No benefit to routine antipsychotic use for treatment or prevention of delirium in hospitalized patients

Référence

Oh ES, Needham DM, Nikooie R et al. Antipsychotics for preventing delirium in hospitalized adults. Ann Intern Med. 3 September 2019 [Epub ahead of print] doi.10.7326/m19-1859 Nikooie R, Neufled KJ, Oh ES et al. Antipsychotics for treating delirium in hospitalized adults. Ann Inter Med. 3 September 2019 [Epub ahead of print] doi.10.7326/M19-1860

Sommaire

These investigators searched multiple databases including PubMed, Embase, and the Cochrane Central Register of Controlled Trials, as well as hand-searched reference lists of included articles, to find randomized controlled trials (RCTs) that compared antipsychotics with either placebo or each other for the prevention of delirium. Two reviewers independently selected articles for inclusion and assessed the risk of bias. Out of the 14 included RCTs (n=4281), 9 had a low risk of bias (n=3407). All were conducted in the inpatient setting, 7 in the intensive care unit. There was heterogeneity amongst the studies in patient populations, settings, antipsychotic medication dosing and administration, and assessment of outcomes. Overall, haloperidol does not affect delirium incidence, delirium duration, mortality, or hospital length of stay (LOS). There is insufficient evidence to make conclusions on the effect of haloperidol on delirium severity or sedation. While there is some evidence that second-generation antipsychotics can reduce delirium incidence in the post-operative setting (RR 0.36, 95% CI 0.26-0.50), these drugs have no effect on hospital LOS or mortality and there is insufficient evidence to determine effect on delirium severity, delirium duration, or sedation. This review found no statistically significant differences between haloperidol or second-generation antipsychotics, as compared with placebo, in rate of arrhythmias, QTc prolongation, or neurologic events. In a similarly conducted systematic review evaluating the use of antipsychotics for the treatment of delirium in hospitalized patients, 16 RCTs and 10 observational studies (n=5607) were included. Haloperidol or second-generation antipsychotics, as compared with placebo, did not affect delirium duration, mortality, sedation status, or hospital LOS. Second-generation antipsychotic use did, however, result in more harmful cardiac effects, specifically QT prolongation, as compared with placebo (RR 1.95, 95% CI 1.03-3.71).