This report describes the design and methodological features of a double-masked, randomized, placebo-controlled trial to determine whether administration of the HMG CoA reductase inhibitor lovastatin retards the progression or facilitates the regression of coronary atherosclerosis. The study population consists of coronary patients with a recent arteriogram that characterizes them as being at high risk for coronary progression and a baseline fasting total serum cholesterol > or = 5.7 mmol/L and < or = 7.8 mmol/L. Lovastatin, or matching placebo, are up-titrated from 20 mg to 40 mg to 80 mg/day during the first 16 weeks of the study in an attempt to attain a fasting serum LDL cholesterol level of 3.4 mmol/L; patients and study personnel remain masked as to cholesterol levels throughout the trial. Coronary arteriography is repeated at 2 years, or earlier if necessitated by worsening symptoms, and all segments are measured quantitatively using a computer-based system. The primary outcome of the trial is a comparison between the lovastatin and placebo groups for coronary change score, defined as the mean of the minimum lumen diameter changes for all lesions (follow-up minus baseline arteriogram) per patient. The advantages and limitations of coronary arteriographic trials and some of the issues related to outcome measurements are discussed. The question posed by this study is of clinical relevance because the consequences of progression of coronary disease, angina, myocardial infarction, and sudden cardiac death are leading causes of morbidity and mortality.