Introduction

Delirium is a syndrome of acute brain dysfunction that is becoming increasingly recognized as an important contributor to both in-hospital and post-discharge morbidity and mortality. Delirium manifests as fluctuations in mental status and may present with a wide spectrum of symptoms. Cardinal symptoms are inattention, disorganized thinking, disorientation, and/or altered levels of consciousness, and additional symptoms can include sleep disturbance, acute psychosis, emotional disturbance, and increased or decreased psychomotor activity [1–3]. In addition, patients may manifest symptoms on a wide spectrum from hypoactivity to hyperactivity and may fluctuate between the two (mixed delirium). The most common form, hypoactive delirium, can also prove to be the most difficult to diagnose [1].

Delirium that develops during an intensive care unit (ICU) stay has been shown to increase the length of stay in the ICU, as well as total hospital length of stay, and increase healthcare costs. Prior estimates have placed the healthcare costs associated with delirium at $164 billion per year in the USA and €140 billion per year in 18 combined European countries [4, 5]. In addition, every day of delirium a patient suffers increases their risk of death by approximately 10 % [6–10]. Similar outcomes are also seen in the operative setting, where patients who develop delirium have increased odds of death both in the hospital and six months after discharge [11•, 12]. Increasing durations of delirium have been shown to be associated with significantly worse cognitive and executive functioning at 3 and 12 months [13••]. This cognitive impairment is independent of severity of illness and age. In addition, postoperative patients who develop delirium have significant declines in physical and social functioning as well as vitality [11•].

Delirium in the ICU is common and likely under recognized. In mixed medical and surgical ICU patients who are mechanically ventilated, 60–80 % develops delirium [14]. This was confirmed in more recent prospective cohort studies of mixed medical and surgical ICU patients, demonstrating 75 % of patients developed delirium during their stay [13••]. Up to a quarter of elderly patients will have delirium upon presentation to the hospital, with an additional third developing it over their stay [15–17]. Given the high incidence of delirium and the tremendous impact delirium has on patient outcomes, clinician education in diagnosis, prevention, and treatment is essential.

Diagnosis of Delirium

The first step toward preventing the deleterious effects of delirium is correctly identifying delirium when it occurs. The cornerstone of diagnosis of delirium is a careful history and physical examination. If a patient is identified as unresponsive (i.e., a Richmond Agitation and Sedation Scale [RASS] of −4 or −5) [18], they are considered to be in coma, and thus cannot be evaluated for delirium. Using the history and examination, providers should seek to identify acute changes and fluctuation in a patient’s mental status. Upon identifying this, providers can use one or two validated tools for diagnosis. Both the Confusion Assessment Method for the ICU (CAM-ICU) [19] and the Intensive Care Delirium Screening Checklist (IDCSC) [20] have been validated as tools for diagnosing delirium in the critically ill. The CAM-ICU assesses four clinical features: acute change or fluctuation in mental status, inattention, disorganized thinking, and altered level of consciousness (Fig. 1). The ICDSC uses eight diagnostic features, with four or more positive features diagnosing delirium. Delirium is especially difficult to detect in the presence of underlying dementia, and the symptoms of delirium can be frequently attributed to pre-existing cognitive dysfunction. The diagnosis of delirium in the demented patient is especially important, however, given that 60–89 % of demented hospitalized patients have delirium [21, 22] and that delirium has been shown to accelerate the process of cognitive decline amongst those with dementia [23•]. Current evidence supports the use of the CAM-ICU as the best tool for evaluating delirium in patients with dementia [24].

Fig. 1
figure 1

The Confusion Assessment Method for the Intensive Care Unit (CAM-ICU). Patients are considered to have delirium if they have a Richmond Agitation–Sedation Scale score of −3 and above, and have features 1 and 2 plus either feature 3 or 4. Reproduced with permission from Dr. E. Wesley Ely (www.icudelirium.org)

Full size image

Risk Factors for Delirium

Identification of patients at highest risk for delirium can help in controlling risk factors and aid in appropriately identifying delirious patients. Risk factors for delirium can be categorized as patient-related factors associated with underlying comorbid disease or iatrogenic risk factors. Patient-related factors can include age, diabetes, heart failure, underlying dementia, electrolyte imbalance, pain, and numerous others [25]. Appropriate management of these chronic underlying conditions is an important facet of preventing delirium. Some iatrogenic factors that can be controlled by clinicians should be considered carefully in the management of the critically ill patient. Sedative medications and sleep hygiene are amongst the two most important.

Studies have consistently shown associations between psychoactive medications, in particular, benzodiazepines, and the development of delirium in the critically ill population. In a cohort of mechanically ventilated medical ICU patients, lorazepam was shown to be an independent risk factor for transition to delirium after controlling for other important patient factors such as age and severity of illness [26]. In medical ICU patients on mechanical ventilation, greater daytime benzodiazepine doses and nighttime increase in benzodiazepine dosing, which occurred in nearly half of patients, were associated with delirium [27•]. Midazolam has also been associated with more delirium among surgical and burn patients [28, 29]. Opioid medications have been shown to decrease delirium when appropriately used to control pain [28, 30, 31]. However, excessive opioid administration has also been associated with increased risk of delirium [31, 32]. In a recent secondary analysis of a prospective observational study, benzodiazepine and/or opioid infusions were strongly associated with transition to delirium in mechanically ventilated ICU patients [33•]. These studies point to the importance of judicious use of sedative medications in the critically ill patients.

Prevention of Delirium

Among ICU patients, several interventions have been shown to have a positive impact in reducing delirium incidence (Table 1). Early mobilization with physical and occupational therapy has been shown to reduce duration of delirium by 50 % and significantly improve functional status at the time of discharge [34]. Improving sleep hygiene by providing ear plugs to patients in the ICU has been shown to reduce the incidence of delirium and improve sleep perception [35]. Choice of sedative agents has also been shown to impact delirium outcomes. Several studies have demonstrated improved delirium endpoints in patients sedated with dexmedetomidine versus benzodiazepines. The Maximizing Efficacy of Targeted Sedation and Reducing Neurological Dysfunction (MENDS) study showed that sedation with dexmedetomidine compared to lorazepam decreased duration of delirium, with lower likelihood of delirium on subsequent days [36]. The Safety and Efficacy of Dexmedetomidine Compared with Midazolam (SEDCOM) study showed a reduction in delirium prevalence in those sedated with dexmedetomidine [37]. In another trial, the DEXmedetomidine COmpared to Morphine (DEXCOM) study demonstrated a reduction in the duration of delirium in patients after cardiac surgery sedated with dexmedetomidine compared to morphine but no reduction in delirium incidence [38]. Another study showed improvements in outcomes related to mental status including arousability, communication, and patient cooperation in patients treated with dexmedetomidine compared to midazolam or propofol [39•].

Table 1 Studies of delirium prevention in the ICU
Full size table

The direct effect of sedation on delirium in the form of “rapidly reversible, sedation-related delirium,” defined as delirium that abates shortly after sedative interruption, was recently studied in a prospective cohort study [40••]. It found delirium to be prevalent, with 89 % of patients developing delirium. A small group of patients (12 %) whose delirium abated after sedation interruption had fewer ventilator days, ICU days, and hospital days, and were also more likely to be discharged home. Those with persistent delirium (77 % of patients) had higher 1-year morality [40••]. This has two potentially important implications. First, recent sedative administration should be carefully considered when evaluating for delirium. Second, given consistent findings of higher mortality, impaired functional status, and cognitive impairment in all patients with delirium, and the likelihood that those with rapidly reversible delirium may be “diluting” the negative outcomes, the effect of persistent delirium is likely greater than has been measured in studies to date.

Studies have investigated the effect of antipsychotics in preventing the incidence of delirium with mixed results. In a blinded, placebo-controlled trial in patients undergoing cardiac surgery, a single dose of sublingual risperidone upon regaining consciousness reduced incidence of delirium compared to placebo [41]. A recent before and after study of haloperidol as prophylaxis in patients at high risk for delirium showed significantly less incidence and duration of delirium [42]. The Haloperidol Effectiveness in ICU Delirium (HOPE-ICU) study, on the other hand, showed no difference in days alive and free of delirium or coma between patients prophylactically treated with intravenous haloperidol compared to placebo [43••].

Pointing to potential neuroinflammation as a mechanism in causing delirium, ongoing statin therapy while in the ICU has been shown in two studies to be associated with lower overall risk of delirium, especially in chronic statin users [44•, 45•]. Further, randomized controlled trial evidence is needed to provide conclusive evidence of the effect of statins on incident delirium. Prior evidence of the association between sleep disturbances and delirium has generated interest in melatonin as an agent to prevent delirium. A recent double-blind randomized controlled trial of melatonin versus placebo in patients with hip fracture did not demonstrate a difference in incidence of delirium [46].

Taken together, this body of literature supports a systematic approach to sedation during critical illness focusing on the ABCDEs—(A) Awakening and (B) Breathing trials, (C) Choice of sedation, (D) Delirium monitoring and management, and (E) early Exercise—in order to improve patient outcomes, including delirium. For example, measurement of RASS and CAM-ICU paired with protocolized de-escalation of sedation reduced mean hourly benzodiazepine dose, duration of mechanical ventilation, ICU length of stay, hospital length of stay, and resulted in a 33 % reduction in odds of delirium [47]. Additionally, a study of the implementation of an ABCDE protocol demonstrated a 3-day increase in ventilator free days, an increased rate of mobilization, and a 45 % reduction in delirium [48••].

Management of Delirium

The cornerstone of delirium management is prevention and correction of underlying processes contributing to brain dysfunction. Large randomized controlled trials do not exist for any pharmacologic treatments for delirium. Numerous smaller studies and reports suggest that antipsychotic medications may be a treatment for delirium, but there are an equal number of studies suggesting no effect (Table 2) [49]. An early study showed that haloperidol and olanzapine were equally efficacious at reducing symptoms of delirium in critically ill patients with olanzapine having fewer extrapyramidal side effects [50]; however, the lack of a control group makes it impossible to draw definitive conclusions on the efficacy of antipsychotics versus spontaneous resolution. The Modifying the Incidence of Delirium (MIND) study compared ziprasidone, haloperidol, and placebo in mechanically ventilated patients with delirium and found no difference in the number of days alive and free of coma or delirium between the three groups [51]. In patients already ordered to receive as needed haloperidol, another study showed that quetiapine was more effective than placebo in time to resolution of first episode of delirium [52]. Based on evidence that impaired cholinergic neurotransmission may play an important role in the development of delirium, rivastigmine was studied as an adjunct treatment to haloperidol versus a combination of placebo and haloperidol [53]. There was no decrease in duration of delirium, and there was a trend toward increased mortality in the rivastigmine group [53]. In delirious mechanically ventilated patients, dexmedetomidine was shown to reduce time to extubation, incidence of tracheostomy, and ICU length of stay compared to haloperidol [54]. Taken together, the body of literature does not support a single pharmacologic approach to treatment of delirium. In general, we recommend treating only when non-pharmacologic measures such as reorientation and mobilization have not succeeded and using the smallest dose of psychoactive agents for the shortest possible duration.

Table 2 Studies of delirium treatment in the ICU
Full size table

Conclusion

Acute brain dysfunction in the form of delirium is a common problem in the critically ill population, and associated with increased healthcare costs and significant increases in morbidity and mortality both in the hospital and after discharge. With a combination of efforts through implementation of best practices, however, we can reduce the burden of delirium. Overall, we recommend an integrated approach to delirium management that involves attention to the ABCDEs—(A) Awakening and (B) Breathing trials, (C) Choice of sedation, (D) Delirium monitoring and management, and (E) early Exercise. In addition, ongoing clinical trials are investigating pharmacologic therapies for delirium and may offer hope for better evidenced-based treatment in the future.