A 42-year-old woman with history of epilepsy was brought to our emergency department after having three witnessed generalized tonic-clonic seizures. Her epilepsy was a result of prior traumatic head injury. Recently the dose of valproic acid had been reduced. The paramedics decided to intubate her prior to transportation due to concerns about the patency of her airway. We are called into the emergency department to evaluate the patient, when she has another generalized tonic-clonic seizure upon our arrival. The seizure lasts 90 seconds and the patient remains unconscious after its conclusion. Her husband informs us that she has not been alert since the first seizure happened. She is intubated and mechanically ventilated with good oxygenation, afebrile, and mildly tachycardic and hypertensive. She has no neck stiffness and no lateralizing signs on motor examination. We lift her eyelids and notice nystagmus-like movements of her eyes.
What do you do now?
Successful treatment of status epilepticus (SE) begins with the recognition that it is a neurological emergency. If you don’t treat it early, SE becomes more refractory over time. This is due to pharmacoresistance and alteration in the GABA receptor sensitivity and availability to agonists, such as benzodiazepines. The longer it lasts the higher the risk of complications, including permanent neuronal damage and dropout.
The questions we should ask ourselves are: First, how do you recognize refractory SE? In patients with continuous generalized convulsions the diagnosis of SE is self-evident (though it still requires differentiation from psychogenic pseudo-SE). But, all too often, there seems to be reluctance to diagnose SE in patients with rapidly repetitive seizures. These patients must be treated for SE if they do not recover full alertness between seizures. The diagnosis is also missed too frequently when the clinical manifestations are subtle (for instance in cases of complex partial SE) and when patients are comatose (in whom the only clinical manifestation, if any, may be nystagmoid eye movements or minimal flickering of a finger or a toe). Generalized convulsive SE becomes nonconvulsive over time, and EEG recording is often needed to find a close correlation between subtle movements and ictal discharges.
Second, how do you treat SE when it becomes refractory? Even the suspicion of SE means you must start antiepileptic treatment immediately. Benzodiazepines are the first-line therapy—intravenous lorazepam being the preferred choice because of its rapid onset of action and longer duration of antiepileptic effect—and should be given while emergently assessing airway patency, adequacy of ventilation and oxygenation, and circulatory status. In all cases a capillary glucose level should be measured to exclude hypoglycemia. Blood should be drawn for measurement of serum electrolytes, lactic acid, creatine kinase, complete cell count, and arterial gases. If seizures stop and the patient wakes up, further progression is not likely. But if the seizures do not stop you need to move to the next line of therapy (fosphenytoin or valproic acid) without delay. In fact, treatment of SE is best optimized by following a clear protocol progressing from one line of therapy to the next until seizures stop. The dose should be adequate, and failure to prescribe the right dose is a common error. Always check if the patient has received an appropriate dose of each drug (e.g., lorazepam 0.1 mg/kg, phosphenytoin 20 mg/kg or higher, valproic acid 30 mg/kg or higher) before concluding it failed. Our current treatment protocol is shown in Figure 11.1.
FIGURE 11.1 Algorithm for the management of status epilepticus. ABC, airway, breathing, circulation; CSF, cerebrospinal fluid; ED, emergency department; EEG, electroencephalography; ICU, intensive care unit; MV, mechanical ventilation.
* The risk of propofol infusion syndrome is substantial and this complication may be fatal.
SE should be considered refractory after failure of two antiepileptic agents. In our practice, the diagnosis of refractory generalized SE means we will need to start a continuous infusion of an anesthetic agent. This decision demands endotracheal intubation for mechanical ventilation and continuous electroencephalographic (EEG) monitoring. There are some exceptions to this rule. In cases of complex partial status epilepticus we try one or two more anticonvulsants before using anesthetics because in these patients there is less evidence that uncontrolled complex partial seizures can produce irreversible brain damage, at least in the short term. For the same reasons, in patients with epilepsia partialis continua we try to avoid intubation and potent anesthetic drugs.
Among anesthetic agents we favor midazolam because of its better safety profile Table 11.1. Midazolam can be effective in aborting status epilepticus when used in high doses. We start with a bolus of 0.2 mg per kilogram of body weight and an infusion of 0.2 mg/kg/hr. However, we rapidly increase the infusion dose until we achieve suppression of the seizures and have reached doses as high as 5 mg/kg/hr in the most recalcitrant cases. Even these very high doses are well tolerated by most patients, although support with vasopressor drugs may be needed. Tachyphylaxis develops quickly with benzodiazepines in general and midazolam in particular. This phenomenon may demand using even higher doses if the infusion needs to be maintained over time.
Propofol is a very effective antiepileptic anesthetic, but we have found it unsafe in the doses necessary to control refractory SE (often higher than 100 micrograms per kg per minute). The main risk is the development of propofol infusion syndrome. This syndrome –albeit rare– is manifested by lactic acidosis, rhabdomyolysis, myocardial depression, and, when most severe, cardiovascular collapse and cardiac arrest. In our experience, even careful monitoring of metabolic changes (serial lactic acid, arterial blood gases, and creatine kinase levels) may fail to recognize the beginning of a fatal form of this complication. Therefore, we rarely use propofol for the treatment of SE and when we do we strictly avoid infusing large doses. More than 80 μg/kg per minute or 3 mg/kg per hour for longer than 48 hours should be avoided.
Continuous infusion of barbiturates, such as pentobarbital, is very effective in aborting SE. Unfortunately, adverse side effects are many and often severe. Hypotension is ubiquitous and requires vasopressors. Infections, especially pneumonia, ileus, and liver toxicity occur in the majority of patients treated with a barbiturate drip for more than 2 days. Consequently, we tend to reserve this option for those patients who fail to be controlled with midazolam.
TABLE 11.1 Doses and Side Effects of Therapeutic Options for Refractory Status Epilepticus
*Usual doses are between 0.2 and 5 mg/kg/hr, but much higher rates of infusion are needed (and can be tolerated) in selected cases.
**Life-threatening risk with prolonged infusion of high doses of the drug; contraindicated in children.
There are other alternatives we have tried with variable success in our most challenging cases. We have not been impressed by the usefulness of lidocaine and ketamine, but others have noticed good results using high doses of ketamine. We have found isoflurane to be the only effective rescue therapy in patients who had become dependent on very high doses of pentobarbital. Our enthusiasm for isoflurane has been tempered, however, by the appearance of marked brain and brainstem abnormalities on the MRI scans in two of our isoflurane-treated patients suggesting a neurotoxic effect.
Induced hypothermia can also be helpful in SE. Our experience with hypothermia for this indication is still limited, but our initial results have been encouraging. Finally, in some cases the seizures can only be controlled by treating the underlying cause that provoked them. Searching for treatable forms of encephalitis, brain lesions amenable to resection, and some specific systemic illnesses (for instance thrombotic thrombocytopenic purpura) is equally important. Last-resort measures have included ketogenic diet or electroconvulsive therapy. How electroconvulsive therapy works is not known and may be simply through a “rebooting” phenomenon. In the worst cases, one must accept at some point that the status epilepticus is untreatably refractory; in these rare instances the status eventually “burns out” at the expense of rapid brain loss which can be documented by the accelerated atrophy on serial neuroimaging.
Back to our patient. After another seizure in the emergency department she was treated with 0.1 mg/kg of intravenous lorazepam. The abnormal eye movements only resolved for a few minutes but then recurred. Because valproate discontinuation likely triggered the seizures, she was loaded with valproic acid (15 mg/kg while awaiting the serum level) and then transferred her to our neurosciences intensive care unit. Emergency EEG showed continuous epileptiform activity. After a bolus of 0.2 mg/kg, midazolam infusion was initiated in the ICU at a rate of 0.2 mg/kg/hr and then titrated until the electrographic seizures abated (infusion rate reached 0.5 mg/kg/hr). We then adjusted the dose of valproic acid according to the serum level, which was subtherapeutic. Within the following 24 hours we were able to wean off the infusion of midazolam. The patient awoke and could be extubated without problems.
Treatment of refractory status epilepticus may be summarized as follows: treat aggressively and early and neurologists should pack a hard punch, protect and support the patient, and try to find a treatable cause. Even when prolonged anesthetic treatment is needed, outcome can still be favorable in young patients if there is no MRI evidence of permanent brain damage.
KEY POINTS TO REMEMBER REGARDING CHOICES IN REFRACTORY STATUS EPILEPTICUS
· Status epilepticus is a neurological emergency.
· Seizures become more resistant to antiepileptics over time, and prolonged status epilepticus can produce irreversible brain damage.
· It is best to use a treatment protocol and to decisively progress from step to step.
· Make sure the right drugs and the right doses are being used.
· If two anticonvulsants fail, consider intubating the patient, use continuous electroencephalographic monitoring, and starting a continuous infusion of an anesthetic agent.
· Among anesthetics, midazolam provides the best balance between safety and effectiveness. Propofol and pentobarbital are also very effective, but their use is associated with a greater risk of severe- and even fatal medical complications.
· Inhaled gases, such as isoflurane, and moderate hypothermia are valuable therapeutic alternatives in recalcitrant cases.
Chen JW, Wasterlain CG. Status epilepticus: pathophysiology and management in adults. Lancet Neurol 2006; 5:246-256.
Cooper AD, Britton JW, Rabinstein AA. Functional and cognitive outcome in prolonged refractory status epilepticus. Arch Neurol 2009; 66:1505-1509.
Fugate JE, Burns JD, Wijdicks EF, Warner DO, Jankowski CJ, Rabinstein AA. Prolonged high-dose isoflurane for refractory status epilepticus: is it safe? Anesth Analg 2010; 111:1520-1524.
Iyer VN, Hoel R, Rabinstein AA. Propofol infusion syndrome in patients with refractory status epilepticus: an 11-year clinical experience. Crit Care Med 2009; 37:3024-3030.
Mayer SA, Claassen J, Lokin J, Mendelsohn F, Dennis LJ, Fitzsimmons BF. Refractory status epilepticus: frequency, risk factors, and impact on outcome. Arch Neurol 2002; 59:205-210.
Meierkord H, Boon P, Engelsen B, G cke K, Shorvon S, Tinuper P, Holtkamp M; European Federation of Neurological Societies. EFNS guideline on the management of status epilepticus in adults. Eur J Neurol 2010; 17:348-355.
Rabinstein AA. Management of status epilepticus in adults. Neurol Clin 2010; 28:853-862.
Shorvon S. The treatment of status epilepticus. Curr Opin Neurol. 2011; 24:165-170.