Spontaneous variability of complex ventricular arrhythmias detected by long-term electrocardiographic recording.

Variations in the frequency of complex ventricular arrhythmias were evaluated by consecutive 24-hour long-term electrocardiographic recordings over 4 days using a two-channel recorder and computer-assisted analysis system with a weighted relative mean error of 7.5 +/ 5% (SD). Twenty patients (mean age 58 +/- 9 years [SD] with various cardiac disorders were selected if they had a daily average of more than 30 ventricular ectopic complexes per hour. Twenty patients had ventricular couplets and 14 patients had ventricular tachycardia (at least triplets). The mean daily number of either couplets or ventricular tachycardia was subjected a four-factor nested analysis of variance to determine the sources of variation in ectopic frequency. Differences in hourly rates accounted for 30% of the variation in the frequency of ectopic complexes. Statistical methods were applied to determine the change in frequency of complex arrhythmias necessary to exceed that attributable to spontaneous variation alone at the p less than 0.05 level. Using a typical protocol, for example, comparing a 24-hour test period with a 24-hour control period would require a 65% decrease in mean hourly frequency of ventricular tachycardia and 75% reduction in the frequency of couplets to demonstrate therapeutic efficacy rather than a reduction due to spontaneous variation alone.