Clinical features and complications of status epilepticus in children

INTRODUCTION — The most common medical neurologic emergency in childhood, status epilepticus (SE) is a serious and often life-threatening medical emergency.

The definition, pathophysiology, classification, risk factors, and outcome of SE in children are reviewed here. Treatment of this disorder is discussed separately. (See "Management of convulsive status epilepticus in children".)

DEFINITION — The duration of continuous seizure activity used to define SE has varied over time. Historically, the International League Against Epilepsy (ILAE) and others defined SE as a single epileptic seizure of >30 minutes duration or a series of epileptic seizures during which function is not regained between ictal events in a 30-minute period [1]. For the purposes of treatment decisions, however, a shorter time window (eg, >5 to 10 minutes of continuous seizures) has been favored and generally accepted in the clinical community, particularly for generalized convulsive seizures.

The ILAE revised its definition of SE in 2015, and the revised definition incorporates both of these time points [2]. Specifically, SE is defined by the ILAE as:

●A condition resulting either from the failure of the mechanisms responsible for seizure termination or from the initiation of mechanisms that lead to abnormally prolonged seizures (after time point t1); and

●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

For generalized convulsive SE, the ILAE definition stipulates that t1 and t2 are 5 and 30 minutes, respectively. The five-minute window corresponds with the time at which urgent treatment should be initiated. At least one study has found that a convulsive seizure lasting more than 5 minutes has a high risk of lasting 30 minutes or more [3]. Also, treatment delay is associated with delayed treatment response [4,5].

For other types of SE, the most appropriate time intervals for t1 and t2 have not been well defined, particularly for nonconvulsive forms of SE. The ILAE suggests using a t1 and t2 of 10 and >60 minutes for focal SE with impaired consciousness, and a t1 of 10 to 15 minutes for absence SE [2].

Electrical (subclinical) seizures — Seizures recognized only on the electroencephalogram (EEG) are referred to as subclinical seizures and can continue after abnormal clinical activity has been stopped by antiseizure medication [6]. Some classifications have considered these seizures to be sustained seizures in comatose patients, whether or not the ictal discharges seen on EEG were accompanied by subtle convulsive activity, such as tonic eye deviation or rhythmic twitching of part of an extremity [7].

Controversy exists about the types of EEG patterns that should be considered ictal in patients without clinical manifestations. Some experts argue that periodic epileptiform activity, a common finding in severe hypoxic-ischemic encephalopathy, should be considered ictal and treated [7,8]. However, most epileptologists consider this activity to be interictal and would not increase anticonvulsant therapy solely on the basis of this type of discharge [9]. By contrast, actual electrographic seizures, which are familiar to the experienced electroencephalographer, should be treated in the absence of concomitant clinical behavior.

PATHOPHYSIOLOGY — SE occurs because of failure of the normal mechanisms that limit the spread and recurrence of isolated seizures [10,11]. Failure occurs because excitation is excessive and/or inhibition is ineffective. Multiple mechanisms probably are involved.

Glutamate is the major amino acid excitatory neurotransmitter in the brain. Its role in the pathogenesis of SE was suggested by an outbreak of illness caused by eating mussels contaminated with domoic acid, an analogue of glutamate [12]. Some affected individuals had prolonged seizures thought to be caused by excessive activation of excitatory amino acid receptors. Other excitatory neurotransmitters that contribute to SE include aspartate and acetylcholine [11].

Gamma-aminobutyric acid (GABA) is the main inhibitory neurotransmitter in the brain, and antagonists to its effects or alterations in its metabolism in the substantia nigra may contribute to SE [13]. In a rat model, for example, the rate of GABA synthesis in the substantia nigra declined significantly during induced SE [14]. Other inhibitory mechanisms include the calcium ion-dependent potassium ion current and blockage of N-methyl-D-aspartate (NMDA) channels by magnesium.

Neuronal loss — Though the outcome of SE generally is favorable in the absence of an underlying neurologic condition, minor amounts of neuronal loss are thought to occur with every episode, particularly if prolonged (see 'Outcome' below). Over time, this loss may accumulate and lead to significant impairment.

Disturbance of the NMDA channels appears to be an important mechanism of neuronal injury in SE [11]. When neurons are depolarized, calcium enters the cell through NMDA channels and causes injury or death. Other possible contributing factors include hypoxia; excessive release of excitatory amino acids and calcium; increases of various proteins, including those that promote apoptosis (programmed cell death); alterations in receptor populations; and, in the temporal lobe, sprouting of dentate granule cells [15,16]. Abnormalities on magnetic resonance imaging (MRI) and release of the neuron-specific enolase (NSE) are markers of neuronal damage.

MRI — Focal radiologic changes are not uncommon occurrences during and after focal SE [17-19]. Some of these abnormalities are related to the underlying cause of the patient's seizures and include focal cortical dysplasias and areas of gliosis.

Other MRI changes appear to be a result of the acute seizure activity, and include mild edema, decreased attenuation with effacement of sulci, and loss of gray-white differentiation. These can resemble changes seen in acute ischemic stroke; however, the abnormalities do not conform to vascular territories, but to the electroencephalogram (EEG) localization of the seizures. In one study of two adults and a four-year-old boy, the MRI findings were consistent with those of cytotoxic and vasogenic edema, hyperperfusion of the epileptogenic region, and alteration of the leptomeningeal blood-brain barrier, the latter manifest by leptomeningeal enhancement on the postcontrast MRI [18]. The acute lesions in this study were followed by regional brain atrophy.

Neuron-specific enolase — Enolase is an enzyme that is part of the glycolytic pathway for the conversion of glucose to pyruvate. It exists in three dimeric forms: alpha, beta, and gamma. The gamma isoform is exclusive to neurons and is termed "neuron-specific enolase" (NSE). This enzyme converts 2-phosphoglycerate to phosphoenolpyruvate. It is released into the cerebrospinal fluid (CSF) and blood after the occurrence of stroke and anoxia.

NSE correlates with the extent and duration of ischemia in animal models of stroke and the outcome in humans, including neonates, with hypoxic-ischemic encephalopathies [20]. Increases after a single seizure have been reported by some researchers, but it is more predictably increased in serum and CSF after SE, particularly complex partial SE [21-23].

Absence status is considered a relatively benign condition. However, serum NSE may be increased in children with prolonged absence status [23], which may be followed by permanent cognitive impairment in selected cases [24].

CLASSIFICATION — The usual classification of SE is similar to that used for individual seizures and includes four major types:

●Focal SE without impairment of consciousness or awareness (simple partial SE) – Continuous or repeated focal motor or sensory seizures without impaired consciousness.

●Focal SE with impairment of consciousness or awareness (complex partial SE) – Continuous or repeated episodes of focal motor, sensory, or cognitive symptoms with impaired consciousness. This is also called nonconvulsive SE. In some patients the clinical manifestations of seizure activity are subtle and not apparent to the clinician.

●Generalized convulsive SE including tonic-clonic, tonic, and clonic – Always associated with loss of consciousness.

●Absence SE – Generalized seizure activity, characterized clinically by altered awareness, but not necessarily unconsciousness.

Numerous less common but important forms, including myoclonic and pseudoseizures or psychogenic seizures, also occur. (See "Management of convulsive status epilepticus in children".)

●Nonepileptic seizures – These seizures (also known as psychogenic seizures) usually are excluded from studies of SE but are a common problem, particularly in teenagers [25]. These individuals often present to emergency departments because of recurrent episodes of abnormal behavior or prolonged seizures, which are unresponsive to standard antiseizure medications. The seizures typically occur in patients with affective and anxiety disorders and a family history of seizures. The clinical characteristics of pseudoseizures and their differentiation from true epileptic seizures are discussed separately. (See "Nonepileptic paroxysmal disorders in children".)

In addition to the clinical classification, SE in children can be classified broadly into categories based on etiology [2]:

●Symptomatic (acute, remote, or progressive) – Acute symptomatic causes include infection, hypoxia, glucose and electrolyte imbalance, trauma, and hemorrhage or stroke. "Remote symptomatic" refers to seizures caused by an insult earlier in life, such as perinatal hypoxic-ischemic injury, trauma, infection, or a congenital brain malformation. "Progressive" refers to etiologies such as brain tumors or progressive myoclonic epilepsy.

●Unknown or cryptogenic – No known or identifiable cause.

One study classified 394 cases of SE in children aged 1 month to 16 years and found important age-related differences [26]. SE was classified as reactive (due to fever), acute symptomatic, remote symptomatic, cryptogenic or idiopathic, and progressive encephalopathy in 29, 28, 24, 15, and 5 percent, respectively. SE occurred more commonly in young children, with more than 40 percent occurring in those younger than two years of age. More than 80 percent of the younger children had SE of febrile or acute symptomatic etiology, whereas older children were more likely to be in the cryptogenic or remote symptomatic categories. Older children also were more likely to be neurologically abnormal (55 versus 21 percent) and to have a history of seizures (64 versus 20 percent) compared with those younger than two years of age. Similar trends are reported in another population-based case series [27].

EPIDEMIOLOGY AND ETIOLOGY — The estimated incidence of childhood SE is between 17 to 23 episodes per 100,000 per year [27,28]. SE can be a complication of acute illness such as encephalitis, or can occur as a manifestation of epilepsy. Incidence rates, causes, and prognosis vary substantially by age. The highest incidence is in the first year of life. Febrile SE is the most common etiology [29]. Approximately 60 percent of children are neurologically healthy prior to the first episode of SE.

Risk factors — Risk factors for SE have been best defined in the setting of established epilepsy. Between 10 and 20 percent of children with epilepsy will have at least one episode of SE [30]. SE occurs as the first seizure in 12 percent of children with epilepsy. Patients with partial seizures that tend to occur in clusters (three or more within 24 hours, with return to baseline between seizures) have a higher incidence of SE compared with those who do not cluster (47 versus 13 percent, in one study) [31]. Other risk factors for SE in children with symptomatic epilepsy include [32]:

●Focal background electroencephalography (EEG) abnormalities

●Focal seizures with secondary generalization

●Occurrence of SE as the first seizure

●Generalized abnormalities on neuroimaging

In a prospective, community-based cohort study of 613 children with newly diagnosed epilepsy, 56 (9 percent) had one or more episodes of SE by the time the diagnosis of epilepsy was established [33]. Similarly, SE was the first presentation of epilepsy in 10 percent of 1382 children in a database of children with new-onset seizures [29]. After a median follow-up of eight years, 18 of the 56 (32 percent) children in the first cohort had at least one recurrence of SE [34]. A first episode of SE occurred in an additional 40 patients. Independent risk factors for SE were [27,34]:

●History of prior SE

●Young age (one year or less) at onset

●Symptomatic etiology of epilepsy

Genetic factors — Studies in rodent models have shown that genetic influences are important in the susceptibility to SE. As an example, both latencies to kainate-induced seizures and SE and the extent of excitotoxic damage to the hippocampus after recurrent seizures vary among different strains of rodents because of polygenic influences [35,36].

Human twin studies also have shown a genetic predisposition for SE. One report examined 39 pairs of twins with a history of seizures in at least one twin [37]. Seizure and SE concordance rates were significantly higher in the 13 pairs of monozygotic twins compared with the 26 pairs of dizygotic twins (0.38 versus 0.00). The concordance rate for SE was highest in the monozygotic twins who also were concordant for seizures (0.58). The nature of the genetic factor(s) is yet to be elucidated.

Some genetic syndromes (eg, Dravet syndrome, generalized epilepsy with febrile seizures plus [GEFS+], Angelman syndrome) tend to present with recurrent episodes of SE.

Causes — SE may occur in the setting of underlying, premorbid epilepsy or as the first manifestation of epilepsy [29]. Virtually any cause of epilepsy may have a first presentation with SE.

SE may also be an acute symptom of medical or neurologic disease. The more common examples of the latter include [29,38]:

●Central nervous system infections

●Acute hypoxic-ischemic insult

●Metabolic disease (eg, hypoglycemia, inborn error of metabolism)

●Electrolyte imbalance

●Traumatic brain injury

●Drugs, intoxication, poisoning

●Cerebrovascular event

FIRES and NORSE — Febrile infection-related epilepsy syndrome (FIRES) refers to a presentation with new-onset refractory SE in the setting of a prodromal febrile illness starting between two weeks and 24 hours prior to onset of refractory SE, with or without fever at onset of SE [39-42].

FIRES is considered a subcategory of new-onset refractory status epilepticus (NORSE), which is characterized by the absence of a clear acute or active structural, toxic or metabolic cause for SE in a patient without active epilepsy or another preexisting relevant neurologic disorder [42]. NORSE has been described in adults as well as children. (See "Refractory status epilepticus in adults", section on 'New-onset refractory status epilepticus'.)

The mechanism of NORSE is unclear, and the infectious agent is often not identified. The evaluation for suspected NORSE is outlined in the table (table 1). Some cases may represent an autoimmune or paraneoplastic encephalomyelitis due to autoantibodies against synaptic proteins such as the N-methyl-D-aspartate (NMDA) receptor, in which case immunomodulatory therapies may be effective [43]. (See "Autoimmune (including paraneoplastic) encephalitis: Clinical features and diagnosis".)

Seizures in patients with NORSE tend to be focal or secondarily generalized and are resistant to therapy, although some patients have responded to the ketogenic diet [44]. Immunotherapies (immune globulin, corticosteroids, plasma exchange, rituximab) are generally not helpful [45]. Mortality and morbidity (impaired cognition, refractory epilepsy) are high. In a study of 279 prospectively enrolled children with refractory SE, criteria for NORSE were met by 46 (16 percent) [46]. The mean age of children with NORSE was 2.4 years, and an extensive evaluation for etiology was unrevealing in 87 percent. Patients presenting with fever at onset (n = 19) were younger, had shorter episodes of SE, and a high rate of recovery (89 percent) to baseline function. Patients with previous fever (n = 16) had more prolonged SE and worse outcomes, with a low rate of recovery (25 percent) to baseline function and a 19 percent mortality.

In another report, seven patients who survived with chronic epilepsy had very similar seizures that were characterized by stereotypical perirolandic and perisylvian features (head and eye deviation, unilateral face and limb jerking, and asymmetric tonic posturing) [45].

SYSTEMIC COMPLICATIONS — Systemic changes frequently accompany prolonged seizures [11]. These complications contribute to the morbidity of SE and can be life threatening.

Hypoxemia — Hypoxemia is thought to account for some of the complications of SE. It can result from impaired ventilation, increased oxygen consumption, or excessive salivation or tracheobronchial secretions [11]. Seizures associated with hypoxemia lead to further metabolic disturbances, including reduced brain glucose levels, lactic acidosis, and depletion of brain adenosine triphosphate (ATP). Severe hypoxemia and acidosis can result in impaired myocardial function, reduced cardiac output, and hypotension, further disrupting cellular function.

Acidemia — Lactic acidosis and respiratory acidosis frequently accompany SE, resulting in one study in pH less than 7.0 in 13 percent of cases [47].

Glucose alterations — Blood glucose concentration typically is elevated at the start of seizures because of catecholamine release and sympathetic discharge. However, prolonged seizures often result in hypoglycemia as the metabolic demands outstrip the supply.

Blood pressure disturbances — Blood pressure, heart rate, and central venous pressure rise at the start of SE as a result of massive release of catecholamine and sympathetic discharge. This rise is accompanied by a large increase in cerebral blood flow (200 to 700 percent in primates) thought to compensate for the brain's increased metabolic needs [48]. However, the blood pressure declines as the seizure persists, often resulting in hypotension. Cerebral blood flow also falls. Although it remains above normal levels, cerebral blood flow is inadequate to meet the increased demands for substrate and oxygen.

Increased intracranial pressure — Intracranial pressure is increased during SE. This increase can further interfere with supply of substrate and oxygen and result in cerebral edema. Factors that contribute to increased intracranial pressure include metabolic acidosis, hypoxemia, and carbon dioxide retention with compensatory cerebral vasodilatation and increased cerebral blood flow [49].

Other findings — An elevated peripheral white blood cell count occurred in 60 percent of children in one series, although some patients had febrile illnesses [47]. This increase probably is attributable to demargination of white cells associated with stress. In the same report, 13 percent had cerebrospinal fluid (CSF) pleocytosis not caused by meningitis or encephalitis.

Generalized muscle contractions can lead to elevated body temperature, and rhabdomyolysis can cause hyperkalemia, increased release of muscle enzymes, and myoglobinuria. Myoglobinuria associated with hypotension can result in acute renal failure. (See "Prevention and management of acute kidney injury (acute renal failure) in children".)

OUTCOME — SE can be fatal or associated with long-term morbidity, including seizure recurrence and neurologic problems. The outcome depends upon the underlying cause, the duration of the seizure, and age of the child [28].

Mortality — Mortality associated with SE can result from the underlying condition or from respiratory, cardiovascular, or metabolic complications of SE [50]. In prospective studies, the mortality rate during hospitalization in high-income countries ranged from 2.7 to 5.2 percent [51,52]. The reported longer-term mortality rates of SE in children varies between 3.8 and 17 percent [52-54]. The underlying etiology is the main predictor of mortality.

Morbidity — Neurologic sequelae of SE include focal motor deficits, intellectual disability, behavioral disorders, and chronic epilepsy. One systematic review found that morbidity other than epilepsy occurred in less than 15 percent of patients [51]. Neurologic sequelae are usually caused by the underlying condition rather than the seizures [51,53,55]. While some have noted that rates of neurologic sequelae are increased in younger patients with a longer duration of seizures, these factors are also linked to and difficult to separate from the underlying cause. Patients with cryptogenic SE and those with febrile SE do not have an increased incidence of recurrent seizures or other neurologic sequelae over baseline [51,56].

Recurrent seizures — The risk for having future seizures of any type is high when SE is the child's first seizure, up to 50 percent in two reports [30,57]. Recurrent SE also was more likely to occur in children who presented with SE (21 versus 1 percent in those with a brief initial seizure) [30]. Other risk factors for recurrence include [30]:

●Remote symptomatic etiology

●Abnormal electroencephalogram (EEG)

●Seizure during sleep

●History of prior febrile seizures

●Focal postictal deficits, including Todd paralysis

In other series, subsequent seizures occurred in 30 percent of 125 children with no history of prior unprovoked seizures, in spite of anticonvulsant treatment in most cases [55,58]. Recurrent SE occurred in 16 percent in the first year of follow-up [52]. Recurrent SE occurred primarily in neurologically abnormal children [27,58].

Neurologic sequelae — Some children, especially those with prolonged and/or inadequately treated SE, have residual neurologic sequelae, although rates are variable [50,54,57,59-62]. As with recurrence of seizure, neurologic outcome depends primarily on the underlying condition [53,63]. In one review, encephalopathy and neurologic deficits occurred in 6 to 15 and 9 to 11 percent, respectively, of children and adults with SE [60].

Refractory status epilepticus — Refractory SE, which is defined as persistent seizure activity despite appropriate therapy, develops in approximately one-third of children with status epilepticus and is associated with high mortality and morbidity [64]. In retrospective reports, the long-term mortality of refractory status epilepticus was 5.4 to 32 percent [50,64,65]. Younger patients (<5 years) and those with multifocal or generalized abnormalities on EEG were more likely to die. In survivors, recurrent seizures were common (31 to 97 percent), as were new neurologic deficits (39 to 100 percent).

SUMMARY AND RECOMMENDATIONS

●The duration of continuous seizure activity used to define status epilepticus (SE) has varied over time. For the purposes of treatment decisions, a generally accepted definition is a seizure that lasts for five minutes or longer or is repeated frequently without regaining consciousness between seizures. (See 'Definition' above.)

●The usual classification of SE is similar to that for individual seizures and includes focal SE without impairment of consciousness or awareness (also known as simple partial SE); focal SE with impairment of consciousness (also known as complex partial SE); generalized convulsive SE including tonic-clonic, tonic, and clonic (these are always associated with loss of consciousness); and generalized nonconvulsive SE including absence SE (characterized by altered awareness, but not necessarily unconsciousness). (See 'Classification' above.)

●SE can be a complication of acute illness such as encephalitis, or can occur as a manifestation of epilepsy. Between 10 and 20 percent of children with epilepsy will have at least one episode of SE. SE occurs as the first seizure in 12 percent of children with epilepsy. (See 'Epidemiology and etiology' above.)

●The main contributor to the morbidity and mortality of SE is the underlying etiology. Systemic complications of SE also contribute to its associated morbidity and mortality; these include hypoxemia, acidemia, glucose alterations, and vascular changes. (See 'Outcome' above and 'Systemic complications' above.)

●Epilepsy is a frequent complication in a patient who presents with SE and may be as high as 20 percent. The presence of an underlying etiology causing neurologic abnormality increases the risk of epilepsy. (See 'Recurrent seizures' above.)