Dealing with ICU Delirium

Abstract & Commentary

By David J. Pierson, MD, Editor, Professor, Pulmonary and Critical Care Medicine, Harborview Medical Center, University of Washington, Seattle, is Editor for Critical Care Alert.

Synopsis: Delirium occurred in about one-third of patients in this study of a mixed medical-surgical ICU population. It was more frequent in more seriously ill patients, and also in those with hypertension, alcoholism, and the effects of sedative and analgesic drugs.

Source: Ouimet S, et al. Incidence, risk factors, and consequences of ICU delirium. Intensive Care Med. 2007;33:66-73.

Ouimet and colleagues at Maisonneuve-Rosemont Hospital in Montreal studied 820 consecutive patients admitted to their mixed medical-surgical ICU to determine the incidence of delirium, factors associated with it, and its clinical consequences. The patients were adults who stayed in the ICU more than 24 hours and survived for at least 1 day. The mean APACHE II score on admission was 16.5, and 79% of the patients were mechanically ventilated. Delirium was assessed daily using the Intensive Care Delirium Screening Checklist (ICDSC), as previously reported by the same group1 and as summarized in Table 1. Patients were considered to have delirium if the score on the 8-point assessment screen was 4 or higher.

After patients who remained comatose throughout their ICU stay (n = 56) were excluded, delirium occurred in 243 of 764 patients (31.8%), during a mean of 5.7±7 days of data collection per patient. Patients who developed delirium had an ICDSC score of 4 or higher for a mean of 38% of their ICU stay, and 10% of them remained delirious at ICU discharge.

Comparing patients with delirium with those who did not develop it according to the ICDSC, initial APACHE scores were higher (mean 18 vs 14, p < 0.0001), but there were no differences in age, sex, or diagnosis. Delirium occurred with equal frequency in medical and surgical patients, and was not more frequent in those with previous neurologic illness. However, it was statistically more frequent in patients with hypertension (odd ratio, 1.88, 95% confidence interval 1.3-2.6) and alcoholism (OR 2.03; 95% CI 1.26-3.25). Delirium was more likely in patients who received sedatives and analgesics when used to induce coma (OR 3.2, 95% CI 1.5-6.8) for procedures, but not when these drugs were used in other circumstances. Patients who developed delirium while in the ICU experienced higher ICU mortality (19.7% vs 10.3%) and overall hospital mortality (26.7% vs 21.4%), as well as longer average stays in both the ICU (11.5 vs 4.4 days) and in the hospital (18.2 vs 13.2 days).


Evidence is accumulating that delirium in ICU patients is an important and detrimental phenomenon. A number of studies have found associations between the development of delirium and increased morbidity and mortality as well as with increased lengths of stay in both ICU and hospital.2 It stands to reason that sicker patients and those with longer ICU stays would be more likely to develop delirium—just as they are more likely to develop nosocomial infections and dysfunction in other organ systems—but multivariate analyses in several studies have documented that the adverse effects of delirium persist when these things are accounted for. Thus, the prevention, prompt diagnosis, and treatment of delirium in the ICU should be prominent in the clinician's mind during the management of patients' primary illnesses.

Two assessment schemes are available for diagnosing and monitoring delirium: the ICDSC, as used in the present study1, and the Confusion Assessment Method for the ICU (CAM-ICU), introduced by Ely and colleagues.3 The incidence of delirium among patients in the ICU in different studies, using these techniques for diagnosis, has ranged from about 10% to more than 80%, likely reflecting differences in severity of illness, case mix, and management in addition to any differences between the methods themselves. The incidence of 32% in the present study, which excluded relatively few patients and included a broad mix of medical and surgical ICU patients, seems reasonable considering all the variables in published reports. In any event, ICU delirium is clearly both common and important, making its recognition and management high priority for all critical care clinicians.

Several other important points are brought out by the Ouimet study and its accompanying editorial. ICU delirium occurs in both "hyperactive" and "hypoactive" forms. While the former is easier to recognize and poses the obvious threats of unintended extubation and dislodgement of vascular lines, studies have shown that delirium in patients lying motionless in bed is also associated with adverse outcomes. Delirium should not be treated with sedatives and narcotics. These agents can mask its manifestations but they do not treat the underlying disorder and in fact may make it worse. The current treatment of choice is haloperidol.

Table 2, adapted in part from the editorial by Polderman2, lists steps clinicians can take to reduce the likelihood that their patients will develop delirium in the ICU.

While few would argue with the importance of preventing a phenomenon so strongly associated with adverse patient outcomes, at present there is little evidence that treating delirium once it is recognized improves those outcomes. Common sense tells us that treatment should help, but well-designed studies of this important aspect of ICU management are sorely needed.


  1. Bergeron N, et al. Intensive Care Med. 2001;27:859-864.
  2. Polderman KH. [editorial] Intensive Care Med. 2007;33:3-5.
  3. Ely EW, et al. JAMA. 2001;286:2703-2710.