The significance of sharp (potentially “epileptiform”) and rhythmic EEG features varies remarkably and ranges from the “irritative” (post-, inter-, or pre-ictal) to the “actively seizing”—along an “ictal-interictal continuum” [42, 43]. Periodic discharges include: periodic lateralized (epileptiform) discharges, traditionally labeled PLEDs, and more recently, LPDs; bilateral independent periodic lateralized (epileptiform) discharges (BiPLEDs, or BIPDs); and generalized periodic (epileptiform) discharges (GPEDs, or GPDs). For a comprehensive review of periodic discharges, see Chap. 5, “Periodic EEG Patterns.”

From the time of their initial report by Chatrian and colleagues [44], periodic lateralized epileptiform discharges have been controversial, but this only intensified as technology improved. They consist of spike, sharp, or polyspike components, with variable following slow wave complexes, usually recurring at 0.5–2 Hz, distributed broadly over most of one hemisphere. Between discharges, the background activity is usually attenuated and slow. Over time, the discharge frequency may decline, and most PLEDs resolve within weeks [44].

PLEDs, or LPDs, occur most commonly after acute large structural lesions such as strokes (the most common cause), tumor, or infection, but also in chronic seizure disorders and static lesions [45, 46]. Patients with PLEDs are often obtunded, with focal neurologic deficits and often, focal motor seizures; epilepsia partialis continua (EPC) is common. Clinical seizures occur in at least 80% of patients with PLEDs and electrographic seizures in even higher proportions; many have had prior SE [46]. Half of patients without prior epilepsy who survive the acute illness develop long-term epilepsy [47].

Most EEG-ers do not consider PLEDs to be a manifestation of ongoing seizures or SE, at least at the time of the recording [47, 48]; they are often considered “the terminal phase of status epilepticus” [46]. Nevertheless, they are highly associated with clinical seizures [46]. Patients with “PLEDs plus,” or LPDs+ (with lower voltage rhythmic epileptiform discharges, or other rhythmic patterns between the higher voltage periodic sharp waves) are more likely to have clinically evident epileptic seizures [49].

Periodic discharges are often considered “ictal” (the sign of an ongoing seizure) if they occur consistently with stereotyped clinical behavior that appears epileptic. More rapid discharges (at least 2 Hz) would also be interpreted as representing seizures (and if prolonged enough, SE) by most EEG-ers. PLEDs are also likely to be “ictal” if the EEG discharges resolve and clinical symptoms improve after ASD treatment. Some studies have found focal hyperperfusion on SPECT scans [50, 51] or evidence of focally increased metabolic activity on PET scans at the time of PLEDs [52], suggesting that PLEDs are the sign of ongoing seizures. Sometimes, PLEDs are definitely seizures, even with no structural lesions and EEG discharge intervals as long as 4 s [53], as shown by resolution of clinical deficits along with slowing of the discharges, spontaneously or in response to BZDs. PLEDs observed during EPC are certainly a manifestation of seizures [54].

Generalized periodic (epileptiform) discharges are continuous generalized spikes, polyspikes, sharp-and-slow waves, or sharp waves, often with a repetition rate >1 Hz, typically arising from a diffusely slow or low voltage background [55]. They are a common finding in C-EEG monitoring in the ICU and highly associated with an abnormal mental status, coma, convulsions, NCSE, and a poor outcome [56] (see Chap. 5). Many are seen following anoxia or other catastrophes, metabolic insults, recent overt seizures, or in the late stages of GCSE [57]. They are highly associated with clinical seizures but not necessarily indicative of ongoing seizures at the time [58, 59]. Some consider most GPDs to be seizures and recommend aggressive ASD therapy [57], while others believe that they are the sign of neuronal injury but not actual seizures and not requiring aggressive treatment; this may depend in large part on the frequency of the GPDs. One study did not find any features that could distinguish clearly between GPDs after anoxia and GPDs after SE [60], even though those conditions have markedly different implications for prognosis and treatment. GPDs tend to persist even with aggressive therapy, and it is not known whether patients benefit from ASD treatment for them.

Some periodic discharges appear to fluctuate and can lie on either side of the “borderline.” Focal or generalized periodic and quasi-periodic discharges can also be elicited in stuporous or comatose patients upon stimulation—“stimulus-induced rhythmic, periodic, or ictal discharges” or SIRPIDs, which usually abate after the stimulus recedes [61]. Hirsch has demonstrated persuasively that some SIRPIDs should be considered a sign of arousal and that others (with clear clinical manifestations) are definite seizures [62].

Another area in which C-EEG monitoring has played an increasingly large role in trying to determine whether a patient is having active seizures is that of neonatal seizures and NCSE—also discovered recently to be much more common than previously suspected [63]. During seizures and NCSE, neonates may exhibit minimal or no change from their clinical baselines [64]. NCSE can remain underdiagnosed if EEG monitoring is not utilized, especially in neonates with severe brain injury [65]. Also, electrographic patterns of neonatal seizures differ from those in older patients, often remaining localized to smaller areas. Some show evolution in frequency, amplitude, morphology, or spatial distribution, but many do not [65]. Many neonatal seizures last 2–3 min but recur frequently; prolonged continuous seizures are less frequent than in adults [66]. Given the clinical subtleties of neonatal seizures and NCSE, clinicians are increasingly dependent on C-EEG monitoring, but with many of the same uncertainties and problems encountered in critically ill adults with markedly abnormal EEGs. Often, it is difficult or impossible to determine if the electrographic abnormalities or epileptiform discharges cause the clinical deficits.

While the ACNS criteria have facilitated studies that have improved understanding of nonconvulsive seizures and NCSE, they do not always determine a clear diagnosis for an individual patient. The criteria have good specificity for nonconvulsive seizures and NCSE and offer greater certainty in their diagnoses. They are often used clinically but were designed primarily as guides for research purposes (thus the need for specificity), but their lack of sensitivity can be a problem, given the wide range of potential findings in patients with (subsequently) clinically confirmed NCSE [70, 71]. There are patients whose EEGs do not meet these criteria who are very likely having seizures or NCSE [71, 73], so clinical judgment remains crucial in diagnosis.

Sometimes, seizures and NCSE are readily apparent on the EEG, with characteristic or even “classic” features. Similar EEG findings, however, can be seen in metabolic and other encephalopathies, and many EEG patterns are neither pathognomonic nor easy to interpret.

Even determining focal versus generalized NCSE patterns can be difficult. During NCSE, the EEG may show epileptiform discharges generalized at the onset; discharges that begin focally, with or without secondary generalization; or both focal and generalized abnormalities [74]. Among pediatric ICU patients, focal or multifocal NCSE (65%) was more common than generalized SE (35%) [63].

Although EEG is the most reliable diagnostic test for nonconvulsive seizures and NCSE, it is not always conclusive. As noted in the ACNS recommendations, a response to treatment (an “ASD challenge”) can be helpful. Some patients with less “classic” EEG findings show a clinical and electrographic response to BZDs or other ASDs, and in the appropriate setting, a rapid response can be diagnostic of NCSE. Lorazepam or another ASD may be given in small sequential doses, while monitoring blood pressure, respiration, and oxygenation. To be diagnostic of ongoing seizure activity, there should be prompt resolution of epileptiform features on the EEG and clear improvement in the patient’s clinical state, or complete cessation of electrographic seizure activity with return of a normal EEG background [75]. Intravenous BZDs may abolish electrographic seizures, but they can also suppress nonepileptic EEG patterns such as TWs [31], so electrographic improvement alone does not prove that a particular EEG pattern was a seizure. The diagnosis must be made taking into account both the clinical context and the EEG findings.

Importantly, lack of prompt improvement with ASDs does not refute a diagnosis of NCSE. Frequently, the response to ASDs is very delayed or inconclusive [36, 75], even in patients with definite NCSE. A rapid clinical response is uncommon, especially in obtunded or comatose patients [76], and sedation from a BZD may also impair a clinical response. When the ASD challenge “works” a diagnosis is made; when not (probably more common) it does not disprove the presence of NCSE.

Probably, the most important “non-classic” NCSE clinically is the continuation of SE after unsuccessful (and often inadequate) treatment of generalized convulsions or GCSE, a relatively common finding on C-EEG monitoring. The EEG may show rapid epileptiform discharges following the apparent control of GCSE, without clinical signs of persistent seizures. In this subtle form, the epileptiform seizure discharges continue on the EEG and the patient remains unresponsive, even while the motor manifestations (convulsions) have ceased. In this setting, a wider range of potentially epileptiform EEG patterns should be considered as indicative of ongoing NCSE. Most epileptologists agree that when these EEG changes are marked, this NCSE should be considered a later stage of GCSE in terms of its pathophysiology and clinical implications and should be treated expeditiously.

ACNS criteria make a diagnosis of NCSE when there are “subtle” motor signs such as blinking or myoclonus. Logically, if patients with these EEG patterns following generalized seizures and GCSE are considered in NCSE when they have clear “subtle signs,” patients with the same EEG patterns in the same setting, even without clinical signs, should probably be diagnosed the same.

These patients with minimal or no motor signs have been said to be in “subtle generalized convulsive” SE [57]. The EEG may show discontinuous epileptiform activity with brief bursts of generalized spikes or GPDs [57, 77]. Some consider these EEG patterns to indicate ongoing SE only if the patient had previous clinical seizures or SE [78]. These patients have high morbidity and mortality [57, 68, 78, 79; see also Chap. 5]; and most agree that the urgency of treatment continues with the persistent electrographic seizures, especially when preceded by generalized convulsions—even while acknowledging that abolition of electrographic seizures does not always lead to clinical improvement.

There are also many other patients with ongoing electrographic seizure activity (whether following GCSE or not), with severe medical illnesses such as cerebrovascular disease, sepsis, or marked metabolic derangements. There is often substantial uncertainty in determining whether the EEG pattern represents seizure activity or not [43]. ACNS criteria help, but they are not always sensitive enough in actual clinical cases. In clinical practice, rhythmic, relatively rapid (>2 Hz) epileptiform discharges, typical for SE on the EEG, without obvious clinical manifestations, should usually be considered SE rather than simply an encephalopathy with epileptiform discharges—even when treatment does not effect a clinical improvement. The EEGs are very similar to those in published NCSE studies [74]; these patterns predict clinically evident seizures when seen on emergence from highly sedating therapy for refractory SE [80, 81]; and some patients with these EEGs respond well to ASDs [82]. How aggressively to treat, however, remains very controversial (see below).

This prolonged electrographic (nonconvulsive, often secondarily generalized, “non-classic”) seizure activity is very common in critically ill patients. In analyzing the correlations of these EEG patterns and corresponding neurologic deficits, many terms have been used. Some are referred to as in “subtle GCSE,” as above [57], and some in electrographic status epilepticus (ESE) [77]. Others refer to “nonconvulsive status epilepticus in coma,” but not all patients with ongoing electrographic seizures are comatose. Some label these patients with severe medical illnesses as having “epileptic encephalopathies,” indicating that the underlying disease causing the discharges is key, and that the epileptic component is not primary and may not respond to ASDs. This term, however, is probably best reserved for childhood conditions such as those associated with ESES (above).

In 2017, there is a much greater appreciation than before of the subtleties and complexities of NCSE, but it can still be difficult to formulate a precise definition, in part because its clinical manifestations are protean and its pathophysiologic implications uncertain. Also, EEG findings, while crucial for diagnosis, are often ambiguous or controversial, and the response to ASDs may be subtle.

What is a seizure? The most recent ILAE definition of an epileptic seizure is “a transient occurrence of signs and/or symptoms due to an abnormal excessive or synchronous neuronal activity in the brain [83].” More briefly, this could be a transient “neurologic clinical deficit due to abnormal, usually hypersynchronous, electrical brain activity.”

What then is nonconvulsive status epilepticus? Gastaut stated that SE was a continuation of one of the many different types of epileptic seizures [84], but it has become clear that SE is, to a large extent, a different pathophysiologic process than is seen in individual seizures—with “a failure of …. seizure termination or …. initiation of mechanisms [leading] to abnormally prolonged seizures” [85], and with manifestations different from those of individual seizures.

A definition of NCSE should include three elements: that there is a clinical abnormality in neurologic function (other than convulsions); that there is persuasive evidence that the dysfunction is due to an epileptic process, i.e., seizure activity; and that it is abnormally prolonged. Briefly, NCSE is “a state of prolonged seizure activity with a resultant change in behavior or level of consciousness [22].”

The ILAE 2015 Task force definition stated: “Status epilepticus is a condition resulting either from the failure of the mechanisms responsible for seizure termination or from the initiation of mechanisms, which lead to abnormally prolonged seizures (after time point t1). It is a condition, that can have long-term consequences (after time point t2), including neuronal death, neuronal injury, and alteration of neuronal networks, depending on the type and duration of seizures [85].” This definition provides pathophysiologic insight and illustrates starkly the difference between individual seizures and SE. It also provides a clear impetus for prompt treatment, though it does not specify the exact nature of NCSE, or what it is.

The clinical deficit typically encompasses an (epileptic) impairment of mental status, cognition, or behavior. During the proliferation of reports of different forms of NCSE a few decades ago, clinical manifestations were recognized as broader and more protean, making it difficult to say exactly what clinical changes should be expected. Beyond a change in alertness or behavior, there may also be a change in sensory perception (e.g., auditory, visual, somatosensory, or psychic) or aphasia, dysmnesia, neglect, alexia, etc. [86]; see also Chap. 21.

Because NCSE is considered an epileptic process without convulsions (or no more than minimal motor manifestations), but with pleomorphic clinical abnormalities, almost all diagnostic criteria insist on EEG evidence of continuing or very frequent epileptiform activity, corresponding closely, or somewhat, to the ACNS criteria. There are also other diagnostic tests to show that a patient is in NCSE, including some imaging techniques (almost always far less convenient and less commonly used) [87, 88] and the “ASD challenge” (described above)—with its own limitations.

Still, controversy persists regarding which EEG patterns are diagnostic of, or consistent with, NCSE, and there are no absolute (or at least, widely agreed upon) criteria for the diagnosis of ongoing seizures by EEG. The determination of whether a particular EEG pattern represents an encephalopathy or NCSE (or both or neither) relies on the determination that certain patterns are “ictal” in nature, i.e., signifying ongoing epileptic seizures. Some EEG findings are persuasive on their own, but in other cases identical EEG findings may indicate ongoing seizures in one patient and not in another—or at least not seizures that a wise neurologist would treat.

In clinical practice, neurologists cannot be restricted to diagnosing NCSE by ACNS criteria alone, helpful as they are, especially in clinical studies. It is not sufficient to report to the requesting physician that the EEG “lies somewhere on the continuum” (the ictal-interictal continuum) or that ASDs are unwarranted because any putative electrographic seizure activity “did not meet ACNS criteria”—especially in the “special case” of possible NCSE after insufficient treatment of GCSE. For a clinical decision on treatment or management, the likelihood of NCSE must be determined by an experienced electroencephalographer who also considers the clinical setting.

One can also question whether it matters if epileptic seizure activity is the (only) cause of the neurologic deficit. In anoxia, for example, the extent of neuronal damage is often great, and that determines the ultimate, usually disastrous, outcome [89]. Similarly, some severe metabolic derangements might suffice to explain a patient’s impairment, but the seizure activity evident on the EEG might do the same in the absence of the primary illness; either process might cause the deficit. Some would label the findings “NCSE” when it is considered that a seizure with the same electrographic appearance likely would cause a clinical deficit if the lesion or illness were not present. Others would not.

Even after diagnosis of NCSE, the same problem persists. It can be difficult or impossible to determine whether the epileptiform waveforms on the EEG considered electrographic seizure activity contribute to or cause the clinical dysfunction.

Many definitions of NCSE include a temporal criterion for the duration of seizure activity, or non-recovery from serial seizures. Long ago, Gastaut defined “fixed and lasting” as 30 min [84], but he was focusing heavily on GCSE. Traditionally, for the “classic” forms of NCSE, epileptologists tended to require EEG evidence of continuous or recurrent seizure activity, along with impaired consciousness, for 30–60 min [19, 90]. For decades, most clinical studies held it useful “in the field” to consider NCSE as seizures persisting for 30 min [20].

Decades after Gastaut’s early definitions, Lowenstein, Bleck and Macdonald [91] offered the first modern temporal criterion of 5 min for GCSE, establishing this as an “operational” definition—to be used largely in helping neurologists make diagnoses and determine treatment and management plans expeditiously. Their proposal was limited to apply to GCSE alone. While most neurologists have adopted this “operational” definition of GCSE, the same 5 min criterion has been used widely (if without demonstrated validity) for all types of SE in some recent publications.

The 2015 ILAE definition built on the same “operational” concept, and accepted the 5 min criterion for GCSE. It was noted, however, that different types of nonconvulsive SE warranted different temporal criteria: “t1” (the time beyond which seizures are unlikely to stop on their own, i.e., become SE, would be 10 min for both absence SE and for CPSE or “focal SE with impaired consciousness” [85]. For NCSE, “t2” (the time after which long-term neurologic injury would be an urgent concern and a reason for more aggressive treatment) would be 60 min for focal SE with impaired consciousness, but “completely unknown” for absence SE. For the (far more common, but quite heterogeneous) “non–classic” NCSE, or “NCSE in coma,” no t1 or t2 was proposed.

Of course, given the burgeoning multiplicity of types of NCSE, the temporal criteria cannot simply be constrained to be exactly 5 or exactly 30 min. As noted by Shorvon, NCSE is a “range of conditions …… dependent largely on the level of cerebral development and integrity, the presence or absence of encephalopathy, the type of epilepsy syndrome, and the anatomical location of the epileptic activity” [22]. Thus, “abnormally prolonged” focal-onset NCSE with dyscognitive features is substantially different in a young person with a genetic epilepsy and in an older patient with multiple medical problems and NCSE due to a stroke and sepsis, though they may appear similar clinically. Proper temporal criteria for the diagnosis of NCSE would be different in these two cases (and in dozens of others!), as might treatment.

In 2017 definitions of NCSE still tend to be imprecise as to its nature, but increasingly helpful “operationally.” Neurologists are much more aware of its many manifestations and thus maintain a better “index of suspicion” that NCSE may be occurring and how to go about evaluating it, probably improving its recognition and diagnosis, and expediting and improving its treatment.