The prognosis of heart failure patients is poor and as many as half of the deaths are sudden and thereby presumably attributable to arrhythmias. In the present study the effect of traditional therapy of mild heart failure with digoxin on arrhythmias was compared with the effect of xamoterol, a cardioselective beta 1 partial agonist, which has in addition beta-blocking properties at higher levels of sympathetic tone. Fifteen patients (NYHA class II-III) were included in the study. After a two-week baseline period they were randomized to digoxin or xamoterol for four weeks followed by a two-week washout and another four weeks of crossover therapy. Heart rate, blood pressure, and the number of complex ventricular premature beats remained essentially unchanged with digoxin. With xamoterol heart rate increased from 86 to 93 (ns) but was significantly higher during the night in comparison with digoxin. The number of ventricular premature beats decreased from 186 +/- 317 to 110 +/- 137 and increased to 130 +/- 175 after treatment. The number of runs decreased from 11 +/- 35 to 2.7 +/- 5 and increased to 5.6 +/- 9 after therapy. In conclusion, no significant effect of digoxin or xamoterol on ventricular arrhythmias was found. However, xamoterol showed a tendency to reduce simple and complex ventricular arrhythmias in patients with mild to moderate heart failure.