Hypothermia in refractory status epilepticus

Abstract

Introduction: Status epilepticus (SE) is a neurological emergency with potentially important mortality and morbidity. After refractoriness to general anesthetics, several pharmacological and nonpharmacological options have been described more or less anecdotally. In this context, and despite animal data supporting neuroprotective actions of brain hypothermia and showing its efficacy in SE models, hypothermia targeting a core temperature of about 33degreeC for at least 24 hours together with pharmacological sedation has been scarcely reported in adults and children. It seems that this approach rarely allows a sustained control of SE, as seizures tend to recur in normothermic conditions. Conversely, hypothermia has a high evidence level and is increasingly used in postanoxic encephalopathy, both in newborns and adults. Due to the thin available clinical evidence, prospective studies are needed to define the value of hypothermia in SE. Refractory status epilepticus and its treatment: SE represents the second most frequent neurological emergency after acute stroke, and bears significant risks of morbidity and mortality [1]. SE persisting despite adequate doses of benzodiazepines and at least one antiepileptic drug (AED) is considered refractory (RSE) [2,3]; this develops in 23 to 43% of patients with SE. RSE is associated with acute, severe and potentially fatal underlying etiologies, such as encephalitis, large stroke, or rapidly progressive primary brain tumors, and may be accompanied by coma; these factors, together with increasing age, represent the most important outcome predictors [1]. After securing pulmonary and cardiac functions, intravenous administration of a sequence of three groups of drugs represents the mainstay of management [4]: benzodiazepines (the only clearly evidence-based step); classical antiepileptic drugs (AED, mostly phenytoin, valproate, or levetiracetam); and general anaesthetics for RSE. Among anaesthetics, midazolam, propofol, or barbiturates represents the most popular agents, without any hard evidence favoring one specific compound [1]. Anesthetic treatment may lead to various complications (infections, metabolic disturbances, ileus, neuropathy, myopathy, thromboembolic events) [5]; it is therefore necessary to balance these risks with the benefit of rapid seizure control. Generalized convulsive RSE should be treated rapidly with general anesthetics, given the danger of systemic and neurological injury with ongoing convulsions; conversely, nonconvulsive SE without marked consciousness impairment can be approached more conservatively, as these forms are probably not associated with the same risk of injury [1,2,6]. RSE that proves refractory to a first course of general anesthetics implies a (repeated) careful search of the underlying etiology. This condition may be managed in several ways, which mostly rely on small series or case reports [1,7]. Briefly, pharmacological options may include further use of anesthetics (the three aforementioned, ketamine, isofluorane), other AED (for example, topiramate, lacosamide), or ketogenic diet. Reported nonpharmacological approaches span from resective surgery, through vagus-nerve, electroconvulsive or transcranial magnetic stimulations, to mild therapeutic hypothermia (TH). While the benefits of hypothermia on patients with head injury were already described by Hippocrates [8], TH enjoys an only evidence-based status in the setting of adult and pediatric postanoxic encephalopathy, and reduction of intracranial pressure [9]. Its indication for the treatment of other acute brain disorders, including SE and traumatic brain injury, is essentially anecdotic. Animal data on hypothermia: Low brain temperature exerts beneficial effects on the cascades involved in acute cerebral injuries; several seminal studies have been recently reviewed [9,10]. Hypothermia reduces brain metabolism and ATP consumption, and leads to decrease of glutamate release, free radicals, oxidative stress, mitochondrial dysfunction and calcium overload. Conversely, brain-derived neuro