Delirium in hospitalized patients: Risks and benefits of antipsychotics

Author and Disclosure Information



Atypical antipsychotics and stroke

The FDA has issued multiple warnings for prescribing antipsychotic medications in the elderly. In 2003, it warned prescribers of increased cerebrovascular adverse events, including stroke, in elderly patients with dementia who were treated with an atypical antipsychotic (risperidone, olanzapine, or aripiprazole) vs placebo.15

Atypical antipsychotics and risk of death

In 2005, the FDA issued a black-box warning about increased all-cause mortality risk in patients with dementia treated with atypical antipsychotics for behavioral disturbance (relative risk 1.6–1.7).16

This warning was likely based on a meta-analysis by Schneider et al17 of trials in which patients with dementia were randomized to receive either an atypical antipsychotic or placebo. The death rate was 3.5% in patients treated with an atypical antipsychotic vs 2.3% in patients treated with placebo, indicating a number needed to harm of 100. The most common causes of death were cardiovascular disease and pneumonia. However, the trials in this meta-analysis included only patients who were prescribed atypical antipsychotics for ongoing management of behavioral disturbances due to dementia in either the outpatient or nursing home setting. None of the trials looked at patients who were prescribed atypical antipsychotics for a limited time in a closely monitored inpatient setting.

Effectiveness of antipsychotics

While several studies since the FDA black-box warning have shown that antipsychotics are safe, the efficacy of these drugs in delirium management remains controversial.

In a 2016 meta-analysis, Kishi et al18 found that antipsychotics were superior to placebo in terms of response rate (defined as improvement of delirium severity rating scores), with a number needed to treat of 2.

In contrast, a meta-analysis by Neufeld et al12 found that antipsychotic use was not associated with a change in delirium duration, severity, or length of stay in the hospital or intensive care unit. However, the studies in this meta-analysis varied widely in age range, study design, drug comparison, and treatment strategy (with drugs given as both prophylaxis and treatment). Thus, the results are difficult to interpret.

Kishi et al18 found no difference in the incidence of death, extrapyramidal symptoms, akathisia, or QT prolongation between patients treated with antipsychotic drugs vs placebo.

In a prospective observational study, Hatta et al19 followed 2,453 inpatients who became delirious. Only 22 (0.9%) experienced adverse events attributable to antipsychotic use, the most common being aspiration pneumonia (0.7%), followed by cardiovascular events (0.2%). Notably, no patient died of antipsychotic-related events. In this study, the antipsychotic was stopped as soon as the delirium symptoms resolved, in most cases in 3 to 7 days.

Taken together, these studies indicate that despite the risk of QT prolongation with antipsychotic use and increased rates of morbidity with antipsychotic use in dementia, time-limited management of delirium with antipsychotics is effective9–11 and safe.


Identifying a single preferred agent is difficult, since we lack enough evidence from randomized controlled trials that directly compared the various antipsychotics used in delirium management.

Both typical and atypical antipsychotics are used in clinical practice to manage delirium. The typical antipsychotic most often used is haloperidol, while the most commonly used atypical antipsychotics for delirium include olanzapine, quetiapine, risperidone, and (more recently) aripiprazole.

The American Psychiatric Association guidelines6 suggest using haloperidol because it is the antipsychotic that has been most studied for delirium,20 and we have decades of experience with its use. Despite this, recent prospective studies have suggested that the atypical antipsychotics may be better because they have a faster onset of action and lower incidence of extrapyramidal symptoms.18,21

Because we lack enough head-to-head trials comparing the efficacy of the 5 most commonly used antipsychotics for the management of delirium, and because the prospective trials that do exist show equal efficacy across the antipsychotics studied,22 we suggest considering the unique pharmacologic properties of each drug within the patient’s clinical context when selecting which antipsychotic to use.

Antipsychotic agents

Table 123–25 summarizes some key characteristics of the 5 most commonly used antipsychotics.


Haloperidol, a typical antipsychotic, is a potent antagonist of the dopamine D2 receptor.

Haloperidol has the advantage of having the strongest evidence base for use in delirium. In addition, it is available in oral, intravenous, and intramuscular dosage forms, and it has minimal effects on vital signs, negligible anticholinergic activity, and minimal interactions with other medications.21

Intravenous haloperidol poses a significant risk of QT prolongation and so should be used judiciously in patients with preexisting cardiac conditions or other risk factors for QT prolongation as outlined above, and with careful cardiac monitoring. Parenteral haloperidol is approximately twice as potent as oral haloperidol.

Some evidence suggests a higher risk of acute dystonia and other extrapyramidal symptoms with haloperidol than with the atypical antipsychotics.21,26 In contrast, a 2013 prospective study showed that low doses of haloperidol (< 3.5 mg/day) did not result in a greater frequency of extrapyramidal symptoms.22 Nevertheless, if a patient has a history of extrapyramidal symptoms, haloperidol should likely be avoided in favor of an atypical antipsychotic.

Next Article:

A large mass in the right ventricle: Tumor or thrombus?

Related Articles