Read this topic

Clinical Psychopharmacology · Paper A

QTc prolongation with psychotropics

The idea

QTc management is not one action — it is a ladder, and the exam tests which rung the vignette is standing on. First ask whether a reversible cause needs fixing. Then ask whether the QT (short for the heart's Q-T interval — the part of the heartbeat this whole family is about) or its rise from baseline has crossed a real threshold. For example, a mild unexplained prolongation calls only for a dose cut or a switch of agent, while a threshold-crossing reading calls for stopping the drug and starting continuous monitoring. Finally, ask whether a symptom shows the heart has already been unstable — in practice, that single detail is what turns a monitored case into an urgent cardiology referral.

How it works

A prolonged QT interval (the part of the heartbeat that measures the heart's electrical reset) reflects delayed cardiac repolarisation. Repolarisation is most vulnerable right as the next beat is due, which is when a torsades-triggering early beat — an 'R-on-T' event — can land.
Hypokalaemia and hypomagnesaemia each independently slow repolarisation further, so they add to any drug-induced effect and are the most common reversible drivers seen on a psychiatric ward (diuretics, poor intake, vomiting, diarrhoea).
Drug risk is a spectrum, not a yes/no. Repeated intravenous haloperidol and high-dose typical antipsychotics sit at the highest-risk end. Citalopram and escitalopram are dose-capped specifically because of QTc risk — citalopram 40 mg/day, 20 mg/day over 65 or with hepatic impairment. Aripiprazole sits at the lower-risk end among antipsychotics.
Two numbers do the escalation workan absolute QTc of roughly 500 ms or more, or a rise of more than about 60 ms from the patient's own documented baseline. Each independently marks the threshold at which the drug is stopped rather than merely reduced.
Torsades de pointes is the named arrhythmia this whole pathway exists to prevent — a self-terminating or sustained polymorphic ventricular tachycardia that can cause syncope or sudden cardiac death. Its specific emergency treatment, intravenous magnesium sulfate, is given even when the measured magnesium level is normal.

How to spot it

A vignette describing a psychotropic patient with hypokalaemia or hypomagnesaemia (diuretics, poor intake, vomiting, diarrhoea) and only a mild-to-moderate QTc rise, with no symptoms, is pointing you to correcting the electrolytes first.
A vignette describing repeated intravenous haloperidol dosing, or any QTc at or beyond roughly 500 ms, or a rise beyond roughly 60 ms from a stated baseline, with no symptoms yet, is pointing you to stopping the drug and starting continuous cardiac monitoring.
A vignette describing a captured run of polymorphic ventricular tachycardia (torsades), with or without dizziness, is pointing you to intravenous magnesium sulfate as the immediate treatment.
A vignette describing a markedly prolonged QTc PLUS syncope, presyncope, palpitations or a 'heart skipping' sensation is pointing you to urgent same-day cardiology discussion, not ward-level monitoring alone.
A vignette describing a patient about to START a QTc-risk drug, with risk factors but no ECG yet done and no current abnormality, is pointing you to a baseline ECG with no other action needed.
A vignette describing REPEATED borderline QTc readings on the same antipsychotic despite dose reduction and normal electrolytes, in a patient who needs ongoing treatment, is pointing you to a switch to a lower-risk agent.

Tell it apart

Reversible-cause case vs high-risk-exposure case: hypokalaemia/hypomagnesaemia with a mild QTc rise is corrected first, whereas a QTc at or beyond ~500 ms (or a rise beyond ~60 ms) is stopped-and-monitored regardless of whether electrolytes are also abnormal.
Stop-and-monitor vs urgent cardiologystopping the drug and starting a monitor is the ward-level response to a high threshold ALONE, whereas urgent same-day cardiology is reserved for a high threshold PLUS a symptom (syncope, presyncope, palpitations) suggesting an arrhythmia has already fired.
IV magnesium vs simple electrolyte correction: IV magnesium sulfate is the specific treatment for a confirmed arrhythmia (torsades) itself, unlike routine electrolyte replacement for a low level with no arrhythmia, which does not need an emergency bolus.
Dose reduction vs drug switchdose reduction plus a repeat ECG suits a single borderline reading with no reversible cause, whereas a drug switch is reserved for REPEATED borderline readings on the same agent despite dose reduction and normal electrolytes.
Baseline ECG vs any active management stepa patient who has not yet started the drug gets a baseline check only, unlike every other option here, which assumes the drug is already running and an abnormality already exists.
Absolute value vs deltaa QTc under 500 ms can still be high risk if it has RISEN by more than ~60 ms from the patient's own baseline — the delta from baseline, not just the raw number, can independently trigger stopping the drug.

Watch for

⚠️ Treating 'the QTc is prolonged' as needing the same single response every time — the correct action depends on severity, cause and symptoms, not on the mere presence of prolongation.
Reaching for stopping the drug as the reflex 'safe' answer even when the vignette hands you a clean reversible cause (hypokalaemia/hypomagnesaemia) and a QTc well under threshold — correcting the cause comes first.
Forgetting that a RISE from baseline (>~60 ms) can be dangerous even when the absolute QTc is still under ~500 ms — checking only the raw number and ignoring the patient's own baseline misses this.
Giving intravenous magnesium sulfate for a merely low magnesium level with a prolonged QTc but no arrhythmia — that dose is for torsades itself, not for correcting a lab value.
Assuming stopping the drug and starting a monitor is always enough — once there is a symptom suggesting an arrhythmia has already occurred, ward-level monitoring alone is not sufficient and same-day cardiology is needed.
Switching antipsychotic at the first sign of any QTc prolongation, rather than reserving a switch for genuinely repeated borderline readings after dose reduction and electrolyte correction have already been tried.

Use it again

First askhas the drug even been started yet? If not, and there is no current abnormality, the answer is a baseline ECG, not an active management step.
If the QTc is prolonged, next askis there an obvious, reversible metabolic cause (low potassium or magnesium)? If yes, and the QTc is not at a red-flag threshold, correct that first before touching the drug.
If there is no reversible cause, or the QTc/exposure has crossed a real threshold (~500 ms absolute, or a ~60 ms rise from baseline), the drug itself is stopped and the patient is monitored — do not settle for a dose reduction once that line is crossed.
If a symptom already suggests an arrhythmia (syncope, presyncope, palpitations) or one has been captured on the monitor, escalate beyond ward-level actions: torsades itself gets IV magnesium, and any symptomatic markedly-prolonged QTc gets urgent same-day cardiology.
Only reach for a drug switch when the same problem (borderline QTc on the same agent) has recurred despite the earlier steps — a switch is a repeated-failure response, not a first response.
Never let the word 'QTc' alone decide the answer — the vignette's specific numbers (absolute value, rise from baseline) and specific symptoms (or their absence) are what decide which rung of the ladder you are on.

In detail

Hook — a number on a printout that changes what happens next

Picture a Tuesday ward round. A routine comes back with one abnormal line: QTc 478 milliseconds (a corrected measure of how long the heart's electrical reset takes). Nothing hurts, nobody has fainted, and the patient feels fine. Two very different things could happen next — correct a blood test and carry on, or stop the psychiatric medicine that took months to work — and which one is right depends entirely on details most people skim past. This page builds the whole decision from zero, so a single abnormal number never gets treated as one fixed instruction again.

Plain definition

The QT interval is the part of an ECG (electrocardiogram, a tracing of the heart's electrical activity) that measures how long it takes the heart's ventricles (the two main pumping chambers) to reset electrically after each beat, a process called repolarisation. Because that reset naturally takes longer when the heart beats slowly and shorter when it beats fast, doctors correct the raw QT for heart rate using a formula (Bazett's correction, the standard one used in practice) to get the QTc — the 'c' stands for corrected. A normal QTc sits below about 440 milliseconds in men and 460 milliseconds in women; many psychiatric drugs, and several everyday factors, can stretch it beyond that.

The critical caveat — the number alone never decides what happens next

Here is the trap almost every learner falls into: treating 'the QTc is prolonged' as if it were itself an instruction. It isn't. The same reading of 478 milliseconds calls for a completely different response depending on why it happened, how far it has risen from the patient's own starting point, and whether anything has actually happened to the heart's rhythm yet. A rise of 60 milliseconds or more from a documented baseline is dangerous even if the absolute number is still under 500 — the direction of travel matters as much as the destination.

First principles — why timing, not just size, is what matters

Start from what the heart's electrical system is actually doing. Each heartbeat is a wave of electrical activity: first the ventricles contract (depolarisation, the contraction phase), then they have to reset before the next beat can safely arrive (repolarisation, the reset phase), largely by pumping potassium ions back out of the heart's muscle cells through dedicated potassium channels (tiny gateways in the cell membrane that let one type of charged particle through). Many drugs — including several antipsychotics and antidepressants — partially block those potassium channels, so repolarisation takes longer.

A longer interval means a longer window during which the heart is only partly reset, and if an early extra beat lands inside that vulnerable window, it can trigger a chaotic rhythm called torsades de pointes (French for 'twisting of the points', describing the rhythm strip's twisting appearance) — a fast, unstable heartbeat that can cause fainting or, at worst, cardiac arrest.

Structured walk-through — the same rise, read two different ways

Suppose a patient's QTc was 420 milliseconds before starting an antipsychotic, and it is 478 milliseconds today — a rise of 58 milliseconds. That is still under both red-flag numbers (an absolute QTc of about 500 milliseconds, and a rise of more than about 60 milliseconds), so if there is an obvious reversible cause such as low potassium or magnesium, correcting that cause comes first, before any change to the drug itself.

Now change one detail: the same patient's QTc has instead risen from 425 to 495 milliseconds — a rise of 70 milliseconds. The absolute number, 495, is still technically under 500, and a candidate reading only that number might reach for a dose reduction. But the rise has already crossed the 60-millisecond threshold on its own, and that threshold is exactly as serious as crossing 500 milliseconds outright — so the drug itself is stopped and the patient goes onto a continuous cardiac monitor, watched until the QTc recovers.

Look-alikes & differentials

Looks like a QTc emergency	How to tell it apart
A simple faint (vasovagal syncope)	Usually has a clear trigger (standing, pain, fear), a slow recovery with pallor and nausea, and a normal QTc
Anxiety-related palpitations	Racing heart with no collapse and no arrhythmia captured on the monitor; QTc is typically normal
A one-off borderline QTc with no rise from baseline	Not itself dangerous; only a genuine rise from the patient's own baseline, or an absolute value near 500, carries the real risk
Torsades de pointes	A captured run of polymorphic ventricular tachycardia (the rhythm strip actually shows it) — this is a true emergency, not a suspicion

Diagnosis — how a prolonged QTc is actually confirmed

There is no blood test for a prolonged QTc — it is read directly off a 12-lead , with the interval measured (usually in lead II or V5) and corrected for heart rate using Bazett's formula to give the QTc. Because measurement can vary slightly between beats and leads, an abnormal-looking reading is usually confirmed by repeating the ECG and, wherever possible, comparing it against the patient's own previous ECG rather than against the population-average cut-off alone — which is exactly why a stated baseline value in a vignette is never a throwaway detail.

Management — what is actually done, and whose guidance it rests on

Maudsley Prescribing Guidelines in Psychiatry sets out the escalation used throughout this page: correct a reversible cause — hypokalaemia (a low blood potassium level) or hypomagnesaemia (a low blood magnesium level) — first if one is present; reduce the dose or switch to a lower-risk agent such as aripiprazole for a persistent borderline reading with no reversible cause; and stop the causative drug, starting continuous cardiac monitoring, once the QTc reaches roughly 500 milliseconds or has risen by more than roughly 60 milliseconds from baseline.

Torsades de pointes itself is treated with intravenous magnesium sulfate, given regardless of the measured magnesium level, because it stabilises the heart's electrical activity directly rather than simply topping up a low level. The MHRA (the UK's medicines and healthcare regulator) capped citalopram, in a 2011 Drug Safety Update, at 40 milligrams a day — 20 milligrams over 65, or with liver impairment — specifically because of this dose-dependent risk.

That same update recommended a baseline before starting citalopram in anyone with cardiac risk factors, which is why a patient who has not yet started the drug gets an ECG rather than a pre-emptive dose change. Once a symptom shows the heart has already been unstable, ward-level monitoring on its own is not enough, and same-day cardiology involvement joins the plan.

Exam pearl

The exam's favourite trap is to hand you a number just under a threshold — a QTc of 495, say — banking on you checking only the absolute value and missing that the rise from a stated baseline has already crossed 60 milliseconds. Before you move on, try explaining to a friend why two patients with the exact same QTc reading today could need two completely different responses.

In one line

A prolonged QTc is managed by asking, in order, whether there is a reversible cause, whether the QTc or its rise from baseline has crossed a real threshold, and whether a symptom shows the heart has already been unstable — never by reacting to the number alone.