Based on potential risk of torsade de pointes (TdeP; a rare arrhythmia), regulatory agencies have issue warning for two specific antidepressants. This has translated in clinic as a class effect in that every antidepressant monography or guideline warns against this side effect. Current data suggest that excessive caution in the face of this undesirable effect could have a deleterious effect on mortality. Most research on antidepressant/antipsychotic drugs and TdeP is based on its intermediate marker, the corrected QT interval (QTc) on the electrocardiogram, or case reports.

Our objective is to measure the contribution of psychotropic drugs (antidepressants and antipsychotics) to the arrhythmia itself, and measure its weight among all other risk factors.

We completed a retrospective case-control study at the Montreal Heart Institute, with a 1:3 ratio (n=440). We performed hierarchical logistic regression for TdeP (cases vs controls), and included the following independent variables: sex, age, hypokalemia, acute myocardial infarction, left ventricular dysfunction, hepatic failure and/or renal failure, other QTc-prolonging drugs, and psychotropic drugs. We then calculated population attributable risks (PAR).

In our final model which adjust risk factors for one another, women, acute myocardial infarction, hypokalemia, left ventricular dysfunction, QTc-prolonging drugs, and use of ≥2 antidepressants were significantly associated with TdeP. Age, hepatic and/or renal failure and antidepressants/antipsychotic drug monotherapy were not. The PAR for use of ≥2 antidepressants was 6.3%, while those of other QTc-prolonging drugs, female sex, and left ventricular dysfunction were 55.1%, 41.8%, and 25.6%, respectively.

When adjusting for concomitant risk factors, monotherapy with antidepressant or antipsychotic drugs is not associated with TdeP. On its side, the use of ≥2 antidepressants is associated with TdeP, with a PAR of lesser magnitude than sex, left ventricular dysfunction, and other QTc-prolonging drugs. This research provides a more nuanced perspective on the relationship between psychotropic drugs and the occurrence of this arrhythmia.

None Declared