Post Coronary Artery Bypass Graft (CABG) Study

BACKGROUND:

A large number of studies have reported that following CABG, the vessels proximal to the grafts demonstrate accelerated atherosclerosis and the grafts themselves may show progressive obstruction due to thrombosis, fibrosis, and graft atheroma. Pathological and clinical studies have documented that graft obstruction during the perioperative period and the first year is usually due to technical problems at surgery or thrombosis and occurs in about 15 to 20 percent of saphenous vein grafts. Antiplatelet drugs such as aspirin have been shown to reduce these early graft occlusions by about 50 percent. In addition to thrombosis, during the first year, most saphenous vein grafts undergo diffuse fibrosis and some distortion. However, the significance of these changes is not known.

After the first year, lipid deposition and changes histologically similar to atherosclerosis have been documented both in experimental studies and in human saphenous vein grafts patent at one year completely occlude and an additional 30 percent narrow over a period of ten years. These angiographic changes in the saphenous vein grafts correlate with high LDL-cholesterol, low HDL-cholesterol and high apolipoprotein B levels. In experimental animals, cholesterol deposition in the grafts has been substantially increased or decreased by increasing or decreasing the saturated fats and cholesterol in the animal's diets. In contrast to changes in the saphenous vein grafts, internal mammary artery grafts have shown substantially lower rates of obstruction. The native vessels (both ungrafted vessels and grafted vessels at distal or proximal sites), however, show evidence of progressive atherosclerosis. The long-term follow-up of grafts also demonstrates thrombotic material and even occlusive thrombus as part of acute events.

Progression of atherosclerotic lesions in grafts and native vessels lead to recurrent angina, unaltered long-term survival and reduced efficacy of repeat CABG surgery. Johnson in an 11 year follow-up of 3,105 post-CABG patients, reported that about 15 percent of patients with preoperative angina have recurrent angina at 1 year; with a further 6 percent developing angina every subsequent year. In this study, patients with recurrent angina had twice the mortality compared to those who were angina-free. Long-term follow-up of the VA Cooperative Trial of CABG shows that the survival of the surgical group appears to be initially superior compared to the medical group but this benefit is diminished by about ten years. This may relate to graft obstruction and progression of native coronary atherosclerosis. Re-operation in these patients carries a higher operative mortality risk and the results are less impressive. It had been estimated that approximately 5 percent of all CABG surgery in 1984 were re-operations and that this percentage would double over the next decade. Therefore, measures to prevent graft occlusion and progression of atherosclerosis in native vessels, if successful, could have substantial clinical and economic importance by reducing mortality, morbidity and the numbers of patients undergoing re-operations.

Apart from trials of aspirin and dipyridamole in post-CABG patients that demonstrated a significant reduction in graft closure within the first year after surgery, there were no large systematic studies of interventions in these patients. Graft occlusion has been shown to correlate with high LDL-cholesterol and low HDL-cholesterol, and the severity of atherosclerosis has been additionally shown to be related to cigarette smoking and increased levels of coagulation factor. Therefore, lowering the LDL cholesterol (by diet and drugs), and anti-thrombotic therapy with warfarin were logical choices for intervention.

Several studies suggested that CABG surgery relieved angina and improved overall quality of life. However, in the available studies, CABG did not consistently appear to be associated with an improvement in employment status, physical recreational activity, or life style. The reasons for these results were not entirely clear and there was a need to identify the biobehavioral and psychosocial factors that predicted successful adjustment after CABG.

The initiative was proposed by Institute staff and approved by the September 1985 National Heart, Lung, and Blood Advisory Council. The Request for Proposals was released in September 1985 and awards made in April 1987.

DESIGN NARRATIVE:

Multicenter, double-blind, randomized, controlled trial. All prospective participants received active warfarin treatment for a month prior to randomization. Only participants demonstrating a minimal reaction to warfarin and consuming over 90 percent of the prescribed medication were randomized. Dietary modification to lower serum cholesterol, an exercise program, and a smoking cessation program were implemented. Patients were randomly assigned in a 2 X 2 factorial design in four treatment groups: aggressive LDL-C lowering with lovastatin 40 to 80 mg/d and, as necessary cholestyramine 8 mg/d to achieve and LDL-C of 60 to 85 mg/dl; moderate LDL-C lowering with lovastatin 2.5 to 5 mg/d, with cholestyramin as needed, to achieve a LDL-C of 130 to 140 mg/dl; warfarin 1 to 4 mg/d to achieve an INR of 1.8 to 2.0; and warfarin-placebo. All participants were followed for five years, at the end of which selective coronary and graft angiography was performed. The primary angiography endpoint was the proportion of major SVG per patient that showed substantial reduction (0.06 mm or greater) in lumen diameter. Biobehavioral studies were conducted in 750 participants.

Planning for the study began in April 1987 and a final protocol was developed by August 1988. Patient recruitment has been completed. Follow-up ended on September 1, 1995 and data analysis continues through December 1998 under contract N01-HC-75076. .

The Post CABG Biobehavioral Study examined a cohort of 759 coronary artery bypass patients (269 women and 490 men) who were enrolled at five clinical centers. Sociodemographic and medical data were obtained by interview and from medical charts. Health-related quality of life and psychosocial data were ascertained preoperatively by interview and questionnaire for those patients whose condition allowed preoperative assessment and was compared among patients from hospitals enrolling both male and female patients.