Spontaneous variability in the frequency of ventricular premature complexes over prolonged intervals and implications for antiarrhythmic treatment.

Spontaneous variability in the frequency of premature ventricular complexes (VPCs) during ambulatory electrocardiography was examined over short (2- to 14-day) and long (6- to 12-month) periods in 20 patients. The coefficient of variation of the mean daily VPC frequency was significantly lower in the short-term periods than the long-term periods. The 95% confidence limit for spontaneous variation in mean daily arrhythmia frequencies was estimated with a linear regression method comparing two 1-day electrocardiographic recordings. The relative reduction in VPCs reaching the confidence limit calculated from the short-term periods was 69% at the beginning and 73% at the end of the 6- to 12-month follow-up. Corresponding values for repetitive VPCs were 78% and 77%. In comparing 2 recordings separated by the long interval, the reductions required were 98 to 100% for all VPCs, and even a 100% reduction was not significant for repetitive VPCs. The number of VPCs fell spontaneously below the initial confidence limit in 8 of 20 patients in later assessments. Thus, apparently positive individual responses are achieved in long-term treatment studies if stringent efficacy criteria are not followed. The variability between recordings over short periods remains fairly constant when reassessed after a long interval, even if the frequency of arrhythmias changes over the long interval. This permits application of conventional percentage limits in evaluation of therapeutic effects if the natural extent of the arrhythmias is periodically reassessed.