REVIEW OF: Andersen-Ranberg NC et al, N Engl J Med 2022;387(26):2425–2435

STUDY TYPE: Randomized double-blind placebo-controlled trial

Antipsychotics are often used for agitation in delirium, but recent studies have called their efficacy into question and some practice guidelines do not recommend them (Devlin JW et al, Crit Care Med 2018;46(9):e825–e873). This study looked at whether haloperidol improved meaningful long-term outcomes. 

This four-year, double-blind study randomized 1,000 ICU patients in five European countries to haloperidol or placebo. After diagnosing delirium through a validated measure, researchers administered 2.5 mg of IV haloperidol or placebo three times daily. Additional PRN doses, up to a maximum daily dose of 20 mg, were given for recurring symptoms. IV placebo was dosed at a similar rate, and doses were stopped when delirium resolved. Patients remained in the same study arm across any subsequent admissions. The primary outcome was days alive and out of the hospital after three months. Secondary outcomes included mortality; days without coma, delirium, or ventilator; serious adverse reactions; and use of restraints or rescue medications for delirium (including other antipsychotics).

There were no statistically significant differences in any outcomes. Adverse reactions were similar between the two groups, although there was a nonsignificant trend toward greater QT prolongation with haloperidol.

Although the researchers did not assess direct improvement in symptoms of delirium, the lack of a difference in the use of rescue medications or restraints raises doubts about haloperidol’s efficacy. A meta-analysis of 11 randomized trials failed to find a difference in “days without delirium” on haloperidol (Andersen-Ranberg NC et al, Crit Care 2023;27(1):329). Also, the results may not fully generalize to the US population. European ICUs have comparatively lower rates of mental illness and use sitters more than restraints. In this study, patients who were already on antipsychotics were excluded, and only 1,000 of the 1,738 screened patients were included. The doses of haloperidol used were also comparatively low.

CARLAT TAKE

This study adds to mounting evidence against antipsychotics in ICU delirium. For mild delirium, nonpharmacologic measures and sleep regulation have better support as first-line interventions.