The value of BZDs as first-line treatment for SE has been confirmed in multiple prospective trials. BZDs may be administered by parenteral routes (rectal, intranasal, buccal, or intramuscular [IM] formulations) when used in-house or in out-of-hospital settings, or if possible, orally. To date, rectal diazepam (DZP) is the only marketed treatment available for use by nonmedical caregivers in the United States, and buccal midazolam (MDZ) is approved for use in the European Union [13]. Various BZDs and their modes of administration are discussed below.

The most common adverse effects of BZDs are respiratory depression and systemic hypotension, and both are dose-related. A less common adverse effect is cardiac arrhythmia. The rate of adverse effects ranges between 12 and 53% in a few studies of generalized CSE [14–16].

DZP is highly lipid soluble and is rapidly absorbed by the mucosa, with good penetrance of the CNS. Peak plasma levels following rectal administration are reached in 10 to 60 min, with a bioavailability of 80–98% [17, 18].

Multiple studies have investigated the efficacy of rectal DZP in the treatment of acute repetitive seizures. The North American Diastat Study Group found that 55% of patients who received rectal DZP became seizure-free, compared to 34% given placebo (p = 0.03) [19]. Another study noted seizure termination in 64% of patients who received rectal DZP, compared to 24% of those given placebo (p < 0.0001) [20]. Both studies showed that rectal DZP was safe, with no significant difference in the rates of adverse events between the two groups, and with drowsiness the most common adverse event.

Buccal Midazolam. Buccal MDZ is another BZD formulation that may be used in out-of-hospital settings—with a practical advantage. It consists of liquid MDZ applied over the gums, which are highly vascularized, allowing for rapid absorption and avoidance of first-pass metabolism [21]. A few studies have compared the efficacy of buccal MDZ to rectal DZP. One found no significant difference in seizure cessation within 10 min between buccal MDZ (75%) and rectal DZP (59%) (p = 0.16) in institutionalized children with severe symptomatic epilepsy [22]. In another study of 177 children presenting to the emergency room with acute seizures, buccal MDZ was more effective than rectal DZP (56% vs. 27%), with similar rates of adverse events [23]. This mode of administration was easier to administer and more acceptable socially than the rectal formulation. Buccal MDZ was compared to intravenous (i.v) DZP in children in one study, with no difference in overall efficacy (defined as complete cessation of seizures 5 min after administration of the study drug) [24].

Intranasal Midazolam. Intranasal MDZ makes use of the large surface area of the mucus membranes of the nasopharynx, allowing for rapid absorption and avoidance of first-pass metabolism [25]. Peak plasma concentrations were achieved about 14 min after administration, with a bioavailability reaching 85%. A prospective randomized trial in children with febrile seizures compared the efficacy of intranasal MDZ to that of IVDZP in the emergency room [26]. The two treatments had equal efficacy controlling seizures (88% with MDZ; 92% with DZP), with no significant adverse effects in either group. Time to administration of intranasal MDZ was faster, but the time period between drug administration and seizure cessation was shorter for the IVDZP group. Another trial compared use of intranasal MDZ to rectal DZP in 92 children for prehospital seizures lasting more than 5 min [27]. There was no difference in the total seizure time after medication administration of the two therapies.

Intranasal Lorazepam. Intranasal lorazepam (LZP) is similar to intranasal MDZ, with peak plasma concentrations achieved in 30 min and a bioavailability of about 80% [28]. A study of children with ongoing seizures in the emergency room compared IVLZP to intranasal LZP, both at a dose of 0.1 mg/kg, with a maximum dose of 4 mg [29]. There was no difference detected between the two formulations of LZP based on clinical seizure remission within 10 min of administration of the study drug, establishing the non-inferiority of intranasal LZP compared to IVLZP.

Sublingual Lorazepam. Sublingual LZP was compared to rectal DZP in a randomized controlled trial conducted across nine hospitals in Sub-Saharan Africa involving 436 children [30]. The efficacy of sublingual LZP (56%) was significantly lower than that for rectal DZP (79%) for terminating seizures within 10 min of study drug administration.

Intramuscular Midazolam. MDZ has a benzene ring that opens at an acidic pH, making it water soluble, such that it is absorbed promptly at injection sites. Once in the blood stream, the benzene ring closes upon exposure to the slightly basic physiologic pH, and MDZ becomes lipid soluble [31]. When injected IM, peak plasma concentration is achieved within 23–40 min, with a bioavailability of about 87% [32, 33]. DZP and LZP are lipid soluble and tend to be absorbed slowly and irregularly after IM injection.

A multicenter double-blind randomized non-inferiority trial (the RAMPART, Rapid Anticonvulsant Medication Prior to Arrival Trial) compared IMMDZ to IVLZP in adults and children with SE [34]. The primary outcome [of seizure cessation without need for additional rescue therapy at the time of arrival in the emergency department] was achieved in 73% of subjects in the IMMDZ group compared with 63% of subjects in the IVLZP group, demonstrating the non-inferiority of IMMDZ. Other studies have compared IMMDZ to IVDZP, with a shorter interval to seizure cessation found with IMMDZ but similar overall efficacy for seizure termination [35].

Intravenous Lorazepam and Diazepam. IVLZP is fast acting, with a median latency of about 3–11 min and a long duration of action (12–24 h) [16, 36, 37]. IVDZP has a median onset of action between 2 and 15 min, with rapid entry into the CNS but has a short duration of action, with a drop of the level by about two-thirds in the first 2 h after administration given its large volume of distribution [16, 38, 39]. IVLZP and DZP are the recommended in hospital first-line therapies for CSE in the various practice guidelines.

Out-of-hospital use of IVDZP and LZP has also been investigated. The San Francisco trial of IVBZD administered by paramedics was a randomized, double-blinded, placebo-controlled trial with 3 arms: IVLZP, IVDZP, or placebo, administered to 205 adults with generalized tonic–clonic seizures lasting more than 5 min [15]. IVLZP was effective in terminating seizures by arrival in the emergency department in 59% of cases, compared with 42.6% with IVDZP, and 21% with placebo. There was no statistically significant difference between IVLZP and IVDZP in efficacy, but both were superior to placebo in efficacy and with respect to cardiorespiratory complications.

The Veterans Affairs Status Epilepticus Cooperative Study included 384 patients in overt generalized CSE who were randomized to one of four IV treatment arms: LZP, phenobarbital (PB), DZP plus phenytoin (PHT), or PHT alone [14]. The primary outcome of this study was cessation of all motor and electrographic seizure activity within 20 min of starting the drug infusion, with no recurrence of seizure activity over the next 40 min. LZP stopped the seizure activity in 65% of cases, phenobarbital in 58%, the DZP and PHT combination in 56%, and PHT alone in 44%. Pairwise comparisons between the different arms showed that LZP was superior to the other treatment arms and was easier to administer, hence the study recommended LZP as the best initial therapy for generalized CSE.

Second-line therapy is used when SE has failed to respond to BZDs. For IV administration, this therapy may include PHT, fosphenytoin (FOS), PB, valproic acid (VPA), levetiracetam (LEV), or lacosamide (LCM).

Phenytoin and Fosphenytoin. Phenytoin has long been used in the treatment of SE. The pharmacokinetic profile includes peaking of brain concentrations within 10 min of IV administration and onset of the anticonvulsant effect within 3 min [40, 41].

Fosphenytoin is a PHT prodrug, metabolized following infusion into the active drug PHT [42]; it may be administered IV or IM. FOS has equal effectiveness to PHT and may be administered by a peripheral IV line, given that it causes less local irritation at the infusion site and is less likely to cause tissue necrosis than does PHT if extravasated [43]. The adverse effects associated with FOS include ataxia, dizziness, nystagmus, somnolence, and hypotension and arrhythmia with rapid infusions, all of which are also attributable to PHT—with the exception of paresthesia of the groin, buttocks, or face, seen more with FOS. The recommended loading dose of PHT is usually 20 mg/kg and that of FOS 20 mg “phenytoin equivalent” (PE) per kg. If SE persists after this loading dose, an additional 5–10 mg/kg may be given and the PHT levels monitored.

In several studies of SE and acute repetitive seizures, IVPHT has terminated seizures in about 40–90% of cases [44–46].

Valproic Acid. Valproic acid (VPA) is water soluble, allowing for its administration at a more physiologic pH along with other medications. It can be delivered rapidly with boluses of 20–30 mg/kg/dose infused in under 10 min, with no significant side effects [47, 48]. It causes minimal sedation, respiratory depression, or hypotension.

There have been several studies comparing the use of IVVPA to that of other ASDs as first-line therapy for SE. In one study, VPA was statistically more effective in aborting seizures in patients with CSE than was PHT (66% vs. 42%) [49], but it was similar to PHT (87.8% vs. 88%) in another [50]. In adults with SE that failed to respond to initial BZD therapy, the efficacy of VPA was similar to that of PHT (88% vs. 84%) in one study [51] and similar to that of continuous DZP infusion (56% vs. 50%) in another [52].