(370010-46) Effect of Antipsychotics on Agitation in ICU Delirium: Secondary Analysis of a Randomized Trial

Introduction: Numerous trials have shown antipsychotics neither prevent nor treat ICU delirium, which is often hypoactive. Nevertheless, patients with hyperactive (i.e., agitated) delirium may benefit from antipsychotics. We performed a secondary analysis of the MIND-USA Study to test the hypothesis that, compared with placebo, haloperidol or ziprasidone reduce agitation in the ICU.

Methods: The multicenter MIND-USA Study randomized 566 ICU patients with delirium to haloperidol (max 20mg daily, N=192), ziprasidone (max 40mg daily, N=190), or placebo (N=184). Trial drug/placebo was administered twice daily for up to 14 days and titrated to presence/resolution of delirium. We examined effects of treatment on agitation using highest daily Richmond Agitation-Sedation Scale (RASS), treating RASS < 0 as 0 to isolate effects on agitation. Separately, we examined effects on sedation using lowest daily RASS, treating RASS >0 as 0. In both analyses, we estimated treatment group effect on highest/lowest daily RASS with proportional odds logistic regression, adjusting for pre-randomization covariates: age, mechanical ventilation, sepsis, Acute Physiology and Chronic Health Evaluation (APACHE) II, Charlson Comorbidity Index, Short Informant Questionnaire on Cognitive Decline in the Elderly (IQCODE), hyperactive delirium, and medical vs surgical ICU admission.

Results: The prevalence of agitation (positive RASS) was high, present on 31% of patient-days in the haloperidol group, 28% in the ziprasidone group, and 31% in the placebo group. During the 14-day study period, neither haloperidol nor ziprasidone affected agitation; compared with placebo, odds of increased highest daily RASS was 1.04 (95% CI, 0.90-1.20) for haloperidol and 0.93 (0.80-1.09) for ziprasidone). Alternatively, odds of increased lowest daily RASS was 0.96 (0.86-1.08) for haloperidol and 0.87 (0.77-0.98) for ziprasidone.

Conclusions: In this multicenter, randomized, placebo-controlled trial of delirious ICU patients, we found no evidence that scheduled haloperidol or ziprasidone affects agitation. Though this study was not designed to examine acute, as needed treatment of agitated delirium with antipsychotics, these results do not support use of scheduled antipsychotics to treat or prevent agitation in the ICU.