STATUS EPILEPTICUS (SE) is a frequent medical emergency that presents to the emergency department ([ED]; Betjemann & Lowenstein, 2015; Gainza-Lein, Fernandez, Ulate-Campos, Loddenkemper, & Ostendorf, 2019; Pallin et al., 2008). Practically, it is defined as any seizure lasting 5 min or more in duration or two or more seizures without a return to baseline neurological status between events (Brophy et al., 2012; Glauser et al., 2016). Time in this condition is exceedingly important, as it has been identified that changes in receptor trafficking and neuropeptide expression occur within minutes and encourage a hyperexcitable state that contributes to the reduced efficacy of benzodiazepines the longer a seizure persists (Gainza-Lein et al., 2019). [gamma]-Aminobutyric acid A (GABA_A) receptors are the site of action for benzodiazepine therapy, and these receptors are internalized during prolonged SE. Conversely, N-methyl-D-aspartic acid (NMDA) receptors facilitate neuronal depolarization in the presence of glutamate, and these receptors are induced during prolonged SE, resulting in subsequent enhanced neuronal excitability. This also concurrently increases intracellular calcium, which activates calcineurin leading to decreases in the number of GABA_A receptors. Sustained and repeated epileptiform activity further modifies subunits of GABA_A and glutamate receptors, resulting in the predomination of neuronal excitation over inhibition. This subsequently makes the brain more susceptible to epileptiform activity and further SE. These pathophysiological changes have been demonstrated to be clinically relevant as research has consistently identified that delays in the time to treatment of SE are associated with longer seizure duration (Gainza-Lein et al., 2019). For example, the absence of prehospital treatment has been independently associated with SE episodes lasting longer than an hour (Chin et al., 2008). In addition, every minute delay from the onset of SE to ED arrival has been shown to confer a 5% increased risk of SE lasting longer than 60 min (Chin et al., 2008). Furthermore, in one study of 218 patients, the authors even identified an independent association between those receiving first-line therapy greater than 10 min after seizure onset and in-hospital mortality (Gainza-Lein et al., 2018).

Available guidelines are consistent in their recommendation of benzodiazepines as first-line therapy (Brophy et al., 2012; Glauser et al., 2016). The Neurocritical Care Society guidelines go on to state that an antiepileptic drug (AED) is required following the administration of benzodiazepines unless a modifiable factor is noted and can be corrected (e.g., hypoglycemia), with a goal of rapid attainment of therapeutic drug concentrations (Brophy et al., 2012). The guidelines do not provide an explicit recommendation for which AED is preferred; however, they do suggest that fosphenytoin, valproate sodium, or levetiracetam are all viable options. This seemingly rapid progression to an AED is reasonable when considering the excessive morbidity and mortality associated with SE and therefore the potential risks associated with ongoing SE may exceed those associated with overtreatment. Some have even suggested that early polytherapy with a benzodiazepine and an AED may potentially be more effective and less toxic than monotherapy alone (Kamppi, Ritvanen, Mustonen, & Soinila, 2015; Wasterlain et al., 2011). In situations of delayed initial therapy, other authors have gone so far as proposing "catch-up dosing" or the combined dosing of first- and second-line medications to improve efficacy (Gainza-Lein et al., 2019). The intent of this review is to assess the available data regarding the timing recommendations for SE management in the ED and outline potential strategies for optimizing the speed of therapy implementation with the goal of achieving earlier seizure cessation.

BENZODIAZEPINES

All available guidelines recommend first-line therapy with benzodiazepines secondary to their rapid onset of action and high rate of efficacy (Brophy et al., 2012; Glauser et al., 2016; Treiman, 1990). In terms of the timing of administration, the Neurocritical Care Society guidelines recommend that a benzodiazepine be administered within the first 5 min of seizure activity (Brophy et al., 2012). The American Epilepsy Society guidelines have similar recommendations, although their time frame is longer, recommending the administration of a benzodiazepine within the first 5-20 min (Glauser et al., 2016). Although all benzodiazepines can be used in this setting, based on individual pharmacokinetic and pharmacodynamic profiles, different agents are preferred on the basis of different routes of administration, with lorazepam being the recommended agent for intravenous administration and midazolam being the suggested agent for intramuscular and intranasal routes (see Table 1; Brophy et al., 2012; Glauser et al., 2016). This is due, in part, to greater lipophilicity of midazolam than that of lorazepam, which makes it more conducive to alternative routes of administration (e.g., intramuscular). These recommendations are supported by the RAMPART trial that prospectively compared the prehospital administration of intramuscular midazolam and intravenous lorazepam in 893 patients for the emergent treatment of SE (Silbergleit et al., 2012). The overall outcome of the study was that the two agents were equivalent in terms of time to seizure cessation; however, the important details were that, even though intramuscular midazolam took longer to work (3.3 min) than intravenous lorazepam (1.6 min), it was the fact that intramuscular midazolam was able to be administered faster (1.2 min) than intravenous lorazepam (4.8 min) that resulted in equivalent results.

Table 1. Recommended status epilepticus therapies

Benzodiazepine Utilization in Real-World Practice

In a review of an SE registry over a 4-year period from multiple hospitals in Europe that included 1,179 episodes of SE and occurring in 1,049 patients, the median time between the onset of SE and first treatment administration was 1 hr, with only 32% being treated within 30 min of the onset of seizure activity (Kellinghaus et al., 2018). This may have been a major contributor to the documented low rate of success of first-line agents in terminating SE (15%). Another review of 17 studies found that in the prehospital arena, 17%-64% of patients had a greater than 30-min delay to first-line therapy, with the median delay to first-line therapy ranging from 30 to 70 min (Hill, Parikh, Ellis, Myers, & Litt, 2017). In addition, between 29% and 61% of patients in the reviewed studies were not treated according to established protocols. Overall, the authors identified a relationship between seizure duration and delayed benzodiazepine therapy that was consistent across all age groups.

One of the most recent large, randomized trials of SE management, the ESETT trial, evaluated 384 patients who continued to have seizures 5-30 min following benzodiazepine therapy, who were then assigned to receive levetiracetam, fosphenytoin, or valproate (Kapur et al., 2019). Patients were eligible for inclusion into the trial if they had been treated with a cumulative dose of benzodiazepines that was consistent with recommendations (e.g., 4 mg) rather than the individual doses being compliant. Interestingly, in a subgroup analysis of 207 of the patients included, the average patient received 2.5 doses of a benzodiazepine, with 78.2% of individual lorazepam doses and 89.3% of individual midazolam doses being below the recommended dose (Sathe et al., 2019). Even in the 102 patients who received their first benzodiazepine dose upon arrival to the ED, 70.2% of those first doses did not meet guideline recommendations. From a timing perspective, these two modalities of benzodiazepine administration, guideline-compliant cumulative doses and guideline-compliant individual doses, are not the same. If a suboptimal individual dose of a benzodiazepine is administered, it takes a longer collective time in order to wait for the suboptimal dose to take effect, then prepare and administer another suboptimal dose, and finally wait on that dose to take clinical effect, rather than administering the guideline-compliant dose at the outset. As previously discussed, during this delay in recommended therapy, the brain is actively adapting and it is becoming more challenging by the minute to terminate the seizure with traditional pharmacotherapy.

Systemic underdosing of benzodiazepines for the treatment of SE has been reported to be a frequent occurrence even outside of clinical trials. In one retrospective analysis of 170 adult and pediatric patients presenting with SE across two EDs, the authors found that only 11% of patients received guideline-recommended benzodiazepine dosing, all of whom were pediatric patients (Langer & Fountain, 2014). In another retrospective analysis of 44 adult patients with SE, the authors found that only one patient received guideline-recommended benzodiazepine dosing (Braun, Gau, Revelle, Byrne, & Kumar, 2017). In a larger retrospective review of 222 adult patients who received benzodiazepines in the ED for the treatment of SE, the authors found that only 1.5% of patients received guideline-recommended benzodiazepine dosing (Weant, Barre, Bruner, Smiley, & Hall, 2021). These data highlight an opportunity to optimize clinical practice through education and the development of systems that reinforce benzodiazepine dosing that is consistent with available guideline recommendations in order to provide the best opportunity for early seizure termination and limit the maladaptive pathophysiological alterations that can occur with continued seizure activity.

ANTIEPILPETIC DRUGS

Considering the multiple receptor and signaling alterations occurring in the actively seizing brain, it would logically follow that a treatment strategy targeted at multiple receptors might be more effective than a singular strategy alone (Wasterlain et al., 2011). The Neurocritical Care Society guidelines in particular recommend the administration of an AED after benzodiazepine administration unless a modifiable factor is noted and can be corrected (Brophy et al., 2012). In animal models, it has been shown that even in the face of both benzodiazepine and AED monotherapy failures to terminate seizures, there was a significant improvement in seizure termination with combination therapy (Wasterlain et al., 2011). In addition, it was noted to allow for lower doses of benzodiazepines to be used, which decreased cognitive side effects. As an extension in the human literature, other authors have identified that an increasing delay from seizure onset to the administration of an AED was significantly correlated with a greater delay in a return of consciousness, independent even of the effect of the first medication administered (Kamppi et al., 2015).

Antiepileptic Drug Utilization in Real-World Practice

In the aforementioned ESETT trial, the authors evaluated 384 patients who continued to have seizures 5-30 min following benzodiazepine therapy who were then randomized to receive levetiracetam, fosphenytoin, or valproate (Kapur et al., 2019). Despite the previously noted suboptimal benzodiazepine administration, and a subsequent median duration of seizure activity of around 60 min, the administration of any of the AEDs led to seizure cessation in approximately half of patients and was nonsignificantly different among the three groups. Although a 50% success rate is far less than optimal, taken within the context of the historical data from our landmark trials looking at benzodiazepine dosing that lead to seizure termination in about 60% of patients, the potential for additive efficacy through the more emergent administration of both classes of agents is compelling (Alldredge et al., 2001; Treiman et al., 1998). In the previously mentioned retrospective review of adult patients who received benzodiazepines in the ED for the treatment of SE, the authors found that despite only 1.5% receiving guideline-recommended benzodiazepine dosing, those who received early AED therapy were significantly less likely to receive additional benzodiazepine doses, be intubated, or be admitted to the intensive care unit (ICU) or hospital (p < 0.05; Weant et al., 2021). Hence, early aggressive combination therapy may be one of the more effective and underutilized strategies to terminate SE.

Antiepileptic Drug-Specific Considerations

Phenytoin

Phenytoin is a traditional second-line AED for the treatment of SE and is a recommended option in both guidelines (Brophy et al., 2012; Glauser et al., 2016). Various studies have shown that phenytoin has clinical efficacy in terminating seizure activity (Agarwal et al., 2007; Alvarez, Januel, Burnand, & Rossetti, 2011; Misra, Kalita, & Patel, 2006; Treiman et al., 1998). However, despite its documented efficacy, the use of phenytoin does present some important challenges to consider. It must be administered no faster than 50 mg/min to limit infusion-related side effects such as hypotension and arrhythmias, and the intravenous line should be flushed with saline following infusion to limit venous irritation secondary to the alkalinity of phenytoin (Pfizer, 2011). Also, because of its poor solubility and risk of precipitation, it must be infused through an inline 0.22- to 0.55-[mu]m filter. In addition, phenytoin has many adverse effects to monitor for during the infusion such as hypotension and bradycardia, as well as the potential for drug interactions that may develop in the acute setting (Brophy et al., 2012; Der-Nigoghossian, Rubinos, Alkhachroum, & Claassen, 2019). Although phenytoin is a cytochrome P450 inducer that can alter the metabolism of other hepatically metabolized medications, this is not an emergent concern in the setting of SE management as it takes days to develop. However, coadministration with valproic acid can result in an immediate drug interaction via multiple mechanisms, including protein binding, which can be exceedingly challenging to manage. Phenytoin is also a narrow therapeutic index medication with a generally accepted therapeutic range of concentrations that identify efficacy (10-20 mcg/ml) and potentially toxicity (>20 mcg/ml) (von Winckelmann, Spriet, & Willems, 2008). Further complicating appropriate dosing of this agent is the substantial interpatient variability that can exist and requires nuanced dosing, particularly in select populations such as the critically ill, those with kidney dysfunction, and obesity (Holder, Bailey, Baum, Justice, & Weant, 2019; Montgomery, Chou, McPharlin, Baird, & Anderson, 2019; von Winckelmann et al., 2008). As a consequence, in the emergent setting of managing SE, some have advocated for the obtainment of postload concentrations (more than 2 hours after the loading dose) to ensure that therapeutic concentrations have been obtained (Selioutski et al., 2017).

Fosphenytoin

Fosphenytoin is also a very effective agent in the treatment of SE and is recommended by existing guidelines (Brophy et al., 2012; Glauser et al., 2016). It is a prodrug of phenytoin, meaning that after it is infused, it is converted to phenytoin and so should not be used adjunctively with phenytoin. It also has an infusion rate limit (150 mg/min) and so peak serum phenytoin concentrations are not reached until completion of the infusion and subsequent conversion to phenytoin. Fosphenytoin is recommended over phenytoin in guideline recommendations due to its improved overall tolerability (Glauser et al., 2016). In a recent trial comparing levetiracetam, fosphenytoin, and valproate for the treatment of benzodiazepine refractory convulsive SE, fosphenytoin was as effective in achieving seizure cessation at 60 min as the other agents studied, although doses were capped at 1,500 mg, which may have resulted in the underdosing of several patients (Kapur et al., 2019). Although a maximum fosphenytoin loading dose of 1,500 mg intravenously is recommended in the American Epilepsy Society guidelines, no such maximum exists in the Neurocritical Care Society guidelines (Brophy et al., 2012; Glauser et al., 2016). Studies evaluating real-world practice have found that obese patients receive significantly smaller loading doses than nonobese patients and that arbitrary total dose limitations in patients result in significantly lower postload concentrations, potentially yielding subtherapeutic phenytoin concentrations that can put the patient at risk for further seizure activity (DasGupta, Alaniz, & Burghardt, 2019; Holder et al., 2019).

Levetiracetam

Levetiracetam is a relatively newer AED that has shown similar efficacy in seizure cessation when compared with other AEDs and is recommended in both guidelines (Brophy et al., 2012; Glauser et al., 2016; Kapur et al., 2019). The notable distinguishing characteristic for this agent is the relative absence of administration side effects or drug interactions in the acute setting of SE. One prospective study looked at the administration of 20, 40, or 60 mg/kg doses of levetiracetam given to 45 patients aged 4-32 years (Wheless et al., 2009). A 100-mg/ml vial of levetiracetam was diluted 1:1 with 0.9% sodium chloride or dextrose 5% in water and infused over 5-6 min. There were no significant changes in blood pressure and no electrocardiogram abnormalities. One patient in the 60-mg/kg group reported a nonprurutic rash and one reported pain at the site of infusion (Wheless et al., 2009). Another retrospective review of 33 adult patients in the ICU showed that there was no difference in side effects in the 16 patients who received an undiluted bolus over 3-5 min and the 17 patients who received an infusion at a rate of 200-400 mg/hr (Burakgazi, Bashir, Doss, & Pellock, 2014). A recent systematic review of nine articles investigated rapid levetiracetam administration over 5 min, with doses ranging from 280 to 4,500 mg (Jense, Douville, & Weiss, 2022). The authors concluded that the administration of intravenous levetiracetam at a fast rate appears to be safe and tolerable via a peripheral line with few adverse effects, suggesting that SE treatment may be safely expedited by administering it more rapidly than other agents.

Valproic Acid

Valproic acid is another traditional AED recommended by both guidelines that has been shown to be effective as a second-line treatment option for patients with SE (Brophy et al., 2012; Glauser et al., 2016). In the ESETT trial, the efficacy of valproic acid in achieving seizure cessation was not significantly different from those of levetiracetam or fosphenytoin (Kapur et al., 2019). Similar to phenytoin and fosphenytoin, as newer AEDs have been developed and studied, it has been favored less as a result of its potential adverse effects and drug interactions (Johannessen & Johannessen, 2003). In another similarity to phenytoin and fosphenytoin, valproic acid is considered a narrow therapeutic index drug with a generally identified therapeutic range (50-100 mcg/ml) for efficacy and so obtainment of postload concentrations (more than 2 hours after the loading dose) to ensure that therapeutic concentrations have been obtained in SE may also be reasonable (Patsalos, Spencer, & Berry, 2018; Van Matre & Cook, 2016). Drug concentration monitoring is particularly recommended in certain situations as other medications can cause decreased valproic acid concentrations (e.g., phenytoin, carbapenem antibiotics; Frigo, Lecchini, Gatti, Perucca, & Crema, 1979; Mori, Takahashi, & Mizutani, 2007; Nau & Loscher, 1984; Pisani, Di Perri, Perucca, & Richens, 1993). Valproic acid does have an advantage over phenytoin and fosphenytoin, however, in that it can generally be safely administered more rapidly at a rate of 6 mg/kg/min to patients with SE (Morton, O'Hara, Coots, & Pellock, 2007; Ramsay et al., 2003; Wheless et al., 2004). One notable consideration regarding valproic acid is that it has a known significant fetal risk and should not be used in pregnancy (Wartman & VandenBerg, 2022). The Neurocritical Care Society guidelines recommend considering the use of levetiracetam for SE not due to eclampsia in this population (Brophy et al., 2012).

Lacosamide

Lacosamide is another newer AED that has been shown to be an effective emergent second-line therapy in the treatment of SE demonstrating noninferiority to fosphenytoin (Der-Nigoghossian et al., 2019; Husain et al., 2018). Although it has not been studied as extensively as some of the other AEDs, and hence a lower recommendation in the guidelines, it has favorable qualities compared with other agents such as minimal drug interactions and a lower incidence of major adverse side effects (i.e., atrioventricular nodal block, somnolence, hypotension, bradycardia; Davidson, Newell, Alsherbini, Krushinski, & Jones, 2018; McLaughlin et al., 2021). Another advantage is that although originally developed to be infused intravenously over 30 min, recent studies have been investigating more rapid administration in the setting of SE to expedite seizure cessation. A retrospective review of 166 adult patients receiving either an undiluted intravenous push at a rate of 80 mg/min or an intravenous infusion over 30 min was compared for safety and efficacy (Davidson et al., 2018). There was no significant difference in adverse effects among the 78 patients who received the undiluted intravenous push and the 88 patients who received an intravenous infusion. This study also showed that there was a significant difference in median time-to-order verification, with a reduced time of 35 min in the intravenous push group compared with 109 min in the intravenous infusion group (p < 0.001; Davidson et al., 2018). Another retrospective cohort analysis of 73 adults receiving an intravenous infusion compared with 102 adults receiving an undiluted intravenous push also showed no significant differences in major adverse effects (McLaughlin et al., 2021). Therefore, as more data become available, this agent may fill a similar role as levetiracetam in the armamentarium of emergent AED therapy that can be rapidly administered at bedside and expedite seizure cessation.

Implementation

The complexity of managing this emergent condition not only exists in understanding the optimal therapies available but also extends to the practical implementation of these therapies in the equally complex environment of the ED. However, there are several groups that have successfully developed processes to ensure that the best possible care is provided in this patient population. For example, one group developed and implemented a generalized convulsive SE management protocol in adult patients (Aranda et al., 2010). Despite a low protocol adherence (38%), following the protocol resulted in a significantly greater likelihood of seizure termination (74% vs. 29%; p < 0.0001) and a significantly lower need for additional benzodiazepine doses (4.7 times; p = 0.0004) and AED administrations (9.1 times; p < 0.0001). Consistent with other studies, early seizure cessation resulted in a significant improvement in clinical outcomes such as decreased ICU (1 vs. 2 days; p < 0.0001) and hospital length of stay (3 vs. 7 days; p = 0.009), as well as a decreased rate of refractory SE and a decreased intubation rate (3.8 times). These authors demonstrated that simply identifying this as an opportunity for improvement, developing a protocol consistent with available guideline recommendations, and implementing it in practice yielded significant benefits in patient outcomes even if it was only followed in one third of patients.

Another group took the approach of developing an inpatient SE alert system in which, following the identification of a patient with suspected convulsive or electrographic SE, a text page could be sent simultaneously to general neurology and pharmacy residents, and the rapid response team to respond at bedside (Villamar et al., 2018). The authors found that in the 19 patients managed with the SE alert system, compared with the 20 patients managed in a traditional fashion, the time to administration of a second-line AED was significantly shorter (22.21 vs. 58.3 min; p < 0.0001, respectively). The authors did not identify any differences in clinical outcomes such as ICU length of stay, mechanical ventilation, or mortality. However, this could be due, in part, to their inability to assess first-line benzodiazepine dosing or timing, which may have impacted outcomes. Nevertheless, these results do support those of the prior study that demonstrated establishing the management of this condition as an emergency and focusing on its emergent management can yield improvements in care.

One unique example identified the care of pediatric patients presenting with seizures as an opportunity and developed a quality improvement program with the intent of reducing treatment times (Ostendorf, Merison, Wheeler, & Patel, 2018). They applied the pharmacotherapy approach of emphasizing intranasal midazolam over intravenous lorazepam, and they simply relocated medications and supplies to more readily available locations. They then ensured the provision of adequate education and training. Treatment was reviewed before and after implementation of this program, and the authors found a significant reduction in the median time to benzodiazepine administration (7.5 vs. 14 min; p = 0.01) and an increase in the percentage of benzodiazepines administered within 10 min (79% vs. 39%). They also confirmed their hypothesis that intranasal midazolam would be faster to administer than intravenous lorazepam (79% vs. 64%), albeit nonsignificant. Interestingly, their time to intravenous fosphenytoin also decreased postimplementation (94 vs. 29 min; p = 0.01). Although impressive, it is important to highlight that these results may have been due to the Hawthorne effect, the education and training provided, or the result of a simplified process that standardized care to one agent. Clinically, this program saw improvements as well, with a significant reduction in patients transferred to the ICU (9% vs. 39%; p = 0.02), which of course has implications for both cost and patient throughput. Although intranasal midazolam would not typically be the drug or route of choice in comparison with intravenous lorazepam, these authors identified that this was an effective approach for their particular situation to optimize care and they demonstrated its efficacy. In fact, in one analysis of 11 studies, non-intravenous benzodiazepine therapy was consistently faster to administer (mean difference = 3.41 min; 95% CI [1.69, 5.13]) and was superior to intravenous therapy for treatment failure (OR = 0.72; 95% CI [0.56, 0.92]) (Alshehri et al., 2017). However, it also took consistently longer to stop seizure activity (mean difference = 0.74 min; 95% CI [0.52, 0.95]). This matches very well with what is known about the pharmacokinetics and pharmacodynamics of these agents (Brophy et al., 2012). In this quality improvement program example, it was conducted in pediatric patients, many of whom do not have intravenous access and there may be a hesitancy at times to administer agents intramuscularly. For this institution and in this population, the functional aspect of administering intranasal midazolam was faster than intravenous lorazepam. Hence, although it is always important to use guideline recommendations to guide care, modifications may be necessary based on individual practice considerations, as time is at a premium in this emergent presentation.

CONCLUSION

SE is a common medical emergency seen in the ED, with a high incidence of morbidity and mortality when seizures are unable to be controlled quickly. Guidelines consistently recommend benzodiazepines as first-line treatment, and studies have helped determine the optimal route of administration to abort seizures rapidly. Ideally, lorazepam is the best option for intravenous administration and midazolam for intramuscular or intranasal administration. In circumstances where it is difficult to obtain intravenous access in order to administer lorazepam, rapid benzodiazepine administration by any route is acceptable when considering the detrimental effects of delaying therapy. However, underdosing of benzodiazepine therapy appears to be a systemic occurrence and may be placing patients at an increased risk of continued seizure activity. Urgent second-line therapy with an AED is also recommended in the guidelines. Although guidelines do not explicitly recommend one agent over another, taking a proactive and systematic approach to treatment that ensures expeditious medication procurement, as well as rapid and safe medication administration, can assist practitioners in identifying which AEDs are optimal to use at their institution for polytherapy that have the greatest potential to enhance the provision of timely care.

REFERENCES