The use of burst suppression has long been considered a therapeutic goal in the management of refractory and super-refractory status epilepticus. Some landmark studies [1] and its association in the guidelines have logically favored this situation [2]. However, there are a few discordant voices expressing surprise at the aggressiveness of these treatments [3], and we are currently seeing a trend toward reevaluating this strategy [4].

First, let’s consider the context in which burst suppression is used. Refractory status epilepticus is a frequent neurological emergency associated with excessively high mortality and morbidity rates of the order of 35% and 50%, respectively. Few randomized trials have focused on this patient population [5, 6]. The question of using burst suppression in refractory status epilepticus nevertheless raises the issue of the immediate ideal therapeutic goal of management and the outcome we wish to achieve for these patients. The most common objective is to probably terminate the ictal activity as quickly as possible, both electrically and clinically, and return the patient to his or her previous state, when the cause of the status epilepticus allows. At the very least, we want to do as little harm as possible. And that is the whole point of burst suppression: are we being excessive in achieving this therapeutic goal? Indeed, in other circumstances in which this intensity of sedation use is not required, the application of burst suppression is associated with greater morbidity, which may be characterized by longer duration of mechanical ventilation, more health-care-associated pneumonia, more delirium, and longer post-intensive-care and hospital length of stay, ultimately resulting in increased mortality [7]. The question therefore deserves careful consideration.

The authors are to be commended for providing such a historical review of the burst suppression paradigm in the treatment of status epilepticus [8]. The elements herein provided are extremely enlightening for the reader and give the youngest among us the necessary hindsight to understand the rationale behind the use of this strategy. This “point of view” is all the more important given that, as the authors rightly point out, the pathophysiological rationale is essentially theoretical and the associated level of evidence is extremely low.

Indeed, we are currently unable to answer the question of the benefit of applying burst suppression in the treatment of status epilepticus. As reported by Hawkes et al. [8], studies addressing this issue are essentially retrospective and therefore are associated with numerous biases. The first is selection bias. Which patients received burst suppression and why? The application of burst suppression requires an important technological environment, combining the use of anesthetic agents with anticonvulsant properties, mechanical ventilation, and often catecholaminergic support. This can be achieved in neurocritical care and/or critical care units, which are sometimes unevenly distributed across the globe. The second is measurement bias. Indeed, to discuss burst suppression as a therapeutic goal, we need to have the resources to evaluate it. Have the patients featured in these publications not been selected? How many of them benefited from continuous EEG monitoring to validate whether this objective has been achieved and maintained over the required time? We could also mention the very high probability of confounding bias. What standardization did these patients receive? Do all anesthetic agents have equivalent pharmacokinetic and pharmacodynamic properties? The answer is obviously no. The rise of therapies such as midazolam, propofol, and, more recently, ketamine and the fact that the use of thiopental has fallen into disuse illustrate this point.

The population in which these anesthetic agents are used is also important. Were the patients known to have epilepsy, or did they present an inaugural form with a high potential for progression to super-refractory status epilepticus? The patients certainly had different volumes of distribution, which were not taken into account. Lastly, the modalities of application are important: the duration of burst suppression is certainly inhomogeneous and deserves to be precisely quantified. The recent evolution of ACNS recommendations to better define this pattern confirms this assertion [9].

This question should also be seen in the context of the availability of continuous EEG monitoring. Indeed, what is the proposed alternative? The guidelines clearly state that the aim is to achieve electrical control of electrical seizure activity (seizure suppression or burst suppression) [2]. Such an objective requires the use of continuous EEG monitoring, and teams trained in its use or at least close collaboration with neurophysiologists teams. The widespread use of burst suppression is certainly due to historical consideration but also to greater ease of use by health care teams. Significant parallel efforts are required to create the conditions necessary for wider use of this monitoring tool [10].

In conclusion, the data provided by Hawkes et al. in this Viewpoint gives us extremely important information from a historical perspective on the use of burst suppression in the treatment of status epilepticus. The available data on this concern do not come from randomized controlled populations, nor do we have any prospective epidemiological studies combining standardized treatments to enable us to conduct high-quality meta-analyses. We cannot correct for the biases inherent in this weakness of information, and the data we interpret therefore do not allow us to know whether burst suppression is a therapeutic goal for the benefit of patients with refractory or super-refractory status epilepticus. As is often the case, the most appropriate response could be the most moderate, and only a personalized approach would be ideal. However, this remains to be demonstrated by the most appropriate tools at our disposal: those of evidence-based medicine.