Inhaled isoflurane for sedation of mechanically ventilated children in intensive care (IsoCOMFORT): a multicentre, randomised, active-control, assessor-masked, non-inferiority phase 3 trial

Background

Inhaled sedation for mechanical ventilation in patients who are critically ill is emerging as an alternative sedative strategy; however, data are scarce on its efficacy and safety in children, compared with intravenous sedation. The IsoCOMFORT trial aimed to compare the efficacy of inhaled sedation with isoflurane versus intravenous midazolam in the paediatric setting.

Methods

IsoCOMFORT was a randomised, active-control, assessor-masked, non-inferiority phase 3 trial conducted across 19 paediatric intensive care units in Spain, France, Germany, and the UK. Children aged 3–17 years who were critically ill and required invasive mechanical ventilation and sedation for an expected duration of at least 12 h were randomly assigned (2:1) via an interactive web-response system to inhaled sedation with isoflurane or to intravenous sedation with midazolam. Randomisation was done in permuted blocks (sizes 3 and 6), stratified by age group, reason for intensive care unit admission (planned or unplanned mechanical ventilation), and country, with treatment allocation masked to outcome assessors. At baseline, a target range for sedation depth was prescribed based on the COMFORT Behaviour (COMFORT-B) scale, and sedation dosing was titrated to reach the target range. Sedative treatment was planned for up to 48 h (±6 h). The primary endpoint was the percentage of time that an adequate sedation depth was maintained, in the absence of rescue sedation, within the individually prescribed target range, as monitored every 2 h for an expected minimum of 12 h (up to 48±6 h) with the COMFORT-B scale. The primary endpoint was assessed for non-inferiority (margin –9·36 percentage points) in the full analysis set (all randomly assigned participants who received ≥6 h of the allocated study sedative and ≥3 masked COMFORT-B assessments), according to intention to treat. Safety was assessed in all participants who received study treatment. The trial was registered with ClinicalTrials.gov, NCT04684238, and EudraCT, 2020-000578-31, and is completed.

Findings

Between Jan 14, 2021, and Jan 19, 2023, 96 children were randomly assigned: 63 to the isoflurane group and 33 to the midazolam group. 92 participants were included in the full analysis set (mean age 7·7 years [SD 4·1]; 35 [38%] female and 57 [62%] male). The least-squares mean percentage of time in the COMFORT-B target range was 68·94% (95% CI 52·83–85·05) in the isoflurane group and 62·37% (44·70–80·04) in the midazolam group. The least-squares mean difference between treatments was 6·57 percentage points (95% CI –8·99 to 22·13), indicating non-inferiority, with the lower bound of the 95% CI exceeding the non-inferiority margin of –9·36 percentage points). In the safety set (n=94), serious adverse events occurred in 19 (31%) of 61 participants in the isoflurane group and eight (24%) of 33 participants in the midazolam group, none of which were considered related to study treatment. Treatment-related severe hypotension occurred in one participant per group, and three participants in the isoflurane group discontinued treatment due to adverse events. No treatment-related deaths were reported.

Interpretation

Among critically ill children, the effectiveness of sedation with inhaled isoflurane was non-inferior to that of intravenous midazolam, offering an alternative medication in children receiving mechanical ventilation.

Funding

Sedana Medical, Stockholm, Sweden.