Abstract 12105: Predictors of Revascularization in the Cohort Assigned to Only Initial Medical Treatment in the BARI 2D Trial
Background: BARI 2D randomized 2368 patients (pts) with type 2 diabetes and stable ischemic heart disease to “prompt revascularization with intensive medical therapy” (REV+MED) versus “intensive medical therapy with delayed REV if clinically indicated” (MED). REV could be either PCI or CABG. Of the 1192 pts randomized to MED, 12.6% of pts had received REV by 6 months and 37.9% by 5 years.
Objective: To develop prognostic models to identify MED pts who would undergo early revascularization (within 6 months) and those who would require REV at any time during 5 years of follow-up.
Methods: Demographic characteristics, cardiac and diabetes clinical factors, angina severity and lesion severity were assessed as predictors of revascularization. Multivariate models were constructed controlling for basic demographic factors.
Results: Baseline angina status prior to stabilization was a strong predictor of early and overall REV with a 9% rate of 6 month REV in pts with no angina at baseline compared to 22% of pts with an initial presentation of unstable angina. The rates for overall REV were 33% and 52% respectively. Pts with proximal LAD disease had a 27.2% rate of overall REV compared to 14.2% in pts with no proximal LAD lesion.
Conclusions: Adjusting for age, sex, race, and geographic region, the strongest predictors of early REV are baseline angina status, low systolic blood pressure, elevated triglycerides and proximal LAD disease. For long-term REV at 5 years, number of diseased coronary regions was also predictive. Even with this model, only 52.1% in the highest-risk quartile of MED patients required REV within the 5 year follow-up period.
- © 2011 by American Heart Association, Inc.