Introduction: The Persistent Challenge of Optimal ICU Sedation
For decades, clinicians managing patients on mechanical ventilation have grappled with the ‘sedation paradox.’ While sedation is necessary to ensure patient comfort, safety, and synchrony with the ventilator, excessive or prolonged sedation is intrinsically linked to delayed weaning, increased rates of delirium, and longer intensive care unit (ICU) stays. Propofol, a GABA-receptor agonist, has long been the cornerstone of ICU sedation due to its rapid onset and short duration of action. However, the emergence of alpha-2 adrenergic receptor agonists—specifically dexmedetomidine and clonidine—offered a theoretically superior alternative. These agents provide ‘cooperative sedation,’ where patients remain rousable and interactive, potentially facilitating earlier liberation from mechanical ventilation.
Despite several small-scale studies suggesting benefits, high-quality evidence comparing these three agents in a large, pragmatic setting remained sparse. The A2B Randomized Clinical Trial, recently published in JAMA, provides a definitive look at whether transitioning from propofol to alpha-2 agonist-based sedation actually delivers on its promise of faster extubation.
Highlights of the A2B Trial
1. The study found no statistically significant difference in the time to successful extubation when comparing dexmedetomidine or clonidine to propofol-based sedation.
2. Alpha-2 agonists were associated with significantly higher rates of clinically relevant adverse events, including severe bradycardia and patient agitation.
3. The findings remained consistent across various patient subgroups, including those with sepsis and high organ-failure scores, suggesting that the results are broadly applicable to a general ICU population.
Study Design: A Pragmatic Multicenter Approach
The A2B trial was a pragmatic, open-label, randomized clinical trial conducted across 41 ICUs in the United Kingdom. Between December 2018 and October 2023, researchers enrolled 1,404 adult patients who had been on mechanical ventilation for less than 48 hours and were expected to require continued ventilation for at least another 48 hours. At the time of enrollment, all patients were receiving propofol and an opioid.
Patients were randomized into three groups:
1. Dexmedetomidine-based sedation
2. Clonidine-based sedation
3. Propofol-based sedation (Usual Care)
The primary objective was to evaluate the time from randomization to successful extubation. Clinicians followed bedside algorithms targeting a Richmond Agitation-Sedation Scale (RASS) score of -2 to 1. In the alpha-2 agonist groups, propofol was tapered off while the study drug was up-titrated. Supplemental propofol was permitted if the primary agent failed to maintain the target sedation depth.
Figure 1. Flow Diagram for Screening, Randomization, and Follow-Up in the Study.
Key Findings: No Shortcut to Extubation
The results of the A2B trial were striking in their lack of superiority for the newer sedation strategies. The primary outcome—time to successful extubation—showed no significant advantage for either dexmedetomidine or clonidine over propofol.
Time to Extubation and Statistical Analysis
The median time from randomization to successful extubation was:
– 136 hours for dexmedetomidine (95% CI, 117-150)
– 146 hours for clonidine (95% CI, 124-168)
– 162 hours for propofol (95% CI, 136-170)
While the absolute median time was numerically shorter for dexmedetomidine, the subdistribution hazard ratio (HR) for time to successful extubation was 1.09 (95% CI, 0.96-1.25; P = .20) for dexmedetomidine vs. propofol and 1.05 (95% CI, 0.95-1.17; P = .34) for clonidine vs. propofol. Neither reached the threshold for statistical significance.
Table 2. Primary and Secondary Outcomes.
| Sedation group | Absolute difference, % (95% CI)a | Relative difference, hazard ratio (95% CI) | |||||
|---|---|---|---|---|---|---|---|
| Dexmedetomidine | Clonidine | Propofol | Dexmedetomidine vs propofol | Clonidine vs propofol | Dexmedetomidine vs propofol | Clonidine vs propofol | |
| No. of patients | 457 | 476 | 471 | ||||
| Primary outcome | |||||||
| Time to successful extubation, median (95% CI), h | 136 (117 to 150) | 146 (124 to 168) | 162 (136 to 170) | 3.13 (−2.33 to 8.43)b | 1.77 (−3.25 to 6.90)b | 1.09 (0.96 to 1.25) | 1.05 (0.95 to 1.17) |
| P value | .20 | .34 | |||||
| Secondary outcomes | |||||||
| Mortality, No./total (%) | |||||||
| In ICU | 96/454 (21) | 103/472 (22) | 105/467 (22) | −1.34 (−6.67 to 4.01) | −0.66 (−5.98 to 4.66) | 0.95 (0.72 to 1.26) | 0.98 (0.74 to 1.28) |
| At 90 d | 122/457 (27) | 138/476 (29) | 135/471 (29) | −1.97 (−7.71 to 3.80) | 0.33 (−5.44 to 6.09) | 0.95 (0.74 to 1.21) | 1.03 (0.82 to 1.31) |
| At 180 d | 132/457 (29) | 145/476 (30) | 141/471 (30) | −1.05 (−6.91 to 4.81) | 0.53 (−5.32 to 6.37) | 0.98 (0.77 to 1.24) | 1.04 (0.82 to 1.31) |
| Time from randomization to ICU discharge, median (95% CI), d | 11 (10 to 12) | 12 (10 to 13) | 12 (11 to 13) | 1.05 (0.92 to 1.19) | 1.01 (0.91 to 1.12) | ||
| Time to optimization of sedation | |||||||
| No. of 12-h nursing shifts from randomization to first RASS score of ≥−2, median (95% CI) | 2 (2 to 2) | 2 (2 to 2) | 2 (2 to 2) | 1.06 (0.96 to 1.17) | 1.06 (0.97 to 1.16) | ||
| Time from randomization to first day with optimum sedation (no recorded agitation, unnecessary deep sedation, or pain behavior), median (95% CI), d | 3 (3 to 4) | 3 (3 to 4) | 3 (2 to 3) | 0.94 (0.83 to 1.07) | 0.95 (0.82 to 1.10) | ||
| Patients with ≥1 event, No. (%) | |||||||
| Serious adverse eventsc | 20 (4.4) | 12 (2.5) | 4 (0.8) | ||||
| Adverse eventsd | 47 (10.3) | 26 (5.5) | 16 (3.4) | ||||
| Long-term patient-centered outcomes at 90 d | |||||||
| Health-related quality of life (EQ-5D-5L) visual analog scale score, mean (SD)e,f | (n = 93) 68 (18) | (n = 105) 60 (21) | (n = 120) 63 (23) | MD, 4.99 (−0.64 to 10.63) | MD, −2.13 (−7.58 to 3.32) | ||
| Health-related quality of life (EQ-5D-5L) index score (excluding deaths), mean (SD)f,g | (n = 92) 0.59 (0.29) | (n = 103) 0.57 (0.28) | (n = 118) 0.54 (0.33) | MD, 0.05 (−0.04 to 0.13) | MD, 0.02 (−0.06 to 0.10) | ||
| Long-term patient-centered outcomes at 180 d | |||||||
| Hospital Anxiety and Depression Scale score, mean (SD)f,h | (n = 69) 13.4 (9.9) | (n = 80) 13.6 (9.6) | (n = 85) 14.8 (9.5) | MD, −1.41 (−4.49 to 1.68) | MD, −1.18 (−4.14 to 1.79) | ||
| Revised Impact of Events Scale (posttraumatic stress) score, mean (SD)f,i | (n = 62) 24.6 (20.8) | (n = 71) 22.7 (21.7) | (n = 71) 30.6 (24.9) | MD, −6.06 (−13.80 to 1.69) | MD, −7.97 (−15.45 to −0.49) | ||
| Telephone Montreal Cognitive Assessment score, mean (SD)f,j | (n = 67) 16.5 (3.5) | (n = 57) 17.0 (3.3) | (n = 63) 16.3 (4.0) | MD, 0.16 (−1.09 to 1.40) | MD, 0.66 (−0.63 to 1.96) | ||
| Health-related quality of life (EQ-5D-5L) visual analog scale score, mean (SD)e,f | (n = 71) 68 (23) | (n = 84) 67 (20) | (n = 90) 66 (23) | MD, 0.82 (−6.00 to 7.64) | MD, 0.28 (−6.22 to 6.79) | ||
| Health-related quality of life (EQ-5D-5L) index score, mean (SD)f,g | (n = 69) 0.61 (0.35) | (n = 83) 0.61 (0.29) | (n = 88) 0.54 (0.34) | MD, 0.06 (−0.04 to 0.17) | MD, 0.07 (−0.03 to 0.16) | ||
Figure 2. Cumulative Incidence Plot for Time From Randomization to Successful Extubation.
The median duration of follow-up was 4.7 days (IQR, 2.0-9.8 days) for dexmedetomidine; 4.9 days (IQR, 2.0-10.8 days) for clonidine; and 5.0 days (IQR, 2.2-11.2 days) for propofol. There were initially 456 patients at risk in the dexmedetomidine group rather than 457 because information on ultimate extubation status was not available for 1 patient
Secondary Outcomes and Safety Profiles
Perhaps more concerning than the lack of efficacy were the safety and tolerance data. Patients in the alpha-2 agonist groups experienced higher rates of agitation and cardiovascular instability:
– Agitation: The risk ratio (RR) for agitation was 1.54 for dexmedetomidine and 1.55 for clonidine compared to propofol. This suggests that while patients are more ‘awake,’ they may also be more distressed or difficult to manage during the weaning phase.
– Severe Bradycardia: Heart rates below 50 beats per minute were significantly more common with dexmedetomidine (RR 1.62) and clonidine (RR 1.58).
– Mortality: There were no significant differences in 180-day mortality between the groups, confirming that while these drugs change the trajectory of sedation, they do not inherently alter the risk of death in this population.
Figure 3. Box-and-Whisker Plots Showing the Highest Richmond Agitation-Sedation Scale (RASS) Scores Achieved.
Includes patients who have not yet achieved the primary outcome.
Expert Commentary: Interpreting the Neutral Result
The A2B trial follows in the footsteps of the SPICE III trial, which also failed to show a mortality benefit for dexmedetomidine. However, A2B is unique in its direct comparison with clonidine and its focus on the pragmatic transition from propofol.
Several factors may explain why alpha-2 agonists did not outperform propofol. First, modern ‘usual care’ with propofol has improved significantly. In the A2B trial, the propofol group was managed with a light sedation target (RASS -2 to 1), which may have minimized the ‘oversedation’ traditionally associated with GABA-agonists. When propofol is used judiciously at low doses to maintain a light state, its rapid clearance makes it a formidable comparator for any weaning study.
Second, the higher rate of agitation in the dexmedetomidine and clonidine groups may have inadvertently delayed extubation. Clinicians may be hesitant to extubate a patient who is agitated or tachypneic, even if those symptoms are a side effect of the sedative transition rather than a sign of respiratory failure.
Third, the pharmacodynamic profile of clonidine—which has a longer half-life and less predictable clearance than dexmedetomidine—may make it less ideal for the dynamic environment of ICU weaning. However, even dexmedetomidine, with its more favorable profile, failed to significantly move the needle on extubation timing.
Biological Plausibility and Mechanistic Insights
Alpha-2 agonists work by inhibiting the release of norepinephrine from the locus coeruleus, inducing a state that mimics natural non-REM sleep. In contrast, propofol enhances GABAergic transmission, leading to a more profound depression of the central nervous system. The biological rationale for using alpha-2 agonists was that by avoiding the global ‘brain shutdown’ of GABA-agonists, patients would maintain better respiratory drive and cognitive clarity.
However, the A2B results suggest that the ‘clarity’ provided by alpha-2 agonists often comes at the cost of sympathetic dysregulation (bradycardia) or inadequate control of the distress associated with an endotracheal tube (agitation). In many patients, the ‘heavier’ but more stable sedation provided by propofol may actually facilitate a more orderly progression toward weaning by preventing the physiological stress of agitation.
Conclusion: Clinical Practice Implications
The A2B Randomized Clinical Trial provides strong evidence that for the general population of critically ill adults requiring mechanical ventilation, dexmedetomidine or clonidine should not be viewed as superior ‘magic bullets’ for reducing ventilation time. Propofol remains a safe, effective, and often more stable option for maintaining light sedation targets.
Clinicians should continue to prioritize the depth of sedation (targeting RASS -2 to 1) over the specific choice of agent. While dexmedetomidine may still have a role in specific scenarios—such as patients with refractory delirium or those who are difficult to wean due to propofol-induced respiratory depression—its routine use as a primary replacement for propofol to speed up extubation is not supported by this high-level evidence.
Funding and Trial Registration
This study was funded by the National Institute for Health and Care Research (NIHR) Health Technology Assessment Programme. The trial is registered at ClinicalTrials.gov (NCT03653832).
Reference:
Walsh TS, Parker RA, Aitken LM, McKenzie CA, Emerson L, Boyd J, Macdonald A, Beveridge G, Giddings A, Hope D, Irvine S, Tuck S, Lone NI, Kydonaki K, Norrie J, Brealey D, Antcliffe D, Reay M, Williams A, Bewley J, Creagh-Brown B, McAuley DF, Dark P, Wise MP, Gordon AC, Perkins GD, Reade MC, Blackwood B, MacLullich A, Glen R, Page VJ, Weir CJ; A2B Trial Investigators. Dexmedetomidine- or Clonidine-Based Sedation Compared With Propofol in Critically Ill Patients: The A2B Randomized Clinical Trial. JAMA. 2025 Jul 1;334(1):32-45. doi: 10.1001/jama.2025.7200 . Erratum in: JAMA. 2025 Dec 4. doi: 10.1001/jama.2025.23530 . PMID: 40388916 ; PMCID: PMC12090071 .
