Promise of Remimazolam in Pediatric Emergence Delirium

Once the basic pharmacology of a new agent is defined, the next step is usually to determine whether the agent improves outcomes. Remimazolam is entering into this phase. It is reasonable to perform these efficacy trials exploring dose or mode of administration in tightly controlled groups with a placebo control. The trial by Cai et al. reported in this issue of Anesthesiology is this type of trial. 

The scenarios where the new therapy is trialed should be in areas where the outcome is important. Ideally, they should also be where there is no completely effective therapy or a lack of agreement over the most effective therapy. Emergence delirium is an important problem in pediatric anesthesia. Its prevalence varies widely in the literature, largely due to great heterogeneity in how it is defined and measured. It is probably not as high as the 50 to 80% that some report, but any practicing pediatric anesthesiologist would likely acknowledge that it occurs not infrequently, and when it does, it causes significant disruption to the child’s recovery. 

Cai et al. explored the hypothesis that either a continuous infusion or a single-bolus dose of remimazolam might decrease the frequency of emergence delirium after laparoscopic inguinal hernia repair. There is mixed evidence that benzodiazepines reduce emergence delirium, and indeed, the mechanism behind emergence delirium remains poorly understood. Nevertheless, it is a plausible hypothesis that remimazolam would reduce emergence delirium. The study included 120 children ranging from 1 to 6 yr of age randomly assigned to receive a continuous remimazolam infusion during surgery, a single-bolus dose of remimazolam at the completion of the procedure, or a placebo. The findings indicated that both methods of remimazolam administration were associated with a reduced incidence of emergence delirium from 35% with placebo to 5% and 7.7% with continuous infusion and a bolus, respectively. Remimazolam was effective not only in lowering the occurrence of emergence delirium but also in reducing the need for propofol intervention in those cases where emergence delirium did manifest. 

The lack of any evidence for a difference between continuous infusion or bolus of remimazolam, and the added complexity and logistical demand of managing an infusion during surgery, suggest that the bolus method might be more practical without compromising efficacy. Moreover, remimazolam in concentrations exceeding 1 mg/ml can precipitate in the intravenous tubing when administered with lactated or acetated Ringer’s solution at infusion rates less than 300 ml/h. 

Both interventions maintained stable hemodynamics and had minimal adverse reactions. However, remimazolam was associated with a small increase in heart rate, the reasons for which are unclear. The relatively small size of the study by Cai et al. does not allow for a thorough safety assessment. 

There are many studies looking at emergence delirium in children. The quality varies tremendously, with major methodologic issues usually related to how the delirium is measured and in particular the muddying problem that pain looks very much like delirium.  Cai et al. should be commended for the thorough methodologic approach used to assess emergence delirium.  To detect emergence delirium, the researchers employed the Pediatric Anesthesia Emergence Delirium (PAED) scale, a tool that has been rigorously validated. The PAED scale encompasses several diagnostic criteria akin to those found in the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition, including disturbances in attention, awareness, and cognition that may present as disorientation or perceptual disturbances. These criteria have been substantiated for the diagnosis of emergence delirium in pediatric patients.  By applying the PAED scale, Cai et al. were able to mitigate the risk of falsely elevated rates of emergence delirium attributed to behaviors stemming from pain, a potential outcome when relying on unvalidated scales. Such nonspecific scales often incorporate behaviors like crying, hyperactivity, and physical agitation, which, although indicative of pain, are not exclusively symptomatic of emergence delirium. This precision in the identification of emergence delirium bolsters the validity of their findings, allowing for a better differentiation between the manifestations of pain and the behavioral construct of emergence delirium. 

Despite the superiority of the PAED score over other scores when it comes to differentiating emergence delirium from pain, it would still have been ideal to have had a tightly controlled environment where efficacy could have been most accurately determined. In this case, that would mean minimal postoperative pain. Cai et al. selected a procedure where they suggest there is minimal nociceptive stimulation, and they employed an extensive, multimodal analgesic protocol to mitigate pain. It could, however, be argued that the study would be more robust had it been conducted in children having nonpainful procedures like diagnostic procedures.

It is recognized that both emergence delirium and pain can manifest at any point during the initial minutes after emergence from anesthesia. Cai et al. entrusted a single researcher blinded to the group assignments to administer the PAED scale and Face, Legs, Activity, Cry, Consolability (FLACC) Behavioral Pain Scale within the first 30 min after surgery. Cai et al. assessed emergence delirium whenever a child displayed behaviors correlating with a PAED score exceeding 10 points, rather than adhering to rigid evaluation intervals. Doing this, they were more effective in capturing the condition and preventing underreporting of its incidence. Additionally, the employment of a single evaluator mitigates the risk of interobserver variability, which could otherwise introduce undue variability.  Cai et al. also correctly reported emergence delirium as the proportion of patients with emergence delirium and not as the difference between groups in mean PAED score. The PAED scale was designed to define emergence delirium as being present or not and should not be used as a continuous scale. 

Given that Cai et al. have demonstrated that remimazolam can reduce emergence delirium compared to placebo, the next step is to compare it to existing therapies. There is good evidence that propofol total intravenous anesthesia or a bolus of propofol at the completion of surgery is an effective way to reduce emergence delirium. Similarly, α² agonists such as clonidine and dexmedetomidine can reduce emergence delirium. Remimazolam should not be recommended as a therapy until it is shown to be superior or at least noninferior to existing methods of reducing emergence delirium. These comparative trials should be done in nonpainful procedures, employ the PAED scale with a single blinded observer, ensure emergence delirium assessment occurs any time during the patient’s postanesthesia care unit stay, and present the incidence of emergence delirium as a proportion of the patient group. The study by Cai et al. provides the impetus to do these trials.