Delirium

Abstract

Delirium, a syndrome characterized by an acute change in attention, awareness and cognition, is caused by a medical condition that cannot be better explained by a pre-existing neurocognitive disorder. Multiple predisposing factors (for example, pre-existing cognitive impairment) and precipitating factors (for example, urinary tract infection) for delirium have been described, with most patients having both types. Because multiple factors are implicated in the aetiology of delirium, there are likely several neurobiological processes that contribute to delirium pathogenesis, including neuroinflammation, brain vascular dysfunction, altered brain metabolism, neurotransmitter imbalance and impaired neuronal network connectivity. The Diagnostic and Statistical Manual of Mental Disorders, 5th edition (DSM-5) is the most commonly used diagnostic system upon which a reference standard diagnosis is made, although many other delirium screening tools have been developed given the impracticality of using the DSM-5 in many settings. Pharmacological treatments for delirium (such as antipsychotic drugs) are not effective, reflecting substantial gaps in our understanding of its pathophysiology. Currently, the best management strategies are multidomain interventions that focus on treating precipitating conditions, medication review, managing distress, mitigating complications and maintaining engagement to environmental issues. The effective implementation of delirium detection, treatment and prevention strategies remains a major challenge for health-care organizations globally.

Figures

Figure 1.. Risk factors for delirium.

Risk factors for delirium relate to premorbid or predisposing factors (that is, a patient’s characteristics) and to precipitating factors, which are factors relating to the presenting illness or that occur after hospital admission.

Figure 2.. Bioenergetic insufficiency may underpin delirium in multiple scenarios.

Neurons and astrocytes both use glucose supplied by the microvasculature to generate ATP by glycolysis. In addition, in the astrocyte–neuron lactate shuttle (ANLS), lactate synthesised by astrocytes after glycolysis can be exported for use by neurons, which convert the lactate to pyruvate that is imported into the mitochondria and used to fuel the tricarboxylic acid (TCA) cycle. There are many ways in which the brain, or regions of the brain, may become dysfunctional due to energy insufficiency and there is support for the idea that these may contribute to precipitation of delirium. First, respiratory distress produces hypoxaemia and can cause brain hypoxia, limiting neuronal energy metabolism. That is, in hypoxic conditions, insufficient oxygen (O2) supply leads to impaired mitochondrial oxidative phosphorylation (OXPHOS) and insufficient generation of energy, in the form of ATP. In these conditions, glycolysis-generated pyruvate (Pyr), instead of being imported into the mitochondria, forms excess lactate (Lac), which can be measured in the extracellular fluid. Second, septic shock reduces blood flow, producing both hypoxia and impaired glucose supply. Third, even with adequate systemic blood flow, brain microcapillary dysfunction may produce brain tissue hypoxia and neuroglycopenia. Fourth, even with normal blood pressure, if neurovascular coupling is impaired, vessels may fail to meet the specific demands of regional neuronal activity and thereby block higher order brain functions. Fifth, systemic hypoglycaemia can lead to insufficient brain glucose supply, delirium and coma. Sixth, even with adequate delivery of glucose to the brain, insulin resistance may result in impaired glucose utilization (not shown). Seventh, altered expression of glucose transporters (GLUT1 and GLUT3), for example, in the degenerating brain, may limit glucose uptake by the endothelium, astrocytes or neurons, thereby limiting the glucose-6-phosphate (G6P) required for glycolysis and limiting the generation of Pyr required for the TCA cycle. Last,impairment of astrocyte function may limit their ability to release glycogen from intracellular stores, metabolize glucose and provide lactate to neurons for energy metabolism.

Figure 3.. Inflammatory mechanisms in delirium.

Inflammatory trauma, surgery, infection and sepsis can trigger delirium. These diverse stressors may share pathogenetic mechanisms, including increased local and circulating levels of damage-associated molecular patterns (in surgery and trauma) and pathogen-associated molecular patterns (in infection and sepsis). These stimuli trigger tissue macrophage and blood monocyte activation and secretion of inflammatory mediators, such as IL-1, IL-1β, IL-6, tumour necrosis factor (TNF), and prostaglandin E2 (PGE2). These molecules may, to a limited extent, cross the blood–brain barrier (BBB) but their production is also induced in the brain endothelium and epithelium (not shown) and secreted directly into the brain parenchyma by endothelial, epithelial and brain perivascular macrophages. By mechanisms that are not entirely clear, microglia are then triggered to produce pro-inflammatory cytokines and reactive oxygen and nitrogen species. If primed by prior pathology in the brain (such as amyloid or prior neurodegeneration), these microglia produce increased levels of these mediators, which affect both astrocytes and neurons. Cytokine-stimulated astrocytes produce increased levels of chemokines, contributing to recruitment of monocytes and other immune cell populations to the brain, but their activation also leads to a loss of metabolic support for neuronal energy metabolism. Microglial-derived inflammatory mediators, such as IL-1β and TNF, directly affect neuronal function to produce both dysfunction and injury or cell death, which collectively may contribute to acute behavioural manifestations in the delirium syndrome but also produce new brain injury that promotes long-term cognitive decline. NO, nitric oxide; ROS, reactive oxygen species.

Figure 4.. Major mechanisms in delirium pathophysiology.

Major perturbations leading to delirium during acute illness include robust acute systemic inflammation (involving increased circulating pro-inflammatory cytokines (such as IL-1, IL-1β and tumour necrosis factor (TNF)), pathogen-associated molecular patterns and damage-associated molecular patterns), hypoxemia, impaired blood flow and tissue perfusion, and impaired metabolism (hyponatremia, hypernatremia and hypoglycaemia). Given the altered arousal states present in different subtypes of delirium, it is widely assumed that alteration of function in the ascending arousal system may be involved. These distributed mid-brain and brainstem nuclei include strong cholinergic drive from the tegmentum to the thalamus to activate cortical arousal and multiple monoaminergic nuclei that activate the cortex to modulate and integrate cortical activation. Medications, prominently including, but not limited to, GABAergic sedatives and anaesthetics and anti-cholinergic and anti-histamine drugs, can therefore substantially alter arousal and are known to contribute to delirium. Microglia can be primed by existing pathology in the brain and further activated by acute inflammatory stimuli, secreting pro-inflammatory cytokines and reactive oxygen and nitrogen species into the surrounding brain tissue. These mediators can directly affect neuronal function but also act directly on astrocytes. Astrocytes can also be primed during chronic brain pathology, becoming hypersensitive to acute inflammatory stimulation and secreting increased levels of chemokines, which can drive recruitment of additional peripheral inflammatory cells to the brain. Activated astrocytes may also lose aspects of the energy metabolism support that they provide to neuronal function. The vasculature may become impaired, both by existing degenerative pathology and by superimposed stressors such as systemic inflammation, leading to endothelial injury and blood–brain barrier (BBB) damage, but vascular supply of oxygen and glucose may also become impaired due to microvascular dysfunction and/or impaired neurovascular coupling, contributing to a metabolic (bioenergetic) insufficiency. All of these mechanisms contribute to the most obvious proximate cause of delirium: acute neuronal dysfunction and network disintegration. 5-HT, 5-hydroxytryptamine; ACh, acetylcholine; BBB, blood–brain barrier; BF, basal forebrain; DA, dopamine; GABA, gamma-aminobutyric acid; His, histamine; LC, locus coeruleus; LDT, laterodorsal tegmental nucleus; LH, lateral hypothalamus; MCH, melanin-concentrating hormone; NA, noradrenaline; NO, nitric oxide; ORX, orexin; PPT, pedunculopontine tegmentum; ROS, reactive oxygen species; TMN, tuberomammillary nucleus; vPAG, ventral periaqueductal gray. Adapted with permission from ref..

Figure 5.. Common tools to screen for delirium in different settings.

Once there is suspicion of delirium based on the presence of symptoms such as acute disturbance of attention, reduced environmental awareness and/or changes in cognition, the choice of screening tool is made based on the setting. The 4 A’s test (4AT) is a 4-item test used in general hospital settings. A score ≥4 indicates delirium is likely. Confusion assessment method (CAM)-based tools, such as the CAM, CAM-ICU, brief-CAM, paediatric-CAM and preschool-CAM, assess four features, Features A–D (or Features 1–4 depending on the specific tool). For delirium to be present according to CAM based tools, A and B must be present, plus either C or D. The Intensive Care Delirium Screening Checklist (ICDSC) is a delirium assessment tool measuring 8 domains, recorded as yes (present; score 1) or no (absent; score 0) answers. Delirium is likely for a score ≥4. The clinical tools described, except for ICDSC, are snapshot assessments. *depending on the tool used, Features A–D may be called 1–4 and will be assessed in different ways. DOB, date of birth; LOC, level of consciousness.

Figure 6.. Associations between performance of the ABCDEF bundle and outcomes.

Results from the Collaborative of over 15,000 patients from 68 academic, community and federal ICUs, which showed percent performance of the ABCDEF bundle and symptom-related outcomes. ABCDEF bundle components include A (Assess, prevent, and manage pain), B (Both spontaneous awakening and breathing trials), C (Choice of analgesia and sedation), D (Delirium: assess, prevent, and manage), E (Early mobility and exercise) and F (Family engagement and empowerment). Each graph shows the relationship between the proportion of eligible elements of the ABCDEF bundle performed on a particular day and the probability that the patient would experience that symptom-related outcome on the following day. The outcomes include requiring mechanical ventilation (part a), significant pain (part b), coma (part c), delirium (part d) and requiring physical constraints (part e). and confidence bands represent the probability of the outcomes and the 95% confidence interval, adjusted for baseline ICU admission characteristics and daily covariates. The relationships between proportion of elements performed and each outcome was significant (P <0.0001) for all outcomes. Adapted with permission from ref..

Figure 7.. Antipsychotic drugs are ineffective in delirium treatment.

In the MIND-USA trial, the efficacy of the antipsychotics haloperidol and ziprasidone was compared to placebo for the treatment of delirium in critically ill patients. There was no significant association between study drug (the atypical antipsychotic ziprasidone, the typical antipsychotic haloperidol or placebo) and days free from delirium or coma (part a), days with delirium (part b) or days with coma (part c). Adapted with permission from ref..

Figure 8.. Proportion of delirious patients according to sedation regimen in the ICU.

ICU patients with a sedation level at Richmond Agitation Sedation Scale (RASS) of –2 or greater were assessed with the Confusion Assessment Method-Intensive Care Unit (CAM-ICU). There were more patients with positive CAM-ICU result during the first 24 hours among ICU patients receiving dexmedetomidine (Dex), an α2-adrenergic receptor agonist, than among those receiving the usual care sedation regimen (mainly the GABAergic agents propofol or midazolam) as directed by the treating physician. CAM-ICU positivity declined significantly faster over the subsequent treatment days in patients receiving dexmedetomidine than in those on usual care regimen (overall P value = 0.01). The data for this figure are from ref..

Figure 9.. Relationship between delirium and post-ICU quality of life.

With critical Illness as the backdrop, this schematic depicts the association between delirium during an intensive care unit (ICU) stay and downstream impairments that lead to poor quality of life. Multiple potential pre-existing and precipitating risk factors can lead to delirium during critical illness. The development and duration of delirium increase the risk of cognitive, psychiatric and physical impairments, which accumulate to reduce a patients’ quality of life. PTSD, post-traumatic stress disorder.