Ventricular arrhythmias are not a short-term reproducible phenomenon--why long recording monitoring is necessary.

ABSTRACT

BACKGROUND: The variability of ventricular arrhythmias (VA) among different days of the week is not well detected by one-day Holter monitoring. AIMS: To evaluate whether there are differences in VA distribution pattern during long recording period. METHODS: The EKG was recorded for 14 h per day during 7 days by Holter system in 34 consecutive pat ventricular couplets and non-sustained ventricular tachycardia (NSVT) recording from patients provided graphic data. We applied the Hurst method (H Coefficient) which evaluates whether a repetitive phenomenon is random or not. When the H is >0.5 and <1 means it is not random and implies a long-term memory effect. Considering the arrhythmic variability, the data were also analyzed by repetitive ANOVA comparing incidence of arrhythmias among the days. RESULTS: Isolated PVCs and ventricular couplets during 98 h recording provided graphic of the occurrence. A trend of increasing and decreasing of arrhythmias was observed which looks erratic. The H coefficient, however, was significantly >0.5 for all patients. Repeated ANOVA showed statistic difference among days in 31 patients with isolated PVCs; in 26 with ventricular couplets and 19 with NSVT when analyzed per hour during week days (p < 0.05). CONCLUSION: PVCs, ventricular couplets and NSVT are not a random phenomenon. Our data suggest the occurrence of ventricular arrhythmias had no similarity among the days, making unlikely that a single Holter recording for 24h may capture this phenomenon.