The proarrhythmic effects of antiarrhythmic drugs are complications which have been described over several decades but the mechanisms (reentry, increased automaticity, ectopic faci, induced repetitive activity, vagal or adrenergic triggers) and the predisposing factors (underlying cardiac disease, previous severe arrhythmia, metabolic disorders, ischaemia, etc...) have only recently been identified. The appreciation of their true frequency poses problems of methodology (mode of recruitment, therapeutic converse proof), of definitions and depends to a great extent on the methods of detection used. Their severity cannot be denied and has been demonstrated both in experience of isolated cases and in recent prospective studies, the conclusions of which must be interpreted critically. Proarrhythmic effects may be observed at atrial (vagal or sympathetic arrhythmias, 1/1 flutter, acceleration of atrial fibrillation in preexcitation syndromes), junctional (artificial unidirectional block created by the antiarrhythmic drug which may be very effective at higher dosages: biphasic effect) or ventricular (aggravation of ventricular extrasystoles, torsades de pointe, ventricular tachycardia/fibrillation) levels. It is curious that no antiarrhythmic drug seems to be statistically less exposed to this type of complication which may result from phenomena of toxicity or idiosyncrasy. Given the potential gravity measures must be taken to prevent this complication, by observing simple rules (respect of contraindication, use of progressive dosage regimens, avoidance of loading doses, elimination of predisposing factors and abstention from dangerous therapeutic associations) and by carefully following up high risk patients.