Dexmedetomidine vs. propofol for sedation in mechanically ventilated ICU patients: a systematic review and meta-analysis of randomized controlled trials
Background Dexmedetomidine and propofol are the two most widely used non-benzodiazepine sedatives for mechanically ventilated ICU patients. Despite guideline recommendations for both agents, the optimal choice remains controversial owing to inconsistent evidence across patient subpopulations. Methods We conducted a systematic review and meta-analysis of randomized controlled trials (RCTs) directly comparing dexmedetomidine (DEX) with propofol (PROP) in adult ICU patients requiring mechanical ventilation. Databases searched included PubMed, Embase, Cochrane CENTRAL, and CNKI. Primary outcomes were duration of mechanical ventilation (MV) and ICU length of stay (LOS). Secondary outcomes included hospital LOS, all-cause mortality, delirium incidence, bradycardia, and hypotension. Random-effects models were applied using the restricted maximum likelihood (REML) estimator with Hartung–Knapp–Sidik–Jonkman (HKSJ) confidence interval correction. Risk of bias was evaluated using the Cochrane RoB 2.0 tool. Pre-specified subgroup analyses by ICU population type and sedation duration were performed. Results Fifteen RCTs (N = 1,603; DEX: n = 815, PROP: n = 788; published 2001–2021) were included in the primary analysis. Four additional studies were included in sensitivity analyses. Dexmedetomidine did not significantly reduce MV duration (MD = −0.58 h; 95% CI: −2.10 to 0.95; P = 0.419; I2 = 67.1%) or ICU LOS (MD = −0.19 days; 95% CI: −0.67 to 0.29; P = 0.388; I2 = 63.3%). No significant difference in hospital LOS (MD = −0.21 days; P = 0.284; I2 = 0%) or all-cause mortality (OR = 0.92; 95% CI: 0.76–1.12; P = 0.354; I2 = 0%) was observed. Among secondary outcomes, dexmedetomidine was associated with lower delirium incidence (OR = 0.52; 95% CI: 0.33–0.80; P = 0.010; I2 = 27.6%). Bradycardia trended higher with dexmedetomidine (OR = 2.40; 95% CI: 0.94–6.17; P = 0.062) without reaching significance. Subgroup analysis revealed that ICU population type significantly moderated MV duration heterogeneity (interaction P = 0.004). Sedation duration also significantly moderated MV results (interaction P < 0.0001). The delirium reduction was directionally consistent across all population subgroups (interaction P = 0.508). Conclusion In mechanically ventilated adult ICU patients, neither primary outcome (MV duration, ICU LOS) differed between dexmedetomidine and propofol. Dexmedetomidine may reduce ICU delirium incidence, but this secondary finding is borderline after correction for multiple comparisons and requires confirmation in dedicated trials. The higher bradycardia risk warrants vigilant haemodynamic monitoring during dexmedetomidine infusion. Systematic review registration https://www.crd.york.ac.uk/PROSPERO/view/CRD420261378906, identifier: CRD420261378906.