Continuous Intravenous Haloperidol Infusions for Agitated Delirium

I rarely get consultation requests to help critical care physicians manage agitation from delirium that is so severe that patients cannot be extubated and removed from deep sedation. Usually, this kind of agitation tends to resolve after the medical causes of delirium are diagnosed, treated, and managed with what would now be called relatively low doses of antipsychotic or Dexmedetomidine.

About 15 to 2o years ago we used to use large doses of intravenous (IV) haloperidol, most often in bolus doses but occasionally in continuous infusions. It sounds archaic and even hazardous now, but protocols similar to the one below were defended in the medical literature because they were effective, and well-tolerated.

IV Haloperidol Drip Protocol


In this case, the primary team physicians have been unable to find a reversible medical cause for the delirium and cannot extubate the patient and remove heavy sedation without incurring the appearance of severe agitation which makes the patient dangerous to self and others. It has been determined that alcohol or sedative-hypnotic withdrawal is not the explanation either. Agitation in this instance may interfere with medical management and therapeutic goals. As far as it’s possible to discern, adequate analgesia is being provided. Other efforts at anxiolysis and control have been ineffective.


It’s important to note that in a small study by Reade et al comparing Dexmedetomidine to IV haloperidol, the former was associated with quicker time to extubation and discharge from the ICU, although it was needed longer than IV haloperidol post-extubation. The QTc interval tended to be longer in patients on IV haloperidol but it’s not clear if this was clinically relevant.


However, the patient has been unresponsive to Dexmedetomidine, the usual drug recommended when agitation from delirium does not respond to bolus IV dosing of haloperidol.


There is no psychiatric medication that is specific for managing this level of agitation. In extreme cases when the only alternatives seem to be severe agitation or deep sedation with anesthetic agents, continuous IV infusion of haloperidol has been reported to be helpful. However, the studies are flawed and our experience with these strategies is limited.


According to Wang et al, “In summary, IV regimens of haloperidol are most useful in the critically-ill given their fast onset of action, ease of administration, and predictable pharmacokinetics. Intermittent IV administration should be the dosing regimen of choice, whereas continuous infusion of haloperidol may be considered for patients failing to respond to large, frequent boluses of IV haloperidol.”


After a discussion of potential risks and benefits, the possibility of using a continuous drip of IV haloperidol could be considered according to the protocol by Riker. This could be adapted as follows:


1. If more than eight 10 mg haloperidol boluses are needed in a 24 hr period or more than 10 mg/hr for more than 5 consecutive hours, consider IV infusion and monitor QTc interval closely, consider stopping if well over 500 ms; although extrapyramidal side effects are reportedly rare with this method, they can occur and could be severe. We’ll advise on management using IV benztropine mesylate if this happens.


2. Bolus dose of 10 mg, followed by 5 mg/hr initial infusion rate.


3. If agitation persists, can repeat bolus with 10 mg every 30 minutes accompanied by an increase in the infusion rate by 5 mg/hr.


4. Consider addition of benzodiazepine, although this has also been identified as a predictor for delirium in the ICU.


5. Once control attained with few or no boluses required for 24 hrs, decrease infusion rate by half or, if the patient achieves a Richmond Agitation and Sedation Scale (RASS) goal score of 0, stop infusion and return to bolus infusions as needed.


6. Monitor on telemetry; be alert for extrapyramidal symptoms and signs of neuroleptic malignant syndrome as well.


The guidelines from the Riker paper for making the haloperidol infusion: “200 mg in 160 mL of 5% dextrose in water (1 mg/mL). Maximum concentration with normal saline to 0.75 mg/mL or 5% dextrose in water to 3 mg/mL (68-70).”





Reade, M. C., K. O’Sullivan, S. Bates, D. Goldsmith, W. R. Ainslie and R. Bellomo (2009). “Dexmedetomidine vs. haloperidol in delirious, agitated, intubated patients: a randomised open-label trial.” Crit Care 13(3): R75.


Wang, E. H., V. H. Mabasa, G. W. Loh and M. H. Ensom (2012). “Haloperidol dosing strategies in the treatment of delirium in the critically ill.” Neurocrit Care 16(1): 170-183.


Riker, R. R., G. L. Fraser and P. M. Cox (1994). “Continuous infusion of haloperidol controls agitation in critically ill patients.” Critical care medicine 22(3): 433-440.

I very much prefer not to recommend this intervention nowadays, but on rare occasions there seems to be few available safer alternatives. There are probably many psychiatrists who remember these protocols. The one above is adapted from the paper by Riker, with updates from the data and monitoring procedures from Wang and Reade. There are only 4 studies extant about using IV haloperidol drips that I’m aware of, although there may be somebody out there who knows more about it.



2 thoughts on “Continuous Intravenous Haloperidol Infusions for Agitated Delirium

  1. Jim,

    30 years ago we used massive doses of PO/IM haloperidol and other potent typicals to treat agitated psychosis. I think Seeman’s work on D2 receptor binding was instrumental in realizing that was not necessary. HIgh dose IV haloperidol has been around for a while. I can recall from the days of the Swan Ganz catheter that there was a study looking at cardiac parameters in ICU patients that they were essentially unchanged before and after haloperidol. That was well before the days of obsessive concern about QTc intervals. You would think that in the days we were referring patients to Cardiology for QTcs > 500 msec due to thioridazine that somebody would have noticed in an ICU setting that haloperidol did the same thing. I am skeptical of the QTc effect of haloperidol for that reason and also because whenever I have looked it up in the literature – they were typically polypharmacy cases with numerous other causes. ICU patients also usually have one or more other potential causes.

    Looking at it from a purely psychiatric perspective, agitation associated with lethal catatonia/delirious mania can be approached from an anesthesiology perspective when all of the usual psychiatric drugs fail. I have treated that with propofol (managed by intensivists and anesthesiology) until the patient could be stabilized on the usual mood stabilizers and antipsychotics. Has light anesthesia with propofol been used to transition agitated critically ill patients?

    I also like to see detailed pharmacokinetic modelling of intravenous haloperidol and if it makes theoretical sense. I have not seen any papers that address that issue.


Comments are closed.