Declining Incidence of Ventricular Fibrillation in Myocardial Infarction

Background. The purposes of the present investigation were 1) to track the incidence of primary ventricular fibrillation (VF) in the control and lidocaine-treated groups in the randomized control trials (RCTs) of lidocaine prophylaxis against primary VF in acute myocardial infarction, with particular emphasis on the time frame of the randomized trial, and 2) to estimate the number of patients who must receive lidocaine currently to prevent one episode of VF. Methods and Results. The following variables from RCTs published between 1969 and 1988 were entered into logistic regression models to predict the percent of patients developing VF: year of publication of the RCT, method of data analysis used in the RCT, route and technique of lidocaine administration, duration of monitoring for VF, and exclusion criteria before randomization (congestive heart failure/cardiogenic shock, ventricular tachycardia/VF, or bradycardia/atrioventricular block). Year of publication was a significant predictor ofVF in both the control and lidocaine groups (p<0.002) even after adjusting for other covariates. Based on a univariate logistic regression model with year as the predictor variable, it was estimated that the incidence of primary VF in the control group fell from 4.51% in 1970 to 0.35% in 1990 and from 4.32% down to 0.11% for the lidocaine group over the same time period. Thus, about 400 patients would currently need prophylaxis with lidocaine to prevent one episode of VF. Conclusions. Present estimates of the risk: benefit ratio of lidocaine prophylaxis should consider the low risk ofVF in control patients and the large number who need lidocaine prophylaxis to prevent one episode ofVF. When added to the previously reported trend toward excess mortality in lidocaine-treated patients, these data argue against the routine prophylactic use of lidocaine in patients with acute myocardial infarction. (Circulation 1992;86:764-773)