Subacute Delirium

Neurologic MONDO:0004629 Pathograph 14 Show in embeddings browser Neurocognitive Disorder

Subacute delirium is delirium — an acute disturbance of attention and awareness with an additional cognitive disturbance, developing over a short period and fluctuating in severity across the day — in which the onset is insidious (days to a week or two rather than hours) and the course is protracted. It is not a mechanistically distinct disease: MONDO:0004629 sits in the same concept space as ICD-10 F05, whose inclusion terms are explicitly "acute or subacute confusional state", "acute or subacute brain syndrome", and "acute or subacute psycho-organic syndrome". This entry therefore curates the delirium pathograph and treats "subacute" as the temporal qualifier on that pathograph — the presentation that is most easily mistaken for evolving dementia and that shades into the persistent delirium recorded under `progression`. The mechanism is convergent rather than linear. A predisposed brain — aged, demented, frail, or otherwise reduced in reserve — meets a precipitant (systemic infection, surgery, metabolic derangement, hypoxia, or a deliriogenic drug). Peripheral inflammatory signalling reaches the CNS, in part through an increasingly permeable blood-brain barrier, activating microglia; neurotransmitter systems (classically a relative cholinergic deficiency with dopaminergic excess) are perturbed; and large-scale cortical networks lose the feedback connectivity that supports attention and perceptual inference. Because the vulnerability and the insult multiply rather than add, a trivial insult suffices in a vulnerable patient and no single node is the disease. Two facts discipline the entry. First, the causal-therapeutic translation of the neurotransmitter model has failed: cholinesterase inhibition did not shorten delirium and signalled harm, and antipsychotics did not alter its duration in a definitive placebo-controlled trial — though dexmedetomidine did benefit the narrower population of non-intubated patients with hyperactive delirium, so the nihilism is about mechanism-directed drugs rather than about pharmacotherapy as such. Second, delirium is not the benign, fully reversible state it was long taken to be — a substantial minority remain delirious months later, and delirium is independently associated with long-term cognitive decline. Prevention by multicomponent non-pharmacological programs remains the only intervention with consistent effect on incidence.

Ask OpenScientist

Ask a research question about Subacute Delirium. OpenScientist will conduct autonomous deep research using the Disorder Mechanisms Knowledge Base and PubMed literature (typically 10-30 minutes).

Submitting...

Do not include personal health information in your question. Questions and results are cached in your browser's local storage.

8
Pathophys.
8
Phenotypes
4
Hypotheses
3
Gaps
14
Pathograph
1
Genes
4
Medical Actions
2
Differentials
1
Trials
2
Models
2
References
1
Deep Research
🏷

Classifications

Harrison's Part
NEUROLOGIC
◈

Mechanistic Hypotheses

4
Delirium as the product of baseline vulnerability multiplied by an acute precipitant
vulnerability_precipitant_model CANONICAL
Evidence balance 2 support
The multifactorial model is the organizing framework for delirium and the one this entry's pathograph is built on: predisposing factors (age, dementia, frailty, sensory impairment) set the threshold, and precipitating insults (infection, surgery, drugs, metabolic derangement) cross it. Its strength is that it explains why a trivial insult suffices in one patient and a series of major ones is needed in another, and why single-target interventions fail while multicomponent ones work. Its weakness as a *mechanistic* model is that it is a risk architecture rather than a molecular pathway — it says which combinations produce delirium, not what the brain does when they do.
Show evidence (2 references)
PMID:29278283 SUPPORT Human Clinical
"Available data suggest that numerous pathological factors may serve as precipitants for delirium, each having differential effects depending on patient-specific patient physiological characteristics (substrate)."
States the precipitant × substrate structure that this hypothesis group organizes the causal edges around.
PMID:23992774 SUPPORT Human Clinical
"In view of the complex multifactorial causes of delirium, multicomponent non-pharmacological risk factor approaches are the most effective strategy for prevention."
The therapeutic corollary of the model: because causation is multifactorial, the intervention that works is multicomponent rather than single-target.
Peripheral inflammation reaching the CNS through a permeable blood-brain barrier
neuroinflammatory_model CANONICAL
Evidence balance 2 support
The best-evidenced molecular account of how a systemic insult becomes a brain state. Peripheral pro-inflammatory signalling — IL-6 prominently — is associated with delirium risk before the insult and with delirium severity after it, and the same studies tie the CNS compartment of that signalling to measured blood-brain barrier permeability. Animal work adds the crucial conditional: the same cytokine challenge is cognitively inert in a normal brain and disabling in a degenerating one, which is the vulnerability × precipitant model expressed at the level of a single molecule.
Pathograph links
Reduced Cognitive Reserve -> Microglial Activation and Neuroinflammation INDIRECT UNKNOWN INTERMEDIATES A primed, aged, or degenerating brain mounts an exaggerated central response to the same peripheral signal, which is the best-supported cellular reading of "reduced reserve".
Systemic Inflammatory Insult -> Blood-Brain Barrier Permeability Increase INDIRECT KNOWN INTERMEDIATES Circulating cytokines act on cerebral endothelium; the resulting permeability change is measurable as a CSF:plasma albumin ratio that tracks central IL-6 receptor signalling.
Systemic Inflammatory Insult -> Microglial Activation and Neuroinflammation INDIRECT KNOWN INTERMEDIATES Peripheral inflammatory signal reaches the CNS by humoral, endothelial, and vagal routes and converts primed microglia to an activated state.
Blood-Brain Barrier Permeability Increase -> Microglial Activation and Neuroinflammation DIRECT Entry of peripheral mediators into the CNS compartment activates parenchymal microglia.
Microglial Activation and Neuroinflammation -> Cortical Feedback Connectivity Loss INDIRECT UNKNOWN INTERMEDIATES The inflammatory and circuit accounts meet here: feedback connectivity falls as systemic inflammation rises, but the pathway from cytokine to connectivity is not resolved.
Microglial Activation and Neuroinflammation -> Neuronal and Astrocytic Injury INDIRECT UNKNOWN INTERMEDIATES Sustained activation is the proposed route from a reversible state to the injury that underlies incomplete recovery.
Protracted and Persistent Course -> Neuronal and Astrocytic Injury INDIRECT UNKNOWN INTERMEDIATES Duration of delirium is the exposure variable that predicts long-term cognitive outcome, so a protracted course is curated as feeding the injury node rather than merely coexisting with it.
Show evidence (2 references)
PMID:34218905 SUPPORT Human Clinical
"Higher preoperative interleukin-6 was associated with postoperative delirium with a standardised mean difference (95% confidence interval) of 0.33 (0.11-0.56) and P=0.003."
Quantifies the inflammation-delirium association in the direction the hypothesis requires — inflammatory tone before the insult predicts the syndrome after it.
PMID:42011187 SUPPORT Human Clinical
"CSF sIL-6R was also significantly correlated with peripheral inflammation (plasma IL-6 and IL-8) and blood-brain barrier permeability (CSF:plasma albumin ratio and plasma S100B)."
Links the peripheral and central inflammatory compartments through a measured barrier-permeability index, which is the step this hypothesis turns on.
Relative cholinergic deficiency with dopaminergic excess
cholinergic_dopaminergic_imbalance_model ALTERNATIVE
Evidence balance 1 support 2 refute
The oldest mechanistic account of delirium, motivated by the deliriogenic potency of anticholinergic and dopaminergic drugs. It is retained here as ALTERNATIVE rather than CANONICAL because its two strongest predictions have failed. Therapeutically, augmenting cholinergic transmission with rivastigmine did not shorten delirium in critically ill patients and the trial was stopped early for excess mortality in the treatment arm; and dopamine-receptor blockade with haloperidol or ziprasidone did not alter delirium duration against placebo. The neurotransmitter disturbance is real and is the best explanation for drug-precipitated delirium; what the negative trials refute is that correcting it downstream reverses the syndrome.
Pathograph links
Microglial Activation and Neuroinflammation -> Cholinergic-Dopaminergic Neurotransmitter Imbalance INDIRECT UNKNOWN INTERMEDIATES Neuroinflammation is proposed to perturb cholinergic and dopaminergic transmission; the intermediates are not established in humans, and the edge is recorded at that strength rather than as a direct effect.
Cholinergic-Dopaminergic Neurotransmitter Imbalance -> Cortical Feedback Connectivity Loss INDIRECT UNKNOWN INTERMEDIATES Cholinergic tone gates cortical signal-to-noise and attention; the link to the measured connectivity failure is plausible but not established in humans.
Show evidence (3 references)
PMID:21056464 SUPPORT Human Clinical
"Impaired cholinergic neurotransmission seems to have an important role in the development of delirium."
States the hypothesis as the trial's own rationale, which is what makes the trial's negative result a test of it rather than an unrelated finding.
PMID:21056464 REFUTE Human Clinical
"Rivastigmine did not decrease duration of delirium and might have increased mortality so we do not recommend use of rivastigmine to treat delirium in critically ill patients."
Refutes the therapeutic prediction of the cholinergic-deficiency model — restoring cholinergic tone neither shortened delirium nor was safe.
PMID:30346242 REFUTE Human Clinical
"The use of haloperidol or ziprasidone, as compared with placebo, in patients with acute respiratory failure or shock and hypoactive or hyperactive delirium in the ICU did not significantly alter the duration of delirium."
Refutes the mirror-image therapeutic prediction — blocking the dopaminergic arm did not change the course of the syndrome.
Loss of large-scale cortical feedback connectivity as the proximate substrate of inattention
network_disconnection_model CANONICAL
Evidence balance 3 support
The circuit-level endpoint on which the other models converge: whatever the upstream insult, the delirious brain shows breakdown of the long-range connectivity that supports attention and predictive processing. Electrophysiological modelling localizes this to reduced feedback connectivity, with the reduction scaling with delirium severity and with systemic inflammation — which is what makes it a shared endpoint rather than a parallel story.
Pathograph links
Microglial Activation and Neuroinflammation -> Cortical Feedback Connectivity Loss INDIRECT UNKNOWN INTERMEDIATES The inflammatory and circuit accounts meet here: feedback connectivity falls as systemic inflammation rises, but the pathway from cytokine to connectivity is not resolved.
Cortical Feedback Connectivity Loss -> Protracted and Persistent Course UNKNOWN Whether failure of the network state to re-establish itself is what makes a delirium subacute or persistent is unresolved; the edge records the candidate relationship without asserting it.
Show evidence (3 references)
PMID:37695013 SUPPORT Human Clinical
"DCM showed that delirium was associated with decreased left-sided superior temporal gyrus (l-STG) to auditory cortex feedback connectivity."
Direct electrophysiological evidence for the connectivity failure this node asserts.
PMID:37695013 SUPPORT Human Clinical
"Feedback connectivity also negatively correlated with delirium severity and systemic inflammation."
Ties the circuit endpoint quantitatively to both the clinical severity and the inflammatory upstream, which is the link the model needs.
PMID:29278283 SUPPORT Human Clinical
"The specific cognitive and behavioral manifestations of the specific delirium picture result from a combination of neurotransmitter function and availability, variability in integration and processing of sensory information, motor responses to both external and internal cues, and the degree of..."
Places network breakdown as the common final term that converts the upstream disturbances into the clinical picture.
?

Discussions and Knowledge Gaps

3
Does "subacute delirium" name anything mechanistically distinct from delirium with a protracted course, or is MONDO:0004629 purely a temporal qualifier?
KNOWLEDGE GAP subacute_vs_acute_delirium_nosology
Attached to
pathophysiology#Protracted and Persistent Course
ICD-10 F05 explicitly bundles "acute or subacute confusional state" under a single code, and no study identified during curation compares the pathophysiology of insidious-onset delirium with that of abrupt-onset delirium. The persistent-delirium literature quantifies how long delirium lasts, but persistence after an acute onset is not the same claim as an insidious onset. Until a study stratifies mechanism or biomarker by onset tempo, this entry curates the delirium pathograph with a temporal qualifier and does not assert a distinct mechanism.
Do rodent inflammatory-challenge models reproduce the attentional disturbance that defines human delirium, or only the sickness behaviour and memory impairment that accompany it?
HUMAN MODEL MISMATCH rodent_models_lack_attentional_readout
Attached to
animal_models#Mouse
The ME7/TNF-alpha model reproduces the conditional structure of delirium convincingly — the same insult, a deficit only on a vulnerable background — but its readout is T-maze working memory, and the accompanying phenotype is sickness behaviour (hypothermia, weight loss, inactivity). Inattention and disturbed awareness, the DSM-defining features, have no established rodent equivalent. Evidence therefore exists in the model, but its translational validity for the cardinal feature specifically remains open. This is a mismatch of readout, not an absence of data.
How does APOE genotype confer delirium risk independently of dementia — is it lipid transport and synaptic resilience, an amplified neuroinflammatory response, or something else?
OPEN QUESTION apoe_mechanism_of_susceptibility
Attached to
genetic#APOE
The multi-ancestry meta-analysis establishes the association and its independence from dementia, and the accompanying proteomics implicate brain vulnerability, inflammation and immune-response processes. But the step from APOE genotype to any node in this pathograph is not established: the entry draws no causal edge from APOE to the microglial or reserve nodes because none is evidenced. Resolving this would connect the strongest genetic signal in delirium to the mechanism the rest of the entry curates, and the druggability analysis in the same paper suggests it is being actively pursued.
⚙

Pathophysiology

8
Reduced Cognitive Reserve
The predisposing half of the model. Advanced age, dementia or milder cognitive impairment, frailty, sensory impairment, and comorbidity burden lower the insult required to produce delirium. This node is a state of the patient before any precipitant arrives; it does not itself produce delirium, and it is the reason the same drug dose is inert in one patient and deliriogenic in another.
Peripheral inflammation reaching the CNS through a permeable blood-brain barrier
Downstream
Microglial Activation and Neuroinflammation A primed, aged, or degenerating brain mounts an exaggerated central response to the same peripheral signal, which is the best-supported cellular reading of "reduced reserve". Peripheral inflammation reaching the CNS through a permeable blood-brain barrier
Show evidence (3 references)
PMID:23992774 SUPPORT Human Clinical
"it serves both as a marker of brain vulnerability with decreased reserve and as a potential mechanism for permanent cognitive damage"
States the reduced-reserve framing that this node represents, and — in the same breath — the injury arm curated downstream.
PMID:39602991 SUPPORT Human Clinical
"The risk factors of delirium included frailty (odds ratio = 2.05), physical restraints (5.01), prior falls (1.99), severe illness (1.32) (evaluated using the Acute Physiology and Chronic Health Evaluation II), and cognitive impairment (2.61)."
Quantifies the predisposing factors — frailty, prior falls, cognitive impairment — that constitute this node, alongside the iatrogenic precipitants curated under environmental.
PMID:25239680 SUPPORT Human Clinical
"In the Vantaa cohort, 465 persons (88.4 ± 2.8 years) completed MMSE at baseline. For every MMSE point lost, risk of incident delirium increased by 5% (p = 0.02)."
Converts this node from a categorical risk factor into a graded one: risk scales continuously with pre-existing cognitive impairment rather than switching on at a dementia diagnosis.
Systemic Inflammatory Insult
The precipitating half. Infection and sepsis, major surgery, trauma, and critical illness raise circulating pro-inflammatory mediators — IL-6, IL-8, TNF-alpha — and this peripheral signal is measurably associated with subsequent delirium. In the subacute presentation the insult is typically indolent (an occult infection, a slowly accumulating drug effect, a gradually worsening metabolic derangement) rather than abrupt, which is why the onset is measured in days.
Peripheral inflammation reaching the CNS through a permeable blood-brain barrier
cytokine-mediated signaling pathway GO:0019221 Gene Ontology (GO) Relation: this pathophysiological event involves this biological process This pathophysiological event involves increased cytokine-mediated signaling pathway (GO:0019221). GO:0019221 is a biological process from the Gene Ontology. ↑ INCREASED
Downstream
Blood-Brain Barrier Permeability Increase Circulating cytokines act on cerebral endothelium; the resulting permeability change is measurable as a CSF:plasma albumin ratio that tracks central IL-6 receptor signalling. Peripheral inflammation reaching the CNS through a permeable blood-brain barrier
Microglial Activation and Neuroinflammation Peripheral inflammatory signal reaches the CNS by humoral, endothelial, and vagal routes and converts primed microglia to an activated state. Peripheral inflammation reaching the CNS through a permeable blood-brain barrier
Show evidence (1 reference)
PMID:34218905 SUPPORT Human Clinical
"Higher preoperative interleukin-6 was associated with postoperative delirium with a standardised mean difference (95% confidence interval) of 0.33 (0.11-0.56) and P=0.003."
Meta-analytic evidence that the systemic inflammatory state indexed by this node predicts the syndrome.
Blood-Brain Barrier Permeability Increase
The permissive step: an inflamed cerebral endothelium lets peripheral inflammatory signalling into the CNS compartment. It is curated as its own node because it is separately measurable in patients — the CSF:plasma albumin ratio and plasma S100B correlate with the CSF soluble IL-6 receptor that indexes central trans-signalling — rather than being inferred from the peripheral cytokine level alone.
Peripheral inflammation reaching the CNS through a permeable blood-brain barrier
regulation of vascular permeability GO:0043114 Gene Ontology (GO) Relation: this pathophysiological event involves this biological process This pathophysiological event involves increased regulation of vascular permeability (GO:0043114). GO:0043114 is a biological process from the Gene Ontology. ↑ INCREASED
brain UBERON:0000955 Uberon multi-species anatomy ontology (UBERON) Relation: this pathophysiological event occurs in this anatomical location This pathophysiological event occurs in brain (UBERON:0000955). UBERON:0000955 is an anatomical location from the Uberon multi-species anatomy ontology.
Downstream
Microglial Activation and Neuroinflammation Entry of peripheral mediators into the CNS compartment activates parenchymal microglia. Peripheral inflammation reaching the CNS through a permeable blood-brain barrier
Show evidence (1 reference)
PMID:42011187 SUPPORT Human Clinical
"CSF sIL-6R was also significantly correlated with peripheral inflammation (plasma IL-6 and IL-8) and blood-brain barrier permeability (CSF:plasma albumin ratio and plasma S100B)."
The measurement that makes this a curatable node rather than an assumed intermediate.
Microglial Activation and Neuroinflammation
Central amplification of the peripheral signal. Microglia switch from a surveillant to an activated phenotype and release IL-1beta, TNF-alpha, and other mediators locally; astrocytic activation accompanies it, and is reflected clinically in the S100B rise curated under biochemical. The conditional nature of this node is what distinguishes delirium from ordinary sickness behaviour: in animals, an identical systemic cytokine challenge leaves working memory intact on a normal background and impairs it acutely on a degenerating one.
Loss of large-scale cortical feedback connectivity as the proximate substrate of inattention Peripheral inflammation reaching the CNS through a permeable blood-brain barrier Relative cholinergic deficiency with dopaminergic excess
microglial cell CL:0000129 Cell Ontology (CL) Relation: this pathophysiological event involves this cell type This pathophysiological event involves microglial cell (CL:0000129). CL:0000129 is a cell type from the Cell Ontology. astrocyte CL:0000127 Cell Ontology (CL) Relation: this pathophysiological event involves this cell type This pathophysiological event involves astrocyte (CL:0000127). CL:0000127 is a cell type from the Cell Ontology.
microglial cell activation GO:0001774 Gene Ontology (GO) Relation: this pathophysiological event involves this biological process This pathophysiological event involves increased microglial cell activation (GO:0001774). GO:0001774 is a biological process from the Gene Ontology. ↑ INCREASED neuroinflammatory response GO:0150076 Gene Ontology (GO) Relation: this pathophysiological event involves this biological process This pathophysiological event involves increased neuroinflammatory response (GO:0150076). GO:0150076 is a biological process from the Gene Ontology. ↑ INCREASED
Animal models
Systemic TNF-alpha challenge on a neurodegenerative background (ME7 prion mouse) Mouse partially recapitulates Reproduces the conditional central inflammatory response — the same peripheral stimulus, a larger central response and an acute cognitive deficit only on the degenerating background.
LPS-induced neuroinflammation and cognitive impairment in mice Mouse recapitulates Reproduces the systemic-insult-to-microglial-activation step directly, with the hippocampal cellular and cytokine readouts that cannot be obtained in patients, and with a TLR4 blockade arm that establishes the direction of causation.
Downstream
Cholinergic-Dopaminergic Neurotransmitter Imbalance Neuroinflammation is proposed to perturb cholinergic and dopaminergic transmission; the intermediates are not established in humans, and the edge is recorded at that strength rather than as a direct effect. Relative cholinergic deficiency with dopaminergic excess
Cortical Feedback Connectivity Loss The inflammatory and circuit accounts meet here: feedback connectivity falls as systemic inflammation rises, but the pathway from cytokine to connectivity is not resolved. Peripheral inflammation reaching the CNS through a permeable blood-brain barrier Loss of large-scale cortical feedback connectivity as the proximate substrate of inattention
Neuronal and Astrocytic Injury Sustained activation is the proposed route from a reversible state to the injury that underlies incomplete recovery. Peripheral inflammation reaching the CNS through a permeable blood-brain barrier
Show evidence (1 reference)
PMID:27633985 SUPPORT Model Organism
"TNF-α (50μg/kg) had no impact on performance of normal animals (normal brain homogenate; NBH) on working memory (T-maze) but produced acute impairments in ME7 animals similarly challenged."
Demonstrates the vulnerability-conditional central response that this node asserts, with the systemic stimulus held constant between groups.
Cholinergic-Dopaminergic Neurotransmitter Imbalance
Relative cholinergic deficiency with dopaminergic excess is the classical proximate account, and it remains the best explanation for drug-precipitated delirium: anticholinergic and dopaminergic agents are among the most reliable precipitants, and their withdrawal is part of effective management. It is curated as a real node with a failed therapeutic corollary — neither cholinesterase inhibition nor dopamine-receptor blockade shortened delirium in randomized trials — so the imbalance should not be read as a validated drug target.
Relative cholinergic deficiency with dopaminergic excess
synaptic transmission, cholinergic GO:0007271 Gene Ontology (GO) Relation: this pathophysiological event involves this biological process This pathophysiological event involves decreased synaptic transmission, cholinergic (GO:0007271). GO:0007271 is a biological process from the Gene Ontology. ↓ DECREASED
Downstream
Cortical Feedback Connectivity Loss Cholinergic tone gates cortical signal-to-noise and attention; the link to the measured connectivity failure is plausible but not established in humans. Relative cholinergic deficiency with dopaminergic excess
Show evidence (1 reference)
PMID:21056464 SUPPORT Human Clinical
"Impaired cholinergic neurotransmission seems to have an important role in the development of delirium."
Supports the existence of the cholinergic disturbance this node names, stated as the rationale of the trial that then failed to exploit it.
Cortical Feedback Connectivity Loss
The circuit-level state of the delirious brain and the proximate substrate of its cardinal sign. Dynamic causal modelling of auditory evoked responses shows reduced top-down (feedback) connectivity during delirium, with increased intrinsic inhibitory tone in the same region; the reduction scales with delirium severity and with systemic inflammation. Framed predictively, the delirious cortex fails to impose top-down expectations on sensory input — which is what inattention, disorganized thinking, and perceptual disturbance look like from the inside.
Loss of large-scale cortical feedback connectivity as the proximate substrate of inattention
cerebral cortex UBERON:0000956 Uberon multi-species anatomy ontology (UBERON) Relation: this pathophysiological event occurs in this anatomical location This pathophysiological event occurs in cerebral cortex (UBERON:0000956). UBERON:0000956 is an anatomical location from the Uberon multi-species anatomy ontology.
Downstream
Protracted and Persistent Course Whether failure of the network state to re-establish itself is what makes a delirium subacute or persistent is unresolved; the edge records the candidate relationship without asserting it. Loss of large-scale cortical feedback connectivity as the proximate substrate of inattention
Show evidence (1 reference)
PMID:37695013 SUPPORT Human Clinical
"Delirium is associated with decreased feedback cortical connectivity, possibly resulting from increased intrinsic inhibitory tone."
The study's own summary of the connectivity state this node describes.
Neuronal and Astrocytic Injury
The arm that makes delirium more than a transient state. Astrocytic (S100B) and axonal (neurofilament light) injury markers rise in association with delirium, and at the clinical level a longer duration of delirium predicts worse global cognition and executive function at 3 and 12 months, with delirium independently associated with long-term cognitive decline across studies. The direction of causation is not fully settled — delirium is at minimum a marker of a vulnerable brain — but the dose-response with delirium duration is what motivates curating injury as a distinct node.
astrocyte CL:0000127 Cell Ontology (CL) Relation: this pathophysiological event involves this cell type This pathophysiological event involves astrocyte (CL:0000127). CL:0000127 is a cell type from the Cell Ontology.
Show evidence (4 references)
PMID:24088092 SUPPORT Human Clinical
"A longer duration of delirium was independently associated with worse global cognition at 3 and 12 months (P=0.001 and P=0.04, respectively) and worse executive function at 3 and 12 months (P=0.004 and P=0.007, respectively)."
The duration-response relationship that argues for delirium contributing to injury rather than only marking prior vulnerability.
PMID:32658246 SUPPORT Human Clinical
"There was a significant association between delirium and long-term cognitive decline, as the estimated effect size (Hedges g) for 23 studies was 0.45 (95% CI, 0.34-0.57; P < .001)."
Pooled magnitude of the long-term cognitive consequence attributed to this node.
PMID:40889075 SUPPORT Human Clinical
"Among CNS injury markers, S100β and neurofilament light chain (NfL) demonstrated the most consistent associations with delirium presence and severity, supporting a role for astrocytic and axonal injury in delirium pathogenesis."
The systematic review that identifies astrocytic and axonal injury markers as the most consistent biomarker signal — the molecular evidence for this node, and the reason it names both cell compartments.
+ 1 more reference
Protracted and Persistent Course
The node that gives this entry its name. Where a classical delirium declares itself in hours and resolves with its precipitant, the subacute presentation evolves over days and often does not resolve on the expected timescale: among older hospital patients who become delirious, nearly half are still delirious at discharge and a fifth remain so at six months. Because the presentation is gradual and the deficits — inattention, disorientation, impaired memory — are the ones dementia also produces, the subacute form is preferentially misread as an evolving dementia rather than as an acute brain state with a treatable cause.
Peripheral inflammation reaching the CNS through a permeable blood-brain barrier
Downstream
Neuronal and Astrocytic Injury Duration of delirium is the exposure variable that predicts long-term cognitive outcome, so a protracted course is curated as feeding the injury node rather than merely coexisting with it. Peripheral inflammation reaching the CNS through a permeable blood-brain barrier
Show evidence (2 references)
PMID:19017678 SUPPORT Human Clinical
"the combined proportions with PerD at discharge, 1, 3 and 6 months were 44.7% (95% CI 26.8%, 63.7%), 32.8% (95% CI 18.4%, 47.2%), 25.6% (95% CI 7.9%, 43.4%) and 21% (95% CI 1.4%, 40.6%), respectively."
Quantifies the protracted course this node stands for, at four follow-up points.
PMID:19017678 SUPPORT Human Clinical
"The outcomes (mortality, nursing home placement, function, cognition) of patients with PerD were consistently worse than the outcomes of patients who had recovered from delirium."
Establishes that the protracted course is prognostically distinct, not just a longer version of the same illness.
⬡

Pathograph

Use the checkboxes to hide or show graph categories. Hover nodes for evidence and cross-linked metadata.
Pathograph: causal mechanism network for Subacute Delirium Interactive directed graph showing how pathophysiology mechanisms, phenotypes, genetic factors and variants, experimental models, environmental triggers, and treatments relate through causal and linked edges.
●

Phenotypes

8
Nervous System 7
Delirium HP:0031258 Human Phenotype Ontology (HP) Relation: this clinical feature is this phenotype This clinical feature is Delirium (HP:0031258), qualified as temporality subacute. HP:0031258 is a phenotype from the Human Phenotype Ontology.
Temporal: SUBACUTE
Show evidence (1 reference)
PMID:23992774 SUPPORT Human Clinical
"Delirium is an acute disorder of attention and cognition in elderly people (ie, those aged 65 years or older) that is common, serious, costly, under-recognised, and often fatal."
Definitional statement of the syndrome this phenotype records.
Confusion and disorientation HP:0001289 Human Phenotype Ontology (HP) Relation: this clinical feature is this phenotype This clinical feature is Confusion (HP:0001289). HP:0001289 is a phenotype from the Human Phenotype Ontology.
Show evidence (1 reference)
PMID:23992774 SUPPORT Human Clinical
"Key diagnostic features include an acute onset and fluctuating course of symptoms, inattention, impaired level of consciousness, and disturbance of cognition (e.g., disorientation, memory impairment, alteration in language)."
Names disorientation among the cognitive disturbances that define the syndrome.
Memory impairment HP:0002354 Human Phenotype Ontology (HP) Relation: this clinical feature is this phenotype This clinical feature is Memory impairment (HP:0002354). HP:0002354 is a phenotype from the Human Phenotype Ontology.
Show evidence (1 reference)
PMID:23992774 SUPPORT Human Clinical
"Key diagnostic features include an acute onset and fluctuating course of symptoms, inattention, impaired level of consciousness, and disturbance of cognition (e.g., disorientation, memory impairment, alteration in language)."
Lists memory impairment among the defining cognitive disturbances.
Perceptual disturbance and hallucinations HP:0000738 Human Phenotype Ontology (HP) Relation: this clinical feature is this phenotype This clinical feature is Hallucinations (HP:0000738). HP:0000738 is a phenotype from the Human Phenotype Ontology.
Show evidence (1 reference)
PMID:23992774 SUPPORT Human Clinical
"Supportive features include disturbance in sleep-wake cycle, perceptual disturbances (hallucinations or illusions), delusions, psychomotor disturbance (hypo- or hyper-activity), inappropriate behavior, and emotional lability."
Names perceptual disturbance and hallucinations as recognized supportive features of delirium.
Psychomotor agitation HP:0000713 Human Phenotype Ontology (HP) Relation: this clinical feature is this phenotype This clinical feature is Agitation (HP:0000713). HP:0000713 is a phenotype from the Human Phenotype Ontology.
Show evidence (1 reference)
PMID:30346242 SUPPORT Human Clinical
"Delirium developed in 566 patients (48%), of whom 89% had hypoactive delirium and 11% had hyperactive delirium."
Ascertains both motor presentations in a defined cohort and gives their relative frequency in critical illness.
Hypoactive presentation with apathy and lethargy VERY_FREQUENT HP:0000741 Human Phenotype Ontology (HP) Relation: this clinical feature is this phenotype This clinical feature is Apathy (HP:0000741). HP:0000741 is a phenotype from the Human Phenotype Ontology.
Show evidence (1 reference)
PMID:30346242 SUPPORT Human Clinical
"Delirium developed in 566 patients (48%), of whom 89% had hypoactive delirium and 11% had hyperactive delirium."
Supports both the predominance of the hypoactive presentation and the VERY_FREQUENT band assigned here, within an ICU population.
Sleep-wake cycle disturbance HP:0006979 Human Phenotype Ontology (HP) Relation: this clinical feature is this phenotype This clinical feature is Sleep-wake cycle disturbance (HP:0006979). HP:0006979 is a phenotype from the Human Phenotype Ontology.
Show evidence (1 reference)
PMID:29278283 SUPPORT Human Clinical
"Delirium is the most common neuropsychiatric syndrome encountered by clinicians dealing with older adults and the medically ill and is best characterized by 5 core domains: cognitive deficits, attentional deficits, circadian rhythm dysregulation, emotional dysregulation, and alteration in..."
Establishes circadian rhythm dysregulation as one of the five core domains of the syndrome.
Other 1
Inattention Short attention span HP:0000736 Human Phenotype Ontology (HP) Relation: this clinical feature is this phenotype This clinical feature is Inattention, annotated with Short attention span (HP:0000736). HP:0000736 is a phenotype from the Human Phenotype Ontology.
HPO has no term for the fluctuating, acquired inattention of delirium; `Short attention span` is the closest available concept and is used with a `preferred_term` of Inattention rather than manufacturing a narrower match.
Show evidence (1 reference)
PMID:23992774 SUPPORT Human Clinical
"Key diagnostic features include an acute onset and fluctuating course of symptoms, inattention, impaired level of consciousness, and disturbance of cognition (e.g., disorientation, memory impairment, alteration in language)."
Lists inattention as a key diagnostic feature of the syndrome.
🧬

Genetic Associations

1
APOE
Gene: APOE hgnc:613 HUGO Gene Nomenclature Committee (hgnc) Relation: this disease-associated gene is this gene This disease-associated gene is APOE (hgnc:613). hgnc:613 is a gene from the HUGO Gene Nomenclature Committee. relationship_type: SUSCEPTIBILITY
Show evidence (2 references)
PMID:41286463 SUPPORT Human Clinical
"Here we conducted a genetic meta-analysis on delirium using multi-ancestry data from the UK Biobank, FinnGen, All of Us Research Program and Michigan Genomics Initiative cohorts (n = 1,059,130; 11,931 cases), yielding the Apolipoprotein E (APOE) gene as a strong delirium risk factor..."
The multi-ancestry meta-analysis establishing APOE as a delirium risk locus, and — crucially for the SUSCEPTIBILITY classification — independently of dementia rather than through it.
PMID:41286463 SUPPORT Human Clinical
"Combining proteins, APOE-ε4 status and demographics significantly improved incident delirium prediction compared to demographics alone."
Shows the association is large enough to add predictive information, which is what distinguishes a susceptibility locus from an incidental signal.
💊

Medical Actions

4
Multicomponent non-pharmacological delirium prevention
Action: supportive careNCI Thesaurus (NCIT) Relation: this treatment is this clinical intervention This treatment is supportive care (NCIT:C15747). NCIT:C15747 is a clinical intervention from the NCI Thesaurus. Ontology label: Supportive Care NCIT:C15747
The Hospital Elder Life Program and its descendants apply standardized protocols to six modifiable risk factors — cognitive impairment, sleep deprivation, immobility, visual impairment, hearing impairment, and dehydration. In the original controlled study delirium developed in 9.9% of the intervention group versus 15.0% of usual care, and pooled across subsequent studies the odds of incident delirium fall by roughly half. It is the only intervention with a consistent effect on incidence, and it is mechanistically a treatment of the predisposing node rather than of any molecular target — which is exactly what the multifactorial model predicts would work.
Mechanism Target:
MODULATES Reduced Cognitive Reserve — Each protocol component removes one of the predisposing or perpetuating factors that constitute this node — immobility, sensory deprivation, dehydration, sleep loss — rather than acting on the inflammatory or neurotransmitter arms.
Show evidence (1 reference)
PMID:10053175 SUPPORT Human Clinical
"The intervention consisted of standardized protocols for the management of six risk factors for delirium: cognitive impairment, sleep deprivation, immobility, visual impairment, hearing impairment, and dehydration."
Enumerates the targets of the intervention, which are the constituents of the node it is linked to.
Show evidence (3 references)
PMID:10053175 SUPPORT Human Clinical
"Delirium developed in 9.9 percent of the intervention group as compared with 15.0 percent of the usual-care group, (matched odds ratio, 0.60; 95 percent confidence interval, 0.39 to 0.92)."
The controlled effect estimate on incident delirium from the defining study.
PMID:25643002 SUPPORT Human Clinical
"Overall, 11 studies demonstrated significant reductions in delirium incidence (odds ratio [OR], 0.47; 95% CI, 0.38-0.58)."
Pooled effect across studies, establishing that the original result generalizes.
PMID:10053175 SUPPORT Human Clinical
"The intervention had no significant effect on the severity of delirium or on recurrence rates; this finding suggests that primary prevention of delirium is probably the most effective treatment strategy."
Bounds the claim: the effect is on incidence, not on the severity or recurrence of an episode that has already begun.
Treatment of the precipitating illness and withdrawal of deliriogenic drugs
Action: supportive careNCI Thesaurus (NCIT) Relation: this treatment is this clinical intervention This treatment is supportive care (NCIT:C15747). NCIT:C15747 is a clinical intervention from the NCI Thesaurus. Ontology label: Supportive Care NCIT:C15747
The mainstay of management of established delirium. No agent has been shown to shorten delirium in an unselected critically ill population — the antipsychotic and cholinesterase-inhibitor trials curated here are both negative — so treatment consists of identifying and correcting the precipitant, reducing sedative and analgesic burden, and applying the same non-pharmacological measures used in prevention. Dexmedetomidine is the exception that defines the rule: it benefited the specific population of non-intubated patients with hyperactive delirium, and is curated separately below.
Mechanism Target:
INHIBITS Systemic Inflammatory Insult — Resolving the infection, metabolic derangement, or surgical stress removes the precipitating insult at its source, which is the only intervention that acts on this node.
Show evidence (1 reference)
PMID:23992774 SUPPORT Human Clinical
"The initial management focuses on three simultaneous priorities: (1) maintaining patient safety; (2) searching for the causes; and (3) managing delirium symptoms."
Places the search for and treatment of the cause at the centre of management, which is what this link asserts.
Show evidence (1 reference)
PMID:23992774 SUPPORT Human Clinical
"Drug reduction for sedation and analgesia and non-pharmacological approaches are recommended."
The recommended management strategy that this treatment entry records.
Antipsychotics for severe agitation
Action: PharmacotherapyNCI Thesaurus (NCIT) Relation: this treatment is this clinical intervention This treatment is Pharmacotherapy (NCIT:C15986). NCIT:C15986 is a clinical intervention from the NCI Thesaurus. NCIT:C15986
Agent: haloperidol CHEBI:5613 Chemical Entities of Biological Interest (CHEBI) Relation: this treatment uses this therapeutic agent This treatment uses haloperidol (CHEBI:5613). CHEBI:5613 is a therapeutic agent from Chemical Entities of Biological Interest.
Haloperidol and atypical antipsychotics remain in use for the behavioural control of severe agitation that threatens patient or staff safety. They are curated here with the evidence that constrains them: in a placebo-controlled trial of 566 delirious critically ill patients, neither haloperidol nor ziprasidone altered the number of days alive without delirium or coma. They are symptomatic agents, not disease-modifying ones, and the entry deliberately does not link them to a mechanism node.
Show evidence (2 references)
PMID:30346242 REFUTE Human Clinical
"The use of haloperidol or ziprasidone, as compared with placebo, in patients with acute respiratory failure or shock and hypoactive or hyperactive delirium in the ICU did not significantly alter the duration of delirium."
Refutes any duration-modifying claim for antipsychotics in delirium, which is why this treatment is curated as symptomatic only.
PMID:23992774 SUPPORT Human Clinical
"No convincing evidence shows that pharmacological prevention or treatment is effective."
The general statement of the evidence position this treatment entry is curated against.
Dexmedetomidine for hyperactive delirium in non-intubated ICU patients
Action: PharmacotherapyNCI Thesaurus (NCIT) Relation: this treatment is this clinical intervention This treatment is Pharmacotherapy (NCIT:C15986). NCIT:C15986 is a clinical intervention from the NCI Thesaurus. NCIT:C15986
Agent: dexmedetomidine CHEBI:4466 Chemical Entities of Biological Interest (CHEBI) Relation: this treatment uses this therapeutic agent This treatment uses dexmedetomidine (CHEBI:4466). CHEBI:4466 is a therapeutic agent from Chemical Entities of Biological Interest.
An alpha-2 adrenergic agonist, and the one pharmacological agent in this entry with a positive randomized result. In a multicentre, double-blind, placebo-controlled trial across nine ICUs, non-intubated critically ill adults with hyperactive delirium randomized to dexmedetomidine did better than placebo on a joint model of agitation duration, delirium duration, and the need for intubation and deep sedation; the trial was stopped for efficacy at the preplanned interim analysis. The effect is real but should not be overread: the median reduction in agitation was about an hour, the other key secondary outcomes were similar between groups, and the population is a specific one — non-intubated patients with the hyperactive presentation, which is the minority form in critical illness.
Show evidence (2 references)
PMID:41160116 SUPPORT Human Clinical
"In non-intubated adult ICU patients with hyperactive delirium, the use of dexmedetomidine was associated with a greater clinical benefit than placebo for the joint modelling of multiple endpoints of agitation control, delirium resolution, and intubation with mechanical ventilation."
The randomized result establishing benefit in the specified population and on the specified composite endpoint.
PMID:41160116 SUPPORT Human Clinical
"Agitation duration was shorter in the dexmedetomidine group (1.0 (1.0-2.0) vs 2.0 (1.0-7.0) hours, absolute difference 95% CI - 1.0 (- 2.0 to - 0.1); ES, - 0.60; 95% CI, -0.92 to -0.27, p = 0.001). Other key secondary outcomes were similar between groups."
Bounds the claim with the effect size and the neutral secondary outcomes, which is why this entry does not present dexmedetomidine as disease-modifying.
🌍

Environmental Factors

2
Physical restraint use
Physical restraints carry the largest odds ratio of any risk factor in the contemporary meta-analysis of hospitalized older patients, and are iatrogenic and modifiable. The direction of causation is not fully separable from confounding by indication — agitated patients get restrained — but restraint use is also a component that multicomponent prevention programs deliberately remove.
Show evidence (1 reference)
PMID:39602991 SUPPORT Human Clinical
"The risk factors of delirium included frailty (odds ratio = 2.05), physical restraints (5.01), prior falls (1.99), severe illness (1.32) (evaluated using the Acute Physiology and Chronic Health Evaluation II), and cognitive impairment (2.61)."
Establishes restraint use as an independently reported risk factor with the largest pooled odds ratio in this analysis.
Mechanism Target:
EXACERBATES Reduced Cognitive Reserve — Restraint enforces immobility and sensory-environmental deprivation, compounding the predisposing state rather than acting as an inflammatory precipitant.
Show evidence (1 reference)
PMID:39602991 SUPPORT Human Clinical
"The risk factors of delirium included frailty (odds ratio = 2.05), physical restraints (5.01), prior falls (1.99), severe illness (1.32) (evaluated using the Acute Physiology and Chronic Health Evaluation II), and cognitive impairment (2.61)."
Gives the restraint-associated odds ratio in the pooled analysis this link rests on.
Deliriogenic medication exposure
Anticholinergic drugs, benzodiazepines, and opioids are among the most reliable precipitants, and the classic demonstration of the vulnerability model is that a single dose of a sedative can precipitate delirium in a patient with dementia and multimorbidity. Reduction of sedative and analgesic burden is a recommended management step.
Show evidence (1 reference)
PMID:23992774 SUPPORT Human Clinical
"Drug reduction for sedation and analgesia and non-pharmacological approaches are recommended."
The management recommendation that follows from treating these agents as modifiable precipitants.
Mechanism Target:
TRIGGERS Cholinergic-Dopaminergic Neurotransmitter Imbalance — Anticholinergic and dopaminergic agents act directly on the transmitter systems this node describes, which is the strongest evidence that the imbalance is mechanistically real even though correcting it downstream does not treat the syndrome.
Show evidence (1 reference)
PMID:23992774 SUPPORT Human Clinical
"The list of potential neurotransmitters involved in delirium is long,60 but a relative cholinergic deficiency and/or dopamine excess are the most commonly inferred,61, 62 correlating with the adverse effects of anticholinergic or dopaminergic drugs."
States the drug-effect reasoning that links this exposure to the transmitter node.
🔬

Biochemical Markers

2
Serum S100B
Show evidence (3 references)
PMID:42011187 SUPPORT Human Clinical
"CSF sIL-6R was also significantly correlated with peripheral inflammation (plasma IL-6 and IL-8) and blood-brain barrier permeability (CSF:plasma albumin ratio and plasma S100B)."
Uses plasma S100B as one of the two indices of barrier permeability, which is the role this marker plays in the entry.
PMID:40889075 SUPPORT Human Clinical
"Among CNS injury markers, S100β and neurofilament light chain (NfL) demonstrated the most consistent associations with delirium presence and severity, supporting a role for astrocytic and axonal injury in delirium pathogenesis."
Establishes S100B as one of the two most consistent CNS-injury markers across the critical-illness literature.
PMID:40889075 SUPPORT Human Clinical
"Despite early promise, serum biomarkers for ICU delirium remain investigational and require further validation before clinical application."
Bounds the claim: this is a research marker, not a diagnostic test, which is how the entry curates it.
Preoperative interleukin-6
Show evidence (2 references)
PMID:34218905 SUPPORT Human Clinical
"Higher preoperative interleukin-6 was associated with postoperative delirium with a standardised mean difference (95% confidence interval) of 0.33 (0.11-0.56) and P=0.003."
The pooled estimate that supports IL-6 as a pre-insult risk marker.
PMID:34218905 SUPPORT Human Clinical
"The association of preoperative blood levels of inflammatory mediators with postoperative delirium may be influenced by the type of surgery and the specific mediator."
Qualifies the marker: the association is conditional on context, which is why it is not curated as diagnostic.
🔬

Diagnosis

2
Confusion Assessment Method (CAM)
The standard bedside instrument, structured around acute onset with a fluctuating course plus inattention, and either disorganized thinking or an altered level of consciousness. Its ICU adaptation, the CAM-ICU, is what ascertains delirium in ventilated patients who cannot be interviewed, and is the instrument used in the major critical-care trials curated here.
Show evidence (2 references)
PMID:23992774 SUPPORT Human Clinical
"validated in high quality studies including over 1000 patients with sensitivity of 94%, specificity of 89%, and high inter-rater reliability."
Reports the CAM's validated operating characteristics in the sentence that names it the most widely used instrument for identifying delirium.
PMID:30346242 SUPPORT Human Clinical
"delirium was detected with the use of the Confusion Assessment Method for the ICU (CAM-ICU),23,24 a validated tool that identifies delirium on the basis of an acute change or fluctuating course of mental status plus inattention and either altered level of consciousness or disorganized thinking"
Describes the ICU adaptation and its operational criteria as applied in a definitive trial.
Search for the precipitating cause
There is no diagnostic test for delirium itself; the diagnostic work is identifying what precipitated it. Because delirium can be the presenting feature of a medical emergency, screening for acute physiological disturbance — hypoxaemia, hypoglycaemia, hypercapnia — precedes any elaborate work-up, and testing is targeted by history and examination rather than run as a battery.
Show evidence (1 reference)
PMID:23992774 SUPPORT Human Clinical
"First, because delirium can be the harbinger of a medical emergency, every patient presenting with delirium should be screened for acute physiologic disturbance such as hypoxemia, low blood glucose, and high arterial carbon dioxide."
States the priority and the specific targets of the initial diagnostic screen.
📈

Progression

2
Persistence at hospital discharge
44.7% of older hospital patients with delirium were still delirious at discharge (95% CI 26.8%-63.7%).
Show evidence (1 reference)
PMID:19017678 SUPPORT Human Clinical
"the combined proportions with PerD at discharge, 1, 3 and 6 months were 44.7% (95% CI 26.8%, 63.7%), 32.8% (95% CI 18.4%, 47.2%), 25.6% (95% CI 7.9%, 43.4%) and 21% (95% CI 1.4%, 40.6%), respectively."
Reports the discharge-time persistence proportion recorded in this phase.
Persistence at six months
21% remained delirious at six months (95% CI 1.4%-40.6%); the wide interval reflects heterogeneity across the 18 pooled reports.
Show evidence (1 reference)
PMID:19017678 SUPPORT Human Clinical
"the combined proportions with PerD at discharge, 1, 3 and 6 months were 44.7% (95% CI 26.8%, 63.7%), 32.8% (95% CI 18.4%, 47.2%), 25.6% (95% CI 7.9%, 43.4%) and 21% (95% CI 1.4%, 40.6%), respectively."
Reports the six-month persistence proportion recorded in this phase.
📊

Prevalence

1
Medically hospitalized older patients
Point Prevalence 23600.0 per 100,000 (19000.0–29000.0) >1 in 1,000
Pooled prevalence 23.6% (95% CI 19%-29%) across 35 studies and 12,097 participants. Reported alongside a pooled incidence of 13.5%, which is a different measure and is not combined with it here.
Show evidence (1 reference)
PMID:39602991 SUPPORT Human Clinical
"A total of 35 studies encompassing 12,097 participants met the inclusion criteria, yielding a 23.6 % pooled prevalence of delirium (95 % confidence interval = 19 % to 29 %) and a 13.5 % pooled incidence of delirium (n = 32, 95 % confidence interval = 11 % to 17 %) among medically hospitalized..."
Source of both the pooled prevalence recorded here and the incidence noted for contrast.
⚖️

Clinical Burden

High
Delirium is common in hospitalized older adults, is independently associated with mortality, and — in the protracted form this entry centres on — persists in a substantial minority for months and carries consistently worse mortality, institutionalization, function, and cognition than a delirium that resolves. It is additionally associated with long-term cognitive decline of moderate effect size.
Show evidence (2 references)
PMID:23992774 SUPPORT Human Clinical
"Delirium is an acute disorder of attention and cognition in elderly people (ie, those aged 65 years or older) that is common, serious, costly, under-recognised, and often fatal."
Summary statement of the burden this assessment records.
PMID:32658246 SUPPORT Human Clinical
"In this meta-analysis, delirium was significantly associated with long-term cognitive decline in both surgical and nonsurgical patients."
Supports the long-term component of the burden assessment.
🔀

Differential Diagnoses

2

Conditions with similar clinical presentations that must be differentiated from Subacute Delirium:

Dementia
Overlapping Features The differential that matters most for the subacute presentation, because an insidious onset over days to weeks removes the abruptness that usually separates the two. Both produce inattention, disorientation, and memory impairment.
Distinguishing Features
  • Delirium is distinguished by its acute change from a known baseline, its fluctuating course across the day, and disturbance of the level of consciousness — none of which characterize uncomplicated dementia. Establishing the patient's baseline from an informant is the decisive step, and dementia is itself the strongest predisposing factor, so the two frequently coexist as delirium superimposed on dementia.
Show evidence (1 reference)
PMID:23992774 SUPPORT Human Clinical
"To differentiate delirium from dementia, obtaining the history is critical to establish the patient's baseline, determine the acuity of mental status change and fluctuations typical of delirium, and to search for etiologic clues."
States the distinguishing procedure and the features it turns on.
Depression
Overlapping Features Hypoactive delirium — withdrawal, psychomotor slowing, reduced engagement — is readily mistaken for depression, particularly in an older inpatient with a depression history.
Distinguishing Features
  • Depression does not produce the fluctuating inattention or the altered level of consciousness that define delirium, and lacks an acute precipitant on work-up. Formal cognitive testing that demonstrates inattention is the practical discriminator.
Show evidence (1 reference)
PMID:30346242 SUPPORT Human Clinical
"hypoactive delirium prevents participation in nursing interventions, physical therapy, and occupational therapy"
Describes the quiet, withdrawn presentation that generates the confusion with depression.
🔬

Clinical Trials

1
NCT03317067 PHASE_III COMPLETED
The 4D trial: a multicentre, double-blind, placebo-controlled, investigator-initiated randomized trial of continuous intravenous dexmedetomidine versus placebo for hyperactive delirium in non-intubated critically ill adults, across nine ICUs, stopped early for efficacy at the preplanned interim analysis.
Target Phenotypes: Agitation HP:0000713 Human Phenotype Ontology (HP) Relation: this clinical trial targets this phenotype This clinical trial targets Agitation (HP:0000713). HP:0000713 is a phenotype from the Human Phenotype Ontology. Delirium HP:0031258 Human Phenotype Ontology (HP) Relation: this clinical trial targets this phenotype This clinical trial targets Delirium (HP:0031258). HP:0031258 is a phenotype from the Human Phenotype Ontology.
Show evidence (2 references)
clinicaltrials:NCT03317067 SUPPORT Other
"The primary purpose of the study is to evaluate whether dexmedetomidine is effective in treating agitated delirium of non-intubated ICI patients"
The registration record establishing the trial's identity and objective; the results are cited from the publication rather than from here.
PMID:41160116 SUPPORT Human Clinical
"The study was stopped for efficacy at the time of the preplanned interim analysis."
Records the trial's early termination for efficacy, which is part of how its result should be interpreted.
🐁

Animal Models

2
Systemic TNF-alpha challenge on a neurodegenerative background (ME7 prion mouse)
The model that operationalizes vulnerability rather than the insult. An identical systemic cytokine challenge is given to normal mice and to mice with established prion neurodegeneration; only the latter show acute working-memory impairment, with higher hippocampal and hypothalamic cytokine transcription despite equivalent plasma TNF-alpha and CCL2. This is the vulnerability × precipitant model reduced to a single controlled experiment.
Species
Mouse
Genotype
ME7 prion-inoculated C57BL/6, challenged with intraperitoneal TNF-alpha
Publication
PMID:27633985
Pathograph links
Microglial Activation and Neuroinflammation partially recapitulates fidelity: moderate Reproduces the conditional central inflammatory response — the same peripheral stimulus, a larger central response and an acute cognitive deficit only on the degenerating background.
Limitations: The cognitive readout is T-maze working memory, not the attentional disturbance that defines human delirium, and the background pathology is prion disease rather than the Alzheimer-type pathology that predisposes most human patients. The model also shows sickness behaviour that overlaps with, but is not, delirium.
Readouts
T-maze working memory performance after systemic TNF-alpha
Direction: DECREASED
Interpretation: Acute cognitive impairment appearing only on the degenerating background is the model's operational stand-in for delirium.
Show evidence (1 reference)
PMID:27633985 SUPPORT Model Organism
"TNF-α (50μg/kg) had no impact on performance of normal animals (normal brain homogenate; NBH) on working memory (T-maze) but produced acute impairments in ME7 animals similarly challenged."
Reports the measurement and its direction, with the crucial negative control on the normal background.
Show evidence (1 reference)
PMID:27633985 SUPPORT Model Organism
"The data indicate that acutely elevated TNF-α has robust acute effects on brain function, selectively in the degenerating brain, but more sustained levels may be required to significantly impact on underlying neurodegeneration."
The authors' own summary of the selectivity that makes this model informative for the vulnerability-conditional node.
Show evidence (1 reference)
PMID:27633985 SUPPORT Model Organism
"Systemic inflammation caused by infection, trauma or co-morbidity can alter the brain's inflammatory status, produce acute cognitive impairments, such as delirium, and drive new pathology and accelerated decline."
States the delirium-relevant rationale that makes this model informative for the entry rather than a general neurodegeneration experiment.
LPS-induced neuroinflammation and cognitive impairment in mice
The dominant experimental proxy for infection-triggered delirium. Lipopolysaccharide produces sickness behaviour with cognitive impairment alongside hippocampal microglial activation, neuronal loss, and a shift in the cytokine milieu — TNF-alpha, IL-1beta, PGE2 and nitric oxide up, IL-4 and IL-10 down — with NF-kB pathway activation. The model earns its place in this entry because it is not merely descriptive: a TLR4-specific inhibitory peptide prevents both the neuroinflammation and the cognitive impairment, which makes the inflammatory step causal rather than correlated.
Species
Mouse
Genotype
Wild-type C57BL/6J challenged with lipopolysaccharide
Publication
PMID:30962497
Pathograph links
Microglial Activation and Neuroinflammation recapitulates fidelity: moderate Reproduces the systemic-insult-to-microglial-activation step directly, with the hippocampal cellular and cytokine readouts that cannot be obtained in patients, and with a TLR4 blockade arm that establishes the direction of causation.
Limitations: The behavioural readouts are Morris water maze and passive avoidance — memory tasks, not measures of attention or awareness, which are what define human delirium. The LPS challenge is also an acute bolus in a healthy young animal, so the model speaks to the precipitant without the reduced-reserve substrate that the human syndrome requires; the ME7 models curated elsewhere in this entry supply that half.
Readouts
Hippocampal microglial activation (IBA-1) and neuronal loss (MAP-2)
Direction: INCREASED
Interpretation: Microglial activation with concurrent neuronal loss in the hippocampus is the cellular signature this node asserts.
Show evidence (1 reference)
PMID:30962497 SUPPORT Model Organism
"We found that LPS treatment leads to sickness behavior and cognitive impairment in mice as shown in the Morris water maze and passive avoidance test, and these effects were accompanied by microglia activation (labeled by ionized calcium binding adaptor molecule-1, IBA-1) and neuronal cell loss..."
Reports the measurement, the markers used, and the accompanying behavioural change.
Cognitive impairment after TLR4 blockade
Direction: ABOLISHED
Interpretation: Preventing the deficit by blocking the receptor that initiates the inflammatory cascade is what makes the model causal rather than descriptive.
Show evidence (1 reference)
PMID:30962497 SUPPORT Model Organism
"Furthermore, VIPER, which is a TLR-4-specific inhibitory peptide, prevented the LPS-induced neuroinflammation and cognitive impairment."
The rescue arm establishing that the inflammatory step is necessary for the cognitive deficit.
Show evidence (1 reference)
PMID:30962497 SUPPORT Model Organism
"These data suggest that LPS induced cognitive impairment and neuroinflammation via microglia activation by activating the NF-kB signaling pathway"
The authors' mechanistic summary, which is the claim this model contributes to the entry's microglial node.
{ }

Source YAML

click to show 
name: Subacute Delirium
creation_date: '2026-08-25T00:00:00Z'
category: Neurologic
synonyms:
- subacute confusional state
- subacute brain syndrome
- subacute psycho-organic syndrome
- delirium, subacute
description: >-
  Subacute delirium is delirium — an acute disturbance of attention and
  awareness with an additional cognitive disturbance, developing over a short
  period and fluctuating in severity across the day — in which the onset is
  insidious (days to a week or two rather than hours) and the course is
  protracted. It is not a mechanistically distinct disease: MONDO:0004629 sits
  in the same concept space as ICD-10 F05, whose inclusion terms are explicitly
  "acute or subacute confusional state", "acute or subacute brain syndrome",
  and "acute or subacute psycho-organic syndrome". This entry therefore curates
  the delirium pathograph and treats "subacute" as the temporal qualifier on
  that pathograph — the presentation that is most easily mistaken for evolving
  dementia and that shades into the persistent delirium recorded under
  `progression`.

  The mechanism is convergent rather than linear. A predisposed brain — aged,
  demented, frail, or otherwise reduced in reserve — meets a precipitant
  (systemic infection, surgery, metabolic derangement, hypoxia, or a
  deliriogenic drug). Peripheral inflammatory signalling reaches the CNS,
  in part through an increasingly permeable blood-brain barrier, activating
  microglia; neurotransmitter systems (classically a relative cholinergic
  deficiency with dopaminergic excess) are perturbed; and large-scale cortical
  networks lose the feedback connectivity that supports attention and
  perceptual inference. Because the vulnerability and the insult multiply
  rather than add, a trivial insult suffices in a vulnerable patient and no
  single node is the disease.

  Two facts discipline the entry. First, the causal-therapeutic translation of
  the neurotransmitter model has failed: cholinesterase inhibition did not
  shorten delirium and signalled harm, and antipsychotics did not alter its
  duration in a definitive placebo-controlled trial — though dexmedetomidine
  did benefit the narrower population of non-intubated patients with
  hyperactive delirium, so the nihilism is about mechanism-directed drugs
  rather than about pharmacotherapy as such. Second, delirium is not
  the benign, fully reversible state it was long taken to be — a substantial
  minority remain delirious months later, and delirium is independently
  associated with long-term cognitive decline. Prevention by multicomponent
  non-pharmacological programs remains the only intervention with consistent
  effect on incidence.
disease_term:
  preferred_term: subacute delirium
  term:
    id: MONDO:0004629
    label: subacute delirium
parents:
- Neurocognitive Disorder
classifications:
  harrisons_chapter:
  - classification_value: NEUROLOGIC
references:
- reference: PMID:23992774
  title: Delirium in elderly people.
- reference: PMID:29278283
  title: 'Delirium pathophysiology: An updated hypothesis of the etiology of acute brain failure.'
mechanistic_hypotheses:
- hypothesis_group_id: vulnerability_precipitant_model
  hypothesis_label: Delirium as the product of baseline vulnerability multiplied by an acute precipitant
  status: CANONICAL
  description: >-
    The multifactorial model is the organizing framework for delirium and the
    one this entry's pathograph is built on: predisposing factors (age,
    dementia, frailty, sensory impairment) set the threshold, and precipitating
    insults (infection, surgery, drugs, metabolic derangement) cross it. Its
    strength is that it explains why a trivial insult suffices in one patient
    and a series of major ones is needed in another, and why single-target
    interventions fail while multicomponent ones work. Its weakness as a
    *mechanistic* model is that it is a risk architecture rather than a
    molecular pathway — it says which combinations produce delirium, not what
    the brain does when they do.
  evidence:
  - reference: PMID:29278283
    reference_title: 'Delirium pathophysiology: An updated hypothesis of the etiology of acute brain failure.'
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      Available data suggest that numerous pathological factors may serve as
      precipitants for delirium, each having differential effects depending on
      patient-specific patient physiological characteristics (substrate).
    explanation: >-
      States the precipitant × substrate structure that this hypothesis group
      organizes the causal edges around.
  - reference: PMID:23992774
    reference_title: Delirium in elderly people.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      In view of the complex multifactorial causes of delirium, multicomponent
      non-pharmacological risk factor approaches are the most effective strategy
      for prevention.
    explanation: >-
      The therapeutic corollary of the model: because causation is multifactorial,
      the intervention that works is multicomponent rather than single-target.
- hypothesis_group_id: neuroinflammatory_model
  hypothesis_label: Peripheral inflammation reaching the CNS through a permeable blood-brain barrier
  status: CANONICAL
  description: >-
    The best-evidenced molecular account of how a systemic insult becomes a
    brain state. Peripheral pro-inflammatory signalling — IL-6 prominently —
    is associated with delirium risk before the insult and with delirium
    severity after it, and the same studies tie the CNS compartment of that
    signalling to measured blood-brain barrier permeability. Animal work adds
    the crucial conditional: the same cytokine challenge is cognitively inert
    in a normal brain and disabling in a degenerating one, which is the
    vulnerability × precipitant model expressed at the level of a single
    molecule.
  evidence:
  - reference: PMID:34218905
    reference_title: 'Preoperative inflammatory mediators and postoperative delirium: systematic review and meta-analysis.'
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      Higher preoperative interleukin-6 was associated with postoperative delirium
      with a standardised mean difference (95% confidence interval) of 0.33
      (0.11-0.56) and P=0.003.
    explanation: >-
      Quantifies the inflammation-delirium association in the direction the
      hypothesis requires — inflammatory tone before the insult predicts the
      syndrome after it.
  - reference: PMID:42011187
    reference_title: 'The association between cerebrospinal fluid interleukin-6, soluble interleukin-6 receptor, and interleukin-6 trans-signalling binary complex with postoperative delirium: an observational cohort study.'
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      CSF sIL-6R was also significantly correlated with peripheral inflammation
      (plasma IL-6 and IL-8) and blood-brain barrier permeability (CSF:plasma
      albumin ratio and plasma S100B).
    explanation: >-
      Links the peripheral and central inflammatory compartments through a
      measured barrier-permeability index, which is the step this hypothesis
      turns on.
- hypothesis_group_id: cholinergic_dopaminergic_imbalance_model
  hypothesis_label: Relative cholinergic deficiency with dopaminergic excess
  status: ALTERNATIVE
  description: >-
    The oldest mechanistic account of delirium, motivated by the deliriogenic
    potency of anticholinergic and dopaminergic drugs. It is retained here as
    ALTERNATIVE rather than CANONICAL because its two strongest predictions
    have failed. Therapeutically, augmenting cholinergic transmission with
    rivastigmine did not shorten delirium in critically ill patients and the
    trial was stopped early for excess mortality in the treatment arm; and
    dopamine-receptor blockade with haloperidol or ziprasidone did not alter
    delirium duration against placebo. The neurotransmitter disturbance is
    real and is the best explanation for drug-precipitated delirium; what the
    negative trials refute is that correcting it downstream reverses the
    syndrome.
  evidence:
  - reference: PMID:21056464
    reference_title: 'Effect of rivastigmine as an adjunct to usual care with haloperidol on duration of delirium and mortality in critically ill patients: a multicentre, double-blind, placebo-controlled randomised trial.'
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      Impaired cholinergic neurotransmission seems to have an important role in
      the development of delirium.
    explanation: >-
      States the hypothesis as the trial's own rationale, which is what makes
      the trial's negative result a test of it rather than an unrelated finding.
  - reference: PMID:21056464
    reference_title: 'Effect of rivastigmine as an adjunct to usual care with haloperidol on duration of delirium and mortality in critically ill patients: a multicentre, double-blind, placebo-controlled randomised trial.'
    supports: REFUTE
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      Rivastigmine did not decrease duration of delirium and might have increased
      mortality so we do not recommend use of rivastigmine to treat delirium in
      critically ill patients.
    explanation: >-
      Refutes the therapeutic prediction of the cholinergic-deficiency model —
      restoring cholinergic tone neither shortened delirium nor was safe.
  - reference: PMID:30346242
    reference_title: Haloperidol and Ziprasidone for Treatment of Delirium in Critical Illness.
    supports: REFUTE
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      The use of haloperidol or ziprasidone, as compared with placebo, in patients
      with acute respiratory failure or shock and hypoactive or hyperactive
      delirium in the ICU did not significantly alter the duration of delirium.
    explanation: >-
      Refutes the mirror-image therapeutic prediction — blocking the dopaminergic
      arm did not change the course of the syndrome.
- hypothesis_group_id: network_disconnection_model
  hypothesis_label: Loss of large-scale cortical feedback connectivity as the proximate substrate of inattention
  status: CANONICAL
  description: >-
    The circuit-level endpoint on which the other models converge: whatever the
    upstream insult, the delirious brain shows breakdown of the long-range
    connectivity that supports attention and predictive processing.
    Electrophysiological modelling localizes this to reduced feedback
    connectivity, with the reduction scaling with delirium severity and with
    systemic inflammation — which is what makes it a shared endpoint rather
    than a parallel story.
  evidence:
  - reference: PMID:37695013
    reference_title: Delirium is associated with loss of feedback cortical connectivity.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      DCM showed that delirium was associated with decreased left-sided superior
      temporal gyrus (l-STG) to auditory cortex feedback connectivity.
    explanation: >-
      Direct electrophysiological evidence for the connectivity failure this
      node asserts.
  - reference: PMID:37695013
    reference_title: Delirium is associated with loss of feedback cortical connectivity.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      Feedback connectivity also negatively correlated with delirium severity and
      systemic inflammation.
    explanation: >-
      Ties the circuit endpoint quantitatively to both the clinical severity and
      the inflammatory upstream, which is the link the model needs.
  - reference: PMID:29278283
    reference_title: 'Delirium pathophysiology: An updated hypothesis of the etiology of acute brain failure.'
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      The specific cognitive and behavioral manifestations of the specific delirium
      picture result from a combination of neurotransmitter function and
      availability, variability in integration and processing of sensory
      information, motor responses to both external and internal cues, and the
      degree of breakdown in neuronal network connectivity, hence the term acute
      brain failure.
    explanation: >-
      Places network breakdown as the common final term that converts the
      upstream disturbances into the clinical picture.
pathophysiology:
- name: Reduced Cognitive Reserve
  biological_scale: ORGANISM
  description: >-
    The predisposing half of the model. Advanced age, dementia or milder
    cognitive impairment, frailty, sensory impairment, and comorbidity burden
    lower the insult required to produce delirium. This node is a state of the
    patient before any precipitant arrives; it does not itself produce
    delirium, and it is the reason the same drug dose is inert in one patient
    and deliriogenic in another.
  evidence:
  - reference: PMID:23992774
    reference_title: Delirium in elderly people.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      it serves both as a marker of brain vulnerability with decreased reserve and
      as a potential mechanism for permanent cognitive damage
    explanation: >-
      States the reduced-reserve framing that this node represents, and — in the
      same breath — the injury arm curated downstream.
  - reference: PMID:39602991
    reference_title: 'Global incidence and prevalence of delirium and its risk factors in medically hospitalized older patients: A systematic review and meta-analysis.'
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      The risk factors of delirium included frailty (odds ratio = 2.05), physical
      restraints (5.01), prior falls (1.99), severe illness (1.32) (evaluated
      using the Acute Physiology and Chronic Health Evaluation II), and cognitive
      impairment (2.61).
    explanation: >-
      Quantifies the predisposing factors — frailty, prior falls, cognitive
      impairment — that constitute this node, alongside the iatrogenic
      precipitants curated under environmental.
  - reference: PMID:25239680
    reference_title: Worsening cognitive impairment and neurodegenerative pathology progressively increase risk for delirium.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      In the Vantaa cohort, 465 persons (88.4 ± 2.8 years) completed MMSE at
      baseline. For every MMSE point lost, risk of incident delirium increased by
      5% (p = 0.02).
    explanation: >-
      Converts this node from a categorical risk factor into a graded one: risk
      scales continuously with pre-existing cognitive impairment rather than
      switching on at a dementia diagnosis.
  downstream:
  - target: Microglial Activation and Neuroinflammation
    causal_link_type: INDIRECT_UNKNOWN_INTERMEDIATES
    hypothesis_groups:
    - neuroinflammatory_model
    description: >-
      A primed, aged, or degenerating brain mounts an exaggerated central
      response to the same peripheral signal, which is the best-supported
      cellular reading of "reduced reserve".
- name: Systemic Inflammatory Insult
  biological_scale: ORGANISM
  description: >-
    The precipitating half. Infection and sepsis, major surgery, trauma, and
    critical illness raise circulating pro-inflammatory mediators — IL-6, IL-8,
    TNF-alpha — and this peripheral signal is measurably associated with
    subsequent delirium. In the subacute presentation the insult is typically
    indolent (an occult infection, a slowly accumulating drug effect, a
    gradually worsening metabolic derangement) rather than abrupt, which is why
    the onset is measured in days.
  biological_processes:
  - preferred_term: cytokine-mediated signaling pathway
    term:
      id: GO:0019221
      label: cytokine-mediated signaling pathway
    modifier: INCREASED
  evidence:
  - reference: PMID:34218905
    reference_title: 'Preoperative inflammatory mediators and postoperative delirium: systematic review and meta-analysis.'
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      Higher preoperative interleukin-6 was associated with postoperative delirium
      with a standardised mean difference (95% confidence interval) of 0.33
      (0.11-0.56) and P=0.003.
    explanation: >-
      Meta-analytic evidence that the systemic inflammatory state indexed by this
      node predicts the syndrome.
  downstream:
  - target: Blood-Brain Barrier Permeability Increase
    causal_link_type: INDIRECT_KNOWN_INTERMEDIATES
    hypothesis_groups:
    - neuroinflammatory_model
    description: >-
      Circulating cytokines act on cerebral endothelium; the resulting
      permeability change is measurable as a CSF:plasma albumin ratio that
      tracks central IL-6 receptor signalling.
  - target: Microglial Activation and Neuroinflammation
    causal_link_type: INDIRECT_KNOWN_INTERMEDIATES
    hypothesis_groups:
    - neuroinflammatory_model
    description: >-
      Peripheral inflammatory signal reaches the CNS by humoral, endothelial,
      and vagal routes and converts primed microglia to an activated state.
- name: Blood-Brain Barrier Permeability Increase
  biological_scale: TISSUE
  description: >-
    The permissive step: an inflamed cerebral endothelium lets peripheral
    inflammatory signalling into the CNS compartment. It is curated as its own
    node because it is separately measurable in patients — the CSF:plasma
    albumin ratio and plasma S100B correlate with the CSF soluble IL-6 receptor
    that indexes central trans-signalling — rather than being inferred from the
    peripheral cytokine level alone.
  locations:
  - preferred_term: brain
    term:
      id: UBERON:0000955
      label: brain
  biological_processes:
  - preferred_term: regulation of vascular permeability
    term:
      id: GO:0043114
      label: regulation of vascular permeability
    modifier: INCREASED
  evidence:
  - reference: PMID:42011187
    reference_title: 'The association between cerebrospinal fluid interleukin-6, soluble interleukin-6 receptor, and interleukin-6 trans-signalling binary complex with postoperative delirium: an observational cohort study.'
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      CSF sIL-6R was also significantly correlated with peripheral inflammation
      (plasma IL-6 and IL-8) and blood-brain barrier permeability (CSF:plasma
      albumin ratio and plasma S100B).
    explanation: >-
      The measurement that makes this a curatable node rather than an assumed
      intermediate.
  downstream:
  - target: Microglial Activation and Neuroinflammation
    causal_link_type: DIRECT
    hypothesis_groups:
    - neuroinflammatory_model
    description: >-
      Entry of peripheral mediators into the CNS compartment activates
      parenchymal microglia.
- name: Microglial Activation and Neuroinflammation
  biological_scale: CELLULAR
  description: >-
    Central amplification of the peripheral signal. Microglia switch from a
    surveillant to an activated phenotype and release IL-1beta, TNF-alpha, and
    other mediators locally; astrocytic activation accompanies it, and is
    reflected clinically in the S100B rise curated under biochemical. The
    conditional nature of this node is what distinguishes delirium from
    ordinary sickness behaviour: in animals, an identical systemic cytokine
    challenge leaves working memory intact on a normal background and impairs
    it acutely on a degenerating one.
  cell_types:
  - preferred_term: microglial cell
    term:
      id: CL:0000129
      label: microglial cell
  - preferred_term: astrocyte
    term:
      id: CL:0000127
      label: astrocyte
  biological_processes:
  - preferred_term: microglial cell activation
    term:
      id: GO:0001774
      label: microglial cell activation
    modifier: INCREASED
  - preferred_term: neuroinflammatory response
    term:
      id: GO:0150076
      label: neuroinflammatory response
    modifier: INCREASED
  evidence:
  - reference: PMID:27633985
    reference_title: Systemic TNF-α produces acute cognitive dysfunction and exaggerated sickness behavior when superimposed upon progressive neurodegeneration.
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    snippet: >-
      TNF-α (50μg/kg) had no impact on performance of normal animals (normal brain
      homogenate; NBH) on working memory (T-maze) but produced acute impairments
      in ME7 animals similarly challenged.
    explanation: >-
      Demonstrates the vulnerability-conditional central response that this node
      asserts, with the systemic stimulus held constant between groups.
  downstream:
  - target: Cholinergic-Dopaminergic Neurotransmitter Imbalance
    causal_link_type: INDIRECT_UNKNOWN_INTERMEDIATES
    hypothesis_groups:
    - cholinergic_dopaminergic_imbalance_model
    description: >-
      Neuroinflammation is proposed to perturb cholinergic and dopaminergic
      transmission; the intermediates are not established in humans, and the
      edge is recorded at that strength rather than as a direct effect.
  - target: Cortical Feedback Connectivity Loss
    causal_link_type: INDIRECT_UNKNOWN_INTERMEDIATES
    hypothesis_groups:
    - neuroinflammatory_model
    - network_disconnection_model
    description: >-
      The inflammatory and circuit accounts meet here: feedback connectivity
      falls as systemic inflammation rises, but the pathway from cytokine to
      connectivity is not resolved.
  - target: Neuronal and Astrocytic Injury
    causal_link_type: INDIRECT_UNKNOWN_INTERMEDIATES
    hypothesis_groups:
    - neuroinflammatory_model
    description: >-
      Sustained activation is the proposed route from a reversible state to the
      injury that underlies incomplete recovery.
- name: Cholinergic-Dopaminergic Neurotransmitter Imbalance
  biological_scale: MOLECULAR
  description: >-
    Relative cholinergic deficiency with dopaminergic excess is the classical
    proximate account, and it remains the best explanation for drug-precipitated
    delirium: anticholinergic and dopaminergic agents are among the most
    reliable precipitants, and their withdrawal is part of effective management.
    It is curated as a real node with a failed therapeutic corollary — neither
    cholinesterase inhibition nor dopamine-receptor blockade shortened delirium
    in randomized trials — so the imbalance should not be read as a validated
    drug target.
  biological_processes:
  - preferred_term: synaptic transmission, cholinergic
    term:
      id: GO:0007271
      label: synaptic transmission, cholinergic
    modifier: DECREASED
  evidence:
  - reference: PMID:21056464
    reference_title: 'Effect of rivastigmine as an adjunct to usual care with haloperidol on duration of delirium and mortality in critically ill patients: a multicentre, double-blind, placebo-controlled randomised trial.'
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      Impaired cholinergic neurotransmission seems to have an important role in
      the development of delirium.
    explanation: >-
      Supports the existence of the cholinergic disturbance this node names,
      stated as the rationale of the trial that then failed to exploit it.
  downstream:
  - target: Cortical Feedback Connectivity Loss
    causal_link_type: INDIRECT_UNKNOWN_INTERMEDIATES
    hypothesis_groups:
    - cholinergic_dopaminergic_imbalance_model
    description: >-
      Cholinergic tone gates cortical signal-to-noise and attention; the link to
      the measured connectivity failure is plausible but not established in
      humans.
- name: Cortical Feedback Connectivity Loss
  biological_scale: ORGANISM
  description: >-
    The circuit-level state of the delirious brain and the proximate substrate
    of its cardinal sign. Dynamic causal modelling of auditory evoked responses
    shows reduced top-down (feedback) connectivity during delirium, with
    increased intrinsic inhibitory tone in the same region; the reduction scales
    with delirium severity and with systemic inflammation. Framed predictively,
    the delirious cortex fails to impose top-down expectations on sensory
    input — which is what inattention, disorganized thinking, and perceptual
    disturbance look like from the inside.
  locations:
  - preferred_term: cerebral cortex
    term:
      id: UBERON:0000956
      label: cerebral cortex
  evidence:
  - reference: PMID:37695013
    reference_title: Delirium is associated with loss of feedback cortical connectivity.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      Delirium is associated with decreased feedback cortical connectivity,
      possibly resulting from increased intrinsic inhibitory tone.
    explanation: >-
      The study's own summary of the connectivity state this node describes.
  downstream:
  - target: Protracted and Persistent Course
    causal_link_type: UNKNOWN
    hypothesis_groups:
    - network_disconnection_model
    description: >-
      Whether failure of the network state to re-establish itself is what makes
      a delirium subacute or persistent is unresolved; the edge records the
      candidate relationship without asserting it.
- name: Neuronal and Astrocytic Injury
  biological_scale: CELLULAR
  description: >-
    The arm that makes delirium more than a transient state. Astrocytic (S100B)
    and axonal (neurofilament light) injury markers rise in association with
    delirium, and at the clinical level a longer duration of delirium predicts
    worse global cognition and executive function at 3 and 12 months, with
    delirium independently associated with long-term cognitive decline across
    studies. The direction of causation is not fully settled — delirium is at
    minimum a marker of a vulnerable brain — but the dose-response with
    delirium duration is what motivates curating injury as a distinct node.
  cell_types:
  - preferred_term: astrocyte
    term:
      id: CL:0000127
      label: astrocyte
  evidence:
  - reference: PMID:24088092
    reference_title: Long-term cognitive impairment after critical illness.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      A longer duration of delirium was independently associated with worse global
      cognition at 3 and 12 months (P=0.001 and P=0.04, respectively) and worse
      executive function at 3 and 12 months (P=0.004 and P=0.007, respectively).
    explanation: >-
      The duration-response relationship that argues for delirium contributing to
      injury rather than only marking prior vulnerability.
  - reference: PMID:32658246
    reference_title: 'Association of Delirium With Long-term Cognitive Decline: A Meta-analysis.'
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      There was a significant association between delirium and long-term cognitive
      decline, as the estimated effect size (Hedges g) for 23 studies was 0.45
      (95% CI, 0.34-0.57; P < .001).
    explanation: >-
      Pooled magnitude of the long-term cognitive consequence attributed to this
      node.
  - reference: PMID:40889075
    reference_title: 'Serum biomarkers of delirium in critical illness: a systematic review of mechanistic and diagnostic evidence.'
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      Among CNS injury markers, S100β and neurofilament light chain (NfL)
      demonstrated the most consistent associations with delirium presence and
      severity, supporting a role for astrocytic and axonal injury in delirium
      pathogenesis.
    explanation: >-
      The systematic review that identifies astrocytic and axonal injury markers
      as the most consistent biomarker signal — the molecular evidence for this
      node, and the reason it names both cell compartments.
  - reference: PMID:38566855
    reference_title: 'Serum NFL and tau, but not serum UCHL-1 and GFAP or CSF SNAP-25, NPTX2, or sTREM2, correlate with delirium in a 3-year retrospective analysis.'
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      The analysis of 71 patients with delirium, Alzheimer's disease (AD), and
      non-AD controls revealed that serum NFL levels are higher in delirium cases
      compared to both AD and non-AD.
    explanation: >-
      Separates delirium-associated neuroaxonal damage from that of the
      underlying dementia, which is the confound this node's causal claim has to
      survive.
- name: Protracted and Persistent Course
  biological_scale: ORGANISM
  description: >-
    The node that gives this entry its name. Where a classical delirium
    declares itself in hours and resolves with its precipitant, the subacute
    presentation evolves over days and often does not resolve on the expected
    timescale: among older hospital patients who become delirious, nearly half
    are still delirious at discharge and a fifth remain so at six months. Because the presentation is
    gradual and the deficits — inattention, disorientation, impaired memory —
    are the ones dementia also produces, the subacute form is preferentially
    misread as an evolving dementia rather than as an acute brain state with a
    treatable cause.
  evidence:
  - reference: PMID:19017678
    reference_title: 'Persistent delirium in older hospital patients: a systematic review of frequency and prognosis.'
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      the combined proportions with PerD at discharge, 1, 3 and 6 months were 44.7%
      (95% CI 26.8%, 63.7%), 32.8% (95% CI 18.4%, 47.2%), 25.6% (95% CI 7.9%,
      43.4%) and 21% (95% CI 1.4%, 40.6%), respectively.
    explanation: >-
      Quantifies the protracted course this node stands for, at four follow-up
      points.
  - reference: PMID:19017678
    reference_title: 'Persistent delirium in older hospital patients: a systematic review of frequency and prognosis.'
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      The outcomes (mortality, nursing home placement, function, cognition) of
      patients with PerD were consistently worse than the outcomes of patients who
      had recovered from delirium.
    explanation: >-
      Establishes that the protracted course is prognostically distinct, not just
      a longer version of the same illness.
  downstream:
  - target: Neuronal and Astrocytic Injury
    causal_link_type: INDIRECT_UNKNOWN_INTERMEDIATES
    hypothesis_groups:
    - neuroinflammatory_model
    description: >-
      Duration of delirium is the exposure variable that predicts long-term
      cognitive outcome, so a protracted course is curated as feeding the injury
      node rather than merely coexisting with it.
phenotypes:
- category: Neurologic
  name: Inattention
  description: >-
    Reduced ability to direct, focus, sustain, and shift attention — the
    cardinal feature, and the one operationalized by every bedside instrument
    (digit span, months of the year backwards). It is also among the symptoms
    most likely to persist.
  phenotype_term:
    preferred_term: Inattention
    term:
      id: HP:0000736
      label: Short attention span
  notes: >-
    HPO has no term for the fluctuating, acquired inattention of delirium;
    `Short attention span` is the closest available concept and is used with a
    `preferred_term` of Inattention rather than manufacturing a narrower match.
  evidence:
  - reference: PMID:23992774
    reference_title: Delirium in elderly people.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      Key diagnostic features include an acute onset and fluctuating course of
      symptoms, inattention, impaired level of consciousness, and disturbance of
      cognition (e.g., disorientation, memory impairment, alteration in language).
    explanation: >-
      Lists inattention as a key diagnostic feature of the syndrome.
- category: Neurologic
  name: Delirium
  description: >-
    The syndrome itself, recorded as a phenotype so that the entry's disease
    concept is also expressible as an HPO annotation for downstream use.
  phenotype_term:
    preferred_term: Delirium
    term:
      id: HP:0031258
      label: Delirium
    temporality: SUBACUTE
  evidence:
  - reference: PMID:23992774
    reference_title: Delirium in elderly people.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      Delirium is an acute disorder of attention and cognition in elderly people
      (ie, those aged 65 years or older) that is common, serious, costly,
      under-recognised, and often fatal.
    explanation: >-
      Definitional statement of the syndrome this phenotype records.
- category: Neurologic
  name: Confusion and disorientation
  description: >-
    Disorientation to time and place with a globally disorganized stream of
    thought, characteristically worse in the evening. In the subacute
    presentation it accumulates over days and is the feature most often
    attributed to dementia instead.
  phenotype_term:
    preferred_term: Confusion
    term:
      id: HP:0001289
      label: Confusion
  evidence:
  - reference: PMID:23992774
    reference_title: Delirium in elderly people.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      Key diagnostic features include an acute onset and fluctuating course of
      symptoms, inattention, impaired level of consciousness, and disturbance of
      cognition (e.g., disorientation, memory impairment, alteration in language).
    explanation: >-
      Names disorientation among the cognitive disturbances that define the
      syndrome.
- category: Neurologic
  name: Memory impairment
  description: >-
    Impaired encoding and recall during the episode; with inattention and
    disorientation it is one of the features most likely to persist beyond the
    acute illness.
  phenotype_term:
    preferred_term: Memory impairment
    term:
      id: HP:0002354
      label: Memory impairment
  evidence:
  - reference: PMID:23992774
    reference_title: Delirium in elderly people.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      Key diagnostic features include an acute onset and fluctuating course of
      symptoms, inattention, impaired level of consciousness, and disturbance of
      cognition (e.g., disorientation, memory impairment, alteration in language).
    explanation: >-
      Lists memory impairment among the defining cognitive disturbances.
- category: Behavioral
  name: Perceptual disturbance and hallucinations
  description: >-
    Illusions and hallucinations, typically visual, more prominent in the
    hyperactive presentation and often accompanied by delusional
    misinterpretation of the care environment.
  phenotype_term:
    preferred_term: Hallucinations
    term:
      id: HP:0000738
      label: Hallucinations
  evidence:
  - reference: PMID:23992774
    reference_title: Delirium in elderly people.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      Supportive features include disturbance in sleep-wake cycle, perceptual
      disturbances (hallucinations or illusions), delusions, psychomotor
      disturbance (hypo- or hyper-activity), inappropriate behavior, and emotional
      lability.
    explanation: >-
      Names perceptual disturbance and hallucinations as recognized supportive
      features of delirium.
- category: Behavioral
  name: Psychomotor agitation
  description: >-
    Restlessness and agitation with pulling at lines and catheters — the
    hyperactive presentation. It is the most visible form and therefore the
    most reliably detected, though it is not the most common in critical
    illness.
  phenotype_term:
    preferred_term: Agitation
    term:
      id: HP:0000713
      label: Agitation
  evidence:
  - reference: PMID:30346242
    reference_title: Haloperidol and Ziprasidone for Treatment of Delirium in Critical Illness.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      Delirium developed in 566 patients (48%), of whom 89% had hypoactive delirium
      and 11% had hyperactive delirium.
    explanation: >-
      Ascertains both motor presentations in a defined cohort and gives their
      relative frequency in critical illness.
- category: Behavioral
  name: Hypoactive presentation with apathy and lethargy
  description: >-
    Withdrawal, reduced spontaneous movement, and lethargy. In critically ill
    patients this is by far the dominant form — 89% of delirious patients in
    the MIND-USA cohort — and it is the form most often missed, because a quiet
    patient does not prompt assessment.
  phenotype_term:
    preferred_term: Apathy
    term:
      id: HP:0000741
      label: Apathy
  frequency: VERY_FREQUENT
  evidence:
  - reference: PMID:30346242
    reference_title: Haloperidol and Ziprasidone for Treatment of Delirium in Critical Illness.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      Delirium developed in 566 patients (48%), of whom 89% had hypoactive delirium
      and 11% had hyperactive delirium.
    explanation: >-
      Supports both the predominance of the hypoactive presentation and the
      VERY_FREQUENT band assigned here, within an ICU population.
- category: Neurologic
  name: Sleep-wake cycle disturbance
  description: >-
    Fragmented nocturnal sleep with daytime somnolence, often to the point of
    day-night reversal; the circadian component is one of the five core domains
    in contemporary syntheses of delirium's phenomenology.
  phenotype_term:
    preferred_term: Sleep-wake cycle disturbance
    term:
      id: HP:0006979
      label: Sleep-wake cycle disturbance
  evidence:
  - reference: PMID:29278283
    reference_title: 'Delirium pathophysiology: An updated hypothesis of the etiology of acute brain failure.'
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      Delirium is the most common neuropsychiatric syndrome encountered by
      clinicians dealing with older adults and the medically ill and is best
      characterized by 5 core domains: cognitive deficits, attentional deficits,
      circadian rhythm dysregulation, emotional dysregulation, and alteration in
      psychomotor functioning.
    explanation: >-
      Establishes circadian rhythm dysregulation as one of the five core domains
      of the syndrome.
biochemical:
- name: Serum S100B
  notes: >-
    Calcium-binding protein released by astrocytes; elevated concentrations
    accompany delirium and, in the perioperative setting, correlate with the
    central IL-6 receptor signalling and blood-brain barrier permeability
    curated in the pathograph. It is a research marker, not a diagnostic test.
  biomarker_term:
    preferred_term: S100B
    term:
      id: NCIT:C98129
      label: Protein S100-B
  evidence:
  - reference: PMID:42011187
    reference_title: 'The association between cerebrospinal fluid interleukin-6, soluble interleukin-6 receptor, and interleukin-6 trans-signalling binary complex with postoperative delirium: an observational cohort study.'
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      CSF sIL-6R was also significantly correlated with peripheral inflammation
      (plasma IL-6 and IL-8) and blood-brain barrier permeability (CSF:plasma
      albumin ratio and plasma S100B).
    explanation: >-
      Uses plasma S100B as one of the two indices of barrier permeability,
      which is the role this marker plays in the entry.
  - reference: PMID:40889075
    reference_title: 'Serum biomarkers of delirium in critical illness: a systematic review of mechanistic and diagnostic evidence.'
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      Among CNS injury markers, S100β and neurofilament light chain (NfL)
      demonstrated the most consistent associations with delirium presence and
      severity, supporting a role for astrocytic and axonal injury in delirium
      pathogenesis.
    explanation: >-
      Establishes S100B as one of the two most consistent CNS-injury markers
      across the critical-illness literature.
  - reference: PMID:40889075
    reference_title: 'Serum biomarkers of delirium in critical illness: a systematic review of mechanistic and diagnostic evidence.'
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      Despite early promise, serum biomarkers for ICU delirium remain
      investigational and require further validation before clinical application.
    explanation: >-
      Bounds the claim: this is a research marker, not a diagnostic test, which
      is how the entry curates it.
- name: Preoperative interleukin-6
  notes: >-
    Circulating IL-6 measured before surgery is higher in patients who go on to
    develop postoperative delirium, making it the best-supported peripheral
    inflammatory predictor. The effect is modest and mediator- and
    surgery-specific, so it is curated as a risk marker rather than a test.
  biomarker_term:
    preferred_term: Interleukin-6
    term:
      id: NCIT:C20451
      label: Interleukin-6
  evidence:
  - reference: PMID:34218905
    reference_title: 'Preoperative inflammatory mediators and postoperative delirium: systematic review and meta-analysis.'
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      Higher preoperative interleukin-6 was associated with postoperative delirium
      with a standardised mean difference (95% confidence interval) of 0.33
      (0.11-0.56) and P=0.003.
    explanation: >-
      The pooled estimate that supports IL-6 as a pre-insult risk marker.
  - reference: PMID:34218905
    reference_title: 'Preoperative inflammatory mediators and postoperative delirium: systematic review and meta-analysis.'
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      The association of preoperative blood levels of inflammatory mediators with
      postoperative delirium may be influenced by the type of surgery and the
      specific mediator.
    explanation: >-
      Qualifies the marker: the association is conditional on context, which is
      why it is not curated as diagnostic.
genetic:
- name: APOE
  notes: >-
    APOE is the most consistently reported common-variant susceptibility locus
    for delirium, acting as a modifier of vulnerability rather than a cause:
    no APOE genotype produces delirium without a precipitant, and no delirium
    requires one. It is curated as SUSCEPTIBILITY for that reason. The same
    meta-analysis reports a multi-trait analysis with Alzheimer disease
    identifying five delirium risk loci and plasma proteins predicting incident
    delirium up to sixteen years ahead — the proteomic counterpart of the
    reduced-reserve node curated in the pathograph. Earlier candidate
    associations at the muscarinic receptor genes CHRM2 and CHRM4 are
    deliberately not curated: they are not among the loci reported by that
    meta-analysis.
  gene_term:
    preferred_term: APOE
    term:
      id: hgnc:613
      label: APOE
  relationship_type: SUSCEPTIBILITY
  evidence:
  - reference: PMID:41286463
    reference_title: Dissecting the genetic and proteomic risk factors for delirium.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      Here we conducted a genetic meta-analysis on delirium using multi-ancestry
      data from the UK Biobank, FinnGen, All of Us Research Program and Michigan
      Genomics Initiative cohorts (n = 1,059,130; 11,931 cases), yielding the
      Apolipoprotein E (APOE) gene as a strong delirium risk factor independently
      of dementia.
    explanation: >-
      The multi-ancestry meta-analysis establishing APOE as a delirium risk locus,
      and — crucially for the SUSCEPTIBILITY classification — independently of
      dementia rather than through it.
  - reference: PMID:41286463
    reference_title: Dissecting the genetic and proteomic risk factors for delirium.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      Combining proteins, APOE-ε4 status and demographics significantly improved
      incident delirium prediction compared to demographics alone.
    explanation: >-
      Shows the association is large enough to add predictive information, which
      is what distinguishes a susceptibility locus from an incidental signal.
environmental:
- name: Physical restraint use
  description: >-
    Physical restraints carry the largest odds ratio of any risk factor in the
    contemporary meta-analysis of hospitalized older patients, and are
    iatrogenic and modifiable. The direction of causation is not fully
    separable from confounding by indication — agitated patients get
    restrained — but restraint use is also a component that multicomponent
    prevention programs deliberately remove.
  effect: Increases risk of delirium in hospitalized older patients
  influences_mechanisms:
  - target: Reduced Cognitive Reserve
    environmental_effect: EXACERBATES
    causal_link_type: INDIRECT_UNKNOWN_INTERMEDIATES
    description: >-
      Restraint enforces immobility and sensory-environmental deprivation,
      compounding the predisposing state rather than acting as an inflammatory
      precipitant.
    evidence:
    - reference: PMID:39602991
      reference_title: 'Global incidence and prevalence of delirium and its risk factors in medically hospitalized older patients: A systematic review and meta-analysis.'
      supports: SUPPORT
      evidence_source: HUMAN_CLINICAL
      snippet: >-
        The risk factors of delirium included frailty (odds ratio = 2.05), physical
        restraints (5.01), prior falls (1.99), severe illness (1.32) (evaluated
        using the Acute Physiology and Chronic Health Evaluation II), and cognitive
        impairment (2.61).
      explanation: >-
        Gives the restraint-associated odds ratio in the pooled analysis this
        link rests on.
  evidence:
  - reference: PMID:39602991
    reference_title: 'Global incidence and prevalence of delirium and its risk factors in medically hospitalized older patients: A systematic review and meta-analysis.'
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      The risk factors of delirium included frailty (odds ratio = 2.05), physical
      restraints (5.01), prior falls (1.99), severe illness (1.32) (evaluated
      using the Acute Physiology and Chronic Health Evaluation II), and cognitive
      impairment (2.61).
    explanation: >-
      Establishes restraint use as an independently reported risk factor with the
      largest pooled odds ratio in this analysis.
- name: Deliriogenic medication exposure
  description: >-
    Anticholinergic drugs, benzodiazepines, and opioids are among the most
    reliable precipitants, and the classic demonstration of the vulnerability
    model is that a single dose of a sedative can precipitate delirium in a
    patient with dementia and multimorbidity. Reduction of sedative and
    analgesic burden is a recommended management step.
  effect: Precipitates delirium, particularly in patients with reduced cognitive reserve
  influences_mechanisms:
  - target: Cholinergic-Dopaminergic Neurotransmitter Imbalance
    environmental_effect: TRIGGERS
    causal_link_type: DIRECT
    description: >-
      Anticholinergic and dopaminergic agents act directly on the transmitter
      systems this node describes, which is the strongest evidence that the
      imbalance is mechanistically real even though correcting it downstream
      does not treat the syndrome.
    evidence:
    - reference: PMID:23992774
      reference_title: Delirium in elderly people.
      supports: SUPPORT
      evidence_source: HUMAN_CLINICAL
      snippet: >-
        The list of potential neurotransmitters involved in delirium is long,60 but
        a relative cholinergic deficiency and/or dopamine excess are the most
        commonly inferred,61, 62 correlating with the adverse effects of
        anticholinergic or dopaminergic drugs.
      explanation: >-
        States the drug-effect reasoning that links this exposure to the
        transmitter node.
  evidence:
  - reference: PMID:23992774
    reference_title: Delirium in elderly people.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      Drug reduction for sedation and analgesia and non-pharmacological approaches
      are recommended.
    explanation: >-
      The management recommendation that follows from treating these agents as
      modifiable precipitants.
prevalence:
- population: Medically hospitalized older patients
  measure_type: POINT_PREVALENCE
  prevalence_class: ABOVE_1_IN_1000
  rate_per_100000: 23600.0
  rate_low: 19000.0
  rate_high: 29000.0
  notes: >-
    Pooled prevalence 23.6% (95% CI 19%-29%) across 35 studies and 12,097
    participants. Reported alongside a pooled incidence of 13.5%, which is a
    different measure and is not combined with it here.
  evidence:
  - reference: PMID:39602991
    reference_title: 'Global incidence and prevalence of delirium and its risk factors in medically hospitalized older patients: A systematic review and meta-analysis.'
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      A total of 35 studies encompassing 12,097 participants met the inclusion
      criteria, yielding a 23.6 % pooled prevalence of delirium (95 % confidence
      interval = 19 % to 29 %) and a 13.5 % pooled incidence of delirium (n = 32,
      95 % confidence interval = 11 % to 17 %) among medically hospitalized older
      patients.
    explanation: >-
      Source of both the pooled prevalence recorded here and the incidence noted
      for contrast.
progression:
- phase: Persistence at hospital discharge
  notes: >-
    44.7% of older hospital patients with delirium were still delirious at
    discharge (95% CI 26.8%-63.7%).
  evidence:
  - reference: PMID:19017678
    reference_title: 'Persistent delirium in older hospital patients: a systematic review of frequency and prognosis.'
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      the combined proportions with PerD at discharge, 1, 3 and 6 months were 44.7%
      (95% CI 26.8%, 63.7%), 32.8% (95% CI 18.4%, 47.2%), 25.6% (95% CI 7.9%,
      43.4%) and 21% (95% CI 1.4%, 40.6%), respectively.
    explanation: >-
      Reports the discharge-time persistence proportion recorded in this phase.
- phase: Persistence at six months
  notes: >-
    21% remained delirious at six months (95% CI 1.4%-40.6%); the wide interval
    reflects heterogeneity across the 18 pooled reports.
  evidence:
  - reference: PMID:19017678
    reference_title: 'Persistent delirium in older hospital patients: a systematic review of frequency and prognosis.'
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      the combined proportions with PerD at discharge, 1, 3 and 6 months were 44.7%
      (95% CI 26.8%, 63.7%), 32.8% (95% CI 18.4%, 47.2%), 25.6% (95% CI 7.9%,
      43.4%) and 21% (95% CI 1.4%, 40.6%), respectively.
    explanation: >-
      Reports the six-month persistence proportion recorded in this phase.
clinical_burden:
  burden_level: HIGH
  rationale: >-
    Delirium is common in hospitalized older adults, is independently
    associated with mortality, and — in the protracted form this entry
    centres on — persists in a substantial minority for months and carries
    consistently worse mortality, institutionalization, function, and
    cognition than a delirium that resolves. It is additionally associated
    with long-term cognitive decline of moderate effect size.
  evidence:
  - reference: PMID:23992774
    reference_title: Delirium in elderly people.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      Delirium is an acute disorder of attention and cognition in elderly people
      (ie, those aged 65 years or older) that is common, serious, costly,
      under-recognised, and often fatal.
    explanation: >-
      Summary statement of the burden this assessment records.
  - reference: PMID:32658246
    reference_title: 'Association of Delirium With Long-term Cognitive Decline: A Meta-analysis.'
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      In this meta-analysis, delirium was significantly associated with long-term
      cognitive decline in both surgical and nonsurgical patients.
    explanation: >-
      Supports the long-term component of the burden assessment.
diagnosis:
- name: Confusion Assessment Method (CAM)
  description: >-
    The standard bedside instrument, structured around acute onset with a
    fluctuating course plus inattention, and either disorganized thinking or an
    altered level of consciousness. Its ICU adaptation, the CAM-ICU, is what
    ascertains delirium in ventilated patients who cannot be interviewed, and
    is the instrument used in the major critical-care trials curated here.
  evidence:
  - reference: PMID:23992774
    reference_title: Delirium in elderly people.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      validated in high quality studies including over 1000 patients with
      sensitivity of 94%, specificity of 89%, and high inter-rater reliability.
    explanation: >-
      Reports the CAM's validated operating characteristics in the sentence that
      names it the most widely used instrument for identifying delirium.
  - reference: PMID:30346242
    reference_title: Haloperidol and Ziprasidone for Treatment of Delirium in Critical Illness.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      delirium was detected with the use of the Confusion Assessment Method for the
      ICU (CAM-ICU),23,24 a validated tool that identifies delirium on the basis of
      an acute change or fluctuating course of mental status plus inattention and
      either altered level of consciousness or disorganized thinking
    explanation: >-
      Describes the ICU adaptation and its operational criteria as applied in a
      definitive trial.
- name: Search for the precipitating cause
  description: >-
    There is no diagnostic test for delirium itself; the diagnostic work is
    identifying what precipitated it. Because delirium can be the presenting
    feature of a medical emergency, screening for acute physiological
    disturbance — hypoxaemia, hypoglycaemia, hypercapnia — precedes any
    elaborate work-up, and testing is targeted by history and examination
    rather than run as a battery.
  evidence:
  - reference: PMID:23992774
    reference_title: Delirium in elderly people.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      First, because delirium can be the harbinger of a medical emergency, every
      patient presenting with delirium should be screened for acute physiologic
      disturbance such as hypoxemia, low blood glucose, and high arterial carbon
      dioxide.
    explanation: >-
      States the priority and the specific targets of the initial diagnostic
      screen.
treatments:
- name: Multicomponent non-pharmacological delirium prevention
  description: >-
    The Hospital Elder Life Program and its descendants apply standardized
    protocols to six modifiable risk factors — cognitive impairment, sleep
    deprivation, immobility, visual impairment, hearing impairment, and
    dehydration. In the original controlled study delirium developed in 9.9%
    of the intervention group versus 15.0% of usual care, and pooled across
    subsequent studies the odds of incident delirium fall by roughly half. It
    is the only intervention with a consistent effect on incidence, and it is
    mechanistically a treatment of the predisposing node rather than of any
    molecular target — which is exactly what the multifactorial model predicts
    would work.
  therapeutic_modality: BEHAVIORAL
  treatment_term:
    preferred_term: supportive care
    term:
      id: NCIT:C15747
      label: Supportive Care
  target_mechanisms:
  - target: Reduced Cognitive Reserve
    treatment_effect: MODULATES
    description: >-
      Each protocol component removes one of the predisposing or perpetuating
      factors that constitute this node — immobility, sensory deprivation,
      dehydration, sleep loss — rather than acting on the inflammatory or
      neurotransmitter arms.
    evidence:
    - reference: PMID:10053175
      reference_title: A multicomponent intervention to prevent delirium in hospitalized older patients.
      supports: SUPPORT
      evidence_source: HUMAN_CLINICAL
      snippet: >-
        The intervention consisted of standardized protocols for the management of
        six risk factors for delirium: cognitive impairment, sleep deprivation,
        immobility, visual impairment, hearing impairment, and dehydration.
      explanation: >-
        Enumerates the targets of the intervention, which are the constituents of
        the node it is linked to.
  evidence:
  - reference: PMID:10053175
    reference_title: A multicomponent intervention to prevent delirium in hospitalized older patients.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      Delirium developed in 9.9 percent of the intervention group as compared with
      15.0 percent of the usual-care group, (matched odds ratio, 0.60; 95 percent
      confidence interval, 0.39 to 0.92).
    explanation: >-
      The controlled effect estimate on incident delirium from the defining study.
  - reference: PMID:25643002
    reference_title: 'Effectiveness of multicomponent nonpharmacological delirium interventions: a meta-analysis.'
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      Overall, 11 studies demonstrated significant reductions in delirium incidence
      (odds ratio [OR], 0.47; 95% CI, 0.38-0.58).
    explanation: >-
      Pooled effect across studies, establishing that the original result
      generalizes.
  - reference: PMID:10053175
    reference_title: A multicomponent intervention to prevent delirium in hospitalized older patients.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      The intervention had no significant effect on the severity of delirium or on
      recurrence rates; this finding suggests that primary prevention of delirium
      is probably the most effective treatment strategy.
    explanation: >-
      Bounds the claim: the effect is on incidence, not on the severity or
      recurrence of an episode that has already begun.
- name: Treatment of the precipitating illness and withdrawal of deliriogenic drugs
  description: >-
    The mainstay of management of established delirium. No agent has been shown
    to shorten delirium in an unselected critically ill population — the
    antipsychotic and cholinesterase-inhibitor trials curated here are both
    negative — so treatment consists of identifying and correcting the
    precipitant, reducing sedative and analgesic burden, and applying the same
    non-pharmacological measures used in prevention. Dexmedetomidine is the
    exception that defines the rule: it benefited the specific population of
    non-intubated patients with hyperactive delirium, and is curated separately
    below.
  therapeutic_modality: BEHAVIORAL
  treatment_term:
    preferred_term: supportive care
    term:
      id: NCIT:C15747
      label: Supportive Care
  target_mechanisms:
  - target: Systemic Inflammatory Insult
    treatment_effect: INHIBITS
    description: >-
      Resolving the infection, metabolic derangement, or surgical stress removes
      the precipitating insult at its source, which is the only intervention
      that acts on this node.
    evidence:
    - reference: PMID:23992774
      reference_title: Delirium in elderly people.
      supports: SUPPORT
      evidence_source: HUMAN_CLINICAL
      snippet: >-
        The initial management focuses on three simultaneous priorities: (1)
        maintaining patient safety; (2) searching for the causes; and (3) managing
        delirium symptoms.
      explanation: >-
        Places the search for and treatment of the cause at the centre of
        management, which is what this link asserts.
  evidence:
  - reference: PMID:23992774
    reference_title: Delirium in elderly people.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      Drug reduction for sedation and analgesia and non-pharmacological approaches
      are recommended.
    explanation: >-
      The recommended management strategy that this treatment entry records.
- name: Antipsychotics for severe agitation
  description: >-
    Haloperidol and atypical antipsychotics remain in use for the behavioural
    control of severe agitation that threatens patient or staff safety. They
    are curated here with the evidence that constrains them: in a
    placebo-controlled trial of 566 delirious critically ill patients, neither
    haloperidol nor ziprasidone altered the number of days alive without
    delirium or coma. They are symptomatic agents, not disease-modifying ones,
    and the entry deliberately does not link them to a mechanism node.
  therapeutic_modality: SMALL_MOLECULE
  treatment_term:
    preferred_term: Pharmacotherapy
    term:
      id: NCIT:C15986
      label: Pharmacotherapy
    therapeutic_agent:
    - preferred_term: haloperidol
      term:
        id: CHEBI:5613
        label: haloperidol
  evidence:
  - reference: PMID:30346242
    reference_title: Haloperidol and Ziprasidone for Treatment of Delirium in Critical Illness.
    supports: REFUTE
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      The use of haloperidol or ziprasidone, as compared with placebo, in patients
      with acute respiratory failure or shock and hypoactive or hyperactive
      delirium in the ICU did not significantly alter the duration of delirium.
    explanation: >-
      Refutes any duration-modifying claim for antipsychotics in delirium, which
      is why this treatment is curated as symptomatic only.
  - reference: PMID:23992774
    reference_title: Delirium in elderly people.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      No convincing evidence shows that pharmacological prevention or treatment is
      effective.
    explanation: >-
      The general statement of the evidence position this treatment entry is
      curated against.
- name: Dexmedetomidine for hyperactive delirium in non-intubated ICU patients
  description: >-
    An alpha-2 adrenergic agonist, and the one pharmacological agent in this
    entry with a positive randomized result. In a multicentre, double-blind,
    placebo-controlled trial across nine ICUs, non-intubated critically ill
    adults with hyperactive delirium randomized to dexmedetomidine did better
    than placebo on a joint model of agitation duration, delirium duration, and
    the need for intubation and deep sedation; the trial was stopped for
    efficacy at the preplanned interim analysis. The effect is real but should
    not be overread: the median reduction in agitation was about an hour, the
    other key secondary outcomes were similar between groups, and the
    population is a specific one — non-intubated patients with the hyperactive
    presentation, which is the minority form in critical illness.
  therapeutic_modality: SMALL_MOLECULE
  treatment_term:
    preferred_term: Pharmacotherapy
    term:
      id: NCIT:C15986
      label: Pharmacotherapy
    therapeutic_agent:
    - preferred_term: dexmedetomidine
      term:
        id: CHEBI:4466
        label: dexmedetomidine
  evidence:
  - reference: PMID:41160116
    reference_title: 'Dexmedetomidine for treatment of hyperactive delirium in non-intubated ICU patients: the 4D randomized clinical trial.'
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      In non-intubated adult ICU patients with hyperactive delirium, the use of
      dexmedetomidine was associated with a greater clinical benefit than placebo
      for the joint modelling of multiple endpoints of agitation control, delirium
      resolution, and intubation with mechanical ventilation.
    explanation: >-
      The randomized result establishing benefit in the specified population and
      on the specified composite endpoint.
  - reference: PMID:41160116
    reference_title: 'Dexmedetomidine for treatment of hyperactive delirium in non-intubated ICU patients: the 4D randomized clinical trial.'
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      Agitation duration was shorter in the dexmedetomidine group (1.0 (1.0-2.0)
      vs 2.0 (1.0-7.0) hours, absolute difference 95% CI - 1.0 (- 2.0 to - 0.1);
      ES, - 0.60; 95% CI, -0.92 to -0.27, p = 0.001). Other key secondary outcomes
      were similar between groups.
    explanation: >-
      Bounds the claim with the effect size and the neutral secondary outcomes,
      which is why this entry does not present dexmedetomidine as
      disease-modifying.
clinical_trials:
- name: NCT03317067
  phase: PHASE_III
  status: COMPLETED
  description: >-
    The 4D trial: a multicentre, double-blind, placebo-controlled,
    investigator-initiated randomized trial of continuous intravenous
    dexmedetomidine versus placebo for hyperactive delirium in non-intubated
    critically ill adults, across nine ICUs, stopped early for efficacy at the
    preplanned interim analysis.
  target_phenotypes:
  - preferred_term: Agitation
    term:
      id: HP:0000713
      label: Agitation
  - preferred_term: Delirium
    term:
      id: HP:0031258
      label: Delirium
  evidence:
  - reference: clinicaltrials:NCT03317067
    supports: SUPPORT
    evidence_source: OTHER
    snippet: >-
      The primary purpose of the study is to evaluate whether dexmedetomidine is
      effective in treating agitated delirium of non-intubated ICI patients
    explanation: >-
      The registration record establishing the trial's identity and objective;
      the results are cited from the publication rather than from here.
  - reference: PMID:41160116
    reference_title: 'Dexmedetomidine for treatment of hyperactive delirium in non-intubated ICU patients: the 4D randomized clinical trial.'
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      The study was stopped for efficacy at the time of the preplanned interim
      analysis.
    explanation: >-
      Records the trial's early termination for efficacy, which is part of how
      its result should be interpreted.
animal_models:
- name: Systemic TNF-alpha challenge on a neurodegenerative background (ME7 prion mouse)
  species: Mouse
  genotype: ME7 prion-inoculated C57BL/6, challenged with intraperitoneal TNF-alpha
  publication: PMID:27633985
  description: >-
    The model that operationalizes vulnerability rather than the insult. An
    identical systemic cytokine challenge is given to normal mice and to mice
    with established prion neurodegeneration; only the latter show acute
    working-memory impairment, with higher hippocampal and hypothalamic
    cytokine transcription despite equivalent plasma TNF-alpha and CCL2. This
    is the vulnerability × precipitant model reduced to a single controlled
    experiment.
  evidence:
  - reference: PMID:27633985
    reference_title: Systemic TNF-α produces acute cognitive dysfunction and exaggerated sickness behavior when superimposed upon progressive neurodegeneration.
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    snippet: >-
      Systemic inflammation caused by infection, trauma or co-morbidity can alter
      the brain's inflammatory status, produce acute cognitive impairments, such as
      delirium, and drive new pathology and accelerated decline.
    explanation: >-
      States the delirium-relevant rationale that makes this model informative for
      the entry rather than a general neurodegeneration experiment.
  modeled_mechanisms:
  - target: Microglial Activation and Neuroinflammation
    relationship: PARTIALLY_RECAPITULATES
    fidelity: MODERATE
    description: >-
      Reproduces the conditional central inflammatory response — the same
      peripheral stimulus, a larger central response and an acute cognitive
      deficit only on the degenerating background.
    limitations: >-
      The cognitive readout is T-maze working memory, not the attentional
      disturbance that defines human delirium, and the background pathology is
      prion disease rather than the Alzheimer-type pathology that predisposes
      most human patients. The model also shows sickness behaviour that overlaps
      with, but is not, delirium.
    readouts:
    - name: T-maze working memory performance after systemic TNF-alpha
      target: Microglial Activation and Neuroinflammation
      direction: DECREASED
      interpretation: >-
        Acute cognitive impairment appearing only on the degenerating background
        is the model's operational stand-in for delirium.
      evidence:
      - reference: PMID:27633985
        reference_title: Systemic TNF-α produces acute cognitive dysfunction and exaggerated sickness behavior when superimposed upon progressive neurodegeneration.
        supports: SUPPORT
        evidence_source: MODEL_ORGANISM
        snippet: >-
          TNF-α (50μg/kg) had no impact on performance of normal animals (normal
          brain homogenate; NBH) on working memory (T-maze) but produced acute
          impairments in ME7 animals similarly challenged.
        explanation: >-
          Reports the measurement and its direction, with the crucial negative
          control on the normal background.
    evidence:
    - reference: PMID:27633985
      reference_title: Systemic TNF-α produces acute cognitive dysfunction and exaggerated sickness behavior when superimposed upon progressive neurodegeneration.
      supports: SUPPORT
      evidence_source: MODEL_ORGANISM
      snippet: >-
        The data indicate that acutely elevated TNF-α has robust acute effects on
        brain function, selectively in the degenerating brain, but more sustained
        levels may be required to significantly impact on underlying
        neurodegeneration.
      explanation: >-
        The authors' own summary of the selectivity that makes this model
        informative for the vulnerability-conditional node.
- name: LPS-induced neuroinflammation and cognitive impairment in mice
  species: Mouse
  genotype: Wild-type C57BL/6J challenged with lipopolysaccharide
  publication: PMID:30962497
  description: >-
    The dominant experimental proxy for infection-triggered delirium.
    Lipopolysaccharide produces sickness behaviour with cognitive impairment
    alongside hippocampal microglial activation, neuronal loss, and a shift in
    the cytokine milieu — TNF-alpha, IL-1beta, PGE2 and nitric oxide up, IL-4
    and IL-10 down — with NF-kB pathway activation. The model earns its place
    in this entry because it is not merely descriptive: a TLR4-specific
    inhibitory peptide prevents both the neuroinflammation and the cognitive
    impairment, which makes the inflammatory step causal rather than
    correlated.
  evidence:
  - reference: PMID:30962497
    reference_title: Neuroinflammation induced by lipopolysaccharide causes cognitive impairment in mice.
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    snippet: >-
      These data suggest that LPS induced cognitive impairment and
      neuroinflammation via microglia activation by activating the NF-kB signaling
      pathway
    explanation: >-
      The authors' mechanistic summary, which is the claim this model contributes
      to the entry's microglial node.
  modeled_mechanisms:
  - target: Microglial Activation and Neuroinflammation
    relationship: RECAPITULATES
    fidelity: MODERATE
    description: >-
      Reproduces the systemic-insult-to-microglial-activation step directly, with
      the hippocampal cellular and cytokine readouts that cannot be obtained in
      patients, and with a TLR4 blockade arm that establishes the direction of
      causation.
    limitations: >-
      The behavioural readouts are Morris water maze and passive avoidance —
      memory tasks, not measures of attention or awareness, which are what define
      human delirium. The LPS challenge is also an acute bolus in a healthy young
      animal, so the model speaks to the precipitant without the reduced-reserve
      substrate that the human syndrome requires; the ME7 models curated
      elsewhere in this entry supply that half.
    readouts:
    - name: Hippocampal microglial activation (IBA-1) and neuronal loss (MAP-2)
      target: Microglial Activation and Neuroinflammation
      direction: INCREASED
      interpretation: >-
        Microglial activation with concurrent neuronal loss in the hippocampus is
        the cellular signature this node asserts.
      evidence:
      - reference: PMID:30962497
        reference_title: Neuroinflammation induced by lipopolysaccharide causes cognitive impairment in mice.
        supports: SUPPORT
        evidence_source: MODEL_ORGANISM
        snippet: >-
          We found that LPS treatment leads to sickness behavior and cognitive
          impairment in mice as shown in the Morris water maze and passive
          avoidance test, and these effects were accompanied by microglia
          activation (labeled by ionized calcium binding adaptor molecule-1,
          IBA-1) and neuronal cell loss (labeled by microtubule-associated protein
          2, MAP-2) in the hippocampus.
        explanation: >-
          Reports the measurement, the markers used, and the accompanying
          behavioural change.
    - name: Cognitive impairment after TLR4 blockade
      target: Microglial Activation and Neuroinflammation
      direction: ABOLISHED
      interpretation: >-
        Preventing the deficit by blocking the receptor that initiates the
        inflammatory cascade is what makes the model causal rather than
        descriptive.
      evidence:
      - reference: PMID:30962497
        reference_title: Neuroinflammation induced by lipopolysaccharide causes cognitive impairment in mice.
        supports: SUPPORT
        evidence_source: MODEL_ORGANISM
        snippet: >-
          Furthermore, VIPER, which is a TLR-4-specific inhibitory peptide,
          prevented the LPS-induced neuroinflammation and cognitive impairment.
        explanation: >-
          The rescue arm establishing that the inflammatory step is necessary for
          the cognitive deficit.
differential_diagnoses:
- name: Dementia
  description: >-
    The differential that matters most for the subacute presentation, because
    an insidious onset over days to weeks removes the abruptness that usually
    separates the two. Both produce inattention, disorientation, and memory
    impairment.
  distinguishing_features:
  - >-
    Delirium is distinguished by its acute change from a known baseline, its
    fluctuating course across the day, and disturbance of the level of
    consciousness — none of which characterize uncomplicated dementia.
    Establishing the patient's baseline from an informant is the decisive step,
    and dementia is itself the strongest predisposing factor, so the two
    frequently coexist as delirium superimposed on dementia.
  evidence:
  - reference: PMID:23992774
    reference_title: Delirium in elderly people.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      To differentiate delirium from dementia, obtaining the history is critical to
      establish the patient's baseline, determine the acuity of mental status change
      and fluctuations typical of delirium, and to search for etiologic clues.
    explanation: >-
      States the distinguishing procedure and the features it turns on.
- name: Depression
  description: >-
    Hypoactive delirium — withdrawal, psychomotor slowing, reduced engagement —
    is readily mistaken for depression, particularly in an older inpatient with
    a depression history.
  distinguishing_features:
  - >-
    Depression does not produce the fluctuating inattention or the altered level
    of consciousness that define delirium, and lacks an acute precipitant on
    work-up. Formal cognitive testing that demonstrates inattention is the
    practical discriminator.
  evidence:
  - reference: PMID:30346242
    reference_title: Haloperidol and Ziprasidone for Treatment of Delirium in Critical Illness.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      hypoactive delirium prevents participation in nursing interventions, physical
      therapy, and occupational therapy
    explanation: >-
      Describes the quiet, withdrawn presentation that generates the confusion
      with depression.
discussions:
- discussion_id: subacute_vs_acute_delirium_nosology
  kind: KNOWLEDGE_GAP
  prompt: >-
    Does "subacute delirium" name anything mechanistically distinct from
    delirium with a protracted course, or is MONDO:0004629 purely a temporal
    qualifier?
  attaches_to:
  - pathophysiology#Protracted and Persistent Course
  rationale: >-
    ICD-10 F05 explicitly bundles "acute or subacute confusional state" under a
    single code, and no study identified during curation compares the
    pathophysiology of insidious-onset delirium with that of abrupt-onset
    delirium. The persistent-delirium literature quantifies how long delirium
    lasts, but persistence after an acute onset is not the same claim as an
    insidious onset. Until a study stratifies mechanism or biomarker by onset
    tempo, this entry curates the delirium pathograph with a temporal qualifier
    and does not assert a distinct mechanism.
- discussion_id: rodent_models_lack_attentional_readout
  kind: HUMAN_MODEL_MISMATCH
  prompt: >-
    Do rodent inflammatory-challenge models reproduce the attentional
    disturbance that defines human delirium, or only the sickness behaviour and
    memory impairment that accompany it?
  attaches_to:
  - animal_models#Mouse
  rationale: >-
    The ME7/TNF-alpha model reproduces the conditional structure of delirium
    convincingly — the same insult, a deficit only on a vulnerable background —
    but its readout is T-maze working memory, and the accompanying phenotype is
    sickness behaviour (hypothermia, weight loss, inactivity). Inattention and
    disturbed awareness, the DSM-defining features, have no established rodent
    equivalent. Evidence therefore exists in the model, but its translational
    validity for the cardinal feature specifically remains open. This is a
    mismatch of readout, not an absence of data.
- discussion_id: apoe_mechanism_of_susceptibility
  kind: OPEN_QUESTION
  prompt: >-
    How does APOE genotype confer delirium risk independently of dementia — is
    it lipid transport and synaptic resilience, an amplified neuroinflammatory
    response, or something else?
  attaches_to:
  - genetic#APOE
  rationale: >-
    The multi-ancestry meta-analysis establishes the association and its
    independence from dementia, and the accompanying proteomics implicate brain
    vulnerability, inflammation and immune-response processes. But the step from
    APOE genotype to any node in this pathograph is not established: the entry
    draws no causal edge from APOE to the microglial or reserve nodes because
    none is evidenced. Resolving this would connect the strongest genetic signal
    in delirium to the mechanism the rest of the entry curates, and the
    druggability analysis in the same paper suggests it is being actively
    pursued.
notes: >-
  Scope. MONDO:0004629 is a course-qualified synonym of delirium rather than a
  separate disease, so this entry curates the delirium pathograph and treats
  "subacute" as a temporal qualifier on it, with a dedicated
  `Protracted and Persistent Course` node and an explicit knowledge-gap
  discussion. If dismech later curates delirium as a top-level entry, this
  entry should be reconciled with it rather than duplicated.

  Deliberate omissions, corrected. An earlier revision of this entry asserted
  that four claims could not be curated because their sources resolved only to
  journal or PMC URLs. That was wrong for three of them: the deep-research run's
  own reference validation had already cached the APOE GWAS (PMID:41286463), the
  S100B/NfL systematic review (PMID:40889075), and the 4D dexmedetomidine trial
  (PMID:41160116), and all three are now curated. The lesson is worth recording:
  check `references_cache/` before concluding a deep-research citation is
  unresolvable, because the validator fetches more than the report cites by
  identifier.

  What genuinely remains uncurated is the thalamic and caudate
  functional-connectivity finding, whose only cached record
  (DOI:10.1176/appi.ajp.2012.11060976) has `content_type: unavailable` and so
  offers no quotable text. It is a real lead and is not curated as evidenced
  content here.
📚

References & Deep Research

References

2
PMID:23992774
Delirium in elderly people.
No top-level findings curated for this source.
PMID:29278283
Delirium pathophysiology: An updated hypothesis of the etiology of acute brain failure.
No top-level findings curated for this source.

Deep Research

1
Claude Code ▸
Subacute Delirium: Comprehensive Research Report
claude-haiku-4-5-20251001, claude-sonnet-5 55 citations 2026-08-25T04:38:44.256131

Subacute Delirium: Comprehensive Research Report

Note on Nosology

"Subacute Delirium" is not an independently defined disease entity with its own genetic or pathophysiological literature — it is a temporal-course descriptor within the single clinical syndrome of delirium. ICD-10-CM code F05 ("Delirium due to known physiological condition") is explicitly defined to include "acute or subacute confusional state," "acute or subacute brain syndrome," and "acute or subacute psycho-organic syndrome," with F05.1 covering delirium superimposed on dementia icd10data.com. MONDO (MONDO:0004629) and UMLS/MedGen (C0154333, SNOMED CT 191507002) carry "Subacute Delirium" as a synonym set ("Delirium, Subacute"; "Subacute confusional state") mapped onto the same underlying concept as delirium generally MedGen C0154333. Consequently, this report treats "subacute delirium" as delirium with a subacute onset/course (onset over days rather than hours, and/or a protracted duration of weeks rather than days) and draws on the delirium literature broadly, flagging the subset of studies that specifically address protracted/persistent/subacute courses (Sections 8 and 11 in particular).

1. Disease Information

Overview. Delirium is an acute neuropsychiatric syndrome of disturbed attention, awareness, and cognition that develops over a short period (hours to days — "acute") or, in the subacute variant, over a somewhat longer interval, and represents a direct physiological consequence of an underlying medical condition, substance intoxication/withdrawal, or medication effect, or multiple combined etiologies [MalaCards / MONDO:0004629 description via search]. Core features include a disturbance of attention and awareness, an additional cognitive disturbance (memory, disorientation, language, visuospatial ability, or perception), development over a short period with a tendency to fluctuate in severity during the day, and evidence that the disturbance is not better explained by a pre-existing/evolving dementia.

Key identifiers: | Ontology | Identifier | |---|---| | MONDO | MONDO:0004629 ("subacute delirium") | | UMLS/MedGen | C0154333 | | SNOMED CT | 191507002 | | ICD-10-CM | F05 (Delirium due to known physiological condition — "acute or subacute confusional state"); F05.1 (superimposed on dementia) | | DSM-5 | Delirium, specified as acute (hours–days) or persistent (weeks–months) |

Synonyms: Subacute confusional state; acute confusional state; toxic-metabolic encephalopathy; organic brain syndrome (historical term); ICU psychosis (informal, ICU-specific); sundowning (informal, when evening-predominant) MedGen C0154333.

Data source note: Most quantitative data below derive from aggregated, cohort-level clinical research (hospital registries, ICU cohorts, meta-analyses) rather than individual EHR mining per se, though several cited studies (e.g., UK Biobank-linked COVID-19 cohort, US World Delirium Awareness Day prevalence study) are EHR/registry-based.

2. Etiology

Causal framework. Delirium arises from an interaction between predisposing (vulnerability) factors and precipitating (insult) factors — a patient with high vulnerability (e.g., advanced age, dementia) can develop delirium from a minor insult, while a resilient patient requires a major insult (multiple simultaneous new medications, major surgery, severe critical illness).

Direct causal factors: - Systemic infection/sepsis and acute critical illness - Major surgery/anesthesia (postoperative delirium) - Metabolic derangement (electrolyte disturbance, hypo-/hyperglycemia, hepatic/renal failure) - Medication effects, polypharmacy, and substance intoxication/withdrawal (including alcohol and benzodiazepine withdrawal) - Hypoxia/hypoperfusion - Primary CNS insults (stroke, seizure, traumatic brain injury, CNS infection)

Genetic risk factors. The largest multi-ancestry GWAS of delirium to date analyzed 1,059,130 individuals (11,931 cases) and identified APOE as a strong risk locus, with the effect persisting after adjustment for dementia/Alzheimer's disease and within dementia-free cohorts — indicating APOE independently confers delirium vulnerability rather than acting solely through dementia risk PLOS Medicine GWAS. A large UK cohort further found that APOE ε4 genotypes increased risk of delirium during COVID-19-related hospitalizations PMC8344705. Additional postoperative-delirium-associated loci include APOC1, TOMM40, and PVRL2, genes previously linked to dementia, cognitive decline, and cerebral imaging phenotypes [Neuroscience News summary]. Notably, earlier candidate-gene associations between the muscarinic cholinergic receptor genes CHRM2 and CHRM4 and postoperative delirium were not replicated in more recent, better-powered analyses, with all three previously identified variants showing null effects medRxiv / Nature Aging, Dissecting the genetic and proteomic risk factors for delirium.

Environmental/clinical risk factors (odds ratios from recent meta-analyses of hospitalized older patients): - Frailty: OR ≈ 2.05 - Physical restraints: OR ≈ 5.01 - Prior falls: OR ≈ 1.99 - Severe illness: OR ≈ 1.32 - Cognitive impairment/dementia: OR ≈ 2.61 ScienceDirect, Global incidence and prevalence of delirium, 2024; PubMed 39602991

Other established risk factors include advanced age, polypharmacy (especially anticholinergics, benzodiazepines, opioids), sensory impairment (vision/hearing), immobility, dehydration, sleep deprivation, indwelling catheters/lines, and mechanical ventilation.

Protective factors. Multicomponent nonpharmacological prevention programs are the best-evidenced protective intervention (see Section 13); no robust genetic protective variant has been established, though the absence of the APOE ε4 allele is associated with lower risk by extension of the GWAS findings above.

Gene-environment interaction. The APOE-delirium association is modulated by acute physiological stressors (e.g., COVID-19 infection, surgery), consistent with a "two-hit" model in which genetic vulnerability (APOE-mediated neuronal/glial resilience) interacts with an environmental/physiological precipitant (infection, surgery, critical illness) to produce clinical delirium PMC8344705.

3. Phenotypes

Delirium's phenotype spans cognitive, behavioral, and psychomotor domains, with laboratory/imaging correlates increasingly recognized as biomarker-level abnormalities.

Phenotype Suggested HP term Notes
Inattention HP:0032341 (Impaired social cognition) / general "attention deficit" — closest specific term is HP:0007018 (Attention deficit) Cardinal feature; assessed via digit span, months-backward
Disorientation HP:0031466 (disorientation) if available, else HP:0000726 (Dementia) subset Fluctuating; worse with evening ("sundowning")
Fluctuating consciousness/awareness HP:0007360 (Aggressive behavior) not applicable; consider HP:0000726 or HP:0002360 (Sleep disturbance) for circadian component Hallmark diagnostic feature (DSM-5 Criterion A/B)
Memory impairment HP:0002354 (Memory impairment) Both encoding and recall affected
Perceptual disturbances (hallucinations, illusions) HP:0000738 (Hallucinations) More common in hyperactive subtype
Psychomotor agitation HP:0000723 (Restlessness) / HP:0100716 (Self-injurious behavior) in severe cases Hyperactive subtype
Psychomotor retardation / lethargy HP:0025336 (Psychomotor retardation) Hypoactive subtype — most common but most underdiagnosed
Sleep-wake cycle disturbance HP:0002360 (Sleep disturbance) Circadian reversal common
Disorganized thinking/speech HP:0031936 (Delusions) or HP:0000750 (Delayed speech and language development, N/A for adults) — best mapped as thought-disorder qualifier
Emotional lability HP:0000712 (Emotional lability)

Motor subtypes and frequency. A subacute-care cohort study of patients ≥65 admitted with delirium found: hyperactive delirium 40.6%, mixed 31%, hypoactive 25.9%, nonmotor 2.6% [PubMed search — persistent delirium subtype study]. Hypoactive delirium, though less prevalent in some settings, is disproportionately missed clinically because of its quiet presentation.

Onset/severity/progression per DSM-5: Delirium severity is formally specified as acute (a few hours to days) or persistent (weeks to months) — the "subacute" category sits at the acute-to-persistent transition, typically representing onset over roughly 1–2 weeks with a course of several weeks [DSM-5 criteria search summary].

Frequency/persistence data: Combined proportions of patients with persistent delirium were 44.7% at discharge, 32.8% at 1 month, 25.6% at 3 months, and 21% at 6 months in an updated systematic review and meta-analysis medRxiv persistent delirium meta-analysis; Delirium Journal version. In patients with and without dementia, delirium symptoms persisted up to 12 months post-diagnosis, with inattention, disorientation, and impaired memory the most persistent individual symptoms in both groups.

Quality of life impact: Persistent/subacute delirium is associated with functional decline, increased nursing-home placement, and worse cognitive trajectory relative to patients whose delirium resolves acutely PMC5506578, subsyndromal delirium meta-analysis.

4. Genetic/Molecular Information

Delirium is not a monogenic disorder; there is no single causal gene analogous to a Mendelian disease. Rather, common-variant susceptibility loci modulate risk in the context of an acute precipitant.

  • APOE (chr19; HGNC:613) — strongest and most replicated genetic risk factor identified in the largest delirium GWAS to date (N=1,059,130, 11,931 cases), independent of dementia status PLOS Medicine.
  • APOC1, TOMM40, PVRL2 (NECTIN2) — chromosome 19 loci in linkage disequilibrium with APOE, associated with postoperative delirium and broader neurocognitive phenotypes.
  • CHRM2, CHRM4 (muscarinic acetylcholine receptors) — earlier candidate-gene hits, not replicated in recent, adequately powered analyses (null effect in the 2024/2025 genetic-proteomic dissection study) Nature Aging 2025.

Proteomic/functional correlates: The same 2024–2025 genetic-and-proteomic dissection identified multiple blood-based proteins predictive of delirium risk years in advance, including markers of brain injury and inflammation not previously linked to delirium — supporting a model in which chronic subclinical neuronal vulnerability (proteomically detectable) interacts with an acute precipitant Nature Aging; PMC12823428.

Functional consequence framing: These findings support a susceptibility/modifier model (GENO: GENO:0000217 susceptibility) rather than a loss/gain-of-function causal mutation model — APOE ε4 carriage confers relative risk via impaired neuronal lipid transport, reduced synaptic resilience, and amplified neuroinflammatory response to systemic insults, rather than through a specific structural variant.

Epigenetics/chromosomal abnormalities: No delirium-specific epigenetic signature or chromosomal abnormality has been established in the literature surveyed; delirium is fundamentally a syndromic, precipitant-driven state superimposed on variable genetic vulnerability rather than a primary genetic or chromosomal disorder.

5. Environmental Information

Environmental/iatrogenic factors: - Polypharmacy — anticholinergics, benzodiazepines, opioids, corticosteroids - Physical restraints (independently associated risk, OR ≈ 5.01) [ScienceDirect 2024 meta-analysis] - Indwelling catheters, intravenous lines, mechanical ventilation - ICU environment: sleep deprivation, sensory overload/deprivation, lack of day-night cues - Surgery/anesthesia exposure (postoperative delirium)

Lifestyle/patient-level factors: Pre-existing frailty, immobility, dehydration, malnutrition, sensory impairment (uncorrected vision/hearing loss), and sleep disruption are all established contributors, consolidated in the "predisposing × precipitating factor" model of delirium.

Infectious triggers: Systemic infection/sepsis is one of the most common precipitants; COVID-19 specifically has been shown to precipitate delirium at elevated rates in APOE ε4 carriers PMC8344705. LPS (bacterial endotoxin)-driven systemic inflammation is the standard experimental proxy for infection-triggered delirium in animal models (see Section 15).

6. Mechanism / Pathophysiology

Delirium's mechanism is multifactorial and converges on acute, reversible cortical-subcortical network dysfunction driven by systemic and neuro-inflammation, neurotransmitter imbalance, and blood-brain barrier compromise.

Causal chain (systemic trigger → clinical syndrome): 1. Systemic insult (infection, surgery, metabolic derangement) → release of peripheral pro-inflammatory cytokines (IL-6, TNF-α, IL-1β) and cortisol. 2. Blood-brain barrier (BBB) dysfunction — increasingly recognized as a key permissive mechanism allowing peripheral inflammatory signals to reach the CNS PMC11622424, Pathophysiology and Biomarkers of Delirium. 3. Neuroinflammation — microglial activation, NF-κB pathway activation, and local cytokine release (TNF-α, IL-1β, PGE2, nitric oxide) in the hippocampus and cortex, demonstrated mechanistically in LPS mouse models Scientific Reports LPS study. 4. Neurotransmitter dysregulation — dysregulated cholinergic (relative deficiency) and dopaminergic (relative excess) signaling is a long-standing mechanistic hypothesis, though the specific cholinergic receptor gene candidates have not held up genetically (see Section 4). 5. Synaptic and network dysfunction — disrupted thalamocortical and frontal-parietal connectivity. 6. Clinical manifestation — inattention, disorientation, and fluctuating consciousness.

Molecular pathways: NF-κB signaling (pro-inflammatory transcription); cortisol/HPA-axis signaling exacerbating neuroinflammation and impairing synaptic function Frontiers postoperative delirium biomarkers review.

Cellular processes: Microglial activation and synaptic remodeling/pruning; astrocytic activation (reflected by S100B elevation); neuronal injury (reflected by NfL, tau, pTau elevation) PMC11622424; PMC10985356, Serum NFL and tau.

Network/circuit-level findings (neuroimaging): - Reduced functional connectivity of the intralaminar thalamic and caudate nuclei with other subcortical regions during a delirium episode, which recovers after resolution AJP, Neural Network Functional Connectivity During and After an Episode of Delirium. - Disrupted reciprocal (inverse) coupling between the dorsolateral prefrontal cortex and posterior cingulate cortex — normally anti-correlated, these regions become abnormally positively correlated during delirium. - EEG shows increased delta/theta power, reduced alpha power, and reduced functional connectivity, and loss of feedback cortical connectivity has been specifically implicated Alzheimer's & Dementia 2024, Gjini et al.; Frontiers EEG systematic review.

Biochemical/biomarker abnormalities: Elevated IL-6, CRP, TNF-α (systemic inflammation, variable specificity); S100B and NfL show the most consistent CNS-injury-marker associations with delirium presence and severity, implicating astrocytic and axonal injury respectively PMC12401822, Serum biomarkers of delirium in critical illness systematic review. A 2024/2025 CSF study found CSF sIL-6R correlated with plasma IL-6/IL-8 and with blood-brain barrier permeability (CSF:plasma albumin ratio, plasma S100B), directly linking peripheral inflammation to BBB compromise in postoperative delirium BJA Open.

GO/CL/UBERON term suggestions: - GO:0034612 (response to tumor necrosis factor), GO:0032496 (response to lipopolysaccharide), GO:0007249 (I-kappaB kinase/NF-kappaB signaling), GO:0006954 (inflammatory response) - CL:0000129 (microglial cell), CL:0000127 (astrocyte) - UBERON:0001954 (Ammon's horn/hippocampus), UBERON:0002435 (striatum, incl. caudate), UBERON:0002444 (thalamus), UBERON:0009834 (dorsolateral prefrontal cortex)

7. Anatomical Structures Affected

Organ level: Primary organ affected is the brain (central nervous system); delirium is by definition a whole-brain functional (not typically focal-structural) disturbance, though it frequently co-occurs with the organ system driving the underlying precipitant (e.g., respiratory failure, sepsis-affected multiple organs, post-surgical state).

Tissue/regional level: - Thalamus (intralaminar nuclei) — reduced functional connectivity during delirium - Basal ganglia / caudate nucleus — reduced subcortical connectivity - Prefrontal cortex (dorsolateral) — disrupted reciprocal connectivity with posterior cingulate - Posterior cingulate cortex — abnormal positive coupling with DLPFC during delirium - Hippocampus — site of microglial activation and neuronal injury in animal models (LPS)

Subcellular level (GO Cellular Component): Synapse (GO:0045202), microglial process, astrocytic end-foot at the blood-brain barrier (relevant to BBB dysfunction mechanism).

Localization: Diffuse/bilateral cortical-subcortical network disturbance rather than a single lateralized lesion; this distinguishes delirium from focal stroke-related confusional states.

8. Temporal Development

Onset: - Acute delirium: onset over hours to a few days (DSM-5). - Subacute delirium: onset over an intermediate timeframe, typically several days to roughly 1–2 weeks, representing an insidious rather than abrupt presentation — often seen with slowly evolving metabolic derangements, indolent infections, or medication accumulation effects, and clinically important because it is more easily mistaken for evolving dementia MedGen C0154333 clinical research summary. - Delirium can occur at any adult age but disproportionately affects older adults; onset in critical illness or postoperative settings is typically within the first several days of the inciting event.

Progression / course pattern: - Fluctuating course is a defining diagnostic feature — symptom severity waxes and wanes over the course of a day, classically worsening in the evening ("sundowning"). - DSM-5 persistence specifier: acute (hours–days) vs. persistent (weeks–months) — subacute delirium occupies the transition zone and often evolves into persistent delirium if the underlying precipitant is not resolved. - Persistent delirium prevalence: 44.7% at hospital discharge, declining to 32.8% (1 month), 25.6% (3 months), and 21% (6 months) in a pooled meta-analysis of older hospitalized patients medRxiv. - Symptoms can persist up to 12 months, with inattention, disorientation, and memory impairment the most persistent individual features [search summary, persistent delirium 12-month follow-up].

Remission patterns: Delirium is classically defined as reversible with treatment of the underlying cause, though full cognitive recovery is not universal — a substantial minority never return to baseline cognition, particularly in patients with pre-existing cognitive impairment.

Critical periods / intervention windows: The subacute-to-persistent transition represents a key intervention window — early identification and treatment of the underlying precipitant during the subacute phase is associated with better functional and cognitive outcomes than intervention after the delirium becomes persistent PMC12404460, Persistent inpatient delirium and increased LOS/mortality.

9. Inheritance and Population

Epidemiology: - Pooled prevalence 23.6% and pooled incidence 13.5% among medically hospitalized older patients (recent systematic review/meta-analysis) ScienceDirect 2024; PubMed 39602991. - US hospitalized-patient prevalence ranges ~16–18%, rising to ~23.6% in older adults specifically. - 2023 US cross-sectional World Delirium Awareness Day study: clinically documented delirium prevalence 16.4% (morning) and 17.9% (evening) assessments ScienceDirect. - Prevalence reaches up to 50% in hospitalized elderly populations broadly, and ~32.5% incidence among critically ill (ICU) patients, rising further with mechanical ventilation and organ dysfunction; ~47.7% prevalence among post-surgical ICU patients [search summary]. - Community-dwelling older adults: systematic review of prevalence/incidence/risk factors in home settings (BJGP 2025) BJGP.org.

Inheritance pattern: Not a Mendelian disorder — complex/multifactorial, with common-variant (APOE) susceptibility contributing modest, non-deterministic risk in the context of an acute precipitant. No formal penetrance, expressivity, anticipation, mosaicism, or founder-effect data are applicable in the Mendelian sense; APOE ε4 allele frequency itself varies by ancestry (well-characterized in gnomAD/population genetics resources independent of delirium).

Population demographics: - Age: strongest independent risk factor — prevalence rises sharply with advancing age, particularly ≥65 years. - Sex: no strong, consistent sex-ratio skew reported across the reviewed epidemiological studies; risk is driven primarily by frailty, comorbidity burden, and acute illness severity rather than sex per se. - Setting-specific demographics: nephrology ward cohorts, pneumonia cohorts (5-decade prevalence/mortality meta-analysis), and emergency-department ambulance-arrival cohorts each show elevated delirium prevalence in older, frailer sub-populations PMC12565244; PMC12148281; PMC11950662.

10. Diagnostics

Clinical/bedside diagnostic tools: - Confusion Assessment Method (CAM) — the most widely used bedside screening/diagnostic algorithm, requiring acute onset/fluctuating course + inattention, plus either disorganized thinking or altered consciousness [search summary, DSM-5/CAM comparison]. - CAM-ICU — ICU-adapted version for non-verbal/ventilated patients. - DSM-5 criteria — considered relatively more restrictive than DRS-R98 or CAM in head-to-head comparisons, identifying a somewhat different (often narrower) case set PMC4207319, DSM-IV vs DSM-5 concordance; PubMed 25601222; PubMed 28903799. - DRS-R98 (Delirium Rating Scale-Revised-98) — severity-graded instrument, used as a comparator/gold standard in validation studies. - Level-of-arousal-inclusive DSM-5 interpretation is argued to be "safer" (more sensitive) for capturing genuine delirium cases PMC4177077.

Laboratory tests: No delirium-specific diagnostic lab test exists; work-up targets the underlying precipitant — CBC, metabolic panel, liver/renal function, TSH, blood/urine cultures, ammonia, toxicology screen as clinically indicated.

Emerging biomarkers (research/investigational, not yet routine clinical diagnostics): - S100B and neurofilament light chain (NfL) — most consistent CNS-injury-marker associations with delirium presence/severity across a systematic review of serum biomarkers in critical illness PMC12401822. - Serum tau and NfL correlated with delirium in a 3-year retrospective analysis, while serum UCHL-1, GFAP, and CSF SNAP-25/NPTX2/sTREM2 did not PMC10985356. - CSF IL-6, soluble IL-6 receptor, and IL-6 trans-signalling complex associated with postoperative delirium in an observational cohort BJA Open. - EEG-based quantitative/functional-connectivity measures — increased delta/theta power, reduced alpha power, reduced connectivity, and loss of feedback cortical connectivity are being explored as objective, non-invasive diagnostic adjuncts Frontiers systematic review; Alzheimer's & Dementia 2024. - Bispectral EEG has also been used as a discovery platform to identify novel protective agents for infection-related delirium in preclinical models Translational Psychiatry 2024.

Genetic testing: Not clinically indicated for delirium diagnosis (it is an acute syndromic diagnosis, not a genetic disease); APOE genotyping is a research risk-stratification tool, not a diagnostic test.

Differential diagnosis: Dementia (chronic, typically non-fluctuating, preserved consciousness), depression (particularly in hypoactive delirium), primary psychiatric psychosis, non-convulsive status epilepticus, stroke/focal neurological deficit, and delirium superimposed on dementia (ICD-10 F05.1) — which is common and requires careful distinction of an acute change from the patient's cognitive baseline.

Screening: No population-level genetic or newborn screening applies (delirium is an acquired, precipitant-driven acute-care condition); "screening" in practice means routine CAM/CAM-ICU or 4AT bedside screening of at-risk hospitalized patients (elderly, ICU, postoperative) rather than genetic carrier screening.

11. Outcome/Prognosis

Mortality: - Persistent delirium is associated with increased 30-day mortality and higher 60-, 90-, 180-, and 360-day mortality, with an incremental increase in mortality risk for each additional day of delirium [search summary, persistent delirium outcomes meta-analysis]; PMC12404460, increased LOS and mortality. - Outcomes (mortality, nursing-home placement, function, cognition) for patients with persistent delirium are consistently worse than for patients who recover from delirium PMC5506578; PubMed 19017678, systematic review of frequency and prognosis.

Long-term cognitive outcomes / dementia risk: - A meta-analysis of 23 studies found delirium significantly associated with long-term cognitive decline (Hedges g = 0.45, 95% CI 0.34–0.57, P<.001) JAMA Neurology meta-analysis; PubMed 32658246. - Patients experiencing in-hospital delirium showed markedly increased dementia risk post-discharge (OR ≈ 5.37, P<.001) compared with those without delirium [search summary]. - A separate estimation study modeled how many dementia cases might be prevented by preventing delirium, underscoring delirium as a modifiable dementia risk factor rather than a purely bystander phenomenon PubMed 37031027. - A recent (2025) systematic review/meta-analysis of long-term clinical outcomes after hospital discharge for delirium further consolidates the adverse-outcome signal Age and Ageing 2025.

Morbidity/function: Increased length of stay, functional decline, and nursing-home placement are consistently reported; hyperactive, mixed, hypoactive, and nonmotor delirium subtypes differ in discharge destination and functional trajectory in subacute-care cohorts.

Prognostic factors: Duration of delirium (each additional day worsens mortality risk), pre-existing dementia, frailty, and severity of the underlying precipitating illness are the most consistently reported prognostic modifiers.

12. Treatment

Pharmacotherapy: - Antipsychotics (haloperidol, ziprasidone): The NEJM MIND-USA trial found no significant difference between haloperidol, ziprasidone, and placebo for delirium duration or severity in critically ill patients NEJM 2018, Haloperidol and Ziprasidone for Treatment of Delirium in Critical Illness(landmark trial; foundational to current guideline skepticism about routine antipsychotic use). A 2024 follow-up analysis of long-term outcomes similarly found no difference in long-term quality of life with haloperidol vs. placebo across the identified trials ScienceDirect 2024, MIND-USA long-term outcomes. - Dexmedetomidine (alpha-2 agonist): The 4D trial (2024/2025) compared haloperidol-first-then-dexmedetomidine strategies for hyperactive delirium in non-intubated ICU patients, finding improved agitation control, reduced need for additional medications, and reduced ICU length of stay/costs with the dexmedetomidine-inclusive strategy Intensive Care Medicine 2025; PMC12678554. - Olanzapine vs. low-dose dexmedetomidine: A 2024 randomized trial in critically ill patients directly compared these agents AJRCCM 2024. - Dexmedetomidine + melatonin for post-CABG delirium: a 2023 RCT (n=80) found delirium occurrence lower with combined dexmedetomidine+melatonin (15%) vs. dexmedetomidine alone (30%) PMC10664157.

Suggested NCIT terms: NCIT:C15986 (Pharmacotherapy) for the generic action, with therapeutic_agent bindings to CHEBI for haloperidol, ziprasidone, olanzapine, dexmedetomidine, and melatonin individually.

Non-pharmacological / supportive care: Reorientation, early mobilization, sleep-hygiene protocols, hearing/vision aid provision, hydration/nutrition support, and minimizing deliriogenic medications and physical restraints — the core components of the Hospital Elder Life Program (HELP) (see Section 13), applied both preventively and as first-line management of established delirium.

Rehabilitation: Physical and occupational therapy to address functional decline associated with persistent delirium; NCIT:C15302 (Physical Therapy).

Experimental/investigational: Bispectral-EEG-guided discovery platforms are being used to screen for novel protective agents against infection-related delirium in preclinical models Translational Psychiatry 2024.

Treatment strategy summary: Current evidence favors treating the underlying precipitant first, applying non-pharmacological multicomponent strategies as first line, and reserving antipsychotics/dexmedetomidine for severe agitation or patient/staff safety concerns rather than as disease-modifying therapy — reflecting the accumulated null results for antipsychotics on delirium duration/long-term outcomes.

13. Prevention

Primary prevention — multicomponent nonpharmacological interventions: - The Hospital Elder Life Program (HELP), developed by Sharon K. Inouye's group at Yale (1999), bundles reorientation, early mobilization, therapeutic activities, hydration/nutrition support, sleep-protocol strategies, and hearing/vision adaptation PMC3724594. - The original NEJM 1999 trial established multicomponent targeted intervention efficacy for preventing delirium onset and reducing total delirium-days in hospitalized older medical patients NEJM 1999. - A 2024 systematic review/meta-analysis of HELP found subgroup effect estimates of RR ≈ 0.65 (HELP programs) and RR ≈ 0.70 (non-HELP multicomponent programs) for delirium incidence reduction ScienceDirect 2024. - A 2024/2025 umbrella review of RCTs for delirium prevention/treatment interventions further consolidates this evidence base ScienceDirect umbrella review. - Nurse-led multicomponent interventions have separately been shown effective in a 2025 systematic review/meta-analysis Journal of Advanced Nursing 2025.

Secondary prevention: Routine bedside screening (CAM/CAM-ICU/4AT) of at-risk hospitalized, postoperative, and ICU patients for early detection, permitting prompt treatment of an emerging precipitant before delirium becomes persistent/subacute-to-chronic.

Tertiary prevention: Once delirium is established, minimizing deliriogenic medications, restraints, and immobility, and aggressively treating the underlying condition, reduces progression to persistent delirium and its associated mortality/cognitive-decline risk.

Pharmacological prophylaxis: Evidence for pharmacological delirium prophylaxis (e.g., prophylactic antipsychotics or dexmedetomidine) remains mixed and is not a first-line recommendation given the null long-term outcome data for antipsychotics; nonpharmacological multicomponent prevention remains the standard of care.

Counseling: Not applicable in the genetic-counseling sense (delirium is not a heritable single-gene disorder); patient/family education regarding delirium risk, expected fluctuating course, and reorientation strategies is a core component of both prevention and management.

14. Other Species / Natural Disease

Delirium as classically defined is a human clinical/DSM construct; there is no widely characterized spontaneous veterinary "delirium" literature comparable to human clinical case series. However: - Veterinary anesthesia/critical-care literature describes post-anesthetic emergence delirium/dysphoria and ICU-associated confusional states in companion animals (dogs, cats) that are mechanistically analogous (systemic inflammation, anesthesia, critical illness precipitating acute confusional states), though this is not formally captured in OMIA as a discrete inherited disease. - The orthologous APOE gene is broadly conserved across mammals (NCBI Gene), consistent with cross-species use of Apoe-knockout or humanized-APOE mouse lines in delirium-adjacent neuroinflammation research (see below).

15. Model Organisms

Primary model: LPS (lipopolysaccharide)-induced systemic inflammation in mice. This is the dominant experimental proxy for infection-triggered delirium: - LPS administration produces acute-onset, fluctuating cognitive impairment meeting DSM-IV-analogous criteria in mice, with microglial activation and neuronal cell loss in the hippocampus, assessed via Morris water maze and passive avoidance tests Scientific Reports 2019; PMC6453933. - LPS increases TNF-α, IL-1β, PGE2, and nitric oxide, with NF-κB pathway activation, directly modeling the neuroinflammatory mechanism described in Section 6. - Interventional studies in this model include deferoxamine (iron chelation) attenuating LPS-induced neuroinflammation/memory impairment PMC4323121, a Banhasasim-Tang (traditional herbal formulation) attenuation study PMC7400939, and hippocampal GPR17 knockdown/inhibition attenuating LPS-induced cognitive impairment PMC10662506.

Perioperative/subclinical infection combination model: A novel adult mouse model combining subclinical infection with surgery to induce cognitive dysfunction has been developed specifically as a model for perioperative neurocognitive disorder, closer to the clinical postoperative-delirium scenario than LPS alone PMC12832885.

Baseline vulnerability models: Mouse models with pre-existing cognitive impairment/neurodegenerative pathology (e.g., amyloid-pathology models) show progressively increased "delirium-like" susceptibility to a superimposed acute inflammatory insult, mechanistically supporting the predisposing-factor × precipitating-factor clinical model PMC4278840, Worsening Cognitive Impairment and Neurodegenerative Pathology Progressively Increase Risk for Delirium.

Model limitations: Mouse "sickness behavior" (reduced activity, reduced food intake) following LPS overlaps with but is not identical to human delirium's core attention/awareness disturbance, which is difficult to operationalize behaviorally in rodents; most LPS studies rely on validated cognitive-impairment proxies (water maze, passive avoidance) rather than a direct delirium-equivalent readout, so translational fidelity for the attentional component of delirium specifically remains an open limitation.

Applications: These models are used to dissect the neuroinflammation → BBB dysfunction → microglial activation → cognitive impairment causal chain, to screen candidate protective/therapeutic agents (e.g., the bispectral-EEG-guided discovery platform noted in Section 12) Translational Psychiatry 2024, and to test genetic-risk-modifying manipulations (e.g., Apoe-genotype mouse lines, though a delirium-specific Apoe-mouse study was not identified in this search and would be a natural translational follow-on to the human GWAS findings in Section 4).

Summary for Knowledge-Base Curation

"Subacute Delirium" should be curated as a temporal/course variant within the single delirium pathophysiology module rather than as a structurally distinct disease mechanism — the causal chain (systemic insult → BBB dysfunction → neuroinflammation → thalamocortical/frontal network dysfunction → clinical syndrome) is shared across acute, subacute, and persistent presentations, with course duration and reversibility as the key differentiating Descriptor (temporality: SUBACUTE / CHRONIC, clinical_course: PROGRESSIVE for cases evolving toward persistent delirium or dementia). APOE (hgnc:613) is the best-supported genetic modifier and should be entered as a SUSCEPTIBILITY relationship type rather than a causal gene. S100B and NfL are the most defensible biomarker phenotypes for a Biochemical/laboratory-abnormality entry given consistency across recent systematic reviews.

Sources

Reference Validation

Checked with linkml-reference-validator 0.2.1.

Outcome Count
References checked 36
Resolved 35
Unresolved (possible confabulation) 1
Unverifiable 0
References weighed for topical relevance 35
On topic 18
Off topic 0

Unresolved references

These identifiers did not resolve to a record and may be fabricated. A lookup that failed for transport reasons is indistinguishable from one that failed because the record does not exist, so spot-check before acting on them:

  • DOI:10.1056/NEJMoa1808217)(landmark (1 mention) - Identifier did not resolve to a record