BARI is a randomized multicenter international clinical trial that compares a strategy of initial PTCA to that of initial CABG for selected patients with CAD. The primary hypothesis tested is that a strategy of initial PTCA is no worse than one of initial CABG when assessed by mortality at 5 years. Other major endpoints include: myocardial infarction (MI), need for repeat procedures and hospitalizations, symptomatic and functional status, radionuclide ejection fraction, quality of life, and economic impact. Beginning in 1987, investigators from 14 primary sites and 4 satellites randomized 1,829 patients (45%). In addition to the randomized cohort, the investigators recruited into a registry 2,013 patients who were eligible but not randomized and a random sample of 422 patients deemed ineligible based on their angiogram. The patients randomized to initial PTCA were similar in severity of disease, concomitant diseases, cardiac risk factors, and demographics to those randomized to CABG. Overall, the cohort had a mean age of 61.5 years and is 27% female, 90% white, and 6% black. Subjects also had somewhat advanced disease with a mean of 3.5 significant lesions which on the average jeopardized 61.5% of left ventricular (LV) territory. The investigators propose to: complete the 5-year followup on all BARI patients, extend the followup of the randomized and registry cohorts to a minimum of 10 years, determine the relative efficacy of PTCA vs. CABG in subgroups of women, blacks, and elderly, and assess the public health impact of BARI. They propose annual telephone interviews for all patients currently enrolled. All randomized patients will receive an annual rest electrocardiogram (ECG) from their primary physician and for any suspected MI and/or revascularization procedure. Instruction for behavior modification in the areas of smoking cessation, exercise, and diet will be reinforced at each followup contact. Periodic monitoring of smoking cessation rates and cholesterol levels are proposed, although no provision for blood testing is provided. The original BARI protocol planned for repeat angiography at year 5; however, due to changes in reimbursement policies, this will be replaced by a radionuclide ventriculogram (RNV) for determination of ejection fraction. The RNVs will be read locally rather than centrally to reduce costs. Certification and quality control procedures are planned to standardize the RNV measurements at each site. A separate application has been submitted as an ancillary project by 6 clinical sites to obtain 5-year angiograms in patients at these sites. However, intercurrent angiograms obtained for clinical indications will be read by the Central Radiographic Laboratory (CRL). In the registry patients, ECGs will be obtained at years 5 and 10 and for any suspected MI or revascularization procedure. Data will be collected from each cardiac hospitalization of >3 days, PTCA, CABG, or other cardiac procedure, and hospitalization for cardiac arrest or complication from a revascularization procedure. An ancillary study funded elsewhere will compare cost of treatment and quality of life issues in the 2 treatment arms. The investigators also propose to continue semiannual surveys of all revascularization procedures performed at BARI sites to provide estimates of the """"""""universe"""""""" of revascularization procedures from which the BARI patients were selected. Data will be collected on the type of procedure, patient demographics, and disease severity. A similar survey will be conducted in a group of hospitals that are representative of institutions that perform both PTCA and CABG in the United States and Canada.
