Dexmedetomidine vs Clonidine vs Propofol: Extubation in ICU Patients

The A2B trial evaluates sedation strategies on extubation timing in critical care.

dexmedetomidineclonidinepropofolsedationcritical carerandomized clinical trial Critical Care / ICU

Authors: 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

Citation: Walsh TS, Parker RA, Aitken LM, et al. Dexmedetomidine- or Clonidine-Based Sedation Compared With Propofol in Critically Ill Patients: The A2B Randomized Clinical Trial. JAMA. 2025;334(1):32-45. doi:10.1001/jama.2025.7200

Article Links: PubMed · Journal / DOI · Free Full Text (PMC)

🎓

Reviewed by a licensed pharmacist

Every analysis is reviewed and approved by a licensed pharmacist before publication, ensuring clinical accuracy and relevance.

Accurate and up-to-date information is our highest priority.

If you spot an error, please suggest a correction. If there is a topic you want covered, tell us and we will look at it.

Suggest a correction Request a topic

⚡ Bottom Line

In this large pragmatic UK RCT (n=1,404), neither dexmedetomidine nor clonidine reduced time to extubation compared with propofol in mechanically ventilated ICU patients. Both α2-agonists caused significantly more agitation and a ~60% higher rate of severe bradycardia (NNH ≈ 8). Routine use of dexmedetomidine or clonidine as primary ICU sedation is not supported by this evidence, propofol remains the preferred agent. Reserve α2-agonists for specific indications (e.g., adjunct for opioid-sparing, alcohol withdrawal, or procedural sedation) where the risk-benefit profile may differ.

Study Design

Pragmatic, open-label randomized clinical trial, duration: 5 years, registered as NCT03653832, funded by National Institute for Health and Care Research.

Trial NameA2B Trial (Alpha-2 Agonist vs Beta: Dexmedetomidine or Clonidine vs Propofol for ICU Sedation)

RegistrationISRCTN16228941

Blinding / AnalysisOpen-label, pragmatic RCT (unblinded). Primary outcome ascertained by unblinded local research teams — acknowledged limitation. Prespecified statistical analysis plan. Fine-Gray subdistribution hazard model for primary outcome (accounts for competing risk of death). ITT analysis population (1,404 of 1,438 randomized); 34 excluded per SAP.

Funding & COIFunded by the National Institute for Health and Care Research (NIHR) Health Technology Assessment Programme (15/57/07). Lead author reported receiving a senior clinical practitioner research award from NIHR. Multiple co-investigators reported consultancy fees from pharmaceutical companies (Bayer, GSK, Boehringer Ingelheim, AstraZeneca). No pharma funding of the trial itself.

Sample Size RationaleAssumed 53% extubation by day 7 in propofol group, 14% pre-extubation death. Expected ~7-day median duration (heavily skewed). Powered for 2-day mean difference = HR ~1.37. Target ≥1,300 patients for 90% power at α=0.025 (one-sided, per comparison). Final enrollment 1,438.

Population

1404 patients; adults within 48 hours of starting mechanical ventilation, expected to require at least 48 hours of ventilation; mean age 59.2 years, 64% male; UK ICU settings.

Baseline Characteristics

Characteristic	Dexmed (n=457)	Clonidine (n=476)	Propofol (n=471)
Age, mean (SD), y	58.8 (14.8)	59.7 (14.5)	59.2 (15.2)
Male sex, n (%)	297 (65%)	308 (65%)	305 (65%)
APACHE II score, mean (SD)	20.0 (8.0)	20.3 (8.1)	20.8 (8.5)
SOFA score, median (IQR)	8 (6–10)	8 (7–10)	8 (7–10)
Sepsis at enrollment, n (%)	297 (66%)	303 (65%)	308 (66%)
PRE-DELIRIC score, median (IQR), %	73 (53–85)	74 (55–86)	72 (51–87)
RASS −4 or −5 at randomization, n (%)	273 (60%)	305 (64%)	285 (61%)
Time MV → randomization, median (IQR), h	20.7 (12.9–31.4)	21.0 (13.3–32.1)	21.0 (13.4–30.5)
Primary diagnosis: Respiratory, n (%)	155 (36%)	158 (36%)	158 (38%)

Primary Outcome Not Met

Time from randomization to successful extubation (extubation + 48 h spontaneous breathing without MV) (Competing risk analysis; death = competing event)

Dexmedetomidine (n=457)

136 h

median (95% CI: 117–150 h)

Clonidine (n=476)

146 h

median (95% CI: 124–168 h)

Propofol (control) (n=471)

162 h

median (95% CI: 136–170 h)

Comparison	Effect Measure	Effect Size (95% CI)	p‑value	Favors
Dexmedetomidine vs Propofol	Subdistribution HR	1.09 (0.96–1.25)	0.20	neither (HR >1 favors dexmed, but NS)
Clonidine vs Propofol	Subdistribution HR	1.05 (0.95–1.17)	0.34	neither

Statistical Methods

Fine-Gray proportional subdistribution hazard regression (marginal model for clustered data) for primary outcome and ICU LOS, handles competing risk of pre-extubation death. Poisson regression for rates (agitation, bradycardia, arrhythmia). Mixed-effects proportional hazards for 180-day mortality. All analyses adjusted for site. Prespecified subgroup interaction tests for 4 variables. Post-hoc unadjusted risk differences calculated with epiR package.

Secondary Outcomes

Outcome	Timepoint	Effect (95% CI)	p‑value	Sig?
ICU mortality	During ICU stay	HR 0.95 (0.72–1.26 (Dexmed vs Prop))	NS	No
90-day all-cause mortality	90 days	HR 0.95 (0.74–1.21 (Dexmed vs Prop))	NS	No
180-day all-cause mortality	180 days	HR 0.98 (0.77–1.24 (Dexmed vs Prop))	NS	No
ICU length of stay	Randomization to ICU discharge	HR 1.05 (0.92–1.19 (Dexmed vs Prop))	NS	No
Agitation rate (RASS ≥+3)Favors propofol	Days 1–7 post-randomization	RR 1.54 (1.21–1.97 (Dexmed vs Prop))	<0.05	Yes
Agitation rate (RASS ≥+3)Favors propofol	Days 1–7 post-randomization	RR 1.55 (1.22–1.97 (Clonidine vs Prop))	<0.05	Yes
Time to first RASS ≥−2 (light sedation)	12-h nursing shifts from randomization	HR 1.06 (0.96–1.17 (Dexmed vs Prop))	NS	No
Time to optimum sedation (no agitation/deep sedation/pain)	Days post-randomization	HR 0.94 (0.83–1.07 (Dexmed vs Prop))	NS	No
PTSD score (IES-R) at 180 daysFavors clonidine	180 days	MD −7.97 (−15.45 to −0.49 (Clonidine vs Prop))	<0.05	Yes

Safety & Tolerability

Adverse Event	Dexmed n (%)	Clonidine n (%)	Propofol n (%)	Effect (95% CI)	p‑value
Severe bradycardia (HR <50 bpm) ⚠	151 (33%)	157 (33%)	94 (20%)	RR (Dexmed vs Prop) 1.62 (1.36–1.93)	<0.001
Cardiac arrhythmia	higher	similar	reference	RR (Dexmed vs Prop) 1.27 (1.15–1.40)	<0.001
Serious adverse events ⚠ ⚠	20 (4.4%)	12 (2.5%)	4 (0.8%)	RR (Dexmed vs Prop) ~5.5× (NR)	<0.05
Any adverse event	47 (10.3%)	26 (5.5%)	16 (3.4%)	RR (Dexmed vs Prop) ~3.0× (NR)	<0.05

⚠ = serious adverse event

Subgroup Analyses

Subgroup	Comparison	Effect (95% CI)	p‑value	Interaction p
Age <64 vs ≥64 years	Dexmed vs Propofol (primary outcome)	Interaction HR 0.90/10-yr increment (0.82–0.99 (continuous age))	NR	Significant (continuous age only)
Sepsis status (yes vs no)	Dexmed & Clonidine vs Propofol	HR NR (NR)	NR	NS
SOFA score (above vs below median of 8)	Dexmed & Clonidine vs Propofol	HR NR (NR)	NR	NS
Delirium risk (PRE-DELIRIC above vs below median of 73%)	Dexmed & Clonidine vs Propofol	HR NR (NR)	NR	NS
Cardiac arrhythmia by age	Dexmed vs Propofol	RR Higher in younger patients (NR)	NR	Significant

Number Needed to Treat / Harm

NNH

Severe Bradycardia (Dexmed)

33% vs 20% in propofol; ARR = 13%; NNH = 1/0.13 ≈ 8. For every 8 patients treated with dexmedetomidine instead of propofol, expect 1 additional severe bradycardia event.

NNH

Severe Bradycardia (Clonidine)

33% vs 20% in propofol; ARR = 13%; NNH = 1/0.13 ≈ 8. For every 8 patients treated with clonidine instead of propofol, expect 1 additional severe bradycardia event.

NNH

Serious Adverse Event (Dexmed)

4.4% vs 0.8% in propofol; ARR = 3.6%; NNH = 1/0.036 ≈ 28. For every 28 patients treated with dexmedetomidine instead of propofol, expect 1 additional serious adverse event.

Limitations

Open-label design, non-significant p-values, potential selection bias due to correction for baseline variable errors.

Clinical Interpretation

Evidence does not support superiority of dexmedetomidine or clonidine over propofol for extubation timing in ICU patients. Adverse event profile requires cautious interpretation.

Clinical Pearls

💊One in 3 patients on dexmedetomidine or clonidine developed severe bradycardia (HR <50 bpm), a 60% higher rate than propofol. This likely caused dose-limiting effects and explains why ~75-77% of α2-agonist patients also required concurrent propofol throughout the trial.

💊Despite pharmacological differences, median time to light sedation (RASS ≥−2) was identical at 2 nursing shifts across all three arms, suggesting that in a pragmatic setting, α2-agonists don't provide faster sedation lightening than propofol.

💊The weak age interaction (HR 0.90 per 10-year increment for dexmed) aligns with the SPICE III signal: older patients may tolerate dexmedetomidine better, while younger ICU patients may be at higher risk of harm. This is hypothesis-generating, not practice-changing.

💊Agitation rates were 54–55% higher with both α2-agonists vs propofol. The most likely explanations: less clinician experience with these agents as primary sedatives, and dose-limiting bradycardia preventing adequate uptitration. Benzodiazepine avoidance (per guidelines) removed the usual safety valve.

💊The clonidine PTSD signal at 180 days (IES-R MD −7.97; CI −15.45 to −0.49) is intriguing but is a secondary outcome with incomplete follow-up (~15% response rate at 180 days), treat as hypothesis-generating for future trials.

Pharmacist Implications

Consider adverse events when selecting sedatives for ICU patients.
Dexmedetomidine and clonidine did not reduce extubation time over propofol.
Review sedation protocols to align with patient safety findings.

Guideline Context

Findings align with existing critical care guidelines on sedation efficacy; specifically, the SCCM guidelines on pain, agitation, and delirium management were not contradicted.

Monitoring Guidance

Monitor heart rate due to risk of bradycardia, assess for agitation frequently during sedation.

Nuanced Considerations

No significant differences found in predefined subgroups (age, sepsis status).

Controversies & Gaps

Need further research on sedation strategies minimizing adverse effects while achieving timely extubation.

Cost & Logistics

Not discussed.

Editor's Note

Extended follow-up and additional subgroup analyses could provide further insights.

Full text source: PubMed Central (Open Access)