We analyzed 24 h ECG recordings of 29 patients with frequent (> 2000/day) ventricular premature contractions (VPCs) to evaluate: 1) the relationship between VPC frequency and heart rate (HR); 2) the relationship between the coupling interval (CI) of VPCs and HR; 3) the time course of changes in VPC frequency following an abrupt and sustained increase in HR; and 4) the effect of oral diltiazem (90 or 180 mg/day for 4 weeks) on VPCs. Based on the patterns of relationship between VPC frequency and HR, patients were divided into 2 groups: 1) 17 patients with a positive correlation, the P group (a linear increase in VPCs with increasing HRs); and 2) 12 patients with a non-positive correlation, the NP group. The NP group showed either a linear decrease in VPCs with increased HRs (4 patients) or an increase in VPCs at low HRs and a decrease at high HRs (8 patients). In all cases, an increased HR was associated with a shortening of CI. In the P group, changes in VPCs after an abrupt increase in HR showed 2 types: 1) delayed-ascent type, in which VPCs increased as the duration of HR increase was prolonged (cumulative effects of heart beat) (7 patients); and non-delayed-ascent type (10 patients). Diltiazem reduced VPCs > or = 75% in all of the 7 delayed-ascent-type patients, compared with 3 of the 10 non-delayed-ascent-type patients (p < 0.05) and none of the 12 patients in the NP group (p < 0.001).(ABSTRACT TRUNCATED AT 250 WORDS)