The Consequences of Unwarranted Pharmacologic Treatment of ICU Delirium

Christina Boncyk, MD, MPH
Assistant Professor of Anesthesiology, Vanderbilt University Medical Center, Nashville, TN

The generally positive results of older controlled studies evaluating antipsychotics for intensive care unit (ICU) delirium treatment helped potentiate the routine use of these agents in critically ill adults despite the serious limitations of these studies.1-3 While more recent publications of far larger and more rigorous randomized trials have demonstrated no benefit with antipsychotic therapy for ICU delirium treatment, 4,5 and ICU practice guidelines recommend the use of non-pharmacologic delirium reduction strategies (and advocating against the use of medication interventions),6 pharmacologic interventions have remained a mainstay of delirium treatment in the ICU.

We sought to describe the pharmacologic prescribing practices for ICU delirium management within our large academic institution that included four diverse ICU populations (medical, surgical, trauma, and cardiovascular).7 We limited our cohort to patients diagnosed with delirium in the ICU (using CAM-ICU assessments) who were not receiving home antipsychotic medications prior to admission to more accurately capture patients initiated on new antipsychotic medications for delirium management. Additionally, we investigated whether the use of these medications was independently associated with delirium continuation, ICU-free days, hospital-free days, and in-hospital mortality.

We chose to specifically focus our investigation on the ICU use of antipsychotics, oral guanfacine, and oral valproic acid from 2013 to 2016 given their perceived frequent use as a delirium treatment. We restricted our antipsychotic analysis to haloperidol, olanzapine, and quetiapine, given these three agents accounted for 98% of all prescribed antipsychotics. We found antipsychotic medications were used by nearly half (45.4%) of patients with ICU delirium; 30.6% received combinations of 2 or more different antipsychotic medications during their ICU admission. Guanfacine and valproic acid were utilized in only 2.2% and 0.84% of patients, respectively.

Antipsychotic medications were initiated early in the course of delirium (median delirium day 2 [1-4]) and continued for almost twice as long as the actual delirium occurrence duration (median prescribing duration 4 [2-10] days vs median delirium duration 2.1 [0.8-5.4] days). Consistent with other reports, we found that more than 20% of patients were continued on a newly-initiated delirium treatment medication(s) at hospital discharge. In the vulnerable post-ICU patient, antipsychotic continuation may be associated with significant downstream safety concerns.

To investigate the association between ICU delirium medication use and in-hospital patient outcomes, we constructed several statistical models. The first included a daily multinomial logistic regression model that accounted for baseline factors like age and daily ICU factors including coma, mechanical ventilation, IV sedative exposure (dexmedetomidine and propofol), and benzodiazepine exposure, to determine the odds of treatment medication use on one ICU day being associated with delirium occurrence the next. For both haloperidol and olanzapine, use of either on one ICU day was associated with an increased odds of delirium the following day (haloperidol: OR 1.48; 95% CI [1.30-1.65]; P<0.001 and olanzapine: OR 1.37; 95% CI [1.20-1.56]; P=0.003, respectively) (Figure 1). A significant association was not found for the other three medications investigated.

Figure 1
Boncyk Figure 1

We then used Cox regression models to investigate the association of use of each medication on one ICU day and in-hospital mortality the next. Haloperidol and olanzapine use on one ICU day were each associated with increased mortality the next ICU day (haloperidol: HR 1.46; 95% CI [1.10-1.93] and olanzapine; HR 1.67; 95% CI [1.14-2.45], respectively). Interestingly, quetiapine was found to be associated with significantly decreased next-day hazard of mortality (HR 0.58; 95% CI [0.40-0.84]). However, the relationship between quetiapine use for delirium and mortality can only be determined in future randomized controlled trials. Guanfacine and valproic acid use were not significantly associated with in-hospital mortality.

Finally, we investigated whether use of these medications impacted ICU- or hospital-free days over the first 30-days following ICU admission. ICU-free days were those days patients spent alive and outside the ICU; hospital-free days were days patients spent alive and outside of the ICU and the hospital. We chose ICU- and hospital-free days over length of stay outcomes to avoid the potential bias when patients who die while still in the ICU are included. None of the medications investigated showed any significant association with ICU-free days. Each antipsychotic medication was associated with fewer hospital-free days (haloperidol: difference in mean, −1.00 d; 95% CI [−1.51 to −0.49]); olanzapine: difference in mean, −1.55 d; 95% CI [−2.26 to −0.84]), and quetiapine: difference in mean, −1.22 d; 95% CI [−1.79 to −0.65]). Guanfacine and valproic acid demonstrated no association. (Figure 2).

Figure 2

Together our results demonstrate how ubiquitously antipsychotic medications (haloperidol, olanzapine, and quetiapine) are used in patients with ICU delirium and echo the results of other cohort studies demonstrating a high prevalence of inappropriate continuation beyond hospital discharge.8-10 Medication reconciliation at all transfer of care points is important for all patients admitted to the ICU. Our results also help shed light on the potential negative sequelae of using pharmacologic therapies to treat ICU delirium in the absence of large, randomized trial(s) demonstrating their benefit. As outlined in practice guidelines6 and the ABCDEF care bundle,11 ICU clinicians should focus on non-pharmacologic strategies to resolve delirium in their patients.


  1. Tahir TA, Eeles E, Karapareddy V, et al. A randomized controlled trial of quetiapine versus placebo in the treatment of delirium. J Psychosom Res. 2010;69(5):485-490.
  2. Skrobik YK, Bergeron N, Dumont M, Gottfried SB. Olanzapine vs haloperidol: treating delirium in a critical care setting. Intensive Care Med. 2004;30(3):444-449.
  3. Devlin JW, Roberts RJ, Fong JJ, et al. Efficacy and safety of quetiapine in critically ill patients with delirium: A prospective, multicenter, randomized, double-blind, placebo-controlled pilot study. Critical Care Medicine. 2010;38(2):419-427.
  4. Girard TD, Exline MC, Carson SS, et al. Haloperidol and Ziprasidone for Treatment of Delirium in Critical Illness. The New England journal of medicine. 2018.
  5. Burry LM, S.; Williamson, D. R.; Hutton, B.; Ely, E. W.; Adhikari, N. K. J.; Egerod, I.; Fergusson, D. A.; Rose, L. Pharmacological interventions for the treatment of delirium in critically ill patients. Cochrane Database of Systematic Reviews. 2015;2015 (6) (no pagination)(CD011749).
  6. Devlin JW, Skrobik Y, Gelinas C, et al. Clinical Practice Guidelines for the Prevention and Management of Pain, Agitation/Sedation, Delirium, Immobility, and Sleep Disruption in Adult Patients in the ICU. Crit Care Med. 2018;46(9):e825-e873.
  7. Boncyk CS, Farrin E, Stollings JL, et al. Pharmacologic Management of Intensive Care Unit Delirium: Clinical Prescribing Practices and Outcomes in More Than 8500 Patient Encounters. Anesth Analg. 2021.
  8. Flurie RWG, J. P.; Tata, A. L.; Millstein, L. S.; Gulati, M. Hospital delirium treatment: Continuation of antipsychotic therapy from the intensive care unit to discharge. American Journal of Health-System Pharmacy. 2015;72(23):S133-S139.
  9. Kram BLK, S. J.; Brooks, K. R. Implications of atypical antipsychotic prescribing in the intensive care unit. Journal of Critical Care. 2015;30(4):814-818.
  10. Marshall J, Herzig SJ, Howell MD, et al. Antipsychotic utilization in the intensive care unit and in transitions of care. J Crit Care. 2016;33:119-124.
  11. Pun BT, Balas MC, Barnes-Daly MA, et al. Caring for Critically Ill Patients with the ABCDEF Bundle: Results of the ICU Liberation Collaborative in Over 15,000 Adults. Crit Care Med. 2018.
Posted in Delirium Research, News, NIDUS Blog and tagged .

Leave a Reply