The natural history of coronary artery disease has to be studied by comparing coronary angiograms of the same patient taken at different times. However, conclusions from repeated angiographic studies are fraught with substantial errors mainly because of: 1. patients selection, 2. variable time interval and 3. technical pitfalls. Despite this bias published interval studies demonstrate that coronary atherosclerosis predominantly is a progressive disease: after 2-3 years 50% to 60%, after 3-4 years 60% to 70% and after 5 years more than 80% of patients demonstrate progressive coronary artery disease at angiography. In addition, quantitative evaluation of coronary angiograms reveals that progression of coronary artery disease: 1. has a variable pattern and pace in each coronary artery and 2. predominantly involves initially normal coronary artery segments. From all clinical and angiographic parameters under scrutiny progressive coronary artery disease is significantly correlated to: abnormal lipid levels at the time of the first angiogram, a period of unstable angina pectoris, interval myocardial infarction and initial severity of coronary artery obstruction. It has to be emphasized, however, that in the individual patient the speed of progression is highly variably supporting the concept of different underlying pathophysiological mechanisms (primary/secondary progression). Regression or coronary stenosis is a rare phenomenon which may occur spontaneously and is anecdotally reported in patients after vigorous treatment of severe hyperlipoproteinemia.