We conclude that in clinically unchanged conditions, the time-domain HR variability measures derived from 24-hour ambulatory recordings of AMI survivors are stable during the early convalescent phase, and the day-to-day differences have only little effect on the result. The only exception appears to be the pNN50 parameter, the use of which should be favorably substituted by the rMSSD measurement. Geometric estimates of HR variability are highly and consistently correlated with statistical measures of overall HR variability, and may be used as substitutes for each other.