Escitalopram and amitriptyline lengthened QTc most, but early cardiac events did not rise

Design

Individual participant data network meta-regression and aggregate meta-analysis of double-blind RCTs

Population

8,679 adults with major depression (QTc); 52,398 across 139 trials (cardiovascular events)

Primary outcome

Change in QTc (Fridericia) over the first 8 weeks

Effect

Escitalopram +8.7 ms (CrI 1.6–15.8), amitriptyline +5.3 ms (0.8–9.8); early MACE risk difference 0.01%

A BMJ analysis pooled individual participant data from 35 double-blind randomised trials (8,679 adults with major depression) to compare the effect of ten antidepressants on QTc in the first eight weeks, adjusting for baseline QTc, age, sex, body mass index, potassium and kidney function. A separate aggregate analysis of 139 trials (52,398 participants) looked at early major cardiovascular events.

Against placebo, escitalopram lengthened QTc by about 8.7 ms (95% credibility interval 1.6 to 15.8) and amitriptyline by 5.3 ms (0.8 to 9.8). Once individual risk factors were modelled, these two were most often the agents with the largest prolongation, and venlafaxine and vortioxetine the smallest. Effects varied considerably between patients, especially for escitalopram, fluoxetine and mirtazapine. Early major cardiovascular events and non-suicidal death were not more common on antidepressants (risk difference 0.01%).

The average changes are small, and most patients will never be at risk from them. The value lies in the person who already has a long QTc, low potassium, kidney impairment or another QT-prolonging drug, where the choice of antidepressant can be made with the ECG in mind.

• Check a baseline ECG before escitalopram or amitriptyline in patients with heart disease, a long QTc or other QT-prolonging drugs.
• Correct low potassium and magnesium before starting an antidepressant in a patient at QT risk.
• Consider venlafaxine or vortioxetine where QT prolongation is a particular concern.
• Do not withhold antidepressants from most patients on cardiac grounds; early cardiac events did not rise.

Why it matters

It gives a way to choose between antidepressants for the patient whose QT is already a worry.

Don't overread it

Eight-week trial data cannot rule out rare arrhythmias or longer-term cardiac harm.

The statistics, in plain English

A credibility interval is the Bayesian equivalent of a confidence interval. For escitalopram it runs from 1.6 to 15.8 ms, so the true average could be trivial or meaningful. Wide variability means some individuals had much larger changes than the average.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.