[High spontaneous variability of ventricular arrhythmias limits the evidence gained by long-term ECG studies].

The spontaneous variability of ventricular arrhythmias was investigated in 8 patients with chronic coronary heart disease. In each patient 6 eight-hour long-term ambulatory ECG recordings were made. During the first three periods patients were without medication and during the following three periods they received a placebo b.i.d. Hour to hour variability was high in all patients. When compared with the first hour of an eight-hour period the number of ventricular premature beats (VPB) per hour in four patients declined by more than 90%, in two patients by more than 70% and in only two by less than 50%. On the other hand, increments of several hundred percent were often observed. Day to day variability of VPBs was also high. When compared with the first documented eight-hour period VPBs either decreased or increased substantially (-84% to +940%). A spontaneous reduction of VPB's for several hours may simulate an antiarrhythmic drug effect. For the group as a whole placebo had a significant effect on the number of VPBs (p less than 0.001). Beside variations of quantity, the complexity of ventricular arrhythmias (Lown classification) spontaneously changed several classes in one patient. Therefore, high spontaneous variability renders it very difficult to estimate the behaviour of ventricular arrhythmias in the individual patient with the Holter-ECG.