Clinical factors predicting successful electrophysiologic-pharmacologic study in patients with ventricular...

Data from 142 patients who had sustained ventricular tachycardia or ventricular fibrillation were analyzed to determine if clinical variables predict response to antiarrhythmic drugs at electrophysiologic study. Effective antiarrhythmic drugs were identified for 43 patients (30%). Ten of 25 variables analyzed were univariate predictors of drug response at the probability (p) level of less than 0.05. Stepwise logistic regression identified three variables independently predictive of drug response: fewer coronary arteries with 70% or greater stenosis (p less than 0.001), female sex (p less than 0.002) and fewer episodes of arrhythmia (p less than 0.03). A function incorporating these three variables was constructed to predict the probability of drug response, and ranges of the predictor function corresponding to high, intermediate and low probabilities of drug response were identified. Response rates in the high (greater than 50%), intermediate and low (less than 10%) probability ranges were 28 (58%) of 48, 10 (27%) of 37 and 5 (9%) of 57, respectively. Thus 40% of the patients who had a less than 10% likelihood of benefit from electrophysiologic-pharmacologic study were classified into the low probability range. When the predictor function was applied prospectively to 25 additional patients, response rates in the three probability ranges were 3 (50%) of 6, 1 (12%) of 8 and 0 (0%) of 11. These data show that analysis of clinical variables can be used to estimate the probability of benefit from electrophysiologic-pharmacologic study.