Valproic Acid for Treatment of Hyperactive or Mixed Delirium in ICU

Overview

Delirium is the most often encountered psychiatric diagnosis in the general hospital, with incidence up to 85% in the intensive care unit (ICU) setting and with significant consequences on patients' morbidity and mortality. Currently, although not FDA approved, antipsychotics are often considered the first-line pharmacological treatment. However, there can be limitations to their use, including side effects or lack of efficacy. Valproic acid (VPA) is one of the alternatives at times used in such patients which from limited case series data appears to be helpful and tolerated. VPA can provide relief from agitation that poses a threat to the safety and recovery of the patient. Moreover, mechanistically it addresses the neurochemical and cellular abnormalities inherent in delirium (it has NMDA-antagonist, anti-dopaminergic, GABAergic,anti-inflammatory, anti-apoptotic, and histone deacetylase inhibitor properties, among others). The purpose of this study is to evaluate the efficacy and tolerability of the VPA in the first known to us randomized controlled trial.

The investigators plan to investigate the efficacy and tolerability of scheduled VPA as compared to placebo with as needed basis (PRN) haloperidol (as a back-up in both arms) for treatment of hyperactive or mixed delirium. Patients will be randomized to scheduled VPA or placebo (normal saline) and both arms will have flexible PRN dosing of haloperidol. Thus, the investigators plan to learn the time to delirium resolution in patients treated with VPA versus placebo; percentage of patients responding to VPA versus placebo; tolerability of VPA versus placebo. If addition of scheduled VPA proves to shorten time to delirium resolution as compared to placebo, lead to less use of haloperidol, and/or have fewer side effects, it would provide a very important addition to our limited evidence-based repertoire of delirium treatment. Moreover, this pilot study would then pave the way for the bigger randomized control trial powered to detect its effect size.