Dear Editor,

Veno‐venous extracorporeal membrane oxygenation (VV-ECMO) in severe acute respiratory distress syndrome (ARDS) can be lifesaving [1]. In these cases, the complex brain-lung crosstalk might lead to a form of acute brain injury that can pose a challenge on sedation strategies [2]. Furthermore, the pharmacokinetic alterations related to the VV-ECMO therapy can lower plasma drug concentrations, making the dose–response relationship of sedatives unpredictable [3]. Complex sedation scenarios in VV-ECMO ARDS patients are thus not rare. Volatile sedation (VS) is a novel therapy in the critical care setting. Different studies demonstrate that VS induces light to deep sedation and is associated with shorter awakening times. Moreover, it could improve oxygenation and decrease the production of alveolar cytokines even in situations where lung mechanics are deeply affected [4].

This retrospective study (January-2021/March-2023), performed in eleven Spanish ECMO centers, included patients with ARDS and VV-ECMO therapy who received concomitant VS for at least 24 h. Patients under 18 years and those who were pregnant were excluded. As the primary objective, we analyzed whether VS led to decreased intravenous (IV) sedation needs. As secondary objectives, we investigated the repercussions of VS in sedation depth and opioid usage, VS practices, and its impact on oxygenation and mechanical ventilation (MV)/ECMO parameters. Data were evaluated before the start of VS (D0), at 24 h (D1), and at 72 h (D3) after its initiation.

Sixty-six patients were included. The patients' general characteristics, main VV-ECMO indications, and VS features are shown in supplementary Table 1S. On D0, patients received a median of 2 (1;2) IV sedatives to achieve a Richmond Agitation Sedation Scale (RASS) score of − 4 (− 4; − 5) and a bispectral index (BIS) level of 45 (40;60). Table 1 summarizes changes in practice after VS. On D1 and D3, there was a decrease in the number of IV sedatives required per patient [2 (1;2) at D0 vs. 1 (0;2) at D1 p = 0.00; 2 (1;2) at D0 vs. 0 (0;1) at D3 p = 0.00] with no significant changes in sedation depth as shown by the BIS [45 (40;60) at D0 vs. 42.5 (39.5;52) at D1 p = 0.32; 45 (40;60) at D0 vs. 50 (40;65) at D3 p = 0.22] and RASS [− 4 (− 4; − 5) at D0 vs. − 5 (− 4; − 5) at D1 p = 0.09; − 4 (− 4; − 5) at D0 vs. − 4 (− 4; − 5) at D3 p = 0.48] values. Daily doses of propofol, midazolam and opioids, excluding remifentanil, remained stable on D3, but the number of patients needing IV sedatives or neuromuscular blockade (NMB) diminished (supplementary Table 2S and Fig. 1S). A slight increase in tidal volume and PaO2/FiO2 ratio (supplementary Fig. 2S) and a decrease in positive end-expiratory pressure (PEEP) were found.

Table 1 Changes in practice after introducing volatile sedation
Full size table

This multicenter study describes VS usage in ARDS patients undergoing VV-ECMO and ultraprotective MV. This study, like the single-center study by Graselli et al., found that VS is feasible in this population and is associated with less need for IV sedatives without compromising the optimal level of sedation or MV efficacy [5]. The introduction of VS can help decrease the usage of continuous NMB and might lead to better oxygenation, although these findings need further study.