Propofol is a sedative-hypnotic medication used for sedation in the ICU; it has various sedative, anxiolytic, and anticonvulsant properties, and its use has been shown to help reduce intracranial pressure [13,14]. Propofol provides a rapid onset of action within seconds after administration and a short duration of action up to 15 minutes. However, propofol has several risks. First, its formulation must be prepared in a lipid solution, which increases the risk of bacterial contamination and could be associated with fatal infectious disease [15,16]. Second, patients have been reported to experience pain on IV injection. Third, propofol is associated with a higher risk of cardiorespiratory depression compared to other IV sedatives like dexmedetomidine, ketamine, and midazolam. Finally, propofol has been associated with fatal "propofol infusion syndrome (PRIS)" [17]. PRIS is a rare syndrome which affects patients undergoing long-term treatment with high doses of propofol and causes cardiac and renal failure, rhabdomyolysis, metabolic acidosis. The safe dose of propofol infusion is considered to be 1–4 mg/kg/hr for sedation in intensive care. However, fatal cases of PRIS have been reported after infusion doses as low as 1.9–2.6 mg/kg/hr. Propofol can be relatively safe and an ideal anesthetics for several minute up to several hours of surgery, but physicians must keep in mind that propofol infusion can cause PRIS in cases of prolonged continuous infusion in the ICU [18].

Midazolam is an amnesic and sedative BDZ, and had been shown to have an onset of 3–5 minutes and a recovery period of 2 hours before remimazolam approval. It is considered safe to use, and the risk of cardiorespiratory depression caused by the use of midazolam is much lower than that of propofol. However, midazolam accumulation can occur via repeated injection or prolonged continuous infusion, leading to unpredictably prolonged sedation associated with delirium [19]. Critically ill patients also sometime present with an altered mental state caused by a neurological event such as intracranial hemorrhage, infarction, or seizure. Intensivists could face challenging situations distinguishing this "real neurological emergency" from the after-effects of prolonged sedation with classic BDZs.

Critically ill patients also commonly have chronic renal failure and experience acute kidney injury. Active metabolites of midazolam can accumulate in patients with renal failure and lead to longer-than-expected sedation of these patients, as conjugated metabolites of midazolam have significant pharmacological activity [20]. Prolonged sedation can occur due to active metabolites of midazolam and an impaired metabolism on the liver enzyme cytochrome P450 3A4 [21].

Dexmedetomidine, a high-affinity adrenergic agonist of the alpha2 receptor, provides light sedation and pain relief in patients in the ICU. Dexmedetomidine is associated with more days free of mechanical ventilation, a shorter time in a coma-like state, and less risk of delirium [22-24]. Additionally, dexmedetomidine has been shown to reduce the incidence of delirium, prevent delirium, and improve mortality [25-28]. However, hypotension and bradycardia are common side effects of dexmedetomidine [29-31]. Dexmedetomidine can be used a sole anesthetic for deep sedation or general anesthesia with a higher dose (5–10 times the maximum recommended dose for procedural sedation). However, deep sedation with a large dose of dexmedetomidine is rarely applied generally due to its associated risks of hypotension and bradycardia, particularly in critically ill patients [32].

Inhalational anesthetic agents are often used for general anesthesia, as they are potent sedatives which show a fast elimination, limited hepatic metabolism, and no accumulation [33]. Volatile sedation in the ICU has been tried more frequently after solving technical problems since the development of inhalational anesthetic devices, such as AnaConDa (SEDANA Medical, Uppsala, Sweden) and Mirus (Pall Medical, Dreieich, Germany) [33,34]. In a systematic review and meta-analysis of randomized controlled trials showed that ICU sedation with volatile anesthetic agents relative to classic IV sedatives, like propofol or midazolam, reduced the awakening time from sedation by 80 minutes and the extubation time by 196 minutes. Despite such benefits, no reductions in the length of stay in the ICU or hospital were reported [34].

However, the use of inhalational sedation in the ICU remains limited. The reasons for this limited usage may be associated with the unfamiliarity of medical staff to inhalational agents and their methods of administration; patients’ higher risk of agitation, nausea, and vomiting after awakening; potential atmospheric contamination; and a rare, but fatal, complication known as malignant hyperthermia. Malignant hyperthermia is a life-threatening reaction to potent inhalation agents (such as halothane, isoflurane, sevoflurane, and desflurane), and the depolarizing muscle relaxant succinylcholine. Malignant hyperthermia show a hypermetabolic crisis such as extremely high body temperature, rigid muscles or muscle spasms, hyperkalemia, high oxygen consumption, high CO₂ production, multiple vital organ failure, and disseminated intravascular coagulation [35].

Remimazolam has a high affinity on GABA-A receptors to bind at the interface between the alpha and gamma subunits, inducing a highly inhibitory central nervous system. It binds to receptors to make the intracellular concentration of chloride ions increase; this is followed by cellular membrane hyperpolarization and inhibitory conduction of the neuron action potentials to enhance the effects of GABA [36].

Remimazolam is characterized by pharmacokinetics (PK) profiles, specifically a high clearance, a small steady-state volume of distribution, a short elimination half-life, a short context-sensitive half-life, and first-order linear PK. Time after administration to onset of remimazolam is 1–3 minutes; this is faster than that of midazolam [37], and the half time of remimazolam is 7–8 minutes, which is much less than that of midazolam [36].

In a phase 1 PK study with healthy volunteers who were given a single dose of remimazolam, the mean residence time for remimazolam was 0.50 hours and the mean residence time for midazolam was 3.56 hours, as the systemic clearance of midazolam is about one-third that of remimazolam and the volume of distribution is more than twice that of midazolam [38,39]. This difference in PK between remimazolam and midazolam could explain one factor in patients’ rapid recovery after receiving remimazolam. There is no clear relationship between body weight and systemic clearance of remimazolam within the studied body weight range (60–100 kg) [39]. There may be no significant benefit for dosing by body weight compared to fixed doses [39], The PK profiles were similar to those in a single-dose phase 1 PK study with continuous infusion of remimazolam and midazolam [40].

Carboxylesterase enzymes can be found in the cytosol and the rough endoplasmic reticulum of tissues [41], and remimazolam is metabolized by this tissue esterase (particularly, the liver) to an inactive carboxy acid metabolite, CNS 7054 [5,12,13]. CNS 7054 has a PK profile with a smaller volume of distribution, slower clearance rate, and a longer mean residence time in comparison to remimazolam [42], and has a 400-fold lower affinity for the GABA-A receptor [43]. An additional study found that there was no significant difference between older adults (median age, 66.0 years) versus younger adults (median age, 21.0 years) in terms of PK profile of remimazolam [44].

One study involved the use of a simulated plasma PK after a 10 mg remimazolam bolus, and revealed no significant different in Cmax values among hepatic impairment patients groups [45]. There was no difference between liver dysfunction and healthy subjects in the incidence and duration of loss of consciousness. However, recovery from sedation was delayed by hepatic impairment. This demonstrated that carboxylesterase enzymes–1A in liver must have an important role in metabolism of remimazolam. As recovery from sedation in severely hepatic impaired patients was delayed, specifically reduced dosing for these patients can be considered. In contrast, no dose adjustment is required for patients with mild or moderate hepatic failure [45]. Remimazolam is not metabolized by cytochrome P-450 isozymes, nor does it inhibit cytochrome P-450 metabolism [46].

In one study, 80% of the dose of remimazolam was excreted in urine as an inactive metabolite, and less than 1% of the dose was detected as unchanged 24 hours after remimazolam injection. Remimazolam indicated no accumulation in patients with renal impairment, and was metabolized at the same rate as that of healthy volunteers. There was no need to adjust the dosing of remimazolam in patients with impaired renal function, as renal function does not affect the PK of remimazolam [45].

To determine a level of sedation in study of anesthetics, researchers use commonly the Modified Observer’s Assessment of Alertness/Sedation Scale (MOAA/S) and Bispectral Index (BIS) of electroencephalogram. MOAA/S is scored 0–5, where 5 represents an alert subject who responds promptly to their name spoken in a normal tone, and 0 represents a patient with no response after a painful trapezius squeeze. BIS index which ranges from 0 to 100, where 0 represents the absence of brain activity, and 100 represents the fully awake state. Generally, BIS values between 40 to 60 meant adequate general anesthesia for a surgery [40].

In the phase 1 pharmacodynamic study with healthy volunteers who were given a single dose of remimazolam, the dose was escalated throughout the cohorts until cohort 9 (0.30 mg/kg) was reached and six of 10 subjects (60%) in this cohort experienced loss of consciousness (MOAA/S scores of <2) for a minimum of five minutes, which was the predefined stopping criterion for dose escalation. In contrast, subjects in the placebo group did not experience sedation [39].

In this study, the onset of sedation was fast for both midazolam and remimazolam (at doses of >0.05 mg/kg). The degree and duration of sedation with remimazolam showed dose dependency, and the peak effect of sedation was observed in 1–4 minutes following the start of the infusion. Administration of a 0.10–0.20 mg/kg dose of remimazolam resulted in deeper sedation and faster recovery from sedation in comparison with administration of a 0.075 mg/kg dose of midazolam [39]. Subjects showed no sedation, or very minimal sedation, at 0.01 and 0.025 mg/kg, and showed small reductions in MOAA/S scores (to 4) and BIS scores (to 75) at 0.05 mg/kg. Doses of 0.075 mg/kg and higher resulted in deeper sedation, as evidenced by MOAA/S scores of <2 and mean BIS scores of 60 [39].

Remimazolam had a sufficiently safe and sedative effect to use for procedures like endoscopy. Induction doses of 5 mg were administered intravenously over a period of 1 minute and 2.5 mg supplemental doses for 15 seconds can be administered within a least 2-minute interval [51]. Compared with the other sedatives, remimazolam demonstrated a higher sedative efficiency than that of midazolam and lower than that of propofol [52-56]. Remimazolam showed a lower incidence of hypotension than midazolam, and no significant differences were observed in hypoxia, bradycardia, nausea, vomiting, and injection site pain. Compared with propofol, remimazolam showed a significantly lower incidence of hypotension, hypoxemia, and pain at injection site; however, there were no differences in the incidence of bradycardia, nausea, and vomiting [57,58].

Remimazolam was used as a general anesthetic, using induction doses of 6 and 12 mg/kg/hr and maintenance rates of 1 mg/kg/hr. Remimazolam was superior to propofol in terms of its efficacy for general anesthesia, and it showed a significantly lower incidence of hypotension and other adverse events [37,59,60].

Phase II trial of Ono pharmaceutical company about 49 ICU patients sedated with remimazolam, all patients were reported to be sedated successfully and have no significant adverse events, and seven out of 49 ICU patients were sedated with remimazolam for >24 hours. However, on analyzing samples 24 hours or more after starting the continuous infusion, a higher concentration of remimazolam was observed than expected [61]. The suitability and safety of remimazolam for prolonged sedation of ICU patients’ needs to be tested in the future. A few clinical trials on sedating patients in ICUs with remimazolam are ongoing or completed to recruit patients.; however, the results of the trials have not been published yet (national clinical trial number: NCT04611425, NCT04815265) [62]. One study involving general anesthesia (5 hours) and postoperative sedation (18 hours) using remimazolam showed that postoperative sedation with a continuous infusion of 0.25 mg/kg/hr remimazolam with infusion of remifentanil provided optimal sedation of ICU patients [47].