Senken Neuroleptika die Mortalität bei Beatmungspatienten?

Haloperidol use in mechanically ventilated patients is associated with lower hospital mortality. (Präsentiert am ESICM-Kongress, Berlin 2004)

Milbrandt E B, Kersten A, Kong L, et al.                                                                                                                                 Crit Care Med in Druck

The CRISMA Laboratory, Dept of CCM, Dept of Biostatistics, Univ Pittsburgh, Pittsburgh, United States.

INTRUDUCTION: Haloperidol, the recommended drug for ICU delirium, has been shown to have anti-inflammatory properties that are protective in animal sepsis models. We hypothesized that haloperidol would improve survival of critically ill, mechanically ventilated (MV) patients.
METHODS: We selected all ICU patients with MV>48h admitted July 1, 2000-June 30, 2001 to the University of Pittsburgh Medical Center. We examined differences in hospital mortality between patients who never received haloperidol (No H) and those who received haloperidol within 2 days of initiation of MV (H) by comparing Kaplan-Meier survival curves using the logrank test. We constructed multivariable logistic regression and Cox proportional hazards models to adjust for admission type (medical vs. surgical), ICU type, age, gender, race, SAPS II, SOFA, and underlying comorbidity.
RESULTS: There were 906 and 83 patients in the No H and H groups, respectively. On the day of intubation, there were no significant differences between groups with the exception that H group patients were more often male (67.5% vs. 55.2%, p=.03). H group patients received a mean (±SD) daily dose of 11.5 (±11.6) mg for 3.5 (±4.6) days. The H group had a lower mortality than the No H group (20.5% vs. 36.1%, p=0.004) (Fig 1). After adjusting for potential confounders, the odds of death was still much higher for the No H group (OR 2.26 [1.26-4.05], p=0.006).
CONCLUSIONS: Haloperidol during MV was associated with significantly reduced hospital mortality. These findings have important implications for critically ill patients and require confirmation in a randomized controlled trial.