Quinidine-related mortality in the short-to-medium-term treatment of ventricular arrhythmias. A meta-analysis

The interim results of the Cardiac Arrhythmia Suppression Trial requires physicians to use a higher threshold for employing antiarrhythmic agents in the treatment of benign or potentially lethal ventricular arrhythmias. Many have managed patients by switching to the traditional class I quinidine despite its known proarrhythmic tendency.

To evaluate the relation between quinidine therapy and mortality in patients with benign or potentially lethal ventricular arrhythmias, we performed a meta-analysis on four randomized double-blind active controlled parallel trials evaluating 1,009 patients in which quinidine (n = 502) was compared to flecainide (n = 141), mexiletine (n = 246), tocainide (n = 67), and propafenone (n = 53). All four trials had similar patient selection, protocols, and methodology (e.g., placebo lead-in and Holter monitoring) but varying lengths of drug exposure. A total of 12 deaths were reported on quinidine and four deaths on the other drugs: two on mexiletine, one on flecainide, and one on tocainide. The statistical analysis of the mortality rates was based on techniques for combining data across separate strata. Based on maximum likelihood estimation, the combined risk of dying on quinidine was statistically significantly higher compared to the other four drugs with a risk difference of 1.6%. The 95% confidence interval was 0-3.1% (p = 0.05). The likelihood ratio test for uniformity of the risk difference across strata showed the trials to be homogeneous (p = 0.88). There was one death recorded for the placebo lead-in period (2 weeks' exposure for 624 patients and 1 week for 385 patients), and seven deaths were reported within 2 weeks on active drug treatment--six on quinidine and one on mexiletine. Furthermore, proarrhythmia was reported in 20 patients on quinidine versus 11 patients on the four other drugs (p = 0.09).

These data suggest that quinidine may have an adverse effect on mortality as compared to other class I antiarrhythmic agents and that individualized patient selection for the use of this agent be carefully weighed relative to its potential for harm and benefit.