a b s t r a c t

Purpose: Although guidelines recommend specific benzodiazepine doses for the treatment of generalized convul- sive status epilepticus (GCSE), underdosing appears to be common. The purpose of this investigation was to as- sess benzodiazepine dosing strategies for the initial management of GCSE in patients presenting to the Emergency Department (ED).

Methods: This was a retrospective review of adult patients who received benzodiazepines in the ED for treatment of GCSE. Characteristics of those achieving seizure cessation following initial benzodiazepine therapy were assessed.

Results: 222 patients presented to the ED and received 403 doses of benzodiazepines, of which 1.5% conformed with recommendations. First-line therapy was successful in 86.8% of patients with an average dose of 1.6 mg (0.02 mg/kg). No difference in dosing was noted between those experiencing early cessation and those that did not (p = 0.132). Patients experiencing early cessation were significantly less likely to receive further doses, be intubated, or be admitted to the intensive care unit (ICU) or hospital (p < 0.05 for all comparisons). Those that received early antiepileptic drug therapy were significantly less likely to receive additional benzodi- azepine doses, be intubated, or be admitted to the ICU or hospital (p < 0.05 for all comparisons).

Conclusions: According to Guideline recommendations, there was consistent underdosing of benzodiazepines noted in both prehospital and ED settings. Early seizure cessation and the early receipt of an antiepileptic drug were found to be associated with multiple significant clinical outcomes. Future investigations should explore op- timal dosing strategies for benzodiazepines as well as the impact of early antiepileptic drug administration.

(C) 2021

1. Introduction

Acute status epilepticus (SE) is a challenging presentation to the Emergency Department (ED) that is associated with substantial mor- bidity and mortality [1-3]. Therefore, its treatment is a medical emer- gency and the early cessation of Seizure activity is associated with improved clinical outcomes [4]. In practice, SE is defined as any seizure lasting >=5 min in duration or >=2 seizures without a return to baseline in between [5,6]. An important subset of SE is generalized convulsive

Abbreviations: ED, Emergency Department; GCSE, Generalized convulsive status epilepticus; IV, Intravenous; IM, Intramuscular; ICU, Intensive Care Unit; SE, Status Epilepticus; EMS, Emergency Medical Services; GCS, Glasgow coma scale; LE, Lorazepam equivalents; REDCap, Research Electronic Data Capture; ESETT, Established Status Epilepticus Treatment Trial.

* Corresponding author.

E-mail address: [email protected] (K.A. Weant).

status epilepticus (GCSE) which is defined as tonic-clonic movements of the extremities [5]. Regardless of the type of SE however, all available guidelines recommend first-line therapy with benzodiazepines due to their rapid onset and ability to terminate seizure activity [5,6]. When used early in SE for seizure control, benzodiazepines have been found to be 79% effective [7]. Although benzodiazepines can be administered through multiple routes, lorazepam is the preferred agent for intrave- nous (IV) administration and midazolam is the preferred agent for in- tramuscular (IM) administration [5,6].

Despite explicit guideline recommendations, the underdosing of these agents appears to occur both in clinical practice as well as in pro- spective, randomized controlled trials [8,9]. Two small, retrospective ob- servational studies have explored this topic in both the Emergency Medical Services (EMS) and ED settings and have found the consistent underdosing of benzodiazepines when used to treat GCSE [10,11]. One analysis of 44 adult patients presenting to the ED found that only one pa- tient received the guideline recommended dosing of a benzodiazepine,

https://doi.org/10.1016/j.ajem.2021.01.094

0735-6757/(C) 2021

which the authors suggested contributed to their low seizure cessation rate of 50% [10]. Another study of 170 adult and pediatric patients pre- senting to the ED found that only 11% of patients received guideline rec- ommended benzodiazepine therapy, all of whom were children [11]. In each of these studies, the small sample size, limited total dose informa- tion, along with limited weight-based dosing information created sub- stantial limitations in ascertaining a more robust understanding of the challenges and opportunities in the management of this condition. The purpose of this study was to assess the benzodiazepine dosing strat- egies for the initial management of GCSE in patients presenting to the ED and the associated compliance with established guideline recommendations.

1. Methods
   1. Study design and setting

This was a single-center, retrospective study evaluating adult (>=18 years) patients who presented to the ED at a large academic Level 1 Trauma center and received IV or IM lorazepam or midazolam for the documented treatment of GCSE between January 1, 2015 and June 1, 2019 (Fig. 1). Patients were excluded if they received a benzodi- azepine for an indication other than GCSE as documented in the ED phy- sician note, were admitted to the hospital prior to receiving therapy, received a benzodiazepine at another facility, had non-convulsive status epilepticus or if the seizure was precipitated by an acute neurological in- jury, a pregnancy-associated condition, hypoglycemia, drug overdose, or cardiac arrest. This investigation was approved by the Institutional Review Board.

• 1. Methods and measurements

Data collection was performed by manual chart review of the elec- tronic medical record (Epic Hyperspace(R) 2018) of patients who re- ceived an IV or IM bolus dose of midazolam or lorazepam and included patient demographics, chief complaint, seizure history, benzo- diazepine regimen, antiepileptic medication administration, seizure characteristics, endotracheal intubation, and hospital admission charac- teristics. Patient noncompliance with antiepileptic Home medications was defined as stated in the medical record or if the chief compliant in- cluded ‘missing doses’, otherwise the patients were assumed to be com- pliant. A seizure was classified as GCSE if there was any mention of “jerking”, “shaking”, or analogous movements or if it was documented as such in the medical record. Glasgow coma scale (GCS) was reported as documented in the medical record and if it was not documented but the patient was described as “fully responsive”, then the patient

was assumed to have a GCS of 15. A seizure that terminated prior to ar- rival was defined as the patient not receiving a benzodiazepine within 5 min of ED arrival. A seizure terminating following any benzodiazepine therapy was defined as the patient not having repeat Benzodiazepine administration within 5 min of the previous dose. Early seizure termina- tion was defined as a patient having seizure cessation following the re- ceipt of their first dose of a benzodiazepine on ED arrival. At the study site, there was no established treatment protocol for SE during the time of the study. Various EMS providers brought patients to the study site and there was no universal protocol for all EMS providers. All ben- zodiazepine doses were converted to lorazepam equivalents (LE): 1 mg of lorazepam = 2 mg of midazolam [12]. Monthly meetings and random patient sample reviews were conducted to ensure adequate ab- stractor monitoring (SLB, SB, RS). All included patients were individu- ally reviewed by the primary investigator (KAW) for accuracy.

• 1. Study outcomes

The primary objective of this study was to assess the benzodiazepine dosing strategies for the initial management of GCSE in patients pre- senting to the ED and associated compliance with established guideline recommendations. guideline compliance was defined as an IV loraze- pam dose of 0.1 mg/kg (maximum: 4 mg/dose) and an IM midazolam dose of 10 mg for patients weighing >40 kg and 5 mg for those weighing 13-40 kg [5,6]. Secondary objectives were to identify benzodi- azepine dosing strategies that resulted in a higher frequency of early sei- zure cessation, identify the contribution of antiepileptic medication administration on seizure termination, and the effect early seizure ter- mination had on subsequent clinical outcomes such has hospital and in- tensive care unit (ICU) admission.

• 1. Statistical analysis

Study data were collected and managed using REDCap (Research Electronic Data Capture). REDCap access was supported by the South Carolina Clinical & Translational Research Institute, with an academic home at the Medical University of South Carolina, through NIH - NCATS Grant Number UL1 TR001450 [13]. Descriptive statistics were used to describe demographic data. Nominal and continuous data were analyzed using chi-square or Fisher’s exact test and Student’s t-test or Mann-Whitney test, respectively. Significance was defined as an alpha = 0.05 and a beta = 0.2. All data analysis was conducted using GraphPad Prism version 8.4.3 for Windows (GraphPad Software, San Diego, California USA, www.graphpad.com).

Fig. 1. Patient flow diagram.

1. Results
   1. Patient population

A total of 222 patients were identified who presented to the ED and received 403 total doses of IV/IM midazolam or lorazepam for the treat- ment of GCSE (Table 1). First-line benzodiazepine therapy in the ED was successful at terminating seizure activity in 86.8% of patients with an av- erage dose of 1.6 mg (0.02 mg/kg) of LE. Patients that experienced sei- zure cessation following their first dose of benzodiazepine upon hospital arrival weighed significantly less (p = 0.037) and were signif- icantly more likely to havea history of seizures (p = 0.013) than those that did not. Patients that achieved early seizure cessation were signifi- cantly less likely be intubated (15.8% vs 50.0%; p < 0.001), be admitted

Table 2

prehospital management undergoing emergency medical services transport.

Characteristics N = 158

Intubation prior to arrival, n (%) 3 (1.9)

Received benzodiazepines prior to arrival, n (%) 41 (25.9) EMS first line benzodiazepine therapy, n (%)

Lorazepam 29 (70.7)

Midazolam 12 (29.3)

EMS first line benzodiazepine therapy route (if documented), n (%)

Lorazepam IV 17 (41.5)

Lorazepam IM 9 (22.0)

Midazolam IV 2 (4.9)

Midazolam IM 3 (7.3)

Mean mg of benzodiazepines per dose, mean mg (SD) 1.95 (0.69) Mean mg/kg per dose of benzodiazepines, mean mg (SD) 0.026 (0.010)

to the ICU (28.8% vs 60.7%; p = 0.002) and be admitted to the hospital

(74.5% vs 92.9%; p = 0.031). A total of 53 patients (23.9%) were intubated, 3 of which were intubated by EMS and 28 were intubated be- fore or after their initial dose of benzodiazepines in the ED.

A total of 158 (71.7%) patients were transported via EMS and, of this subgroup, 41 patients (25.9%) received a benzodiazepine en route (Table 2). Of those that received benzodiazepine therapy prior to their

Mean total benzodiazepines doses administered by EMS per patient, mean doses (SD)

Mean total mg of benzodiazepines administered by EMS per patient, mean mg (SD)

Mean total mg/kg of benzodiazepines administered by EMS per patient, mean (SD)

Seizures terminated prior to ED arrival following benzodiazepine therapy, n (%)

1.34 (0.62)

2.74 (1.78)

0.037 (0.025)

27 (65.9)

arrival to the ED, 65.9% were noted to have seizure termination prior to arrival. Lorazepam (70.7%) was the most frequently administered agent with an average dose of 1.95 mg (0.026 mg/kg). Patients that

EMS: Emergency Medical Services; IV: Intravenous; IM: Intramuscular; SD: Standard Deviation; ED: Emergency Department.

Table 1

Baseline demographics.

Characteristics Total

(n = 222)

Seizure cessation following first ED dose of benzodiazepines?^a

		Yes (n = 184)	No (n = 28)
Age, mean years (SD)	48.8 (16.9)	48.4 (17.0)	51.39 (17.2)	0.382
Total body weight, mean kg (SD)	78.8 (22.5)	77.4 (21.9)	86.99 (26.0)	0.037
Ideal body weight, mean kg (SD)	64.4 (11.1)	65.0 (11.2)	66.34 (10.2)	0.566
Female (%) Ethnicity, n (%) Caucasian/White	89 (40.1) 88 (39.6)	73 (39.7) 74 (40.2)	10 (35.7) 9 (32.1)	0.836 >0.999
Black/African American	130 (58.6)	108 (58.7)	17 (60.7)
Other	4 (1.8)	2 (1.1)	2 (7.1)
History of seizures, n (%)	125 (56.3)	113 (61.4)	10 (35.7)	0.013
History of antiepileptic drug noncompliance, n (%) Precipitating cause of status epilepticus, n (%) Noncompliance	38 (17.1) 34 (15.3)	34 (18.5) 30 (16.3)	2 (7.1) 2 (7.1)	0.180 0.267
Coexisting condition	33 (14.9)	25 (13.6)	6 (21.4)	0.261
Malignancy	12 (5.4)	12 (6.5)	0 (0)	0.374
Stroke	15 (6.8)	11 (6.0)	2 (7.1)	0.683
Other	41 (18.5)	34 (18.5)	7 (25.0)	0.443
Unknown	87 (39.2)	72 (39.1)	11 (39.3)	>0.999
Transport by EMS, n (%)	158 (71.2)	125 (67.9)	23 (82.1)	0.184
Glasgow Coma Score at	15 (3-15)	15 (3-15)	13 (3-15)	0.177
arrival, median (range) Endotracheal Intubation,	53 (23.9)	29 (15.8)	14 (50.0)	<0.001
n (%)
ICU admission, n (%)	80 (36.0)	53 (28.8)	17 (60.7)	0.002
Length of ICU admission,	5.0 (6.3)	4.9 (5.9)	6.7 (8.7)	0.345

p-value

had benzodiazepines administered by EMS received a significantly higher average dose (1.95 mg vs 1.60 mg; p = 0.001) and weight-based dose (0.026 mg/kg vs 0.022 mg/kg; p = 0.023) per ad- ministration en route, compared to the doses received upon hospital ar- rival. No patients that received benzodiazepines prior to hospital arrival received guideline recommended dosing.

The most common first-line benzodiazepine administered upon hospital arrival was IV lorazepam (93.9%; Table 3). A total of 6 doses (1.5%) were found to conform with guideline recommendations, all of which were IV lorazepam administered upon ED arrival. All patients that received guideline recommended doses experienced seizure cessa- tion following the dose and only one required further benzodiazepine dosing while in the ED. Overall, no significant difference was noted be- tween the first-line total doses (1.56 mg vs 1.73 mg; p = 0.132, respec- tively) or weight-based doses (0.022 mg/kg vs 0.022 mg; p = 0.761, respectively) of those experiencing early seizure cessation and those that did not. Patients that achieved early seizure cessation were signifi- cantly more likely to receive fewer benzodiazepine doses in the ED (1.5 vs 2.3; p < 0.001).

The most common IV antiepileptic drug administered was levetirac- etam (60.9%). No significant difference was found in early seizure termi- nation between those receiving an IV antiepileptic drug and those that did not (60.9% vs 57.1%; p = 0.836). Although no significant difference was found between the first dose of benzodiazepine therapy in those that had early seizure termination and those that did not (1.59 mg vs 1.58 mg; p = 0.883), those that received a dose of an antiepileptic drug before or immediately after first-line benzodiazepine therapy in the ED received significantly fewer subsequent benzodiazepine doses (1.24 vs. 1.83; p < 0.001). They were also significantly less likely to be intubated (9.30% vs. 27.78%; p = 0.001), be admitted to the ICU (22.09% vs. 40.48%; p = 0.007), and be admitted to the hospital (67.44% vs. 83.33%; p = 0.008) than those that did not.

mean days (SD)

Hospital Admission, n (%) 173 (77.9) 137 (74.5) 26 (92.9) 0.031

1. Discussion

Length of hospital admission, mean days (SD)

5.8 (7.6) 5.161 (6.7) 8.923 (11.4) 0.022

Our analysis of 222 patients presenting to the ED with GCSE who re- ceived 403 total doses of benzodiazepines found that only 1.5% received

ED: Emergency Department; SD: Standard Deviation; EMS: Emergency Medical Services;

ICU: Intensive Care Unit.

^a 10 patients excluded due to intubation prior to arrival or intubated prior to first dose of benzodiazepines in the Emergency Department.

guideline appropriate dosing of with an average dose of <2 mg. The achievement of early seizure cessation was found to be associated with multiple significant clinical outcomes, including a lower rate of

Table 3

Emergency department arrival therapy.

Characteristics Seizure cessation following first ED dose of

benzodiazepines?^a

p-value

	Yes (n = 184)	No (n = 28)
Received guideline appropriate benzodiazepine dosing, n (%)	4 (2.2)	2 (7.1)	0.180
Pre-hospital Benzodiazepine Doses	0 (0)	0 (0)
First Benzodiazepine Dose	2 (1.1)	0 (0)
Second Benzodiazepine Dose	1 (0.5)	1 (3.6)
Third Benzodiazepine Dose	0 (0)	1 (3.6)
Fourth Benzodiazepine Dose	1 (0.5)	0 (0)
ED first benzodiazepine dose route, n (%)			0.387
Lorazepam IV	174 (94.6)	25 (89.3)
Lorazepam IM	6 (3.3)	1 (3.6)
Midazolam IV	2 (1.1)	1 (3.6)
Midazolam IM	2 (1.1)	1 (3.6)
Mean mg per first benzodiazepine dose, mean mg (SD)	1.56 (0.58)	1.732 (0.48)	0.132
Mean mg/kg per first benzodiazepine dose, mean mg (SD)	0.022 (0.013)	0.022 (0.011)	0.761
Patients that had >=2 benzodiazepine doses administered in the ED, n (%)	n = 166b; 62 (37.3)	n = 25b; 24 (96.0)	<0.001
Mean total benzodiazepines doses administered in the ED per patient, mean doses (SD)	1.49 (0.78)	2.25 (0.80)	<0.001
Mean total mg of benzodiazepines per dose administered in the ED, mean mg (SD)	1.57 (0.57)	1.90 (0.73)	0.008
Mean total milligrams/kg per dose administered in the ED, mean mg (SD)	0.022 (0.012)	0.024 (0.012)	0.283
Mean total mg of benzodiazepines administered in the ED per patient, mean mg (SD)	2.36 (1.58)	4.11 (1.75)	<0.001
Mean total mg/kg of benzodiazepines administered in the ED per patient, mean (SD)	0.033 (0.024)	0.054 (0.032)	<0.001
IV Antiepileptic drugs administered, n (%)	112 (60.9)	16 (57.1)	0.836
Timing of IV antiepileptic drugs administered, n (%)
Before/After First Benzodiazepine Dose	84 (75.0)	2 (12.5)	<0.001
After Second Benzodiazepine Dose	25 (22.3)	10 (62.5)	0.002
After Third Benzodiazepine Dose	2 (1.8)	4 (25.0)	0.002
After Fourth Benzodiazepine Dose	1 (0.9)	0 (0)	>0.999
IV antiepileptic drugs administered, n (%)c Levetiracetam	70 (62.5)	8 (50.0)	0.323
Fosphenytoin/Phenytoin	42 (37.5)	11 (68.8)	0.028
Valproic acid	9 (8.0)	0 (0)	0.601
Lacosamide	3 (2.7)	0 (0)	>0.999

ED: Emergency Department; IV: Intravenous; IM: Intramuscular; SD: Standard Deviation.

^a 10 patients excluded due to intubation prior to arrival or intubated prior to 1st ED dose of benzodiazepines.

^b Patients excluded due to sequential intubation during their ED stay.

^c Patients could have received multiple antiepileptic drugs.

intubation, ICU admission, and hospital admission. Further, the receipt of an antiepileptic drug before or immediately after first-line benzodiaz- epine therapy was noted to result in significantly fewer subsequent benzodiazepine doses and was associated with a significantly lower in- cidence of intubation, ICU admission, and hospital admission.

To our knowledge, this is the largest analysis of current real-world practices in the ED regarding benzodiazepine dosing for GCSE. This re- search expands upon, and compliments, the work of previous investiga- tors who also identified the consistent underdosing of benzodiazepines in the ED setting [10,11]. In their analysis of 170 adult and pediatric pa- tients from two EDs, Langer et al. found that only 11% of patients were compliant with guideline recommendations for benzodiazepine dosing, all of whom were pediatric patients [11]. First-line benzodiazepine ther- apy was effective in 56% of the patients reviewed, however no informa- tion is provided regarding the average and weight-based doses of these agents. The majority of the patients analyzed were admitted to the ICU (63%) with 44% of patients requiring intubation. Although this study sheds light on the frequent underdosing of benzodiazepines in clinical practice, it is challenging to interpret the extent of guideline deviation in the absence of benzodiazepine doses and patient weights. To further characterize these results, Braun et al. analyzed 44 adult patients who presented to the ED for GCSE [10]. They also found the consistent underdosing of benzodiazepines, with only one patient receiving guide- line appropriate therapy and an overall mean lorazepam dose in the ED of 2 mg. Similar to Langer et al., the authors observed a seizure cessation rate with initial benzodiazepine doses of only 50% and an Intubation rate of 52.3%. This study provided more extensive data regarding aver- age benzodiazepine dosing and patient weights; however, its small sample size limited the analysis of dosing details, including mg/kg doses, and did not include a review of antiepileptic drug administration.

Our analysis similarly identified a high incidence of benzodiazepine underdosing by both EMS and ED providers. However, it expands upon the existing data by providing a larger and more expansive analy- sis of benzodiazepine dosing details in combination with the contribu- tion of antiepileptic drug administration. The overall mean lorazepam dose in our analysis was <2 mg, or a mean 0.022 mg/kg of total body weight. Subsequent benzodiazepine doses followed along a similar dos- ing scheme. Those that did not have seizure cessation following their first benzodiazepine dose received significantly more benzodiazepine doses, significantly more total milligrams of benzodiazepines, and a sig- nificantly higher weight-based total dose of benzodiazepines.

We did note a much higher early seizure cessation rate for first-line benzodiazepine therapy in the ED (86.8%) and a comparatively low rate of intubation (23.9%) to that of the above studies. The seizure cessation rates we identified are more in line with prospective investigations ex- ploring the emergent treatment of SE, where early seizure cessation rates have been documented between 59.1 and 73.4% with benzodiaze- pine therapy [14-16]. The limited dosing information provided in both investigations makes it challenging to fully ascertain the underlying eti- ology for this variance, however it may be attributable to the confound- ing factors of the pediatric population in the Langer et al. study and prehospital care in the Braun et al. study that had a much higher rate of prehospital IV midazolam usage compared with our results (48% vs 4.9%). The relatively low intubation rate noted in our analysis is likely a consequence of the comparatively high early seizure cessation rate as it has been found that the rate of respiratory depression in patients with SE is more likely a consequence of untreated status epilepticus than benzodiazepine therapy [6]. Our intubation rate is also more con- sistent with prospective studies of prehospital benzodiazepine adminis- tration for SE that have reported intubation rates <20% [14,15]. Braun

et al. attributed their higher intubation rates to their patients having a median GCS of 5, which is substantially lower than our median GCS of

15. This is congruent with a review of patients admitted to the ICU for SE who had a median GCS of 6 and a concomitant 60.2% intubation rate [17].

Unlike many of the emergent Disease states encountered in the ED, there exist multiple clinical guidelines for the acute management of SE [5,6]. The guidelines put forth by the Neurocritical care Society endorse benzodiazepines as the treatment of choice for initial management [5]. Lorazepam IV is the preferred agent and route, however midazolam IM is also recommended if IV access is unavailable. They recommend a lorazepam dose of 0.1 mg/kg IV (maximum: 4 mg/dose) repeated every 5-10 min as needed. For midazolam, they recommend

0.2 mg/kg IM up to a maximum of 10 mg. These recommendations are also echoed by the American Epilepsy Society [6]. However, it is impor- tant to note that although both organizations provide dosing recom- mendations, there are no dose-ranging studies available identifying the ideal dose and so these recommendations are based largely on ob- servational data and expert opinion. Nevertheless, the consistent find- ing of benzodiazepine underdosing in studies despite clear guidance is perplexing. One theory put forth is that underdosing is secondary to the perceived risk of the cardiorespiratory compromise associated with benzodiazepines [14]. Interestingly the converse is probably more likely, with one randomized, double-blind trial exploring the prehospital administration of IV benzodiazepines and finding that respi- ratory and circulatory complications in the treatment groups to be 10.3-10.6% compared with 22.5% in the placebo group (p = 0.08) [14]. Another possible explanation for the lack of compliance with avail- able guidelines may be the discrepancies in dosing noted across the pri- mary literature supporting benzodiazepines as first-line agents with associated limited variations in seizure cessation success. Although the VA Cooperative Study utilized the recommended dosing of IV loraze- pam 0.1 mg/kg (maximum: 4 mg/dose) and achieved a seizure cessa- tion rate of 64.9% and the RAMPART trial used a flat dose of IV lorazepam 4 mg and achieved a seizure cessation rate of 63.4%, the Out-Of-Hospital study used a flat dose of IV lorazepam of 2 mg and achieved a seizure cessation rate of 59.1% [14-16]. Providing even fur- ther dosing variation, the more recent publication of the Established Status Epilepticus Treatment Trial (ESETT), comparing levetiracetam, fosphenytoin, or valproate in the treatment of GCSE that was unrespon- sive to treatment with benzodiazepines, considered the Cumulative dose administered rather than individual doses when determining if benzodiazepine dosing was consistent with guideline recommenda- tions [18]. One group analyzed the details of the actual benzodiazepine dosing of 207 adult and pediatric patients enrolled in ESETT and found 511 benzodiazepine administrations, an average of 2.47 doses per sub- ject [9]. In adults specifically, only 14.7% of initial lorazepam doses and 15.4% of initial midazolam doses were consistent with guideline recom- mendations. It is important to note that all of these patients were en- rolled in ESETT because they were refractory to benzodiazepine therapy. Nevertheless, the sanctioned underdosing of benzodiazepines in the ESETT trial further contributes to the confusion surrounding the optimal dosing of benzodiazepines in the treatment of SE.

The Neurocritical Care Society guidelines go on to state that an anti-

epileptic drug is required following the administration of benzodiaze- pines unless a modifiable factor is noted and can be corrected (e.g., hypoglycemia), with a goal of rapid attainment of therapeutic drug concentrations [5]. The guidelines do not provide an explicit rec- ommendation for which antiepileptic drug is preferred, however they do suggest that fosphenytoin, valproate sodium, or levetiracetam are vi- able options. This seemingly Rapid progression to an antiepileptic drug may be reasonable when considering the excessive morbidity and mor- tality associated with SE, and therefore the potential risks associated with ongoing SE may exceed those associated with overtreatment [14]. Some have even suggested that early polytherapy may potentially be more effective and less toxic than monotherapy [19,20]. In situations

of delayed initial therapy, other authors have proposed “catch-up dos- ing” or the combined dosing of first and second line medications [3]. Our analysis found that, despite no significant difference existing be- tween initial benzodiazepine doses, those that received an antiepileptic drug before or immediately after first-line benzodiazepine therapy re- quired significantly fewer subsequent benzodiazepines doses (p < 0.001). They were also significantly less likely to be intubated (p = 0.001), admitted to the ICU (p = 0.007), or the hospital (p = 0.008). The early combination of a first-line benzodiazepine with an an- tiepileptic drug may help mitigate some of the consequences of the con- sistent underdosing of benzodiazepines in the management of GCSE in the ED. An analogous situation was seen with ESETT. Despite the benzo- diazepine underdosing noted earlier, and a median duration of seizure activity of around 60 min prior to enrollment, the administration of an- tiepileptic drugs still led to seizure cessation in approximately 50% of patients [18].

The primary objective of our study was to assess benzodiazepine dosing strategies for patients with GCSE in our ED and we found that out of 403 doses, 98.5% were inconsistent with guideline recommenda- tions. As discussed, this underdosing appears to be a consistent oppor- tunity across EDs and some have attempted to address this issue by developing quality improvement methodologies within the hospital setting in order to improve the adherence to guidelines and protocols [21,22]. Following the implementation of this methodology, one hospi- tal was able to reduce their time to benzodiazepine therapy and reduce ICU admissions [22]. Another group developed a simple linear protocol that resulted in a 92.7% compliance rate and an 86.3% seizure termina- tion rate following 1-2 benzodiazepine doses [23]. Our analysis also noted that none of the prehospital benzodiazepine doses were compli- ant with guideline recommendations, which also appears to be a consis- tent opportunity across the nation. One analysis of Treatment protocols from 33 EMS systems found that only 2 EMS system protocols recom- mended the appropriate IM midazolam dose and only 2 protocols in- cluded IV lorazepam, and they did so at a dose of 2-4 mg [24]. Guideline compliant protocol development is also necessary for EMS providers and it directly impacts subsequent treatment in the ED, as prehospital treatment is associated with shorter SE duration and fewer recurrent seizures upon arrival in the ED [25,26].

4.1. Limitations

There are several limitations to our analysis, largely deriving from the retrospective study design and the Emergent care of the condition investigated. The dosing and sequence of therapy could have been af- fected by the post-event documentation process that is frequently uti- lized in the ED. Successful early seizure cessation may also be related to the duration of the seizure prior to hospital arrival rather than the dose of the benzodiazepine, something we were unable to capture ret- rospectively. There was no defined protocol for GCSE care in our ED and so this allowed for a wide variety of medications, doses, and routes of administration which made analysis and interpretation challenging. However, this also provides for a better reflection of physician decision-making without external influences. As this study was focused on emergent ED care, an encephalogram was not readily available for all patients to aid in Treatment decisions, allowing for the possibility of un- diagnosed treatment failures if patients progressed to nonconvulsive status epilepticus following initial therapy. Further, this study relied on the diagnosis of the treating physician at time of therapy and there- fore it is possible that some included patients had non-epileptic activity, however this would not have altered the primary assessment of guide- line compliance.

5. Conclusions

In our analysis of 222 patients presenting to the ED for GCSE and re- ceiving 403 total doses of benzodiazepines, we found that only 1.5% of

doses were complaint with existing guidelines. The achievement of early seizure cessation was found to be associated with multiple signif- icant clinical outcomes, including a lower rate of intubation, ICU admis- sion, and hospital admission. Regardless of benzodiazepine dosing, the early receipt of an antiepileptic drug resulted in significantly fewer sub- sequent benzodiazepine doses and was associated with a significantly lower incidence of intubation, ICU admission, and hospital admission. Future investigations should explore the optimal total and weight- based dosing strategies for benzodiazepines as well as the impact of early antiepileptic drug administration. In the interim, there exists a clear opportunity to develop systems to ensure benzodiazepine dosing is consistent with currently available evidence and recommendations.

Funding

This research did not receive any additional grants from funding agencies in the public, commercial, or not-for-profit sectors.

Declaration of Competing Interest

The authors have none to disclose.

References

1. Pallin DJ, Goldstein JN, Moussally JS, et al. Seizure visits in US emergency depart- ments: epidemiology and potential disparities in care. Int J Emerg Med. 2008 Jun;1 (2):97-105. https://doi.org/10.1007/s12245-008-0024-4 (PubMed PMID: 19384659; PubMed Central PMCID: PMCPMC2657249).
2. Betjemann JP, Lowenstein DH. Status epilepticus in adults. Lancet Neurol. 2015 Jun; 14(6):615-24. https://doi.org/10.1016/S1474-4422(15)00042-3 (PubMed PMID:

25908090).

1. Gainza-Lein M, Fernandez IS, Ulate-Campos A, et al. Timing in the treatment of sta- tus epilepticus: from basics to the clinic. Seizure. 2019 May;68:22-30. https://doi. org/10.1016/j.seizure.2018.05.021 (PubMed PMID: 29884518).
2. Claassen J, Goldstein JN. Emergency neurological life support: status Epilepticus. Neurocrit Care. 2017 Sep;27(Suppl. 1):152-8. https://doi.org/10.1007/s12028-017-

0460-1 (PubMed PMID: 28913605).

1. Brophy GM, Bell R, Claassen J, et al. Guidelines for the evaluation and management of status epilepticus. Neurocrit Care. 2012 Aug;17(1):3-23. https://doi.org/10. 1007/s12028-012-9695-z (PubMed PMID: 22528274).
2. Glauser T, Shinnar S, Gloss D, et al. Evidence-based guideline: Treatment of convul- sive status epilepticus in children and adults: Report of the Guideline Committee of the American Epilepsy Society. Epilepsy Curr. 2016 Jan-Feb;16(1):48-61. https://doi. org/10.5698/1535-7597-16.1.48 (PubMed PMID: 26900382; PubMed Central PMCID: PMCPMC4749120).
3. Treiman DM. The role of benzodiazepines in the management of status epilepticus. Neurology. 1990 May;40(5 Suppl 2):32-42 (PubMed PMID: 2159132).
4. Alvarez V, Lee JW, Drislane FW, et al. Practice variability and efficacy of clonazepam, lorazepam, and midazolam in status epilepticus: A multicenter comparison. Epilepsia. 2015 Aug;56(8):1275-85. https://doi.org/10.1111/epi.13056 (PubMed PMID: 26140660; PubMed Central PMCID: PMCPMC4877129).
5. Sathe AG, Tillman H, Coles LD, et al. Underdosing of benzodiazepines in patients with status Epilepticus enrolled in established status epilepticus treatment trial. Acad Emerg Med. 2019 Aug;26(8):940-3. https://doi.org/10.1111/acem.13811 (PubMed PMID: 31161706).
6. Braun J, Gau E, Revelle S, et al. Impact of non-guideline-based treatment of status ep- ilepticus. J Neurol Sci. 2017 Nov 15;382:126-30. https://doi.org/10.1016/j.jns.2017.

09.031 (PubMed PMID: 29111005).

1. Langer JE, Fountain NB. A retrospective observational study of current treatment for generalized convulsive status epilepticus. Epilepsy Behav. 2014 Aug;37:95-9. https://doi.org/10.1016/j.yebeh.2014.06.008 (PubMed PMID: 25010323).
2. Chouinard G. Issues in the clinical use of benzodiazepines: potency, withdrawal, and rebound. J Clin Psychiatry. 2004;65(Suppl. 5):7-12 (PubMed PMID: 15078112).
3. Harris PA, Taylor R, Thielke R, et al. Research electronic data capture (REDCap)-a metadata-driven methodology and workflow process for providing translational re- search informatics support. J Biomed Inform. 2009 Apr;42(2):377-81. https://doi. org/10.1016/j.jbi.2008.08.010 (PubMed PMID: 18929686; PubMed Central PMCID: PMCPMC2700030).
4. Alldredge BK, Gelb AM, Isaacs SM, et al. A comparison of lorazepam, diazepam, and placebo for the treatment of out-of-hospital status epilepticus. N Engl J Med. 2001 Aug 30;345(9):631-7. https://doi.org/10.1056/NEJMoa002141 (PubMed PMID: 11547716).
5. Silbergleit R, Durkalski V, Lowenstein D, et al. Intramuscular versus intravenous therapy for prehospital status epilepticus. N Engl J Med. 2012 Feb 16;366(7): 591-600. https://doi.org/10.1056/NEJMoa1107494 (PubMed PMID: 22335736; PubMed Central PMCID: PMCPMC3307101).
6. Treiman DM, Meyers PD, Walton NY, et al. A comparison of four treatments for gen- eralized convulsive status epilepticus. Veterans affairs status Epilepticus cooperative study group. N Engl J Med. 1998 Sep 17;339(12):792-8. https://doi.org/10.1056/ NEJM199809173391202 (PubMed PMID: 9738086).
7. Tobochnik S, Gutierrez C, Jacobson MP. Characteristics and acute outcomes of ICU patients with initial presentation of seizure. Seizure. 2015 Mar;26:94-7. https:// doi.org/10.1016/j.seizure.2015.01.016 (PubMed PMID: 25799909).
8. Kapur J, Elm J, Chamberlain JM, et al. Randomized trial of three anticonvulsant med- ications for status epilepticus. N Engl J Med. 2019 Nov 28;381(22):2103-13. https:// doi.org/10.1056/NEJMoa1905795 (PubMed PMID: 31774955; PubMed Central PMCID: PMCPMC7098487).
9. Wasterlain CG, Baldwin R, Naylor DE, et al. Rational polytherapy in the treatment of acute seizures and status epilepticus. Epilepsia. 2011 Oct;52(Suppl. 8):70-1. https:// doi.org/10.1111/j.1528-1167.2011.03243.x (PubMed PMID: 21967369; PubMed Central PMCID: PMCPMC4540642).
10. Kamppi L, Ritvanen J, Mustonen H, et al. Delays and factors related to cessation of generalized convulsive status epilepticus. Epilepsy Res Treat. 2015;2015:591279. https://doi.org/10.1155/2015/591279 (PubMed PMID: 26347816; PubMed Central PMCID: PMCPMC4546976).
11. Hill CE, Parikh AO, Ellis C, et al. Timing is everything: Where status epilepticus treat- ment fails. Ann Neurol. 2017 Aug;82(2):155-65. https://doi.org/10.1002/ana.24986 (PubMed PMID: 28681473; PubMed Central PMCID: PMCPMC5823514).
12. Ostendorf AP, Merison K, Wheeler TA, et al. Decreasing seizure treatment time through quality improvement reduces critical care utilization. Pediatr Neurol. 2018 Aug;85:58-66. https://doi.org/10.1016/j.pediatrneurol.2018.05.012 (PubMed PMID: 30054195).
13. Tourigny-Ruel G, Diksic D, Mok E, et al. Quality assurance evaluation of a simple lin- ear protocol for the treatment of impending status epilepticus in a pediatric emer- gency department 2 years postimplementation. CJEM. 2014 Jul;16(4):304-13. https://doi.org/10.2310/8000.2013.131131 (PubMed PMID: 25060084).
14. Betjemann JP, Josephson SA, Lowenstein DH, et al. Emergency medical services pro- tocols for generalized convulsive status epilepticus. JAMA. 2019 Mar 26;321(12): 1216-7. https://doi.org/10.1001/jama.2019.0441 (PubMed PMID: 30912825; PubMed Central PMCID: PMCPMC6439684).
15. Alldredge BK, Wall DB, Ferriero DM. Effect of prehospital treatment on the outcome of status epilepticus in children. Pediatr Neurol. 1995 Apr;12(3):213-6. https://doi. org/10.1016/0887-8994(95)00044-g (PubMed PMID: 7619187).
16. Aranda A, Foucart G, Ducasse JL, et al. Generalized convulsive status epilepticus man- agement in adults: a cohort study with evaluation of professional practice. Epilepsia. 2010 Oct;51(10):2159-67. https://doi.org/10.1111/j.1528-1167.2010.02688.x

(PubMed PMID: 20738380).