The Agitated Patient with a Silent ECG
A 54-year-old man with no psychiatric history is brought to the emergency department by police after he was found wandering, shouting, and striking walls. He is diaphoretic, tremulous, and disoriented to place and time. Vital signs: HR 128, BP 168/96, T 38.1°C, RR 24. He is sweating profusely and has dilated pupils. He cannot give a coherent history. A relative arrives and reports he has been drinking 'a lot' daily for years and stopped about 36 hours ago after a minor argument; she also mentions he takes 'some tablets for nerves' but does not know which. The team wants to start intramuscular haloperidol and a benzodiazepine immediately for agitation.
1. Before sedation, what is the single most important diagnostic step, and why does the clinical picture make this mandatory rather than optional?
Read the worked answer
The most important step is to obtain an ECG and check the QTc (and electrolytes, especially potassium and magnesium), plus a urine/blood toxicology screen, before giving any QT-prolonging agent.
The picture is a sympathomimetic-toxic or withdrawal state with autonomic instability (tachycardia, hypertension, fever, mydriasis, diaphoresis, delirium). Two diagnoses must be actively excluded because they change management and can be rapidly fatal:
- Anticholinergic or sympathomimetic toxicity (e.g., from 'nerve tablets', stimulants, or a mixed overdose) - here the dilated pupils and dry flushed skin point away from pure alcohol withdrawal and toward a toxidrome.
- Alcohol withdrawal delirium (delirium tremens) - the 36-hour timeline and heavy chronic use fit, but DTs alone rarely produce mydriasis and the relative's 'nerve tablets' raise the possibility of a co-ingested agent.
Why the ECG is non-negotiable: haloperidol and many antipsychotics prolong QTc. In an undifferentiated agitated, febrile, tachycardic patient with possible electrolyte disturbance or drug effect, a prolonged QTc turns a 'routine' IM antipsychotic into a torsades risk. The ECG also screens for ischemia or arrhythmia driving the agitation. Sedation should not proceed until the rhythm and QTc are known and reversible causes (hypoglycemia, hypoxia, sepsis, head injury) are excluded with a bedside glucose and focused exam.
2. The ECG shows QTc 520 ms. Outline your pharmacological plan and justify each choice against the likely underlying diagnoses.
Read the worked answer
With QTc 520 ms, avoid haloperidol and other QT-prolonging antipsychotics as first-line. Recommended approach:
- Benzodiazepines are the treatment of choice for alcohol withdrawal and are relatively QT-neutral. Use a symptom-triggered regimen (e.g., diazepam or lorazepam titrated to a validated scale such as CIWA-Ar), with lorazepam preferred if there is hepatic impairment or if the patient cannot be monitored closely.
- Thiamine before or with glucose to prevent Wernicke encephalopathy; give parenteral thiamine in a malnourished chronic drinker.
- Correct magnesium and potassium - hypomagnesemia both perpetuates hypokalemia and predisposes to torsades; repletion is part of treating the QTc.
- Fluid and temperature management, and treat any intercurrent infection.
- If a toxidrome is confirmed (e.g., anticholinergic), physostigmine is considered only with cardiology/tox guidance and never with a widened QRS; for sympathomimetic toxicity, benzodiazepines plus supportive care are first-line, and avoid antipsychotics.
- Environmental measures: quiet, low-stimulation room, continuous cardiac monitoring, frequent reassessment.
Justification: the dominant reversible driver is withdrawal/toxicity, both of which respond to GABAergic sedation with benzodiazepines rather than dopamine blockade. Antipsychotics treat agitation symptomatically but do not treat withdrawal and add arrhythmia risk in a patient with an already prolonged QTc and possible electrolyte derangement.
3. Twelve hours later he is calm but confused and says he 'sees bugs on the wall.' How does this change your formulation, and what is your next step?
Read the worked answer
Persistent confusion with visual hallucinations after the autonomic storm settles shifts the differential toward an organic psychotic state rather than a primary psychotic disorder. Key considerations:
- Alcohol withdrawal delirium can persist for days and is the most parsimonious explanation given the timeline, but it should be improving with adequate benzodiazepine coverage.
- Wernicke encephalopathy must be considered - the triad is confusion, ataxia, and ophthalmoplegia, but the full triad is often incomplete. Give parenteral thiamine if not already done; do not wait for the triad.
- Hepatic encephalopathy in a chronic drinker - check ammonia, liver function, and look for asterixis.
- Anticholinergic or other drug effect lingering, or a co-ingested substance.
- Delirium from infection, hypoxia, or metabolic derangement - recheck vitals, glucose, electrolytes, and screen for urinary or respiratory infection.
Next step: do not label this as schizophrenia or a primary psychosis. Reassess with a full delirium workup (vitals, glucose, electrolytes, LFTs, ammonia, urinalysis, oxygen saturation, medication review), continue monitored benzodiazepine taper, ensure thiamine, and reorient the patient. Visual hallucinations with fluctuating confusion in a medically ill patient are a delirium signature, and antipsychotics should be used only cautiously and briefly for distressing symptoms once QTc and causes are addressed.