Delirium: Risperidone vs haloperidol vs placebo. An unexpected end.


Efficacy of Oral Risperidone, Haloperidol, or Placebo for Symptoms of Delirium Among Patients in Palliative Care: A Randomized Clinical Trial, by Meera Agar et al [1].


As I told you guys before (link), delirium treatment demands two things: patient’s safety and patience. There isn’t still a drug that I can give and relax, because it will shorten delirium duration, and here I’m talking about any well studied drug with strong evidence behind it. Now, one of the few studies that showed any benefit in reducing delirium duration was published in 2010, by Devlin and colleagues [2], but, it was a small study, with only 18 patients per arm.

Despite delirium is a theme surrounded by immense debates, its pharmacological treatment hasn’t improved very much during the last years. Today, the cornerstone treatment is prevention and a multidisciplinary approach. In cases of hyperactive or mixed delirium, culminating in agitation, unfortunately, there isn’t much science, we get our blowpipe loaded with haloperidol, aim good, and hope for the best. FYI, here in Brazil, we abandoned the use of blowpipes in critical care a long ago! And also, we don’t have pet monkeys. Just in case you were wondering!


This study is one of many that shows us how important is to choose wisely our patients before prescribing any antipsychotics to them. This study population does not include ICU patients. But this is an ICU blog, one might say! Hey! Relax, and open your horizons! Only patients under palliative care, some of them in hospices, with progressive diseases that were no longer curable were selected. The inclusion criteria were (patients should meet all 3 criteria):

-Delirium diagnosed via DSM IV (link) criteria

-Memorial Delirium Assessment Scale (link)

-Presence of delirium symptoms associated with distress, using the NuDESC scale (link)

Patiens were excluded if: regular use of antipsychotic drugs within 48 hours or any contraindication to its use.


They randomized patients in a 1:1:1 ratio to placebo, haloperidol or risperidone. Patients < 65y/o received a loading dose of 0.5mg plus a first dose of 0.5mg, then, 0.5mg every 12h to a maximum dose of 4mg/day. For patients older than 65y/o, the doses were halved. Patients with hallucinations/agitation could receive 2.5mg of midazolam SC every 2h, for symptom control. All participants received individualized treatment plans despite the intervention. The overall dose of haloperidol used in this study was a bit low, however, since the average age of patients was around 75y/o, maybe it was fair and smart to use a smaller dose, in order to avoid any stiff drooling grandpa at the hospices. The drugs were given every 12h, which didn’t sound too appropriate for me. If I think my patient has any indication for haloperidol use for symptoms control, I will not give him a drug and go for a 12h coffee until the next re-evaluation. The solution authors came with, without giving more antipsychotics, was to use midazolam instead. Ok, in a palliative care way of thinking, the use of midazolam for symptoms control makes sense, but, if somehow we will try to extrapolate these results for an ICU environment, we know that the use of benzos is associated with delirium. Another downside was the time to access the primary outcome (average of the last 2 delirium symptom scores), on day 3. Maybe we are not able to see any benefit of the intervention in only 3 days. Let’s see this table from one of delirium studies I like the most, published by Sharon Inouye in NEJM [3].

If Sharon had stopped her study, which compared a multicomponent intervention vs standard of care to prevent delirium, on day 3, we probably would say that the intervention was not different from usual care. Instead, when you look through time, we can see the difference. Today, we all know the importance of a interdisciplinary team approach in delirium prevention adn treatment. Despite of that, it was a well written study, since one of the authors critics to other studies were their methodological flaws. Previous studies used delirium symptoms scores to measure the primary outcome, that included symptoms that are not treated with antipsychotic drugs, to avoid this, the authors decided to use the NuDESC-derived total score on items 2, 3, and 4 ( inappropriate behavior, inappropriate communication, illusions/hallucinations). A difference of 1 unit between baseline and follow-up was considered clinically significant. Here I have an issue, because, if the patient was non-communicative, he would receive 1 point, and it’s hard to believe that this is clinically significant. Sample size calculation came up with 80 patients/arm. They used ANCOVA for primary analysis, and missing data was handled with multiple imputation. Unfortunately, 15 patients in placebo group, 18 in haloperidol and 31 in risperidone group had the treatment discontinued, all for good reasons, however, 18 patients in risperidone group had the treatment suspended due patient deterioration, twice as much as the other groups. Also, I really don’t think is a good idea to input data for primary outcome when you have so many drop-outs and small sample sizes.

This table from the supplement shows us interesting data. First, we can see the dropouts, also we can see the mean number of NuDESC score (items 2, 3, 4) through time. Patients began with a score around 2.55, and ended up with scores no bigger than 1.3 (placebo 0.71). Is a score of 1.3 something that bothers me? Probably not! I don’t even know if the baseline scores bother me. Maybe they could have studied patients with more pronounced symptoms.


What was found

Regarding the primary outcome, the authors highlighted that risperidone not only didn’t decrease delirium symptoms, but it made them worse. The ITT analysis showed that risperidone arm had greater delirium symptoms than placebo (0.48 Units, 95% CI, 0.09-0.86; P = .02), the same for haloperidol arm vs placebo (0.24 Units, 95% CI, 0.06-0.42; P = .009). Secondary analysis corroborated with these results, as shown below:

The authors also found similar results with other analysis. These findings are in concordance with a recent published review [4]. However, despite the conservative doses of haloperidol and the time to measure the primary outcome, I think this can be considered one of the most important studies evaluating delirium treatment in recent years, showing us that drugs we use in our daily practice might be worthless, but also can be dangerous. For me, it’s clear that the focus of delirium treatment should be prevention. Since 99 we know that multidisciplinary interventions work for delirium prevention and control. That’s what we should be focusing on, and let aside this pursuit of a miraculous drug to “fix” a dysfunction, which sometimes is consequence of numerous predisposing factors. I don’t fully agree with the authors when they say that a bigger dose could lead to more side effects. It might help to treat more severe symptoms. Lot of studies showed the safety of delirium treatments, like the MIND trial [5].

In the end, I’m left with these feelings: For patients under palliative care, we should think twice before giving any antipsychotic expecting to decrease delirium symptoms. On the other hand, we should have in mind that antipsychotics can help us in treating other symptoms, like nausea (haloperidol/chlorpromazine), insomnia (olanzapine/quetiapine) and hyporexia (olanzapine). This study might tell us that we can increase delirium in our patient by giving them antipsychotics, but, we should always remember that in palliative care, we should treat the symptoms that brings suffering to our patients, not to us.


1.     Agar MR, Lawlor PG, Quinn S, et al. Efficacy of Oral Risperidone, Haloperidol, or Placebo for Symptoms of Delirium Among Patients in Palliative Care: A Randomized Clinical Trial. JAMA Intern Med. 2016.

2.     Devlin JW, Roberts RJ, Fong JJ, et al. Efficacy and safety of quetiapine in critically ill patients with delirium: a prospective, multicenter, randomized, double-blind, placebo-controlled pilot study. Crit Care Med. 2010;38(2):419-427.

3.     Inouye SK, Bogardus ST, Charpentier PA, et al. A multicomponent intervention to prevent delirium in hospitalized older patients. N Engl J Med. 1999;340(9):669-676.

4.     Neufeld KJ, Yue J, Robinson TN, Inouye SK, Needham DM. Antipsychotic Medication for Prevention and Treatment of Delirium in Hospitalized Adults: A Systematic Review and Meta-Analysis. J Am Geriatr Soc. 2016;64(4):705-714.

5.     Girard TD, Pandharipande PP, Carson SS, et al. Feasibility, efficacy, and safety of antipsychotics for intensive care unit delirium: the MIND randomized, placebo-controlled trial. Crit Care Med. 2010;38(2):428-437.



Photo credit


What's Your Reaction?

Confused Confused
Damn Damn
Geeky Geeky
Like Like
Scary Scary
Win Win
Angry Angry
Happy Happy

Delirium: Risperidone vs haloperidol vs placebo. An unexpected end.

log in

Become a part of our community!

reset password

Back to
log in