BACKGROUND:

The Lipid Research Clinics Coronary Primary Prevention Trial and the Coronary Drug Project had shown that morbidity and mortality from ischemic heart disease were reduced by blood cholesterol-lowering therapy. Although blood cholesterol reduction ameliorated experimental atherosclerosis in animal models, the two largest human studies with angiographic observation of arterial lesion change, the NHLBI Type II Coronary Intervention Study and a study by Cohn et al, had not demonstrated significant treatment effects. Favorable, but inconclusive, treatment trends were observed in four unrandomized angiographic trials and one trial too small for evaluation of randomized groups.

The clinical trial was supported by a subproject within a program project grant.

DESIGN NARRATIVE:

CLAS-I was randomized and selectively-blinded. Screening for the trial consisted of five clinic visits, at which baseline data, including angiographic data, were obtained and a prerandomization trial of the study drugs was conducted. One hundred eighty-eighty subjects were randomized to either 30 grams (g) of colestipol hydrochloride plus 3 to 12 g of niacin daily or to placebo. Both groups received diet intervention. The drug group received less than 125 milligrams (mg) of cholesterol daily, 22 percent of energy as fat, 10 percent as polyunsaturated fat, and 4 percent as saturated fat. The placebo group received less than 250 mg of cholesterol per day, 26 percent of energy as fat, 10 percent as polyunsaturated fat, and 5 percent as saturated fat. The different diet composition for drug and placebo groups was to enhance the differential in blood cholesterol responses between the two groups. Study subjects and clinic staff were blinded to the prerandomization study drug trial lipid responses. Subjects were blinded to treatment assignments. Subjects and staff were not blinded to on-trial lipid values. The primary endpoint, the global change score, was change in atherosclerosis observed by angiography of native coronary arteries and aorta coronary bypass grafts. Evaluation of study end-points was performed by staff and consultants who were blinded to treatment group assignments, as well as to the temporal ordering of angiographic data. Subjects were seen monthly for the first six months and then at two-month intervals. A repeat angiogram was performed at two years. Of the 188 randomized subjects, 162 completed the study.

On completion of CLAS-I, participants not requiring further bypass surgery were invited to continue in CLAS-II for an additional two years on their previously assigned treatment. Blinded study methods were maintained; there was no crossover between treatments. One hundred thirty-eight subjects continued in CLAS-II; 103 completed a third angiogram before the CLAS-I outcome was known and CLAS-II terminated. The CLAS-II clinical procedures, lipid, lipoprotein-cholesterol, and apolipoprotein analyses were the same as in CLAS-I. The CLAS-II angiographic and file evaluation procedures exactly replicated those in CLAS-I.