Delirium is defined as an acute decline in cognitive functioning and should be considered a medical emergency as it is often the result of a noxious disruption to equilibrium. Delirium is common in the hospitalized older adult, with some studies reporting incidence rates of 29% to 64%.[1] The consequences of delirium are significant and include associations with increased mortality, cognitive and functional decline, falls, and admission to long-term care.[1] When delirium does occur, the medical team and patient’s family should be aware that further support on discharge may be needed given the risk of cognitive and functional decline associated with delirium.

Delirium is considered a quality indicator in the care of hospitalized older patients.[2] A better understanding of the pathophysiology of delirium and some effective strategies for diagnosis, prevention, and management can help clinicians ensure that patients affected by delirium receive the care they need.

Risk factors and pathophysiology
Delirium can arise following one single destabilizing medical event, but more often multiple factors contribute to its initiation and development. In addition, some patients are more vulnerable than others. In one study, Inouye and colleagues identified the most important preadmission risk factors for delirium to be visual impairment, severe illness, cognitive impairment, and volume depletion.[3] The greatest in-hospital risk factors found in another study were the use of physical restraints, malnutrition, polypharmacy, the use of a urinary catheter, and any iatrogenic event.[4] With each additional predisposing factor, the likelihood of delirium grows and should prompt a clinician to consider prevention strategies. 

The pathogenesis behind delirium is not fully understood, but several mechanisms have been postulated. Systemic inflammatory cytokine response and neurotransmitter disruption are the two main mechanisms thought to link multifactorial medical conditions and delirium.[5] During a systemic inflammatory response such as sepsis, proinflammatory cytokines released in the peripheral circulation enter the central nervous system, altering endothelial function, diminishing perfusion, activating microglia, and causing neuronal apoptosis and neurotoxicity. This process is self-propelling and can last for months after the initial event.[5] 

Anticholinergic medications are known to induce or worsen delirium by binding to nicotinic and muscarinic receptors in the brain, modulating cognition and arousal, and lending support to the concept of cholinergic deficiency in delirium.[6] In addition, metabolic abnormalities may lead to decreased acetylcholine synthesis and synaptic transmission. Dopamine, norepinephrine, and serotonin can also play a role in arousal, and their levels are affected by the cholinergic pathway.[6] The function of other neurotransmitters in delirium, such as glutamate, melatonin, and gamma-aminobutyric acid, is less well understood.[6]

The presence of delirium may indicate underlying brain vulnerability and should therefore increase suspicion for an underlying cognitive impairment or dementia, especially if the precipitating insult appears disproportionately minor.[7] The pathophysiological changes described here also suggest that delirium itself can cause permanent neuronal damage, thereby contributing to the risk of cognitive decline.[8] Clinically, it can take up to 12 months for patients to return to their cognitive baseline after an episode of delirium, and some never regain their previous cognitive function.[9] 

Diagnosis
One commonly used diagnostic definition of delirium is based on DSM-5 criteria and requires a new acute disturbance in cognition, fluctuating attention, and alteration of the sleep-wake cycle with changes that are primarily related to an underlying medical cause and not better explained by an evolving dementia.[10] Patients may present with hyperactive delirium, which is associated with agitation and hypervigilance, or they may present with hypoactive delirium, which is associated with drowsiness and apathy. It is not uncommon for patients to have both forms at various times during the course of the same illness. It is particularly easy to miss a patient with hypoactive delirium as they do not call attention to themselves, perhaps explaining why the hypoactive form is associated with a poorer prognosis.[1] 

The diagnosis of delirium requires a patient interview, a physical examination, cognitive testing, and a review of the medical chart and any collateral information. Screening tools are an attractive adjunct to clinical assessment, especially if time is limited. One of the most widely used is the confusion assessment method (CAM), validated in various settings, including intensive care units, emergency departments, and nursing homes, and proven to have high interobserver reliability.[11,12] The abbreviated 3-minute 3D-CAM, consisting of clinical observation and a one-page patient questionnaire, can be used at the bedside.[13] Any assessment should focus on the main features of delirium. A diagnosis is suggested by acute-onset confusion and inattention plus disordered thinking or altered level of consciousness. Many hospitals now include a version of the CAM on nursing assessment flow-sheets, helping physicians to quickly scan the chart for signs of delirium.

Because a multitude of underlying causes may be involved, a broad diagnostic approach is needed and clinicians should keep an open mind about the possibility of more than one contributing medical problem. The workup should include a detailed assessment to search for the precipitant, whether it is an acute medical illness, a change in therapy, or the destabilization of a chronic condition. Causes of physical discomfort such as constipation and urinary retention are common precipitants but are often overlooked. During the physical examination, the clinician should look for evidence of occult infection, volume depletion, abdominal pathology, deep vein thrombosis, and a neurological cause.[1] 

The investigations ordered will depend on the assessment, but generally include an electrocardiogram, a complete blood count with differential, and testing for electrolytes (including calcium, phosphate, and magnesium), liver enzymes, thyroid function, troponin, and vitamin B12. A chest X-ray or urinalysis and culture should be considered if symptoms suggest infection, keeping in mind that elderly patients can fail to manifest typical signs of infection (elevated white blood cell count, fever, or focal symptoms). Neuroimaging is indicated for patients with focal neurological deficits, unexplained confusion, or suspected head trauma.[1] Worsening severity or a prolonged course should prompt a repeat workup for ongoing medical instability, new precipitants, or less common causes of delirium such as encephalitis, rapidly progressive dementia, or seizure.[1] Neuroimaging, lumbar puncture, and electroencephalogram should be considered in these situations.

Nonpharmacological strategies for prevention
Primary prevention is the cornerstone of delirium management and has the best evidence for success of any intervention. Some institutions have adopted the Hospital Elder Life Program (HELP) to address six risk factors in older patients that commonly contribute to delirium: cognitive impairment, sleep deprivation, immobility, visual impairment, hearing impairment, and dehydration (Table 1 [10]).[14] Addressing these risk factors results in fewer days of delirium and fewer episodes of delirium, but does not reduce delirium severity and recurrence rates,[14] suggesting that once delirium develops, it is difficult to intervene. 

A recent meta-analysis has shown that nonpharmacological interventions not only reduce the incidence of delirium but also prevent falls.[15] Other important strategies include managing pain, maintaining nutrition, and performing a thorough medication review.[16] Discontinuation of urinary catheters whenever possible is encouraged given the association of catheters with urinary tract infections. Involving the patient’s family, primary bedside nurse, and clinical nurse leader in the creation of a nursing care plan can also be instrumental in the success of these nonpharmacological delirium prevention strategies. Other important aspects of the care plan include assisted feeding and positioning in bed to prevent aspiration, frequent turning to prevent skin breakdown, and minimizing the use of restraints given the association of restraints with injury and worsened delirium.[17] If restraints must be used, the patient should be supervised vigilantly and the restraints discontinued as soon as possible. 

Given the complex nature of caring for older adults, a comprehensive geriatric assessment (CGA) by a geriatrician or a physician with expertise in geriatric care can help identify treatment goals and coordinate care. A number of studies in the orthotrauma population have shown that a proactive perioperative CGA can reduce the incidence of delirium.[18,19] 

Treatment
Sometimes a patient’s behaviors continue to be concerning despite management of the underlying cause of the delirium. Treatment with medication can be an option if nonpharmacological strategies such as one-on-one reorientation from a family member or trained care aide cannot prevent aggression or severe agitation and patient or staff safety is threatened. Medication is also an option if there is sleep-wake disturbance or psychosis. However, it is important to remember that not all behaviors need to be controlled by medication, nor do medications address the underlying cause of the behavior.

Typical and atypical antipsychotics may be considered for management of delirium even though there is inconclusive evidence that antipsychotic use reduces delirium severity or duration.[20] Studies of antipsychotics in delirium have largely been small and of varying quality. The types and doses of antipsychotics, as well as the populations studied, have been heterogeneous, making it difficult to generalize from the findings. Potential side effects include extra-pyramidal symptoms (EPS), including parkinsonism, akathisia, and dystonia, and prolongation of the QTc interval, particularly with use of haloperidol and quetiapine.[21] While the atypical antipsychotics are generally believed to have a lower propensity to induce EPS, prolonged use in patients with dementia has been associated with increased mortality.[22] 

Despite these drawbacks, antipsychotics can be used in limited circumstances with the understanding that the potential for side effects and the lack of robust evidence means that use of antipsychotics for symptoms of delirium is considered off-label. Risks and benefits of treatment should be discussed before initiating antipsychotic therapy, and families should be involved to see if alternative nonpharmacological strategies can be used instead. Treatment should be for a short time, and down-titration or discontinuation should be considered on a daily basis. Typically, we start geriatric patients on lower doses and titrate as appropriate (Table 2 [11]). Consultation with geriatric medicine, care of the elderly services, or a geriatric psychiatry physician, if available, can help guide management when medications are being considered.

Recent studies suggest that antipsychotics are useful in delirium prophylaxis, particularly in the postsurgical period. However, there is a large variation in the surgical patient population studied as well as in the regimens used. A recent meta-analysis showed that perioperative olanzapine and risperidone were useful in delirium prevention, while no difference was found when haloperidol was compared with a placebo.[24] Because of the heterogeneity of results and small effect size, the off-label use of prophylactic antipsychotics perioperatively has not become standard of care. 

Similarly, there have been several studies looking at the effects of melatonin on the prevention of delirium. Al-Aama and colleagues studied 145 elderly medical patients in a randomized double-blinded placebo-controlled trial and found that the melatonin group had a significantly lower risk of delirium incidence,[25] while de Jonghe and colleagues studied the effect of melatonin on the incidence of delirium in older patients with hip fracture but did not find melatonin efficacious.[26] Overall, there is weak evidence that melatonin is effective for the prevention of delirium. However, melatonin has a good safety profile and may be a reasonable choice to start in hospitalized patients at high risk for delirium, or in established delirium as a sleep aid. Doses between 0.5 and 9.0 mg have been studied in delirium and in dementia patients, with 3.0 to 6.0 mg being used commonly in clinical practice.[27] 

Given the cholinergic deficit theory in the pathogenesis of delirium, there have also been trials with acetylcholinesterase inhibitors, medications that increase the amount of acetylcholine in the synapse. Donepezil and rivastigmine have been studied in randomized controlled trials for both the prevention and the treatment of delirium, but the results have been disappointing. One of the largest studies of rivastigmine as an adjunct to usual care with haloperidol in intensive care unit patients was stopped early after higher mortality in the rivastigmine group was observed, even though the difference was not statistically significant.[28] Similarly, studies show that donepezil is not effective for delirium prophylaxis and treatment.[29,30] 

Summary
Delirium during hospitalization indicates medical instability that can affect a patient’s independence after discharge. While the exact pathophysiology of delirium is not completely understood, certain well-established in-hospital risk factors can be identified and addressed. These include cognitive impairment, sleep deprivation, immobility, visual impairment, hearing impairment, and dehydration. Ideally, nonpharmalogical strategies should be used to address these risk factors. The most effective approach is prevention, followed by treatment of the underlying cause. Drug therapies can be used if the safety of the patient and staff is a concern, but caution is advised given the side effects associated with certain medications. Typical and atypical antipsychotics may be used on an off-label basis for a short time and if down-titration and discontinuation are considered on a daily basis. 

In some institutions, efforts have been made to systematically implement measures to prevent and treat delirium. When delirium does occur, the medical team and family should be aware that the patient may need further support on discharge given the risk of cognitive and functional decline associated with delirium. 

Competing interests
None declared.

This article has been peer reviewed.

Dr Wan is a staff geriatrician at Providence Health Care. Dr Chase is a clinical instructor in the Division of Geriatric Medicine at the University of British Columbia and a staff geriatrician at St. Paul’s Hospital.