The most widely used definition for SE is "a seizure lasting more than 30 minutes or recurrent seizures for more than 30 minutes during which the patient does not regain consciousness" [11,12]. More recently, an operational definition has also been suggested for adults and children older than 5 years [13] (Box 1). If we consider the duration for which most new-onset seizures in children last, once a seizure lasts for more than five to ten minutes, it is unlikely to stop spontaneously within the next few minutes, and intervention is indicated [14]. The use of the operational definition allows early treatment (starting at 5-10 min) [15]. However, in view of most previous studies on SE having been done using the 30-minute definition, the group suggests that for research purposes, both the definitions be considered and data provided with respect to both time durations.

SE in children is commonly due to cryptogenic or remote symptomatic causes in older children, and febrile or acute symptomatic cause in younger children [9,16]. Majority of childhood convulsive SE in a UK study (56%) occurred in previously neurologically healthy children, a quarter of SE were prolonged febrile seizures, and 17% were acute symptomatic [17]. Epidemiological data on SE in India is limited to a few single-center studies [18-20], with only one providing exclusive pediatric data [18]. The high proportion of acute symptomatic etiology, delayed presentation and poor outcome are the commonly reported findings. In an Indian pediatric intensive care unit (PICU) study over seven years, 53% had SE as their first seizure and only 60% had received any treatment prior to coming to the PICU [18]. A recent multi-centric study on SE in children across nine centers in India also reported similar findings: 82% acute symptomatic, <3% pre-hospital treatment, <20% deficit-free survival, and no uniform management protocol [unpublished data].

Treatment of SE needs to be initiated as early as possible since once seizures persist for 5 to 10 minutes, they are unlikely to stop on their own in the subsequent few minutes [21]. Moreover, the longer an episode of SE continues, the more refractory to treatment it becomes and the greater is the likelihood of complications [22]. Thus, the need for early treatment, preferably pre-hospital, is clear.

Pre-hospital management includes both first-aid during seizures, and pharmacotherapy. The initial care of a child with convulsions/coma is adequately described in Facility-based Integrated Management of Neonatal and Childhood Illnesses (F-IMNCI) guidelines of the Government of India [23] and will not be elucidated here further. Decision about pharmacotherapy must consider the drug and also the route of drug delivery (Box 2).

Benzodiazepines are the drugs that are currently in use for pre-hospital therapy for SE and include diazepam, lorazepam and midazolam. Pre-hospital treatment with benzodiazepines has been shown to reduce seizure activity significantly compared with seizures that remain untreated until the patient reaches the emergency department [24]. The various routes employed include per-rectal (diazepam, lorazepam, paraldehyde), intranasal (midazolam), buccal (midazolam, lorazepam) and intramuscular (midazolam).

Rectal diazepam is an approved out-of-hospital treatment for acute repetitive seizures in children. Response rates have been demonstrated to be similar to intravenous diazepam [25]. Multiple randomized, double-blind, placebo controlled studies have demonstrated that rectal diazepam given by caregivers at home is an effective and safe treatment for acute recurrent seizures [26-28]. Rectal administration may be difficult with wheel-chair users and larger patients. It can be socially unacceptable, and there is increasing concern about risk of sexual abuse allegations [29,30]. Therefore, non-rectal routes are gradually gaining favor for use by relatives/health workers in out-of-hospital settings. Rectal diazepam is the recommended drug for control of seizures in the F-IMNCI guidelines, in situations where intravenous access is not available [23].

In the past decade, research evidence has shown that buccal midazolam is more than or equally effective to rectal diazepam for children presenting to hospital with acute seizures, and is not associated with an increased incidence of respiratory depression [29,31,32]. Therefore, it may be considered as an acceptable alternative to rectal diazepam [30]. Intranasal midazolam has been shown to be as effective as intravenous diazepam in the treatment of prolonged febrile convulsions [4,33-35], and may also be an alternative. More recent data from the RAMPART study [36] of pre-hospital management of SE in children and adults has shown intramuscular midazolam to be as safe and effective as intravenous lorazepam for pre-hospital seizure cessation.　This may therefore emerge to be the agent of choice for out-of-hospital management of seizure by trained personnel.

Rectal and intranasal lorazepam have also shown efficacy for termination of acute convulsive seizures　in children [4,37]. However, non-availability of a commercial preparation in India precludes any firm guidance on non-parenteral use of lorazepam in India.

Parents of all children at risk of seizure recurrence should be counseled for appropriate home management for seizure [38].

Although convulsive seizures are the most obvious manifestation, SE is in fact a multisystem phenomenon i.e. prolonged and ongoing SE affects multiple organ systems. Hence, apart from attempts to rapidly control seizures, important goals of therapy are neuro-protection, and prevention and treatment of systemic complications associated with intravenous AEDs, anesthetic drugs and prolonged unconsciousness [39]. The supportive care should be tailored to the health care setting, the clinical presentations of SE, degree of encephalopathy, and degree of impairment of vital functions.

Assessment and care of vital functions is essential at all stages of managing any child with SE [40]. Adequate care of airway, breathing and circulaion takes precedence over any pharmacological therapy.

Airway: It is essential to maintain a patent airway during all stages of management of SE.

Breathing: Hypoxemia may result from respiratory depression/apnea, aspiration, airway obstruction, and neurogenic pulmonary edema [41].

Circulation: Continuous monitoring of pulse, blood pressure and perfusion should be done in all SE patients.

The treating team should anticipate one or more of the below mentioned problems depending on the duration of SE, age, underlying etiology, and the associated systemic co-morbidities. The cerebral and systemic metabolism undergoes changes described as initial phase of ‘compensation’, and if SE is sufficiently prolonged, later phase of ‘decompensation’ [42,43]. During the initial phase, prolonged seizures result in increased cerebral blood flow and metabolism, excessive catecholaminergic activity and cardiovascular changes. These in turn result in hyperglycemia, hyperpyrexia, tachycardia, sweating, hypertension, incontinence, cardiac arrhythmias, and lactic acidosis [43-46]. If the SE is prolonged, the cerebral autoregulation progressively fails and cerebral perfusion becomes dependent on systemic blood pressure resulting in hypoxia, cerebral ischemia, hypoglycemia, and lactic acidosis [42,43]. Management of these conditions is detailed in Web Table I. Both hypernatremia (serum sodium >145 meq/L) and hyponatremia (<135 meq/L) are deleterious for the brain. The major risks associated with hypernatremia are intracranial hemorrhage (subdural, subarachnoid and intraparenchymal) and osmotic demyelination (pontine or extra-pontine) with rapid correction.

Risk of infections is greatly increased in those with SE, especially when the duration is prolonged. Ventilator-associated pneumonia, urinary tract infection, pseudomembranous colitis, oral candidiasis, and speticemia are the common infections [47,48]. Commonest organisms are P. aeruginosa, A. spp, K. pneumoniae, and Entero bacteriaceae [48]. Hyperpyrexia, rhabdomyolysis, and raised intracranial pressure are the other common accompaniments [43,49,50]. Rarely, SE is associated with ictal bradycardia, stress cardiomyopathy, neurogenic pulmonary edema, rhabdomyolysis and related renal failure, or bone fractures [46]. Hypotension is common due to prolonged seizures, IV benzo-diazepines, or anesthetic agent infusions, and stress cardiomyopathy (Takotsubo cardiomyopathy) [47,51-53]. Cardiac arrhythmias are also common (up to 58%), with higher mortality in these patients [54,55]. Management depends on the cause of hypotension, hypovolemic shock, distributive shock, cardiomyopathy, or cardiac arrhythmias (WebTable I).

Although in most cases it is mild, early identification and aggressive treatment of rhabdomyolysis prevents complications like renal failure and compartment syndrome. The initial fluids for resuscitation may include normal saline or 5% dextrose in water (approximately 2-3 times the daily maintenance). Sodium bicarbonate may be added to IV fluids, especially if there is associated metabolic acidosis and/or hyperkalemia [56,57].

The clinical scenario, including the history and physical examination, is the most important factor guiding the specific evaluation that each child will require [58]. The investigations usually considered include blood chemistries, complete blood count, antiepileptic drug (AED) levels, toxicological studies, lumbar puncture, electroencephalography, and neuroimaging (Computed tomography [CT] scan and Magnetic resonance imaging [MRI]). The major part of evaluation can be performed after the child has been stabilized in an intensive care setting, and the seizures have been completely or partially controlled [58,59].

The investigations done are primarily to (i) determine the cause of status epilepticus, (ii) to look for complications of status epilepticus per se, and (iii) to identify the side-effects of drugs. Early identification of the etiology can result in aggressive specific management of cause. The investigations may vary depending on whether it is the first episode of SE in a normal child, or SE in a child with pre-existing epilepsy and already receiving AEDs [16,58]. The tests are detailed below (in no specific order), and listed in Table I in the order of importance.

TABLE I Investigations in a Child With Status Epilepticus 

Electrolyte and glucose abnormalities have been reported to be present in 1-16% of children with SE, although it is unclear whether they were the etiology in all and did treatment lead to cessation of the SE [16].

Serum calcium: Hypocalcemia as a cause of seizures is common in our country [60,61], usually due to vitamin D deficiency, and presents as a cluster of seizures in infancy. Early recognition avoids unnecessary treatment with AEDs and other interventions. Ionic calcium is more reliable as a guide for treatment and levels are usually <0.8 µmol/L in symptomatic children. However, all children with SE and subnormal ionized calcium levels (<1.2 µmol/L) should be treated. Total serum calcium, if done, should always be combined with estimation of phosphorous, serum alkaline phosphatase and serum albumin, for proper interpretation. Serum calcium estimation is an essential investigation for all children younger than 2 years with status epilepticus, irrespective of presence or absence of suggestive features.

Random blood sugar: Should be done in all children at presentation (especially in children less than 5 years) [58], as hypoglycemia may be responsible for seizures, and both hypo- or hyper-glycemia cause brain damage. When hypoglycemia is documented, urine ketones and reducing sugar should also be evaluated.

Serum sodium: Hyponatremia has been reported to be a cause in 1% of new-onset childhood convulsive SE [62]. However, most children with this abnormality were found to have suggestive features on history and clinical examination [63]. As this finding has important therapeutic implications [58], serum sodium estimation should be done in all, if feasible.

Metabolic disorders: Metabolic disorders are reportedly present in around 4.2% of children with SE [16], though their etiological significance is unclear. Routine metabolic workup therefore appears unwarranted. However, arterial blood gas estimation should be done in all children with established SE, if facilities are available; or when transferring to the Intensive Care Unit (ICU).

Blood counts: May be done routinely in children presenting with SE [16], especially those with associated fever. Infants with infection may not have fever and blood counts should be considered in them, even if afebrile.

Similarly, send blood cultures if the child is febrile (and above 6 months), or in a younger child, even if afebrile, if an infection is suspected.

Cerebrospinal fluid (CSF) examination: A central nervous system (CNS) infection is reported in 12.5% of pediatric convulsive SE [16]. CNS infections are also an important cause of SE in Indian children [18]. A CSF examination should be done by lumbar puncture in a febrile child, after stabilizing the child and excluding raised intracranial tension [16]. In infants younger than 6 months, signs of meningitis may not be clearly demonstrated and fever also may not be present. In such a situation, whenever there is a clinical suspicion of a CNS infection or sepsis, lumbar puncture should be done.

If done, CSF should be subjected at least to cell count (total and differential), biochemistry (protein, sugar, CSF: blood sugar ratio), bacterial culture, and gram stain. CSF pleocytosis, if present, should not be ascribed to a febrile SE [64]. Additional investigations on the CSF should be individualized. Systemic illness is a common trigger for convulsive SE in a patient who is already at risk, and, therefore, fever itself is not an indication to perform a lumbar puncture in a patient with epilepsy presenting with SE [58].

Inadequate AED drug levels (whether due to non-compliance, missed dose, or recent drug-dose alterations) are associated with a significant proportion of SE in children [9], although some studies found contradictory results [65]. Low AED levels were found in more than 30% childhood SE, although this was not necessarily the cause of SE [16].

AED levels should be done, if feasible, in all patients receiving AED and presenting with SE, as it has both etiologic (non-compliance/low drug-level as a cause) and therapeutic (loading dose of the previously effective drug for management) implications. However, availability of required facilities is likely to act as a bottleneck.

While considering EEG in SE, two situations need to be considered viz., an isolated, short-duration single EEG recording, or continuous EEG monitoring. No Indian studies on usefulness of EEG in pediatric SE are available. EEG abnormalities have been reported in ~90% children presenting with SE, though these were done hours to days later [16]. The information whether the seizure is focal or generalized is an important one when deciding chronic AED therapy for the patient.

EEG monitoring has been shown to be extremely useful, but under-utilized in SE management. After convulsive SE, one-third of　children　who undergo EEG monitoring are reported to have electrographic seizures, and among these, one-third experience entirely electrographic-only seizures [66,67].

An EEG should be considered in every child presenting with new-onset SE, although it can be delayed till the control of SE. An EEG should also be done if there is suspicion of non-convulsive SE (child not returning to the pre-SE state or remaining persistently encephalo-pathic even after the control of convulsive SE) or pseudostatus is suspected. Continuous EEG monitoring optimizes the management of SE and should be used, if feasible.

Neuroimaging can identify structural causes for SE, especially to exclude the need for neurosurgical intervention in children with new-onset SE without a prior history of epilepsy, or in those with persistent SE despite appropriate treatment. MRI is more sensitive and specific than CT scanning, but CT is more widely available and quicker in an emergency setting.

A meta-analysis reported structural lesions in 7.8% of childhood SE, commonly CNS malformations, trauma, and stroke/hemorrhage [66]. In a more recent study [68], the yield of MRI to detect structural lesions in convulsive SE was 31%. In the Indian setting, where inflammatory granulomas are a common cause of seizures [69], neuroimaging is likely to provide a higher yield.

Neuroimaging should be done, if feasible, in all children with SE, in whom no definitive etiology has been found. It should only be done after the child is appropriately stabilized and the seizure activity controlled. Emergent neuroimaging may be considered if there are clinical indications (new-onset focal deficits, persistent altered awareness, fever, recent trauma, history of cancer, history of anticoagulation, or a suspicion of AIDS).

Management of febrile SE is similar to that recommended for SE in these guidelines; however, there is evidence to show that phenytoin is less efficacious in this situation [116].

It is well-established that the duration of the first seizure does not affect the risk of recurrence, whether it is a single seizure or a status epilepticus. Moreover, remission rates are also not different in those who present with an episode of SE [117]. Brief recommendations for follow-up management after control of SE are provided in Box 4.

BOX 4 Guidelines for Follow-up Management of Children with Status Epilepticus

During the deliberations, the group also tried to identify the areas requiring research in the Indian context. These are listed in Box 5, and are expected to provide guidance to the researchers about issues needing evidence-support.

Contributors: All members of the writing group were involved in all aspects of manuscript preparation, approval of the final manuscript and the decision to publish.

Funding: Association of Child Neurology, Indian Council of Medical Research, and Sanofi-India Pvt Ltd.

Competing interests: None stated.

Important information: The participation in the meeting and its deliberations by all the invited experts was done in their individual capacity, and should not be considered as the official position of their respective Institutions, or Professional bodies to which they belong as office-bearers or members.

Experts (in alphabetical order): Anju Aggarwal, UCMS, Delhi; Satinder Aneja, LHMC, Delhi (Convener); B Chakravarty, AIIMS, Delhi; A Chattopadhyay, Apollo hospital, Kolkata; JS Goraya, Dayanand Medical College, Ludhiana; Rahul Jain, Chacha Nehru Bal Chikitsalaya, Delhi; Sourabh Jain, SZ Hospital, Bhopal; Urmila Jhamb, MAMC, Delhi; Veena Kalra, Delhi; Mahesh Kamate, JNMC, Belgaum; Sujata Kanhere, KJ Somaiya Medical College, Mumbai; Praveen Khilnani, BLK Memorial hospital, Delhi; Ramesh Konanki, Hyderabad; Rashmi Kumar, KGMC, Lucknow; PAM Kunju, Thiruvanantpuram; Lokesh Lingappa, Hyderabad; MM Mehndiratta, JSSH, Delhi; Rekha Mittal, Max hospital, Delhi; D Mishra, MAMC, Delhi (Co-convener); V Murugan, Chennai; Rajniti Prasad, BHU, Varanasi; Ashalatha Radhakrishnan, SCTIMST, Trivandrum; Col. KS Rana, Military Hospital, Jabalpur; Naveen Sankhyan, PGIMER, Chandigarh; Suvasini Sharma, LHMC, Delhi; Sunit Singhi, PGIMER, Chandigarh; Sanjib Sinha, NIMHANS, Bangalore; Bibek Talukdar, CNBC, Delhi; Manjari Tripathi, AIIMS, Delhi; Vrajesh Udani, Hinduja Hospital, Mumbai; and Nitish Vora, Ahmedabad.

Rapporteur: Rachna Sehgal, Safdarjung Hospital, Delhi.

Observers: Puneet Jain, Bijoy Patra, Dinesh Raj and Harikishan Suthar (all Delhi); Neetu Sharma (Gwalior).

Invited but could not attend the meeting: CP Bansal, President, Indian Academy of Pediatrics; Virender Kumar, LHMC, Delhi; Sheffali Gulati, AIIMS, Delhi; Rakesh Lodha, AIIMS, Delhi; Pratibha Singhi, PGIMER, Chandigarh; Kalpana V, GMC, Thiruvanathapuram; and Jitendra Sahu, PGIMER, Chandigarh.