Abstract
Background:
Previous retrospective studies involving mixed intensive care unit (ICU) populations found antipsychotic continuation rates ranging from 21% to 61% at hospital discharge. Surgical ICU admission was cited as a risk factor for antipsychotic continuation, although data in the trauma population are limited.
Objective:
The purpose of this study was to evaluate the number of critically ill trauma patients with escalated psychotropic medications in the surgical ICU and the rate of continuation at ICU transfer and hospital discharge.
Methods:
This single-center, retrospective, Institutional Review Board (IRB)-approved study examined adult trauma patients admitted to the surgical ICU at an urban Level 1 Trauma and Academic Medical Center from December 1, 2021, to May 31, 2023. Patients were excluded if they had a history of select psychiatric disorders or were in the ICU for less than 48 hours. Prior psychotropic medication use was noted, and escalation was defined as a new or increased dose of psychotropic medication. The incidence of delirium and agitation was recorded to assess the indication for psychotropic medication escalation. The primary and secondary outcomes were the percentage of patients with escalated psychotropic medications in the ICU who were continued on therapy at the time of ICU transfer and hospital discharge, respectively.
Results:
Four hundred and twenty-four patients admitted to the surgical ICU were included; 51.4% were escalated on a psychotropic medication while in the ICU. Nearly 35% and 31.8% of the overall population were continued on psychotropic medication at ICU and hospital discharge, respectively. Of patients on psychotropic medication at hospital discharge, 55.6% were discharged to acute rehabilitation, 28.9% to home, and 13.3% to long-term care.
Conclusion and relevance:
Escalation of psychotropic medications was common in trauma patients admitted to the surgical ICU. Further investigation into the appropriateness of psychotropic medication prescription during care transitions is needed.
Background
Antipsychotic medications are frequently started in the intensive care unit (ICU) to treat symptoms of acute delirium and agitation and are appealing due to their sedative effects that do not hinder respiratory drive. However, the cited efficacy of antipsychotics for the treatment of delirium is inconsistent, and the 2025 Pain, Agitation/Sedation, Delirium, Immobility, and Sleep Disruption in Adult Patients in the ICU Guidelines (PADIS Guidelines) state that a recommendation neither for nor against the use of antipsychotics can be made over usual care for the treatment of delirium in adult patients admitted to the ICU. Nonetheless, short-term use may help manage delirium manifestations that put the patient or caregiver at risk, but antipsychotic discontinuation immediately after resolution of the patient’s distressful symptoms is recommended. 1
A potential consequence of antipsychotic initiation in the ICU is sustained use during transitions of care to less acute settings, especially concerning at the time of hospital discharge. Continued antipsychotic use is associated with risks including cardiac effects (prolonged QT interval and increased likelihood of ventricular arrhythmias), neuroleptic malignant syndrome, extrapyramidal symptoms, and sudden cardiac death.2-4 The incidence of antipsychotic continuation at hospital discharge ranged from 21% to 61% in previous retrospective studies but focused primarily on mixed or medical ICU patients.2,5-7 One analysis found admission to the surgical ICU increased the risk for antipsychotic continuation, but no previous evaluation has focused on the trauma population 2 despite delirium complicating over 20% of critically ill trauma patients. 8
The purpose of this study was to evaluate the number of critically ill trauma patients with escalated psychotropic medications in the surgical ICU and the rate of continuation at ICU transfer and hospital discharge.
Methods
This single-center, retrospective study was conducted at an urban Level 1 Trauma and Academic Medical Center. The institution is a safety-net hospital and serves a diverse and socially complex patient population. Adult trauma patients admitted to the surgical ICU between December 1, 2021, and May 31, 2023, were included. Patients were excluded if they were in the ICU for less than 48 hours, died during the hospitalization, or had a select psychiatric disorder (schizophrenia, schizoaffective disorder, or bipolar disorder) that may warrant long-term use of psychotropic medications.
Psychotropic escalation was defined as a new initiation or increased dose of a home psychotropic medication. Psychotropic medication use included administration of haloperidol, olanzapine, droperidol, quetiapine, valproic acid, ziprasidone, or risperidone during the hospitalization. These agents were selected for analysis based on inclusion on the hospital formulary and practice patterns within the surgical ICU. Continuation of prior-to-admission psychotropic medications at the same dose or orders for as needed medications that were not administered were not included in the evaluation. Initiation of psychotropic medication was left to the discretion of the provider; no electronic decision support is used for treatment of ICU delirium or agitation.
Patient demographics were collected and included risk factors for psychotropic medication use including history of substance use disorder, injury severity score (ISS), and diagnosis of agitation or delirium. Self-identified race was collected from the medical record to detect racial disparities in prescribing patterns. Toxicology tests screened for the following substances: acetaminophen, amphetamines, barbiturates, benzodiazepines, buprenorphine, cocaine, ethanol, fentanyl, lysergic acid diethylamide (LSD), methadone, opiates, phencyclidine (PCP), salicylates, and tetrahydrocannabinol (THC).
Delirium was defined as an acute onset of behaviors characterized by restlessness, illusions, and incoherence of thought and speech. Patients were classified as having delirium based on positive Confusion Assessment Method for the ICU (CAM-ICU) screening or if there was a diagnosis of delirium symptom onset after hospital arrival documented in the patient’s medical record. Patients were excluded from a delirium diagnosis if symptoms were due to alcohol withdrawal. Agitation was defined as a Richmond Agitation-Sedation Scale (RASS) score greater than 2 at least twice in 24 hours. Additional variables assessed included psychotropic medication regimen, duration of ICU and hospital stay, and hospital disposition.
The primary outcome was the percentage of patients escalated on psychotropic medication in the ICU who were continued on therapy after transferring to a medical ward. The secondary outcome was the percentage of patients escalated on a psychotropic medication in the ICU who were continued on therapy after hospital discharge. Subgroup analyses included incidence of continuation at ICU and hospital discharge for patients who received psychiatry consultation, were greater than 65 years of age, or were admitted with head trauma that included any of the following injuries: stroke, traumatic brain injury, or intracranial hemorrhage (including traumatic, spontaneous, and aneurysmal hemorrhages) secondary to an insult.
Categorical data were analyzed using Chi-Square or Fisher’s Exact tests, as appropriate. All continuous data were analyzed using Mann-Whitney U and reported as median and interquartile range (IQR). Separate backward logistic regression analyses were performed to identify independent predictors of antipsychotic (AP) continuation at ICU and hospital discharge in patients with escalated psychotropic regimens in the surgical ICU. Variables that demonstrated a P-value of <0.20 in univariate analyses were placed into the model. Hosmer-Lemeshow scores of >0.05 were considered to signal good model fit.
P-values less than 0.05 were considered statistically significant. Statistical analyses were completed using IBM SPSS Statistics, Version 30. Armonk, New York: IBM Corporation.
This study was approved by the Hennepin Healthcare Human Research Institutional Review Board (IRB) with the reference number of IRB—FY2023-687.
Results
Between December 1, 2021, and May 31, 2023, 424 trauma patients who were admitted to the surgical ICU were included. Overall, patients were predominantly white males in their early 50s with 17.9% experiencing delirium and 32.7% experiencing agitation.
Quetiapine was the most prescribed agent (63.9%), followed by olanzapine (50.5%), haloperidol (33.3%), valproic acid (25.9%), risperidone (2.7%), and droperidol (1.2%). No patients were prescribed ziprasidone.
Of the 424 patients in the overall study population, 218 (51.4%) had psychotropic escalation while in the ICU. Of these 218 patients, 148 (67.9%) or 34.9% of the overall study population had their psychotropic regimen continued at the time of ICU discharge. These patients were more likely to have an alcohol screen performed, experience a longer hospital length of stay as well as be identified as having agitation, delirium, or a positive toxicology screen (Tables 1 and 2).
Baseline Characteristics—AP Escalated in ICU and Continued vs Not at ICU Discharge.
IQR = interquartile range, SUD = substance use disorder.
Hospitalization Characteristics—AP Escalated in ICU and Continued vs Not at ICU Discharge.
Note. Bolded values are statistically significant. IQR = interquartile range.
One hundred thirty-five (61.9%) of the 218 patients, or 31.8% of the overall study population, with ICU psychotropic escalation had their regimen continued at time of hospital discharge. This population was more likely to demonstrate longer ICU and hospital durations. They also had a higher incidence of positive toxicology screens, psychiatry consultation, and delirium or agitation diagnoses. Their ISSs were higher than patients without AP continuation, and they were more likely to be discharged to acute rehabilitation (Tables 3 and 4). There were no differences detected in prescribing patterns based on patient race.
Baseline Characteristics—AP Escalated in ICU and Continued vs Not at Hospital Discharge.
Note. Bolded values are statistically significant. IQR = interquartile range.
Hospitalization Characteristics—AP Escalated in ICU and Continued vs Not at Hospital Discharge.
Note. Bolded values are statistically significant. IQR = interquartile range.
The findings of the subgroup analyses may be found in Tables 5 to 7. Psychiatry consults and age greater than 65 were both statistically significant factors in psychotropic escalation in the ICU. There was a statistically significant difference between groups with regard to psychiatry consults as it related to psychotropic continuation both at ICU and hospital discharge. Neurological injury was statistically significantly higher in those patients with psychotropic continuation at hospital discharge but not at ICU discharge.
Subgroup Analysis—Psychiatry Consult.
Note. Bolded values are statistically significant.
Subgroup Analysis—Age Greater Than 65 Years.
Note. Bolded values are statistically significant.
Subgroup Analysis—Neurological Injury.
Note. Bolded values are statistically significant.
The backwards logistic regression analysis aimed at identifying independent predictors of psychotropic continuation at ICU discharge in those with psychotropic escalation in the ICU was deemed a poorly fit model based on the Hosmer-Lemeshow value of 0.034, and thus the findings were not reported.
A second backwards logistic regression analysis aimed at identifying independent predictors of psychotropic continuation at hospital discharge in those with psychotropic escalation in the ICU was deemed a model with a good fit based on the Hosmer-Lemeshow value of 0.422. Neurological injury, delirium, and discharge to home were deemed independent predictors, as was age, each additional year conferring an increased risk of 2.8% (see Table 8).
Backwards Logistic Regression Analysis—Independent Predictors of AP Prescription at Time of Hospital Discharge in Patients with AP Escalation in the ICU.
Note. Bolded values are statistically significant.
Discussion
This study found the continuation of psychotropic medications for the management of delirium and agitation following ICU and hospital discharge occurred in approximately one third of the critically ill trauma population started on psychotropic medications. Without sufficient evidence that patients receive more than symptomatic relief related to their acute delirium and agitation, the continuation of these medications is potentially inappropriate during transitions of care and may contribute to chronic polypharmacy, prescription inertia, unnecessary costs, and the potential for serious adverse effects.
Perhaps the most compelling reason to limit unnecessary psychotropic use is their potential to cause serious cardiac adverse effects, specifically drug-induced ventricular arrhythmia related to QT interval prolongation and an increased risk of sudden cardiac death. Wu and colleagues published a retrospective analysis demonstrating antipsychotic exposure for as little as 7 days can lead to an increase in arrhythmias and is more common with first-generation than with second-generation antipsychotics. 3 Similarly, a study by Andersen-Ranberg et al 9 compared haloperidol to placebo for ICU delirium and found no difference in days alive and out of the hospital at 90 days after randomization but noted that more patients in the haloperidol group had to discontinue therapy for a clinically relevant QT interval prolongation.
Psychotropic utilization is particularly concerning in the elderly. In addition to the drug class’s inclusion in the Beer’s criteria, 10 the American College of Surgeons (ACS) Best Practice Guidelines for Geriatric Trauma Patients recommend thorough medication reconciliation at each phase of care as well as the avoidance of antipsychotics due to the risk for falls in the elderly patient population. 11 The need for systematic psychotropic de-prescribing was highlighted in a retrospective cohort of 260 elderly patients released from the hospital with newly prescribed antipsychotics during the hospital stay that were continued on discharge. The study found that at the time of first readmission, 65% of patients were still taking the same antipsychotic on which they were discharged, reinforcing the real-world risk of prolonged unnecessary psychotropic therapy if not addressed at time of care transitions. 4 Our study included 116 patients greater than 65 years of age. Of those patients, 43.1% were escalated on psychotropic medications and 52% discharged on psychotropic medications, putting them at risk for long-term consequences (see Table 6).
Within our health care system, trauma patients are cared for by a wide variety of providers offering inpatient services such as daily interdisciplinary rounds including a clinical pharmacist, pharmacist-led discharge medication reconciliation, and trauma psychology consultation. Even with these safeguards in place, the findings of this study are consistent with previously described observations.2,5,7 Several factors may play a role in inappropriate psychotropic medication continuation including patient acuity, drug selection, clinical knowledge gaps, insufficient oversight, and inadequate communication during care transitions. Transitions of care programs and communication tool improvement may be needed to better address these issues. 5 The implementation of an antipsychotic discontinuation bundle based on CAM-ICU scores in a medical ICU population resulted in a decrease in antipsychotic continuation at ICU discharge from 27.9% to 17.7%. 6 An alternative strategy utilized a collaborative practice agreement between physicians and pharmacists, allowing the ICU and Internal Medicine pharmacists prescriptive authority to discontinue or taper antipsychotics at the time of delirium resolution and found a similar reduction in antipsychotic continuation. 12 Prompted by the findings of the present evaluation, we have developed a multidisciplinary committee to examine our organization’s management of delirium and agitation. We are currently in the process of developing an institutional practice guideline for the neurotrauma population, including both ICU and non-ICU patients.
This study has several strengths. First, to our knowledge, this is the first retrospective observation highlighting the trauma population in an urban Level 1 Trauma and Academic Medical Center. Second, despite the common development of acute delirium in the critically ill trauma population, the majority of ICU literature on the topic exists within mixed or medical ICU settings.6-8 Our study provides insight into an underreported group. Furthermore, the study included a unique subgroup of 134 patients admitted with neurological injury, a population at additional risk for agitation and delirium. We found this population is statistically significantly more likely to experience psychotropic continuation at hospital discharge than the overall general trauma population, highlighting the difficult management of agitation and delirium in these complex patients. Including these patients offers unique insight into highly complex clinical courses that often warrant an increase in psychotropic medications. However, 56.9% of patients without neurological injury were discharged with a psychotropic, demonstrating that this problem is not isolated to patients with neurological injuries.
This study is not without limitations. First, it was a retrospective, single-center observational evaluation completed at a tertiary care center and may not be generalizable to other settings. Second, select evaluations were beyond the scope of this study: psychotropic medication indication (delirium, agitation, sleep, or psychiatric disorder), differentiation between delirium types, and correlation between RASS and CAM-ICU scores with type of delirium. In addition, neither CAM-ICU scores nor time to agitation or delirium onset in relation to time to psychotropic initiation or escalation were captured; this may have shed light on the agent’s indication and on the subsequent appropriateness of continuation at care transitions. Third, patients who received psychiatry consults were statistically significantly more likely to have psychotropic initiation or escalation in the ICU as well as continuation at unit transfer and hospital discharge. However, this may have been for an appropriate indication unrelated to agitation or delirium. Next, our institution frequently utilizes valproic acid for delirium and agitation, with approximately 15% of the population prescribed valproic acid escalation. This may partially account for the overall incidence of escalation and continuation being amongst the highest of those previously described and offers insight into the unique practice of valproic acid prescription for ICU delirium and agitation.13,14 Finally, positive toxicology screens may have been due to appropriate medication administration by pre-hospital or emergency department teams.
Conclusion and Relevance
This study demonstrated the frequent occurrence of psychotropic escalation in critically ill trauma patients in the surgical ICU (51.4%); these agents were commonly continued from the ICU (34.9%) and the hospital (31.8%). These findings highlight the importance of appropriate psychotropic de-prescribing during transitions of care for critically ill trauma patients to decrease the risk of potential adverse effects.
Footnotes
Authors’ Note
Presented at the International Symposium on Intensive Care & Emergency Medicine, March 2024.
Ethical Considerations
This study was approved by the Hennepin Healthcare Research Institute.
Consent to Participate
The requirement for informed consent was waived, as it was deemed impracticable due to the retrospective nature of the study.
Consent for Publication
Not applicable.
Author Contributions
JS, JJ, and RN contributed to conception and study design.
JS, JJ, and HR contributed to the literature review.
JS and JJ contributed to data acquisition.
JS, JJ, RN, and HR contributed to data analysis and interpretation.
JS and HR contributed to the drafting of the manuscript.
JS, JJ, RN, and HR contributed to critical revision.
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Data Availability Statement
The data set is not available due to institutional preferences.