Haloperidol
Kevin Willmann
A meta-analysis of mostly short-term studies published in the June 2025 issue of PLOS One found no increased mortality with haloperidol use, nor that the drug was arrhythmogenic, despite decades of warnings about it causing QT prolongation (QTP).
“The main message is not that haloperidol is risk-free, but that the cardiovascular story is more nuanced than previously assumed,” said lead author Michael Cristian Garcia, who is affiliated with the Research Institute of St. Joe’s Hamilton in Canada and the Temerty Faculty of Medicine at the University of Toronto. “Clinicians still need to weigh patient context, dose, and duration and to monitor closely.”
Background
Discovered in 1958, haloperidol is a first-generation antipsychotic drug that has been used to treat psychotic disorders such as schizophrenia, delirium, some dementia behaviors, plus nausea and vomiting to a lesser extent. The drug generates around 1 million prescriptions in the United States.
However, it is commonly a source of medication safety alerts because it is listed as a “known risk” QT interval-prolonging medication. As such, FDA recommends electrocardiogram assessments for patients taking haloperidol, in particular for I.V. formulations, due to case reports of QTP, sudden death, and torsades de points (TdP).
Garcia said haloperidol has been categorized as a riskier medication because it “is a high-potency dopamine blocker—effective for symptoms, but that same potency drives parkinsonism, other movement disorders, and potential QT-prolongation issues. Newer agents balance dopamine with serotonin effects, which often softens those trade-offs.”
Study method and results
Garcia and coauthors, all based in Canada, formulated their results from a search for randomized controlled trials involving patients 18 years or older, comparing haloperidol use to placebo. The authors searched Medline, Embase, International Pharmaceutical Abstracts, and Cochrane Central during the week of June 4, 2023, and repeated the search the week of August 2, 2024, using search strategies drafted in collaboration with medical science librarians at the University of Toronto.
“Comparing haloperidol to placebo shows its isolated effect—does the drug actually work, and what risks come directly from the drug rather than the underlying illness? That establishes efficacy and absolute safety signals,” said Garcia.
From the 84 trials included in the review, the study team found non-death events were uncommon, with only two reports of TdP in the haloperidol arm, 22 ventricular arrhythmias, two seizures, and one syncope. While most major adverse cardiovascular events were observed in critical care trials, the authors did not find an increased risk of mortality in that subgroup, which was consistent with previous meta-analyses on critically ill patients and adult in-patients. The results among the subgroup of patients with dementia or delirium also corroborate previous meta-analyses that did not report increased risk of mortality.
The authors’ review also found no statistical difference in the risk of major adverse cardiovascular events between trials that did or did not conduct ECG assessments at screening, baseline, or follow up. The authors noted that the risk of serious cardiac harm appears to be rare.
The team’s findings showed only one major adverse cardiovascular event—a death in the haloperidol arm—observed among 37 trials done in psychiatric populations, and one major adverse cardiovascular event—a death in the placebo—among 14 randomized controlled trials that occurred in surgery or perioperative populations.
The team did note that the mean ages in these groups skewed younger.
Going forward
Garcia said that FDA guidance for haloperidol is specific to behavioral problems in dementia, but the drug is often used for overlapping issues such as delirium, agitation, or even refractory nausea in end-of-life care in those with dementia.
“Our study does not change regulatory labeling, but it adds nuance: For patients without significant cardiac disease, the cardiovascular risk may be less worrisome than previously thought, which can help clinicians weigh its role in these other scenarios,” said Garcia.
The authors recommended further research be done to clarify actual clinical outcomes related to QTP meds to inform safe prescribing practices.
Garcia said the team is currently conducting a large cohort study using hospital data to identify the major drug-induced QT-prolongation offenders and their relationship with major adverse cardiac events.
“This is a massive task, and we are also working on validation studies to outline for future researchers how to approach these pharmacological studies both methodologically and analytically,” he said. ■