The statistical relationship between hypercholesteremia and the incidence and mortality rate of coronary artery disease has long been recognized. Epidemiologic studies have shown that hypercholesteremia is one of several factors (such as obesity, hypertension, cigarette smoking and lack of physical exercise) with an important bearing on the occurrence of myocardial infarction and death from coronary atherosclerosis.It has not yet been established that reduction of the serum cholesterol level will retard or prevent the development of new atheromatous plaques or will lead to disappearance of plaques which have already formed. Nevertheless, informed opinion has swung more and more to the view that it is better to have a normal serum cholesterol level than an elevated one, and that some form of therapy is advisable if hypercholesteremia exists. A doctor interested in preventing heart attacks in his patients must take note of the etiologic factors and decide what to recommend regarding each of them. A decision to do nothing about any of the factors might be dficult to justify on the basis of available statistics. This is especially true if two or more of these abnormalities are present, placing the patient in a high-risk group.To reduce the serum cholesterol level, one may employ dietary measures, chemotherapy, or both. It is important, however, to distinguish between treatment of obesity and treatment of hypercholesteremia. If a patient is overweight, no one would dispute the fact that caloric restriction (which implies restriction of fats) is necessary for weight reduction. Our obese, hypercholesteremic patients are told that first the body weight should be brought to normal by caloric restriction, after which the serum cholesterol level may be normal. If it is still elevated while the patient is receiving a diet which maintains ideal weight, chemotherapy is offered. If weight is not excessive in a hypercholesteremic subject, either dietary manipulation or chemotherapy may be utilized. Many dietary programs for hypercholesteremia which are highly effective in a metabolic unit or under the circumstances of a short-term study in a research center are neither practical for home use nor acceptable to the patient as a way of life. As in diabetes, in hypercholesteremia there may be a metabolic defect which is too severe to permit normalization of the blood levels by dietary measures-or at least by any which are practical for use at home. For this reason,