FormalPara Key Points
Table 1
Full size table

1 Introduction

Status epilepticus (SE) is a life-threatening condition and medical emergency associated with long-term consequences, including “neuronal death, neuronal injury, and alteration of neuronal networks, depending on the type and duration of seizures” [1], and a risk of mortality around 20% [2] increasing up to 33% in patients with impaired consciousness [3]. It is currently defined 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, prolonged seizures” [1].

The pharmacological management of SE follows a stepwise approach. Fast-acting benzodiazepines are administered as first-line treatments, leading to SE cessation in about 60–70% of cases [4,5,6]. In benzodiazepine-resistant cases, an intravenous (IV) administration of antiepileptic drugs (AEDs) is required to control SE and prevent or minimize the risk of negative long-term systemic or neuronal consequences, particularly in the case of convulsive SE [4, 5]. The AEDs commonly used as second-line treatments for SE are phenytoin, phenobarbital, valproate, levetiracetam (LEV), and lacosamide [4, 5, 7]. If generalized tonic–clonic (convulsive) SE persists despite administration of IV AEDs, anesthetic therapy with all its complications is recommended.

Brivaracetam (BRV) is a high-affinity synaptic vesicle glycoprotein 2A ligand that is currently licensed as a treatment for focal-onset seizures in patients aged ≥ 4 years with epilepsy as monotherapy or adjunctive therapy [8]. After oral administration, BRV is rapidly and completely absorbed; it has low (< 20%) plasma protein binding and a linear and predictable pharmacokinetic profile [9, 10]; furthermore, it carries minimal risks of drug–drug interactions [11].

Although structurally related to LEV, BRV has higher brain permeability, faster brain synaptic vesicle glycoprotein 2A occupancy, and more rapid onset of action [12, 13]. These properties make IV BRV potentially an ideal compound in the emergency setting, particularly in the treatment of SE. Although BRV is currently not labeled to treat SE, preclinical studies have been encouraging and showed efficacy in animal models of SE [12, 14]. The aim of our study was to systematically review the evidence about the clinical efficacy and tolerability of IV BRV in the treatment of SE.

2 Methods

The results of the present systematic review were reported according to the recommendations of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement [15]. The review protocol was not previously registered.

We included any study evaluating IV BRV as treatment for SE of any type in patients of any age, irrespective of definition of SE adopted and neurological outcomes assessed.

The following electronic databases and data sources were systematically searched using the following search strategy: Brivaracetam AND “status epilepticus”:

  1. 1.

    MEDLINE (January 1966–3 December, 2018), accessed through PubMed;

  2. 2.

    EMBASE;

  3. 3.

    Google Scholar;

  4. 4.

    ClinicalTrials.gov (available at: https://clinicaltrials.gov/);

  5. 5.

    Opengrey.eu (available at: www.opengrey.eu).

To minimize publication bias, we also searched the conference proceedings of international congresses by the International League Against Epilepsy and the American Epilepsy Society from 2016 onwards. All searches were conducted on 3 December, 2018. All resulting titles and abstracts were evaluated, and any relevant article was considered. No language restrictions were adopted. Retrieved articles were independently assessed for inclusion by two review authors (FB, RN); any disagreement was resolved through discussion.

The following data were independently extracted by two review authors (FB, RN) for any included study: main study author and date of publication; type of study; total number, age, and sex of participants; type and etiology of SE; previous or concomitant drugs, including order of administration and maximal dosage; dosage of BRV (initial dose, titration interval, maximal dose); use of BRV as last medication; number of drugs administered prior to BRV to treat SE; time from SE onset to BRV administration; number of patients achieving SE cessation when BRV was administered as the last drug; time from BRV administration to SE cessation (only for responders); neurological outcomes; occurrence and/or type of adverse effects. Limitations of included studies were discussed narratively. We did not plan the quantitative synthesis of data as we expected to find great clinical and methodological heterogeneity between studies.

3 Results

A total of 326 records was identified (14 MEDLINE; 83 EMBASE; 227 Google Scholar; 0 ClinicalTrials.gov; 0 Opengrey.eu; 1 abstract proceedings of international congresses of the International League Against Epilepsy; 1 abstract proceedings of international congresses of the American Epilepsy Society). After removal of duplicates and reading title and abstracts, seven studies were eventually included [16,17,18,19,20,21,22] (Fig. 1).

Fig. 1
figure 1

Study inclusion flow diagram

Full size image

Overall, 37 patients (21 were female) with ages ranging from 22 to 85 years were included. Characteristics of the included studies and participants are summarized in Table 1. There was a great heterogeneity in the type and etiology of SE across studies. Details on BRV administration and efficacy/tolerability outcomes are reported in Table 2. The number of drugs used prior to BRV to treat SE ranged from one to eight. The time from SE onset to BRV administration ranged from 0.5 h to 105 days. The initial BRV dose ranged from 50 to 400 mg. The proportion of patients achieving clinical SE cessation when BRV was administered as the last drug varied from 27 to 50%. In case reports, all patients achieved SE cessation. The time from BRV administration to SE cessation ranged from 15 min to 94 h. No serious adverse effects were reported.

Table 1 Study design, clinical characteristics of patients, and the type and etiology of status epilepticus (SE)
Full size table
Table 2 Details of brivaracetam (BRV) treatment in included studies
Full size table

For six responders in whom BRV was used as the last medication, individual patient data were available; in these patients (age: 61 ± 25 years; maximal median BRV dose: 200 mg, range 100–400 mg), the median time from BRV administration to SE cessation was 15 h (range 15 min–27 h) [18, 21]. In a case series not reporting individual data, the median time from BRV given as the last AED (numbers of prior AEDs: 1–6) to SE cessation was 22 h (range 5–96 h) [22]. In this study, responders (seven patients) received a significantly greater median loading dosage per body weight compared with non-responders (3.3 mg/kg vs. 1.5 mg/kg; p = 0.02); all responders had loading doses above 1.9 mg/kg [22].

4 Discussion

In this systematic review, we identified and critically appraised the currently available evidence on the use of IV BRV for the treatment of SE in a real-world clinical setting. The efficacy in case series varied considerably with cessation rates ranging from 27 to 50%. No adverse events have been reported, suggesting that BRV can be safely administered to patients presenting with SE of different types and etiologies. The wide range of efficacy observed across the studies can find different explanations. The response rate of 100% may reflect publication bias, i.e., the higher likelihood of studies with positive results to be submitted and eventually published compared to studies with negative results [23]. To take into account and minimize this risk, we extended the searches to the gray literature looking for abstracts published in conference proceedings, but not yet published as full-length reports. At the same time, however, study results reported only as abstracts are inevitably less complete or preliminary compared with those published in final full-length articles. This could hamper the opportunity to establish a straight causal correlation between the intervention and clinical response. More specifically, it is not always possible to ascertain whether changes in co-medications were made, and details on the use of simultaneous or sequential drugs were not always reported.

The small number of patients may have increased the imprecision of results, and the great clinical heterogeneity of the included patients and the different dosages administered may account for the wide clinical response following BRV administration. The lack of information on the type and etiology of SE, both major determinants of treatment response [3, 24], as well as comorbidities prevented us from analyzing the contribution of these factors to the overall efficacy of BRV. Furthermore, the time from SE onset to BRV administration was extremely wide, ranging from 0.5 h to 105 days, and in many cases BRV was given in patients with super-refractory SE and after several attempts (up to eight) with other antiepileptic or anesthetic drugs. In these patients, BRV was probably used to treat SE associated with severe underlying brain dysfunction. The type and the etiology of SE were extremely heterogeneous across included studies, further limiting the robustness of any conclusions on the clinical role of BRV in the treatment of this condition.

The IV BRV doses used to replace oral therapy (100 mg BRV intravenously) were unlikely to represent a loading dosage enough to control SE in most patients [22]. In this regard, a recent study assessed the correlation between BRV exposure and clinical response. Interestingly, patients achieving SE cessation after being administered BRV as the last AED received a significantly greater median loading dosage per body weight compared with non-responders, and a minimum loading dose of 2 mg/kg has been proposed to be safe and likely advisable in the treatment of SE in adults [22]. The benefit of a further increase in loading doses was not investigated, and it is, hence, unclear whether IV BRV shows a ceiling effect in terms of response, similar to that demonstrated with the use of LCM at loading doses higher than 9 mg/kg [25].

Owing to high lipophilicity, BRV crosses the blood–brain barrier rapidly and has a tissue distribution similar to that of fast-acting benzodiazepines [26]. Remarkably, BRV has an entry half-time faster than 1 min, with a barely detectable distribution phase, and reaches its maximal brain concentration within 10 min after IV administration [12]. In animal models, BRV has been also shown to have a faster onset of anticonvulsant activity than LEV [12] and a supra-addictive efficacy with diazepam to control SE [14, 27].

However, the limited available data prevent us from drawing definitive conclusions about the onset of antiepileptic activity of BRV and comparative response with other AEDs. In the six patients in whom BRV was used as the last medication and for whom individual data were reported, the time from BRV administration to SE cessation ranged from 15 to 27 h. Although these figures might confirm the rapid onset of antiepileptic action for BRV, in a case series not reporting individual details, the median time from BRV given as the last AED to SE cessation (seven patients) was longer (22 h; range 5–96 h) [22].

The very low amount of clinical data on BRV in SE prevents us making a comparison with LEV. Although LEV is being increasingly used as a second-line treatment for SE [28], it has also been tested as first-line treatment for generalized convulsive SE in a prehospital randomized controlled trial aimed at determining the efficacy of adding intravenous LEV (2.5 g) to clonazepam (1 mg) [29]. This study did not demonstrate an additional benefit of adding LEV to clonazepam compared to clonazepam treatment alone in the prehospital control of SE (74% vs. 84%, respectively; percentage difference − 10.3%, 95% confidence interval − 24.0 to 3.4). However, it is “unlikely that the added benefit of LEV was adequately assessed in this trial, as any effect was likely overshadowed by the high success rate of clonazepam” [30]. Furthermore, the duration of observation (SE cessation was assessed at 15 min of drug injection) was probably not long enough to fully detect the antiepileptic activity of LEV.

The current evidence on BRV in SE is not enough to justify its use as a first-line drug, although preclinical data have provided promising results of a faster onset of anticonvulsant activity than LEV. If further studies will provide robust evidence showing a fast onset of antiepileptic activity for BRV, its use as a first-line treatment for SE, as a benzodiazepine substitute, could be considered. However, more data need to be collected, also considering that the large multicenter prospective registry, SENSE (Sustained Effort Network for treatment of Status Epilepticus), demonstrated that treatment failure is higher if AEDs are used as first-line treatment instead of benzodiazepines [31].

5 Conclusions

The information on the use of IV BRV in the treatment of SE is currently scarce and of low quality. It is based only on a few case reports and small case series and, therefore, hampered by several confounding factors and a high risk of biases. The evidence available so far does not support the use of BRV for the treatment of SE, unless more clinical data are prospectively collected in observational studies (large case series or registries). Hence, further controlled prospective studies in clinically homogeneous and larger cohorts are warranted to explore the efficacy and tolerability of IV BRV for the treatment of SE and to investigate whether higher loading doses can be more efficacious. Based on pharmacological properties and promising preclinical data, more clinical information should be also gathered on the use of BRV administered as second-line or even as first-line treatment for SE, as a benzodiazepine substitute, and the possible synergistic interactions with other antiepileptic agents.