Published in the November 2006 issue of Today’s Hospitalist
Given the pervasiveness of delirium among hospitalized patients, many hospitalists make a habit of simply sedating delirious patients. But according to Jeffrey Wiese, MD, that approach may be inappropriate on several levels.
While the conventional wisdom may say that treating the patient’s underlying pneumonia or post-operative pain will help the delirium resolve on its own, Dr. Wiese, director of the hospitalist service at Tulane University Health Sciences Center in New Orleans, urged hospitalists to consider a different approach.
Not only do inpatient physicians need to give patients with delirium special attention, he explained, but they need to adopt a new paradigm to diagnose and manage the condition.
“Delirium is its own marker for mortality and morbidity, and it needs to be managed with as much urgency as the underlying disease,” Dr. Wiese told a group of hospitalists at the Fall 2006 Hospitalist CME Series meeting in Chicago. “Letting delirium persist untreated, even for fixed periods of time, may accelerate the development of dementia. It definitely increases the mortality associated with the disease that induced the delirium.”
He called on hospitalists to take a new approach to managing delirium, one that includes the extensive use of non-invasive, nonsedating therapies. In many patients with delirium, he said, these non-invasive techniques will not only resolve delirium faster, but they can stop an acute episode from progressing to persistent delirium or even dementia.
A new approach
According to some estimates, delirium accounts for up to 49 percent of hospital days. It can also affect as many as 24 percent of patients in a general medicine ward and up to 87 percent of patients in the intensive care unit (ICU). That adds up to about $2,500 per patient, or a whopping $7 billion in Medicare spending every year.
As Dr. Wiese pointed out, delirium’s cost in terms of mortality is even more staggering. One study published in the March 16, 2006, issue of the New England Journal of Medicine found that hospital admission mortality with delirium ranged between 22 and 76 percent.
Despite those statistics, most doctors don’t approach delirium with the same urgency as they do other serious mortality and morbidity markers. Take, for instance, a patient with acute renal insufficiency. “You would immediately begin intravenous fluids and try to augment blood flow and support for the kidneys,” noted Dr. Wiese, who is also associate chair of medicine at Tulane. And if you saw a patient with chest pain, he added, you’d give fluids and reduce stress on the heart to maintain cardiac cell function and integrity.
He said that when treating patients with delirium, it’s helpful to think of the disease in its most basic clinical terms, which one expert has dubbed “acute cerebral insufficiency.” Viewed from that perspective, Dr. Wiese said, delirium requires ensuring that the brain has adequate oxygen, hydration and glucose.
“More than anything else,” he explained, “these therapies stave off some of the mortality and morbidity that goes with acute delirium.”
Looking for the cause
To root out the cause of delirium in your patients, Dr. Wiese outlined the diagnostic algorithm that he and his colleagues use at Tulane. That algorithm calls for ordering tests according to pre-test probabilities, and it includes tests for common abnormalities like electrolytes, toxicology and blood cultures.
Because time is so critical in treating delirium, you want to start by ruling out meningitis and encephalitis. “We’ll first consider a meningitis diagnosis,” Dr. Wiese said, “and order a lumbar puncture and antibiotics if that seems to be present.”
Once you’ve removed meningitis from the differential diagnosis, consider other infections and order antibiotics if they’re warranted. You also want to look for evidence of trauma and order a head CT if there is a probability of trauma or intravascular hemorrhage.
Also look for evidence of intoxication from illicit drugs and consider the possibility of alcohol withdrawal and seizures. Finally, Dr. Wiese advised making a comprehensive review of the patient’s medications, looking “particularly for those drugs, like narcotics, that have anticholinergic effects.”
Delirium vs. dementia
In making your diagnosis, Dr. Wiese said, you need to distinguish delirium from dementia, which is a leading risk factor for delirium, and from primary psychiatric diseases.
A good starting point is to assess the rate of change in the patient’s altered mental status, something that can be difficult without the input of family members or caregivers. Because that support is often not available, Dr. Wiese has come to rely on the confusion assessment method, which he said is much more specific than the mini-mental status examination.
The confusion assessment method focuses on several factors “including disorientation, memory impairment, perceptual disturbances and agitation “that can help you distinguish delirium from dementia.
And if you tend to order an electroencephalogram (EEG) to help make that distinction, think again. Dr. Wiese pointed out that an EEG does not have “sufficient sensitivity and specificity to give you the positive and negative likelihood ratios you’d like to have to distinguish delirium from dementia.”
The test, however, can be helpful in two scenarios. It can help you diagnose nonconvulsive status epilepticus, which is a concern for cancer patients, and it can help distinguish delirium from depression and other primary psychiatric diseases.
“For the most part, primary psychiatric conditions have a normal EEG,” Dr. Wiese said. “The slowing of the EEG, especially as it pertains to alpha waves, is a hallmark of delirium and can be a distinguishing feature.”
To manage delirium, Dr. Wiese said hospitalists should first ask a sentinel question: Is the patient a danger to himself or to others? If the answer is “no,” try to make much greater use of something Dr. Wiese said is one of the most important recent advances in delirium management: evidence supporting the benefit of non-invasive, nonsedating therapies.
The New England Journal of Medicine study earlier this year, for example, randomized patients with delirium to receive either usual care or a non-sedating intervention designed to reduce cognitive impairment and sleep deprivation. Within the intervention group, the incidence of delirium dropped from 15 percent to 10 percent. Researchers also found a 35 percent decrease in the total number of days that patients had delirium.
Non-invasive therapies are so effective, Dr. Wiese said, because they don’t introduce more substances to a vulnerable population. “The medications that we use to treat delirium can induce further delirium,” he explained. “The fewer confounders that we have in patient management, the better.”
The delirium management protocol he has helped design at Tulane addresses several elements, including the following:
“¢ Orientation and impairment. Give patients visual orientation cues in their room, including photos of each of their medical team members. You also need to vigilantly guard against visual and hearing impairment. Make sure patients, particularly those transferred from nursing homes, have their eyeglasses and hearing aids. Also check for impacted ear wax.
“¢ Cognitive stimulation. Patients should have cognitively stimulating activities, such as regular daytime visits from family and hospital staff, at least two or three times a day.
“¢ Sunlight and mobility. Patients with delirium should be placed near windows so they get sunlight for most of the day. They should also be mobilized frequently and as soon as possible during their hospital stay.
“¢ Protected sleep periods. Look very carefully, Dr. Wiese advised, at when you order vital sign checks, medication dosing and lab draws. “You want to make sure that patients get eight hours of protected time to sleep at night, without people coming in and out,” he said. By the same token, take care that patients with delirium aren’t rooming with other patients who need care during nighttime hours.
When patients need sedation
For patients who do need sedation, Dr. Wiese urged hospitalists to make sure that they have “the luxury of good blood pressure” to institute the medications they want to use.
“This is usually the limiting feature of going straight to benzodiazepines, which reduce the preload to the heart,” he said. That makes it hard to continue these medications in patients with hypotension.
If you’re going to consider sedation, he added, the obvious place to start is haloperidol. This agent has fewer delirium-inducing effects than benzodiazepines, and it has less of a preload reduction that can be a problem with patients with low blood pressure. It’s also easier to administer than similar medications. Dr. Wiese noted that halperidol, however, does have one major downside: It can exacerbate arrhythmias.
While atypical antipsychotics such as olanzapine and risperidone are not first-line therapies for managing delirium, they may have a place in treating acute delirium, particularly among patients in the ICU who don’t have the cardiovascular support for benzodiazepines.
And benzodiazepines are the culprit in another form of delirium: benzodiazepine withdrawal in patients who are being treated for alcohol withdrawal. In these cases, Dr. Wiese advised physicians to consider managing patients with carbamazepine, even though it hasn’t been approved by the FDA for that use.
One study of carbamazepine found that patients suffering from benzodiazepine withdrawal benefited from 800 mg a day of the drug in divided doses, with every third dose reduced by 100 mg. “For that alcohol withdrawal patient who is just not getting better,” said Dr. Wiese, “this is an option worth considering.”
Finally, Dr. Wiese pointed out, hospitalists need to increasingly move to team-based care for managing delirium, particularly as their own coverage migrates to being shift-based.
“The hallmark of delirium is that it waxes and wanes during different times of the day,” he said. As one physician shifts signs off to another, “having team-based conferences that can ensure continuity in management is essential for coordinating this care.”
Michael Krivda is a freelance writer specializing in health care. He is based in Perkasie, Pa.
Sun-downing: another form of delirium
In his presentation on delirium at the Fall 2006 Hospitalist CME Series in Chicago, Jeffrey Wiese, MD, dispelled some major myths about “sun downing,” the syndrome in which patients become agitated only late in the day.
“People consider sun-downing to be a natural feature of being old, and it’s not,” Dr. Wiese said. “Sun-downing is its own form of delirium.”
Part of the problem is that sun-downing patients, when they’re not agitated, frequently appear to be quiet, withdrawn or sleepy. That leads many physicians to prescribe medications such as diazepam on a PRN basis to treat the sun-downing episodes as isolated incidents.
The truth is that physicians need to realize that these patients are actively suffering from delirium, even when they’re not displaying agitation.
“Pay particular attention to their medication list,” said Dr. Wiese, who is director of the hospitalist service at Tulane University Health Sciences Center in New Orleans. Find out what anticholinergic drugs the patient is taking, he advised, and eliminate as many medications as possible.