Risk Factors for Tachycardia Events Caused by Antiarrhythmic Drugs: Experience From the ESVEM Trial

Abstract

Background: In the Electrophysiology Study versus Electrocardiographic Monitoring (ESVEM) trial, up to seven antiarrhythmic drugs were randomly assigned to 486 patients with a history of sustained ventricular arrhythmia. At baseline, all the patients had inducible sustained ventricular tachycardia (VT) and had ≥10 premature ventricular beats (PVBs) per hour on 48-hour Holter monitoring. A total of 1,229 drug trials were performed. Antiar rhythmic drugs were discontinued during hospitalization because of ventricular tachy arrhythmias thought to be a proarrhythmic effect of the antiarrhythmic drugs in 96 of 479 patients (20%) who received drugs. Proarrhythmic effects were defined as sustained VT, ventricular fibrillation or arrhythmic death, torsade de pointes, or distinct intolerable wors ening of the baseline arrhythmia after at least three doses of the drug.

Methods and Results: Eighteen baseline characteristics were analyzed for factors that would predict a higher incidence of proarrhythmia. These included type of heart disease, previous myocardial infarction, symptom activity scale, gender, type of arrhythmia, VT/ ventricular fibrillation, age, left ventricular ejection fraction (LVEF), PVB frequency, heart rate, QRS duration, and QT interval. Multiple logistic regression analysis identified increased mean PVB frequency (P = .003) and increased heart rate (P = .026) as significant predic tors of proarrhythmia. Decreased LVEF (<25%) exhibited only a trend toward significance ( P = .073). When proarrhythmia was redefined as sustained VT, cardiac arrest or arrhyth mic death, or torsade de pointes (n = 59), PVB frequency (P = .003) and heart rate (P = .034) were still the only significant baseline predictors.

Conclusion: In the ESVEM study, higher PVB frequency and higher heart rate were signif icant predictors of drug-induced proarrhythmia.