Treating Delirium: When the Brain Goes Off Track

October 1, 2006

Typically, delirium worsens at night ("sundowning"), with lucid intervals often present in the morning. It is important to realize that delirium may appear before any abnormal laboratory values are detected and may persist after the resolution of these abnormalities.


The great Talmudic sage Hillel was once challenged to sum up his faith while standing on one foot. He replied, "That which is hateful to you, do not do to your neighbor. All the rest is commentary." If a modern-day sage of neuropsychiatry were asked to sum up the treatment of delirium, he or she might reply, "Prevent delirium in the first place, but if it occurs, treat the underlying cause and do not make matters worse. All the rest is commentary."

Indeed, any treatment for delirium that is not aimed at reversing the underlying pathophysiology could be considered a form of "cosmetic psychopharmacology," to use Peter Kramer's famous expression--or at best, a kind of symptomatic damage control. Obviously, it is far better to prevent delirium in the first place than it is to treat it after it presents, and a variety of primary and secondary delirium-prevention strategies have been devised.1,2 Nonetheless, the effectiveness of such strategies is still unclear, and psychiatrists faced with agitated or disturbed delirious patients need to know the advantages and disadvantages of pharmacologic treatment approaches. First, though, I will briefly review some clinical aspects of delirium and a few important new research findings.

Out of the furrow

The term "delirium" is derived from the Latin, de lira, meaning "out of the furrow"--perhaps an allusion to the old Roman plow blade going "off track." Delirium may be defined as an acute or subacute disturbance of consciousness characterized by reduced ability to focus, sustain attention, or shift attention appropriately. Its hallmark, however, is a fluctuating level of consciousness, often changing from hour to hour. A prodrome of restlessness, insomnia, and nightmares sometimes precedes frank delirium. The clinical picture may sometimes include slowed or slurred speech, irritability, combative behavior, impaired short-term memory, sensory distortions, psychotic phenomena, and a reversed sleep-wake cycle.

Typically, delirium worsens at night ("sundowning"), with lucid intervals often present in the morning. It is important to realize that delirium may appear before any abnormal laboratory values are detected and may persist after the resolution of these abnormalities. Thus, the diagnosis is made on clinical, not biochemical, grounds.3,4 Delirium may be found in as many as 11% to 16% of medical inpatients, and rates may climb to as high as 65% among elderly patients admitted for acute hospital care.5 Delirium may be superimposed on preexisting dementia, and it is sometimes not appreciated in elderly patients with premorbid Alzheimer disease or other dementing illnesses. Often, a sudden burst of yelling, agitation, or aggressive behavior will flag the demented patient's new onset of delirium.

I sometimes tell residents that delirium is caused by Harrison's Principles of Internal Medicine, since virtually any of the major medical disorders discussed therein may present with an "acute organic brain syndrome"--the now outmoded term for delirium. Thus, delirium may be seen as a complication of primary neurologic diseases, intercurrent infection, hypoxia, dehydration, orthopedic or cardiac surgery, or a plethora of other medical conditions.6 Both prescription and over-the-counter drugs are also frequent inciting factors. For all of these reasons, a careful search for the underlying causes of delirium is essential.

But is there a common pathophysiology that might unify these diverse causes of delirium? The honest answer is that we don't know. However, the leading candidate appears to be a massive failure in cholinergic neurotransmission, leading to the disruption of pathways linking cortical and subcortical structures.6 While this hypocholinergic hypothesis is certainly a gross oversimplification, it is both heuristic and testable. We can show, for example, that anticholinergic drugs induce or worsen delirium in human subjects and that serum anticholinergic activity is increased in patients with delirium. Conversely, physostigmine (Antilirium, others)--a reversible cholinesterase inhibitor that boosts acetylcholine--has FDA-approved labeling for the reversal of delirium due to anticholinergic toxicity. As we will see a bit later, the principle underlying this drug's action is now being exploited, using newer and better-tolerated agents.

Pharmacologic management of delirium

Nonpharmacologic approaches to the delirious patient are an important part of good care, and they include such interventions as creating a calming environment with "orienting cues," such as clocks and calendars. However, our focus is on pharmacologic interventions, which are usually required for patients with delirium whose behavior jeopardizes their safety or the safety of others (eg, patients who are pulling out their intravenous lines or assaulting others).

Historically, the mainstay of pharmacologic management has been the drug haloperidol (Haldol)--and arguably, this agent remains the gold standard despite great interest in atypical antipsychotics, such as risperidone (Risperdal), olanzapine (Zyprexa), and quetiapine (Seroquel).1,6 Indeed, Inouye7 insists that "haloperidol remains the only recommended agent for the treatment of delirium in older persons. . . . The use of atypical antipsychotic drugs is not recommended, since the efficacy of these drugs remains un-clear." On the other hand, a recent double-blind comparison (N = 28) of haloperidol and risperidone found no significant difference in the efficacy or response rate between the 2 drugs.8 Both agents seemed to be well tolerated (akathisia developed in 1 haloperidol-treated patient), but no objective ratings of side effects were used. Open studies suggest that other atypical antipsychotics may be useful in treating delirium, but controlled data and head-to-head comparisons with haloperidol are lacking.

The benzodiazepine lorazepam (Ativan) seems to be a popular agent for managing delirious patients in emergency department settings, and it is sometimes combined with haloperidol to create the drug cocktail known facetiously as "Halivan."9 And yet, as Carnes and associates9 pointedly note, "the common selection of lorazepam to treat delirium is troubling because benzodiazepines themselves are implicated in delirium." Indeed, Inouye7 points out that "lorazepam may prolong and worsen symptoms of delirium." With the exception of delirium due to alcohol withdrawal, neuroleptic malignant syndrome, or Parkinson disease, it is usually best to avoid benzodiazepines in managing patients with delirium.6 (Benzodiazepines, moreover, can sometimes worsen the motoric component of Parkinson disease in patients taking levodopa, perhaps by interfering with dopaminergic functioning.10)

New directions

If the hypocholinergic hypothesis were correct, one would expect that agents that boost acetylcholine might actually reverse, reduce, or even prevent delirium. Recently, clinicians have reported encouraging findings using so-called Alzheimer medications in the treatment of various kinds of delirium. For example, Kalisvaart and coauthors11 recently reported on 3 elderly men (aged 85, 79, and 81 years) who were suffering from prolonged delirium and were unresponsive to haloperidol or other atypical antipsychotic drugs. Within days of starting treatment with the cholinesterase inhibitor rivastigmine (Exelon), all 3 responded well. Similarly, Kobayashi and colleagues12 reported on the successful use of donepezil (Aricept) for severe, intractable delirium associated with a right basal forebrain vascular lesion after surgery for craniopharyngioma. The patient had been treated with antipsychotics, antidepressants, and hypnotics with little improvement. Donepezil administration "dramatically improved" his intractable delirium at the 19th post-donepezil-administration day, but this was followed by amnestic symptoms. Nevertheless, the authors concluded that the case supported the hypocholinergic theory of delirium.

Such positive case reports, however, have been overshadowed recently by a negative controlled study. Liptzin and colleagues13 studied 80 patients in a randomized, double-blind, placebo-controlled trial of donepezil. Each patient received donepezil or placebo for 14 days before surgery and 14 days afterward. Postoperative delirium was assessed using the Delirium Symptom Interview and other instruments. The study found that delirium, diagnosed by DSM-IV criteria, was present on at least 1 postoperative day in 18.8% of the patients, but there were no significant differences between the donepezil and placebo groups. There was also no difference between the groups in the occurrence or duration of subsyndromal delirium. The authors concluded that they were "unable to demonstrate a benefit for donepezil in preventing or treating delirium in a relatively young and cognitively intact group of elderly patients undergoing elective orthopedic surgery."13

Although discouraging, this finding does not mean that donepezil or related agents are ineffective in treating all types of delirium. Indeed, local experience (S.A. Jacobson, MD, personal communication, October 2005) suggests that cholinesterase inhibitor therapy is useful in various types of delirium, particularly acute hepatic encephalopathy. Large-scale randomized, controlled studies are clearly needed; however, in the meantime, there are compelling theoretical reasons to consider a cholinesterase inhibitor for delirious patients who do not respond to haloperidol.

Summing up

Delirium is a syndrome in search of a diagnosis.Psychiatrists must not only recognize delirium but must also make vigorous efforts to determine its causes. Treatment is directed at reversing these underlying factors whenever possible. Symptomatic management is usually best undertaken using the gold standard agent haloperidol even though controlled studies comparing haloperidol with other antipsychotics are lacking. With respect to newer atypical antipsychotics, the proverbial jury is still out. Benzodiazepines--although popular--have significant drawbacks in treating delirium, except in cases involving delirium from alcohol withdrawal. The use of cholinesterase inhibitors is an exciting area of research and represents a reasonable treatment option in refractory cases.

Note: Drugs discussed for the treatment of delirium are all used "off-label" (without FDA approval), with the exception of physostigmine.

Dr Pies is clinical professor of psychiatry at Tufts University. His most recent books include Creeping Thyme, a collection of poetry (Brandylane Publishers); Zimmerman's Tefillin, a short story collection (PublishAmerica); and Handbook of Essential Psychopharmacology, 2nd edition, (American Psychiatric Publishing).


References:1. Leentjens AF, van der Mast RC. Delirium in elderly people: an update. Curr Opin Psychiatry. 2005;18: 325-330.
2. Inouye SK, Bogardus ST Jr, Charpentier PA, et al. A multicomponent intervention to prevent delirium in hospitalized older patients. N Engl J Med. 1999;340: 669-676.
3. Trzepacz PT, Wise MG. Neuropsychiatric aspects of delirium. In: Yudofsky SC, Hales RE, eds. The American Psychiatric Press Textbook of Neuropsychiatry. 3rd ed. Washington, DC: American Psychiatric Press; 1997: 447-470.
4. Lipowski ZJ. Delirium: Acute Confusional States. New York: Oxford University Press; 1990.
5. Levkoff S, Cleary P, Liptzin B, Evans DA. Epidemiology of delirium: an overview of research issues and findings. Int Psychogeriatr. 1991;3:149-167.
6. Inouye SK. Delirium in older persons. N Engl J Med. 2006;354:1157-1165.
7. Inouye SK. Response to letter. N Engl J Med. 2006; 354:2510-2511.
8. Han CS, Kim YK. A double-blind trial of risperidone and haloperidol for the treatment of delirium. Psychosomatics. 2004;45:297-301.
9. Carnes M, Howell T, Rosenberg M, et al. Physicians vary in approaches to the clinical management of delirium. J Am Geriatr Soc. 2003;51:234-239.
10. Yosselson-Superstine S, Lipman AG. Chlordiazepoxide interactions with levodopa. Ann Intern Med. 1982; 96:259-260.
11. Kalisvaart CJ, Boelaarts L, de Jonghe JF, et al. Successful treatment of three elderly patients suffering from prolonged delirium using the cholinesterase inhibitor rivastigmine [in Dutch]. Ned Tijdschr Geneeskd. 2004;148:1501-1504.
12. Kobayashi K, Higashima M, Mutou K, et al. Severe delirium due to basal forebrain vascular lesion and efficacy of donepezil. Prog Neuropsychopharmacol Biol Psychiatry. 2004;28:1189-1194.
13. Liptzin B, Laki A, Garb JL, et al. Donepezil in the prevention and treatment of post-surgical delirium. Am J Geriatr Psychiatry. 2005;13:1100-1106.