Delirium

01Pathophysiology

Delirium is fundamentally an acute failure of global cerebral function. Unlike focal neurological deficits (stroke, mass lesion), delirium reflects a diffuse breakdown in the brain's ability to maintain attention, integrate sensory input, and sustain organized cognition. The hallmark is inattention, which is the single most important clinical feature and the feature that separates delirium from nearly all psychiatric mimics.

The leading neurochemical model centers on cholinergic deficiency and dopaminergic excess. Acetylcholine is the primary neurotransmitter for sustained attention and memory consolidation. When cholinergic transmission drops, whether from anticholinergic medications, metabolic insults, or systemic inflammation, attention collapses. This is why anticholinergic drugs are among the most common pharmacologic triggers. Conversely, dopamine excess contributes to the psychomotor agitation, hallucinations, and perceptual disturbances seen in the hyperactive subtype. This dopaminergic-cholinergic imbalance is also the rationale for using dopamine-blocking agents (antipsychotics) as symptomatic treatment.

A second critical mechanism is neuroinflammation. Systemic illness (sepsis, pneumonia, UTI) triggers peripheral cytokine release (IL-1, IL-6, TNF-alpha), which crosses the blood-brain barrier and activates microglial cells. This inflammatory cascade impairs synaptic function and neurotransmitter balance. This explains why even a "simple" UTI in an elderly patient with low cognitive reserve can produce a dramatic delirious state. The concept of cognitive reserve is central: a patient with pre-existing dementia has a lower threshold for delirium because the brain has fewer compensatory mechanisms to absorb additional insults.

The fluctuating course ("waxing and waning") directly reflects the dynamic interplay between the precipitating insult and the brain's attempts at compensation. This fluctuation is a defining feature and one of the strongest clinical clues on examination questions.

Visual hallucinations in delirium, as opposed to auditory hallucinations in primary psychotic disorders, arise from disruption of the visual association cortex and thalamic relay pathways. This is a high-yield discriminator: an elderly patient with acute-onset visual hallucinations is far more likely to have delirium than schizophrenia.

02Classification and Clinical Manifestation

Additional clinical features by domain:

03Diagnostic Workup

The diagnosis of delirium is clinical. There is no single laboratory test or imaging study that "confirms" delirium. The Confusion Assessment Method (CAM) is the best initial screening tool and is considered the standard bedside instrument. It requires four features, of which three must be present:

1. Acute onset and fluctuating course (required)

2. Inattention (required)

3. Disorganized thinking

4. Altered level of consciousness

Features 1 and 2 are mandatory. The presence of either feature 3 or 4 alongside them satisfies the diagnostic criteria. Pay close attention to this: inattention is the cardinal feature. If the patient can attend normally, the diagnosis is not delirium regardless of how confused they appear.

After establishing the diagnosis, the workup shifts entirely to identifying the precipitating cause. This is the part that exam questions focus on most heavily. A reasonable first-pass panel for any delirious patient includes CBC, BMP, urinalysis, chest X-ray, and a medication review. This combination covers the most common triggers: infection, metabolic derangement, and drug effect.

Brain imaging (CT or MRI) is not routine. It is ordered only when there are focal neurological signs (hemiparesis, cranial nerve palsies, new-onset seizures), a history of recent head trauma, or when no metabolic or infectious cause can be identified. An exam question that presents delirium with a clear UTI source and then asks "what is the next step" is testing whether you will waste time ordering a head CT. The answer in that scenario is to treat the UTI, not image the brain.

Lumbar puncture follows similar logic: it is reserved for patients with meningeal signs, immunocompromised states, or unexplained persistent encephalopathy.

EEG shows diffuse slowing in delirium (generalized theta and delta waves), which can help confirm the diagnosis in ambiguous cases and is particularly useful for detecting nonconvulsive status epilepticus, a treatable mimic.

04Management and Treatment

The foundational principle of delirium management is that delirium is a symptom, not a disease. The treatment is to find and fix the cause. An exam question that offers haloperidol as the "next best step" before a cause has been identified or addressed is almost always wrong.

Step 1: Medication reconciliation. Review every medication the patient is taking. Anticholinergics (diphenhydramine, oxybutynin, promethazine), benzodiazepines (lorazepam, diazepam), opioids, corticosteroids, and fluoroquinolones are among the most common drug triggers. Stopping or substituting the offending agent is often curative. This step is tested repeatedly: a postoperative patient on scheduled diphenhydramine for sleep who becomes acutely confused should have the diphenhydramine discontinued as the next best step.

Step 2: Address physiologic triggers. Treat infections with appropriate antibiotics. Correct electrolyte abnormalities. Ensure adequate oxygenation. Relieve urinary retention (this is a frequently overlooked but very testable cause). Manage undertreated pain, because both uncontrolled pain and excessive opioid use can trigger delirium.

Step 3: Non-pharmacologic interventions. These are always first-line and should be implemented in every case. Reorientation cues (visible clock, calendar, familiar family photos), minimizing room changes, maintaining the sleep-wake cycle (no unnecessary nighttime vitals or lab draws), early ambulation, and ensuring patients have their glasses and hearing aids. Physical restraints should be avoided because they worsen agitation and prolong delirium.

Step 4: Pharmacologic management. Medications are reserved only for patients whose agitation poses a safety risk to themselves or staff (pulling out central lines, endotracheal tubes, or physically combative behavior). Haloperidol remains the most widely referenced agent at a starting dose of 0.5 to 1 mg. It can be given IV, IM, or PO. Monitor the QTc interval, because haloperidol can prolong QT and precipitate torsades de pointes. Second-generation antipsychotics like quetiapine (starting at 12.5 to 25 mg PO at bedtime) are increasingly used, particularly in patients with Parkinson disease or Lewy body dementia where haloperidol is contraindicated due to severe sensitivity to dopamine blockade (can cause irreversible parkinsonism and neuroleptic malignant syndrome).

Critical exception: If the cause of delirium is alcohol withdrawal or benzodiazepine withdrawal, antipsychotics alone are inappropriate and dangerous. These patients require benzodiazepines (lorazepam or chlordiazepoxide) because the underlying mechanism is GABA receptor downregulation. Using only an antipsychotic in alcohol withdrawal can mask symptoms while the patient progresses to withdrawal seizures or delirium tremens. This is one of the most commonly tested "traps" surrounding delirium management.

Duration: Delirium typically resolves within days once the underlying cause is treated, but in elderly patients, it can persist for weeks. Antipsychotics should be tapered and discontinued as soon as agitation resolves; they are not continued long-term.

05Differential Diagnosis and Distractors

06Traps and High-Yield Pearls

The single most common way students lose points on delirium questions is by failing to recognize hypoactive delirium. Exam vignettes deliberately present a quiet, lethargic, withdrawn elderly postoperative patient and offer depression, dementia, or "normal postoperative fatigue" as answer choices. The correct answer is delirium, and the giveaway is that the change was acute (over hours to days) and the patient demonstrates inattention on bedside testing. Remember: hypoactive delirium is the most commonly missed subtype in real life and therefore the most commonly tested subtype on examinations.

The second major trap involves management sequencing. When a question asks for the "next best step" in a delirious patient, the answer is almost never "give haloperidol." The answer is to find and treat the cause: check a urinalysis, discontinue the anticholinergic medication, correct the sodium, obtain a chest X-ray. Jumping to pharmacologic management before addressing the underlying trigger is the classic wrong-answer pathway. Antipsychotics are a bridge for safety, not a treatment for delirium itself.

Third, never forget the alcohol and benzodiazepine withdrawal exception. If the vignette describes a patient 48 to 72 hours after hospital admission or after abruptly stopping alcohol/benzodiazepines, with tremors, autonomic instability (tachycardia, hypertension, diaphoresis), agitation, and visual hallucinations, the treatment is benzodiazepines, not antipsychotics. Using haloperidol alone in this setting is a patient safety error and a reliably tested wrong answer.

Finally, understand the concept of delirium superimposed on dementia. A patient with known Alzheimer disease can absolutely develop delirium. Many students see "dementia" in the history and assume the confusion is baseline. The discriminator is the tempo of change: if the family says "she has been like this for years," it is dementia. If the family says "she was fine yesterday and today she does not recognize me," it is delirium on top of dementia, and you must search for a precipitant.

The core competency being tested across all delirium questions is the ability to recognize an acute, fluctuating change in attention as a medical emergency requiring cause-directed evaluation, not simply symptomatic sedation.