The heart effects of antidepressants may depend as much on the patient as the drug

A major analysis of randomized trials tracked changes in heart rhythm across 10 antidepressants and examined how age, baseline QTc, potassium, body mass index, and kidney function shaped treatment effects.

Study: QTc changes and early major adverse cardiovascular events associated with antidepressant treatment for major depressive disorder in adults: individual participant data network meta-regression of double blind randomised trials. Image Credit: Photoroyalty / Shutterstock

In a recent study published in the British Medical Journal, a group of researchers compared antidepressant-specific effects on the heart rate-corrected QT interval (QTc) and examined associations between antidepressant treatment and early major cardiovascular adverse events (MACE) in adults with major depressive disorder.

Background

Antidepressants are widely used for depression, but could a treatment that improves mood also affect the heart’s electrical activity?

In England, 8.89 million people received an antidepressant during 2024-2025, while antidepressant use has increased worldwide. QTc prolongation reflects delayed ventricular repolarization and may provide an early warning of proarrhythmic states, but it does not necessarily lead to major cardiovascular events.

Concerns have focused on certain antidepressants and patient characteristics such as age, baseline QTc, potassium, body mass index, and kidney function. Further research is needed to clarify how these factors interact with individual antidepressants.

About the study

Researchers updated a previous systematic-review search to 18 March 2026 and included published and unpublished double-blind randomized controlled trials of adults aged 18 years or older with major depressive disorder.

Eligible trials compared antidepressants as monotherapy with each other or placebo. At least two trained researchers independently screened records and full texts, with disagreements resolved through discussion or a senior researcher.

For QTc analyses, researchers obtained individual participant data (IPD) from 130 trials involving 40,013 participants, of which 35 trials involving 8,679 participants contained individual-level QTc data and contributed to the primary QTc analysis. They also examined trial documents to extract QTc measurements and prespecified covariates.

The primary outcome was the change in QTc Fridericia (QTcF) at approximately 4 weeks, with secondary outcomes being QTcF above 500 milliseconds and increases of at least 60 milliseconds.

Analyses considered age, sex assigned at birth, baseline QTc, serum potassium, body mass index, and estimated glomerular filtration rate (eGFR). Two-stage network meta-regression used a multivariate mixed-effects Bayesian model, with sensitivity analyses using the QTc Bazett (QTcB) correction.

Early MACE were examined in the IPD, while non-suicidal sudden death was assessed using aggregate trial data. Pooled risk differences with 95% credibility intervals (CrIs) and prediction intervals (PIs) were calculated.

Missing pretreatment covariates were multiply imputed as needed, and extraction inconsistencies were resolved independently through consensus or senior review.

Study results

Of 524 eligible randomized controlled trials involving 116,791 participants, IPD were obtained from 130 trials. Thirty-five trials provided QTc data for 8,679 participants; 33 trials were rated as low overall risk of bias and 2 as moderate.

The participants had a mean age of 47 years; 63.0% were female; and 10 antidepressants were used: amitriptyline, bupropion, duloxetine, escitalopram, fluoxetine, mirtazapine, paroxetine, trazodone, venlafaxine, and vortioxetine. Baseline mean QTcF was 403 milliseconds, body mass index was 26, serum potassium was 4.3 mmol/L, and eGFR was 96 mL/min/1.73 m².

When treatment effects were considered independently of other covariates, mirtazapine was associated with a 27.68-Millisecond shortening of QTcF compared with placebo, whereas fluoxetine showed the largest estimated QTcF prolongation, though this estimate was not statistically significant.

After adjusting for mean age, sex, baseline QTcF, body mass index, potassium, and eGFR, amitriptyline and escitalopram were associated with significant increases in mean QTcF of 5.3 and 8.7 milliseconds, respectively.

Bupropion, duloxetine, paroxetine, venlafaxine, and vortioxetine showed smaller changes in QTcF than amitriptyline and escitalopram in head-to-head comparisons. Similar patterns were observed using the QTcB formula.

In age-specific model estimates, mirtazapine and amitriptyline significantly increased QTcF at younger ages compared with placebo, whereas fluoxetine and escitalopram significantly increased QTcF at older ages.

Across 216 combinations of age, sex, baseline QTcF, body mass index, and serum potassium, with eGFR fixed at 96 mL/min/1.73 m², escitalopram increased QTcF in 94.0% of combinations, trazodone in 91.7%, and amitriptyline in 85.7%.

Escitalopram, amitriptyline, and fluoxetine were most often associated with the highest QTcF prolongations. Among antidepressants that were never associated with the highest QTcF increase, venlafaxine, duloxetine, and vortioxetine most often showed QTcF decreases across the modeled profiles.

Average effects were generally small, but some modeled QTc changes across individual profiles exceeded the 30-millisecond threshold the European Medicines Agency regards as a level of potential clinical interest.

The planned inferential analyses for QTcF above 500 milliseconds and increases of at least 60 milliseconds could not be performed due to limited IPD.

Seven participants had QTcF above 500 milliseconds,onds and 56 had increases above 60 milliseconds. No participant in the IPD sample experienced torsade de pointes or sudden non-suicidal death, while 10 experienced syncope. In aggregate data from 139 trials involving 52,398 participants, 12 sudden non-suicidal deaths occurred: nine among 34,575 antidepressant-treated participants and three among 17,823 placebo-treated participants.

The risk difference was 0.01%, with a 95% CrI of −0.01% to 0.02%, and no class-specific effect was detected for the antidepressant classes examined.

Conclusions

The study found that antidepressants differed in their effects on the corrected QT interval, and that these effects varied according to individual characteristics including age, baseline QTc, potassium level, body mass index, and estimated glomerular filtration rate. Small average increases in QT were observed with amitriptyline and escitalopram after adjustment for covariates.

The trial data did not detect evidence of increased early MACE or non-suicidal sudden death with antidepressants compared with placebo, but the analyses were underpowered for rare events, and the short trial duration means uncommon or later-occurring MACE could have been missed. The findings support considering individual risk factors during treatment selection and shared decision-making.

The authors also cautioned that trial participants may not fully represent routine clinical populations; that escitalopram was included in the QTc analysis; and that escitalopram data came from only 186 participants. 

Real-world data are needed to define these risks more precisely, particularly in patients with comorbidities or concurrent QT-prolonging medications. 

The study received NIHR funding and a grant from Angelini Pharma; the paper states that the funder and sponsor had no role in study design, data collection, analysis, interpretation, or manuscript preparation. Several authors also reported industry research or consultancy relationships.