Portal de Periódicos da CAPES

Delirium in critically ill patients: current knowledge and future perspectives

2019; Elsevier BV; Volume: 19; Issue: 12 Linguagem: Inglês

10.1016/j.bjae.2019.09.004

2058-5357

Mark van den Boogaard, A. J. C. Slooter,

Anesthesia and Neurotoxicity Research

Key points•Delirium occurs frequently in critically ill patients and is associated with serious consequences.•The aetiology of delirium is heterogeneous and multifactorial.•Non-pharmacological multicomponent interventions seem to be most effective at preventing delirium.•Treatment of delirium should focus on the treatment of underlying conditions and the symptoms of delirium.Learning objectivesBy reading this article, you should be able to: •Recall the features and different subtypes of delirium in the critically ill patient.•Assess delirium in the ICU and explain its consequences.•Analyse risk factors and predictors for delirium.•Evaluate the ways that delirium in the ICU can be best prevented and treated. •Delirium occurs frequently in critically ill patients and is associated with serious consequences.•The aetiology of delirium is heterogeneous and multifactorial.•Non-pharmacological multicomponent interventions seem to be most effective at preventing delirium.•Treatment of delirium should focus on the treatment of underlying conditions and the symptoms of delirium. By reading this article, you should be able to: •Recall the features and different subtypes of delirium in the critically ill patient.•Assess delirium in the ICU and explain its consequences.•Analyse risk factors and predictors for delirium.•Evaluate the ways that delirium in the ICU can be best prevented and treated. Delirium is derived from the Latin word Lira meaning 'track' or 'trail', and therefore, 'de-lirium' can be best translated in terms of 'derailment' or 'getting off track'. Delirium is an acute disorder of attention and awareness. In addition, there are disturbances in cognition and consciousness, but the level of arousal is not severely compromised as in coma. Delirium is not solely caused by a pre-existing neurocognitive disorder, but is caused by another medical condition.1American Psychiatric AssociationDiagnostic and statistical manual of mental disorders (DSM-5). American Psychiatric Association, Arlington, VA2013Crossref Google Scholar Delirium is the manifestation of an acute encephalopathy, and is expressed as a continuum of symptoms and signs. Incorrectly, the term delirium may be used dichotomously, as if these symptoms and signs are either present or absent, ignoring more subtle features. In patients suffering from some but not all features of delirium, the term subsyndromal delirium is being used. The aetiology of delirium is very heterogeneous, although the presentation is rather homogeneous. Delirium affects many patients who have been exposed to a variety of both predisposing and precipitating risk factors. Therefore, it is usually impossible to assign one specific cause for delirium.2Ely E.W. Gautam S. Margolin R. et al.The impact of delirium in the intensive care unit on hospital length of stay.Intensive Care Med. 2001; 27: 1892-1900Crossref PubMed Scopus (793) Google Scholar Delirium occurs, with an average incidence rate of 29% during an ICU stay.3Rood P. Huisman-de Waal G. Vermeulen H. Schoonhoven L. Pickkers P. van den Boogaard M. Effect of organisational factors on the variation in incidence of delirium in intensive care unit patients: a systematic review and meta-regression analysis.Aust Crit Care. 2018; 31: 180-187Abstract Full Text Full Text PDF PubMed Scopus (30) Google Scholar Half of these cases become apparent within the first 2 days after admission to the ICU. The duration of delirium varies strongly between patients with a median duration of 2–3 days (inter-quartile range: 2–7 days). In exceptional cases delirium may persist for weeks or months.4Kiely D.K. Jones R.N. Bergmann M.A. Murphy K.M. Orav E.J. Marcantonio E.R. Association between delirium resolution and functional recovery among newly admitted postacute facility patients.J Gerontol A Biol Sci Med Sci. 2006; 61: 204-208Crossref PubMed Scopus (81) Google Scholar In rare cases, delirium never resolves.4Kiely D.K. Jones R.N. Bergmann M.A. Murphy K.M. Orav E.J. Marcantonio E.R. Association between delirium resolution and functional recovery among newly admitted postacute facility patients.J Gerontol A Biol Sci Med Sci. 2006; 61: 204-208Crossref PubMed Scopus (81) Google Scholar Multiple factors increase the risk of delirium. Pre-existing factors that increase a patient's vulnerability are termed predisposing risk factors. Factors that trigger the onset of delirium are classified as precipitating risk factors. Both predisposing and precipitating risk factors interact; the more or the stronger predisposing factors are present, the fewer or weaker precipitating factors are required to develop delirium, and vice versa. For instance, an older patient with symptoms of cognitive impairment can develop delirium with only mild hypoxia, or a urinary tract infection, whilst a young, healthy person can only develop delirium in case of serious illnesses, such as sepsis or multiple trauma (Fig. 1). The established predisposing risk factors are increased age, cognitive impairment, and pre-existing hypertension.5Zaal I.J. Devlin J.W. Peelen L.M. Slooter A.J. A systematic review of risk factors for delirium in the ICU.Crit Care Med. 2015; 43: 40-47Crossref PubMed Scopus (319) Google Scholar Other suggested predisposing risk factors are nicotine use, alcohol use, increased ASA grade, and cardiac disease, although the literature on these factors is scarce.5Zaal I.J. Devlin J.W. Peelen L.M. Slooter A.J. A systematic review of risk factors for delirium in the ICU.Crit Care Med. 2015; 43: 40-47Crossref PubMed Scopus (319) Google Scholar The triggering or precipitating factors can be classified in three domains:(i)Acute illness-related factors, such as previous coma; increased severity of illness, mostly expressed in the Acute Physiology and Chronic Health Evaluation (APACHE) score; multiple trauma; sepsis; need for ventilatory support; pain; and systemic hypoperfusion with metabolic acidosis(ii)Medication-related factors: benzodiazepines increase the risk in a dose-dependent manner; anticholinergic drugs, opioids, and corticosteroids are probably associated with the development of delirium, although some studies contradict this.5Zaal I.J. Devlin J.W. Peelen L.M. Slooter A.J. A systematic review of risk factors for delirium in the ICU.Crit Care Med. 2015; 43: 40-47Crossref PubMed Scopus (319) Google Scholar(iii)Environmental factors, such as increased noise, lack of daylight, and admission to a ward (compared with a personal room) increase the risk of delirium.6Van Rompaey B. Elseviers M.M. Schuurmans M.J. Shortridge-Baggett L.M. Truijen S. Bossaert L. Risk factors for delirium in intensive care patients: a prospective cohort study.Crit Care. 2009; 13: R77Crossref PubMed Scopus (316) Google Scholar It is important to note that most triggering factors are modifiable, whilst predisposing risk factors are all non-modifiable. Despite homogenous presentation and the possibility of a final common pathway, the pathophysiology of delirium is unclear. Neuroinflammation, an aberrant stress response, imbalance of neurotransmitters, and alterations in neural networks have all been considered as main underlying hypotheses for the pathophysiology of delirium..7Maldonado J.R. Neuropathogenesis of delirium: review of current etiologic theories and common pathways.Am J Geriatr Psychiatry. 2013; 21: 1190-1222Abstract Full Text Full Text PDF PubMed Scopus (354) Google Scholar Critical illness is usually associated with an inflammatory response, for example, trauma, complicated surgery, or sepsis. Peripheral pro-inflammatory cytokine signals transmitted to the brain can lead to neuroinflammation that persist for months. Moreover, the systemic inflammatory response may diminish as a result of sepsis. Many other factors play a role in delirium in sepsis, such as reduced cerebral perfusion pressure, ischaemia caused by systemic hypotension, hypoxemia and microcirculatory alterations (including endothelial dysfunction).8Ebersoldt M. Sharshar T. Annane D. Sepsis-associated delirium.Intensive Care Med. 2007; 33: 941-950Crossref PubMed Scopus (196) Google Scholar These factors contribute to the reduction in ceebral blood flow that has been observed during delirium.9Shioiri A. Kurumaji A. Takeuchi T. Matsuda H. Arai H. Nishikawa T. White matter abnormalities as a risk factor for postoperative delirium revealed by diffusion tensor imaging.Am J Geriatr Psychiatry. 2010; 18: 743-753Abstract Full Text Full Text PDF PubMed Scopus (50) Google Scholar The aberrant stress response hypothesis addresses the adverse effects of acute stress. Several stressors, such as surgery, systemic inflammation, and pain, cause the brain to activate the limbic–hypothalamic–pituitary–adrenal axis with associated increased concentrations of cortisol. In healthy individuals, this response is adaptive and has adequate feedback regulation. Cognitive decline and ageing are associated with impaired feedback regulation of the stress response pathway, resulting in sustained high cortisol concentrations that contribute to the development of delirium. It is further presumed that delirium is associated with reduced activity of neurones that communicate with the neurotransmitter acetylcholine.10Hshieh T.T. Fong T.G. Marcantonio E.R. Inouye S.K. Cholinergic deficiency hypothesis in delirium: a synthesis of current evidence.J Gerontol A Biol Sci Med Sci. 2008; 63: 764-772Crossref PubMed Scopus (345) Google Scholar Acetylcholine plays a central role in attention and consciousness; it is particularly affected in delirium as there are interactions with other pathways such as the dopaminergic system. In addition, delirium may be associated with the production of a random and loose brain network, which means that there is reduced concomitant activity of brain regions.11van Dellen E. van der Kooi A.W. Numan T. et al.Decreased functional connectivity and disturbed directionality of information flow in the electroencephalography of intensive care unit patients with delirium after cardiac surgery.Anesthesiology. 2014; 121: 328-335Crossref PubMed Scopus (64) Google Scholar As delirium is multifactorial, it is assumed that the interaction of different pathways leads to disruption of large-scale neural networks, and that this disruption causes attention deficits, a decreased level of consciousness, and other features of delirium. Studies with preoperative brain MRIs found cerebral atrophy, white matter hyperintensities, and cortical infarcts were risk factors for postoperative delirium.9Shioiri A. Kurumaji A. Takeuchi T. Matsuda H. Arai H. Nishikawa T. White matter abnormalities as a risk factor for postoperative delirium revealed by diffusion tensor imaging.Am J Geriatr Psychiatry. 2010; 18: 743-753Abstract Full Text Full Text PDF PubMed Scopus (50) Google Scholar,12Omiya H. Yoshitani K. Yamada N. et al.Preoperative brain magnetic resonance imaging and postoperative delirium after off-pump coronary artery bypass grafting: a prospective cohort study.Can J Anaesth. 2015; 62: 595-602Crossref PubMed Scopus (27) Google Scholar These structural changes affect neural network function and may thus predispose to delirium. Delirium has an acute onset, and occurs within a few hours or days after the onset of the underlying illness. The features of delirium are changes in consciousness level and inattention, which manifest as difficulty to sustain attention. However, distraction by irrelevant stimuli may occur as well. The sleep–wake rhythm is often disturbed, resulting in sleeplessness. Another frequent sign is disorientation, particularly in time and place. The subtypes of delirium are categorised according to psychomotor alterations into hypoactive, hyperactive, and mixed motor subtypes, of which the mixed subtype is the most frequent type.13Peterson J.F. Pun B.T. Dittus R.S. et al.Delirium and its motoric subtypes: a study of 614 critically ill patients.J Am Geriatr Soc. 2006; 54: 479-484Crossref PubMed Scopus (471) Google Scholar The hypoactive subtype is particularly underdiagnosed and is characterised by reduced alertness, motor activity, and speech. Patients with hypoactive delirium are often misdiagnosed as having depression, although depression is not characterised by a decreased level of consciousness. Conversely, the hyperactive subtype is characterised by restlessness, increased and inappropriate motor activity, and sometimes agitation. This form of delirium is commonly caused by withdrawal of alcohol or other drugs; it is rarely observed in patients in the ICU, possibly because of the administration of sedative drugs. The remaining patients have the mixed motor subtype, and alternate between the hyper- and hypoactive type. Importantly, patients with ICU-acquired weakness may not be able to show motor hyperactivity with their limbs. In these patients, delirium may be underdiagnosed and undertreated. Perceptual disturbances (i.e. hallucinations and delusions) are often thought to be present in delirium, but they seem to occur in fewer than 40% of cases and are not limited to hyperactive delirium. The severity and symptoms of delirium may fluctuate during the day, typically with restlessness in the evening and at night ('sundowning'), and daytime sleepiness. The management of delirium comprises diagnosis, prediction, prevention, and treatment. The diagnosis of delirium is not difficult in critically ill patients who develop a slightly altered level of consciousness with disturbed attention in the course of hours to a few days. However, despite this, delirium is often underdiagnosed, and around 60–70% of cases are missed by ICU nurses and physicians.14van Eijk M.M. van Marum R.J. Klijn I. de Wit N. Kesecioglu J. Slooter A.J.C. Comparison of delirium assessment tools in a mixed intensive care unit.Crit Care Med. 2009; 37: 1881-1885Crossref PubMed Scopus (244) Google Scholar Lack of recognition may be because delirium is so common that it is not regarded as being unusual. Several tools have been developed to improve the detection of delirium, of which the Confusion Assessment Method for the ICU (CAM-ICU) and the Intensive Care Delirium Screening Checklist (ICDSC) are the most commonly used (Table 1).15Devlin J.W. Skrobik Y. Gelinas C. et al.Clinical practice guidelines for the prevention and management of pain, agitation/sedation, delirium, immobility, and sleep disruption in adult patients in the ICU.Crit Care Med. 2018; 46: e825-e873Crossref PubMed Scopus (1276) Google Scholar Whilst the CAM-ICU is a brief assessment based on formal testing, the ICDSC is based on observations during a nursing shift. Guidelines recommend the use of one of these instruments, and their sensitivity in research settings is adequate (80% for the CAM-ICU and 74% for the ICDSC).15Devlin J.W. Skrobik Y. Gelinas C. et al.Clinical practice guidelines for the prevention and management of pain, agitation/sedation, delirium, immobility, and sleep disruption in adult patients in the ICU.Crit Care Med. 2018; 46: e825-e873Crossref PubMed Scopus (1276) Google Scholar,16Gusmao-Flores D. Salluh J.I. Chalhub R.A. Quarantini L.C. The confusion assessment method for the intensive care unit (CAM-ICU) and intensive care delirium screening checklist (ICDSC) for the diagnosis of delirium: a systematic review and meta-analysis of clinical studies.Crit Care. 2012; 16: R115Crossref PubMed Scopus (306) Google Scholar However, in routine clinical practice, the sensitivity of the CAM-ICU appears to be lower than this (47%), whilst the ICDSC has not yet been evaluated.17van Eijk M.M. van den Boogaard M. van Marum R.J. et al.Routine use of the confusion assessment method for the intensive care unit: a multicenter study.Am J Respir Crit Care Med. 2011; 184: 340-344Crossref PubMed Scopus (197) Google ScholarTable 1The Confusion Assessment Method for ICU (CAM-ICU) and the Intensive Care Delirium Screening Checklist (ICDSC). RASS; Richmond Agitation-Sedation ScaleCAM-ICUICDSCCriteriaPositive score ifCriteriaYESNO1. Acute onset or fluctuating courseOne or both features present1. Altered level of consciousnessYES: RASS score other than zeroNO: RASS = 0 or recent sedative use102. InattentionRead the following series of 10 letters and let patient squeeze on the letter 'A'S A V E A H A A R T†More than two errors2. InattentionDifficulty in following a conversation or instructions. Easily distracted by external stimuli. Difficulty in shifting focuses. If present103. Altered level of consciousnessBased on the RASS scoreRASS score other than zero3. DisorientationDisoriented in name, place and/or date104. Disorganized thinkingAsk the following questions:- Will a stone float on water?- Are there fish in the sea?- Does one pound weigh more than two pounds?- Can you use a hammer to pound a nail?CommandSay to patient: "Hold up this many fingers" (hold two fingers in front of the patient), "Now do the same with the other hand" (without repeating the number of fingers)More than 1 mistake on the questions or commands4. Hallucination, delusion or psychosisAsk if present105. Psychomotor agitation or retardationEither hyperactive, requiring sedatives or restraints, or hypoactive106. Inappropriate speech or moodIf present107. Sleep-wake cycle disturbanceEither frequent awakening/ <4 hours of sleep or sleeping during most of the day108. Symptom fluctuationFluctuation of any of the above symptoms over a 24 hour period10Positive CAM-ICU: Criterion 1 plus 2 and either criterion 3 or 4No delirium: 0 pointsSubsyndromal delirium: 1-3 pointsDelirium: 4-8 pointsThe CAM-ICU can only be administered when the RASS score < -3.† If the patient has a neuromuscular disease and squeezing is impossible, eye blinks can be used.The Intensive Care Delirium Screening Checklist (ICDSC). It is only possible to assess the ICDSC if the RASS score < -3. The first four criteria are based on a bedside assessment, the other four on observations throughout the entire shift.† If the patient has a neuromuscular disease and squeezing is impossible, eye blinks can be used. Open table in a new tab A key feature of delirium is inattention, which can be easily assessed using the attention screening examination that is part of the CAM-ICU. The patient is instructed to squeeze the hand of the assessor, or close the eyes, when he or she hears the letter 'A' out of a series of 10 letters. Disturbed attention during this test is defined as more than two errors. Other important features of delirium, such as disorientation, emotional problems, and hallucinations, can be assessed with closed questions in patients who cannot communicate verbally. Not all 'confused' patients are, in fact, delirious. A characteristic of delirium is a slightly decreased level of consciousness. Other patients who appear to be confused have a normal level of consciousness, such as patients with isolated sensory dysphasia, dementia, or a psychosis resulting from other conditions. Importantly, restlessness and agitation do not always indicate delirium, and can also be observed in patients with pain, itching, or constipation. Akathisia is an important but rare relevant condition, defined as an urge to move, and which may be an adverse effect of neuroleptic drugs. In assessing patients for delirium, it is important to look for symptoms and signs of 'delirium mimics', and for specific causes of delirium that may require a specific treatment. Caution is needed in patients with focal neurological signs, an atypical presentation (e.g. psychosis in a patient with normal attention and consciousness), an atypical onset (either very acute or chronic), and a paucity of delirium risk factors. Agitation with dilated pupils, clonus, hyperthermia, and hyperactive bowel sounds may indicate a serotonin syndrome. Hyperthermia is also a prominent feature of neuroleptic malignant syndrome, with extreme rigidity and increased concentrations of serum creatine kinase. In typical cases, the diagnosis of delirium can be made without imaging or neurophysiological tests. CT or MRI of the brain is required in patients with new focal signs or other features that are uncommon in delirium and that suggest a structural cause. EEG changes during delirium include diffuse slowing of background activity. In addition, EEG may show electrographic seizures and periodic epileptiform discharges in some patients with delirium, particularly in the context of sepsis, and especially in patients with subtle convulsive signs, such as nystagmus and chewing movements. EEG is a very sensitive but less specific technique for detecting delirium.18van der Kooi A.W. Leijten F.S. van der Wekken R.J. Slooter A.J. What are the opportunities for EEG-based monitoring of delirium in the ICU?.J Neuropsychiatry Clin Neurosci. 2012; 24: 472-477Crossref PubMed Scopus (31) Google Scholar Future studies are needed to define the best approach to detection and diagnosis. Several models are available to help predict the onset of delirium. The Early Prediction of Delirium ICU (E-PRE-DELIRIC) tool predicts delirium at the time of admission to the ICU, and is correct in 68% of cases. This model consists of nine predictive factors: age, history of cognitive impairment, history of alcohol abuse, ICU admission category, urgent admission, mean arterial BP, use of corticosteroids, respiratory failure, and serum urea concentration at ICU admission. The second model, PRE-DELIRIC, predicts delirium 24 h after admission to the ICU, and consists of 10 predictors: age, APACHE II score, coma, ICU admission category, infection, presence of metabolic acidosis, use of morphine, use of a sedative drug, urea concentration, and urgent admission. Although the PRE-DELIRIC model is a significantly better predictor, clinicians found the E-PRE-DELIRIC more feasible.19Wassenaar A. Schoonhoven L. Devlin J.W. et al.External validation of two models to predict delirium in critically ill adults using either the Confusion Assessment Method-ICU or the Intensive Care Delirium Screening Checklist for delirium assessment.Crit Care Med. 2019; 47: e827-e835Crossref Scopus (13) Google Scholar Recently, a new prediction model has been developed to predict mental status the following day. Importantly, ICU physicians' and nurses' estimation of a patient's risk is significantly less accurate (correct prediction of area under the curve [AUC]: 0.59; 95% confidence interval [CI]: 0.49–0.70) compared to the PRE-DELIRIC model (correct prediction of AUC: 0.84; 95% CI: 0.82–0.87), indicating that it would be better to use an ICU delirium prediction model in daily practice. Non-pharmacological, multicomponent interventions seem to be effective in patients not in the ICU but need to be further studied in the ICU. As patients in the ICU are exposed to numerous risk factors for delirium, a multifactorial approach (instead of a single intervention) seems to be appropriate.15Devlin J.W. Skrobik Y. Gelinas C. et al.Clinical practice guidelines for the prevention and management of pain, agitation/sedation, delirium, immobility, and sleep disruption in adult patients in the ICU.Crit Care Med. 2018; 46: e825-e873Crossref PubMed Scopus (1276) Google Scholar This may include early mobilisation, use of earplugs, reductions of night-time light and noise, increased exposure to daylight, and a reorientation programme with cognitive training.15Devlin J.W. Skrobik Y. Gelinas C. et al.Clinical practice guidelines for the prevention and management of pain, agitation/sedation, delirium, immobility, and sleep disruption in adult patients in the ICU.Crit Care Med. 2018; 46: e825-e873Crossref PubMed Scopus (1276) Google Scholar Studies showing beneficial effects of single interventions were mostly not well designed or were underpowered, whereas well-designed and well-powered studies showed no beneficial effect.20Bannon L. McGaughey J. Verghis R. Clarke M. McAuley D.F. Blackwood B. The effectiveness of non-pharmacological interventions in reducing the incidence and duration of delirium in critically ill patients: a systematic review and meta-analysis.Intensive Care Med. 2019; 45: 1-12Crossref Scopus (35) Google Scholar Studies of multicomponent interventions focus on cognitive impairment (as reorientation, cognitive stimulation, music, and use of clocks), sedation/sleep impairment (reducing sedation, and minimising external light and noise), immobility (early mobilisation), and hearing and visual impairment (encouraging the use of hearing aids and glasses). Overall, the use of such strategies reduces delirium. Another multi-intervention approach is the use of the awakening and breathing coordination, delirium monitoring/management, and early exercise/mobility/family engagement bundle, which has been significantly associated with decreased incidence of delirium and reduced hospital mortality. Although all studies seem promising, the evidence is limited and not supported by well-designed and well-powered studies.20Bannon L. McGaughey J. Verghis R. Clarke M. McAuley D.F. Blackwood B. The effectiveness of non-pharmacological interventions in reducing the incidence and duration of delirium in critically ill patients: a systematic review and meta-analysis.Intensive Care Med. 2019; 45: 1-12Crossref Scopus (35) Google Scholar However, the use of this bundle of non-pharmacological interventions is recommended by the recently updated recently updated guideline (Table 2) of the Society of Critical Care Medicine (SCCM) on Pain, Agitation/Sedation, Delirium, Immobility, and Sleep Disruption in Adult Patients in the ICU (PADIS).15Devlin J.W. Skrobik Y. Gelinas C. et al.Clinical practice guidelines for the prevention and management of pain, agitation/sedation, delirium, immobility, and sleep disruption in adult patients in the ICU.Crit Care Med. 2018; 46: e825-e873Crossref PubMed Scopus (1276) Google ScholarTable 2Delirium statement derived from the PADIS guideline. 15Devlin J.W. Skrobik Y. Gelinas C. et al.Clinical practice guidelines for the prevention and management of pain, agitation/sedation, delirium, immobility, and sleep disruption in adult patients in the ICU.Crit Care Med. 2018; 46: e825-e873Crossref PubMed Scopus (1276) Google Scholar HMG-CoA, 5-hydroxy-3-methylglutaryl-coenzyme A.Delirium monitoring(i)Critically ill adults should be regularly assessed for delirium using a valid tool (good practice statement).(ii)The level of arousal may influence delirium assessments with a validated screening tool (ungraded statement).Outcomes associated with delirium(i)Strong association with long-term cognitive impairment and may be with a longer hospital stay(ii)Not associated with post-traumatic stress disorder or post-ICU distress(iii)Not associated with ICU length of stay, discharge disposition to a place other than home, depression, functionality/dependence, or mortalityPharmacological delirium prevention(i)It is suggested not to use haloperidol, an atypical antipsychotic, dexmedetomidine, an HMG-CoA reductase inhibitor (i.e. statin), or ketamine to prevent delirium in all critically ill adults (conditional recommendation: very low to low quality of evidence).Delirium treatment(i)Regarding subsyndromal delirium, it is suggested not to use haloperidol or an atypical antipsychotic (conditional recommendation: very low to low quality of evidence).(ii)Regarding delirium, it is suggested not to routinely use haloperidol, an atypical antipsychotic, or an HMG-CoA reductase inhibitor (i.e. a statin) to treat delirium (conditional recommendation: low quality of evidence).(iii)It is recommended to use dexmedetomidine for delirium treatment in mechanically ventilated patients, in which agitation is precluding weaning or extubation (conditional recommendation: low quality of evidence).Non-pharmacological delirium prevention and treatment(i)It is suggested not to use bright light therapy as single intervention to prevent delirium (conditional recommendation: moderate quality of evidence).(ii)It is suggested to use a multicomponent, non-pharmacological intervention focusing at (but not limited to) reducing modifiable risk factors; improving cognition; and optimising sleep, mobility, hearing, and vision in critically ill adults (conditional recommendation: low quality of evidence).Multicomponent interventions include (but are not limited to) strategies to reduce or shorten delirium (e.g. reorientation, cognitive stimulation, and use of clocks), improve sleep (e.g. minimising light and noise), improve wakefulness (i.e. reduced sedation), reduce immobility (e.g. early rehabilitation/mobilisation), and reduce hearing or visual impairment (e.g. enable the use of devices, such as hearing aids or eyeglasses). Open table in a new tab Small RCTs in patients that underwent operation and in patients in the ICU with low disease severity have suggested a beneficial effect of prophylactic haloperidol, but a large-scale RCT, an RCT on both prophylaxis and early treatment for delirium, and an RCT to prevent delirium in patients with subsyndromal delirium in the ICU all showed no beneficial effect.21van den Boogaard M. Slooter A.J.C. Bruggemann R.J.M. et al.Effect of haloperidol on survival among critically ill adults with a high risk of delirium: the REDUCE randomized clinical trial.JAMA. 2018; 319: 680-690Crossref PubMed Scopus (156) Google Scholar,22Al-Qadheeb N.S. Skrobik Y. Schumaker G. et al.Preventing ICU subsyndromal delirium conversion to delirium with low-dose IV haloperidol: a double-blind, placebo-controlled pilot study.Crit Care Med. 2016; 44: 583-591Crossref PubMed Scopus (72) Google Scholar It is therefore unlikely that delirium in patients in the ICU can be prevented with haloperidol. Conversely, promising preventive effects have been found with dexmedetomidine, a highly selective α2-receptor agonist in reducing the incidence of delirium.15Devlin J.W. Skrobik Y. Gelinas C. et al.Clinical practice guidelines for the prevention and management of pain, agitation/sedation, delirium, immobility, and sleep disruption in adult patients in the ICU.Crit Care Med. 2018; 46: e825-e873Crossref PubMed Scopus (1276) Google Scholar There is no evidence that statins or cholinesterase inhibitors can prevent delirium in the ICU. Interestingly, the melatonin receptor agonist ramelteon had an exceptionally strong effect on the prevention of delirium in a small RCT in non-intubated patients in the ICU or the acute ward.23Hatta K. Kishi Y. Wada K. et al.Preventive effects of ramelteon on delirium: a randomized placebo-controlled trial.JAMA Psychiatry. 2014; 71: 397-403Crossref PubMed Scopus (238) Google Scholar When reducing delirium, it is important to minimise the effects of pharmacological risk factors that may induce or prolong ICU delirium. Benzodiazepines seem to be particularly associated with the development of delirium in the critically ill. The risk of delirium increases by 4% with every 5 mg equivalent dose of midazolam, with a plateau for doses of ≥150 mg in patients who are awake.24Zaal I.J. Devlin J.W. Hazelbag M. et al.Benzodiazepine-associated delirium in critically ill adults.Intensive Care Med. 2015; 41: 2130-2137Crossref PubMed Scopus (139) Google Scholar The evidence regarding opioids is contradictory. As undertreated severe pain is a presumed risk factor for delirium, adequate pain management is important. Non-opioid analgesia may be provided, for example, with ketamine, but this may also lead to hallucinations and delirium, and the routine use of ketamine is not recommended. There is sparse and conflicting evidence on the effects of drugs with anticholinergic effects and corticosteroids, and therefore, it is not possible to make strong recommendations. As dexmedetomidine has sedative and analgesic effects, it may be an alternative drug to both benzodiazepines and opioids. Haloperidol is commonly prescribed for the treatment of delirium. However, there is no evidence that haloperidol reduces the duration of delirium, and the SCCM guideline on delirium recommends that haloperidol should be restricted to treatment of ICU delirium with agitation and psychosis.15Devlin J.W. Skrobik Y. Gelinas C. et al.Clinical practice guidelines for the prevention and management of pain, agitation/sedation, delirium, immobility, and sleep disruption in adult patients in the ICU.Crit Care Med. 2018; 46: e825-e873Crossref PubMed Scopus (1276) Google Scholar Recently, a large RCT comparing haloperidol and ziprasidone with a placebo in delirious patients in the ICU with acute respiratory failure or shock showed no difference in delirium days between the three groups.25Girard T.D. Exline M.C. Carson S.S. et al.Haloperidol and ziprasidone for treatment of delirium in critical illness.N Engl J Med. 2019; 380: 1778-1779Crossref Scopus (1) Google Scholar This confirms the recommendation to restrict the use of haloperidol for symptomatic treatment in delirium. There is limited evidence that atypical antipsychotics have fewer adverse effects than haloperidol, but atypical antipsychotics cannot be administered intravenously. Rigidity is a common adverse effect, whilst akathisia, dystonia, and prolongation of the QTc interval are rare. Cholinesterase inhibitors are not recommended to treat delirium in all a patients. Psychotic features include hallucinations and delusions. These features are not limited to hyperactive patients with delirium, but can also occur in hypoactive delirium. There are no studies that have specifically focused on treatment of psychosis in ICU delirium. Based on studies on other psychotic illnesses, such as schizophrenia, and clinical experience, treatment with antipsychotics is warranted. The preferred agent for treatment of psychosis in delirium is haloperidol because of its rapid effects and the possibility of i.v. administration (0.5–5 mg repeated three times, depending on age). The first step in the management of agitated patients is to evaluate and treat underlying causes, such as pain, itching, constipation, or bladder retention. Further, reassurance and implementation of non-pharmacological measures, such as relaxing music, are warranted. In cases that agitation needs to be treated pharmacologically, dexmedetomidine and clonidine may be superior to haloperidol. Delirium may be associated with anxiety that does not respond sufficiently to non-pharmacological measures, such as reassurance and music. In those cases where anxiety is extreme, it is unclear whether benzodiazepines should be started: they have anxiolytic properties and could prolong delirium.24Zaal I.J. Devlin J.W. Hazelbag M. et al.Benzodiazepine-associated delirium in critically ill adults.Intensive Care Med. 2015; 41: 2130-2137Crossref PubMed Scopus (139) Google Scholar It should be noted that anxiety in delirium may be caused by hallucinations and delusions, but these are difficult to assess in patients who cannot communicate verbally and who do not respond to closed questions. Importantly, α2-agonists, such as dexmedetomidine and clonidine, have anxiolytic properties as well. Benzodiazepines are often prescribed to promote sleep. However, as stated earlier, benzodiazepines may induce delirium, and although they may promote light sleep, they suppress deep sleep and rapid-eye-movement sleep, and, therefore, recovery. Patients receiving antipsychotic treatment may be switched to quetiapine (25–50 mg at night). It has sedative effects and does not induce natural sleep. Theoretically, α2-agonist drugs may promote sleep, but this treatment has not yet been studied extensively. A hyperadrenergic state presents with hypertension, tachycardia, spontaneous hyperventilation, and sweating, and is particularly common in patients with substance withdrawal, such as withdrawal of alcohol. Benzodiazepines are often prescribed in these cases. However, evidence that benzodiazepines are beneficial in alcohol withdrawal is limited and based on older literature. As benzodiazepines also increase the risk of ICU delirium, it is currently unclear whether benzodiazepines are superior to other approaches in patients with a hyperadrenergic state, such as α2-agonists and anticonvulsant drugs. Delirium can be very distressing for the patient and for family members and caregivers. Delirium may impair outcome because of the increased frequency of self-extubation, removal of catheters, (micro) aspiration, and muscle weakness from immobility. Delirium is further consistently associated with prolonged durations of mechanical ventilation, ICU and hospital admission, and therefore higher medical costs.15Devlin J.W. Skrobik Y. Gelinas C. et al.Clinical practice guidelines for the prevention and management of pain, agitation/sedation, delirium, immobility, and sleep disruption in adult patients in the ICU.Crit Care Med. 2018; 46: e825-e873Crossref PubMed Scopus (1276) Google Scholar The higher death rate in patients with delirium compared with those without delirium does not appear to result from severity of underlying illness and is thus an indirect effect.26Klein Klouwenberg P.M. Zaal I.J. Spitoni C. et al.The attributable mortality of delirium in critically ill patients: prospective cohort study.BMJ. 2014; 349: g6652Crossref PubMed Scopus (135) Google Scholar Besides the serious short-term consequences, delirium is also associated with long-term impaired cognitive functioning.15Devlin J.W. Skrobik Y. Gelinas C. et al.Clinical practice guidelines for the prevention and management of pain, agitation/sedation, delirium, immobility, and sleep disruption in adult patients in the ICU.Crit Care Med. 2018; 46: e825-e873Crossref PubMed Scopus (1276) Google Scholar This association may be causal because of prolonged neuro-inflammation. In addition to cognitive problems, delirium was found to be associated with more problems in activities of daily living. Delirium in the ICU seems not to be associated with long-term anxiety, depression, or post-traumatic stress disorder, but many ICU survivors can vividly remember their delirium experience during the ICU stay even after many years. Current investigations aim to develop a detection tool for delirium that is both accurate and easy to use. Another important area of research is focused on the underlying substrate, with the objective to identify factors that predispose an individual to develop delirium. A better understanding of the pathophysiology is crucial to develop rational interventions to prevent and treat this important problem. MvdB declares no conflicts of interest. AJCS is an advisor for Prolira, a start-up company that develops an EEG-based delirium monitor. Any (future) profits from EEG-based delirium monitoring will be used for future scientific research only. The associated MCQs (to support CME/CPD activity) will be accessible at www.bjaed.org/cme/home by subscribers to BJA Education. Mark van den Boogaard RN PhD is a senior researcher at the ICU of the Radboud UMC with special interests in ICU delirium, its prediction and prevention, and long-term outcomes after admission to ICU. He is a board member of the European Delirium Association (EDA), and was a member of the delirium section of the Society of Critical Care Medicine (SCCM) 2018 PADIS guideline committee. Arjen Slooter MD PhD is a professor in intensive care neuropsychiatry at UMC Utrecht Brain Center. His focus is on delirium and on neuropsychiatric outcome after ICU admission. He works on an EEG-based device to detect delirium, functional connectivity in delirium, drug adverse effects and delirium, postoperative delirium, and long-term outcome after admission to ICU. He is President of the EDA and was chairman of the delirium section of the SCCM 2018 PADIS guideline committee.