Clinical manifestations

Presentation and course

According to the ILAE task force report on classification of status epilepticus, the first axis of classification is based on semiology (47). The two main criteria in this axis are presence or absence of prominent motor symptoms and degree of unconsciousness. Those with prominent motor symptoms are categorized into convulsive status epilepticus, myoclonic, and focal motor. Under the category of convulsive, there are generalized convulsive status epilepticus, focal onset evolving into bilateral convulsive status epilepticus, and unknown. The reported percentage that starts with a focal onset varies significantly depending on the age and population being studied. If generalized convulsive status epilepticus is untreated or inadequately treated, the motor manifestations become increasingly subtle the longer it persists. Impaired consciousness is always present from the time the seizure generalizes.

Prognosis and complications

Generalized convulsive status epilepticus is the most dramatic and life-threatening form of status epilepticus. The reported case-fatality ratio varies among the studies and is closely linked to the underlying etiology, which can be different in different age groups. Low level of antiseizure medications, trauma, alcohol, and fever are usually associated with lower mortality rate whereas hypoxia, stroke, and CNS infection are associated with higher rates. In a meta-analysis of 61 studies, the pooled mortality rate in adult studies was 15.9% (95% CI, 12.7%-19.2%), 13% (95% CI, 7.2%-19%) for all-aged population studies, and 3.6% (95% CI, 2%-5.2%) for pediatric studies (35). There are variations in the reported estimate of mortality, ranging from 21% in the first 30 days after an episode of status epilepticus to 31.2% 10-year mortality. These estimates are lower in reports that are based on in-hospital mortality due to shorter observation periods (32). A retrospective study of 70 patients with generalized convulsive status epilepticus in Finland showed that severe form of status epilepticus, delayed diagnosis, high number of complications during treatment, and poor condition at hospital discharge were independent predictors of long-term mortality (48). In general, mortality is highest in the elderly, up to 50% in the age group of 80 years and above, which is probably because of presence of more comorbid conditions in this age group and because the etiology is often of more ominous causes such as anoxia, hypoxia, and severe stroke (29).

Refractory status epilepticus is defined as status epilepticus that persists despite use of an appropriate dose of a first-line agent (benzodiazepines) and a second-line antiseizure medication, and usually requires a third-line medication (anesthetics). Super-refractory status epilepticus (SRSE) refers to status epilepticus that persists for more than 24 hours following the introduction of anesthetics. About 29% to 43% of patients with status epilepticus progress to refractory status epilepticus despite treatment. The studies investigating the relation of refractoriness to mortality at times have revealed conflicting results. In a study of patients with status epilepticus with significant motor symptoms the overall mortality was high, at 28.4%, but no clear relationship between the refractoriness and mortality was found. They identified elderly age, continuous clinical seizure activity, absence of former seizure, infection, prolonged ICU stay, use of anesthesia, and cardiac comorbidity to be predictors of mortality in refractory status epilepticus. On the other hand, a cohort of 804 patients with status epilepticus reported a significantly higher mortality rate in refractory status epilepticus (24.5%) and super-refractory status epilepticus (37.9%) compared to the patients with nonrefractory status epilepticus (9.8%) (12). Another study of 68 patients with refractory status epilepticus found predictors of poor outcome were low GCS at the time of admission, need for mechanical ventilation and intravenous anesthesia, and long duration of refractory status epilepticus before recovery (22). In another retrospective cohort of 148 patients with status epilepticus, 14.2% of patients were found to have super-refractory status epilepticus, with a mortality rate of 28.6%. Factors that were associated with increased mortality in super-refractory status epilepticus were younger age, absence of prior systemic disease or epilepsy, generalized convulsive status epilepticus, unsuccessful anesthetics taper-off, need for a second anesthetic agent and increased duration of its use, and multiple systemic complications (13). In a cohort of pediatric patients with refractory convulsive status epilepticus, the longer time to onset of treatment with benzodiazepines was associated with higher mortality (16).

Other complications during generalized convulsive status epilepticus include respiratory failure and hypoxia, hyperpyrexia, excessive sweating, salivary and tracheobronchial hypersecretion, tachycardia, bradycardia, cardiac arrhythmias, pulmonary edema, respiratory or metabolic acidosis, hyperkalemia, hypoglycemia, hyponatremia, rhabdomyolysis, renal failure, hepatic failure, and intracranial hypertension. About 30% of patients with refractory status epilepticus show neurologic deficits at long-term follow-up (14; 23).

Patients with status epilepticus are at risk of developing subsequent epilepsy and the reported risk in different studies has a wide range of about 12% to 40% (20). However, most of these studies are not limited to generalized convulsive status epilepticus but include nonconvulsive status epilepticus as well. In a cohort of children with generalized convulsive status epilepticus, the cumulative incidence of epilepsy was 24.7%, mostly in those with remote symptomatic generalized convulsive status epilepticus (38).

Status Epilepticus Severity Score (STESS) is a 6-point scoring that was originally developed to orient early treatment strategy (39) and was shown to have a negative correlation with survival and return to baseline clinical condition. In the STESS scoring system, 1 point is assigned for being stuporous or comatose, 1 point for generalized convulsive seizure and 2 points for non-convulsive status epilepticus in coma, 2 points for age of 65 or higher, and 1 point for no (or unknown) history of previous seizures.

The Epidemiology-based Mortality Score in Status Epilepticus (EMSE) is another scoring system based on four factors—etiology, age, comorbidity, and EEG (EACE)—that has a prognostic value in mortality (28). In a study comparing the STESS with modified EMSE score (using three factors: etiology, age, and comorbidity-EAC) the cutoff for STESS was 4 points or higher and for EMSE-EAC was 34 points or higher. Both scores were found to have useful short-term mortality prediction and high negative predictive value (40). A review of prognostic scores in status epilepticus is provided by Yuan and colleagues (54).

Biological basis

Etiology and pathogenesis

According to etiology (axis 2) classification of status epilepticus in a 2015 report of the ILAE task force, the underlying cause of status epilepticus can be divided into symptomatic or known and cryptogenic or unknown. In their report, it is argued that use of the term idiopathic or genetic is not applicable to the underlying etiology of status epilepticus as there is a known underlying metabolic, toxic, or intrinsic factor (such as sleep deprivation) that may lead to status epilepticus in patients with idiopathic or genetic epilepsy syndromes (47). In the category of known etiologies, based on the temporal relationship, the terms acute (such as stroke), remote (such as postencephalitic), and progressive (such as brain tumor) can be applied. In those with preexisting epilepsy, low serum level of antiseizure medications or breakthrough seizures are the most common etiologies. In those without preexisting epilepsy, the most common etiologies are acute stroke, anoxic or hypoxic injury, alcohol intoxication or withdrawal, drug-induced, infection, trauma, and metabolic derangements.

In children, the reported etiology varies among reports. In data from a cohort of 665 patients, febrile seizure was the cause in about 41% of children mostly between age 1 to 5, more than 55% had unknown cause, and only slightly over 2% had acute symptomatic causes (34). This is in contrast to some other studies where the proportion of known causes was higher (09; 52).

New-onset refractory status epilepticus (NORSE) is a rare clinical condition when status epilepticus develops in a patient without history of active epilepsy or other preexisting relevant neurologic disorder and there is no clear acute or active structural, toxic, or metabolic causes found. Etiology of new-onset refractory status epilepticus is unknown. If a cause is eventually identified, autoimmune encephalitis is the most common. Antibodies targeting N-methyl-D-aspartate (NMDA) receptor and the voltage-gated potassium channel (VGKC) complex are the most commonly found antibodies (43). About half of the cases of new-onset refractory status epilepticus remain cryptogenic despite workup. However, the clinical manifestations in most of these cases resemble the cases with autoimmune etiology, and, therefore, it is possible that they are of autoimmune etiology with antibodies that are not yet identified.

Pathophysiology

Most evidence on status epilepticus pathophysiology is derived from animal models.

Several mechanisms are suggested to be involved in initiation and maintenance of seizure activity in status epilepticus. The initiation of seizure activity is mediated by an increase in excitatory input through glutamate excitation, a decrease in inhibitory input through GABA-mediated inhibition, or a change in both inputs to a group of neurons that usually leads to hypersynchronous neuronal firing. Other contributing mechanisms to synchronous neuronal firing include direct electrical coupling via gap junctions, shifts in extracellular ion concentrations, electrical field potentials mediated by extracellular current flow, and cellular swelling with narrowed intercellular space.

Animal studies show that changes in neuronal membrane receptors with prolonged seizures play a role in maintaining the seizures and make aborting them more difficult. One of these changes is internalization and decrease in membrane density of GABA-A receptors. This means that benzodiazepines that work on GABA-A receptors will be less effective in aborting seizures as the time lapses. On the other hand, there is externalization of the NMDA receptors and increase in their membrane density as soon as 5 minutes after onset of seizure activity. Other changes that lead to decreased inhibition include decreased adenosine, galanin, and dynorphin. Changes that can lead to increased inhibition include increased glutamate release, inflammation and breakdown of the blood-brain barrier, and increase in substance P (42; 21).

Repetitive muscle contraction during status epilepticus significantly increases the metabolic demand of the body and can lead to anaerobic metabolism and increase in metabolic acidosis and changes in heart rate, blood pressure, respiratory rate, blood glucose, electrolytes, and body temperature. Continued seizure can result in hypoxia, pulmonary edema, respiratory acidosis, and rhabdomyolysis. Eventually hypoxia, hypoglycemia, hyperthermia, and acidosis can lead to neuronal injury in prolonged seizures. This injury in neurons in the dentate nucleus and pyramidal layer of hippocampus can lead to hippocampal sclerosis, which, in turn, can be a focus of seizures and generate a vicious cycle (51).

Epidemiology

The reported incidence of status epilepticus has a wide range, from 9.1 to 41 per 100,000 per year. This variation is explained by differences in study design, use of diverse timing definition of status epilepticus, and differences between study populations, to name a few (02). Estimates suggest that the incidence might be higher in females than in males, contrary to what previous studies found.

Differential diagnosis

Confusing conditions

The diagnosis of overt generalized tonic-clonic status epilepticus is not difficult and ordinarily should not be confused with any other condition. The impairment of consciousness differentiates the diagnosis of generalized tonic-clonic status epilepticus from focal-onset motor status epilepticus.

Generalized tonic-clonic status epilepticus does need to be differentiated from nonepileptic status epilepticus, which is often prolonged and can be confused with epileptic status epilepticus. In a study, about 16% of patients who were intubated for a convulsive activity were found to have nonepileptic seizure. These patients were more likely to have a history of psychiatric disorder; have no history of intracranial abnormality; be younger than 50 years of age, white, and female; and have a systolic blood pressure of less than 140 mmHg (49). Clinically, initially true overt generalized tonic-clonic seizures should exhibit recognized patterns of evolution with initial tonic stiffening followed by clonic jerks that increase in amplitude but decrease in frequency as the seizure progresses. Each seizure should be stereotypically similar to others in the series. If the seizures exhibit a Jacksonian spread of motor involvement, the pattern of spread should be consistent with anatomical representation in the cortex. In addition, tonic-clonic activity should be sustained throughout the duration of the seizure without stopping and starting pauses.

Associated and underlying disorders

Generalized tonic-clonic status epilepticus most often starts with focal-onset seizures. However, syndromes of generalized seizures can also present with generalized tonic-clonic status epilepticus, especially in the setting of noncompliance with antiseizure medications. Refer to Presentation and course under Clinical manifestations and Etiology and pathogenesis, above.

Diagnostic workup

The diagnostic workup depends on the patient's history and on the response to treatment of the acute episode of generalized tonic-clonic status epilepticus. If the patient has a preexisting epilepsy, the precipitating cause of the current episode of generalized tonic-clonic status epilepticus is commonly low antiseizure medication serum concentrations or intercurrent infection. This should be determined by obtaining blood for serum chemistries, complete blood count, and analysis of toxicologic agents.

A known history of intracranial abnormality may require neuroimaging with CT or preferably MRI to detect changes in the lesion-like recurrence or enlargement of the mass.

If the episode of generalized tonic-clonic status epilepticus is the patient's first seizure, then the patient should be evaluated for causes of new-onset epilepsy or for the possibility of an underlying CNS condition. Depending on age, CNS infection, cerebral infarction, mass lesion, systemic metabolic disorders, substance abuse, or other toxins must be considered. Specifically, the following tests can be helpful in most cases:

EEG. Many cases of initial generalized tonic-clonic status epilepticus can be diagnosed and managed only on the basis of history and neurologic examination. However, data obtained from an emergent EEG, serial EEGs, or video-EEG monitoring are required, especially in cases in which there is suspected ongoing nonconvulsive status epilepticus after generalized convulsive status epilepticus (15).

Typical EEG in generalized tonic-clonic status epilepticus may show continuous bilateral though frequently asymmetric ictal discharges. However, some ictal EEG patterns are difficult to recognize as subtle generalized tonic-clonic status epilepticus, and a routine EEG can miss intermittent ictal activity. Clinical guidelines for the indication of long-term video-EEG monitoring among critically ill patients, including status epilepticus, are published (19).

The meaning of other EEG patterns such as periodic discharges with or without lateralizing or epileptiform features depends on the clinical setting of status epilepticus; if post-anoxic EEG shows only periodic discharges without evolution, aggressive pharmacologic treatment should generally be avoided. Latest revision of guidelines for terminology of critical care EEG features have been developed by the American Clinical Neurophysiology and can be accessed at the following website: https://www.acns.org/UserFiles/file/ACNSStandardizedCriticalCareEEGTerminology_rev2021.pdf.

General anesthetics and other sedatives used in the ICU may abolish all motor output before they terminate subtle generalized convulsive status epilepticus; so without EEG, physicians will not be able to determine whether status epilepticus is controlled or how much medication is needed. This information is also important in the titration of the anesthetics to achieve desired EEG patterns (burst suppression, seizure cessation, or complete electrocerebral silence). With digital EEG, proprietary mathematical algorithms to process EEG to simplified, rapidly reviewable information are adventitious for management of refractory status and related cases in the neuro ICU. Compressed spectral array (CSA), rhythmic run detectors (R2D2), amplitude integrated EEG (aEEG), and various other paradigms are useful to detect potential seizures for detailed review.

Blood tests. Some of the useful blood tests include antiseizure medication levels, CBC, renal function, electrolytes, glucose, calcium, magnesium, liver function, blood gases, blood-clotting measures, and toxicology. Serum should be tested for presence of onconeuronal and cell-surface neuronal antibodies in cases of new-onset status epilepticus of unknown cause.

Lumbar puncture. Lumbar puncture for CSF analysis may be performed if CNS infection is suspected. CSF may be checked for onconeuronal and cell-surface neuronal antibodies in cases of new-onset status epilepticus of unknown cause, particularly if suspicion for autoimmune etiology is high.

Brain imaging. MRI of the brain with gadolinium enhancement is recommended for first-time new-onset status epilepticus and may reveal the underlying etiology. Status epilepticus itself can result in increased signal intensity on FLAIR, T2-, or diffusion-weighted imaging sequences, along with areas of low value on ADC map. These changes are nonspecific and variably reversible on recovery from status epilepticus.

Management

Management of generalized tonic-clonic status epilepticus should proceed simultaneously along three fronts: termination of status epilepticus, prevention of recurrence, and treatment of complications. Traditionally, status epilepticus has been divided into four time-based stages. In 2012, the Neurocritical Care Society published guidelines for the management of status epilepticus and revised the conventional treatment stages to emergent initial therapy (first-line), urgent control therapy (second-line), and refractory therapy (third- and fourth-line). This was done simply to emphasize the importance of rapid treatment of status epilepticus. The American Epilepsy Society also published guidelines for status epilepticus management in which they divided management into four phases of time-based stabilization phase: initial, second, and third therapy phases (18). Stabilization of the patient is an integral part of treatment.

Table 1. First- and Second-Line Therapy Medications in Status Epilepticus

Termination of status epilepticus. The management of status epilepticus starts with the arrival of paramedics. Benzodiazepines can be given safely and effectively out of the hospital by paramedics. Results of the RAMPART trial involving 79 hospitals and more than 4000 paramedics showed that intramuscular midazolam not only was noninferior but was superior to intravenous lorazepam (44). The superior efficacy of intramuscular midazolam was due to more rapid administration of the drug in a prehospital setting. Buccal or nasal midazolam appeared to be at least as effective as rectal diazepam with little or no side effects. Meta-analysis showed the buccal and nasal administration was easy to handle and socially more acceptable than the rectal route (05).

In-hospital treatment usually begins with an intravenous benzodiazepine as first-line therapy. This should be followed with second-line therapy as either maintenance therapy or an attempt to terminate the status epilepticus if benzodiazepines fail. In a meta-analysis on cost-effectiveness of second-line antiseizure medications in 2019, it was found that the most effective medications were phenobarbital, valproate, lacosamide, levetiracetam, and phenytoin. The most cost-effective medications in the same study were levetiracetam followed by valproate (41). In a randomized controlled clinical trial (ESETT-Established Status Epilepticus Treatment Trial), the efficacy of intravenous levetiracetam, phenytoin, and valproate were compared, and no statistically significant differences were seen between these three medications as all three resulted in seizure cessation and improved consciousness by 60 minutes in approximately half of the patients (24). In two randomized trials in children (ConSept and EcLiPSE) it was shown that levetiracetam is not superior to phenytoin as second-line treatment in pediatric status epilepticus (11; 33). Valproate should be considered first when idiopathic (genetic) generalized epilepsy is the cause of status epilepticus. The evidence for using topiramate is scant and inconclusive. There is increasing evidence about the efficacy of lacosamide in both convulsive and nonconvulsive status epilepticus with an overall efficacy of about 57% in a systematic review of the literature (45). There is little controlled data about brivaracetam, but there are some promising observational data about the efficacy of this medication (04; 37).

When available, the intravenous form of the patient’s routine medications should be given as low levels are a common cause of status epilepticus, and lab results of serum levels may not be available in a timely manner. Oral replacement of routine medications can also be given but may have a delayed effect.

In cases of refractory status epilepticus, third-line coma-inducing agents should be started, aiming for burst suppression or seizure suppression on continuous EEG. There is far less unanimity for the choice of agents for third-line therapy. Although many epilepsy specialists prefer nonanesthetic antiseizure medications, such as barbiturates or benzodiazepines, many ICU specialists prefer anesthetic agents like propofol. Currently, the best evidence supports continuous intravenous infusions of short-acting barbiturates (pentobarbital), benzodiazepine (midazolam), or propofol. Pentobarbital was superior to both propofol and midazolam in effectively controlling refractory status epilepticus (10; 30); however, it was significantly more likely to lead to hypotension. Prolonged use of propofol comes with higher risk of propofol infusion syndrome, which can be fatal. The cardiorespiratory side effects of midazolam are less compared to pentobarbital or propofol. If patients cannot tolerate hemodynamic instability or use of vasopressors, intravenous ketamine is an option without risk of hypotension. Ketamine can also be used for super-refractory status epilepticus (01). In a literature review published by the treatment committee of the American Epilepsy Society, it was noted that in the case of convulsive refractory status epilepticus there is not sufficient evidence on the efficacy of antiseizure medications such as lacosamide, levetiracetam, brivaracetam, and valproate, as well as anesthetic agents such as pentobarbital, propofol, midazolam, and ketamine (50).

Burst-suppression or complete seizure suppression should be maintained for 24 to 48 hours. One or more antiseizure medications should be optimized with high or supra therapeutic doses/levels during this time in order to increase likelihood of a successful wean from coma-inducing or anesthetic agents. The patient’s usual antiseizure medications with suspected or known low levels should be replenished and maintained. If seizures recur on weaning from anesthetic agents, additional antiseizure medication and other agents may be added. If maximal medication treatment is unsuccessful to escape anesthetic-sedative therapy, alternative treatments should be considered, and longer duration burst-suppression or seizure suppression with an anesthetic agent should be continued, along with alternate therapies.

Table 2. Treatments for Refractory Status Epilepticus

Alternative treatments for super refractory status epilepticus (SRSE). Alternative therapies should be considered in cases of refractory status epilepticus that prove refractory to all treatments. Most of these cases occur in young adults with new-onset refractory status epilepticus. Treatment recommendations currently are mostly based on case series and clinical experience rather than randomized clinical trials. If immune causes are suspected, for acute treatment, high-dose intravenous steroids (IV methylprednisolone 1 g/day for 3 to 5 days), with or without intravenous immunoglobulin (IVIG) (0.4 g/kg per day for 5 days) or plasma exchange (alternate days for five sessions) are recommended (46). If these first-line treatments fail or there is relapse, second-line immunosuppressive therapy such as rituximab or cyclophosphamide can be considered (08).

Therapeutic hypothermia has received attention as an alternative treatment for status epilepticus, refractory status epilepticus, and super-refractory status epilepticus mostly based on the animal models; however, the evidence for its clinical use has not been strong. The HYBERNATUS (Hypothermia for Brain Enhancement Recovery by Neuroprotective and Anticonvulsant Action after Convulsive Status Epilepticus) trial is a randomized, open-label prospective trial looking at the use of moderate hypothermia (32° to 34° C) for 24 hours, which found that hypothermia added to standard care was not associated with significantly better 90-day outcome than standard care alone in patients with convulsive status epilepticus (27). In a few pediatric case reports, there was significant control of seizures with therapeutic hypothermia but many of them experienced recurrence after rewarming (26).

A phase I/II multicenter study on ketogenic diet that involved 15 patients with super refractory status epilepticus showed resolution of seizure in 11 (73%) of the patients. Side effects included metabolic acidosis, hyponatremia, hypoglycemia, hyperlipidemia, and constipation. Five patients ultimately died (33%) (07).

Surgical treatment of refractory status epilepticus has been documented in small case series and case reports that involved cortical resection, callosal sectioning, multiple subpial resection, hemispherectomy, and vagus nerve stimulation. There is no protocol on how or when to surgically manage this medical emergency. In a retrospective study on pediatric status epilepticus in Miami Children’s hospital, all 15 patients who had refractory status epilepticus underwent ictal SPECT and FDG-PET imaging with intraoperative electrophysiological studies and surgery had seizure control and transitioned out of the intensive care unit. Four patients incurred worsened neurologic function, developing hemiparesis and dysphasia (03).

Prevention of recurrence. Identifying the etiology and treating the underlying cause is crucial for the proper management of generalized tonic-clonic status epilepticus. Many etiologies, such as intracranial infections, metabolic derangements, and stroke, are treatable. Treatment of alcohol or benzodiazepine withdrawal is essential for generalized tonic-clonic status epilepticus treatment.

Treatment of complications. Management of the complications of generalized tonic-clonic status epilepticus involves understanding and managing the physiological consequences of tonic-clonic seizures. Greater duration of anesthesia and coma is associated with increased morbidity and mortality. Following status epilepticus, hyperthermia may persist for some time. Temperature elevation may contribute to brain damage. Elevations in epinephrine may trigger cardiac arrhythmias and ischemic changes as well as elevations in plasma glucose. Peripheral accumulation of lactate may result in acidosis. Surprisingly, status epilepticus-induced brain injury does not seem to be exacerbated by acidosis; protection seems to be because acidosis itself may have an anticonvulsant effect, interceded by the proton-mediated inhibition of glutamate at the N-methyl D-aspartate (NMDA) channel (25).