Article Text

Observational study
Among patients admitted to ICU, delirium is most common in those with neurological diagnoses, and is associated with adverse health outcomes
  1. Margaret A Pisani,
  2. Lauren Ferrante
  1. Department of Internal Medicine, Yale University, New Haven, Connecticut, USA
  1. Correspondence to : Dr Margaret A Pisani
    Yale University School of Medicine, Pulmonary & Critical Care, PO Box 208057, New Haven, CT 06520–8057, USA; margaret.pisani{at}

Statistics from

Commentary on OpenUrlCrossRefPubMed

Implications for practice and research

  • Delirium in an intensive care unit (ICU) population is associated with adverse short-term outcomes and increased mortality.

  • Patients with mixed and hyperactive delirium subtypes have worse short-term outcomes than those with hyperactive delirium.

  • Patients admitted to an ICU with neurological diagnoses have a high incidence of delirium.


The study by van den Boogaard and colleagues confirms prior literature that delirium can be detected using validated screening methods and that it is prevalent in critically ill patients and is associated with adverse outcomes. This study delves further into delirium subtypes comparing hyperactive, hypoactive and mixed delirium subtypes among different types of ICU admissions. One notable difference when comparing this work to prior studies is that ICU length of stay with a median stay of 1 day and length of mechanical ventilation are both shorter than that described in other similar cohorts.


This prospective observational cohort enrolled 1613 patients admitted to a 33-bed university hospital ICU in the Netherlands. The ICU admitted surgical, cardiac, cardiac surgery, neurosurgical, medical, neurological and trauma patients. The bedside nurses, using the confusion assessment method for the ICU (CAM-ICU), performed delirium screening three times a day. Patients were considered delirious if they had one positive CAM-ICU during their ICU stay. They categorised delirium as hypoactive, hyperactive or mixed, which has features of both hyperactive and hypoactive. They excluded patients with a sustained Richmond Agitation Sedation Scale of −4/−5; if they were delirious before admission, or if the nurses screened for delirium less than <80% of the time.


Delirium incidence was 26% and increased to 53% for patients who spent >2 days in the ICU. The mixed subtype was most common with an incidence rate of 53%. Patients with a mixed subtype had the most short-term adverse consequences. Hyperactive delirium was greatest in the cardiac surgery group, which had the shortest duration of delirium and fewer short-term consequences. The incidence of hypoactive delirium was greatest in neurology and neurosurgical patients. After adjusting for covariates, delirium was significantly associated with duration of mechanical ventilation, ICU and hospital length of stay and in-hospital mortality.


This study highlights important differences in the incidence of delirium between ICU patient subgroups. The incidence of delirium varied according to patient subgroup, with 64% of neurology, 40% of medical, 12% of cardiac surgery and 10% of neurosurgical patients developing delirium. A total of 135 patients who were delirious before admission, 134 comatose patients and 88 patients who were missing CAM-ICU scores were excluded. It would be interesting to see which subgroup of diagnoses these patients belonged to. Is it possible that these patients were over-represented in one of the patient subgroups, such as the neurosurgical patients, where nurses were unable to perform the CAM-ICU?

Importantly, this study confirms the adverse outcomes associated with ICU delirium demonstrated in prior studies.1 ,2 The incidence of delirium doubled for patients who spent >2 days in the ICU, suggesting that delirium may be preventable. This finding is important because prior studies have demonstrated that delirium duration is associated with adverse outcomes including 1 year mortality.3 In this study, the mixed subtype of delirium was most common, similar to a previously published cohort.4 Unlike the previous report, age was not independently associated with delirium subtype. Patients with a mixed subtype had a longer duration of delirium and a greater number of short-term consequences than those with hypoactive or hyperactive delirium. This is important because mixed delirium subtype has features of both hyperactive and hypoactive delirium, and it is possible that hyperactive patients receive psychoactive medications that convert them to a hypoactive delirium. In essence, agitation has been treated, but not delirium. This question has not been addressed in the literature, but one study demonstrated that psychoactive medications were associated with delirium persistence.5 Further studies on delirium prevention and the impact of reducing delirium duration on outcomes are needed.

View Abstract


  • Competing interests None.

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.