional electrode location repeatedly. To qualify as present, a single frequency, i. Plus (1) ¼ additional feature which renders the pattern more morphology, or location must persist at least three cycles ictal-appearing (i.e., more closely resembling an EEG pattern (e.g., 1 Hz for 3 seconds, or 3 Hz for 1 second). seen during seizures) than the usual term without the plus. This modifier applies only to PDs and RDA, not SW.  The following would not qualify as fluctuating: 2 Hz for 30 Subtyping of “1”: all cases with “1” should be categorized as seconds, then 1.5 Hz for 30 seconds, then 2 Hz for 3 minutes, follows into 1F, 1R, 1FS, or 1FR: clinicalneurophys.com Journal of Clinical Neurophysiology Volume 38, Number 1, January 2021 17 Copyright © by the American Clinical Neurophysiology Society. Unauthorized reproduction of this article is prohibited.  L. J. Hirsch, et al. Standardized Critical Care EEG Terminology FIG. 31. Rhythmic Delta Activity PLUS fast activity (RDA1F). If a pattern qualifying as RDA or PDs has associated continuous fast frequencies (theta or faster), this can and should be coded as 1F if the fast activity is not present in the background activity when the RDA or PDs is not present. fast activity cycling with the rhythmic delta and having a stereotyped relationship to the delta wave. EDB ¼ Extreme Delta Brush. i. “1F”: with superimposed (some prefer the synonyms of NOTE: If a pattern qualifies as both PDs and RDA admixed or associated) fast activity, defined as theta or faster, simultaneously with approximately equal prominence, it should whether rhythmic or not. “1F” can be applied to PDs (see Supp be coded as PDs1R rather than RDA1S. EEGs 18 and 19, Supplemental Digital Content 1, http://links. NOTE: Re: Bilateral “1” versus unilateral: If a pattern is lww.com/JCNP/A134) (Fig. 30) or RDA (Figs. 31A and 31b). bilateral and qualifies as plus on one side, but not on the other, ii. “1R”: with superimposed rhythmic or quasi-rhythmic delta the overall main term should include the plus (although one activity; can be applied to PDs only (Fig. 32). side does not warrant a plus). For example, bilateral indepen- iii. “1S”: with associated sharp waves or spikes, or sharply dent periodic discharges with fast activity in one hemisphere contoured; can be applied to RDA only (see Supp EEG 20, only (PDs on one side, and independent PDs1F on the other) Supplemental Digital Content 1, http://links.lww.com/ would qualify for BIPDs1F (see Supp EEG 21, Supplemental JCNP/A134). The sharp contour, sharp waves, or spikes Digital Content 1, http://links.lww.com/JCNP/A134) (Fig. need to occur at least once every 10 seconds but not as part 34). Similarly, generalized rhythmic delta activity with of an SW pattern (Fig. 33). associated spikes in one hemisphere only (RDA on one side NOTE: It is possible to have “1FR” for PDs or “1FS” for and synchronous RDA1S on the other) would qualify for RDA. GRDA1S. NOTE: The wave within periodic spike-wave discharges NOTE: Re: 1F: If a pattern qualifying as RDA or PDs has (spike-wave-interval-spike-wave-interval-spike-wave-interval.) associated continuous fast frequencies (theta or faster), this does not qualify a pattern as PD1R because the wave is simply can and should be coded as 1F if the fast activity is not part of the spike-wave complex (which is the periodic discharge present in the background activity when the RDA or PDs is itself). However, RDA occurring at the same time as PDs but not present. In other words, code as 1F if the fast activity is without time-locked association with the PDs would qualify as part of the RDA or PD pattern and not simply part of the PD1R. background activity (Figs. 30B and 30C). When referring to 18 Journal of Clinical Neurophysiology Volume 38, Number 1, January 2021 clinicalneurophys.com Copyright © by the American Clinical Neurophysiology Society. Unauthorized reproduction of this article is prohibited.  Standardized Critical Care EEG Terminology L. J. Hirsch, et al. FIG. 32. Periodic Discharges PLUS RDA (PDs1R). RDA occurring at FIG. 34. Bilateral Independent Periodic Discharges PLUS fast the same time as PDs but without time-locked association with the activity (BIPDs1F). BIPDs with fast activity in one hemisphere only PDs would qualify as PDs1R. (PD on one side, and PD 1F on the other) would qualify for BIPDs1F. PDs1F, the fast activity can either be continuous (as long as the fast activity has a stereotyped relationship to the delta the fast activity was not present when the PDs were not wave (i.e., periodic delta brushes) (see Supp EEG 22, present) or can occur with each discharge in a regular fashion Supplemental Digital Content 1, http://links.lww.com/ (regardless of background). JCNP/A134) (Figs. 35A and 35C). Possible EDB: NOTE: “Extreme Delta Brush (EDB)”: A specific subtype Satisfying criterion A) or B) above EXCEPT either: of 1F (Table 2): i. only occasional or frequent (rather than abundant or  Definite EDB: Consists of either abundant or continuous: continuous) OR A. RDA1F, in which the fast activity has a stereotyped ii. the superimposed fast activity lacks a stereotyped relationship to the delta wave (e.g., always maximal relationship to the delta wave; continuous, invariant on the upstroke, crest, or downstroke of the wave) fast activity during RDA would fall into this category (Figs. 35A and 35B); OR (Figs. 35B and 35C). B. PDs1F, in which each PD consists of a single blunt NOTE: EDB is a subtype of 1F, therefore it must qualify as delta wave with superimposed fast activity, and in which 1F for it to also be considered as EDB. RDA with fast activity in the background and not associated with the pattern does not qualify as 1F and therefore cannot qualify as EDB. TABLE 2. Relationship between RDA1F, PDs1F and Extreme Delta Brush (EDB) RDA1F; or PDs1F if (and only if) the PDs are blunt delta waves Continuous/ Frequent/ Abundant Occasional (‡50% of (‡1 to 49% of record/epoch) record/epoch) Fast activity WITH Definite EDB Possible EDB stereotyped relationship to delta wave Fast activity Possible EDB RDA1F or PDs1F, WITHOUT but NOT EDB FIG. 33. Generalized Rhythmic Delta Activity PLUS Spikes stereotyped (GRDA1S). Generalized rhythmic delta activity with associated relationship to delta spikes in one hemisphere only (RDA on one side and synchronous wave RDA 1S on the other) would still qualify as GRDA1S. clinicalneurophys.com Journal of Clinical Neurophysiology Volume 38, Number 1, January 2021 19 Copyright © by the American Clinical Neurophysiology Society. Unauthorized reproduction of this article is prohibited.  L. J. Hirsch, et al. Standardized Critical Care EEG Terminology FIG. 35. Extreme Delta Brush (EDB). A. This is a subset of 1F, with abundant or continuous RDA1F or PDs1F (only if the PDs are blunt delta waves), where the fast activity has a stereotyped relationship to each delta wave. B. Extreme delta brush (EDB): RDA subtype. Examples of RDA that meet criteria for definite and possible EDB (A-C); with an example that does not (D), as the fast activity is part of the background (therefore this pattern does not count as RDA1F). C. Extreme Delta Brush (EDB): PD subtype. Examples of PDs (where the PDs are blunt delta waves) that meet criteria for definite and possible EDB. Example B is not EDB as the 1F is not on the delta wave (i.e., it is not delta brushes). Example E is not 1F as the fast activity is part of the background, therefore it can not be EDB. 20 Journal of Clinical Neurophysiology Volume 38, Number 1, January 2021 clinicalneurophys.com Copyright © by the American Clinical Neurophysiology Society. Unauthorized reproduction of this article is prohibited.  Standardized Critical Care EEG Terminology L. J. Hirsch, et al. FIG. 36. A. Generalized Extreme Delta Brush (GEDB). GRDA1F also qualifies as definite GEDB if the RDA1F is abundant or continuous; and as possible GEDB if the RDA1F is occasional or frequent. B. Lateralized Extreme Delta Brush (LEDB). LRDA1F also qualifies as definite LEDB if the LRDA1F is abundant or continuous; and as possible LEDB if the LRDA1F is occasional or frequent. C. Bilateral Independent Extreme Delta Brush (BIEDB). BIRDA1F also qualifies as definite BIEDB if the BIRDA1F is abundant or continuous; and as possible EDB if the BIRDA1F is occasional or frequent. clinicalneurophys.com Journal of Clinical Neurophysiology Volume 38, Number 1, January 2021 21 Copyright © by the American Clinical Neurophysiology Society. Unauthorized reproduction of this article is prohibited.  L. J. Hirsch, et al. Standardized Critical Care EEG Terminology second (positive) phase of highest voltage (see Supp EEG 23, Supplemental Digital Content 1, http://links.lww.com/JCNP/ A134); or the same but with the first (negative) phase of sufficiently low voltage to be obscured by background activity, leaving a biphasic waveform, positive-negative in polarity. Note that a biphasic waveform may be categorized as “triphasic” by this definition. The phrase “with triphasic morphology” should be added to the appropriate term when this modifier applies. This modifier applies to PDs and SW, but not RDA; it can also be used to describe sporadic discharges. c. “Anterior-posterior lag” or “posterior-anterior lag”: A lag is present if there is a consistent measurable delay of.100 ms from the most anterior to the most posterior derivation in which it is seen, or vice versa (see Supp EEG 23, FIG. 37. Anterior-posterior (AP) lag. Supplemental Digital Content 1, http://links.lww.com/ JCNP/A134) (Fig. 37); specify typical delay in millisec- onds from anterior to posterior (negative ¼ posterior to NOTE: The only periodic pattern that can qualify for anterior lag) in both a longitudinal bipolar and a referen- EDB is periodic delta brushes. Any other periodic pattern with tial montage, preferably with an ipsilateral ear reference. superimposed fast activity remains PD1F only. By similar This applies to PDs or the spike/sharp wave component of notion PD1F with the fast in between periodic delta waves SW. would also not qualify as EDB. This is because the fast d. Polarity: Specify for the dominant phase (phase with the greatest activity is not associated with the wave (i.e., it is not periodic voltage) only. Should be determined in a referential montage. delta brushes). Describe the typical discharge. Applies only to PDs and the spike/ NOTE: EDB can be in any location as any other form of RDA sharp component of SW, not RDA. Categorize as or PDs (i.e., generalized [Fig. 36A], lateralized [Fig. 36B], bilateral i. Positive independent [Fig. 36C], unilateral independent or multifocal). ii. Negative NOTE: There are multiple other features that may make a pattern iii. Dipole, tangential more “ictal-appearing,” such as increased sharpness, higher iv. Unclear voltage (amplitude), and fluctuation, but these are already accounted for in the other modifiers. 4. Minor Modifiers (Most of the following section D. ELECTROGRAPHIC AND can be applied to any EEG phenomenon) ELECTROCLINICAL SEIZURES a. “Sudden onset” versus “gradual onset” (“sudden onset” 1. Electrographic Seizures (ESz) preferred over “paroxysmal”). Sudden onset is defined as ESz (largely based on the Salzburg criteria)11,12 is defined as progressing from absent to well-developed within 3 seconds. either: b. “Triphasic morphology”: Three phases, negative-positive- a. Epileptiform discharges* averaging.2.5 Hz for $ 10 seconds negative, with each phase longer than the previous, and the (.25 discharges in 10 seconds), OR FIG. 38. Electrographic seizure (ESz). 22 Journal of Clinical Neurophysiology Volume 38, Number 1, January 2021 clinicalneurophys.com Copyright © by the American Clinical Neurophysiology Society. Unauthorized reproduction of this article is prohibited.  Standardized Critical Care EEG Terminology L. J. Hirsch, et al. FIG. 39. Electroclinical seizure (ECSz). b. Any pattern with definite evolution as defined above and literature to change this, we maintained the status quo in this lasting $ 10 seconds (see Supp EEG 24a, b, and c, matter. Hopefully, future investigations will help determine Supplemental Digital Content 1, http://links.lww.com/ the proper minimum duration for defining a seizure, if there JCNP/A134) (Fig. 38). is one. *NOTE: Electrographic seizures can consist of sharply contoured discharges that are not technically “epileptiform.” 2. Electrographic Status Epilepticus (ESE) For example,.25 sharply contoured discharges in 10 seconds is ESE is defined as an ESz for $ 10 continuous minutes or still a seizure, although each discharge can be.200 ms duration for a total duration of $ 20% of any 60-minute period of (and therefore technically not “epileptiform”). recording. The 10 minute cutoff matches the definition of focal NOTE: Whether to maintain or eliminate the “10 second status epilepticus with impaired consciousness by the Inter- rule” (clearly an arbitrary cutoff) was a matter of significant national League Against Epilepsy.17 The 20% cutoff, lowered debate among the authors and the greater EEG community from the previous 50%, is based on expert consensus and on surveyed during creation of this version of the nomenclature. one study in critically ill children in whom the risk of However, because there was no consensus or convincing new neurological decline was significantly greater when the FIG. 40. Electroclinical seizure (ECSz)dfor patients with previous known epileptic encephalopathy. clinicalneurophys.com Journal of Clinical Neurophysiology Volume 38, Number 1, January 2021 23 Copyright © by the American Clinical Neurophysiology Society. Unauthorized reproduction of this article is prohibited.  L. J. Hirsch, et al. Standardized Critical Care EEG Terminology maximum hourly seizure burden was.20%.23 A similar 4. Electroclinical Status Epilepticus (ECSE) cutoff was identified in neonates with hypoxic-ischemic ECSE is defined as an electroclinical seizure for $ 10 encephalopathy.24 continuous minutes or for a total duration of $ 20% of NOTE: “Possible electrographic seizure” and “possi- any 60-minute period of recording. An ongoing seizure ble electrographic SE”: These terms are synonyms for with bilateral tonic-clonic (BTC) motor activity only needs patterns on the ictal-interictal continuum (IIC); see section to be present for $ 5 continuous minutes to qualify as F below, page 25. For the sake of standardized reporting, ECSE. This is also referred to as “convulsive SE,” a subset the pattern should be described using the RPP modifiers of “SE with prominent motor activity.” 17 In any other (section C) and identified as meeting the criteria for the IIC. clinical situation, the minimum duration to qualify as SE For this reason, “possible ESz” and “possible ESE” have is $ 10 minutes. not been defined but can be used synonymously with IIC in 4b. Possible ECSE: Possible ECSE is an RPP that qualifies EEG impressions or when communicating with referring for the IIC that is present for $ 10 continuous minutes or for a clinicians. total duration of $ 20% of any 60-minute period of recording, which shows EEG improvement with a parenteral antiseizure medication BUT without clinical improvement. This remains 3. Electroclinical Seizure (ECSz) largely in line with “possible NCSE” as defined by the Salzburg ECSz is defined as any EEG pattern with either: criteria. NOTE: Possible ECSE cannot include patterns that already a. Definite clinical correlate* time-locked to the pattern (of any qualify as ESz/ESE. duration) (see Supp EEG 25, Supplemental Digital Content 1, NOTE: If parenteral antiseizure medication leads to http://links.lww.com/JCNP/A134) (Fig. 39), OR resolution of ESz/ESE AND clinical improvement, then b. EEG AND clinical improvement with a parenteral (typically these should be reported as ESz/ESE AND ECSz/ECSE IV) antiseizure medication (see Supp EEG 26a and b, (similar to how an isolated seizure can be both an ESz and Supplemental Digital Content 1, http://links.lww.com/JCNP/ an ECSz). A134) (Fig. 39). NOTE: For patients with prior known epileptic encephalop- NOTE: The EEG pattern during an “electroclinical athy, to qualify as ECSE, the EEG pattern needs to represent seizure” does not necessarily need to qualify as an “electro- either: graphic seizure.” For example, if static 1-Hz PDs have a a. an increase in prominence or frequency of epileptiform clinical correlate, this would not qualify as an ESz but would discharges compared with baseline, with an observable qualify as an electroclinical seizure (ECSz). Many seizures decline in clinical state, OR would however qualify for both “electrographic” and “elec- b. EEG and clinical improvement with a parenteral troclinical” seizures, and these should be reported under both (typically IV) antiseizure medication (Fig. 40). terms. NOTE: An electroclinical seizure (ECSz) can be of any NOTE: As a principle, all EEG data have to be put into duration, including ,10 seconds, if (and only if) there is a clinical (history, clinical presentation, physical examination) definite clinical correlate. By definition, electrographic and paraclinical (laboratory, toxicology, cerebral imaging) seizures must be $ 10 seconds duration. An evolving context to help establish or reject the diagnosis of status electrographic pattern lasting ,10 seconds qualifies as either epilepticus. an evolving RPP (e.g., “evolving RDA”) or “evolving NOTE: If any of these phenomena (ESz, ECSz, ESE, ECSE) BIRDs” (see section E, page 24). are stimulus-induced (reproducibly brought about or exacerbated *NOTE: A “definite clinical correlate” can be subtle, by an alerting stimulus), then they warrant an “SI-” prefix, as including face twitching, eye deviation, or nystagmus. Provided described in section C 3g above, page 14. that the clinical sign is clearly time-locked to the EEG pattern (and absent when the pattern is absent), then it should be considered an electroclinical seizure. NOTE: Any seizure or status epilepticus without prom- E. BRIEF POTENTIALLY ICTAL RHYTHMIC inent motor activity can also be referred to as “nonconvul- DISCHARGES (BIRDs) sive.” The term “nonconvulsive” is preferred over (Largely based on Yoo JY et al., JCN 201714) “subclinical” because it is usually unclear if the electrographic BIRDs are defined as focal (including L, BI, UI or Mf) or activity is contributing to the patient’s impaired mental status; generalized rhythmic activity.4 Hz (at least six waves at a if it were contributing, it would still be nonconvulsive but regular rate) lasting $ 0.5 to ,10 seconds, not consistent would not be subclinical. with a known normal pattern or benign variant, not part of NOTE: The term “nonconvulsive” can be applied to both burst-suppression or burst-attenuation, without definite clin- electrographic and electroclinical seizures. All ESz and ESE ical correlate, and that has at least one of the following alone (without clear clinical correlate) would be nonconvulsive. features: However, any ECSz or ECSE without prominent motor activity could also be termed nonconvulsive. a. Evolution (“evolving BIRDs,” a form of definite BIRDs) (Fig. 41A) 24 Journal of Clinical Neurophysiology Volume 38, Number 1, January 2021 clinicalneurophys.com Copyright © by the American Clinical Neurophysiology Society. Unauthorized reproduction of this article is prohibited.  Standardized Critical Care EEG Terminology L. J. Hirsch, et al. FIG. 41. Brief Potentially Ictal Rhythmic Discharges (BIRDs). A. BIRDs with evolution, aka "evolving BIRDs" (a form of definite BIRDs). B. BIRDs with a similar morphology and location as interictal epileptiform discharges in the same patient (definite BIRDs). C. BIRDs with a similar morphology and location as seizures in the same patient (definite BIRDs). D. BIRDs that are sharply contoured but without the above features (possible BIRDs). b. Similar morphology and location as interictal epileptiform chance that it may be contributing to impaired alertness, discharges or seizures in the same patient (definite BIRDs) (see causing other clinical symptoms, and/or contributing to Supp EEG 27, Supplemental Digital Content 1, http://links. neuronal injury. Thus, it is potentially ictal in at least some lww.com/JCNP/A134) (Figs. 41B and 41C) sense and often warrants a diagnostic treatment trial, typically with c. Sharply contoured but without (a) or (b) (possible BIRDs) a parenteral antiseizure medication. Although this is a concept under (Fig. 41D) development and with no broad consensus, the following patterns can be considered to be on the IIC: NOTE: Paroxysmal fast activity lasting $ 0.5 to ,10 seconds qualifies as BIRDs, whether generalized (also known as generalized a. Any PD or SW pattern that averages.1.0 and #2.5 Hz over paroxysmal fast activity, or GPFA) or focal. 10 seconds (.10 and # 25 discharges in 10 seconds) (see NOTE: Although they are termed “brief,” technically all Supp EEG 28, Supplemental Digital Content 1, http://links. BIRDs are “very brief” because they are ,10 seconds. lww.com/JCNP/A134) (Fig. 42A); or b. Any PD or SW pattern that averages $ 0.5 Hz and #1.0 Hz over 10 seconds ( $ 5 and #10 discharges in F. THE ICTAL-INTERICTAL CONTINUUM (IIC) 10 seconds) and has a plus modifier or fluctuation (see This term is synonymous with “possible ESz” or “possi- Supp EEG 29a–c and 30, Supplemental Digital Content ble electrographic SE.” The IIC is a purely electrographic term 1, http://links.lww.com/JCNP/A134) (Figs. 42B and that is not a diagnosis; it requires careful interpretation in 42C); or the full clinical context. A pattern on the IIC is a pattern that c. Any lateralized RDA averaging.1 Hz for at least 10 seconds (at does not qualify as an ESz or ESE, but there is a reasonable least 10 waves in 10 seconds) with a plus modifier or fluctuation clinicalneurophys.com Journal of Clinical Neurophysiology Volume 38, Number 1, January 2021 25 Copyright © by the American Clinical Neurophysiology Society. Unauthorized reproduction of this article is prohibited.  L. J. Hirsch, et al. Standardized Critical Care EEG Terminology (Figs. 42D and 42E). This includes any LRDA, BIRDA, UIRDA, and MfRDA, but not GRDA. AND d. Does not qualify as an ESz or ESE (Section D above, page 22). NOTE: If treatment of a pattern on the IIC with a parenteral antiseizure medication leads to improvement in the EEG AND definite clinical improvement, this would meet criterion B of an ECSz or ECSE. If treatment of an IIC pattern with a parenteral antiseizure medication leads to improvement in the EEG BUT NOT clinical improvement, this would be possible ECSE. NOTE: If the IIC pattern is stimulus-induced (reproducibly brought about or exacerbated by an alerting stimulus), then it warrants an “SI-” prefix, as described in section C 3g above, page 14. G. MINIMUM REPORTING REQUIREMENTS IN CLINICAL CARE The recommendations from the ACNS Consensus Statement on Continuous EEG in Critically Ill Adult and Children, Part II,25 are repeated here for convenience: 1. First 30 to 60 minutes (equivalent to a “standard” or “routine” EEG). This should be reviewed as soon as possible and reported to the clinical team. 2. Each 24-hour period. A written report should be completed at least once per day. If significant changes occur in the record during this period, then additional epochs should be reported separately as needed, either verbally or in writing. NOTE: We recommend communicating updates to the clinical team at least twice per day except in unusually stable circumstances. H. OTHER TERMS 1. Daily Pattern Burden is defined as the total duration of a pattern per 24 hours. For example, if GPDs were present for 33% of the record for 12 hours, then 10% of the record for 12 hours, the Daily GPD Burden would be 4 hours 1 1.2 hours ¼ 5.2 hours. FIG. 42. The Ictal-Interictal Continuum (IIC). Does not qualify as an electrographic seizure or electrographic status epilepticus but can be considered with any of the following features: A. Any PD or SW pattern that averages.1.0 Hz and #2.5 Hz over 10 seconds (.10 and #25 discharges in 10 seconds). B and C. Any PD or SW pattern that averages $ 0.5 Hz and #1.0 Hz over 10 seconds ( $ 5 and #10 discharges in 10 seconds), AND has a plus modifier or fluctuation. D and E. Any lateralized RDA (LRDA, BIRDA, UIRDA, MfRDA) averaging.1 Hz for $ 10 seconds (at least 10 waves in 10 seconds) with a plus modifier or fluctuation. 26 Journal of Clinical Neurophysiology Volume 38, Number 1, January 2021 clinicalneurophys.com Copyright © by the American Clinical Neurophysiology Society. Unauthorized reproduction of this article is prohibited.