The International League Against Epilepsy (ILAE) task force on classification of status epilepticus defined ‘Convulsive Status Epilepticus’ (CSE) as a condition resulting either from the failure of the mechanisms responsible for seizure termination or from the initiation of mechanisms, which lead to abnormally and prolonged seizures (after time point t1; 5 minutes). CSE is a condition, which can have long-term consequences (after time point t2; 30 minutes), including neuronal death, neuronal injury, and alteration of neuronal networks, depending on the type and duration of seizures (Trinka, 2015). Initial treatment of a convulsive status epilepticus consists of first-line therapy with benzodiazepines, followed by second-line therapy with anti-epileptic drugs. Whereas evidence for first-line therapy with benzodiazepines is rather strong, large, randomized trials studying the effect of anti-epileptic drugs were lacking until 2019.

This module evaluates which initial treatment is most effective and beneficial for generalized convulsive status epilepticus in adults. This module contains an update of the module ‘Status Epilepticus bij volwassenen: initiële behandeling’ dated from June 2020 (including a search strategy dated from 1 January 2019).

1. Intravenous levetiracetam versus intravenous phenytoin

1.1. Seizure cessation (critical)

1.2. Termination of SE after drug administration (critical), 1.4. Requirement of ventilatory support (important), 1.5. Hypotension (important), 1.6. Length of stay (important)

1.3. Mortality (important)

2. Intravenous levetiracetam versus intravenous valproate

2.1. Seizure cessation (critical)

2.2. Termination of SE after drug administration (critical)

2.3. Mortality (important)

2.4. Requirement of ventilatory support (important)

2.5. Hypotension (important)

2.6. Length of stay (important)

3. Intravenous fosphenytoin versus intravenous valproate

3.1. Seizure cessation (critical)

3.2. Termination of SE after drug administration (critical)

3.3. Mortality (important)

3.4. Requirement of ventilatory support (important)

3.5. Hypotension (important)

3.6. Length of stay (important)

4. Intravenous lorazepam versus intravenous diazepam

4.1. Seizure cessation (critical)

4.3. Mortality (important)

4.4. Requirement of ventilatory support (important)

4.5. Hypotension (important)

4.2. Termination of SE after drug administration (critical) and 4.6 Length of stay (important)

5. Intravenous levetiracetam versus intravenous lorazepam

5.1. Seizure cessation (critical)

5.2. Termination of SE after drug administration (critical), 5.3 Mortality (important), and 5.6 Length of stay (important)

5.4. Requirement of ventilatory support (important)

5.5. Hypotension (important)

6. Intravenous lorazepam + intravenous valproate versus intravenous lorazepam + intravenous levetiracetam

6.1. Seizure cessation (critical)

6.2. Termination of SE after drug administration (critical), 6.4. Requirement of ventilatory support (important), 6.5. Hypotension (important)

6.3. Mortality (important)

6.6. Length of stay (important)

7. Intravenous levetiracetam + clonazepam versus intravenous placebo + clonazepam

7.1. Seizure cessation (critical)

7.2. Termination of SE after drug administration (critical), 7.4. Requirement of ventilatory support (important), and 7.5. Hypotension (important)

7.3. Mortality (critical)

7.6. Length of stay (critical)

The conclusions below remain maintained (2020), as no new literature was found for the following comparisons:

intravenous diazepam vs. intravenous valproate

intravenous diazepam + intravenous phenytoin vs. intravenous phenobarbital

intravenous lorazepam vs. intravenous phenytoin

intravenous lorazepam vs. intramuscular midazolam

intravenous valproate vs. intravenous phenytoin

intravenous valproate vs. intravenous phenobarbital

intravenous valproate vs. intravenous phenytoin

Description of studies

Chu (2020) performed a systematic review and meta-analyses to evaluate the efficacy and safety of intravenous levetiracetam in patients with status epilepticus. Only RCTs, comparing intravenous levetiracetam to other antiepileptic drugs were included in this systematic review. Six RCTs with a total of 543 patients were included. The definition of ‘convulsive status epilepticus’ varied slightly from study to study. All studies included generalized convulsive status epilepticus participants, whereas three studies (Gujjar, 2017; Misra, 2012; Chakravarthi, 2015) included several patients presenting with focal or subtle convulsive status epilepticus. Four comparisons were available:

• intravenous levetiracetam vs. intravenous phenytoin (Gujjar, 2017; Chakravarthi, 2015);
• intravenous levetiracetam vs. intravenous valproate (Tripathi, 2009);
• intravenous levetiracetam vs. intravenous lorazepam (Misra, 2012); and
• intravenous levetiracetam plus clonazepam vs. intravenous placebo plus clonazepam (Mundlamuria, 2015).

The dose, maintenance dose, and injection timing of levetiracetam and control drugs varied between studies. Mundlamuri (2015) followed their participants for one month after discharge from hospital), and in other studies patients were followed only during their hospital stay. Regarding risk of bias, in five RCTs was selection bias detected and in five RCTs was selective reporting bias detected, but all studies mentioned adequate blinding.

Relevant outcome measures included drug administration to seizure cessation (critical), termination of status epilepticus after drug administration (critical), mortality (important), and requirement for ventilatory support (important).

Kobata (2020) performed a systematic review to compare intravenous diazepam and intravenous lorazepam for the emergency treatment of adult status epilepticus. Two RCTs (Leppik, 1983; Alldredge, 2001) with a total of 204 adult status epilepticus patients at an emergency setting were included. Leppik (1983) compared intravenous diazepam (10 mg) to intravenous lorazepam (4 mg) and Alldredge (2001) compared intravenous diazepam (5 mg) to intravenous lorazepam (2 mg). For both RCTs, when the seizures did not terminate or recurred, second injections of identical doses of the same benzodiazepines were given. The study populations differed: Leppik (1983) included convulsive and non-convulsive SE, whereas Alldredge (2001) included only generalized tonic-clonic seizure. Risk of bias was assessed as ‘high’ for Leppik (1983) and assessed as ‘low’ for Alldredge (2001). Relevant outcome measures include seizure cessation (critical), mortality (important), hypotension (important), respiratory depression (important).

Kapur (2019) performed an RCT to compare the efficacy and safety of intravenous levetiracetam, intravenous fosphenytoin, and intravenous valproate in children and adults with convulsive status epilepticus that were unresponsive to treatment with benzodiazepines. Patients (n=384) were randomized into the intravenous levetiracetam group (n=145, mean age ± SD: 33.3 ± 26.0, 53% male), intravenous fosphenytoin group (n=118, mean age ± SD: 32.8 ± 25.4, 60% male), or intravenous VPA group (n=121, mean age ± SD: 32.2 ± 25.4, 54% male). A total of 39% of the patients who were enrolled were children and adolescents (up to 17 years of age), 48% were adults (18 to 65 years of age), and 13% were older adults (>65 years of age). To answer our clinical question, we extracted the results from the adult subgroup analysis intention to treat analyses were performed. Relevant outcome measures include seizure cessation (critical), mortality (important), hypotension (important), respiratory depression (important).

Nene (2019) conducted an RCT to compare the efficacy of valproate and levetiracetam following initial intravenous lorazepam in elderly patients (>60 years) with generalized convulsive status epilepticus. Intravenous lorazepam was followed by administration of intravenous valproate or levetiracetam based on the treatment arm to which the patient was randomized. All patients received initial intravenous lorazepam (0.1 mg/kg) followed by one of the two antiepileptic drugs: parenteral valproate (20-25 mg/kg) or intravenous levetiracetam (20-25 mg/kg). A total of 118 patients who had generalized convulsive status epilepticus were randomized (valproate: n=60, mean age: 66.6 ± 6.7, 31% female; levetiracetam: n=58, mean age: 68.5 ± 8.0, 41% female). A total of 100 patients (valproate: n=50, mean age: 66.6 ± 6.7, 11% female; levetiracetam n=50) completed the study. Intention to treat analyses were performed. Relevant outcome measures include seizure cessation (critical), mortality (important), hypotension (important), respiratory depression (important).

This module is an update from the module dd. June 2020 (including a search until 1 January 2019). No new literature was found regarding the following comparisons:

• intravenous diazepam + intravenous phenytoin vs. intravenous phenobarbital
• intravenous diazepam + intravenous phenytoin vs. intravenous phenytoin
• intravenous diazepam vs. intravenous valproate
• intravenous lorazepam vs. intramuscular midazolam
• intravenous lorazepam vs. intravenous phenobarbital
• intravenous valproate vs. intravenous phenobarbital
• intravenous valproate vs. intravenous phenytoin

Thus, the previous conclusions remain valid. The description of previous included studies is presented as Appendix 1 (2020).

Results

1. Intravenous levetiracetam versus intravenous phenytoin

1.1 Seizure cessation (critical)

For the comparison between levetiracetam and phenytoin, three studies (Chakravarthi, 2015; Mundlamuri, 2015; Gujjar, 2017) with 196 participants were combined by Chu (2020) in a meta-analysis. Analyses showed that seizure cessation was present in 70 out of 94 (74%) patients in the levetiracetam-group and in 71 out of 102 (70%) phenytoin-group patients, resulting in a risk ratio of 1.09 (95%CI 0.92 to 1.29), see Figure 1. The difference between the groups was not statistically significant nor clinically relevant.

1.2 Termination of SE after drug administration (critical)

Not reported for the comparison between levetiracetam and phenytoin.

1.3 Mortality (important)

Chu (2020) defined ‘mortality’ as in hospital mortality. Analyses showed that hospital mortality was present in 9 out of 94 patients (10%) patients in the levetiracetam-group and in 11 out of 102 patients (11%) phenytoin-group patients, ressulting in a risk ratio of 0.88 (95%CI 0.38 to 2.04), see Figure 2. This difference was not statistically significant but clinically relevant.

The outcomes 1.4 Requirement of ventilatory support (important), 1.5 Hypotension (important), and 1.6 Length of stay (important) were not reported for the comparison between levetiracetam and phenytoin.

Figure 1 Studies comparing intravenous levetiracetam and intravenous phenytoin: seizure cessation.

Figure 2 Studies comparing intravenous levetiracetam and intravenous phenytoin: hospital mortality.

2. Intravenous levetiracetam versus intravenous valproate

2.1 Seizure cessation (critical)

For the comparison between levetiracetam and valproate, two studies (Tripathi, 2010; Mundlamuri, 2015) with 182 participants were combined in a meta-analysis (Chu, 2020). One additional RCT (Kapur, 2019) including 266 patients was added to this comparison. Kapur (2020) assessed cessation of seizures within 60 minutes.

Seizure cessation was present in 135 out of 236 (57%) patients in the levetiracetam-group and in 116 out of 212 (55%) patients in the valproate-group, resulting in a risk ratio of 1.08 (95%CI 0.93 to 1.26), see Figure 3. The difference between the groups was not statistically significant nor clinically relevant.

2.2 Termination of SE after drug administration (critical)

For the comparison between levetiracetam and valproate, one RCT (Kapur, 2019) assessed the time to termination of seizures. This outcome was defined as the interval from the start of infusion of the trial drug to the cessation of clinically apparent seizures.

Data were available for 24 patients (n=14 levetiracetam, n=10 valproate). Median time from start of trial-drug infusion to termination of seizures for patients with treatment success was 10.5 minutes (interquartile range: 5.7 to 15.5) in the levetiracetam-group and 7.0 minutes (interquartile range: 4.6 to 14.9) in the valproate-group. If the median is interpreted as a mean value, then there could potentially be a clinically relevant difference. However, since this is a median (data is skewed), no statement can be made about the clinically relevant difference.

2.3 Mortality (important)

Chu (2020) defined ‘mortality’ as in hospital mortality, including two studies (Tripathi, 2010; Mundlamuri, 2015). One additional RCT (Kapur, 2019) assessed mortality at hospital discharge or 30 days, whichever came first.

Mortality was present in 14 out of 241 (5.8%) patients in the levetiracetam-group and in 8 out of 216 (3.7%) patients in the valproate-group, resulting in a risk ratio of 1.56 (95%CI 0.65 to 3.71), see Figure 4. The difference between the groups was not statistically significant but clinically relevant. Herewith, the very low number of events and the wide confidence interval should be considered.

2.4 Requirement of ventilatory support (important).

Kapur (2019) assessed acute respiratory depression at hospital discharge or 30 days, whichever came first. In the levetiracetam-group in 12 out of 150 patients (8%) and in the valproate-group 10 out of 125 patients (8%) experienced acute respiratory depression. This resulted in a risk ratio is 1.00 (95%CI 0.45 to 2.24). This difference was not statistically significant nor clinically relevant. The very low number of events and the wide confidence interval should be considered.

2.5 Hypotension (important)

For the comparison between levetiracetam versus valproate, Kapur (2019) assessed life-threatening hypotension within 60 min after start of trial drug infusion. In the levetiracetam-group in 1 out of 150 patients (1%) and in the valproate-group 2 out of 125 patients (2%) experienced hypotension.

2.6 Length of stay (important)

For the comparison between levetiracetam versus valproate, one RCT (Kapur, 2019) assessed the length of stay. Length of stay was defined as admission to the ICU; the length of ICU stay, and length of hospital stay.

In the levetiracetam-group 87 out of 145 patients (60%) and in the valproate-group 71 out of 121 patients (59%) were admitted to the ICU, resulting in a risk ratio of 1.02 (95%CI 0.84 to 1.25). Median length of ICU stay was 1 day (interquartile range: 0 to 3) for both groups.

Median length of total hospital stay was 3 days (interquartile range: 1 to 7) in the levetiracetam-group and 3 days (interquartile range: 2 to 6) in the valproate-group. These differences were not statistically significant nor clinically relevant.

Figure 3 Studies comparing intravenous levetiracetam and intravenous valproate: 
seizure cessation.

Figure 4 Studies comparing intravenous levetiracetam and intravenous valproate: 
hospital mortality.

3. Intravenous fosphenytoin versus intravenous valproate

3.1 Seizure cessation (critical)

For the comparison between fosphenytoin and valproate, one RCT (Kapur, 2019) assessed cessation of seizures within 60 minutes. Seizure cessation was found in 53 of 118 (45%) in the fosphenytoin-group and 56 of 121 (46%) in the valproate-group. This resulted in a risk ratio of 0.97 (95%CI 0.74 to 1.28). This difference was not statistically significant nor clinically relevant.

3.2 Termination of SE after drug administration (critical)

For the comparison between fosphenytoin and valproate, one RCT (Kapur, 2019) assessed the time to termination of seizures. This outcome was defined as the interval from the start of infusion of the trial drug to the cessation of clinically apparent seizures.

Data were available for 25 patients (n=15 fosphenytoin, n= 10 valproate). Median time from start of trial-drug infusion to termination of seizures for patients with treatment success was 11.7 minutes (interquartile range: 7.5 to 20.9) in the fosphenytoin-group and 7.0 minutes (interquartile range: 4.6 to 14.9) in the valproate-group. No statement can be made about the clinically relevant difference as these are median values (data is skewed).

3.3 Mortality (important)

For the comparison between fosphenytoin and valproate, one RCT (Kapur, 2019) assessed mortality at hospital discharge or 30 days, whichever came first. In the fosphenytoin-group 3 out of 125 patients (2%) died and in the valproate-group 2 out of 125 patients (2%) died. When valproateis compared to fosphenytoin, the risk ratio is 1.50 (95%CI 0.26 to 8.82). Herewith, the very low number of events (5 in total) and the wide confidence interval should be considered.

3.4 Requirement of ventilatory support (important).

Kapur (2019) assessed acute respiratory depression at hospital discharge or 30 days, whichever came first. In the fosphenytoin-group 16 out of 125 patients (13%) and in the valproate-group 10 out of 125 patients (8%) experienced acute respiratory depression. When fosphenytoin compared to valproate, the risk ratio is 1.60 (95%CI 0.76 to 3.39). This difference was statistically significant but clinically relevant. Caution is needed due to the wide confidence interval.

3.5 Hypotension (important)

For the comparison between fosphenytoin and valproate, Kapur (2019) assessed life-threatening hypotension within 60 min after start of trial drug infusion. In the fosphenytoin-group 4 out of 125 patients (3%), and in the valproate-group 2 out of 125 patients (2%) experienced hypotension.

3.6 Length of stay (important)

For the comparison between fosphenytoin and valproate, one RCT (Kapur, 2019) assessed the length of stay. Length of stay was defined as admission to the ICU; the length of ICU stay, and length of hospital stay.

In the fosphenytoin-group 70 out of 118 patients (59%) and in the valproate-group 71 out of 121 patients (59%) were admitted to the ICU, resulting in a risk ratio of 1.75 (95%CI 1.31 to 2.34). This difference was not statistically significant but clinically relevant.

Median length of ICU stay was 1 day (interquartile range: 0 to 3) in both groups. Median length of total hospital stay was 3 days (interquartile range: 1 to 6) in the fosphenytoin-group and 3 days (interquartile range: 2 to 6) in the valproate-group. These differences were not statistically significant nor clinically relevant.

4. Intravenous lorazepam versus intravenous diazepam

4.1 Seizure cessation (critical)

For the comparison between diazepam and lorazepam, two studies (Leppik, 1983; Alldredge, 2001) with 204 participants were combined in a meta-analysis (Kobata, 2020).

Seizure cessation was present in 54 out of 101 patients (53%) patients in the diazepam-group and in 72 out of 103 patients (70%) lorazepam-group patients, resulting in a risk ratio of 1.24 (95%CI 1.03 to 1.49), see Figure 5. The difference between the groups was statistically significant and clinically relevant.

4.2 Termination of SE after drug administration (critical)

Not reported for the comparison diazepam versus lorazepam.

4.3 Mortality (important)

For the comparison between diazepam and lorazepam, two studies (Leppik, 1983; Alldredge, 2001) with 204 participants were combined in meta-analysis (Kobata, 2020). Kobata (2020) defined ‘mortality’ as mortality at discharge. Mortality at discharge was present in 3 out of 101 patients (3%) in the diazepam-group and in 5 out of 103 patients (5%) lorazepam-group patients, resulting in a risk ratio of 1.72 (95%CI 0.43 to 6.90), see Figure 6. This difference was not statistically significant but clinically relevant. Caution is needed due to the very low number of events and the wide confidence interval.

4.4 Requirement of ventilatory support (important)

Kobata (2020) compared diazepam to lorazepam and assessed ventilatory support based on two studies (Leppik, 1983; Alldredge, 2001), representing 204 patients. Analyses showed that ventilatory support was needed in 10 out of 101 patients (10%) in the diazepam-group and in 11 out of 103 patients (11%) lorazepam-group patients, resulting in a risk ratio of 1.07 (95%CI 0.48 to 2.41), see Figure 7. This difference was not statistically significant nor clinically relevant.

4.5 Hypotension (important)

One study (Leppik, 1983) included in Kobata (2020) assessed ‘hypotension’. Leppik (1983) assessed hypotension in 70 patients. In the levetiracetam-group and diazepam-group, hypotension occurred in 1 out of 37 patients (2.7%) and 0 out of 33 patients (0%), respectively.

4.6 Length of stay (important)

Not reported for the comparison diazepam versus lorazepam.

Figure 5 Studies comparing intravenous Lorazepam and intravenous Diazepam: 
seizure cessation.

Figure 6 Studies comparing intravenous Lorazepam and intravenous Diazepam: 
mortality.

Figure 7 Studies comparing intravenous Lorazepam and intravenous Diazepam: 
requirement of ventilatory support.

5. Intravenous levetiracetam versus intravenous lorazepam

5.1 Seizure cessation (critical)

For the comparison between levetiracetam and lorazepam, one study (Misra, 2012) was selected from the meta-analysis (Chu, 2020). Misra (2012) included 79 patients and defined termination after drug administration at 10 minutes. Seizure cessation was present in 29 out of 38 patients (76%) in the levetiracetam-group and in 31 out of 41 patients (76%) in the lorazepam-group patients, resulting in a risk ratio of 1.01 (95%CI 0.79 to 1.29). This difference between the groups was not statistically significant nor clinically relevant.

5.2 Termination of SE after drug administration (critical)

Not reported for the comparison levetiracetam versus lorazepam.

5.3 Mortality (important)

Not reported for the comparison levetiracetam versus lorazepam.

5.4 Requirement of ventilatory support (important)

Analyses showed that in the levetiracetam-group 4 out of 38 patients (11%) and in the lorazepam-group 10 out of 41 patients (24%) used ventilatory support, resulting in a risk ratio of 0.43 (95%CI 0.15 to 1.26). This difference was not statistically significant but clinically relevant. Caution is needed due to the very low number of events.

5.5 Hypotension (important)

Analyses showed that in the levetiracetam-group 2 out of 38 patients (5%) and in the lorazepam-group 8 out of 41 patients (20%) had hypotension, resulting in a risk ratio of 0.27 (95%CI 0.06 to 1.19). This difference was not statistically significant but clinically relevant. Caution is needed due to the very low number of events.

5.6 Length of stay (important)

Not reported for the comparison levetiracetam versus lorazepam.

6. Intravenous lorazepam + intravenous valproate versus intravenous lorazepam + intravenous levetiracetam

6.1 Seizure cessation (critical)

For the comparison between valproate and levetiracetam following initial intravenous lorazepam, one RCT (Nene, 2019) assessed termination of SE as no recurrence of seizures after half an hour of infusion of study drugs. Intention to treat analysis (n=118) showed that valproate following lorazepam led to seizure control in 41 out of 60 patients (68%), while levetiracetam following lorazepam led to seizure control in 43 out of 58 patients (74%).

This resulted in a risk ratio of 0.92 (95%CI 0.73 to 1.16), this difference was not statistically significant nor clinically relevant.

6.2 Termination of SE after drug administration (critical)

Not reported for the comparison lorazepam + valproate versus lorazepam + levetiracetam.

6.3 Mortality (important)

Nene (2019) assessed 1-month mortality comparing valproate and levetiracetam. The 30-day mortality was comparable in both groups (valproate: n=13, 22.4%; levetiracetam: n=10, 18.2%). this difference resulted in a risk ratio of 1.26 (95%CI 0.60 to 2.64). This difference was not statistically significant but clinically relevant. Herewith, the very low number of events and the wide confidence interval should be considered.

6.4 Requirement of ventilatory support (important)

Not reported for the comparison lorazepam + valproate versus lorazepam + levetiracetam.

6.5 Hypotension (important)

Not reported for the comparison lorazepam + valproate versus lorazepam + levetiracetam.

6.6 Length of stay (important)

Nene (2019) assessed the length of hospital stay comparing valproate and levetiracetam. Mean length of hospital stay was 3.6 days (SD: 1.9) in the valproate-group and 4.4 days (SD: 3.9) in the levetiracetam-group.

7. Intravenous levetiracetam + clonazepam versus intravenous placebo + clonazepam

7.1 Seizure cessation (critical)

For the comparison between adding levetiracetam to clonazepam vs. adding placebo to clonazepam, one study (Navarro, 2016) was selected from the meta-analysis (Chu, 2020). Navarro (2016) included 136 patients and defined termination of SE after drug administration within 15 minutes. Status epilepticus was controlled by levetiracetam in 57 out of 68 patients (84%) and by placebo in 50 out of 68 patients (74%), which resulted in a risk ratio of 1.14 (95%CI 0.96 to 1.36). This difference was not statistically significant, but clinically relevant.

7.2 Termination of SE after drug administration (critical)

Not reported for the comparison levetiracetam and clonazepam vs. adding placebo to clonazepam.

7.3 Mortality (important)

Navarro (2016) assessed mortality at 15 days after admission to hospital or earlier if discharged from hospital. Mortality was comparable in both groups (clonazepam plus levetiracetam: n=3, 5%; clonazepam plus placebo: n=4, 6%). This difference resulted in a risk ratio of 0.74 95%CI 0.17 to 3.17). This difference was not statistically significant but clinically relevant. Herewith, the very low number of events and the wide confidence interval should be considered.

7.4 Requirement of ventilatory support (important)

Not reported for the comparison levetiracetam and clonazepam vs. adding placebo to clonazepam.

7.5 Hypotension (important)

Not reported for the comparison levetiracetam and clonazepam vs. adding placebo to clonazepam.

7.6 Length of stay (important)

Navarro (2016) assessed the length of hospital stay comparing levetiracetam and clonazepam vs. adding placebo to clonazepam. Median length of hospital stay was 10 days (range 1 to 15) in the levetiracetam-group and 10 days (range 1 to 15) in the placebo-group.

Median length of ICU stay was 3 days (range 0 to 15) in the levetiracetam-group and 3 days (range 1 to 15) in the placebo-group.

Level of evidence of the literature

1. Intravenous levetiracetam versus intravenous phenytoin

1.1 Seizure cessation (critical)

Regarding the comparison levetiracetam and phenytoin, the level of evidence regarding the outcome measure ‘seizure cessation’ started as high was downgraded by 3 levels to very low because of study limitations (-1; risk of bias), imprecision (-1; confidence interval crosses boundaries of clinical important difference), and conflicting results between studies (-1; heterogeneity).

1.3 Mortality (important)

Regarding the comparison intramuscular levetiracetam and phenytoin, the level of evidence regarding the outcome measure ‘mortality’ started as high and was downgraded by 3 levels to very low because of study limitations (-1; risk of bias) and imprecision (-2; confidence interval crosses boundaries of clinical important difference and relatively small number of patients).

The outcomes 1.2 Termination of SE after drug administration (critical), 1.4 Requirement of ventilatory support (important), 1.5 Hypotension (important), and 1.6 Length of stay (important) were not reported and could not be graded.

2. Intravenous levetiracetam versus intravenous valproate

2.1 Seizure cessation (critical)

Regarding the comparison between levetiracetam and valproate, the level of evidence regarding the outcome measure ‘seizure cessation’ started as high and was downgraded by 3 levels to low because of study limitations (-1; risk of bias) and imprecision (-2; because of crosses boundaries of clinical important difference).

2.2 Termination of SE after drug administration (critical)

Regarding the comparison between levetiracetam and valproate, the level of evidence regarding the outcome measure ‘termination of SE after drug administration’ started as high and was downgraded by 3 levels to very low because of study limitations (-1; risk of bias) and imprecision (-2; because of a small number of patients from only one study).

2.3 Mortality (important)

Regarding the comparison between levetiracetam and valproate, the level of evidence regarding the outcome measure ‘mortality’ started as high and was downgraded by 3 levels to low because of study limitations (-1; risk of bias) and imprecision (-2; very low number of events, only one study, large confidence interval crosses boundaries of clinical important difference).

2.4 Requirement of ventilatory support (important).

Regarding the comparison between levetiracetam and valproate, the level of evidence regarding the outcome measure ‘ventilatory support’ started as high and was downgraded by 3 levels to low because of study limitations (-1; risk of bias) and imprecision (-2; very low number of events, only one study, large confidence interval crosses boundaries of clinical important difference).

2.5 Hypotension (important).

Regarding the comparison between levetiracetam and valproate, the level of evidence regarding the outcome measure ‘hypotension’ started as high and was downgraded by 3 levels to low because of study limitations (-1; risk of bias) and imprecision (-2; very low number of events, a small number of patients from only one study).

2.6 Length of stay (important).

Regarding the comparison between levetiracetam and valproate, the level of evidence regarding the outcome measure ‘length of stay’ started as high and was downgraded by 3 levels to low because of study limitations (-1; risk of bias) and imprecision (-2; confidence interval crosses boundaries of clinical important difference, a small number of patients from only one study).

3. Intravenous fosphenytoin versus intravenous valproate

3.1 Seizure cessation (critical)

Regarding the comparison between fosphenytoin and valproate, the level of evidence regarding the outcome measure ‘seizure cessation’ started as high and was downgraded by 3 levels to very low because of study limitations (-1; risk of bias) and imprecision (-2; because of crosses boundaries of clinical important difference).

3.2 Termination of SE after drug administration (critical)

Regarding the comparison between fosphenytoin and valproate, the level of evidence regarding the outcome measure ‘termination of SE after drug administration’ started as high and was downgraded by 3 levels to very low because of study limitations (-1; risk of bias) and imprecision (-2; because of a small number of patients from only one study).

3.3 Mortality (important)

Regarding the comparison between fosphenytoin and valproate, the level of evidence regarding the outcome measure ‘mortality’ started as high and was downgraded by 3 levels to very low because of study limitations (-1; risk of bias) and imprecision (-2; very low number of events, only one study, large confidence interval crosses boundaries of clinical important difference).

3.4 Requirement of ventilatory support (important)

Regarding the comparison between fosphenytoin and valproate, the level of evidence regarding the outcome measure ‘ventilatory support’ started as high and was downgraded by 3 levels to very low because of study limitations (-1; risk of bias) and imprecision (-2; very low number of events, only one study, large confidence interval crosses boundaries of clinical important difference).

3.5 Hypotension (important)

Regarding the comparison between fosphenytoin and valproate, the level of evidence regarding the outcome measure ‘hypotension’ started as high and was downgraded by 3 levels to very low because of study limitations (-1; risk of bias) and imprecision (-2; very low number of events, a small number of patients from only one study).

3.6 Length of stay (important)

Regarding the comparison between fosphenytoin and valproate, the level of evidence regarding the outcome measure ‘length of stay’ started as high and was downgraded by 3 levels to very low because of study limitations (-1; risk of bias) and imprecision (-2; confidence interval crosses boundaries of clinical important difference, a small number of patients from only one study).

4. Intravenous lorazepam versus intravenous diazepam

4.1 Seizure cessation (critical)

Regarding the comparison between diazepam and lorazepam, the level of evidence regarding the outcome measure ‘seizure cessation’ started as high and was downgraded by 3 levels to very low because of study limitations (-1; risk of bias) and imprecision (-2; relatively small study population).

4.3 Mortality (important)

Regarding the comparison between diazepam and lorazepam, the level of evidence regarding the outcome measure ‘mortality’ started as high and was downgraded by 3 levels to very low because of study limitations (-1; risk of bias) and imprecision (-2; very low number of events, large confidence interval crosses boundaries of clinical important difference).

4.4 Requirement of ventilatory support (important)

Regarding the comparison between diazepam and lorazepam, the level of evidence regarding the outcome measure ‘ventilatory support’ started as high and was downgraded by 3 levels to very low because of study limitations (-1; risk of bias) and imprecision (-2; very low number of events, large confidence interval crosses boundaries of clinical important difference).

4.5 Hypotension (important)

Regarding the comparison between diazepam and lorazepam, the level of evidence regarding the outcome measure ‘hypotension’ started as high and was downgraded by 3 levels to very low because of study limitations (-1; risk of bias) and imprecision (-2; very low number of events, relatively small study population).

The outcomes 4.2 Termination of SE after drug administration (critical) and 4.6 Length of stay (important) were not reported and could not be graded.

5. Intravenous levetiracetam versus intravenous lorazepam

5.1 Seizure cessation (critical)

Regarding the comparison between levetiracetam and lorazepam, the level of evidence regarding the outcome measure ‘seizure cessation’ started as high and was downgraded by 3 levels to very low because of study limitations (-1; risk of bias) and imprecision (-2; relatively small study population, confidence interval crosses boundaries of clinical important difference).

5.4 Requirement of ventilatory support (important)

Regarding the comparison between diazepam and lorazepam, the level of evidence regarding the outcome measure ‘ventilatory support’ started as high and was downgraded by 3 levels to very low because of study limitations (-1; risk of bias) and imprecision (-2; relatively small study population, confidence interval crosses boundaries of clinical important difference).

5.5 Hypotension (important)

Regarding the comparison between diazepam and lorazepam, the level of evidence regarding the outcome measure ‘hypotension’ started as high and was downgraded by 3 levels to very low because of study limitations (-1; risk of bias) and imprecision (-2; relatively small study population, low number of events, confidence interval crosses boundaries of clinical important difference).

The outcomes 5.2 Termination of SE after drug administration (critical), 5.3 Mortality (important), and 5.6 Length of stay (important) were not reported and could not be graded.

6. Intravenous lorazepam + intravenous valproate versus intravenous lorazepam + intravenous levetiracetam

6.1 Seizure cessation (critical)

Regarding the comparison between valproate and levetiracetam, the level of evidence regarding the outcome measure ‘seizure cessation’ started as high and was downgraded by 3 levels to very low because of study limitations (-1; risk of bias) and imprecision (-2; relatively small study population from only one study, confidence interval crosses boundaries of clinical important difference).

6.3 Mortality (important)

Regarding the comparison between valproate and levetiracetam, the level of evidence regarding the outcome measure ‘mortality’ started as high and was downgraded by 3 levels to very low because of study limitations (-1; risk of bias) and imprecision (-2; relatively small study population from only one study).

6.6 Length of stay (important)

Regarding the comparison between valproate and levetiracetam, the level of evidence regarding the outcome measure ‘length of stay’ started as high and was downgraded by 3 levels to very low because of study limitations (-1; risk of bias) and imprecision (-2; relatively small study population from only one study).

The outcomes 6.2. Termination of SE after drug administration (critical). 6.4. Requirement of ventilatory support (important), 6.5. Hypotension (important) were not reported and could not be graded.

7. Intravenous levetiracetam + clonazepam versus intravenous placebo + clonazepam

7.1 Seizure cessation (critical)

Regarding the comparison between adding levetiracetam to clonazepam vs. adding placebo to clonazepam, the level of evidence regarding the outcome measure ‘seizure cessation’ started as high and was downgraded by 3 levels to very low because of study limitations (-1; risk of bias) and imprecision (-2; relatively small study population from only one study, confidence interval crosses boundaries of clinical important difference).

7.3 Mortality (important)

Regarding the comparison between adding levetiracetam to clonazepam vs. adding placebo to clonazepam, the level of evidence regarding the outcome measure ‘mortality’ started as high and was downgraded by 3 levels to very low because of study limitations (-1; risk of bias) and imprecision (-2; relatively small study population from only s study, very low number of events).

7.6 Length of stay (important)

Regarding the comparison between adding levetiracetam to clonazepam vs. adding placebo to clonazepam, the level of evidence regarding the outcome measure ‘length of stay’ started as high and was downgraded by 3 levels to very low because of study limitations (-1; risk of bias) and imprecision (-2; relatively small study population from only one study).

The outcomes 7.2. Termination of SE after drug administration (critical), 7.4. Requirement of ventilatory support (important), and 7.5. Hypotension (important) were not reported and could not be graded.

A systematic review of the literature was performed to answer the following question: 
Which type, dosage and form of medication is most effective in the treatment of status epilepticus in adults with generalized convulsive status epilepticus?

 

P: adults with status epilepticus

I: anti-epileptic medication (such as diazepam, lorazepam, midazolam, brivaracetam, lacosamide, levetiracetam, phenytoin, valproate and other, new anti-epileptic medication)

C: same medication but other dosage and/or form, other type of medication, no treatment/placebo

O: (time from) drug administration to seizure cessation; (duration of) persistent seizure cessation (dichotomous and continuous); requirement for ventilatory support, hypotension, patient satisfaction, mortality, length of hospital stay

Relevant outcome measures

The working group defined convulsive status epilepticus as stated in the ILAE classification of 2015: “Status epilepticus is a condition resulting either from the failure of the mechanisms responsible for seizure termination or from the initiation of mechanisms, which lead to abnormally, prolonged seizures (after time point t₁). It is a condition, which can have long-term consequences (after time point t₂), including neuronal death, neuronal injury, and alteration of neuronal networks, depending on the type and duration of seizures.” (Trinka, 2015). For convulsive status epilepticus time point t₁ is 5 minutes and time point t₂ is 30 minutes.

The working group considered (time from drug administration to) seizure cessation and (duration of) continued seizure cessation as a critical outcome measure for decision making; mortality, requirement for ventilatory support, hypotension, length of hospital stay, and patient satisfaction as an important outcome measure for decision making.

A priori, the working group did not define the outcome measures listed above but used the definitions used in the studies. Patient satisfaction had to be assessed using validated instruments. Length of hospital stay is defined as the overall length of time a patient is hospitalized.

Per outcome, the working group defined the following differences as a minimal clinically (patient) important differences:

Dichotomous outcomes (relative risk; yes/no):

• From drug administration to seizure cessation within a certain time period: ≥10%
• Persistent seizure cessation at least one hour: ≥10%
• Respiratory depression (requirement for ventilatory support): ≥10%
• Mortality: ≥10%

Continuous outcomes:

• Time from drug administration to seizure cessation: ≥2 minutes
• Duration of persistent seizure cessation: ≥30 minutes
• Hypotension: ≥10%
• Patient satisfaction: ≥10% difference on a validated scale
• Length of hospital stay: ≥2 days

Search and select (Methods)

The search strategy, which dated from 01-01-2019 was updated. The databases Medline (via OVID) and Embase (via Embase.com) were searched with relevant search terms from 01-01-2019 (last search) until 21-06-2021. The detailed search strategy is depicted under the tab Methods. One search was conducted for all modules concerning the treatment of status epilepticus. The search was limited to systematic reviews and randomized controlled trials. The systematic literature search resulted in 305 hits. Studies for this module were selected based on the following criteria:

• systematic review (searched in at least two databases, and detailed search strategy, risk of bias assessment and results of individual studies available) or randomized controlled trial;
• full-text English language publication;
• adults aged ≥ 18 years;
• studies including ≥ 20 (ten in each study arm) patients; and
• studies according to the PICO.

After reading the full text five studies were included in the literature summary of this module.

Results

Five studies were included in the analysis of the literature, of which two systematic reviews and three randomized controlled trials. Important study characteristics and results are summarized in the evidence tables. The assessment of the risk of bias is summarized in the risk of bias tables.