Secondary Logo

Journal Logo

A Placebo- and Midazolam-Controlled Phase I Single Ascending-Dose Study Evaluating the Safety, Pharmacokinetics, and Pharmacodynamics of Remimazolam (CNS 7056): Part I. Safety, Efficacy, and Basic Pharmacokinetics

Antonik, Laurie J. MD*; Goldwater, D. Ronald MD; Kilpatrick, Gavin J. PhD; Tilbrook, Gary S. PhD; Borkett, Keith M. BSc

doi: 10.1213/ANE.0b013e31823f0c28
Anesthetic Pharmacology

BACKGROUND: A new benzodiazepine, remimazolam, metabolized by tissue esterases to an inactive compound, CNS 7054, has been developed to permit a fast onset, a short and more predictable duration of sedative action, and a more rapid recovery profile than with currently available benzodiazepines. We report on the safety and efficacy of the first human study.

METHODS: A phase I, single-center, double-blind, placebo- and active-controlled, randomized, single-dose escalation study was conducted. Up to 10 cohorts of healthy subjects were scheduled to receive a single 1-minute IV infusion of remimazolam, midazolam, or placebo. In the 10 possible cohorts, remimazolam doses were from 0.01 to 0.35 mg/kg. In cohorts 1 to 3, 6 subjects received remimazolam and 1 placebo. From cohort 4 onward, an additional 3 subjects in each cohort received midazolam (0.075 mg/kg). Safety, pharmacokinetics, and pharmacodynamics were measured. A stop criterion of loss of consciousness for >5 minutes in >50% of subjects was predefined.

RESULTS: The stop criterion was reached in cohort 9 (0.30 mg/kg remimazolam) so that 81 subjects were enrolled. Remimazolam was well tolerated in all dose cohorts, and no serious adverse events (AEs) were reported. Three AEs of mild (SpO2 85%–88%) hemoglobin desaturation (2 in the remimazolam groups and 1 in the midazolam group) resolved spontaneously, and 1 AE of moderate hemoglobin desaturation (SpO2 75%) resolved with a chin lift in the highest remimazolam dose group. No supplemental oxygen or manual ventilation was required. Vital signs remained stable throughout, although there was an increase in heart rate 2 minutes postdose for both remimazolam and midazolam. There were no reports of hypo- or hypertension. The pharmacokinetic behavior of remimazolam was linear and its systemic clearance approximately 3 times that of midazolam. Clearance was essentially independent of body weight. A rapid onset and dose-dependent sedation was observed after administration of remimazolam at 0.05 mg/kg and higher. Remimazolam (0.075 to 0.20 mg/kg) induced peak sedation levels similar to or higher than those achieved with midazolam (0.075 mg/kg). Median recovery times after approximately equieffective doses of remimazolam (0.10 and 0.15 mg/kg) and midazolam (0.075 mg/kg) were 10 and 40 minutes, respectively.

CONCLUSIONS: Remimazolam provided sedation with rapid onset and offset, and was well tolerated. There was no supplemental oxygen or ventilation required. On the basis of these data, further studies on the potential utility of remimazolam for sedation/anesthesia are warranted.

Published ahead of print December 20, 2011 Supplemental Digital Content is available in the text.

From the *Department of Anesthesiology and Critical Care Medicine, Johns Hopkins University School of Medicine, Baltimore, Maryland; PAREXEL International, Early Phase Clinical Unit, Harbor Hospital, Baltimore, Maryland; PAION UK Ltd., Cambridge, United Kingdom.

Funding: This work was funded by PAION UK Ltd.

Conflict of Interest: See Disclosures at the end of the article.

Reprints will not be available from the authors.

Address correspondence to Keith M. Borkett, BSc, PAION UK Ltd., Compass House, Vision Park, Histon, Cambridge CB24 9ZR, UK. Address e-mail to

Accepted September 30, 2011

Published ahead of print December 20, 2011

Remimazolam is an ester that is designed to undergo rapid hydrolysis in the body by nonspecific tissue esterases to an inactive carboxylic acid metabolite, CNS 7054.1 This mechanism of deactivation should result in a very rapid and predictable offset profile. In vitro, remimazolam binds to brain benzodiazepine sites (γ-aminobutyric acid A [GABAA] receptor) with high affinity, while the carboxylic acid metabolite (CNS 7054) shows >300 times lower affinity. No evidence for off-target activities of remimazolam or CNS 7054 has been observed in vitro. Remimazolam enhances GABA currents in cells stably transfected with subtypes of the GABAA receptor. Remimazolam, like other “classic” benzodiazepines, does not show clear selectivity between subtypes of the GABAA receptor.1

Remimazolam has sedative actions in rodents, inducing a reduction of locomotor activity at lower doses, and a loss-of-righting reflex at higher doses. The duration of loss-of-righting reflex was short (<15 minutes) and was reversed by treatment with the benzodiazepine antagonist flumazenil.1 The sedative profile of remimazolam has also been examined in 2 larger species, pigs and sheep.26 In both species, IV dosing of remimazolam induced rapid sedation of short duration. We report the results from a first-in-human, phase I, double-blind, single-dose escalation study of remimazolam that was designed to determine its safety, pharmacokinetics (PK), and pharmacodynamics (PD) in healthy adults. We provide the safety and efficacy results, and an accompanying study published in this issue of the Journal will provide the detailed PK and PD of this compound.7

Back to Top | Article Outline


Study Conduct

This study was conducted at PAREXEL International, Early Phase Clinical Unit, in Baltimore, Maryland, in accordance with the Declaration of Helsinki, in compliance with Good Clinical Practice, and local and Food and Drug Administration regulatory requirements. The clinical study protocol and informed consent forms were reviewed and approved by an IRB (Chesapeake Research Review, Columbia, Maryland) before study start. Written informed consent was obtained from all subjects before the start of the study.

Back to Top | Article Outline


Subjects eligible to take part in this study were healthy males or females, ages 18 to 55 years inclusive, weighing between 60 and 100 kg, with a body mass index (BMI) of 18 to 30 kg/m2. The subjects had a grade 1 or 2 Mallampati score,8 and were nonsmokers for at least the previous 6 months. Female subjects were not pregnant, were nonlactating, and were using medically acceptable forms of birth control, or were of nonchild-bearing potential.

Back to Top | Article Outline


This was a single-center, double-blind, randomized, single ascending-dose study of remimazolam administered as a 1-minute IV injection, compared with midazolam (0.075 mg/kg), and placebo. Up to 91 healthy male and female adults were planned for enrollment in up to 10 cohorts. In the first 3 cohorts, subjects were randomized to receive either a single dose of remimazolam or placebo in a 6:1 ratio. From cohort 4 onward, an additional 3 subjects were randomized to receive midazolam (Table 1). No formal sample size calculations were performed.

Table 1

Table 1

Subjects were admitted to the Clinical Pharmacology Unit the day before study drug administration. They were required to fast (no food or liquids of any kind) for at least 8 hours before drug administration, which took place in the morning. Subjects remained supine throughout the procedures. During study drug administration, a board-certified anesthesiologist was present and remained until the subject recovered consciousness and had normal cardiovascular and respiratory function. Oxygen and the benzodiazepine reversal drug flumazenil were immediately available if required, but were not to be administered unless necessary. Subjects were allowed clear fluids 2 hours after dosing, and lunch was served approximately 4 hours postdose. Subjects were discharged from the unit after all 24-hour postdose assessments had been completed on day 2. On day 8 (±2 days), subjects returned to the unit for safety assessments.

Because this was a double-blind study, the subjects and Clinical Pharmacology Unit personnel remained blinded to treatment assignment throughout. After completion of dosing for each cohort, a blinded data monitoring committee met to consider the safety data and allow the study to continue to the next dose cohort, if appropriate. There was also a predefined stopping rule: should no safety issues prevent dose escalation, the study would stop if loss of consciousness (Modified Observer's Assessment of Alertness/Sedation9 [MOAA/S] of <2) for ≥5 minutes was observed in >50% of subjects in a single cohort. Table 2 details the MOAA/S score used.

Table 2

Table 2

Back to Top | Article Outline

Measurements: Efficacy

During the study, sedation was measured by bispectral index (BIS) monitoring,10 and MOAA/S score assessments. The BIS numeric value was displayed on a BIS A-2000 monitor, which was set to the time interval default value of 1 minute to allow for signal averaging and artifact detection. The BIS monitoring units did not have a continuous recording capability. The confounding effects of stimulation producing movement were mitigated by recording the BIS value immediately before the MOAA/S assessment. The MOAA/S was used as the primary indication of sedation. The BIS and MOAA/S scores were recorded predose and at 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 12, 15, 20, 25, 30, 35, 40, 45, 50, and 60 minutes after the start of the infusion. An additional MOAA/S assessment was performed at 120 minutes postdose.

The time to fully alert was measured from the time of first postdose MOAA/S score of <5 to the first of 3 consecutive MOAA/S scores of 5.

Back to Top | Article Outline

Measurements: Plasma Sampling, Bioanalysis, and Pharmacokinetics

One hour before dosing and under local anesthetic, an arterial line for PK sample collection was placed in the distal radial artery of the nondominant arm. Radial arterial PK samples were collected 1, 2, 3, 4, 6, 8, 10, 12, 15, 20, 30, 45, and 60 minutes after and 2, 3, and 4 hours after the start of the infusion and the arterial line then removed. In the opposite (dominant) arm, 2 IV cannulae were placed: one for study drug administration, and one for PK sampling. The cannula for study drug administration was upstream of the cannula for PK sampling and was removed after study drug dosing. Venous blood samples were collected 2, 3, 4, 6, 8, and 12 hours postdose. Plasma concentrations of midazolam, remimazolam, and CNS 7054 were measured by high-performance liquid chromatography (HPLC) with tandem mass spectrometric detection using a SCIEX API 3000 instrument with a Turbo IonSpray interface in positive mode at Aptuit (Research Avenue South, Heriot Watt University Research Park, Riccarton, Edinburgh, UK). Deuterium labeled D4-remimazolam and D4-CNS 7054 were used as internal standards. The compounds were extracted from plasma using solid-phase methodology (Waters Sep-Pak C18 100 mg, 2 mL, 96-well plates). The HPLC system used a Varian Pursuit 3 μ Diphenyl 50 × 2.0 mm column and an isocratic mobile phase (75% aqueous 0.1% formic acid and 25% acetonitrile) at a flow rate of 350 μL/min and a temperature of 55°C. The approximate elution times and nominal mass transitions for remimazolam and CNS 7054 were 1.55 minutes, 439.1 to 407.2, and 2.5 minutes, 425.1 to 362.9, respectively. The method was validated over the range of 1 to 1000 ng/mL for both remimazolam and CNS 7054 using a sample volume of 200 μL. A separate assay was validated over the range 0.2 to 100 ng/mL for remimazolam only. A Phenomenex Luna 3 μ, 50 × 2.0 mm PFP HPLC column was used for the midazolam bioanalysis at a temperature of 35°C and a mobile phase of 0.1% formic acid in aqueous methanol at a flow rate of 300 μL/min and a step-gradient (5% methanol for 2 minutes, 95% methanol for 0.6 minutes). The approximate elution times and mass transitions for midazolam and the internal standard, alprazolam, were 2.1 minutes, 326.4 to 291.2, and 2.3 minutes, 309.4 to 281.0, respectively, and the method was validated over the range of 1 to 100 ng/mL using 100-μL samples.

Because the venous plasma levels of remimazolam tended to be higher than the arterial ones at the same time points (2, 3, and 4 hours; see accompanying paper for details7), the noncompartmental PK analysis of this compound was undertaken with only the arterial concentrations. The midazolam and CNS 7054 venous and arterial plasma levels were similar, and so all concentrations were used in the analysis, the 2-, 3-, and 4-hour data being averaged. Systemic exposure (AUC(0−t)) was estimated from the time of dosing to the last measurable time point (t) by application of the trapezoidal rule, assuming an exponential change in concentration between adjacent time points. The terminal rate constant (k) was obtained from the log-linear slope of the later datapoints by weighted linear regression analysis and the terminal half-life (T½) as ln(2)/k. Exposure was extrapolated to infinity according to the following equation:

where conc(t) is the last measurable concentration. The area under the first moment curve was estimated similarly and extrapolated to infinity according to the following equation:

Clearance (CL) was calculated as Dose/AUC(0 − ∞), mean residence time (MRT) as AUMC(0−∞)/AUC(0−∞), and steady state volume of distribution as CL × MRT. The maximum concentration and the time to peak concentration were observed parameters.

Back to Top | Article Outline

Measurements: Safety

Safety was monitored by physical examination (including an airway assessment and Doppler ultrasound or Allen Test11 at the screening visit only, and brief neurological assessments at all examinations), safety laboratory tests (biochemistry, hematology, and urinalysis), vital signs (heart rate, blood pressure [BP], and respiration rate), continuous 2-lead cardiac monitoring, 12-lead electrocardiograms (ECGs), continuous pulse oximetry, assessments of concomitant medications and AEs, and monitoring of airway management. Subjects also completed pain and nausea visual analog scales to assess their levels of comfort.

Twelve-lead ECG recordings were obtained using Cardiosoft CAM14 (G.E. Medical Systems, Milwaukee, WI). Standard 12-lead ECGs were performed after the subject had been resting supine ≥5 minutes at screening, at baseline, at 1 hour predose, and at 5 and 65 minutes postdose. To allow for QTc variability, at the 1-hour predose time point only, we collected triplicate ECGs 5 minutes apart, and all measurements averaged from these 3 ECGs. The following ECG parameters were recorded: PR interval, QRS interval, QT interval, and QTc interval (Bazett and Fridericia). All ECGs were evaluated by a qualified physician for the presence of abnormalities, and any clinically significant findings were documented as adverse events (AEs). Because the best QT correction approach for heart rate is still subject to discussion, both Bazett and Fridericia have been presented here.

Back to Top | Article Outline

Data-Monitoring Committee

The data-monitoring committee reviewed physical and neurological examination findings, safety laboratory variables (hematology, biochemistry, and urinalysis), 12-lead ECGs (including a cardiologist report on all ECGs performed on day 1), and any findings of clinical significance from a 2-lead cardiac telemetry (monitored continuously from 1 hour before to 4 hours after dosing). In addition, listings of vital signs (supine BP, heart rate, respiration rate, temperature) and SpO2 values were provided along with any airway management procedures that were performed. Any treatment-emergent AE reports were presented with date and time of onset and resolution, treatment given, severity, relatedness to study drug, action taken in relation to study drug, and outcome. An AE was defined as any untoward medical occurrence (sign, symptom, or laboratory finding) in a subject who was administered study drug.

The data-monitoring committee paid particular attention to the following events had they occurred:

  1. Dose-limiting toxicity: defined as a grading of 3 or 4 on the National Cancer Institute Common Terminology Criteria for Adverse Events v3.0.
  2. Clinically significant vital sign findings, confirmed by immediate repeat measurement, defined as the following:
    • (a) tachycardia: >100 beats per minute (bpm);
    • (b) bradycardia: <45 bpm;
    • (c) hypertension: an increase in systolic BP to ≥150 mm Hg and diastolic BP to ≥100 mm Hg, or an increase of BP necessitating medical intervention;
    • (d) hypotension: a decrease in systolic BP to ≤80 mm Hg and diastolic BP to ≤40 mm Hg, or a decrease of BP necessitating medical intervention;
    • (e) respiratory rate decrease: <8 breaths per minute;
    • (f) oxygen saturation (SpO2) <90%;
    • (g) other safety events of concern to the data-monitoring committee.

Unexpected serious SAEs considered related to remimazolam were to be considered a dose-limiting toxicity.

The decision to move to the next cohort and escalate the dose were based on the following criteria:

  • [Black Square] Any of the above events experienced by ≤1 of 6 remimazolam-treated subjects: enroll a new cohort of subjects at the next higher dose level
  • [Black Square] Any of the above events experienced by 2 of 6 remimazolam treated–subjects: repeat that dose level with a new cohort of subjects. The number of additional subjects was to be determined by the data-monitoring committee at the time of the meeting.
  • [Black Square] Any of the above events experienced by ≥3 of 6 remimazolam-treated subjects: that cohort defines the maximum tolerated dose, and no further subjects will be enrolled.
Back to Top | Article Outline

Dose Selection

The dose of 0.01 mg/kg was chosen as a conservative starting point for dosing in this study. On the basis of preclinical data and modeling, it was expected that minimal or no sedation would be seen in the first 3 cohorts (0.01 to 0.05 mg/kg remimazolam). The 0.075 mg/kg remimazolam dose level (cohort 4) was the first at which sedation was expected to be observed. Hence, from this cohort onward, midazolam (0.075 mg/kg) was administered to an additional 3 subjects per cohort, taking the total subject numbers up to 10 per cohort from cohort 4 onward. Midazolam was given as a 1-minute infusion to maintain the study blind, and the dose of 0.075 mg/kg was chosen as being in the center of the total dose range used clinically to achieve conscious sedation.a

Back to Top | Article Outline



Eighty-one subjects completed the study, of whom 54 received remimazolam, 18 midazolam, and 9 placebo. All subjects who enrolled completed the study; there were no withdrawals. The range of cohort means for weight (mean range: 75.4 to 81.9 kg) and BMI (mean range: 24.4 to 26.5 kg/m2) were similar among treatment groups (Table 3). The mean age ranged from 29 years (0.25 mg/kg treatment group) to 41 years (0.025 mg/kg treatment group); however, since all subjects were healthy adults, analyses of safety, PD, and PK were not expected to be affected by the age differences.

Table 3

Table 3

No treatment group had >50% females, and 3 treatment groups were all male (remimazolam: 0.01, 0.05, and 0.075 mg/kg cohorts).

Back to Top | Article Outline

Pharmacokinetic Data

The key PK parameters for remimazolam, CNS 7054, and midazolam are shown in Tables 4 and 5, and the arterial plasma PK profiles for remimazolam and CNS 7054 are shown in Figures 1 and 2, respectively. The consistent values of the systemic clearance of remimazolam and the apparent clearance of CNS 7054 showed that their PK were linear across the dose range studied. The clearance of remimazolam was rapid (overall mean = 70.3 ± 13.9 L/h) and distribution moderate (mean steady state volume of distribution = 34.8 ± 9.4 L). Systemic clearance of midazolam was approximately one-third that of remimazolam (mean 23.0 ± 4.5 L/h) and steady state volume of distribution over twice as large (mean 81.8 ± 27.1 L). As a result, the mean residence time of midazolam was 7 times that of remimazolam (3.6 hours in comparison with 0.51 hour). The apparent clearance of the metabolite, CNS 7054, was relatively slow at 4.22 ± 1.25 L/h in comparison with the systemic clearance of the parent compound (70 L/h) and its apparent volume of distribution small at 17.5 ± 3.8 L. The terminal half-life of the metabolite (2.89 ± 0.65 hours) was nearly 4 times that of remimazolam (0.75 ± 0.15 hour).

Table 4

Table 4

Table 5

Table 5

Figure 1

Figure 1

Figure 2

Figure 2

A plot of body weight against the systemic clearance of remimazolam suggests no clear relationship (Fig. 3).

Figure 3

Figure 3

Back to Top | Article Outline

Pharmacodynamic Data

Dose escalation proceeded as planned throughout the cohorts until cohort 9 (0.30 mg/kg) was reached. Six of 10 subjects (60%) in this cohort experienced loss of consciousness (MOAA/S scores of <2) for a minimum of 5 minutes. The predefined stopping criterion was therefore met, and further dose escalation was stopped. After the blind was broken, it was revealed that 5 of these 6 subjects had been treated with remimazolam and 1 with midazolam.

No sedation was observed in the placebo group. The onset of sedation was rapid for both midazolam and remimazolam (at doses of 0.05 mg/kg and above), with the median MOAA/S scores for remimazolam decreasing below 4 at doses of ≥0.075 mg/kg either during the infusion of study drug or within 60 seconds thereafter (Fig. 4). The degree and duration of sedation with remimazolam were dose dependent, with the peak effect of sedation being observed approximately 1 to 4 minutes after the start of the infusion. At the higher doses of 0.10 to 0.20 mg/kg, remimazolam appeared to produce greater sedation, while still maintaining quicker recovery from sedation, when compared with 0.075 mg/kg midazolam.

Figure 4

Figure 4

As shown in Figure 4, the mean duration of sedation increased with increasing remimazolam dose; from approximately 8 minutes (0.075 mg/kg) to approximately 40 minutes (0.25 mg/kg). Durations of loss of consciousness also increased with remimazolam dose, with median MOAA/S scores of <2 for approximately 0, 1, 3, 5, 9, and 12 minutes in 0.075, 0.1, 0.15, 0.2, 0.25, and 0.3 mg/kg remimazolam treatment groups, respectively.

Variable levels of sedation were observed in subjects treated with 0.075 mg/kg midazolam. In comparison with remimazolam dosed at 0.10 to 0.20 mg/kg, subjects treated with midazolam were not as deeply sedated, but were generally sedated for longer. Mean times from first MOAA/S of <5 to fully alert (defined as the first of 3 consecutive MOAA/S scores of 5 after onset of sedation) showed a dose-dependent increase across remimazolam treatment groups (Table 6). Subjects treated with 0.075, 0.1, 0.15, 0.2, 0.25, and 0.3 mg/kg remimazolam were fully alert after 5.5, 10.5, 10.0, 20.0, 34.0, and 31.5 minutes (medians) of sedation, respectively. In comparison, subjects treated with midazolam were sedated for a median of 40 minues.

Table 6

Table 6

The BIS values also reflected the quick onset of sedation of both remimazolam and midazolam, and a more rapid recovery from sedation after remimazolam than after midazolam treatment, as shown in Figure 5.

Figure 5

Figure 5

Back to Top | Article Outline

Safety Data

Most of the AEs were considered mild in intensity and unrelated to trial medication. Seventy-eight percent of both placebo and midazolam subjects reported AEs, compared with 33% (0.075 mg/kg) to 100% (0.1 mg/kg) for the various cohorts treated with remimazolam. There was no apparent remimazolam dose response in the number of AEs, or frequency of subjects reporting AEs.

The most commonly occurring AEs were secondary to the arterial catheterization, and were, hence, considered unrelated to trial medication. Treatment-related AEs were experienced by 17% to 33% of subjects in most treatment groups, including placebo and midazolam. The most frequently observed AEs with a suspected relationship to remimazolam were headache (4 of 54 subjects, 7%) and somnolence (3 of 54 subjects, 6%). These were each reported in 1 of 18 subjects (6%) receiving midazolam. No cases of headache or somnolence were reported in subjects treated with placebo.

There were 4 AEs of hemoglobin desaturation reported in 3 subjects during the trial: subjects 702 (0.2 mg/kg remimazolam; 1 AE), 906 (0.3 mg/kg remimazolam; 2 AEs), and 704 (midazolam; 1 AE). The AEs were of 1- to 10-minute duration. Three of them were considered mild (O2 saturation 85%–88%) and resolved spontaneously. One AE in subject 906 (0.3 mg/kg remimazolam) was considered moderate in intensity (O2 saturation 75%) and resolved after 2 minutes. This AE required the only instance of airway management (chin lift) during the study. All hypoxic AEs were considered probably or definitely related to trial medication.

In the midazolam group and remimazolam treatment groups of 0.075 mg/kg and above, mean heart rate was increased by approximately 15 to 20 bpm at 2 minutes postdose. These changes were not considered clinically significant.

Some QTcB intervals were prolonged in both the midazolam and higher-dose remimazolam treatment groups. Fourteen subjects had prolonged QTcB of >30 ms, postdose QTcB values of ≥450 ms, or both: 4 of 18 subjects treated with midazolam (22%) and 10 of 54 treated with remimazolam (19%). These changes were not considered clinically relevant and in most cases relate to changes in heart rate. There were no significant changes in QTcF intervals.

The BP was stable, with 4 reports of changes (1 low systolic BP at 8 hours postdose, and 3 elevations) that were considered to be mild and unrelated to the study medication. There were no changes in clinical laboratory values, temperature, or respiration rate that were considered clinically significant during the study.

Both midazolam (0.075 mg/kg) and remimazolam induced only mild and transient nausea. Neither midazolam nor remimazolam caused notable pain on injection. There were no serious AEs reported.

Back to Top | Article Outline


Midazolam is currently the most commonly used drug to sedate patients for procedures such as upper gastrointestinal endoscopy and colonoscopy. A drawback of midazolam is that the recovery from sedation can be prolonged because of the production of active metabolites and a reliance for metabolism on the liver enzyme cytochrome P450 3A4. Extended sedation results in additional resources being required to monitor the patient as well as reducing the throughput of patients in clinics undertaking procedures that require sedation.

Remimazolam was designed to be metabolized rapidly to an inactive metabolite by widespread tissue esterases and therefore to have a predictable and short duration of sedation that is independent of organ function. Here we provide data from the first human study of remimazolam in comparison with placebo and midazolam. Remimazolam was examined at a range of doses and midazolam was used at a single dose in the center of the range used clinically for procedural sedation.

As predicted from preclinical studies,14 remimazolam showed dose-related sedative effects in human subjects. Onset of sedation occurred within 60 seconds of termination of drug infusion. The depth of sedation induced by remimazolam increased with dose. No sedation, or very minimal sedation, was observed at 0.01 and 0.025 mg/kg. Remimazolam dosed at 0.05 mg/kg resulted in small reductions in MOAA/S scores (to 4) and BIS scores (to 75). Doses of 0.075 mg/kg and above resulted in deeper sedation, as evidenced by MOAA/S scores of <2 and mean BIS scores of 60 to 70 soon after dosing. Administration of midazolam (0.075 mg/kg) resulted in minimum MOAA/S scores generally in the range of 3 to 4; however, there was some variability in MOAA/S scores, with some subjects being deeply sedated and others showing minimal sedation. The lowest mean BIS score after midazolam treatment was 73.

The duration of sedation induced by remimazolam also appeared to be dose dependent. In the dose range of 0.075 to 0.2 mg/kg, a considerably quicker recovery was observed in patients treated with remimazolam than with midazolam (0.075 mg/kg), with a median time for return to fully alert ranging from 5.5 to 20 minutes for remimazolam in comparison with 40 minutes for midazolam. Doses of remimazolam in the range 0.075 to 0.2 mg/kg may be appropriate for procedural sedation. Recovery took up to 50 minutes with higher doses of remimazolam (0.25 and 0.30 mg/kg). Such doses are unlikely to be useful for procedural sedation but may be used for the induction of anesthesia. As with duration of sedation, recovery time also appeared to be dose dependent.

The PK of remimazolam were linear across the dose range studied. Systemic clearance of remimazolam was rapid and distribution was moderate. As levels of remimazolam declined, those of the carboxylic acid metabolite, CNS 7054, increased. CNS 7054 showed a slower clearance and smaller volume of distribution than the parent compound.

The mean residence time of remimazolam was 0.50 hours in comparison with 3.56 hours for midazolam. The reason for this difference in residence time is that systemic clearance of midazolam was approximately one-third that of remimazolam with a volume of distribution more than twice as large. This difference in PK between remimazolam and midazolam is likely to be one factor in the more rapid recovery of the subjects receiving remimazolam. Of note is the observation that there was no clear relationship between weight and systemic clearance of remimazolam, within the weight range (60 to 100 kg) studied. This indicates that dosing by body weight may not offer significant advantage in terms of variability in systemic exposure in comparison with fixed doses, but further work is needed to confirm this.

Treatment-related AEs were similar across all 3 study treatments (remimazolam, midazolam, and placebo). The frequency or severity of AEs in the remimazolam groups was not dose dependent. Headache and somnolence were the most common, occurring with similar frequency in both remimazolam and midazolam groups. There were no clinically significant effects on heart rate, ECG, BP, respiratory rate, temperature, or clinical laboratory values.

The subjects were breathing room air throughout the trial. Three subjects experienced hemoglobin desaturation, 2 with remimazolam and 1 with midazolam. These AEs were short lived, and all but 1 resolved spontaneously. One subject treated with remimazolam (0.3 mg/kg) showed evidence of upper airway obstruction, which resolved with a chin lift. No supplemental oxygen or other airway manipulation was required. In practice, the episodes of hemoglobin desaturation during deep sedation should be preventable with supplemental oxygen, which is typically administered during procedures requiring sedation. In addition, deep sedation may be avoided by use of more appropriate bolus/top-up dosing regimes that may be identified in further clinical trials of remimazolam.

Hemodynamics appeared to be stable throughout, and on the basis of this initial study, remimazolam shows no evidence of significant cardiovascular effects. There was a mild increase in heart rate that was similar between remimazolam and midazolam.

From this first-in-human study the initial side effect profile of remimazolam appears similar to that of midazolam.

Back to Top | Article Outline


Remimazolam provided sedation with rapid onset and offset, and was well tolerated. Overall, the hemodynamics were generally stable throughout, although there was an increase in heart rate 2 minutes postdose, which was comparable between both remimazolam and midazolam. At the lower remimazolam doses (up to 0.15 mg/kg), SpO2 remained stable throughout. At the higher remimazolam doses (0.20 and 0.30 mg/kg), 3 incidences of hemoglobin desaturation were observed from 12 subjects (1 requiring a chin lift), in comparison with 1 incidence after midazolam. There was no supplemental oxygen or ventilation required. On the basis of these data, further studies on the potential utility of remimazolam for sedation/anesthesia are warranted.

Back to Top | Article Outline


Name: Laurie J. Antonik, MD.

Contribution: Conduct of study.

Attestation: Laurie J. Antonik has approved the final manuscript.

Conflict of Interest: Laurie J. Antonik has no connection with or financial interest in PAION. Laurie J. Antonik has acted in an advisory capacity to PAION as part of an expert panel.

Name: D. Ronald Goldwater, MD.

Contribution: Conduct of study.

Attestation: D. Ronald Goldwater has approved the final manuscript.

Conflict of Interest: D. Ronald Goldwater has no connection with or financial interest in PAION.

Name: Gavin J. Kilpatrick, PhD.

Contribution: Study design.

Attestation: Gavin J. Kilpatrick has approved the final manuscript.

Conflict of Interest: Gavin J. Kilpatrick was an employee of PAION at the time of conduct of the study. He owns shares of and share options in PAION.

Name: Gary S. Tilbrook, PhD.

Contribution: Data analysis.

Attestation: Gary S. Tilbrook has approved the final manuscript.

Conflict of Interest: Gary S. Tilbrook was an employee of PAION at the time of conduct of the study.

Name: Keith M. Borkett, BSc.

Contribution: Study design and manuscript preparation.

Attestation: Keith M. Borkett has seen the original study data, reviewed the analysis of the data, approved the final manuscript, and is the author responsible for archiving the study files.

Conflict of Interest: Keith M. Borkett was an employee of PAION at the time of conduct of the study.

This manuscript was handled by: Tony Gin, FANZCA, FRCA, MD.

Back to Top | Article Outline


The authors would like to thank Michele Field and James Lees for their support in conducting this study and Karin Wilhelm-Ogunbiyi, MD, Hugh Wiltshire, PhD, and Prof. Peter S. Glass for constructive review of the manuscript.

a Midazolam package insert. Hospira Inc., 2008.
Cited Here...

Back to Top | Article Outline


1. Kilpatrick GJ, McIntyre MS, Cox RF, Stafford JA, Pacofsky GJ, Lovell GG, Wiard RP, Feldman PL, Collins H, Waszczak BL, Tilbrook GS. CNS 7056: a novel ultra-short-acting benzodiazepine. Anesthesiology 2007;107:60–6
2. Mutter C, Rudolf G, Diemunsch P, Tilbrook GS, Borgeat A. CNS 7056X: an ultra-short acting benzodiazepine: pharmacokinetic and pharmacodynamic study in pig. Anesthesiology 2006;105; A1610
3. Upton RN, Ludbrook GL. A model of the kinetics and dynamics of induction of anaesthesia in sheep: variable estimation for thiopental and comparison with propofol. Br J Anaesth 1999;82:890–9
4. Upton RN, Martinez AM, Grant C. A dose escalation study in sheep of the effects of the benzodiazepine CNS 7056 on sedation, the EEG and the respiratory and cardiovascular systems. Br J Pharmacol 2008;155:1–10
5. Upton RN, Martinez AM, Grant C. Comparison of the sedative properties of CNS 7056, midazolam, and propofol in sheep. Br J Anaesth 2009;103:848–57
6. Upton RN, Somogyi AA, Martinez AM, Colvill J, Grant C. Pharmacokinetics and pharmacodynamics of the short-acting sedative CNS 7056 in sheep. Br J Anaesth 2010;105:798–809
7. Wiltshire HR, Kilpatrick GJ, Tilbrook GS, Borkett KM. A placebo- and midazolam-controlled phase I single ascending dose study evaluating the safety, pharmacokinetics, and pharmacodynamics of remimazolam (CNS 7056): Part II. Population pharmacokinetic and pharmacodynamic modeling and simulation. Anesth Analg 2012;115:284–96
8. Samsoon GL, Young JR. Difficult tracheal intubation: a retrospective study. Anaesthesia 1987;42:487–90
9. Chernik DA, Gillings D, Laine H, Hendler J, Silver JM, Davidson AB, Schwam EM, Siegel JL. Validity and reliability of the Observer's Assessment of Alertness/Sedation Scale: study with intravenous midazolam. J Clin Psychopharmacol 1990;10:244–51
10. Bower AL, Ripepi A, Dilger J, Boparai N, Brody FJ, Ponsky JL. Bispectral index monitoring of sedation during endoscopy. Gastrointest Endosc 2000;52:192–6
11. Husum B, Berthelsen P. Allen's test and systolic arterial pressure in the thumb. Br J Anaesth 1981;53:635–7
© 2012 International Anesthesia Research Society