Insights into the Electrophysiology Study Versus Electrocardiographic Monitoring Trial: its programmed stimulation protocol may introduce bias when assessing long-term antiarrhythmic drug therapy
LA Biblo,
MD Carlson,
and
AL Waldo
Department of Medicine, Case Western Reserve University/University Hospitals of Cleveland, Ohio, USA.
OBJECTIVES. We hypothesized that if the Electrophysiology Study Versus Electrocardiographic Monitoring (ESVEM) trial programmed stimulation protocol misclassified some drug trials as effective, then the misclassification rate would be proportionally greater for drugs other than sotalol. BACKGROUND. In the ESVEM trial, patients treated with sotalol had fewer arrhythmic recurrences than those treated with other antiarrhythmic drugs despite similar efficacy predictions during electrophysiologic testing. METHODS. We retrospectively compared the standard programmed stimulation protocol used at Case Western Reserve University, which used three extrastimuli during all follow-up studies, with the ESVEM protocol in 176 antiarrhythmic drug trials: sotalol (n = 54), procainamide (n = 73) and quinidine/mexiletine (n = 49). RESULTS. Predictions of efficacy were higher in the sotalol trials (14 of 54 standard, 20 of 54 ESVEM) than in procainamide trials (7 of 73 standard, 14 of 73 ESVEM) or quinidine/mexiletine trials (1 of 49 standard, 7 of 49 ESVEM). Thus, the two protocols classified 19 of 176 trials differently: not effective by the standard protocol but effective by the ESVEM trial. Discordant predictions of drug efficacy constituted a smaller proportion of ESVEM protocol efficacy predictions for sotalol (6 [30%] of 20) than for the other drugs (13 [62%] of 21, p < or = 0.05). CONCLUSIONS. In the present study, the ESVEM programmed stimulation protocol predicted efficacy more often than the standard protocol. Discordant predictions represented a smaller portion of efficacy predictions for sotalol than for the other drugs. Thus, in the ESVEM trial, the superior long-term follow-up observed in patients assigned to sotalol may have been an artifact of the stimulation protocol utilized by the ESVEM investigators.
