Balancing the risks and benefits of anesthetics in status epilepticus

PURPOSE
According to international guidelines, status epilepticus refractory to first- and second-line antiseizure medication should be treated with anesthetics. Therefore, continuously delivered intravenous midazolam, propofol, or barbiturates are recommended as third-line therapy. While electroencephalographically (EEG)-controlled titration of anesthetics to seizure termination or to the emergence of an EEG burst-suppression pattern makes sense, evidence of the efficacy and tolerability of such third-line treatment is limited and concerns regarding the risks of anesthesia remain. The lack of treatment alternatives and persistent international discord reflecting contradictory results from some studies leave clinicians on their own when deciding to escalate treatment. In this conference-accompanying narrative review, we highlight the challenges of EEG-monitored third-line treatment and discuss recent studies that examined earlier administration of anesthetics.


RESULTS
Based on the literature, maintaining continuous burst suppression is difficult despite the constant administration of anesthetics, and the evidence for burst suppression as an adequate surrogate target is limited by methodological shortcomings as acknowledged by international guidelines. In our Swiss cohort including 102 patients with refractory status epilepticus, burst suppression as defined by the American Clinical Neurophysiology Society's Critical Care EEG Terminology 2021 was established in only 21%. Besides case reports suggesting that rapid but short-termed anesthesia can be sufficient to permanently stop seizures, a study including 205 patients revealed that anesthesia as second-line treatment was associated with a shorter median duration of status epilepticus (0.5 versus 12.5 days, p < 0.001), median ICU (2 versus 5.5 days, p < 0.001) and hospital stay (8 versus 17 days, p < 0.001) with equal rates of complications when compared to anesthesia as third-line treatment.


CONCLUSIONS
Recent investigations have led to important findings and new insights regarding the use of anesthetics in refractory status epilepticus. However, numerous methodological limitations and remaining questions need to be considered when it comes to the translation into clinical practice, and, in consequence, call for prospective randomized studies. This paper was presented at the 8th London-Innsbruck Colloquium on Status Epilepticus and Acute Seizures held in September 2022.


Introduction
Status epilepticus (SE) is a life-threatening neurologic emergency with ongoing epileptic seizures [1] that comes along with high morbidity and mortality [2][3][4]. When emergency medical treatment with the administration of first-and second-line anti-seizure drugs fails to terminate seizures, SE is defined as treatment-refractory (RSE). A patient with RSE faces a greatly increased risk of death which is mirrored by an in-hospital case fatality rate of up to 40% [4]. This calls for an urgent transfer of the patient to an intensive care unit (ICU) to closely monitor and stop seizures by the induction of deep coma and to prevent and treat disease-related complications.
International guidelines recommend the continuous intravenous administration of anesthetic drugs, including midazolam, propofol, or barbiturates for 24 to 48 hours as a third-line treat-  [5,6]. To achieve such deep coma, the dosage of anesthetics must be steadily increased during electroencephalographic (EEG) monitoring, until surrogates for seizure termination are noted, such as either persistent cessation of electrographic seizures, or the emergence of a burstsuppression pattern or an isoelectric EEG. However, the continuous and high-dose administration of anesthetics may be accompanied by potentially severe and harmful side effects. Several studies have reported unfavorable outcomes of patients with RSE independently associated with continuously administered anesthetics [7][8][9][10][11]. These reports call for heightened awareness of appropriate patient selection due to potential adverse events that may be linked to high doses and prolonged duration of anesthetic treatment [12,13]. The limitations of such studies, including the retrospective design, along with the limited body of evidence for the use of anesthetics to treat refractory SE, and the fact that recommendations for anesthetics are primarily based on expert opinion, fuel an ongoing debate when it comes to the optimal usage of third-line treatment. Therefore, these decisions are often left to the physician's judgment when the question of further therapy escalation arises after the failure of first-and second-line treatment.
In this conference-accompanying narrative review, we highlight the challenges of EEG-monitored third-line treatment of RSE and discuss recent findings regarding the effect of earlier administration of anesthetic treatment. We further examine whether certain subgroups of SE patients might benefit more or less from the administration of anesthetics as second-line treatment.

Challenges of third-line treatment of refractory SE
Reports and studies have uncovered and discussed many potential adverse events of anesthetics, including respiratory tract infections that are associated with prolonged mechanical ventilation [8,14], propofol infusion syndrome [15][16][17], a rare syndrome leading to cardiac failure, rhabdomyolysis, metabolic acidosis, and kidney failure, as well as severe arterial hypotension, cardiotoxicity, and gastrointestinal paralysis with barbiturates [18,19]. These complications come along with the clinical challenges when it comes to the adequate and rapid titration of anesthetics to establish a sufficient treatment response in RSE. At first, glance, following and strictly adhering to the treatment guidelines [5,6] seems to be straightforward and not very challenging. A multi-national survey identified EEG burst suppression as the preferred titration target for anesthetic treatment for RSE [20] with the experts' opinion that an interburst interval of 10 seconds should be established [21]. Further, burst suppression seems to be easily recognizable even for the untrained EEG reader after a short training session [22]. However, at second glance, the presumed effortlessness of these treatment procedures turns out to be rather deceptive. Maintaining a continuous sufficient burst-suppression pattern on EEG is difficult despite constant intravenous administration of anesthetics [23], and the evidence for burst-suppression as an adequate surrogate target is limited by methodological shortcomings as acknowledged by international guidelines themselves [5,6]. Furthermore, the current American Clinical Neurophysiology Society's (ACNS) Standardized Critical Care EEG Terminology definition of 2021 of a burst suppression requires a suppression or attenuation proportion of 50% of the EEG recording which seems to be a rather arbitrary cut-off [24]. Preliminary results of our upcoming observational study including a semiquantitative EEG analysis of 102 adult patients treated with anesthetics in RSE, showed that such ''complete" burst-suppression with the recommended 50% suppression or attenuation proportion was established in 21% of patients within a median of 51 hours (IQR 16.1-103.7). In another 14% of the 102 patients, burst suppression was achieved earlier (within a median of 23 hours [IQR 1.0-28.9]) but with an attenuation or suppression proportion of <50% (i.e., ''incomplete" burst suppression). Hence, in 65% of the 102 patients, no burstsuppression (''complete" or ''incomplete") could be achieved and in 79% of the 102 patients ''complete" burst-suppression could not be established. These results are in line with previous results that underscored the difficulties to induce and maintain a burstsuppression pattern: a retrospective study meticulously quantitatively analyzed the proportion of achieved burst-suppression pattern and demonstrated a remarkable inter-patient and intrapatient variability of suppression proportions despite a constant and continuous administration rate of anesthetic with most patients not meeting the titration goal of a predefined burstsuppression suppression rate of 80% despite continuous EEG monitoring [23]. Even in a prospective randomized study, burst suppression as the main intervention could not be established in several patients [25]. If such inconsistencies of burst suppression are clinically relevant in our cohort and whether they are associated with specific patient characteristics and outcomes is currently under investigation. Prior studies, however, revealed contradictory results regarding the associations with different titration goals, such as with EEG-detected seizure termination, burstsuppression, or isoelectric EEG, and outcomes [18,[26][27][28][29][30][31]. Unfortunately, despite the results of these previous studies, many questions remain and should be addressed in future studies: Should a burst-suppression pattern be preferred instead of an isoelectric EEG? Can the titration phase of anesthetics and the stability of burst suppression be optimized by quantitative EEG analysis and an increased number of EEG-trained nurses or physicians in the ICU? Is the quality of burst suppression dependent on the type of anesthetic drug? Should we keep the titration phase as short as possible when administering anesthetics and does this influence treatment success?

Early administration of anesthetics in RSE and outcomes
Although treatment guidelines for SE recommend starting continuously administered anesthetics as a third-line treatment, little is known regarding the importance and effect of the exact time of the initiation of anesthesia in SE patients. Previous case reports suggested that in selected patients with RSE and a reversible cause, a rapid but short-termed anesthetic induction is sufficient to permanently stop seizures [32]. A recent study including 205 adult SE patients from two Swiss academic medical care centers, aimed to explore the safety and efficacy of anesthetics when administered immediately after first-line antiseizure treatment [33]. Seventythree percent were treated according to the guidelines and 27% were treated with artificial coma immediately after the failure of first-line treatment. While the primary outcome (i.e., in-hospital death) did not differ between the two groups, coma induction after first-line treatment was associated with a shorter median SE duration (0.5 days with anesthetics as second-line versus 12.5 days with anesthetics as third-line, p < 0.001) and a shorter median ICU and hospital stay (ICU stay: 2 days with anesthetics as second-line versus 5.5 days with anesthetics as third-line, p < 0.001; hospital stay: 8 days with anesthetics as second-line versus 17 days with anesthetics as third-line, p < 0.001), without increasing the rate of complications. The result that nonadherence to treatment guidelines when it comes to the administration of antiseizure drugs did not have a negative impact on outcomes in terms of increased death and complications or reduced rate of return to premorbid neurologic function after SE is in line with prior studies [34]. Importantly, the duration of ICU stay and especially mechanical ventilation and extubation delay are strongly associated with medical in-hospital complications, including delirium and the duration of administered anesthetics has been correlated with the emergence of progressive brain atrophy [35][36][37][38][39]. Hence, it seems more than plausible that shorter ICU and hospital stays by direct and early coma induction may reduce complication rates and costs related to critical care as well as improve the availability of ICU beds. However, subgroup analyses revealed that improvement of outcome with such early coma induction after first-line treatment was restricted to patients without presumed fatal etiologies of SE. This more aggressive approach is in accordance with the idea that patients who continue to have clinical or EEG evidence of seizures after first-line treatment with benzodiazepines should already be considered to have RSE and should in consequence be treated with anesthetics [40]. Although the results with such an unconventional approach are promising, the limitations (mainly resulting from the semiquantitative EEG analyses and the retrospective nature of the study) call for further studies to confirm or disprove these findings and the mechanistic hypothesis that earlier treatment escalation with third-line medication may shorten RSE duration, ICU and hospital stay, as well as reduced complication rates in SE patients without any underlying fatal etiology (Fig. 1). However, as the current studies cannot exclude that some patients treated with early anesthesia (i.e., as second-line treatment) would have responded to non-anesthetic second-line antiseizure drugs and not have required anesthesia, randomized controlled trials are needed in this context.

Conclusions
While the first studies raised serious concerns about the continuous administration of anesthetics as third-line treatment in RSE, recent investigations have led to important findings and indications related to the management and use of anesthetics in refractory SE. However, numerous methodological limitations and remaining questions need to be considered when it comes to trans-lation into clinical practice, and, in consequence, urgently call for prospective randomized studies. Meanwhile, clinicians treating patients in RSE must continue to carefully balance the risks and benefits that treatment escalation may bring to the individual patient.

Submission declaration and verification
The authors confirm that the work has not been published previously, that it is not under consideration for publication elsewhere, that its publication is approved by all authors, and tacitly or explicitly by the responsible authorities where the work was carried out, and that, if accepted, it will not be published elsewhere in the same form, in English or any other language, including electronically without the written consent of the copyright-holder.

Declaration of Competing Interest
The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: Raoul Sutter received research grants from the Swiss National Foundation (No 320030_169379), the Research Fund of the University Basel, the Scientific Society Basel, and the Gottfried Julia Bangerter-Rhyner Foundation. He received personal grants from UCB-pharma and holds stocks from Novartis, Roche, Alcon, and Johnson & Johnson. Anja Jünger reports no disclosures. Sira M. Baumann reports no disclosures. Pascale Grzonka reports no disclosures. Pia De Stefano reports no disclosures. Urs Fisch reports no disclosures.