Preventing and treating delirium in clinical settings for older adults

Abstract

Delirium is a serious consequence of many acute or worsening chronic medical conditions, a side effect of medications, and a precipitant of worsening functional and cognitive status in older adults. It is a syndrome characterized by fluctuations in cognition and impaired attention that develops over a short period of time in response to an underlying medical condition, a substance (prescribed, over the counter, or recreational), or substance withdrawal and can be multi-factorial. We present a narrative review of the literature on nonpharmacologic and pharmacologic approaches to prevention and treatment of delirium with a focus on older adults as a vulnerable population. Older adult patients are most at risk due to decreasing physiologic reserves, with delirium rates of up to 80% in critical care settings. Presentation of delirium can be hyperactive, hypoactive, or mixed, making identification and study challenging as patients with hypoactive delirium are less likely to come to attention in an inpatient or long-term care setting. Studies of delirium focus on prevention and treatment with nonpharmacological or medication interventions, with the preponderance of evidence favoring multi-component nonpharmacological approaches to prevention as the most effective. Though use of antipsychotic medication in delirium is common, existing evidence does not support routine use, showing no clear benefit in clinically significant outcome measures and with evidence of harm in some studies. We therefore suggest that antipsychotics be used to treat agitation, psychosis, and distress associated with delirium at the lowest effective doses and shortest possible duration and not be considered a treatment of delirium itself. Future studies may clarify the use of other agents, such as melatonin and melatonin receptor agonists, alpha-2 receptor agonists, and anti-epileptics.

Introduction and pathophysiology

Among older adults, incidence of delirium in studies varies greatly but may be as high as 70%, 60%, and 65% in long-term care, the frail and hospitalized, and postsurgical settings, respectively. ¹ A recent prospective study of geriatric inpatients shows delirium rates of 32%, increasing to 80% in critical care settings. ² This corroborates a 2020 meta-analysis of nine studies investigating delirium prevalence finding a range of 9–32%. ³ Consequently, the potential public health implications of delirium in older adults include a healthcare cost burden estimated to be $38 billion to $152 billion adjusted for inflation (2005 dollars) each year in the United States. ⁴

Delirium itself is a neuropsychiatric syndrome that is associated with fluctuating mental status, abrupt in onset and transient in nature, though cognitive effects can persist. ¹ The Diagnostic and Statistical Manual of Mental Disorders (DSM), fifth edition defines delirium as (a) a disturbance of attention and consciousness; (b) disturbance developing within a short period of time with a change in the usual level of attention and consciousness and fluctuating in severity throughout the day; and (c) other cognitive symptoms such as memory disturbance, disorientation, speech disturbance, disturbances in visuospatial abilities, or perception not better explained by other pre-existing or developing neurocognitive disorders or coma, and with evidence that the disorder is a direct result of somatic disease, substance intoxication or withdrawal, or toxin exposure. ⁵

The pathophysiology of delirium, especially hypoactive delirium, is complex and not fully understood. Pathophysiological causes of delirium are theorized to include neurotransmitter imbalance; delirium is conceptualized as a hypocholinergic, hyperdopaminergic state. ¹ Other theories include electrolyte or metabolic derangements and neuroinflammation. ¹ The SARS-CoV (COVID-19) pandemic has provided further evidence of an inflammation-mediated pathway of delirium. ⁶ Aging, in addition to the neurophysiologic changes mentioned previously, is associated with altered neurotransmitter balance and exaggerated inflammatory responses. ⁷ Pharmacokinetic changes associated with aging include decreased renal and hepatic clearance and increases in volume of distribution with shifts in fat-to-lean muscle ratio resulting in higher elimination half-lives of medications. Pharmacodynamic changes also occur with aging with increased sensitivity to medications. These will not only predispose individuals to becoming delirious from medication effects, but will also guide management. ⁸

Delirium is underrecognized and carries increased risk of mortality, long-term care placement, and chronic cognitive sequelae. ¹ In one observational survey of a hospital consultation-liaison service in Japan, approximately 30% of cases of delirium diagnosed in referred patients were missed, and 14% of delirium cases were misdiagnosed as depression by the referring provider. ⁹ Long-term care residents have increased rates of cognitive and functional impairment, malnutrition, and frailty, which may place them at higher risk for delirium while making delirium more challenging to identify. Other significant risk factors associated with delirium in these settings include dehydration, falls, use of physical restraints, and SARS-CoV-2 infection. In nursing homes, ¹⁰ delirium increases risks of re-hospitalization, mortality, and worse functional outcomes, yet delirium remains understudied in these settings. To address delirium burden in older adults, national guidelines in the US and UK outline risk factors, clinical indicators, prevention strategies, and management recommendations. Specifically, the American Geriatrics Society provides expert recommendations for postoperative delirium and the UK’s NICE guidelines provide a broad overview of delirium identifying individuals over 65 as being a vulnerable group and advocating for screening.^11,12

Methods

For this narrative review, electronic database searches were conducted on PubMed and Google Scholar using different combinations of the search terms ‘delirium’, ‘geriatric’, ‘late life’, ‘elderly’, and ‘encephalopathy’. Abstracts were reviewed and relevant articles pulled for further study. Articles investigating human studies in English language only were included. Key points from these articles are describes in narrative format, grouped by theme. In instances where multiple studies have been published on a single topic area, articles were selectively chosen which provided an appropriate representation of the topic as opposed to a systematic inclusion of all articles. As the aim of this review is to provide an overall narrative review and not to utilize systematic methodology, all studies related to topics highlighted may not be included.

Delirium prevention in older adults

Prevention of delirium is most often accomplished by identifying and addressing the many vulnerabilities that increase risk for delirium. These risk factors can be broken down into predisposing factors, which increase vulnerability to delirium, and precipitating factors, which are the noxious insults that trigger the onset of delirium. In the case of someone with multiple predisposing factors and thus a higher vulnerability, even relatively minimal precipitating factors may be enough to cause delirium. Conversely, someone with relatively few predisposing factors may not develop delirium without exposure to significantly higher levels of precipitating factors. ¹³

Predisposing and precipitating factors

Dementia is one such predisposing factor in the general medical setting [relative risk (RR): 2.3–4.7]. In a comprehensive review of 11 studies consisting of 5166 participants ranging from age 65 to 85 across various settings, the presence of an underlying dementia diagnosis was associated with a two- to fivefold increase in delirium risk. ¹⁴ Dementia is not only a predisposing factor for delirium, but also there seems to be a bidirectional relationship as episodes of delirium have been associated with long-term cognitive impairment. There have been multiple proposed mechanisms including neurotoxicity, inflammation, chronic stress, direct neuronal damage, and acceleration of underlying dementia pathology. ¹⁴ Other predisposing factors include cognitive impairments excluding dementia (RR: 2.1–2.8), functional impairment (RR: 4.0), vison impairment (RR: 2.1–3.5), medical comorbidity or severe illness (RR: 1.3–5.6), depression (RR: 3.2), alcohol abuse (RR: 5.7), and older age (RR: 4.0). ¹³

Precipitating factors for delirium in the general medical setting include: multiple medications started during the admission (RR: 2.9), use of centrally active medications (RR: 4.5), use of physical restraints (RR: 3.2–4.4), bladder catheterization (RR: 2.4), elevated serum urea (RR: 5.1) or elevated blood urea nitrogen to creatinine ratio (RR: 2.0), and adverse events during the hospitalization (RR: 1.9). ¹³

Many agents that affect the central nervous system (CNS) have been associated with delirium, whether prescribed, obtained over the counter, or used recreationally. The most common are those affecting γ-aminobutyric acid, muscarinic acetylcholine, dopamine, or opioid receptors. Many of these are used as sedatives or anxiolytics in the inpatient setting (benzodiazepines, nonbenzodiazepine sedative hypnotics, anticonvulsants, barbiturates). Others, used for a variety of intended purposes, increase anticholinergic (specifically antimuscarinic) burden [antihistamines, antispasmodics, some antiparkinsonian agents, tricyclic antidepressants, second-generation antipsychotics (SGAs)]. ¹⁵ Beers criteria may be used as a reference to avoid prescribing medications likely to cause delirium, as well as other medications considered to have high risk/benefit ratio in older adults. ¹⁶

Likewise, intoxication or withdrawal from recreational substances may present with delirium, with the most concerning scenario being alcohol or sedative-hypnotic withdrawal due to the potential for high morbidity and mortality. Prevention consists of symptom triggered or scheduled treatment with several protocols, with benzodiazepines and barbiturates being the most used agents. This choice of medication use contrasts with the management of other types of delirium, in which benzodiazepines may precipitate or worsen symptoms. ¹⁵ Treatment with IV thiamine is also indicated in the setting of known or suspected heavy alcohol use to prevent the development or exacerbation of Wernicke encephalopathy or progression to Korsakoff dementia. ¹⁷

Nonpharmacological prevention strategies

Prevention strategies can be divided into both nonpharmacological and pharmacological interventions, ¹⁷ and several multi-component nonpharmacological approaches to delirium management focus on prevention. The most well studied are the Hospital Elder Life Program (HELP) and the ABCDEF bundle. HELP is an approach that uses a multidisciplinary team and trained volunteers to target delirium risk factors to prevent functional and cognitive decline in older persons. The interventions include reorientation, early mobilization, hydration, nutrition, sleep, and hearing and/or vision adaptation. ¹⁸ A 2015 meta-analysis found that 11 studies with 4267 patients demonstrated that multi-component nonpharmacological delirium interventions reduced delirium incidence (odds ratio: 0.47; 95% CI: 0.38–0.58) with a number needed to treat of 14.3 (95% CI: 11.1–20.0). When limited to four randomized or matched trials with 977 intervention patients, delirium incidence was decreased by 44% (RR: 0.56; 95% CI: 0.42–0.76) with an NNT of 20.0 (95% CI: 12.5–33.3). ¹⁹ The ABCDEF bundle is a guide to coordinate care in the intensive care unit (ICU) that involves assessment and prevention of pain, spontaneous breathing trials, selection of analgesia and sedation, identification and management of delirium, early mobilization, and family engagement. ²⁰ As such, it is a practical implementation of the Society of Critical Care Medicine’s clinical practice guideline regarding prevention of delirium. ²¹ Two large studies showed increased use of the bundle improves delirium outcomes.^22,23 Early mobilization when combined with interruption of sedation has also been shown to decrease duration of delirium in ICU patients. ²⁴

Early identification of developing delirium is often part of multi-component nonpharmacological interventions. Nursing initiatives can play a key role in identification as nurses are often best positioned to identify components of delirium such as fluctuation throughout the day and changes from baseline mental status. Nevertheless, the evidence for nursing initiatives for early identification of delirium is uncertain. In one study, a nursing screening initiative was shown to result in increased Confusion Assessment Method (CAM) screening and odds of delirium decreased by 25.3% among high-risk patients. ²⁵ In another study conducted in the ICU, no changes in delirium-free or coma-free days were observed with addition of nursing interventions. ²⁶ Nursing initiatives for screening and prevention of delirium have the most support for use in the context of multi-component bundles.

Pharmacological prevention strategies

In contrast to the success of nonpharmacological approaches, pharmacological interventions have proven less helpful in preventing delirium. Several pharmacological strategies to prevent delirium in high-risk settings have been studied, with antipsychotics, cholinesterase inhibitors, alpha-2 agonists, and melatonin receptor agonists (including melatonin itself) among those with the most available evidence. The outcomes are similar for surgical, critically ill, and medically hospitalized patients, with no clear benefit for the use of pharmacotherapy to prevent delirium.

Surgical patients

Clinical trials investigating cholinesterase inhibitors in surgical patients, including rivastigmine and donepezil, have not demonstrated a reduction in the incidence of delirium after elective total major joint-replacement surgery,^27,28 surgery for hip fracture, ¹⁵ or elective cardiopulmonary bypass surgery. ²⁹ A 2013 systematic review and meta-analysis of six studies of prophylactic use of antipsychotics versus placebo in surgical patients suggested that SGAs were more beneficial than placebo for preventing the incidence of delirium but did not decrease severity of delirium or change the rates of adverse events. ³⁰ In contrast melatonin, ³¹ and the melatonin receptor agonist ramelteon ³² have shown efficacy in some but not all studies in preventing delirium in postoperative patients. Dexmedetomidine has shown mixed results in preventing delirium when used as sedation. In surgical patients, several studies have shown decreased delirium incidence^33–35 with dexmedetomidine while others have not found this association. ³⁶

Critically ill patients

Prophylactic use of low-dose antipsychotic medications has also been studied in critical care settings with inconsistent results. In critically ill patients, the Haloperidol Effectiveness in ICU Delirium (HOPE-ICU) and Prophylactic Haloperidol Use for Delirium in ICU Patients at High Risk for Delirium (REDUCE) randomized placebo-controlled trials both found no difference in survival or delirium incidence with prophylactic haloperidol use.^37,38 A 2019 Cochrane review found that in critically ill patients dexmedetomidine may shorten delirium duration, although noted that this was based on a single study. ³⁹

General medical patients

In medically hospitalized patients aged 70 or older at high risk of delirium, prophylactic haloperidol did not reduce delirium incidence in the placebo-controlled randomized controlled HARPOON study. ⁴⁰ Studies have shown decreased delirium incidence with use of melatonin receptor agonists in medically hospitalized patients;^41,42 however, multiple other studies have shown no reduction in delirium incidence with these agents.^43–45 Several meta-analyses and systematic reviews also show conflicting results of melatonin agonist efficacy.^46–50 Despite evidence in some studies for the use of melatonin agonists, current evidence does not support using these or other pharmacologic strategies to prevent delirium. ⁵⁰

Identification and treatment of delirium in older adults

Prompt identification of delirium is crucial in hospitalized and postoperative populations. The most used clinical instrument to identify delirium is the CAM which is easily administered at the bedside and has been validated in multiple settings.^51,52 CAM involves identifying an acute onset and fluctuating course as well as evidence of inattention. In addition, one of either disorganized thinking or an altered level of consciousness must be present. The CAM has a sensitivity of 84–100% and specificity of 90–95%. ⁵¹ The CAM-ICU is an adaptation of the CAM for mechanically ventilated patients who are unable to speak which has been validated for use in the ICU. ⁵³ Another tool for screening for delirium in the ICU is the Intensive Care Delirium Screening Checklist (ICDSC), which uses a screening checklist of DSM criteria including altered level of consciousness, inattention, disorientation, hallucination or delusion, psychomotor agitation or retardation, inappropriate mood or speech, sleep/wake cycle disturbance, and symptom fluctuation. ⁵⁴ The ICDSC has been compared to the CAM-ICU in critically ill patients with high agreement rates. ⁵⁵

Despite protocols and guidelines with evidence for prevention of delirium,^18,19,56,57 in many cases, patients develop delirium despite these efforts, making it important to understand options for treatment of delirium. In some instances, the interventions used to prevent delirium are also successful at treating delirium. For example, one study found assessing nonpharmacological risk factors such as immobility, functional decline, visual or hearing impairment, dehydration, and sleep deprivation are effective for both delirium prevention and treatment. ¹⁷ Here, we will review the evidence available for a variety of medications that have been commonly used to treat delirium, or for which there is some theoretical rationale for their use.

Antipsychotics

Antipsychotics are commonly used to treat delirium in older adults ⁵⁸ despite the United States Food and Drug Administration black box warning of increased risk of death in older adults with dementia-related psychosis. ⁵⁹ The efficacy of antipsychotics for the treatment of delirium in any age group is unclear, with most meta-analyses pointing to no significant benefit over placebo;^{39,58,60–62} however, heterogeneity of medications, study populations, and measured endpoints are important complicating factors. In a study by Nikooie et al., no differences emerged in delirium severity or cognitive functioning for haloperidol versus SGA use. This study also found insufficient or no evidence for the use of antipsychotics to treat delirium versus placebo, and the authors conclude that evidence does not support routine use of haloperidol or SGAs to treat delirium in adult inpatients. Another meta-analysis by Burry et al. ³⁹ found antipsychotics did not reduce delirium symptoms compared to non-antipsychotics or placebo. In two meta-analyses studying the use of haloperidol in critically ill patients, no benefit was found for any measured endpoint.^61,62 Conversely, in another analysis of antipsychotics in delirium, the authors conclude their use may worsen symptoms of delirium. ⁶³ The populations studied in these analyses are heterogenous and thus applicability of the conclusions to the specific population of hospitalized older adults is of uncertain validity. Additionally, in two of the studies,^39,61 the comparator was placebo or non-antipsychotic medication, with the rationale that no medication had been proven superior to placebo.

Two meta-analyses draw the conclusion that specific antipsychotic medications or classes may have utility in the treatment of delirium. ^{64 65} In a network meta-analysis (NMA), Wu et al. concluded that haloperidol alone and haloperidol plus an initial dose of lorazepam were superior to placebo; however, only one study included in the analysis used lorazepam plus haloperidol, and the diversity of patient populations and study parameters in the included studies are likely to interfere with the ability to perform a valid NMA. Kishi et al. concluded in their meta-analysis of haloperidol and SGAs that antipsychotics in general are superior to placebo or usual care and that SGAs are associated with shorter time to recovery and lower incidence of extrapyramidal symptoms compared to haloperidol.

Overall, the weight of the evidence is against antipsychotics providing clinically significant benefit compared to placebo or usual care. Future trials can address differences in SGAs in treatment of delirium, as limited data were published on individual agents and receptor profiles vary considerably within the class. For example, no RCTs examine the use of aripiprazole despite case reports and case series providing evidence for aripiprazole use in delirium in older adults. ⁶⁶

Melatonin and melatonin agonists

Guidance on the use of melatonin and melatonin agonists for prevention of delirium^41,67,68 is limited by a lack of published trials to support or refute their utility in the treatment of delirium in older adults. One study shows efficacy in the setting of delirium in older adults when the etiology is due to organophosphate poisoning. ⁶⁹ The theoretical basis for the involvement of melatonin in delirium is illustrated by studies of postoperative older adults who were found to have abnormal melatonin secretion ⁷⁰ and mechanically ventilated patients with low measured melatonin metabolites. ⁷¹ Recently published study designs aim to clarify a possible role for melatonin in the treatment of delirium in older adults,^72,73 one of which has initial data published demonstrating feasibility, so more data are expected in the near future.

Cholinesterase inhibitors

The neurotransmitter acetylcholine is thought to play a key role in the pathogenesis of delirium, and anticholinergic medications are among the most implicated in precipitation of iatrogenic delirium. ⁷⁴ Despite this, no current evidence from controlled trials support efficacy of agents that increase acetylcholine in the treatment of delirium. In two studies, donepezil specifically was found to have no effect on incidence, severity or duration of delirium, nor adverse events related to delirium.^46,75 One randomized controlled trial (RCT) comparing rivastigmine with placebo in patients who were also prescribed haloperidol was discontinued early due to higher mortality in the rivastigmine group, and it was incidentally found that median duration of delirium was longer in the rivastigmine group, though this did not achieve statistical significance. ⁷⁶ Additionally, one small RCT evaluating citicoline (a precursor of acetylcholine) found no significant difference in delirium outcomes. ⁷⁷

Alpha-2-agonists

Alpha-2 agonists have been evaluated as possible treatments for delirium and agitation related to delirium given anxiolytic and sedative properties. Dexmedetomidine is a selective, centrally acting alpha-2-agonist often used in the ICU for sedation. Data reporting the effects of dexmedetomidine on delirium are mixed, however, overall favor a reduction in delirium. ⁷⁵ When compared to midazolam and lorazepam in use for sedation in critically ill patients, dexmedetomidine is associated with lower prevalence, incidence, and shorter duration of delirium.^78,79 This improved performance may be due to lack of midazolam or lorazepam, as benzodiazepines are known be deliriogenic, ⁸⁰ though some evidence suggests that this effect is related to a difference in how alpha-2-agonists act on sleep. ⁸¹ Guanfacine is a centrally acting alpha-2 agonist that is available in an oral formulation, in contrast to dexmedetomidine, which until recently has only been available for intravenous use and limited to the intensive care setting. A small case series of seven patients explored the use of guanfacine as an adjunctive medication in the management of agitation associated with delirium, and it was found to be both safe and effective though more formal, wider scale investigations are needed to draw generalizable conclusions. ⁸² Clonidine has also been investigated as an agent to shorten the course of delirium, though one small (total n = 20) study found no effect. ⁸³

Anti-seizure medications

Limited studies investigate the role of anti-seizure medications in treatment and prevention of delirium. In one meta-analysis evaluating pharmacologic interventions for prevention of delirium in hospitalized patients above the age of 65, no clear evidence of beneficial effect was found with anti-seizure medications. ⁷⁵ Gabapentin has been evaluated in multiple RCTs for prevention of postoperative delirium, and in one pilot study was found to reduce the incidence of postoperative delirium, though in other trials was not found to reduce rates, severity, or duration of delirium. ⁸⁴ Valproate has been of interest particularly as adjunctive treatment for delirium and for use in critically ill patients. Some preliminary studies have shown reduction of agitation, delirium, and/or concomitant psychoactive medication use,^85–89 though it is ultimately difficult to draw definitive conclusions due to small sample sizes, observational study design, and confounding variables such as patient factors influencing decision to give valproate, natural symptom course, and differential contributions of concurrent treatments.

Benzodiazepines

Benzodiazepines are strongly associated with the precipitation of delirium and published guidelines recommend against the use of benzodiazepines as first-line agents for the treatment of delirium outside of certain circumstances such as alcohol/sedative-hypnotic withdrawal. ⁹⁰ Numerous studies have shown that benzodiazepines are an independent risk factor for the development of delirium and can exacerbate delirium when used for agitation. ⁸⁰ Multiple proposed mechanisms for this include interference with sleep patterns, disruption of the circadian rhythm and melatonin release, decrease in acetylcholine release, and thalamic dysfunction. ⁸⁰

To conclude the discussion of delirium pharmacotherapy, it is worth emphasizing that delirium is a syndrome and not a discrete phenomenon. The heterogeneity of pathophysiology and various precipitating causes and presentations complicates studies on delirium treatment as a medication or class of medications may target one hypothesized cause but not others. For example, a subset of patients with delirium primarily mediated by increase in dopamine signaling may benefit from treatment with antipsychotics; however, there is currently no way to separate these patients out from the population of delirious patients as a whole. Thus, medications that may be helpful to some patients show no separation from placebo in studies. Nevertheless, the development of further subtyping, perhaps biomarker identification of more discrete different deliria, there is reason to hope that targeted pharmacotherapy will become a useful tool.

Special topics in delirium treatment

Medical complexity in management/formulation of delirium

Delirium is a multifactorial syndrome and thus optimal treatment will naturally vary depending on etiology and course. The underlying causes of the changes in mentation and behavior that constitute delirium must be considered when developing both pharmacologic and behavioral management strategies. Here, we will touch only briefly on several special topics in treatment and long-term management, with the knowledge that this listing is neither exhaustive nor comprehensive.

Decreased hepatic function, end-stage renal disease, and cardiac disease

Delirium from hepatic dysfunction (i.e. hepatic encephalopathy – HE) comes with a unique set of challenges and may have complex etiology. The clinical presentation of HE ranges from somewhat insidious, odd behavior to overt alterations in behavior and cognition. When hepatic function is impaired, working with the patient and other physicians/providers to optimize treatment of underlying liver disease can ultimately minimize ammonia and other neurotoxic metabolites.^91,92 If a medicine is necessary to address symptoms, choosing one that will have the least impact on an already stressed system is preferred. Opting for medications that do not undergo initial metabolism in the liver (or, ideally, are not metabolized in the liver at all) and lowering starting doses to account for impaired metabolism can decrease side effect burden while controlling symptoms effectively.

Renal disease can also cause alterations in mentation through a variety of mechanisms including electrolyte derangements and buildup of urea and other renally excreted waste products. Poor excretion of medications such as lithium, gabapentin, duloxetine, and venlafaxine and their metabolites as well as many medications commonly used in the critical care setting, such as morphine and midazolam among others, may also contribute. Consideration of a patient’s renal function when evaluating a delirious patient includes reducing polypharmacy and ensuring that necessary medications are dosed appropriately. ⁹³ If a necessary medication is contraindicated in end-stage renal disease or dialysis, working with the patient and their nephrologist to adjust the dosing schedule may allow the patient to benefit while minimizing side effects.

Cardiac disease may prompt the development of alternative strategies to manage delirium. Some antipsychotic medications can alter the corrected QT interval (QTc) to varying degrees, increasing risk of torsades de pointes and other arrhythmias but are they are usually not the cause of clinically significant increases in QTc. ⁹⁴ Most patients with delirium are already at increased risk of arrhythmias due to their underlying medical conditions or co-occurring cardiac problems unrelated to their hospitalization. Physicians and other providers should balance risks and benefits of using medications to treat delirium while also maximizing nonpharmacologic strategies.

Oncology

Delirium commonly occurs in individuals with cancer, ranging from 10% to 30% of hospitalized patients. ⁹⁵ In cancer care, delirium can be related to direct effects (e.g. primary CNS cancer or disease metastatic to the CNS), or secondary effects (e.g. paraneoplastic syndromes, cancer treatment, ⁹⁶ or complications from cancer treatment such as chemotherapy-associated dehydration and electrolyte abnormalities). Up to half of patients experience delirium in the first month following stem cell transplant.^96,97 Over 80% of patients with cancer will experience delirium at the end of life. ⁹⁵

Management of delirium in cancer populations is similar to the management of delirium in other medical conditions, including identification of the underlying medical cause, addressing modifiable risk factors, and managing symptoms with pharmacotherapy as indicated. ⁹⁸

Palliative care

Delirium is prevalent in palliative care populations. Over half of patients in the inpatient palliative care setting will develop delirium. ⁹⁹ At the end of life, terminal delirium occurs in over 75% of patients. ¹⁰⁰ Symptoms of terminal delirium include non-purposeful movements, changes in consciousness, facial grimacing, moaning, myoclonus, agitation, and anxiety. ¹⁰¹ Changes in respiratory pattern, temperature, and skin mottling are common symptoms at the end of life, and often accompany terminal delirium. ¹⁰² Delirium is a useful prognosticator for end of life and should be considered in conversations with family or in goals of care discussions. ¹⁰²

There is limited clinical guidance on managing delirium at the end of life. ¹⁰¹ Management includes use of antipsychotics for agitation, and low dose, short-acting benzodiazepines for anxiety. ¹⁰² Case reports suggest dexmedetomidine as an alternative for managing agitation at end of life. ¹⁰³ Judicious use of medications like benzodiazepines can be considered to reduce distress, manage agitation, and allow for a more peaceful death. ¹⁰⁴

Prevention of complications of delirium

In a meta-analysis, delirium was associated with long-term cognitive decline, at least 3 months after the acute episode, in both medical and surgical patients. ¹⁰⁵ Delirium is not a marker of pre-existing cognitive impairment but may accelerate decline and those with cognitive impairment are at greater risk of developing delirium. When comparing patients with and without delirium, both patients who were cognitively healthy and those with dementia prior to developing delirium had greater cognitive decline than patients who did not develop delirium. ¹⁰⁵ In patients with pre-existing dementia, delirium is associated with further increases in cognitive decline and increased mortality. ¹⁰⁶ The oldest patients have an eightfold increase in dementia associated with delirium. As existing dementia is frequently not identified prior to acute medical illness or surgery, the patients most vulnerable to delirium may not have pre-existing factors proactively addressed, thus worsening outcomes associated with delirium.

In addition to cognitive decline, patients who develop delirium have more severe sequelae than those who do not, ¹ including increased mortality, longer hospital admissions, and longer time of mechanical ventilation. They are also at increased risk of other complications including urinary tract infections, falls/immobility, pressure ulcers, sleep disturbances, and problems with feeding/nutrition/electrolyte balance. A bidirectional relationship between delirium and frailty in older adults suggests that promptly addressing delirium similarly addresses progression to frailty, with interventions in nutrition, mobility/physical conditioning, activities of daily living, and cognitive rehabilitation. ¹⁰⁷ Patients with delirium are more likely to be discharged to an acute nursing facility versus home and to be institutionalized. ¹⁰⁸

Multidisciplinary nonpharmacologic intervention involving physicians, nurses, physical, occupational and speech therapists, care coordinators, and family members can moderate the risk of complications and worsened outcomes associated with delirium. Recording daily urine output and bowel movements, removal of urinary catheters, instituting a regular toileting regimen, and preventing constipation are all essential components of a delirium prevention strategy. Early mobility in the ICU and medical wards – getting patients into a chair, walking on the unit, and participating in physical therapy can prevent many of the commonly seen complications including pressure ulcers and frailty as well as respiratory complications. Management of light, sound, and sleep-wake cycles with potential inclusion of melatonin can address common sleep issues in hospitalized patients.^20,57

Family and caregiver experiences

In one study of patients with terminal delirium, 70% of families experienced distress related to observing delirium in their relatives. ²⁰ Caregivers are at risk of developing behavioral health complications, including anxiety and post-traumatic sequelae, after caring for patients with delirium; ¹⁰⁹ however, family members who are provided psychoeducation about the diagnosis of delirium find the experience as caregivers less distressing. ¹¹⁰ After educational interventions, families reported reduced distress and increased preparedness to provide care upon discharge. ¹¹¹

Conclusion

Delirium is a serious possible complication of many illnesses and a possible side effect of many medications with significant long-term sequelae. It contributes to longer hospitalizations, more time spent in ICUs, accelerated cognitive decline, and other increased morbidity and mortality. Worsening cognition is both a risk factor for and a consequence of delirium. Older adults are more susceptible to delirium than other age cohorts and more at risk of lingering effects due to the biological effects of aging. Prevention and treatment of delirium in this population is of critical importance.

Despite the significance of delirium, there are few proven strategies for treatment once it develops, and thus interventions aimed at prevention should be prioritized. The HELP and ABCDEF bundle have been shown to prevent delirium as has avoidance of drugs that are known to precipitate it. Despite efforts at prevention, some patients will become delirious, and then nonpharmacological interventions used to prevent delirium may also be used to treat it. Unfortunately, the evidence for efficacy of pharmacologic interventions is weak, with most studies showing little to no benefit over placebo. Antipsychotics are commonly used to treat the symptoms of hyperactive delirium in clinical practice and though there is some evidence in favor of this practice, it is neither robust nor convincing when taken as a whole. There is less evidence available regarding melatonin, melatonin agonists, and alpha-2 agonists though some of the existing evidence is positive.

In short, more evidence supports the prevention of delirium than treatment, and nonpharmacologic interventions are preferred. Medication use is most appropriately targeted at treating symptoms of hyperactive delirium to promote patient safety and minimize distress, using agents that will not exacerbate the delirium. Care must be taken to account for the patient’s co-occurring conditions when devising interventions to avoid worsening outcomes. More studies on both prevention and treatment can help to improve care for all patients, with older adults likely to benefit most substantially because of increased risk and more serious consequences when delirium develops.