Remimazolam for intravenous anaesthesia in children: reassessing the dose‐onset relationship

Although the new trial from Fang et al. suggests there are haemodynamic advantages associated with the use of remimazolam in children [1], the conclusion that it has a slower onset than propofol warrants reconsideration. We argue the delay observed (45 s vs. 25 s) more likely reflects subtherapeutic dosing, using half the established paediatric ED95, rather than a pharmacological limitation of the drug itself. Fang et al. administered 0.3 mg.kg-1 of remimazolam, which is less than half of the required dose suggested by recent studies. Three independent investigations have converged on remarkably similar ED95 values. Cai et al. found 0.57 mg.kg-1 was required for loss of consciousness in young children [2]. Li et al. likewise reported 0.55 mg.kg-1 [3]. Chen et al. determined 0.56 mg.kg-1 was necessary for tracheal intubation, despite the concurrent administration of both remifentanil and esketamine [4]. While these studies researched different endpoints, the reported dosing requirements are consistent (Table 1). None reported haemodynamic instability at these higher doses. Time-to-event modelling shows that a 6 mg.kg-1.h-1 incremental increase in remimazolam dosing speeds onset by 15–20 s (hazard ratio 1.80, 95%CI 1.57–2.07) [5]. This reflects basic receptor pharmacology, as subtherapeutic doses fail to achieve the high receptor occupancy required for rapid and reliable hypnosis. For a typical 20 kg child, Fang et al. deliver just 6 mg when the evidence suggests 11 mg would be optimal. This 5 mg deficit explains both the slow onset and why 4.3% of patients required rescue medication. The selection of 0.3 mg.kg-1 likely reflects an extrapolation from adult dosing, a practice that compromises the validity of onset comparisons in this study. Whilst a propofol dose of 2.5 mg.kg-1 is appropriate for children, the remimazolam dose fell significantly below the established therapeutic window. The rigorous methods and safety findings reported by Fang et al. should be applauded. However, misattributing dose-related delays to the drug could lead clinicians to abandon a potentially valuable option. Underdosing creates its problems: prolonged induction times; more rescue medications; and higher costs. Whilst the cited studies reported no haemodynamic instability at doses of 0.5–0.6 mg.kg-1, these were relatively small studies. Larger safety trials specifically examining haemodynamic effects at these higher doses would provide additional reassurance before widespread adoption. We recommend starting with 0.5 mg.kg-1 for pre-school-aged children, with titration between 0.45 and 0.60 mg.kg-1 as needed. Individual patient factors, age-specific pharmacokinetics, comorbidities and concurrent medications should guide dose selection within this range. Younger children may require doses toward the higher end of this spectrum, while those with compromised cardiovascular function might benefit from more conservative dosing. The delayed onset observed by Fang et al. appears to be a correctable dosing issue rather than an inherent limitation of remimazolam. While their safety data are reassuring, interpreting efficacy outcomes in the context of subtherapeutic dosing may inadvertently discourage adoption of a valuable peri-operative option. Future studies using paediatric-specific doses (0.5–0.6 mg.kg-1) should establish definitively whether remimazolam can achieve competitive onset times while maintaining its haemodynamic advantages.