Haloperidol for the treatment of delirium in ICU patients: a systematic review and meta‑analysis

Abstract

Objectives

Haloperidol is the most frequently prescribed medication for managing delirium in the intensive care unit (ICU). However, there is limited and inconclusive evidence regarding its efficacy. A meta-analysis was conducted by pooling data from recent clinical randomized controlled trials to assess the effectiveness of haloperidol in adult ICU patients with delirium.

Methods

Studies were searched in PubMed, Embase and Cochrane Library databases on August 10, 2024. We performed a meta-analysis to estimate the efficacy of haloperidol for the treatment of ICU adult patients with delirium. This study is registered with INPLASY, number 202480104. The estimates are expressed as odds ratio (OR) or mean difference (MD) with a 95% confidence interval (CI).

Results

A total of 2863 patients were included in the analyses. All the included studies were randomized controlled trials. The frequency of patients diagnosed with delirium used both confusion assessment method of intensive care unit (CAM-ICU) and intensive care delirium screening checklist (ICDSC) was 34% (n = 2863), and used CAM-ICU only was 66% (n = 2863). There was no difference in short-term (28–30 days) mortality between the two groups [OR = 0.89, 95% CI 0.60–1.32, P = 0.56] and long-term (90 days to 1 year) mortality [OR = 0.87, 95% CI 0.70–1.07, P = 0.19]. Furthermore, the haloperidol group demonstrated an advantage in reducing the length of ICU stay [MD = -1.13, 95% CI − 1.93–− 0.32, P < 0.05] compared to the placebo group, with no statistically significant difference in length of hospital stay [MD = − 0.24, 95% CI -1.71–1.24, P = 0.75].

Conclusions

Haloperidol showed a significant trend in reducing the length of ICU stay. However, there was no statistical difference between the two groups in terms of delirium reduction.

Introduction

Delirium is a prevalent manifestation of acute brain dysfunction in critically ill patients, representing a clinical syndrome resulting from various underlying causes rather than a singular disease entity. It encompasses cognitive impairment and is not limited to a specific etiology [1]. ICU delirium is a prevalent medical issue among patients in the intensive care unit (ICU) [2]. Patients with delirium are at an elevated risk for adverse outcomes, including increased case fatality and hospitalization costs, as well as the potential for long-term cognitive dysfunction, imposing a substantial burden on both patients and their families. Numerous factors can contribute to the development of delirium, including non-modifiable predisposing factors, triggers, and medication-related factors. The incidence of delirium is high among hospitalized patients in general hospitals. However, there is a lack of unified and standardized criteria for diagnosis and treatment.

Delirium is characterized by disturbances in attention and consciousness, which are central to its field of cognitive change. In addition, it includes fundamental symptoms in two other domains: alterations in the sleep–wake cycle and difficulties in comprehension (thought process and language). A comprehensive understanding of these clinical characteristics is crucial for effective diagnosis and management. Recent studies also provided valuable insights into the clinical characterization and phenotype of delirium [3, 4]. Based on the recent research findings, two criteria diagnosis of delirium that should be included in: a specified 24-h diagnostic period and a severity threshold. Changes in activity level or circadian rhythm patterns should be re-evaluated for inclusion as core features in future diagnostic systems [3].

The Pain, Agitation/sedation, Delirium, Immobility and Sleep disruption guidelines (PADIS) do not advocate the routine use of any pharmacological agents for the prevention or treatment of delirium [5]. The evidence regarding the efficacy of haloperidol as a commonly prescribed medication for managing delirium in ICU patients has been limited and conflicting [6]. Despite previous studies evaluating the effectiveness of haloperidol, there is still a lack of existing evidence, and the results are not statistically significant. More trial data are needed to provide higher certainty evidence. In recent years, there are six clinical studies have compared the effects of haloperidol and placebo on mortality and length of stay in patients with ICU delirium [7,8,9,10,11,12]. These studies are all randomized clinical trials (RCTs). The strength of this study lies in its systematic updating of existing systematic reviews to incorporate additional randomized clinical trials and conduct a comprehensive assessment of the effects of haloperidol versus placebo.

Methods

This meta-analysis was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [13]. The pre-registration for our meta-analysis can be accessed via the International Platform Database for Registration Systems Evaluation and Meta-Analysis Protocols (reference number: INPLASY202480104). The full text is available for download from inplasy.com ( https://inplasy.com/inplasy-2024-8-0104 ). It should be noted that ethical approval is not required for this study.

Search strategy

A comprehensive electronic search of articles published in the field was conducted by three researchers (Yue Zhao, Qing Wang, Biao Sun) before 10 August 2024. A comprehensive manual search of the PubMed, Embase and Cochrane databases was conducted in order to select relevant randomized controlled trials. For further details regarding the precise literature search strategies employed, please consult the Appendix (Supplementary File).

Inclusion and exclusion

Document management uses EndNote (X9) software, and two investigators independently evaluate the project qualifications to verify the project’s eligibility. First, the title and abstract are screened, and then the relevant articles are comprehensively reviewed, and the eligible articles are reserved for comprehensive review. The studies included in this review were all randomized controlled trials. The following inclusion criteria were met: (1) ICU adult patients with delirious. (2) Treatment with haloperidol or placebo or conventional therapy. (3) Outcome indicators: death from any cause/length of ICU stays/length of hospital stays. We excluded animal trials, studies that included patients < 18 years of age, and there was not enough data to extract, such as summaries of some meetings, literature such as reviews, and pharmacological presentations, in addition, literature materials such as review and meeting reports. Unrelated to the study topic, inappropriate intervention and control will also be excluded. We contacted the authors if associated data from their studies were required.

Bias and quality assessment

The two researchers conducted independent evaluations, preliminary selections and verifications of the literature in accordance with a unified and standardized methodology. The literature was included or excluded in accordance with the pre-established criteria, and the data were subsequently collected. The quality of the selected articles was evaluated in accordance with the Cochrane Reviewer Handbook 5.1.0 [14].

Data synthesis and analysis

The meta-analysis was conducted utilizing RevMan 5.4 software. Data that met the pre-established criteria of homogeneity (P > 0.10 and I² ≤ 50%) as determined by the heterogeneity test were analyzed using the fixed effects model (M-H). In the event that the aforementioned homogeneity criteria (P ≤ 0.10 or I² > 50%) are not met and heterogeneity cannot be excluded, a random effects model is employed to consolidate the effect [15]. The estimates are expressed as odds ratio (OR) or mean difference (MD) with a 95% confidence interval (CI). A p value of less than 0.05 was deemed to be statistically significant.

Results

The flow chart (Fig. 1) provides a summary of the search and research selection process. A total of 1333 related literatures were retrieved, of which 412 were excluded due to duplication. 868 studies were also excluded after reading the titles and abstracts. A comprehensive evaluation was conducted on the remaining 53 studies, based on a full reading of the full text. Data from 6 trails evaluating the efficacy of haloperidol for the treatment of delirium in critically ill patients were included.

Fig. 1

Selection process for studies included in the meta-analysis

The principal characteristics of the included trials are set forth in Table 1. A total of 2863 patients were included in the analyses. All the included studies were randomized controlled trials. The estimates are expressed as odds ratios (OR) or mean differences (MD) with a 95% confidence interval (CI). All studies conducted a comparative analysis of the effectiveness of haloperidol for treating delirium in adult ICU patients. The mortality outcomes are summarized in Fig. 2. There was no difference in short-term (28–30 days) mortality (A) between the two groups [OR = 0.89, 95% CI 0.60–1.32, P = 0.56] and long-term (90 days to 1 year) mortality (Fig. 2B) [OR = 0.87, 95% CI 0.70–1.07, P = 0.19]. Furthermore, the haloperidol group demonstrated an advantage in reducing the length of ICU stay (Fig. 3A) [MD = − 1.13, 95% CI − 1.93–− 0.32, P = 0.006] compared to the placebo group, with no statistically significant difference in length of hospital stay (Fig. 3B) [MD = − 0.24, 95% CI − 1.71–1.24, P = 0.75].

Table 1 Design and outcomes of the studies included in the meta-analysis

Fig. 2

Comparisons of the short-term (28–30 days) mortality and long-term (90 days–1 year) mortality between haloperidol and placebo

Fig. 3

Comparisons of the length of ICU stay and the length of hospital stay between haloperidol and placebo

This study used RevMan software to assess the influence of an individual study on the overall combined estimate for each predefined outcome. The cost-effective risk evaluation findings from these experiments are presented in Supplement File. The randomized controlled trials (RCTs) included in this study were of high quality and deemed to have a low risk of bias.

Discussion

At present, the pathophysiological mechanism of delirium is not well understood, and it is very important to identify and treat the patients with high risk of delirium and delirium [1]. Delirium may be related to the interaction of multiple factors that cause dysfunction of the brain’s neural networks [16]. In addition to its mild antihistamine and antiserotonin effects, haloperidol has potent antiadrenergic properties, as well as some peripheral anticholinergic activity. Therefore, haloperidol can counteract a number of pathways that lead to delirium [17]. Although haloperidol is widely used to treat delirium in the ICU, its efficacy remains uncertain, and many trials have reached inconsistent conclusions about its effectiveness in preventing delirium in ICU patients [18,19,20,21,22].

In this systematic review comparing the efficacy of haloperidol versus placebo in the treatment of critical adult delirium, we analyzed data from six randomized controlled trials published between 2018 and 2024, involving a total of 2,863 patients. Our findings suggest that haloperidol has no tendency to reduce long-term mortality and short-term mortality but significantly shorten ICU stay compared to placebo. A systematic review conducted by Wu published in 2019 concluded that the use of haloperidol did not lead to a reduction in all-cause mortality among ICU patients with delirium [23]. A RCT study by Van Den Boogard published in 2018, similarly confirmed that prophylactic haloperidol use did not improve 28-day survival compared with placebo in critically ill adults at high risk of delirium, and this finding do not support the use of prophylactic haloperidol to reduce mortality in critically ill adults [24]. In our study, for the mortality analysis, we used 28–30 days mortality and 90 days to 1-year mortality. The findings suggest that haloperidol has no tendency to improve long-term mortality (90 days to 1 year) in ICU patients compared to placebo, also has no effect on short-term mortality (28–30 days) in ICU patients compared with placebo. Due to the limited number of RCTs meeting the inclusion criteria, this study did not conduct a meta-analysis comparing the mortality of ICU patients receiving different doses of haloperidol. In a post hoc analysis involving a total of 1,459 delirium-free patients admitted to the ICU, it was confirmed that delirious patients who received haloperidol treatment experienced prolonged stays in the ICU [25]. In our research, we observed that the use of haloperidol led to a significant improvement in ICU stay duration compared to placebo, although it did not affect the overall length of hospital stay for patients. We hypothesized that variations in trial outcomes may be attributed to the severity of ICU-admitted patients, risk factors for delirium, and varying haloperidol dosages.

Many researchers and clinicians rely on screening instruments for the diagnosis of delirium, such as the confusion assessment method of intensive care unit (CAM-ICU), rather than applying the diagnostic criteria of the diagnostic and statistical manual of mental disorders (DSM) or intensive care delirium screening checklist (ICDSC) for assessment. Although CAM-ICU are more convenient compared to DSM, they have been criticized for lacking rigor, which can lead to false-positive diagnoses of delirium [26]. In the RCTs we included, only one RCT research used both the CAM-ICU screening instrument and the ICDSC diagnostic criteria for identifying delirium patients. This approach may result in the inclusion of patients with severe arousal disturbances or multiple neurocognitive changes and neuropsychiatric disorders in clinical trials focused on delirium treatment. This could introduce negative bias in such studies, where some patients may be deemed unresponsive to delirium treatment, even though they do not actually have delirium but rather arousal, neurocognitive, or neuropsychiatric disorders. For these conditions, antipsychotic medications should not be used and/or would be ineffective. Therefore, future clinical trials should employ more appropriate reference standards for diagnosing delirium, such as the diagnostic criteria of DSM or ICD, to avoid the generation of negative bias results.

Limitations of this study need to be noted. First, in this study, different haloperidol application strategies, different severity of patients, and different evaluation systems for delirium may increase the heterogeneity of outcome measures. In addition, although the exclusion criteria of different studies have made certain introductions about the cognitive level of patients, there are differences in the exclusion criteria of different studies, which may also lead to unstable outcomes. Second, due to the limited sample size of the study, the small number of included studies may also affect the accuracy of the results. In summary, multi-center, large-scale randomized controlled trials are still the future research direction. The trial focused only on haloperidol, and the results cannot be extrapolated to other antipsychotics that have been studied. Of course, the exploration of combined use with other drugs, including the combination of necessary care means, may also be the direction of future research.

Conclusions

Haloperidol showed a significant trend in reducing the length of ICU stay, and there was no significant difference in mortality. Future randomized controlled trials with large samples are needed for further validation.