Delirium Preventionand Treatment

Yoanna Skrobik, MD, FRCP(C)a,b

KEYWORDS

� Delirum � Critical care � Prevention � Pharmacology� Non-pharmacologic management

Delirium is characterized by an acute change or fluctuation in mental status, inatten-tion, and either disorganized thinking or an alteration in level of consciousness. Itoccurs in 35% to 80% of critically ill hospitalized patients. The variability in deliriumrates described in the surgical or medical critically ill depends partly on the severityof illness and partly on the type of instrument used to screen for delirium.

Brain dysfunction (delirium and coma) in nonneurologic intensive care unit (ICU)patients has been the subject of increased study in recent years; coma and deliriumappear to be independent predictors of longer hospital stay, higher hospital costs,and higher mortality.1–3 Delirium may be associated with prolonged cognitive impair-ment, impaired activities of daily living, and decreased quality of life in survivors of crit-ical illness.

DEFINITIONS AND CATEGORIES OF DELIRIUM

Delirium can be diagnosed in ICU settings by psychiatrists and trained nonpsychiatricpersonnel. It can be detected in mechanically ventilated and other nonverbal patientsusing validated instruments such as the Confusion Assessment Method for the ICU4

and the Intensive Care Delirium Screening Checklist (ICDSC).5 Verbal communicationlimits descriptions of specific symptoms in intubated patients. Agitation and slowingoccur frequently (in more than 90% of patients), and specific symptoms may bemarkers of prognosis.6

SUBSYNDROMAL DELIRIUM

Subsyndromal delirium occurs when patients have one or more delirium symptomsbut do not meet criteria for full-blown clinical delirium.7 This clinical syndrome iswell described in the geriatric literature. In the ICU, its identification is possible withthe ICDSC, a graded scale at the patient’s bedside, with clinical criteria rated fromzero to eight.7 Delirious patients have four clinical abnormalities or more. Patientswith no abnormalities (ie, an ICDSC of 0) are considered cognitively normal. Those

a Department of Medicine (critical care), Universit�e de Montr�eal, Montr�eal, Qu�ebec, Canadab Intensive care unit, Hopital Maisonneuve Rosemont, Montr�eal, Qu�ebec, CanadaE-mail address: skrobik@sympatico.ca

Crit Care Clin 25 (2009) 585–591doi:10.1016/j.ccc.2009.05.003 criticalcare.theclinics.com0749-0704/09/$ – see front matter ª 2009 Elsevier Inc. All rights reserved.

Skrobik586

with ICDSC ratings of one to three items are labeled as having ‘‘subsyndromaldelirium.’’7 Subsyndromal delirium affects roughly one-third of the critically ill, as re-ported in the only paper published to date on the subject. This incidence, whencombined with the number of patients considered delirious with the ICDSC, adds toa combined 70% incidence of delirium-like cognitive abnormalities. It is possiblethat the discrepancies in incidence descriptions for delirium (35% by some and80% by others) are related to separate identification or confounding of these twosyndromes. Whether subsyndromal delirium constitutes a graded step in the spec-trum of brain dysfunction severity (from normal to subsyndromal delirium to delirium)or not is unclear. Although subsyndromal delirium resembles delirium in that it is asso-ciated with longer hospital stay and a higher probability of dependent living uponhospital discharge, it does not have the same risk factors and is probably preventable.

PREVENTION

Delirium has untoward consequences, on the one hand, and potentially preventableassociated factors, on the other. Studies addressing delirium prevention and prophy-laxis can be separated into those evaluating nonpharmacologic, pharmacologic, andcombined (pharmacologic and nonpharmacologic) approaches.

Nonpharmacologic Delirium Prevention

Controversy exists as to whether factors associated with delirium are causative. Non-pharmacologic intervention aims to prevent or reverse these potential contributors.Although no studies have been published to date on nonpharmacologic deliriumprevention in the ICU, two prevention strategy studies in non-ICU patients deservemention. The first8 to successfully demonstrate the effectiveness of nonpharmaco-logic intervention targeted specific aspects of care in a high-risk medical geriatric pop-ulation. Treated patients underwent systematic orientation; therapeutic activitiesdesigned to lessen cognitive impairment; early mobilization; nonpharmacologic mini-mization of psychoactive drug use; prevention of sleep deprivation; enhancement ofcommunication, including provision of eyeglasses and hearing aids; and early inter-vention for volume depletion.8 The incidence of delirium was 9.9% with this interven-tion compared with 15.0% in the group receiving usual care (odds ratio [OR], 0.60;95% confidence interval [CI], 0.39, 0.92).8 In those patients who did develop delirium,exposure to ‘‘preventive’’ measures did not change the severity or the duration ofdelirium or any of the complications associated with it.8

The second nonpharmacologic intervention study,9 the first randomized clinical trial,allocated hip fracture patients to ‘‘standard care’’ versus care in which treating physi-cians requested a routine geriatric consultation. The consultation incorporated stan-dardized recommendations targeting 10 domains: systemic oxygenation monitoring,as a surrogate for oxygen delivery to the brain; fluid and electrolyte balance; painmanagement; minimization of psychoactive drug use; optimal bowel and bladderfunction; nutrition; early mobilization; prevention of postoperative complications;appropriate environmental stimuli; and treatment of delirium symptoms.9 A 77%adherence to the geriatric consultant’s recommendations was achieved in this study,and the total cumulative incidence of delirium during hospitalization was 32% in theproactive geriatric-consultation group versus 50% in the ‘‘usual care’’ group (OR,0.48 [95% CI, 0.23, 0.98]; relative risk [RR], 0.64 [95% CI, 0.37, 0.98]).9

One study documented the incidence of delirium in 49 patients randomized toearly physiotherapy and occupational therapy versus 55 patients randomized toroutine care.10 There were fewer days of delirium in the ICU [2.0 (0.0, 6.0) vs 4.0

Delirium Prevention and Treatment 587

(2.0, 7.0); P 5 .034] and in the hospital [2.0 (0.0, 6.0) vs 4.0 (2.0, 8.0); P 5 .017] in theearly mobilization group.10 Likewise, there were a lower percentage of ICU days withdelirium [33.0% (0.0%, 58.0%) vs 57.0% (33.0%, 69.0%); P 5 .015] and percentage ofhospital days with delirium [25.0% (0.0%, 43.0%) vs 39.0% (22.0%, 62.0%); P 5 .009]in the early mobilization group.10 Risk factors for delirium between groups were similar.

The features that link these studies are careful and systematic assessments ofnon–life-threatening patient characteristics, medications, and needs. Most of theinterventions focus on dimensions that are not part of the culture of most ICUs.Some interventions, for instance prevention of sleep deprivation, may be challengingto implement in an ICU where sleep is abnormal in the majority of patients.11 Further, itis a secondary consideration in the context of caring for acute emergencies. Otherinterventions, such as early mobilization, have been shown to be of benefit but arenonetheless not yet part of routine practice.

Until ICU-based delirium prevention studies are available, it seems reasonable toimplement patient-focused care, and to broaden this perspective to include reorienta-tion, communication, mobilization, and minimization of pharmacologic exposure. Atthe very least, it is difficult to see how such approaches, when administered bywell-educated health care professionals, could place the patient at risk.

Pharmacologic Delirium Prevention

Aizawa and colleagues12 hypothesized that sleep disturbances are critical factors inthe etiology of postoperative delirium. In this study, patients were admitted to theICU after abdominal surgery and were then randomized to pharmacologic sleep-wake cycle adjustment or to conventional care. The ‘‘Delirium Free Protocol’’ involvednightly routine administration of intramuscular diazepam at 20:00 hours and intrave-nous 8-hour long infusions of flunitrazepam and meperidine. The incidence of deliriumin the 7 days after surgery was significantly lower in the intervention group (5% vs thecontrols’ 35%) (OR, 0.10 [95% CI, 0.01, 0.89]; RR, 0.14 [95% CI, 0.02, 1.06]).12

However, the protocol caused some sedation upon waking, and this may have inter-fered with delirium assessment.

There are no trials evaluating antipsychotic drugs or anticholinesterase inhibitors fordelirium prevention in the ICU. Outside the ICU, a trial evaluating prophylactic haloper-idol was not effective in preventing delirium but did reduce its severity and duration.13

A second study that compared donepezil with placebo reported an incidence ofdelirium of 18.8% after surgery in treatment and placebo groups (RR, 1.2 [95% CI,0.48, 3.00]).14

Dexmedetomidine has been compared with midazolam in the sedation of critically illpatients.15 The prevalence of delirium during sedative treatment was strikingly lower indexmedetomidine-treated patients: 54% (n 5 132/244) versus 76.6% (n 5 93/122) inmidazolam-treated patients ([95% CI, 14%–33%]; P<.001).15 Dexmedetomidine-treated patients were more likely to develop bradycardia (42.2% [103/244] vs18.9% [23/122]; P<.001).15 It can be speculated that dexmedetomidine, a potentcentral alpha-2-receptor antagonist, which is described for the treatment of alcoholwithdrawal, has a potential role to play in delirium prevention or treatment. Conversely,the benzodiazepines used in the nondexmedetomidine group may have contributed tothe increased incidence of delirium.16

Combined Pharmacologic and Nonpharmacologic Prevention

The use of sedatives16,17 and analgesics18 has been linked to delirium, particularly inthe context of excessive sedation. The author’s group implemented and compared, ina PRE-POST fashion, a symptom-driven protocol in a single tertiary ICU.19 They

Skrobik588

routinely distinguished among the clinical features of pain, agitation, and delirium andprovided protocolized pharmacologic and nonpharmacologic approaches that wereindividualized to specific symptoms. After the institution of this protocol (POST),more patients remained cognitively intact (32.5% PRE vs 41.2% POST; P 5 .004).The rate of delirium was similar (34.7% PRE vs 34.2% POST; P 5 .9), but the numberof patients manifesting subsyndromal (ie, subclinical) delirium was significantly less inPOST. There was no difference in antipsychotic use between the PRE and POSTgroups.

Combined pharmacologic and nonpharmacologic prevention strategies are chal-lenging and onerous to implement in any inpatient unit. Problems with adherence toprotocols have been described, and they likely limit the effectiveness. Further studiesare needed but, of necessity, are likely to be limited in scope and in number. Risk strat-ification may aid in identifying patients most likely to benefit from multimodal interven-tions. Strategies to improve intervention implementation and adherence, such asshared ‘‘ownership’’ and interactive education, may also serve to improve interdisci-plinary protocols. Such combined approaches are likely to be an integral part of anydelirium intervention package.

ALCOHOLIC PATIENTS

Alcoholic patients deserve special mention. About 1 in 10 North Americans purport-edly consumes excess alcohol and is therefore at risk for alcohol withdrawal. In addi-tion, alcoholism contributes to as many as 21% of admissions to ICU.20,21 Validatedquestionnaires (eg, CAGE questionnaires or Clinical Institute Withdrawal Assessmentfor Alcohol Scales) are not routinely administered in the critical care setting. Alco-holism doubles the incidence of delirium without necessarily developing into alcoholwithdrawal syndrome.22,23 Screening for alcohol consumption should be part of everyICU admission case history, whenever feasible.

DELIRIUM TREATMENTNonpharmacologic

Addressing delirium management in the ICU requires routine screening for its pres-ence. Validated pedagogic interventions targeting nurses are useful in establishingreliable delirium screening.24 Early delirium recognition and management in themedical and surgical ICU setting may aid in delirium resolution and shorten the lengthof stay. Although interventions focused on increasing exposure to daylight, avoidinguse of restraints, and boosting family contact are endorsed by intensivists and othercaregivers, none of these interventions have been rigorously evaluated.25

PharmacologicDelirium is distressing for patients, families, and caregivers. Pharmacologic sedationof the agitated or frightened patient is therefore broadly perceived as desirable.Although no double-blind, randomized, placebo-controlled trial has ever establishedthe efficacy or safety of any antipsychotic medication in the management of delirium,administration of antipsychotics is endorsed by guideline recommendations, and it ispart of routine clinical practice for the majority of intensive care specialists.26,27

Any discussion of antipsychotic medication as a category is complicated by thevariety and the differences in the receptor-adherence profiles characteristic toeach.28,29 All conventional and atypical antipsychotics appear to be equally effica-cious in the treatment of psychosis, and at present there is no evidence of differentialeffects on delirium. Two studies evaluating antipsychotic use in the ICU have been

Delirium Prevention and Treatment 589

published. In the first,30 delirious patients able to tolerate enteral nutrition wererandomized to receive enteral olanzapine or intravenous haloperidol. Both groups ofpatients improved with time in their delirium severity score. Benzodiazepine require-ments, which reflected the need for sedation, decreased equally in both groups.30

Patients in the olanzapine arm had less extrapyramidal side effects than thosereceiving haloperidol.30 In the second study,31 30 delirious patients able to tolerateenteral nutrition were randomized to receive quetiapine (50 mg every 12 h) or placeboin addition to as-needed intravenous haloperidol. Patients receiving quetiapineachieved a ‘‘nondelirium’’ score (ie, Intensive Care Delirium Checklist score <4) faster,had a shorter ICU stay (54 vs 138 hours, P 5 .007), spent less time agitated (6 vs 36hours, P 5 .03), and required a shorter treatment duration. It is not clear whether thesefindings are generalizable, as the studies are limited by their inclusion of patients ableto tolerate enteral drugs, thereby limiting the generalizability of the findings.

There are no intravenous second-generation antipsychotics available for ICU use.The use of other sedatives, such as benzodiazepines, has generated much scientificand academic interest over the last 10 years; however, the focus has been the titrationof sedation and the disadvantages of its excess rather than optimal therapy for agita-tion. Well-established sedation scales, such as the Richmond Agitation and SedationScale,32 are better validated for sedation than for agitation. Optimal pharmacologictreatment of delirium in the ICU thus remains to be established.

THE SPECIFIC CASE OF DELIRIUM ANDALCOHOLWITHDRAWAL

Pharmacologic interventions shown to be beneficial in the prevention of alcohol with-drawal have been best described with the administration of sedatives.33 Althoughproviding alcohol to patients stratified to be at high risk for alcohol withdrawal seemseffective, the heterogeneity in the design of the studies and the potential risks ofalcohol administration preclude the recommendation of this practice.21,34 Benzodiaz-epines, major tranquilizers, and central alpha-blockers, alone or in combination, arethe cornerstone of conventional alcohol withdrawal symptom management. The addi-tion of phenobarbital and propofol, in addition to titrated benzodiazepines, not onlybenefits patients but also reduces the necessity for intubation in patients admittedto the ICU with alcohol withdrawal syndrome (AWS).35 Titrating sedative drugs topatient need benefits the critically ill.36,37 For alcoholic patients, this is particularlytrue; however, screening for patients at risk and use of an appropriate alcohol with-drawal scale (AWS or Clinical Institute Withdrawal Assessment) is essential. In thecontext of careful titration, administering additional sedation early in the course ofalcohol withdrawal symptoms in ICU patients yields a better outcome.21

SUMMARY

Little is known of nonpharmacologic and pharmacologic delirium prevention and treat-ment in the critical care setting. Trials emphasizing early mobilization suggest that thisnonpharmacologic approach is associated with an improvement in delirium incidence.Titration and reduction of opiate analgesics and sedatives may improve subsyndromaldelirium rates. All critical care caregivers should rigorously screen for alcohol abuse,apply alcohol withdrawal scales in alcoholic patients, and titrate sedative drugsaccordingly. No nonpharmacologic approach or drug has been shown to be beneficialonce delirium is established. Considering the importance and the consequences ofdelirium in the critical care setting, studies to further address prevention and rigoroustrials addressing pharmacologic intervention are urgently needed.

Skrobik590

REFERENCES

1. Ely EW, Shintani A, Truman B, et al. Delirium as a predictor of mortality in mechan-ically ventilated patients in the intensive care unit. JAMA 2004;291:1753–62.

2. Ouimet S, Kavanagh BP, Gottfried SB, et al. Incidence, risk factors and conse-quences of ICU delirium. Intensive Care Med 2007;33:66–73.

3. Thomason JW, Shintani A, Peterson JF, et al. Intensive care unit delirium is anindependent predictor of longer hospital stay: a prospective analysis of 261non-ventilated patients. Crit Care 2005;9:R375–81.

4. Ely EW, Inouye SK, Bernard GR, et al. Delirium in mechanically ventilatedpatients: validity and reliability of the confusion assessment method for the inten-sive care unit (CAM-ICU). JAMA 2001;286:2703–10.

5. Bergeron N, Dubois MJ, Dumont M, et al. Intensive care delirium screening check-list: evaluation of a new screening tool. Intensive Care Med 2001;27:859–64.

6. Marquis F, Ouimet S, Riker R, et al. Individual delirium symptoms: do they matter?Crit Care Med 2007;35:2533–7.

7. Ouimet S, Riker R, Bergeon N, et al. Subsyndromal delirium in the ICU: evidencefor a disease spectrum. Intensive Care Med 2007;33:1007–13.

8. Inouye SK, Bogardus ST Jr, Charpentier PA, et al. A multicomponent interventionto prevent delirium in hospitalized older patients. N Engl J Med 1999;340:669–76.

9. Marcantonio ER, Flacker JM, Wright RJ, et al. Reducing delirium after hip fracture:a randomized trial. J Am Geriatr Soc 2001;49:516–22.

10. Schweickert WD, Pohlman MC, Pawlik A, et al. The impact of early mobilization onICU delirium in mechanically ventilated (MV) patients. Lancet 2009;373:1874–82.

11. Cooper AB, Thornley KS, Young GB, et al. Sleep in critically ill patients requiringmechanical ventilation. Chest 2000;117:809–18.

12. Aizawa K, Kanai T, Saikawa Y, et al. A novel approach to the prevention of post-operative delirium in the elderly after gastrointestinal surgery. Surg Today 2002;32:310–4.

13. Kalisvaart KJ, de Jonghe JF, Bogaards MJ, et al. Haloperidol prophylaxis forelderly hip-surgery patients at risk for delirium: a randomized placebo-controlledstudy. J Am Geriatr Soc 2005;53:1658–66.

14. Liptzin B, Laki A, Garb JL, et al. Donepezil in the prevention and treatment ofpost-surgical delirium. Am J Geriatr Psychiatry 2005;13:1100–6.

15. Riker RR, Shehabi Y, Bokesch PM, et al. Dexmedetomidine vs midazolam forsedation of critically ill patients: a randomized trial. JAMA 2009;301:489–99.

16. Pandharipande P, Shintani A, Peterson J, et al. Lorazepam is an independent riskfactor for transitioning to delirium in intensive care unit patients. Anesthesiology2006;104:21–6.

17. Pandharipande P, Cotton B, Shintani A, et al. Prevalence and risk factors fordevelopment of delirium in surgical and trauma intensive care unit patients.J Trauma 2008;65:34–41.

18. Marcantonio ER, Juarez G, Goldman L, et al. The relationship of postoperativedelirium with psychoactive medications. JAMA 1994;272:1518–22.

19. Ouimet, et al. Delirium prevention strategies: targeting and improving outcomes.ICM 2006;32(Suppl 1):S141.

20. de Wit M, Wan SY, Gill S, et al. Prevalence and impact of alcohol and other druguse disorders on sedation and mechanical ventilation: a retrospective study.BMC Anesthesiol 2007;7:3.

21. Moss M, Burnham EL. Alcohol abuse in the critically ill patient. Lancet 2006;368:2231–42.

Delirium Prevention and Treatment 591

22. Blondell RD, Powell GE, Dodds HN, et al. Admission characteristics of traumapatients in whom delirium develops. Am J Surg 2004;187:332–7.

23. Marcantonio ER, Goldman L, Mangione CM, et al. A clinical prediction rule fordelirium after elective noncardiac surgery. JAMA 1994;271:134–9.

24. Devlin JW, Marquis F, Riker RR, et al. Combined didactic and scenario-basededucation improves the ability of intensive care unit staff to recognize deliriumat the bedside. Crit Care 2008;12:R19.

25. Cheung CZ, Alibhai SM, Robinson M, et al. Recognition and labeling of deliriumsymptoms by intensivists: does it matter? Intensive Care Med 2008;34:437–46.

26. Jacobi J, Fraser GL, Coursin DB, et al. Clinical practice guidelines for the sus-tained use of sedatives and analgesics in the critically ill adult. Crit Care Med2002;30:119–41.

27. Seitz DP, Gill SS, van Zyl LT. Antipsychotics in the treatment of delirium: a system-atic review. J Clin Psychiatry 2007;68:11–21.

28. Kapur S, Seeman P. Antipsychotic agents differ in how fast they come off thedopamine D2 receptors. Implications for atypical antipsychotic action. J Psychi-atry Neurosci 2000;25:161–6.

29. Nordstrom AL, Farde L, Wiesel FA, et al. Central D2-dopamine receptor occu-pancy in relation to antipsychotic drug effects: a double-blind PET study ofschizophrenic patients. Biol Psychiatry 1993;33:227–35.

30. Skrobik YK, Bergeron N, Dumont M, et al. Olanzapine vs haloperidol: treatingdelirium in a critical care setting. Intensive Care Med 2004;30:444–9.

31. Devlin, et al. Crit Care Med 2008;36(12):A17.32. Ely EW, Truman B, Shintani A, et al. Monitoring sedation status over time in ICU

patients: reliability and validity of the Richmond Agitation-Sedation Scale(RASS). JAMA 2003;289:2983–91.

33. Spies CD, Otter HE, Huske B, et al. Alcohol withdrawal severity is decreased bysymptom-orientated adjusted bolus therapy in the ICU. Intensive Care Med 2003;29:2230–8.

34. Hodges B, Mazur JE. Intravenous ethanol for the treatment of alcohol withdrawalsyndrome in critically ill patients. Pharmacotherapy 2004;24:1578–85.

35. Gold JA, Rimal B, Nolan A, et al. A strategy of escalating doses of benzodiaze-pines and phenobarbital administration reduces the need for mechanical ventila-tion in delirium tremens. Crit Care Med 2007;35:724–30.

36. Kress JP, Vinayak AG, Levitt J, et al. Daily sedative interruption in mechanicallyventilated patients at risk for coronary artery disease. Crit Care Med 2007;35:365–71.

37. Girard TD, Kress JP, Fuchs BD, et al. Efficacy and safety of a paired sedation andventilator weaning protocol for mechanically ventilated patients in intensive care(awakening and breathing controlled trial): a randomised controlled trial. Lancet2008;371:126–34.