Emergency Medicine Reports July 7, 1997

Status Epilepticus: A Systematic Approach To Assessment, Differential Diagnosis, and Outcome-Effective Management

Author:J. Stephen Huff, MD, ssociate Professor of Emergency Medicine and Neurology, Department of Emergency Medicine, University of Virginia Health Sciences Center, Charlottesville.

Peer Reviewer:Steven M. Winograd, MD, FACEP,Attending Physician, Department of Emergency Medicine, Battle Creek Health System, Battle Creek, MI.

The patients are very sick—on occasion, they are in extremis. The history is often sketchy or misleading, and the pharmacological treatment options are potentially toxic and complicated. In short, when it comes to status epilepticus (SE), the heat is always on, the clinical territory is treacherous, and there is little tolerance for error. Among the most challenging, life-threatening, and complex conditions encountered by the emergency physician, generalized convulsive status epilepticus (GCSE) demands prompt etiologic characterization and a logically sequenced approach to pharmacotherapeutic stabilization.

Not surprisingly, management of these patients frequently mandates a team approach. Moreover, the clinical mission statement and treatment goals for this condition are rarely ambiguous. From a practical clinical perspective, the emergency practitioner must put an end to the seizures, stabilize the patient's overall medical condition, identify an etiology for the epileptic episode, coordinate care with appropriate subspecialists, and make a disposition to the appropriate service and medical unit.

Interventions must be aimed at maximizing the opportunity for seizure cessation while minimizing the risks from complications of drug therapy and invasive procedures designed to control the airway. The balancing act requires clinical acumen, a thorough understanding of seizure pathophysiology and classification, and balanced assessment of the risk of medication-related adverse effects. Because unusual, multi-factorial, or chronic etiologies are frequently the precipitating factor in GCSE, comprehensive care of the patient inevitably involves interaction with a number of specialty and primary care physicians, including neurologists, neurosurgeons, internists, pediatricians, family practitioners, and/or intensivists.

Finally, nothing looms with greater importance and urgency in the mind of the ED physician than the heat-of-battle questions—and corresponding solutions—that surface while managing patients with GCSE. For example, the clinician will want to pin down the etiology of the seizure. Is it caused by medication noncompliance? An ischemic event? Infection? And how will management and assessment be tailored according to specific precipitating factors? When is a stat EEG useful, and what are the indications for computerized tomography (CT) scanning? Second, what seizure-specific and systemic-related interventions minimize the risk of cerebral ischemia and neuronal damage? And finally, what are the appropriate treatment algorithms and drugs of choice for managing GCSE in the emergency setting?

As you might expect, the literature on epilepsy is vast, and the clinical trials cited in this monograph are oriented toward issues and strategies that affect the ED management. However, with these clincal issues in clear focus, this review will provide a targeted and clinically oriented approach to the diagnosis and management of GCSE in the adult patient.

—The Editor

Historical Features and Current Classification of Status Epilepticus

Baffled by the unusual and violent movements that characterize epilepsy, the ancients attributed SE to possession by spirits or demons. In fact, the earliest surviving reference to SE is found in Babylonian cuneiform stone tablets dating to the middle of the first millennium BC.1 Interestingly, these tablets note the mortality associated with repeated seizures. Later, Hippocrates referred to epilepsy as "the sacred disease." The modern description of continuous seizure actvity has been attributed to a French physician who published his obervations in 1824, with the term "status" appearing in an English translation of this work.2 The first American descriptions of clinical SE date from the early 1900s. The syndrome of GCSE is clearly described in these reports, which emphasize the morbidity and mortality associated with this clinical syndrome.3,4

Current Classification. urrently, SE is a term that may be used to describe any continuing seizure acitvity regardless of its etiology. From a clinical perspective, an extremely useful approach to categorizing SE is to divide SE into classifications that are similar to those used for seizure disorders. As might be expected, these classifications are based on information and neuronal discharge patterns observed on video-electroencephalograms (VEEG).5,6

Simple Seizures.Using this scheme, the term "simple" implies that an isolated area of the cortex is involved. Clinically speaking, simple SE patterns will produce unremitting focal motor or sensory phenomena, with full preservation of consciousness. For example, the twitching of an extremity or small area of the body in the setting of preserved consciousness (and confirmatory electroencephalogram [EEG] evidence of focal electrical abnormality) would be categorized as a simple motor seizure.

Generalized Seizures. he term "generalized" indicates that abnormal electrical activity involves all areas of the cerebral cortex. In this case, generalized activity may spread from a single, focal area of electrical abnormality, or it may originate in all areas of the cortex simultaneously. When all cortical areas are observed (by EEG) to be involved at once, the condition is characterized as a primarily generalized seizure. This epileptic pattern may occur with petit mal seizures and in some (primary) generalized tonic-clonic seizures. Absence seizures (also known as petit mal) are a primarily generalized seizure type involving all cortical areas at once. If diffuse involvement of the cortex results from a localized or focal cortical area, this is "secondary generalization;" this spread may be observed at the bedside or by EEG. Accordingly, GCSE, refers to a continuing generalized seizure, regardless of its underlying eliptogenic trigger or pattern of electrical migration.

In adults, most generalized seizures originate from a focal area of abnormal electrical activity that then spreads through secondary generalization to involve all areas of the cortex. Often, this electrical migration occurs too rapidly to be observed clinically at the bedside. On occasion, the typical aura of a generalized convulsive seizure will persist in the absence of any associated motor movements. It should be stressed that this aura represents ongoing seizure activity that is limited to a focal area of the cortex. In this case, however, the abnormal, localized electrical activity associated with the seizure does not spread or generalize to the entire cerebral cortex but remains circumscribed in one abnormal focus.

Complex Seizures. ith respect to seizure classification, the term "complex" signifies that consciousness has been altered. For example, a simple motor seizure may evolve into a complex partial seizure with altered consciousness. Occasionally, this state may persist for hours or days with minimal or no associated motor activity. The terminology that applies to this clinical presentation is partial complex SE. Absence SE and complex partial SE are often grouped under the term "nonconvulsive status epilepticus" and are sometimes referred to as twilight or fugue states.

Nonconvulsive Status Epilepticus. he ED physician should be aware that there is controversy surrounding the term, "nonconvulsive status epilepticus."7 In the past, this terms has been used to describe a state characterized by the absence of clinically apparent convulsive seizures in conjunction with EEG activity indicating that generalized seizures are continuing throughout the cortex.8 Typically, this is a seizure disorder of childhood and produces a characteristic EEG pattern. At times, absence seizures persist with minimal motor movements and altered consciousness for as long as hours or days. Currently, the term nonconvulsive status epilepticus should be reserved for absence SE and partial complex SE. The term "subtle status epilepticus" is more correctly used to indicate patients that have evolved from GCSE or are in a comatose state with epileptiform activity.9-13 (See Table 1.)

The Electroencephalogram.Although a detailed discussion of EEG findings in in GCSE is not necessary—and interpretation of the EEG is not within the diagnostic province of emergency physicians—it should be stressed that some patients, in whom the diagnosis of SE is not clear on the basis of clinical findings alone, may require emergent EEG testing. In this vein, a predictable evolution of EEG patterns has been observed in clinical and experimental studies of GCSE, progressing from discrete EEG seizure activity to periodic epileptiform discharges on a flat background. These changes seem correlated to bedside observations in which continuous seizure activity evolves into subtle GCSE.9,14,15

Clinical Correlations. linical management of SE has been complicated by debate surrounding its precise definition, a universal consensus of which, unfortunately, has not been forthcoming. From a clinical perspective, there is an urgent need to clarify the relationship between seizure duration and the likelihhood of self-termination. Generally speaking, seizures with a sufficiently short duration frequently cease on their own and require less aggressive emergency intervention, whereas those lasting for longer periods are more resistant to self-termination and require more aggressive pharmacological intervention.

The World Health Organization (WHO), for example, defines SE as "a condition characterized by an epileptic seizure that is sufficiently prolonged or repeated at sufficiently brief intervals so as to produce an unvarying and enduring epileptic condition."6 Typically, SE has been defined as 30 minutes of continuous seizure activity or a series of seizures without return to full consciousness between the ictal episodes.16 Unfortunately, this definition is ambiguous and imprecise and, therefore, academic clinicians in the area often use their own criteria. It should be noted that the WHO definitions are based on clinical observations rather than EEGs or any other physiologic monitoring.

In response to the WHO criteria, a number of experts have cited landmark pathophysiologic studies demonstrating that a shorter period (i.e., < 30 minutes) of seizure activity may be associated with neuronal injury, which makes seizure self-termination unlikely. They suggest that SE be defined as a continuous convulsion lasting 20 minutes or longer.17 Another consensus panel states that aggressive treatment for GCSE should be initiated when a seizure has persisted for just 10 minutes or longer and recommends that patients who are still seizing upon arrival to the ED be aggressively treated.16 Other neurological investigators have noted that the likelihood that a seizure will self-terminate diminishes after only four minutes of seizure activity.18

To clarify the relationship between seizure duration and ictal self-termination, one study evaluated a series of patients with frequent secondarily generalized tonic-clonic seizures documented by VEEG monitoring.19 This investigation revealed that the mean duration of such seizures was one minute, and seizures that ceased spontaneously usually terminated within a two-minute window. Based on these results, the authors of this study recommended intravenous anticonvulsant drug administration for generalized tonic-clonic seizures that last longer than two minutes.19 Althugh a consensus has not been achieved, emergency physicians require temporal endpoints to guide management of SE. Consequently, for purposes of this review, a duration of five minutes or greater of continuous generalized convulsive seizure activity will arbitrarily be used as the definition of GCSE and to initiate diagnostic and treatment pathways.


As would be expected, epidemiologic study of SE presents formidaable reporting problems,since a significant percentage of episodes of SE occur with an acute systemic or neurologic insult. Because in many hospitals, emergency physicians, internists, and intensivists care for patients wth SE without neurologic consultation, epidemiologic studies that depend exclusively on neurologists for identification of SE by neurologists will underestimate the prevalence of this clinical syndrome.20 Reporting is also complicated by the fact that epidemiologic investigators usually employ a seizure duration of 30 minutes to define SE, a time period that does not conform to all current expert definitions. The arbitrary 30 minute duration criterion employed in these studies clearly underestimates the clinical incidence of SE.

With these caveats and inclusionary criteria as possible disclaimers, one report estimates that SE occurs in approximately 50,000-60,000 individuals in the United States annually.21 In contrast, another estimate generated by an ongoing, prospective study suggests that 126,000-195,000 cases occur each year.20 Distribution of SE by age is bimodal, with the greatest number of cases occurring in children and in inviduals older than 60 years of age.21 The overall mortality of SE in this study was 22%, with a 3% mortality rate in the pediatric group (excluding infants less than one month of age) and 26% in the adult group. The highest mortality rates were observed in the elderly patients in the setting of hypoxic, hemorrhagic, or ischemic insults.20

Etiologies: Triggers, Toxins, and Trauma

In a very broad sense, the "imperfect rule of thirds" applies to the etiology of SE. That is, about one-third of all episodes of SE occur in patients with known seizure disorders, about one-third of episodes occur in patients as the initial manifestation of a seizure disorder, and the remainder result from a myriad of other medical, toxicologic, neurologic, and neurosurgical conditions. Specific incidence rates vary among studies, as do age groups evaluated, inclusionary criteria, and definitions of specific etiologies.

Causative Factors. n a 10-year retrospective review of patients 14 years of age or older at an urban hospital, the causes of GCSE included in descending order: discontinuation or irregularity of anticonvulsant drug regimen; alcohol-related seizures; cerebrovascular disease; drug overdose; metabolic disorders; cardiac arrest; and CNS infections, trauma, and cerebral tumors. A few patients had no discernible cause.22 A similar review performed a decade later (1980-1990) at the same institution demonstrated that these incidence rates remained essentially unchanged.23 (See Table 2.)

An initial report from another ongoing, prospective epidemiologic study in Richmond, VA, identified three major etiologies for SE in adults: low antiepileptic drug levels; remote (i.e., secondary symptomatic) etiologies associated with a previous CNS insult including stroke, hemorrhage, or tumor, and; other precipitants, including alcohol, hypoxia, and metabolic derangments. Almost half of the adult cases with SE were caused by acute or remote cerebrovascular disease.20 The prognosis in adults was linked primarily to underlying medical conditions, among them anoxia (71%) and hypoxia (53%), which were associated with mortality rates in excess of 25%. In contrast, SE resulting from low or indaequate antiepileptic drug levels was associated with only a 4% mortality.20

Lacking consistency in study design and terminology, these studies highlight one of the problems with assigning an etiology to episodes of SE. For example, anticonvulsant drug irregularity or noncompliance is the terminology used for the most common cause of SE in one study,22 anticonvulsant withdrawal is the most common assigned cause in another,23 and low antiepileptic drug levels is used in another.20 Without question, these studies are using different terminology to refer to seizures related to irregular medication use. What is clear from these studies, despite their inconsistencies, is that most patients presenting with SE who have been prescribed anticonvulsant medications have no detectable blood levels of anticonvulsants—as many as 75% in one study 23—and that acute treatment in the majority of these patients will not need to be altered because of concerns about anticonvulsant drug toxicity.23

Toxins and Antidotes. lthough an in-depth discussion of toxin-induced seizures is beyond the scope of this article, common etiologies encountered by the ED physician will be highlighted. In this regard, theophylline, isoniazid (INH), antidepressants, and sympathomimetics including cocaine are among the common ingestions that may precipitate seizures and SE. In particular, the ED physician must maintain a high index of suspicion for INH-induced seizures, inasmuch as a specific antidote is available for what might otherwise apprear to be intractable seizures: one-time administration of high-dose pyridoxine in a dose equivalent to the gram amount of isoniazid ingested is useful for preventing recurrent seizures.24 All patients with toxin-induced, GCSE require aggressive supportive care, including endotracheal intubation. In addition, repetitive charcoal dosing or hemoperfusion may be useful for some ingestions. The reader is referred to recent reviews for more information.25,26

Neurological and Systemic Consequences

of Status Epilepticus

The necessity for targeted and aggressive management of SE is best explained against the backdrop of neurological and systemic derangements that occur with prolonged seizure activity. Because GCSE can produce permanent neuronal damage, prompt and definitive treatment in the ED is mandatory. The extent to which, and the reasons for why brain tissue is damaged are not fully characterized, but extensive animal studies—as well as more limited pathologic studies in humans following GCSE—consistently demonstrate neuronal damage.27,28 In fact, the causes of GCSE-mediated CNS damage have been under investigation for more than 25 years, and, recently, a clearer pathophysiological picture has emerged.

Systemic Derangements. irst, it should be stressed that significant physiologic changes accompany GCSE, most of which can be corrected with treatment directed at the seizure disorder. Although body temperature varies among patients with SE, there is a clear tendency for hyperpyrexia, even when infectious causes are excluded. Temperature elevation is thought to follow vigorous muscle activity associated with SE, with elevations greater than 41 C commonly observed. Peripheral leukocytosis is common as CSF pleocytosis occurs, which must be distinguished from infectious causes.22,29,30 Hypertension, tachycardia, cardiac arrhythmias, and hyperglycemia are among the systemic effects caused by a marked increase in catecholamines that accompany GCSE.31

An important diagnostic clue suggesting the presence of recent seizure activity, lactic acidosis commonly occurs following a single generalized motor seizure and resolves upon termination of the seizure.32 In fact, profound metabolic acidosis in SE has been reported with a pH less than 7.1.33 Many of these systemic responses are believed to result from catecholamine surges that follow seizure actvity and which also accompany GCSE.34 As a rule, the aforementioned effects are seen early in the course.31 Increased pulmonary transcapillary fluid flux may produce pulmonary edema, although this is thought to occur from mechanisms unrelated to the increased sympathomimetic activity.35 Prolonged GCSE can produce a variety of other clinical consequences, including hypotension, hypoglycemia, rhabdomyolysis, and CNS damage, which results primarily from ischemia and other tissue-compromising factors.31,35,36

As would be expected, cerebral metabolic demand increases greatly with GCSE. Surprisingly, however, cerebral blood flow and oxygenation are thought to be preserved or even elevated, early in the course of GCSE.37 As long as compensatory hemodynamic responses are in place, cerebral oxygenation appears to be adequate to forestall neuronal damage. However, late-stage descompensation of cardiovascular hemostasis observed in GCSE, which usually is characterized by hypotension, can occur after hours of elevated catecholamine levels.37 Experts believe that systemic hypotension that results from prolonged GCSE may contribute to the late development of cerebral ischemia, a period characterized by diminished perfusion at a time during which cellular energy demands remain high.31 In other words, the transition from early to late status epilepticus is often characterized by a shift from adaptive to maladaptive mechanisms.28,26

A marker of brain injury, serum neuron-specific enolase, has been noted to be elevated in patients with SE, and may serve as a marker of brain injury.38 Morphologic changes in humans following severe SE have also been reported.39 Moreover, experiments in paralyzed and artificially ventilated animals in which many of the systemic metabolic changes seen in GCSE have been artificially induced suggest that neuronal loss after focal or generalized SE is linked to the abnormal neuronal discharges and not simply to the systemic effects of GCSE.40 In this regard, excitotoxic mechanisms involving glutamate and, possibly, aspartate have been postulated, the end result of which is calcium entry into the neuron, followed by activation of calcium-dependent enzymes that produce irreversible neuronal damage. Finally, changes in membrane lipids, free radicals, second messengers, protein kinases, and immediate-gene expression are other mechanisms postulated to cause neuronal injury. The reader is referred to recent reviews for detailed information.27,36,41-44

Clinical Presentation and Differential Diagnosis

Although GCSE can usually be recognized by the clinician without difficulty at the bedside, there is a limited, but important, differential diagnosis that must be considered in the patient with SE. (See Table 3.)For example, on occasion, psychogenic seizures may be difficult to distinguish from GCSE by appearance alone. In this regard, one study has noted that when confronted by pseudoseizures, a significant percentage of emergency physicians immediately assumed that a neurological emergency was present, and, in response, embarked on an aggressive course of pharmacotherapy. In many cases, important elements of the physical examination and historical information were neglected. In this study, unresponsiveness without movement was the most common presentation.45

Other presentations of psychogenic SE may include asynchronous extremity movement, forward pelvic thrusting, and geotropic eye movements, a physical finding in which the eyes point in a nonphysiologic manner toward the ground, whether the head is turned left or right.46,47 The gold standard test is confirmatory EEG monitoring at the time of seizure activity. Another potentially confusing presentation is seen in patients with acute injuries or insults to the CNS who may demonstrate rapidly repeating extensor or flexor posturing, which may be confused with convulsive activity.48

Diagnostic Evaluation.The initial laboratory database for patients presenting with presumed GCSE should include a CBC, blood glucose level, serum electrolytes, anticonvulsant drug levels, and urinalysis. A lumbar puncture will be necessary if meningitis is suspected. Toxicologic screening and other blood work are indicated for specific situations. A CT scan is indicated if tumor, abscess, hemorrhage, or trauma is suspected, or if the patient has no prior history of seizures.

Electroencephalogram.Although an EEG currently is not available on a STAT basis in most EDs, evidence is accumulating that rapid access to this diagnostic modality may be required on an individualized basis. For example, if the patient fails to awaken promptly after clinical termination of the seizures, or if there is evidence of subtle SE, arrangements should be made for a prompt EEG. CT scanning might also be conisidered if the history suggests an expanding intracranial lesion. The time frame for obtaining an EEG has not been formalized, but some authors recommend 20 minutes as a decision point. In one study conducted in a medical center that offered EEG monitoring to the ED on a 24-hour basis, 37% of patients referred for emergency EEG had combined EEG and clinical evidence of SE that was not tonic-clonic in nature and that would have gone undetected without EEG.11

In the recent randomized investigation evaluating treatment for GCSE, an EEG was requested as soon as a patient was entered into the study. In this trial, more than 20% of patients had ictal activity on EEG, even though tonic-clonic motor movements had ceased.49 Clearly, EEGs may be helpful in unmasking atypical presentations of GCSE. In this regard, delays in initiating treatment for GCSE have been noted in patients who arrive in the ED with an altered state of consciousness that is ascribed to a postictal state, when, in fact, these individuals are in an interictal state. In these cases, treatment was delayed until further seizures were observed in the ED. STAT EEGs in the ED are recommended for the aforementioned presentations as well as for nonconvulsive SE (absence or partial complex SE), which often goes unrecognized or is mistaken for behavioral disturbance.15,50

Current Management Standards and Sequenced Pharmacotherapy

The rationale for emergent and aggressive treatment of generalized convulsions is based on evidence reported in clinical trials and on the basis of clinical experience. (See Table 4.)First of all, that seizures beget more seizures is a well-accepted clinical axiom. Accordingly, it follows that early treatment of an seizure episode may play a role in the prevention of SE.

Despite the life- and brain-threatening consequences of GCSE, there is no universally accepted treatment algorithm. Although many recent reviews of SE have been published, each targeting a slightly different audience, 16,51-57 until recently, none represented findings gleaned from a double-blind, randomized, multicenter trial. In contrast, the recommendations for initial treatment of SE in this article are based on the definition of GCSE as presented above and on evidentiary findings reported in a recently completed, randomized, multicenter VA trial evaluating GCSE in adults.49

In this study, GCSE was defined as overt (i.e., two or more seizures without complete recovery of consciousness or continuous seizure activity for 10 minutes) or subtle (coma following seizures with epileptiform activity detected on EEG). Treatment success was judged on the basis of both clinical and EEG absence of seizure activity 20-60 minutes after medication administration. Importantly, four different intravenous drug regimens were evaluated in a blinded, randomized manner for the initial treatment of GCSE: phenytoin at 18 mg/kg; diazepam 0.15 mg/kg; phenobarbital 15 mg/kg; and lorazepam 0.1 mg/kg. Lorazepam was the most successful initial medication for halting SE (65% resolved with lorazepam alone). In addition, either lorazepam or diazepam followed by phenytoin was more effective than phenytoin alone in the initial treatment of GCSE.49

General Principles. anagement of these patients must be systematic, logical, and sequential. After the differential diagnosis of GCSE is considered, pertinent history from EMS and the patient’s family is obtained, and a brief physical examination is performed, the type of SE should be determined. (See Table 5.) nitial stabilization consists of aggressive supportive care, including airway management, rapid glucose testing, intravenous access (ideally, a freely flowing catheter into a large vein), as well as cardiac and hemodynamic monitoring. The precise etiology of the GCSE must be considered as treatment commences. Rather than enumerate a deterministic time line for treatment, all options are presented on an "ASAP" basis, although intervals between different interventions will depend on the individual clinical presentation.

Initial Pharmacotherapy.After the diagnosis of GCSE is confirmed, clinical data and expert recommendations indicate that the initial medication should be lorazepam (0.1 mg/kg, or 8 mg for an 80 kg adult) infused at 2 mg/min. Higher doses are unlikely to be effective. (See Table 6.) iazepam (0.15 mg/kg or 12 mg in an 80 kg adult) may be used if lorazepam is not available, as is sometimes the case in the prehospital setting. Some experts recommend a maximum diazepam dose of 20 mg/kg in the adult if seizures are not controlled.

If the seizures have not stopped within five minutes of benzodiazepine administration, a second drug should be administered. Current recommendations suggest that the next agent should be a phenytoin compound, either phenytoin or the pro-drug fosphenytoin. No data are available that suggest a clear preference of one drug over another. However, there are important differences between these two antiepileptic agents. Phenytoin (which contains a propylene glycol diluent and a high pH) is associated with soft tissue irritation, inflammation, and rarely, tissue loss. An equivalent dose of fosphenytoin may be administered at a more rapid rate and achieves a peak level in the therapeutic range more quickly. The clinical significance of these benefits is not entirely clear, although these features of fosphenytoin would seem to have advantages. In addition, soft tissue injury is less common with water-soluble fosphenytoin.

The favorable features of fosphenytoin must be weighed against its higher cost. In fact, access to fosphenytoin is limited at many institutions because of the cost alone (approximately $70 for 1000 mg fosphenytoin vs. $3 for generic phenytoin). Studies indicate that fosphenytoin may be administered safely at 150 phenytoin equivalents (PE)/min. The FDA has determined that fosphenytoin will be dosed in PE; a loading dose for an 80 kg person (20 PE/kg) could be given in as little as 10 minutes and 40 seconds following these guidelines.58,59 Phenytoin administered at that dose (20 mg/kg) at the maximum recommended rate of 50 mg/min would require 32 minutes to complete the infusion. Dextrose-containing solutions must be avoided with phenytoin; fosphenytoin may be administered in any common intravenous solution. Cardiac monitoring and blood pressure monitoring should be in place and the need for aggressive supportive care continually reassessed regardless of which agent is employed. Hypotension and cardiac arrhythmias may be encountered at rapid infusion rates and usually respond to slowing or halting the infusion.60

If seizures continue after a benzodiazepine has been administered and a phenytoin (or fosphenytoin) infusion has been given at a dose consistent with 20 mg(PE)/kg, many experts recommend additional phenytoin to achieve a total of 30 mg (or PE)/kg of body weight.16 Although this seems like a reasonable recommendation, there are no data to support this recommendation. However, it is consistent with a general principle to push each drug to a maximum or ideal dose before adding additional medications.

Refractory Generalized Convulsive Status Epilepticus. efractory GCSE is defined in various ways, but most often, is characterized as persistent epileptic seizures that fail to respond to recommended doses of benzodiazepine, phenytoin, and perhaps, phenobarbital.61 For the purposes of this discussion, refractory SE is defined as a failure to halt clinical seizure activity at the termination of benzodiazepine and phenytoin administration, or, persistent seizure activity (clinical or EEG) persisting 60 minutes after the onset of seizure activity and after adequate doses of a benzodiazepine and a phenytoin have been administered. The treatment of repetitive myoclonus in a comatose patient following diffuse hypoxic brain injury is controversial, since the response to antiepileptic drugs is poor as is the prognosis. The physiologic origin of this syndrome may not be cortical. The myoclonus is usually limited in duration and often is not treated as aggressively as GCSE unless an epileptiform EEG pattern is present.11,62,63

Following benzodiazepine and phenytoin administration, many options have been recommended, even though there is no data confirming the relative advantage of one additional agent over another. If seizure activity persists through and/or following the maximal therapeutic doses of a benzodiazepine and phenytoin, the likelihood of stopping the seizures with any additional agent diminshes. Additional drugs may need to be administered to the point of coma and cessation of epileptiform EEG activity. Again, there are no controlled studies or data for this group of patients and all treatments are based on consensus opinion, personal statements, case series, or isolated case reports.

Supportive care is key. The patient will most likely require intubation at this point, if not already performed, and blood pressure support with fluids or pressors is likely to be needed. Arrangements should be in progress for transfer to an ICU or other appropriate facility. EEG monitoring may become necessary and is desirable since the end point may be EEG termination of electrical seizure activity in a paralyzed patient. Of course, the etiology of the GCSE should be investigated with consideration of electrolyte abnormalities, CNS infection, or mass lesion.

Second-Line Antiepileptic Therapy.Intravenous midazolam has been recommended by some as the next choice if seizures continue following optimal dosing of lorazepam and a phenytoin. A 0.2 mg/kg loading dose is administered followed by a continuous infusion of 0.1-0.4 mg/kg/h, as determined by clinical activity and EEG monitoring. Doses may have to be increased beyond this range.64-67Blood pressure and respiratory support will be required. Intravenous diazepam infusion has also been employed for refractory GCSE.68

Third-Line Antiepileptic Therapy. n intravenous barbiturate is often recommended as the additional drug following lorazepam and a phenytoin, or it may be useful following midazolam. The choice of many intensivists is pentobarbital (5-12 mg/kg followed by 0.5-5 mg /kg/h). EEG monitoring is necessary to monitor cortical activity; intubation and pressure support will be required as well.51,64,69,70 Traditionally, phenobarbital has been the barbiturate of choice. Recommended initial loading dosage is 20 mg/kg (no faster than 100 mg/min); again, blood pressure support, ventilation support, and EEG monitoring will be necessary. Phenobarbital may be continued at an additional 10 mg/kg every 30 minutes until the seizures stop.51

Miscellaneous Antiepileptic Drugs. ntravenous propofol is reported as being successful in terminating GCSE in a handful of cases with tricyclic antidepressant overdose, stroke, acute brain injury, or encephalopathy.71,72-74 In all cases, other anticonvulsants had been administered. A bolus dose of 0.2 mg/kg may be given initially followed by infusion of 0.1-2.0 mg/kg/h with the dose titrated to clinical and EEG response. Doses as high as 7.5 mg/kg/h have been reported. Significant volume and caloric loads will accrue at this dose; expense may range to greater than $1000/d for this dose. In one study, a patient was maintained on propofol infusion as well as other anticonvulsants for eight days with eventual good recovery.74

Etomidate infusion has been used in the setting of refractory GCSE to terminate convulsive activity. An initial bolus is given (0.3 mg/kg) followed by additional bolus administered empirically based on continuing seizure activity. Once seizure cessation has been achieved, an infusion (20 mcg/kg/min) can be administered, which is then adjusted with EEG monitoring. In a study evalutating etomidate, patients had a variety of neurologic and medical conditions and were receiving other anticonvulsants.75

Anecdotal reports of intravenous lidocaine terminating refractory GCSE are scattered in the literature; undoubtedly, many unreported cases exist.73,74 The dose in an adult is typically 100 mg repeated once if necessary; other anticonvulsants typically have been administered in varying doses. Lidocaine toxicity may cause seizures and, therefore, some consider it contraindicated for use in SE.

Neuromuscular Blocking Agents.Traditional clinical axioms state that neuromuscular (NM) blocking agents are contraindicated in patients with SE. There is certainly no merit to this unqualified statement. Short-acting NM agents may be necessary for airway management and ventilation support. The use of longer-acting agents to control motor activity necessitates EEG monitoring to verify that SE with continuing epileptiform electroencephalograhic activity is not continuing.17,61,65

Alternative Treatment In The Absence of Intravenous Access.Occasionally, GCSE may be encountered in a patient in whom rapid intravenous access is not possible. A variety of approaches to this problem have been described, primarily in case reports. These options are offered for information only; no recommendations can be made regarding efficacy of one treatment over another.

Diazepam administered via the rectal route has been reported effective in a number of series involving pediatric patients.78-79 Doses of 0.5 mg/kg with a repeat dose of 0.25 mg/kg in 10 minutes is recommended, if necessary, until intravenous access can be achieved. Administration through a soft rubber catheter or using a small syringe inserted 4-5 cm inside the rectum is the described technique. In another pediatric study, a small number of children received rectal lorazepam at a dose of 0.1 mg/kg; the protocol was to repeat the dose in 15 minutes if seizures did not stop, but this did not prove necessary because all seizures were clinically terminated in the children who received one dose of lorazepam by the rectal route.81

A 10 mg IM dose of midazolam has been reported to terminate seizures in adults.82 Intranasal midazolam has recently been reported to terminate SE in two cases where intravenous access was problematic. The dose employed for an adult with a renal failure, diabetes, and a history of stroke was 10 mg; for a 2-year-old child with two episodes of GCSE, doses of 1.6 mg (repeated 5 minutes later when seizures recurred) and on another occasion 4.0 mg intranasally were employed. The child had a chronic seizure disorder and was on phenobarbital chronically.83

IM fosphenytoin may be safely administered with peak levels occurring about 30 minutes after injection. Cardiovascular monitoring is not a label requirement. The volume of injection will be large and some prefer to divide the injection into multiple sites.84 Although no reports have evaluated fosphenytoin administered by the IM route for terminating SE, this route of administration can be considered.

Disposition. atients with generalized convulsive status epilepticus should be admitted to a hospital unit where vigilant observation is available and rapid intervention is possible should seizures recur. Frequently, coexisting medical conditions or advanced supportive care will necessitate ICU admission. If SE cannot be stopped, or there is evidence or suspicion of subtle SE, the patient should receive neurologic consultation in the ED or ICU, and an EEG should be ordered on a STAT basis.


The emergency physician will maximize clinical outcomes by using an expanded definition of GCSE. In this regard, GCSE should be considered if continuous seizure activity is present for five minutes or longer and if the patient does not awaken between seizures. Historical information, including prehospital history, should be included in the timing of onset of GCSE. Experimental evidence suggests that the abnormal electrical activity and associated changes in CNS environment makes seizures more difficult to stop once duration extends beyond 5-10 minutes.

From a practical clinical perspective, GCSE is a brain-injuring emergency that demands aggressive treatment. Put simply, neuronal injury will occur if GCSE is not terminated. Overall prognosis is most closely related to the etiology of the seizure. Lorazepam is the most effective first drug in terminating SE and should be followed by a phenytoin compound. No clear data exist for additional recommendations, although a course of action has been outlined. It should be emphasized that SE may be secondary to a systemic or new CNS precipitant, in which case, treatment of other medical conditions is necessary.


. Wilson JVK, Reynolds EH. Translation and analysis of a Cuneiform text forming part of a Babylonian treatise on epilepsy. Med Hist 990;34:185-198.

2. Calmeil J. De l’epilepsie, etudieè sous le rapport de son siege et de son influence sur la production de alienation mentale. Universitè de Paris. 1824.

3. Clark LP, Prout TP. Status epilepticus: A clinical and pathological study in epilepsy. Am J Insanity1904;60:645-699.

4. Clark LP, Prout TP. Status epilepticus: a clinical and pathological study in epilepsy. Am J Insanity1904;61:81-108.

5. Gastaut H. Classification of status epilepticus. Adv Neurol1983;34:15-35.

6. Gastaut H. Dictionary of Epilepsy. Part I: Definitions. Geneva: World Health Organization; 1973.

7. Huff JS. Nonconvulsive status epilepticus in adults (letter). Am J Emerg Med 989;7:44.

8. Fagan KJ, Lee SI. Prolonged confusion following convulsions due to generalized nonconvulsive status epilepticus. Neurology 990;40:1689-1694.

9. Treiman DM, Walton NY, Kendrick C. A progressive sequence of electroencephalographic changes during generalized convulsive status epilepticus. Epilepsy Res1990;5:49-60.

10. Rutecki P, Sutula T. Status epilepticus. In: Johnson RT, Griffin JW, eds. Current Therapy in Neurologic Disease.St. Louis: Mosby-Year Book, Inc.; 1997:41-46.

11. Privitera MD, Strawsburg RH: Electroencepahlographic monitoring in the emergency department. Emerg Med Clin N Am 994; 12:1089-1100.

12. Treiman DM, DeGiorgio CM, Salisbury S, et al. Subtle generalized convulsive status epilepticus. Epilepsia1984;25:653.

13. Treiman DM. Epileptic emergencies and status epilepticus. In: Grotta JC, ed. Management of the Acutely Ill Neurological Patient.New York: Churchill Livingstone, 1993;111-121.

14. Treiman DM. Generalized convulsive status epilepticus in the adult. Epilepsia1993;34:S2-S11.

15. Jordan KG. Status epilepticus: A perspective from the neuroscience intensive care unit. Neurosurg Clin N Am1994;5:

16. Epilepsy Foundation of America. Treatment of convulsive status epilepticus: Recommendations of America’s Working Group on status epilepticus. JAMA1993;270:854-859.

17. Bleck TP, Stefan H. Status epilepticus. In: Hacke W, Hanley DF, Einhäupl KM, et al, eds. Neurocritical Care. ew York: Springer-Verlag, 1994;761-769.

18. Bleck TP. Unpublished data. 1997.

19. Theodore WH, Porter RJ, Albert P, et al. The secondarily generalized tonic-clonic seizure: A videotape analysis. Neurology 994;44:1403-1407.

20. DeLorenzo RJ, Hauser WA, Towne AR, et al. A prospective population-based epidemiological study of status epilepticus in Richmond, Virginia. Neurology 996;46:1029-1035.

21. Hauser WA. Status epilepticus: Epidemiologic considerations. Neurology 990;40:9-13.

22. Aminoff MJ, Simon RP. Status epilepticus: Causes, clinical features and consequences in 98 patients. Am J Med 980; 69:

23. Lowenstein DH, Alldredge BK. Status epilepticus at an urban public hospital in the 1980’s. Neurology 993;43:483-488.

24. Wasson S, et al. Single high-dose pyridoxine treatment for isoniazid overdose. JAMA 981;246:1102-1104.

25. Kunisaki TA, Augenstein WL. Drug-and toxin-induced seizures. Emerg Med Clin N Am 994;12:1027-1056.

26. Olsen KR, Kearney TE, Dyer JE, et al. Seizures associated with poisoning and drug overdose. Am J Emerg Med 992; 11:

27. Fountain NB, Lothman EW. Pathophysiology of status epilepticus. J Clin Neurophysiol 995;12:326-342.

28. Lothman E. The biochemical basis and pathophysiology of status epilepticus. Neurology 990;40:13-23.

29. Schmidley JW, Simon RP. Postictal pleocytosis. Ann Neurol 981;9:81-84.

30. Devinsky O, Nadi NS, Theodore WH, et al. Pleocytosis following simple, complex partial, and generalized tonic-clonic seizures. Ann Neurol 988;23:402-403.

31. Walton NY. Systemic effects of generalized convulsive status epilepticus. Epilepsia 993;34:S54-S58.

32. Orringer CE, Eustace JC, Wuncsh CD, et al. Natural history of lactic acidosis after grand mal seizures. N Engl J Med 977;297:796-799.

33. Tawadros HH, Thiagarajan D, Mukherji BB. Life-threatening lactic acidosis in status epilepticus. J Kansas Med Soc 980;81:

34. Simon RP. Changes in plasma catecholamines after tonic-clonic seizures. Neurology 984;34:255-257.

35. Simon RP. Physiologic consequences of status epilepticus. Epilepsia 985;26S:58-S66.

36. Wasterlain CG, Fujikawa DG, Penix L, et al. Pathophysiological mechanisms of brain damage from status epilepticus. Epilepsia 993;34:S37-S53.

37. Benowitz NL, Simon RP, Copeland JR: Status epilepticus: divergence of sympathetic activity and cardiovascular response. Ann Neurol1986;19:197-199.

38. DeGiorgio CM, Corrale JD, Gott PS, et al: Serum neuron-specific enolase in human status epilepticus. Neurology1995;45:

39. DeGiorgio CM, Tomiyasu U, Gott PS, et al. Hippocampal pyramidal cell loss in human status epilepticus. Epilepsia 992;33:

40. Meldrum BS. Metabolic factors during prolonged seizures and their relation to nerve cell death. Adv Neurology 983;34:

41. Young D, Dragunow M. Status epilepticus may be caused by loss of adenosine anticonvulsant mechanisms. Neuroscience1994;58:

42. Represa A, Jorquera L, Le Gal La Salle G, et al. Epilepsy-induced collateral sprouting of hippocampal mossy fibers: Does it induce the development of ectopic synapses with granule cell dendrites? Hippocampus 993;3:257-268.

43. Sloviter RS, Dean E, Sollas AL, et al. Apoptosis and necrosis induced in different hippocampal neuron populations by repetitive perforant path stimulation in the rat. J Comp Neurol 996;366:516-533.

44. Lothman EW, Bertram EH. Epileptogenic effects of status epilepticus. Epilepsia 993;34:S59-S70.

45. Young GB, Gilbert JJ, Zochodne DW. The significance of myoclonic status epilepticus in postanoxic coma. Neurology1990;40:1843-1848.

46. Jagoda A, Richey-Klein V, Riggio S. Psychogenic status epilepticus. J Emerg Med 995;13:31-35.

47. Chabolla DR, Krahn LE, Elson LS, et al. Psychogenic nonepileptic seizures. Mayo Clin Proc 996;71:493-500.

48. Haines SJ. Decerebrate posturing misinterpreted as seizure activity. Am J Emerg Med 988;6:173-177.

49. Treiman DM, Meyers PD, Walton NY, et al. Treatment of generalized convulsive status epilepticus: A randomized double-blind comparison of four intravenous regimens. N Engl J Med 997; (in press).

50. Kaplan PW. Nonconvulsive status epilepticus in the emergency room. Epilepsia 996;37:643-650.

51. Cascino GD. Generalized convulsive status epilepticus. Mayo Clin Proc 996;71:787-792.

52. Delgado-Escueta AV, Bajorek JG. Status epilepticus: mechanisms of brain damage and rational management. Epilepsia 982;23:S29-S41.

53. Walsh GO, Delagado-Escueta AV. Status epilepticus. Neurolgic Clin 993;11:835-855.

54. Shepherd SM: Management of status epilepticus. Emerg Med Clin N Am 994;12:941-961.

55. Chang CWJ, Bleck TP. Status epilepticus. Neurologic Clin1995;13:529-548.

56. Ramsey RE. Treatment of status epilepticus. Epilepsia1993; 34:S71-S81.

57. Runge JW, Allen FH. Emergency treatment of status epilepticus. Neurology1996;46(supp 1):S20-S23.

58. Knapp LE, Kugler AR, Eldon MA. Fosphenytoin: pharmacokinetics and administration. Emerg Med 996;28:9-16.

59. Ramsey RE, DeToledo J. Intravenous administration of fosphenytoin: options for management of seizures. Neurology 996;46:

60. Earnest MP, Marx JA, Drury LR. Complications of intravenous phenytoin for acute treatment of seizures. JAMA1983;249:

61. Jagoda A, Riggio S. Refractory status epilepticus in adults. Ann Emerg Med 993;22:1337-1348.

. Leis AA, Ross MA, Summers AK. Psychogenic seizures: Ictal characteristics and diagnostic pitfalls. Neurology 992;42:95-99.

. Kumar A, Bleck T. Intravenous midazolam for the treatment of refractory status epilepticus. Crit Care Med1992;20:483-488.

64. Bleck TP. Refractory status epilepticus. Neurol Chronicle 992;2:1-4.

65. Gavrilescu T, Otis JAD. Efficacy of midazolam in pentobarbital-resistant status epilepticus. Neurology 995;45:A346.

. Parent JM. Treatment of refractory status epilepticus with continuous infusion of midazolam. Neurology 994;44:1837-1840.

. Bell HE, Bertino JS. Constant diazepam infusion in the treatment of continuous seizure activity. Drug Intell Clin Pharm1984;18: 965-970.

68. Lowenstein DH, Aminoff MJ, Simon RJ. Barbiturate anesthesia in the treatment of status epilepticus: Clinical experience in 14 patients. Neurology1988;38:395-400.

69. Rashkin MC, Youngs C, Penovich P. Pentobarbital treatment of refractory status epilepticus. Neurology1987;37:500-503.

70. Borgeat A, Wilder-Smith OH, Jallon P, et al. Propofol in the management of refractory status epilepticus. Intensive Care Med1994;20:148-149.

71. Pitt-Miller P, Elcock BJ, Maharaj M. The management of status epilepticus with a continuous propofol infusion. Anesth Analg1994;78:1193-1194.

72. Merigian KS, Browning RG, Leeper KV. Successful treatment of amoxapine-induced refractory status epilepticus with propofol (Diprivan). Acad Emerg Med1995;2:128-133.

73. Huff JS, Bleck TP. Propofol and midazolam in status epilepticus (letter). Acad Emerg Med1996;3:179.

74. Yeoman P, Hutchinson A, Byrne A, et al. Etomidate infusions for the control of refractory status epilepticus. Intensive Care Med 989;15:225-259.

75. Pascual J, Sedano MJ, Polo JM, et al. Intravenous lidocaine for status epilepticus. Epilepsia1988;29:584-589.

76. Aggarwal P, Wali JP: Lidocaine in refractory status epilepticus: a forgotten drug in the emergency department. Am J Emerg Med 993;11:243-244.

77. Albano A, Reisdorff EJ, Wiegenstein JG. Rectal diazepam in pediatric status epilepticus. Am J EmergMed 989;70:168-172.

78. Seigler RS. The administration of rectal diazepam for acute management of seizures. J Emerg Med 990;8:155-159.

79. Dieckmann RA. Rectal diazepam for prehospital pediatric status epilepticus. Ann Emerg Med1994;23:216-224.

80. Appleton R, Sweeney A, Choonara I, et al. Lorazepam versus diazepam in the acute treatment of epileptic seizures and status epilepticus. Dev Med Child Neuro1995;37:682-688.

81. Mayhue FE. IM midazolam for status epilepticus in the emergency department. Ann Emerg Med 988;17:643-645.

82. Kendall JL, Reynolds MRG. Intranasal midazolam in patients with status epilepticus. Ann Emerg Med 997;29:415-417.

83. Uthman BM, Wilder BJ, Ramsey RE. Intramuscular use of fosphenytoin: an overview. Neurology1996;46:S24-S28.

Physician CME Questions

. Neuronal damage from status epilepticus is thought to result
primarily from:

A. hypoxia.

B hypotension.

C. abnormal CNS electrical activity.

D. acidosis.

2. Population studies reveal that the most common cause of status epilepticus in the adult population is:

A. alcohol-related

B. anticonvulsant discontinuation

C. meningitis

D. acute or remote stroke

3. Studies show that the most effective anticonvulsant in terminating status epilepticus that is refractory to optimal doses of lorazepam and phenytoin is:

A. phenobarbital

B. midazolam

C. propofol

D. No clear data exist for next choice

4. Indications for request for stat EEG in treating a patient with seizures include:

A. suspicion of subtle generalized convulsive status epilepticus in an unresponsive patient with small twitching movements.

B. failure of a patient to improve level of consciousness 20 minutes following clinical termination of seizures

C. a patient with generalized convulsive status epilepticus when treatment included airway management, intubation, mechanical ventilation, and maintenance of neuromuscular blockade for adequate ventilation

D. All of the above

5. Subtle generalized status epilepticus may be used to best describe which of the following?

A. A comatose patient who recently had convulsive seizures and now has only intermittent arm twitching, but generalized seizure activity continues on EEG

B. Partial complex status epilepticus or absence status epilepticus

C. The quiet period between seizures

D. Psychogenic seizures

6. Nonconvulsive status epilepticus is a term best employed for?

A. A comatose patient who recently had convulsive seizures and now has only intermittent arm twitching, but generalized seizure activity continues on EEG

B. Partial complex status epilepticus or absence status epilepticus

C. The quiet period between seizures

D. Psychogenic seizures

. The following drugs may be useful in treating refractory generalized convulsive status epilepticus:

A. propofol.

B. midazolam.

C. pentobarbital.

D. All of the above

. A clinical definition for GCSE includes continuous generalized convulsions lasting for:

A. two hours.

B. one hour.

C. 30 minutes.

D. five minutes.