Status epilepticus is the most common neurological emergency in childhood and continues to be associated with significant mortality and morbidity. Although the main determinant of adverse outcome following status epilepticus is etiology, there continues to be concern that status epilepticus itself contributes to the outcome. Therefore, an understanding of the types of status epilepticus, the mechanisms, and frequencies of adverse outcomes and appropriate treatments is essential for defining strategies that aim to reduce morbidity associated with status epilepticus.

Status epilepticus is traditionally divided into convulsive and nonconvulsive forms. The aim of the current chapter is to provide an overview of nonconvulsive status epilepticus (NCSE) in terms of definition, clinical features, outcomes, treatments, and its relationship with epileptic encephalopathy. However, many of these aspects are not universally agreed and therefore there will also be discussion of the controversies surrounding NCSE.

In April 2004, a group of physicians with an interest in NCSE, representing a spectrum of opinion, met in Oxford¹ to discuss and debate the definition, diagnosis, and treatment of NCSE. There was no consensus on a specific definition and therefore the following broad definition was made; NCSE is a range of conditions in which electrographic seizure activity is prolonged and results in nonconvulsive clinical symptoms.

The International League Against Epilepsy suggest that status epilepticus should be defined as a seizure that shows no clinical signs of arresting after a duration encompassing the great majority of seizures of that type in most patients or recurrent seizures without interictal resumption of baseline central nervous system function. However, these definitions are not particularly useful in clinical practice as there is too much breadth; that is, what is prolonged?, and the definitions about clinical features are vague.

A time definition that is frequently used for NCSE is 30 minutes, which is the same as that commonly used to define convulsive status epilepticus. This construct appears to work for convulsive status epilepticus; after 30 minutes the seizure becomes self-sustaining, is more likely to cause brain injury and is associated with systemic decompensation, that is, hypotension, hypoglycemia, etc. However, the construct is less applicable to NCSE as 30 minutes may not describe a point following which there are potential pathophysiological changes that separate NCSE from short seizures with similar clinical manifestations.

The proposed definition from the Oxford meeting uses the term nonconvulsive clinical symptoms, which potentially means that any clinical feature (e.g., dizziness and headache) could be considered as a feature of NCSE if there were consistent electroencephalogram (EEG) features. The agreed component of the definition is that the features need to be associated with EEG abnormalities, but if any neurological symptom can be a manifestation of NCSE then many patients with neurological symptoms will require an EEG to rule out NCSE. It is likely that this is neither practical nor necessary.

Over the years, various authors have suggested definitions for the term NCSE and are inclusive of clinical changes such as impaired consciousness, and associated ictal electroencephalographic abnormalities, and response to treatment.^2,3,4,5,6 Most authors agree that alterations in the clinical state and associated plausible electroencephalographic changes should be the basis of the definition.⁷ Clinical changes alone are not sufficient because these may be very subtle and sometimes hard to differentiate from normal behavior or nonepileptic medical disorders.⁸ Therefore, at the current time, there is no universally accepted definition of NCSE, but the most consistent aspect of any proposed definition is the presence of electrical discharges continuing for a “prolonged” period of time.

NCSE in the adult population generally falls into two groups. The first consists of patients with generalized spike-wave discharge, often in the context of primary generalized epilepsy, known as absence status epilepticus. The second includes patients with focal epileptic discharges known as complex partial status epilepticus. It is widely accepted that these two phenomena have differing clinical manifestations and outcomes, raising the issue as to whether NCSE can reasonably be considered a single entity. This issue is more difficult in children as the types of NCSE described in adults are less common in children, but frequent epileptic discharges that amount to NCSE in electrical terms are seen in other situations. Children with hypsarrhythmia,⁹ Panayiotopoulos syndrome,^10,11 electrical status epilepticus in slow wave sleep (ESES),¹² benign Rolandic epilepsy, Landau–Kleffner syndrome (LKS), and Lennox–Gastaut syndrome may all meet electrical criteria but do not constitute a single pathophysiological entity.

There are many shared clinical features in absence and complex partial status epilepticus but the frequency of individual clinical features differ in either.

ABSENCE STATUS EPILEPTICUS

Patients with short absence seizures characteristically have complete loss of consciousness without recall of the event. However, in absence status epilepticus there is a spectrum in the degree of alteration of consciousness ranging from slight clouding, manifesting as difficulties in carrying out particular activities that are usually not problematic for the patient, to an epileptic stupor in which the patient is completely unresponsive. However, some degree of clouding of consciousness is almost universal in absence status epilepticus. Patients may also manifest motor phenomena¹³ (myoclonus, atonia, blinking of eyelids, and pseudoataxia) and rarely psychiatric features including psychosis. Episodes may last hours or days and frequently terminate following a tonic–clonic convulsion. The majority of patients with absence status epilepticus have an existing diagnosis of absence epilepsy.⁵

There is a spectrum of electrographic abnormalities in patients with absence status epilepticus. The classical diagnostic pattern consists of continuous or near-continuous bilaterally synchronous and symmetrical spike-wave activity. The frequency of the discharges can be the typical 3 Hz spike and wave of a classical absence seizure,¹⁴ or an atypical pattern with frequencies of 1.5–4 Hz.¹⁴ In many patients, the EEG pattern is inconstant over time and the discharge frequency fluctuates.

COMPLEX PARTIAL STATUS EPILEPTICUS

The manifestations of complex partial status epilepticus are heterogeneous and typically consist of more than the patient’s habitual focal seizure. Similar to absence status epilepticus, the seizures typically last for hours and occur most frequently in patients carrying a diagnosis of focal epilepsy.

Confusion is the cardinal manifestation of complex partial status epilepticus^{15,16,17,18,19} and is more likely to be nonfluctuating than absence status epilepticus. Motor manifestations¹⁵ are more obvious than in absence status epilepticus and include alterations in posture, unsteadiness, convulsive movements, and tonic spasms. There is a tendency to adversion and the patients may walk in circles; complex automatisms are more often seen in shorter seizures.¹⁵ The patients may have behavioral changes including agitation, excitement, or psychomotor retardation and may have psychotic features. Speech patterns may be markedly abnormal and patients may become mute. In addition, there may be autonomic manifestations including pupillary dilatation, vomiting, borborygmi, and change in color.

Complex partial status epilepticus has very varied electroencephalographic manifestations.^{2,14,16,19,20,21} Some patients have very few epileptic discharges and the EEG may not differ from the interictal state. Other ictal patterns include frequent (almost continuous) spike, spike-and-slow wave discharges, or desynchronizations that can be focal or more widespread. The most common anatomical location for focal discharges is the temporal lobe, although frontal lobe status epilepticus is not uncommon. Frontal lobe status epilepticus²² may be associated with spike-wave discharges similar to absence status epilepticus.

Although these two types seem distinct there may be significant overlap in the clinical features^{5,16,17,18,23} of absence^23,24 and complex partial types of NCSE. Fear, anxiety, irritability, and aggression are more frequent and also seen are automatisms, eye deviation, and nystagmus in those with complex partial NCSE. With both, complex partial seizures and absence NCSE total unresponsiveness, speech arrest, cyclic behavior, and stereotypic nonlateralized automatisms are common. Thus, these subtypes of NCSE may be difficult to distinguish on the basis of clinical features alone. As a rule, complex partial NCSE is more common in clinical practice than the absence type of NCSE.^17,25

MANIFESTATIONS OF NCSE IN PEDIATRIC PRACTICE

Absence and complex partial status epilepticus in their purest forms are rare in childhood and NCSE in children is more typical of defined epilepsy syndromes. A variety of ictal manifestations that differ from adults have been described and include alterations in cognitive and behavioral status. Long-lasting epileptic EEG abnormalities may be observed in benign and severe epilepsy syndromes. Benign syndromes include Panayiotopoulos syndrome and Benign–Rolandic epilepsy. The predominant clinical feature in Panayiotopoulos syndrome is ictal vomiting although other autonomic manifestations¹¹ such as color change, pupillary abnormalities, cardiorespiratory, and thermoregulatory alterations are also common. When these manifestations last at least 30 minutes they may be considered a form of NCSE.^11,26,27

As with the adult forms of NCSE, treatment of the NCSE results in improvement in the EEG and a return of the child to their preictal state that in Panayiotopoulos syndrome is usually a return to normal. Children with benign Rolandic epilepsy¹ may have frequent Rolandic discharges in the EEG in sleep but there is seldom evidence for a clinical manifestation that could be considered to be status epilepticus, that is, the ictal clinical manifestations are usually short.

In a setting of acute change in behavior from a baseline as seen in developmentally normal children, NCSE may be relatively easy to diagnose because of the clear contrast between the two states and in presence of a clearly abnormal EEG. However, in children with severe epilepsy associated with learning and behavioral difficulties the diagnosis of NCSE is extremely difficult, and this is probably the most likely setting for NCSE in children.^28,29

Severe syndromes in which there are epileptic discharges that may persist for at least 30 minutes include Ohtahara syndrome, West syndrome, Dravet syndrome, LKS, Lennox–Gastaut syndrome, and ESES; see Figure 39–1. The conditions are frequently referred to as epileptic encephalopathies although the terminology “encephalopathy with epilepsy” may be preferable, as the latter term does not imply that the observed encephalopathy is caused by epileptic discharges. The relationships between frequent epileptic discharges in the EEG and the clinical manifestations in the patients with these syndromes are not entirely clear. In these syndromes, there is often a complex interaction between the effects of the underlying etiology, the seizures, and the electrical abnormality such that it is not possible to discern individual effects.²⁹

In West syndrome⁹ there may be ongoing epileptic discharges (hypsarrhythmia) associated with functional deficits such as loss of normal visual behavior which could be considered as a clinical manifestation of NCSE. The infantile spasms are each short but may occur in clusters that last 30 minutes and it is uncertain whether this constitutes a separate manifestation of NCSE or is a form of convulsive status epilepticus.

In LKS the main clinical feature is loss of language although manifestations in other domains of functioning (motor, behavior, and learning) are common. However, in LKS, it is not clear whether these are ictal manifestations or a function of an associated encephalopathy. Successful treatment of the epileptic discharges may result in some recovery but it is not usually complete and occasionally there is an apparent response to treatment with little change in the EEG abnormality. In addition, there is no obvious relationship between the severity of the EEG abnormality and the clinical features. Similarly, in ESES, the EEG may be normal in the waking state, but it is in this state that the predominant clinical features of behavior and learning impairments are most common. It is in the sleep state that the EEG is consistent with NCSE.

Understanding the relationship between EEG abnormalities and ictal clinical manifestations is even more complex in children with Lennox–Gastaut syndrome and other encephalopathies with epilepsy. This is more glaring when we have a “boundary syndrome” in the classification scheme as suggested by Shorvon.¹ The term boundary syndrome includes cases of epilepsy with encephalopathy in which it is unclear as to the extent of contribution of the electrographic activity to the observed clinical impairment. This may induce a subjective bias to the diagnosis of NCSE in such a clinical setting.⁹

The EEG abnormality that may look like atypical absence status epilepticus may persist for many years. Clinical correlates include learning and behavioral impairments. In these children, it is often difficult to assess the baseline behavior because of the associated impairments and behavior may continue to worsen at a steady-state unrelated or independent of the EEG, so when it comes to defining change from baseline, it is a challenge to ascribe such changes to NCSE.²⁹

Children with Lennox–Gastaut syndrome have periods when they lose further function and become unsteady, less responsive, and drool. They are sluggish, uncooperative, and often drowsy, walk with an unsteady gait and are known to have frequent falls.¹ These features are said to be descriptive of NCSE in children with learning and behavioral difficulties.^1,16,28,29 Periods of worsening may or may not be associated with a worsening of the EEG (see Figs. 39–2 and 39–3), but treatment as if this worsening was an episode NCSE may return the child to their previous state. If the background loss of function is considered to be NCSE then it is unclear what the period of worsening is other than “worse” NCSE. Thus, in the setting of encephalopathy with epilepsy dissecting NCSE from seizures/nonepileptic events may be difficult.