Heart rate variability (HRV) reflects autonomous activity that influences the heart. It has been shown that HRV is depressed during acute myocardial infarction (AMI) and that it recovers with time. Beta-blockers reduce mortality after AMI and changes in sympathico-vagal activity have been suggested to be of importance. Under certain animal experimental conditions, metoprolol has been reported to increase vagal tone more than atenolol, which could have clinical implications. The purpose of the present study was to compare the effects of atenolol and metoprolol treatments on HRV during 6 weeks after AMI and to follow the post MI changes in HRV in patients on betablockers. METHODS: In an open, randomised cross-over study, 28 patients were randomised to 3+3 weeks\' treatment with atenolol or metoprolol starting 2-5 days after AMI. Twenty-four hour Holter recordings were made before randomisation and after 3 and 6 weeks. HRV was analysed as HR, SDRR, SDANN, SD, rMSSD and pNN50 in the time domain and as coefficient of component variance (CCV) of HF and LF, and as LF/HF ratio in the frequency domain. RESULTS: The average daily dose in our study population was 106 mg of metoprolol and 54 mg of atenolol. There were trends toward lower heart rates daytime, lower LF/HF ratio daytime and higher rMSSD on atenolol compared to metoprolol. In the total group of 28 patients we found during the first 3 weeks, a significant increase of SDNN, SDANN (p<0.0001) and LF/HF ratio daytime and CCV-HF night-time (p<0.01). All differences and trends were unchanged between 3 and 6 weeks. CONCLUSIONS: There was no evidence of more increased vagal tone with metoprolol compared to atenolol as has been suggested from animal models. In patients also on chronic treatment with beta blockers, an increase of HRV was seen during the first weeks post MI.