Short-term outcomes across sedation practice patterns in mechanically ventilated patients with bacterial pneumonia: a nationwide cohort study

Background Although guidelines recommend light sedation with propofol or dexmedetomidine in mechanically ventilated patients, appropriate management with monotherapy is often challenging. Whether combined sedation reduces dose-dependent adverse effects of each agent or causes additive harm remains unclear. We investigated short-term outcomes associated with early sedation practice patterns in mechanically ventilated patients with bacterial pneumonia. Methods Using the Japanese Diagnosis Procedure Combination database, we identified adult intensive care unit patients with bacterial pneumonia requiring mechanical ventilation between April 2018 and March 2023 and classified them into combined sedation, propofol-based sedation, and dexmedetomidine-based sedation groups. The composite outcome included in-hospital death and functional decline at discharge. Propensity-score matching-weight analyses adjusted for measured confounders. Results Among 3,302 eligible patients, 1,135 (34.4%) were classified into the combined-sedation group, 1,067 (32.3%) into the propofol-based sedation group, and 1,100 (33.3%) into the dexmedetomidine-based sedation group. Combined sedation was not associated with a lower risk of the composite outcome compared with propofol-based sedation (26.3 % vs. 30.0 %; risk difference [RD] −3.7 %; 95% confidence interval [CI] −8.0 to 0.7) but was associated with a lower risk of the composite outcome compared with dexmedetomidine-based sedation (26.3 % vs. 33.1 %; RD − 6.7%; 95% CI − 11.2 to −2.3). In-hospital mortality was lower with combined sedation than propofol-based (22.4% vs. 26.7%; RD, −4.2%; 95% CI, −8.5 to −0.03) or dexmedetomidine-based (22.4% vs. 30.1%; RD, −7.7%; 95% CI, −12.1 to −3.3) sedation, although the comparison with propofol-based sedation was not robust in sensitivity analyses and was not supported by 28-day or 90-day in-hospital mortality. The in-hospital mortality difference between combined sedation and dexmedetomidine-based sedation was consistent across sensitivity analyses and 28-day and 90-day in-hospital mortality. Functional decline at discharge did not differ among groups. Tracheostomy was more frequent with combined sedation than with both comparators, whereas 28-day ventilator-free days did not differ. Conclusions Combined sedation was not associated with a lower risk of the composite outcome of in-hospital death and functional decline at discharge, compared with propofol-based sedation; however, a lower risk was observed compared with dexmedetomidine-based sedation.