Status epilepticus: Why, what, and howPP Nair, J Kalita, UK Misra
Department of Neurology, Sanjay Gandhi PGIMS, Lucknow, Uttar Pradesh, India
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0022-3859.81807
Source of Support: None, Conflict of Interest: None
Status epilepticus (SE) is an important neurological emergency with high mortality and morbidity. The first official definition of SE was the product of 10 th Marseilles colloquium held in 1962 which was accepted by International League Against Epilepsy in 1964. There are as many types of SE as of seizures. SE is supposed to result from failure of normal mechanisms that terminate an isolated seizure. In half of the cases, there is no history of epilepsy and SE is precipitated by some intercurrent infection. In children, it is often infection, whereas in adults, the major causes are stroke, hypoxia, metabolic derangements, and alcohol intoxication or drug withdrawal. The treatment of SE aims at termination of SE, prevention of seizure recurrence, management of precipitating causes, and the management of complications. The extent of investigations done should be based on the clinical picture and cost benefit analysis. The first line antiepileptic drugs (AED) for SE include benzodiazepines, phenytoin, phosphenytoin, and sodium valproate. Mortality of SE ranges between 7 and 39% and depends on underlying cause and response to AEDs.
Keywords: Electroencephalography, status epilepticus, etiology, treatment
Status epilepticus (SE) is an important neurological emergency with high mortality and morbidity. The incidence of SE varies depending on the population studied. ,,,, In India, the incidence of SE is likely to be high due to high prevalence of epilepsy, central nervous system (CNS) infections, and treatment gap. Considering the 18.3 per 100 000 population incidence of SE in USA and extrapolating it to India, it is estimated that the incidence of convulsive SE is likely to exceed 183 000, resulting in heavy burden on the society. This review focuses on the management of SE with theoretic and clinical considerations involved in choosing antiepileptic drugs (AED).
The annual incidence of SE in US is 18.3 to 41 per 100 000 population and in Europe, 10.3 to 17.1 per 100 000 population and that of nonconvulsive SE is 2 to 8 per 100 000. ,,,,, SE incidence is probably higher in poorer populations.  In half of the cases, there is no history of epilepsy and SE is precipitated by some intercurrent infection. In children, the etiology of SE is usually infection and in adults, the major causes are stroke, hypoxia, metabolic derangements, and alcohol intoxication or drug withdrawal.  In patients with epilepsy, SE is often precipitated by drug withdrawal due to noncompliance with antiepileptic drugs. SE of focal onset accounts for the majority of cases of SE. An epidemiological study on SE reported that 69% of episodes in adults and 64% of episodes in children were of focal onset; these resulted in secondary generalization in 43% in the adults and 36% in children. The incidence of SE was bimodally distributed, occurring most frequently in the first year of life and after 60 years of age. Males are affected more frequently than the females, which is partly attributed to lower seizure threshold in males compared with females. , Patients above 60 years of age are at the highest risk of developing SE, with an incidence of 86 per 1 000 000 population per year. , A hospital-based study from India, however, revealed peak incidence during fourth and fifth decades. In about 53.8% of patients, SE was due to either CNS or systemic infections.  However, etiology may vary depending on the population studied. Studies from intensive care units (ICUs) have demonstrated that SE is a major cause of coma. , Fourteen percent of patients with convulsive SE remained in altered sensorium after control of convulsions; on Electroencephalography (EEG), they were found to have nonconvulsive SE, thus highlighting the importance of either regaining the consciousness or absence of epileptiform discharges on EEG. 
Epilepsy is known since antiquity. The term epilepsy is derived from Greek word Epilambenein meaning "to seize" or "to attack." In Ayurvedic literature, epilepsy is mentioned as Apasmara (apa - negation or loss, smara - recollection or consciousness) and it is mentioned in Chinese literature as early as 770 to 221 BC. At around 400 BC, epilepsy was described as "sacred disease"; its first description is in the writings of Babylon during 300 to 600 BC.  The term SE first appeared in English literature in the translation of the lecture of Desire Bournville who gave the detailed clinical description.  The term has evolved from the phrase "etat de mal" which was a slang used by epileptic patients in Paris.
The first definition of SE probably comes from Clark and Prout: "maximal development of epilepsy in which seizures are so frequent that coma and exhaustion are continuous between the seizures."  In his text book of neurology, Kinner Wilson described SE as the severest form of epilepsy.  The first official definition of SE was the product of 10 th Marseilles colloquium held in 1962 which was accepted by ILAE (International League Against Epilepsy) in 1964.  SE is a seizure that persists for a significant length of time or is repeated frequently enough to produce a fixed and enduring epileptic condition. This definition was retained in the revised classification published in 1971 and was slightly modified a decade later as "a seizure that persists for sufficient length of time or repeated frequently enough that recovery between attacks does not occur."  The official definition did not specify the duration of seizures. Over the years, various investigators have tried to define the duration of 30 minutes. The basis for this lies in the animal studies during 1970s and 1980s, which revealed significant brain damage after 30 minutes of seizure, despite control of blood pressure (BP), respiration, and temperature. ,,,
The mechanism of neuronal injury in human beings is complex and includes factors other than the duration of SE such as systemic derangements due to persistence of seizures.  Moreover, the duration and therapeutic response of SE may depend upon the underlying etiology.  It has been observed that spontaneous cessation of generalized convulsive seizures is unlikely after 5 minutes  ; hence, for defining SE, duration as short as 5 minutes has been suggested.  The duration of typical isolated seizure is easy to define. In a study on thousands of patients with generalized tonic clonic seizures, it was found that tonic phase lasted for 1 to 20 seconds; clonic phase, 30 seconds; and postictal tonic contraction, up to 4 minutes.  Using Video EEG analysis of 47 patients, the mean duration of tonic clonic phase of seizure was 62 seconds (range, 16-108 seconds).  In a patient with continuous seizures, it is unreasonable and impractical to wait for 30 minutes before treatment is initiated. In the prehospital treatment of SE, the duration of SE was defined as seizures persisting for more than 5 minutes.  In Veterans Affairs cooperative trial on treatment of generalized convulsive SE, the duration of 10 minute was used to define SE. 
Operational definition for generalized SE in adults and older children (>5 years) refers to >5 minutes of continuous seizure or 2 or more discrete seizures between which there is incomplete recovery of consciousness. Mechanistic definition of SE refers to a condition in which there is failure of normal factors that serve to terminate a typical generalized tonic clonic seizure.  Some studies have compared seizures of 10 to 29 minutes duration with seizures lasting >30 minutes and found that half of the seizures of 10 to 29 minutes stop spontaneously even without AED. 
Classification of SE is necessary for the appropriate management. There are several schemes for classifying SE. Seizure type by the ILAE  and its dichotomy of focal and generalized onset are also used to categorize SE on the basis of the assumption that there is a status epilepticus equivalent for every seizure type. Using both clinical and EEG criteria, SE is subdivided into generalized convulsive status (tonic-clonic, tonic, clonic, and myoclonic), generalized nonconvulsive (absence) status, elementary partial status (with several subtypes), and complex partial status. SE can also be categorized according to patients' age into SE confined to early childhood, later childhood, childhood and adult life, and confined to adult life [Table 1].  Epilepsy Research Foundation Workshop defines nonconvulsive SE as a term used to denote a range of conditions in which electrographic seizure activity is prolonged and results in nonconvulsive clinical symptoms.  A convenient formulation of non convulsive status epilepticus (NCSE) is the alteration of consciousness or behavior from baseline state for at least 30 minutes without convulsive movements, and the presence of one or more of the following epileptiform patterns:
A simple classification of NCSE may be into (1) complex partial SE (CPSE) and (2) generalized nonconvulsive SE. 
However, NCSE is best subdivided by age, and further subdivided into the forms of NCSE seen in the epileptic encephalopathies, acute brain injury, and those with a history of epilepsy without encephalopathy and some boundary syndromes. 
SE is thought to result from failure of mechanisms that normally terminate an isolated seizure. This failure can arise from abnormal persistence of excessive excitation or ineffective recruitment of inhibition. Experimental studies suggest that there is an induction of reverberating seizure activity between hippocampus and parahippocampal structures and seizure progresses through a sequence of distinct electrophysiological changes.  The proposed mechanisms of SE are as follows:
Excitatory amino acids are thought to be involved in the pathogenesis of SE, which is supported by the fact that prolonged SE occurs following ingestion of mussels containing domoic acid, an analogue of glutamate.  In experimental SE, seizures rapidly become self-sustaining and continue long after the withdrawal of the epileptogenic stimulus whether chemical or electrical. ,,,,, Vicedomini and Nadler in their self-sustaining model of SE showed that a series of ten after-discharges were sufficient to trigger self-sustaining SE and once self-sustaining SE is established, it is easily stopped only by drugs which directly or indirectly inhibit glutamatergic neurotransmission. , Barbiturates and other GABAergic drugs never become totally ineffective but lose potency and may require high doses resulting in cardiovascular depression. 
It is suggested that in the first few seconds, there is protein phosphorylation, opening and closure of the ionic channels, release of neurotransmitters and modulators, and receptor desensitization. In seconds to minutes, receptor trafficking results in the movement of the existing receptors from the synaptic membrane into the endosomes, or their mobilization from storage sites to the synaptic membrane. This process drastically changes excitability by altering the number of inhibitory and excitatory receptors available in the synaptic cleft.  In minutes to hours time range, there are plastic changes in neuropeptide modulators which are often maladaptive, leading to a state of raised excitability. This is supported by immunocytochemical and confocal microscopy studies which have revealed a decrease in the number of GABA-A subunits present on the synaptic membrane and an increase inside the cell.  Endocytosis of the GABA-A receptors might partly explain the failure of GABA-A inhibition and the progressive pharmacoresistance to benzodiazepines (BDZ) as self-sustaining SE proceeds. ,,, Other mechanisms like accumulation of intracellular chloride or higher bicarbonate might also play a part in loss of GABA-mediated inhibition. , Interestingly, extrasynaptic GABA-A receptors do not endocytose, raising the possibility that stimulation of those extrasynaptic receptors might be useful in the treatment of SE. At the same time, a-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid and N-methyl D-aspartate (NMDA) receptor subunits move to the synaptic membrane where they form additional excitatory receptors [Figure 1]. This change further increases excitability during uncontrolled seizures.  Neuronal damage in SE result from sustained NMDA-mediated neuronal stimulation which leads to apoptosis.  When these neuronal cells are depolarized, the Mg 2+ ions blocking the channel diffuses outward, allowing sodium ions and Ca 2+ to flood the cell, resulting in a cascade of Ca +2 -mediated cytotoxic events, leading to neuronal injury, cell lysis, and cell death. The cell destruction triggered in this manner may be reversible if SE is terminated within the first hour. It has been shown that heat-shock protein (72 kDa, HSP-72) is induced in some neurons in SE and that it may have a neuroprotective role. 
In experimental studies following generalized tonic clonic SE, brain damage has been demonstrated; hyperthermia was demonstrated to result in hippocampal damage.  In animal studies, despite optimal control of BP, temperature, partial pressure of oxygen, and partial pressure of carbon di oxide, SE resulted in neuronal damage in substantia nigra pars reticularis after 30 minutes and in third and fourth layer of neocortex as well as in CA1 and CA4 pyramidal neurons of hippocampi after 45 to 120 minutes. 
In the early stage of SE, there is massive release of catecholamines  which result in tachycardia, arrhythmias, high systemic, pulmonary, and left atrial pressure, and occasionally pulmonary edema. ,,,, Blood glucose is elevated.  Respiratory failure and lactic acidosis result in metabolic acidosis. Hyperpyrexia with increased white cell counts in SE may be mistaken for infection.  Low-grade cerebrospinal fluid (CSF) pleocytosis may follow SE. , BP declines 15 to 30 minutes after SE and may be markedly low after 2 hours of continuous seizure activity. Besides, there may be hypoglycemia.  Renal failure may occur because of rhabdomyolysis and myoglobinuria.  In the initial phase of SE, cerebral blood flow increases to meet the elevated demands, thereby increasing the intracranial pressure.  Later, cerebral edema ensues; in focal seizures, focal brain edema may be present. In general, the physiological changes in SE can be divided into two phases. In phase 1, compensatory mechanisms prevent cerebral damage. In phase 2, these mechanisms fail and there is an increasing risk of cerebral damage as the status progresses. The transition from phase 1 to phase 2 occurs after about 30 to 60 minutes of continuous seizures. These physiological changes do not necessarily occur in all the patients. The type and extent of the changes depend on etiology, clinical circumstances, and the methods of treatment used. 
The first step in managing SE is to ascertain that the patient has seizures. A single generalized seizure with complete recovery may not require treatment. Once diagnosis of SE is made, the treatment should be started immediately. Though the duration of seizure for diagnosis of SE is controversial, any patient who arrives in emergency and is convulsing would require aggressive treatment. The treatment of SE should proceed with the following 4 aims:
Maintaining the patient's airway and oxygenation is most important step in the management of SE. Intubation is not required if airway is patent. BP and pulse should be checked. In patients with history of seizures, one should check if patient has missed the medication recently. A screening neurological examination is necessary to check for focal intracranial lesion.
One should promptly obtain IV access and send the blood for blood sugar, serum electrolyte, and AED levels (if the patient was on treatment with AED), intoxicants screen (if no cause is found), and blood counts wherever indicated. Isotonic saline infusion should be started; however, if hypoglycemia is suspected, 100 mg thiamine followed by 50 ml 25% glucose solution is infused. Thiamine is used to avert Wernicke's encephalopathy in a susceptible patient. Blood gases should be determined to ensure adequate oxygenation.
Acidosis, hypoxia, and hyperpyrexia are commonly associated and generally resolve with control of SE. Pharmacological treatment of SE is initiated according to the accepted guidelines. Imaging (computed tomography [CT] scan) is recommended after stabilization of patient. If imaging is normal, CSF examination is performed to rule out CNS infection.
Utility of different tests in SE
A number of tests such as lumbar puncture, metabolic and toxin screening, genetic studies, and imaging studies are recommended in the management of SE. These tests should not be blindly recommended but should be obtained depending on the clinical need.
EEG: In a patient with convulsive SE, where the diagnosis is clinically apparent, performing EEG may be difficult and may not be required. However, it may be useful in a patient with SE who remains comatose, despite control of convulsions (coma in such patients may be due to drug overdose or ongoing seizures) and to rule out nonconvulsive SE.
The following four patterns of EEG changes have been reported in SE: discrete, merging, continuous, and periodic lateralizing epileptiform discharges. In human beings, however, the sequential transition of EEG is not always found. In a study on 70 patients with SE, EEG revealed discrete pattern in 4, merging in 5, continuous in 5, periodic in 6, multiple patterns in 2, and slowing in 29 patients. Clinical seizure recurred within 24 hours in 38 patients, 15 of them had epileptiform discharges at 1 hour EEG.  Early EEG after 1 hour of control of convulsive SE predicted seizure recurrence. EEG can help to confirm that the episode of SE has ended, especially when there are doubts about ongoing subtle seizures. EEG monitoring of patients up to 24 hour after clinical signs of SE had ended, revealed that nearly half of their patients continued to demonstrate electrographic seizures that often had no clinical correlation. EEG monitoring of SE patients after clinical control of SE is thus regarded important for optimal management.  The patients with SE who fail to recover rapidly and completely should be monitored by EEG for at least 24 hour to ensure that the recurrent and/or subtle SE are not missed. Monitoring is also advised if there are periodic discharges on EEG in SE patients who are in altered sensorium, though they do not have obvious seizures. Periodic discharges in such a situation predict SE recurrence. In comatose patients without clinical signs of seizure activity, up to 8% met criteria for nonconvulsive SE.  The duration and delay in diagnosis of nonconvulsive SE was strongly linked to mortality in another ICU-based study.  American Academy of Neurology has made specific recommendation regarding various investigations in SE, including EEG in children.  An EEG may be considered in a child presenting with new onset SE as it may determine whether there are focal or generalized abnormalities that may influence diagnostic and treatment decisions (Level C, class III evidence) ,,,,, (Levels and classes of evidence explained in [Table 2] and [Table 3]).  Although NCSE occurs in children who present with SE, there are insufficient data to support or refute recommendations regarding whether an EEG should be obtained to establish this diagnosis (Level U). An EEG may be considered in a child presenting with SE if the diagnosis of pseudo SE is suspected (Level C, class III evidence). 
During SE, neuroimaging may be used to exclude other neurologic conditions. Therefore, it is important to identify the neuroimaging features that are associated with SE. In a study on three patients, the magnetic resonance imaging (MRI) and CT findings during partial SE mimicked those of acute ischemic stroke. Diffusion weighted imaging (DWI) and T2-weighted MRI showed cortical hyperintensity with a corresponding low apparent diffusion coefficient, and CT showed an area of decreased attenuation with effacement of sulci and loss of gray-white differentiation. However, the lesions did not respect vascular territories; there was increased signal of the ipsilateral middle cerebral artery on MRA, and leptomeningeal enhancement appeared on postcontrast MRI. On follow-up imaging, the abnormalities had resolved.  Several case studies have demonstrated reduced Apparent Diffusion Co efficients (ADCs) in patients with focal SE. ,, Combined use of DWI and perfusion imaging in a series of 10 patients presenting with CPSE showed regional hyperintensity on DWI, and a reduction of the ADC in (i) hippocampal and pulvinar region of the thalamus in six, (ii) pulvinar and cortical regions in 2, (iii) only the hippocampal region in 1, and (iv) hippocampal, pulvinar, and cortex in 1. In all the patients, a close spatial correlation of focal hyperperfusion with areas of ADC/DWI changes was present. Follow-up MRI examinations showed partial or complete resolution of diffusion and perfusion abnormalities, depending on the length of the follow-up interval.  DWI and T2-weighted changes were seen throughout the cerebral cortex, hippocampus, amygdale, and medial thalamus in rat brains after 4 hours of SE.  According to the guidelines, neuroimaging may be considered for the evaluation of the child with SE if there are clinical indications or if the etiology is unknown (Level C, class III evidence). ,,,, If neuroimaging is done, it should only be done after the child is appropriately stabilized and the seizure activity controlled. There is insufficient evidence to support or refute recommending routine neuroimaging in SE (Level U).
Recommendations for other investigations
There are insufficient data to support or refute whether blood cultures should be done on a routine basis in children in whom there is no clinical suspicion of infection (Level U).
There are insufficient data to support or refute whether LP should be done on a routine basis in children in whom there is no clinical suspicion of a CNS infection (Level U).
AED levels should be considered when a child with epilepsy on AED prophylaxis develops SE (Level B, class II and III evidence). ,,,,,,
Toxicology testing may be considered in children with SE when no apparent etiology is immediately identified. It should be noted that a specific serum toxicology level is required, rather than simply urine toxicology screening (class III). ,,,,,,,,
Studies for inborn errors of metabolism may be considered when the initial evaluation reveals no etiology, especially if there is a preceding history suggestive of a metabolic disorder (Level C, class III evidence). ,,,,,,,
There are insufficient data to support or refute whether genetic testing (chromosomal or molecular studies) should be done routinely in children with SE (Level U).
For routine practice, one can follow some simple measures; CSF examination is indicated if CNS infection is suspected. AED levels should be tested if the patient was on AED and there is a likelihood of drug default. EEG is useful in the first seizure and for the diagnosis of nonconvulsive SE. Genetic and metabolic studies are recommended if there is family history of inborn errors of metabolism or clinical suggestion of a hereditary metabolic or genetic disease. Toxicology tests are recommended if no other cause of SE could be found out.
Pharmacologic therapy of SE
The goal of pharmacologic therapy is rapid termination of clinical and electrical seizures. The drugs for controlling SE should be administered parenterally. Midazolam and paraldehyde can be given intramuscularly; diazepam, midazolam, and paraldehyde by the rectal route; but all others must be given by IV injection [Table 4].
In a survey among neurologists in USA, the first choice for the treatment of generalized convulsive SE was lorazepam (76%) followed by fosphenytoin or phenytoin (PHT) (95%) if the first line therapy fails. When both failed, 43% physician would choose phenobarbital, 19% would use either infusion phenobarbitone, midazolam, or propofol, and 16% would choose IV sodium valproate (SVA).  In VA co-operative study, the success rate of different drugs was 64.9% for lorazepam, 58.2% for phenobarbitone, 55.8% for PHT/diazepam, and 43.6% for PHT alone. In this study, lorazepam demonstrated statistically significant advantage over PHT (P = 0.02). There was no significant difference among other agents. AEDs had been categorized into first line, 2 nd line, and 3 rd line. Aggregate response to 2 nd line was 7% and 3rd line was 2.3%.  Earlier, 85% success rate with lorazepam has been reported in convulsive SE. 
The benzodiazepines (BDZ) are among the most effective AEDs in the treatment of SE. The commonly used BDZ are diazepam, lorazepam, and midazolam. This class of drugs enhances GABA and barbiturate receptor complex.
Diazepam is one of the drugs of choice for first line management of SE, as evidenced by randomized controlled trials (RCTs). ,, It enters CNS rapidly because of high lipid solubility but after 15 to 20 minutes is redistributed in the body, thus reducing its clinical efficacy. , Sufficient cerebral levels are reached within one minute of a standard IV injection and rectal administration produces peak levels at about 20 minutes. Despite this fast distribution half life, the elimination half life is about 24 hours; the sedative effect, therefore, could accumulate with repeated administration. Diazepam in dose of 5 to 10 mg/min controls seizures in 75% patients. ,, Adverse effects include respiratory suppression, hypotension, and sedation. Diazepam can be administered by intramuscular, IV, and per rectal routes. Bolus IV doses of diazepam should be given in an undiluted form at a rate not exceeding 2 to 5 mg/min. The adult bolus IV or rectal dose in SE is 10 to 20 mg; additional 10 mg doses can be given at 15 minute intervals, to a maximum of 40 mg. In children, the equivalent bolus dose is 0.2 to 0.3 mg/kg.
Lorazepam has emerged as preferred BDZ for management of SE based on RCT.  Lorazepam is less lipid soluble than diazepam with distribution half life of 2 to 3 hours compared with 15 minutes for diazepam. Hence, it has a longer duration of action. Lorazepam binds more tightly to GABA receptors compared with diazepam, resulting in longer duration of action. The usual dose of lorazepam is 4 to 8 mg and the anticonvulsant effect lasts 6 to 12 hours. This can be repeated once after 20 minutes if no effect has been observed. Its main disadvantage is the rapid development of tolerance, repeated doses being much less effective, and the drug has no place as long-term therapy. It has a broad spectrum of activity and terminates seizures in 75 to 80% cases.  Sudden hypotension or respiratory collapse is less likely because of its relative lipid-insolubility and the lack of accumulation after single bolus injections.
Midazolam is commonly used as first choice BDZ for treating SE in Europe. ,,,, It can be given by intramuscular injection, as well as by the rectal or IV routes. Midazolam has short-lived action and there is tendency to relapse following a single bolus injection. It is cleared from the body faster than diazepam with less tendency to accumulate. Midazolam is usually given intramuscularly or rectally in premonitory status, in a dose of 5 to 10 mg (in children, 0.15-0.3 mg/kg), which can be repeated once after 15 minutes. 5 to 10 mg IV bolus injection can also be given (repeated to a maximum of 0.3 mg/kg in adults). There is limited experience of an IV infusion.
Which of the BDZ to choose: Comparison of IV diazepam (10 mg) and lorazepam (4 mg) as first line treatment of SE in a randomized double blind trial on 78 patients did not show any difference in efficacy or latency of action, but the number of patients was too small to define true significance. Seizures were terminated 58% in diazepam and 78% in lorazepam group at latency of 2 and 3 minutes, respectively. 
Phenytoin (PHT) is one of the most effective drugs in treating SE. PHT causes relatively little respiratory or cerebral depression, although hypotension is more common. The initial infusion of PHT takes 20 to 30 minutes in an adult, and the onset of action is slow. The usual PHT solutions have a pH of 12, and if added to large volumes of fluid at lower than physiological pH (5% glucose), precipitation may occur in the bag or tubing; normal saline is safer. The rate of infusion of PHT solution should not exceed 50 mg/min, and it is prudent to reduce this to 20 to 30 mg/min in the elderly. The adult dose is 15 to 18 mg/kg and main advantage of PHT is lack of sedative side effect; however, a number of potential serious side effects may occur. Cardiac arrhythmia and hypotension have been reported in individuals above 40 year of age. It is likely that these side effects are due to rapid administration and propylene glycol which is used as its diluent. In addition, local irritation and dizziness may accompany IV administration.
Fosphenytoin was approved by Food and Drug Administration (FDA) in 1996 for the treatment of SE. It is a water-soluble prodrug of PHT which is completely converted into PHT in 8 to 15 minutes following IV administration; hence, the side effects related to propylene glycol are avoided.  Fosphenytoin is 100% bioavailable and is rapidly and completely converted to PHT in adults after IV and intramuscular administration. , Conversion is mediated by both alkaline and acid phosphatases present ubiquitously on cell surfaces of vascularized tissue  and the rate of conversion is independent of age, race, and gender. In patients with impaired renal and hepatic function, the conversion is actually rapid as a result of self displacement from plasma proteins and increased clearance to PHT; therefore, 10 to 20% dose reduction may be considered in these patients and also in patients with hypoalbuminemia. Fosphenytoin, like PHT, is used for treating acute partial and generalized tonic clinic seizures. It is metabolized by liver and its half life is 14 hours. Therapeutic plasma PHT concentration is achieved faster with fosphenytoin than PHT. This may be related to several factors such as faster rate of infusion, rapid conversion, and displacement of PHT by fosphenytoin from the plasma protein binding site. 1.5 mg fosphenytoin is equivalent to 1 mg of PHT. The dose concentration and infusion rates are expressed as PHT equivalent (PE). The initial dose of fosphenytoin is 15 to 20 mg PE/kg and can be infused 150 mg PE/min (3 times faster than IV PHT). Fosphenytoin can be given intramuscularly, but therapeutic concentration is not achieved till 30 minutes. 
Phenobarbital is used in SE when BDZ and PHT have failed to control seizures. Initial loading dose is 15 to 20 mg/kg. Since high dose of phenobarbital is sedating, airway protection is important to prevent aspiration. IV phenobarbital is also associated with hypotension. It may take about 30 minutes for the therapeutic dose to be infused. It is diluted in polyethylene glycol which results in complications such as renal failure, myocardial depression, and seizures. These limitations restrict the use of phenobarbital in patients who have not responded to other drugs.
FDA approved sodium valproate (SVA) for the use in SE in 1997. Parenteral SVA is used primarily for rapid loading and when oral therapy is not possible. It has a broad spectrum and may also be useful in absence and myoclonic SE. In a study on 60 SE patients, majority of whom had convulsive SE, seizures were aborted by SVA in 66% and by PHT in 42%. SVA as a second choice was effective in 79% and PHT in 25% patients, suggesting the higher efficacy of SVA compared with PHT.  Patients not responding to 2 first line AEDs may be given parenteral SVA, especially in the setting where intubation and artificial ventilation are not feasible. It is given as a bolus of 30 mg/kg and can be infused at a rate of 3 to 6 mg/kg/min. 
Paraldehyde ,, is a drug still widely used in the treatment of SE, especially as an alternative to diazepam in early stage of SE, where IV administration is difficult or conventional AEDs are contraindicated or proved ineffective. The drug is given rectally or intramuscularly and absorption by both routes is fast and complete. The onset of action is rapid and paraldehyde is effective for many hours. Seizures tend to recur after initial control. Toxicity is unusual provided, the correct dose of paraldehyde is not exceeded, it is freshly reconstituted and is not decomposed. Inappropriately diluted or decomposed paraldehyde is highly toxic by any route of administration. The intramuscular injection must be deep into the muscle (usually gluteal), well away from the sciatic nerve. IV infusion is now rarely recommended. Paraldehyde also reacts with rubber and plastic, but a plastic syringe is acceptable if given rapidly after drawing the solution. It can be given at a dose of 10 to 20 ml of 50% solution rectally or intramuscularly (children, 0.07-0.35 ml/kg), which can be once repeated after 15 to 30 minutes. For rectal or intramuscular administration, it is diluted in equal measure with normal saline or arachis oil. For IV administration, it should be given as a 5% infusion in 5% dextrose, freshly made up every three hours.
Thiopentone ,,, is a barbiturate anesthetic. It should be given in ICU settings as patients require intubation and artificial ventilation. The most troublesome side effect is persisting hypotension and many patients require pressor therapy. Thiopentone has a strong tendency to accumulate and may also cause acute hypersensitivity. It should be administered cautiously in the elderly, and in those with cardiac, hepatic, or renal diseases. Formal clinical trials of its safety and effectiveness in either adults or children are few. Thiopentone can react with polyvinyl infusion bags or infusion sets. The continuous infusion should be made up in normal saline. The aqueous solution is unstable if exposed to air. Thiopentone is infused as 100 to 250 mg bolus over 20 seconds, with further 50 mg boluses every 2 to 3 minutes until seizures are controlled, with intubation and artificial ventilation. The IV infusion is then continued at the minimum dose required to control seizure activity (burst suppression on the EEG), usually between 3 and 5 mg/kg/h. After 24 hours, the dose should be controlled by blood level monitoring. Thiopentone should be continued for no less than 12 hours after seizure activity has ceased, and then slowly discontinued.
Propofol ,, is a nonbarbiturate anesthetic agent. Although it is widely used in the management of refractory SE, published experience is limited. Propofol is highly lipid soluble and has a high volume of distribution resulting in rapid action. Its effects are maintained only whilst the infusion is continued. Propofol administration causes profound respiratory and cerebral depression, requiring assisted ventilation. Long-term administration causes marked lipaemia and may result in acidosis. It may cause involuntary movements which should not be mistaken for seizures. Its safety in young children has not been established. It can be given as 2 mg/kg bolus dose which can be repeated if seizures continue, succeeded by an infusion of 5 to 10 mg/kg/h guided by EEG. The dose should be gradually reduced, and the infusion tapered 12 hours after seizure remission. In the elderly, the dose should be lowered.
Levetiracetam (LEV) is the (S) enantiomer of piracetam. It was initially approved by FDA as an add-on therapy for the treatment of patients with partial onset seizures in 1999. The mechanism of action differs from other AEDs and appears to be related to its binding to the synaptic vesicle protein 2A. It was introduced for the treatment of SE in 2006 and there are insufficient data on the safety and efficacy of this drug in SE. However, there are several case reports and retrospective analysis supporting its use in SE. ,, One ICU-based prospective study has shown that LEV is useful in refractory SE.  European Federation of Neurological Societies proposes the usefulness of LEV for the treatment of refractory complex partial SE. 
This drug does not seem to act by any of the mechanisms of currently available AEDs, but the exact molecular mechanisms of action of lacosamide have not yet been clarified. In addition to exerting anticonvulsant activity, it has shown to have the potential to retard kindling-induced epileptogenesis.  One case report has shown that IV lacosamide is effective in the treatment of nonconvulsive SE.  Another case report showed efficacy of oral lacosamide in refractory convulsive SE.  However, further large randomized studies are required to prove the efficacy and safety of this drug for the treatment of SE.
The drug treatment of tonic clonic SE can be divided into different stages like premonitory stage, stage of early SE (0-30 minutes), stage of established SE (30-90 minutes), and stage of refractory SE (after 60/90 minutes).
In patients with established epilepsy, usually, a prodromal phase (the premonitory stage) presages status, during which seizures become increasingly frequent or severe. The early treatment gives better result and prevents the evolution of seizures to SE. In this stage, diazepam, midazolam, or paraldehyde is highly effective. These drugs may be administered at home under careful supervision. Midazolam has the advantage that it can be given by intramuscular injection or buccal instillation. A randomized trial has shown that buccal midazolam has equal efficacy and is as rapid as rectal diazepam.  It is more convenient, potentially faster to administer, and less stigmatizing. The dose used is 10 mg drawn up into a syringe and instilled into the mouth between the cheeks and gums.
Stage of early SE (0-30 minutes)
Once SE has developed, treatment should be carried out in hospital, under close supervision. Fast-acting BDZ such as IV lorazepam or diazepam are the drugs of choice. Rectal or intramuscular paraldehyde is a useful alternative to BDZ in early status, where facilities for IV injection or for resuscitation are not freely available. Majority of patients respond to these treatments. Patients with SE should be observed for 24 hours in the hospital following remission.
Stage of established SE (30-90 minutes)
The stage of established SE can be operationally defined as SE which has continued for 30 minutes despite treatment in early-stage. By this time, the physiological decompensation begins. Intensive care facilities are desirable. The first-line treatment options at this stage are PHT, fosphenytoin, or phenobarbitone, loading intravenously followed by repeated oral or IV supplementation. IV valproate or LEV may be considered particularly when intensive care facilities and ventilatory support are not available.
Stage of refractory SE (after 60/90 minutes)
If seizures continue for 60 to 90 minutes after the initiation of therapy, the stage of refractory status is reached and full anesthesia is required in ICU care. Prognosis of these patients is poor with high mortality and morbidity. A number of anesthetic agents have been administered, although few have been subjected to formal evaluation and all have drawbacks. The most commonly used anesthetics are as follows: the IV barbiturate thiopentone or propofol with full range of intensive care facilities, including EEG monitoring.
The short-term mortality of SE has been reported as 7.6 to 39%. ,,,, Such a wide range is probably due to the differences in the study design, cause of SE, and healthcare facility. Underlying etiology is a powerful predictor of mortality, but not the seizure semiology. The predictors of mortality from SE included duration of seizures, age of onset, and etiology. , The patients with cerebral anoxia, stroke, and CNS infection have high mortality. , Patients with SE due to alcohol withdrawal and AED default or low therapeutic dose have relatively low mortality rate. In Richmond study, mortality was higher in adults (26%) compared with children (3%).  In a hospital-based study, mortality was 15.6% among 96 patients with a first SE episode. For the first SE episode, death was associated with potentially fatal etiology, age ≥65 years, and stupor or coma at presentation, but not with gender, history of epilepsy, SE type, or if treated late (≥1 hour).  A recent study on NCSE found extent of impairment of consciousness as the predictor of mortality.  In a retrospective review of patients of neuro ICU during 1993 to 2002, 43% patients were refractory to first line drugs and encephalitis was the most important cause of refractory SE. However, low levels of AED were associated with nonrefractory SE. Hyponatremia in the first 24 hours was significantly associated with refractory SE.  In nonfatal cases, SE is associated with significant morbidity. Studies have shown cognitive decline, as documented by neuropsychological testing after prolonged secondarily generalized partial SE.  SE is a neurological emergency and early treatment results in lesser mortality and better recovery.
We thank Mr. Rakesh Kumar Nigam for secretarial help.
[Table 1], [Table 2], [Table 3], [Table 4]