Trial Comparing Haloperidol, Quetiapine and Placebo in the Pharmacological Treatment of Delirium

Background:

Delirium is an important problem in critical care. Its prevalence often reaches 75% in intensive care patients. Its occurrence is associated with numerous complications and deleterious consequences such as death, longer stay, higher cost, and long-term cognitive impairment. Delirium treatment entails correcting its underlying causes and usually initiating a pharmacological intervention with an antipsychotic. Typical antipsychotics, particularly haloperidol, are commonly used to treat delirium although few placebo-controlled trials of pharmacological treatments for delirium have been conducted. Furthermore, appropriate doses for delirium treatment have yet to be established. In critical care, two pilot studies provided the first randomized, placebo-controlled evidence for the pharmacologic treatment of ICU delirium. One found that neither haloperidol nor ziprasidone significantly reduced the incidence or duration of delirium compared with placebo whereas the other one found that quetiapine added to as-needed haloperidol resulted in faster delirium resolution.

Objective:

The goal of this study is to determine the effectiveness of antipsychotics in regular dosage regimen (quetiapine group and haloperidol group) compared to as-needed haloperidol (placebo group) in the pharmacological treatment of delirium. We will conduct a three-arm randomized controlled trial to achieve this goal.

Materials and Methods:

During one year, 45 delirious patients from three intensive care units will be recruited and randomized into one of three groups. Randomization will be performed in blocks of 9 by the pharmacy department, using a random numbers table.

Patients will be continuously screened for delirium using the Intensive Care Delirium Screening Checklist (ICDSC) as part of routine care. A positive screening score (≥4) will warrant confirmation of delirium diagnosis by the treating physician. Treatment will begin according to randomization group, provided that informed consent has been obtained. Delirium status will be monitored during the episode using the Nursing Delirium Screening Scale (Nu-DESC). When the Nu-DESC monitoring will become negative for delirium (total score below 2), the resolution of the episode will be confirmed by the treating physician. A clinical evaluation by a psychiatrist will be performed within 24-48 hours of each of the two evaluations made by the treating physician (beginning and end of the delirium episode).

The treating physician will initiate twice-daily treatment at the first of five levels for each of the three groups: 1) 1 mg of intravenous (IV) haloperidol + oral (PO) placebo, 2) 50 mg of PO quetiapine + IV placebo, or 3) IV + PO placebo. Therapy will be titrated upwards on a daily basis by increments of 1) 1 mg of IV haloperidol or 2) 50 mg of PO quetiapine, or 3) IV + PO placebo every 12 hrs, respectively, if the subject received at least two doses of as-needed haloperidol in the previous 24 hrs. As-needed (PRN) doses of 2 mg of IV haloperidol q 30 minutes will be available to patients from all three groups and administered by nurses until symptoms associated with delirium resolve. In case of unsuccessful as-needed treatment, rescue (STAT) doses of 5 mg of IV haloperidol q 30 minutes will be available to patients from all three groups and will be administered by nurses if agreement is reached with the treating physician that the situation indeed calls for it. The treatment level of patients requiring a STAT dose will immediately be raised to the above level. The treatment will stop when one of the following occurs: (1) the subject is deemed by the treating physicians, based on their clinical judgment, to no longer demonstrate signs of delirium and, therefore, to no longer require scheduled therapy with an antipsychotic agent; (2) 21 days of therapy has elapsed; (3) ICU discharge occurred; or (4) a life-threatening adverse event potentially attributable to the study drug occurred that warranted discontinuation of the study drug.

Adverse effects will be closely monitored: extrapyramidal reactions, neuroleptic malignant syndrome, drowsiness, hypotension, QTc prolongation. The treatment level of patients presenting a non life-threatening adverse event will immediately be lowered to the level directly below.

The sample size was calculated for a 2-tailed test with an alpha of .05 and a power of .80.

The primary statistical analysis will involve Cox proportional time to event analysis comparing the three groups. Secondary analysis will use T-test comparisons for continuous variables and chi square for proportional analysis.