The term nonconvulsive status epilepticus (SE) is used to describe a situation in which there is prolonged electrographic seizure activity that results in an alteration of consciousness or behavior, without convulsive movements.¹ Because many other conditions can result in altered mentation, an electroencephalographic (EEG) pattern consistent with nonconvulsive status is needed to confirm this diagnosis. However, the exact EEG changes necessary for diagnosis must be clearly stipulated, as EEG may show epileptiform activity associated with similar cognitive and behavioral changes in cases that are not in nonconvulsive SE.

Kaplan proposed the following definition of nonconvulsive SE, which takes into account both clinical and electrographic criteria:²

1. 
   

   Altered consciousness or behavior from the baseline state for at least 30 minutes without convulsive movements

   
   

   and

   
   
2. 
   

   The presence of one or more of the following epileptiform patterns:

   
   
   
   
   1. 
      

      Repetitive focal or generalized epileptiform activity (spikes, sharp waves, spike and wave, sharp and slow-wave complexes) or rhythmic theta or delta at more than 2 per second

      
      
   2. 
      

      The above EEG patterns at less than one per second, but with improvement or resolution of epileptic activity and improvement of the clinical state following intravenous (IV) injection of a rapidly acting antiepileptic drug (AED), such as benzodiazepine

      
      
   3. 
      

      A temporal evolution of epileptiform or rhythmic activity at more than one per second with change in location or frequency over time

Nonconvulsive SE encompasses both generalized nonconvulsive SE, including absence SE and myoclonic SE, and localization-related or focal subtypes. This chapter will focus on absence SE.

The differentiation of complex partial SE from absence SE on clinical grounds alone can be difficult; however, some clinical distinctions have been described. In absence SE, alteration of contact with the environment is more continuous, in contrast to the cyclic variations in responsiveness levels that may be seen in complex partial SE.³ Motor signs in absence SE consist of generalized myoclonus compared with more focal or lateralized findings, such as head or eye version, focal limb posturing, or focal weakness in complex partial SE. Prominent fear, anxiety, irritability, and speech and memory difficulties are more suggestive of complex partial than absence SE. Finally, whereas automatisms may be seen in both absence and complex partial SE, prominent lip smacking and lateralized automatisms would suggest the latter.

Patients with absence SE can be divided into several distinct subgroups:

1. 
   

   Persons with typical absence SE and a history of idiopathic generalized epilepsy

   
   
2. 
   

   Persons with atypical absence SE and a history of symptomatic generalized epilepsy

   
   
3. 
   

   Adults who present de novo with typical absence SE and who frequently have coexistent metabolic dysfunction, infection, medication-related toxicity, or atrophy

   
   
4. 
   

   Borderline cases, which look like generalized nonconvulsive SE but have some focal components, often in the context of a history of focal epilepsy, particularly frontal lobe epilepsy

Because these groups are clinically quite distinct, the epidemiology, clinical and EEG features, possible causes or triggers, and treatment will be addressed separately for each group.

Epidemiology

In the idiopathic generalized epilepsies, nonconvulsive SE is much more common than is convulsive SE, and the risk of absence SE varies considerably between different idiopathic generalized epilepsy syndromes.⁴ Overall, the risk of absence SE appears somewhat greater in adults with absence seizures (24.4%)⁵ than in young children (16–20%).^6–8 In a study by Agathonikou et al., only 29% of adults with absence SE had onset of absence epilepsy before the age of 10 years.⁵

Table 26-1 shows the reported risk of absence SE in the various idiopathic generalized epilepsy syndromes. Syndromes with the greatest risk of absence SE include two that are not yet officially recognized by the International League Against Epilepsy: phantom absences with generalized tonic-clonic seizures and perioral myoclonia with absence. The entity of phantom absences with generalized tonic clonic seizures was reported by Panayiotopoulos et al. in 13 adults, representing 9.6% of all adults with idiopathic generalized epilepsy.⁹ Absences were very mild, usually lasting 2 to 4 seconds, and associated with inconspicuous impairment of cognition. Generalized tonic-clonic seizures occurred without consistent circadian distribution or specific provoking factors, and appeared later (mean age 32 years) than in most other idiopathic generalized epilepsies. Patients were not photosensitive and did not have associated myoclonic jerks. One or more bouts of absence SE occurred in half of all patients.

Perioral myoclonia with absence has a relatively broad range of onset from 2 to 13 years.¹⁰ Absences vary in frequency but are typically brief (<4–8 s) and associated with rhythmic contractions of the orbicularis oris, depressor anguli oris, or muscles of mastication, resulting in twitching of the mouth, protrusion of the lips, and jerking of the jaw. Generalized tonic-clonic seizures occur in nearly all patients but are infrequent. Over half will develop one or more episodes of absence SE, which may end with a generalized tonic-clonic seizure. The interictal EEG shows brief bursts or fragments of spike wave. Ictally, irregular, generalized polyspike-wave discharge at 3 to 4 Hz without photosensitivity is seen. This syndrome also persists into adulthood and tends to be more resistant to antiepileptic therapy.

Genton et al. described 11 adults with a distinct subtype of idiopathic generalized epilepsy that they termed absence status epilepsy.¹¹ Patients presented after puberty or in early adulthood with recurrent, unprovoked typical absence SE but also had infrequent generalized tonic-clonic seizures and, in some, rare absences. Family history was negative, neurologic examination and imaging were normal, and the EEG showed generalized spike and polyspike-wave discharges on a normal background. Photosensitivity was not present. Absence SE responded variably to benzodiazepines, but valproate afforded good seizure control.

Possible Mechanisms of Absence Status

Absence seizures are dependent on generation of abnormal oscillatory rhythms in the corticothalamic network. Oscillatory activity is dependent on low-threshold (or transient) calcium currents that are activated by hyperpolarization and deactivated by depolarization. The oscillation produced by such activity is fundamental to the sleep spindle. In absence epilepsy, abnormal excitation (possibly related to impaired inhibition) exists that includes the neocortex and neurons within the reticular nucleus of the thalamus. This excitation produces hyperpolarization of thalamocortical neurons and results in activation of T-type calcium channels and action potentials that activate neocortical neurons. These neocortical neurons in turn repeat the cycle, and a resonant, abnormal rhythm is manifested.

The spike in calcium current is associated with activation of the hyperpolarization-activated (I_h) depolarizing current, which slowly brings the thalamocortical neuron into a hyperpolarized state.⁴ The risk of absence SE may be increased by genetic or environmental changes that reduce this Ih current or by factors that result in increased thalamocortical hyperpolarization, including decreased cortical inhibition, activation of neurons in the nucleus reticularis thalami, or increased gamma-aminobutyric acid (GABA) type B receptor activation on the thalamocortical neurons.

Clinical Manifestations: Ictal Behavior and EEG

Typical absence SE in the idiopathic generalized epilepsies affects patients without neurologic or intellectual handicaps. Patients present with variable degrees of clouding of consciousness; in most cases, they are confused but responsive. They may display inappropriate behaviors and be unable to identify familiar people. Approximately half of all patients have some motor signs, including facial and eyelid myoclonus, eye blinking, decreased speech, and slowed responses. The duration of impairment varies from hours to days. Frequently, an episode of absence SE will culminate with a generalized tonic-clonic seizure (Figure 26-1).

The EEG usually shows trains of 3 Hz spike-wave discharge, although the actual frequency may vary from 1 to 4 Hz; polyspike-wave discharges may also be seen. With increased duration of absence SE, the frequency of spike-wave discharges decreases, and discharges become more irregular.¹²

Triggers

Absence SE may be brought on by either abrupt withdrawal of effective AEDs or use of inappropriate medication. Tiagabine,^13,14 vigabatrin,^15,16 and carbamazepine^16,17 should not be used in primary generalized epilepsy, as they may trigger absence SE. However, even medications that are effective for generalized seizures, such as lamotrigine,¹⁸ valproate,¹⁹ and levetiracetam,²⁰ have been reported to exacerbate absence SE. Other possible triggers include physiological stressors (e.g., sleep deprivation, fatigue, hypoglycemia, and menstruation), psychological stressors, hyperventilation, photic stimulation, and other medications (e.g., baclofen, third-generation cephalosporins, lithium, chloroquine, and immunomodulating drugs, such as tacrolimus and ifosfamide).

Treatment

Inappropriate AEDs or other agents that may be exacerbating absence seizures should be discontinued. Absence SE in the idiopathic generalized epilepsies is usually very responsive to benzodiazepines, either given intravenously or by oral, buccal, or rectal routes (diazepam [0.2–0.5 mg/kg to maximum of 20 mg], lorazepam [0.05–0.1 mg/kg to maximum of 4 mg], or midazolam [0.1–0.2 mg/kg to maximum of 10 mg]). In the rare case that the condition does not respond to benzodiazepines, IV valproate (20 mg/kg) may be effective.

Although absence SE should be treated promptly after diagnosis, it is debatable whether this condition causes brain damage. In an animal model of absence SE, Wong et al. found no evidence of brain damage or behavioral change.²¹ Shirasaka found that cerebrospinal fluid (CSF) neuron-specific enolase levels were normal in two cases of absence SE, suggesting that this form of SE does not result in neuronal damage.²² There is no evidence that patients with idiopathic generalized epilepsy have a more refractory form of epilepsy after the occurrence of absence SE. However, the idiopathic epilepsy syndromes in which absence SE is more commonly encountered (i.e., perioral myoclonia with absence and idiopathic generalized epilepsy with phantom absences) may be more resistant to AEDs than syndromes with less frequent absence SE (i.e., childhood absence epilepsy, juvenile absence epilepsy, and juvenile myoclonic epilepsy).

Epidemiology

Nonconvulsive SE epilepticus is common and probably underrecognized in the epileptic encephalopathies. It is seen in multiple types of symptomatic generalized epilepsy.

Lennox-Gastaut syndrome typically presents in the preschool years in children who usually have a prior history of neurologic disease. Multiple seizure types can be seen, including atonic, atypical absence, myoclonic, and partial. However, nocturnal tonic seizures are the most specific type for this syndrome. Children are usually developmentally delayed, and the interictal EEG shows background slowing with frontally predominant, slow (usually <2 Hz) spike-wave discharges. Nonconvulsive SE occurs in 50 to 75% of these children, usually presenting as recurrent atypical absences that are often interspersed with brief tonic seizures.

Myoclonic astatic epilepsy of Doose also affects preschoolers. However, children are neurologically normal prior to presentation. Seizure types are multiple, including myoclonic, atonic, and generalized tonic-clonic. However, the myoclonic-astatic seizure is the most characteristic of this syndrome. The interictal EEG shows slow spike-wave (usually 2–3 Hz) with theta rhythms in the parietal regions. Nonconvulsive SE is seen in 36% and manifests with staring and/or stupor, often with subtle myoclonus of the face and limbs.²³

Dravet syndrome begins in infancy with recurrent, often prolonged, focal or generalized seizures triggered by low-grade fever. Myoclonus appears closer to 2 years of age, and children ultimately may develop other seizure types, including atypical absence and complex partial seizures. Over three quarters of children have a detectable mutation in the SCN1A gene. Forty percent have one or more episodes of nonconvulsive SE, and present with stupor and erratic myoclonus.

Angelman syndrome is caused by lack of UBE3A gene expression from the maternally inherited chromosome 15 due to various 15q11-q13 abnormalities.²⁴ It characteristically presents with severe global developmental delay, absence of language, ataxia, hypotonia, happy demeanor, and sleep disruption. Over 90% of children have epilepsy (generalized tonic-clonic and myoclonias), and absences are the most frequent seizure types. Generalized, nonconvulsive SE is frequent and usually takes the form of exceedingly frequent absences accompanied by subcontinuous myoclonias involving the face and/or distal muscles.

Children with ring chromosome 20 present with frequent episodes of nonconvulsive SE. Developmental delay varies in severity, and dysmorphic features are described but may be absent early in life. Children may present with a characteristic ictal behavior consisting of terror and hallucinations²⁵ and frequent subtle nocturnal seizures with frontally predominant beta evolving into bifrontal high-voltage delta waves.²⁶ Chromosome 20 involves known epilepsy genes, including the nicotinic cholinergic receptor (ACHA4) and the potassium channel gene (KCNQ2).

Wolf-Hirschhorn (4p-) syndrome presents with dysmorphic features, including hypertelorism, microcephaly, high nasal bridge, micrognathia, and major malformations of the skeletal system, heart, and kidneys. Seizures, which are often triggered by fever, usually begin in the first 2 years of life and consist of generalized or unilateral tonic-clonic seizures, atypical absences, and myoclonias. Nonconvulsive SE may be seen.²⁷

Possible Mechanisms of Atypical Absence Status Epilepticus

The etiology of epileptic encephalopathies is heterogeneous and includes structural and genetic causes, but it is often unknown. The seizures are possibly due to the maturational phenomena of hyperexcitability combined with inhibition of the cortex. This would lead to decreased cortical inhibition of the subcortical regions, which may cause subtle seizures, such as atypical absence seizures.²⁸

Clinical Manifestations: Ictal Behavior and EEG

Clinically, children present with varied degrees of impairment of awareness. Additionally, ataxia, drooling, and frequent myoclonias of the face or distal muscles often accompany absence SE in this population. Furthermore, in children with Lennox-Gastaut syndrome, brief tonic seizures frequently are interspersed within the nearly continuous absence seizures. However, recognition of absence SE requires a high degree of suspicion to appreciate the change in mental awareness in a child who is usually already moderately impaired. In 50 children with nonconvulsive SE, of whom 31 had either Lennox-Gastaut or myoclonic astatic epilepsy, Stores et al. found that the change in behavior was obvious in only 64%; these children showed decreased activity, slowed responsiveness, and impaired level of consciousness, with or without motor manifestations of poor balance or intermittent bilateral limb myoclonus.²⁹ Consequently, in children with symptomatic generalized epilepsies, nonconvulsive SE is often diagnosed late, and, even when recognized, it is often not aggressively treated. In the study by Stores et al., only 5 of the 50 cases were recognized within hours to days after onset, and only 20 of the 50 patients were suspected to be in nonconvulsive SE when the EEG was ordered.²⁹ In a smaller study of eight patients, the duration of nonconvulsive SE prior to diagnosis ranged from 3 days to 4 weeks.³⁰

The EEG during SE in children with atypical absence SE is variable, depending on the underlying disorder. It typically shows diffuse, slow, rhythmic activity that is less well formed than typical absence SE. In Lennox-Gastaut syndrome, the ictal EEG shows rhythmic slow spike waves (<3 Hz), and in some children, diffuse high-amplitude theta/delta waves are present (Figure 26-2). Bursts of faster 10 Hz activity are superimposed on the slow activity. Frontally predominant, 10 Hz recruiting rhythms are frequently seen on EEG in children with Lennox-Gastaut syndrome who are in absence SE.¹² The EEG during nonconvulsive SE in children with myoclonic astatic epilepsy of Doose shows irregular slow waves and spike-wave complexes that occur in long bursts. Slow spike waves may also occur. The EEG can become so disorganized that it is similar in appearance to hypsarrhythmia.³¹ As with children with Lennox-Gastaut or myoclonic astatic epilepsy, the EEG of children with Dravet syndrome consists of diffuse slow waves with rare spikes and sharp waves that can be either focal or diffuse. The spikes and sharp waves have a higher voltage over the anterior head regions and the vertex.³² In Angelman syndrome, the EEG characteristically shows rhythmic patterns that may be of three types: (1) prolonged runs of high-amplitude, frontally predominant delta at 2 to 3 Hz with superimposed sharp waves; (2) rhythmic 4 to 6 Hz, high-amplitude activity; and (3) posterior predominant, 3 to 4 Hz, high-amplitude spike and sharp waves seen mainly with eye closure.²⁴ In children with ring chromosome 20 syndrome, the EEG may demonstrate continuous rhythmic theta/delta activity over the bifrontal head regions.¹² The EEG in Wolf-Hirschhorn (4p-) syndrome may show two patterns: (1) slow 2 to 3 Hz high-voltage, rhythmic notched waves or waves with superimposed spikes, often beginning asynchronously in either the frontocentral or posterior regions; or (2) bursts of posterior rapid repetitive spikes.²⁷ This paroxysmal activity is activated by slow-wave sleep.

There are other epileptic syndromes with continuous epileptiform activity on EEG that may be mistaken for absence SE. Electrical SE in sleep (ESES) is an EEG finding described by Patry et al. in reference to six children with nearly continuous activation that began with sleep onset, continued throughout the night, and resolved when the children awakened.³³

Children with ESES on EEG typically have one of two clinical syndromes: either continuous spike-wave activity during slow-wave sleep (CSWS) or Landau-Kleffner syndrome (LKS). The EEG during wakefulness in CSWS may show brief bursts of spike-wave activity of the frontotemporal or centrotemporal regions, or it may be diffuse. During non–rapid eye movement (NREM) sleep, the discharges are significantly activated and become nearly continuous.^34,35 The EEG during wakefulness in LKS is variable and may contain focal or generalized epileptiform discharges, or it may be normal.^36,37 Like CSWS, during sleep there is marked activation, although the spike-wave activity is maximal over the centrotemporal head regions.³⁸ In contrast to atypical absence SE, the discharges in both of these syndromes disappear upon awakening.^34,35