Propensity Analysis of Long-Term Survival After Surgical or Percutaneous Revascularization in Patients With Multivessel Coronary Artery Disease and High-Risk Features - Propensity Analysis of Long-Term Survival After Surgical or Percutaneous Revascularization
The goal of the study was to evaluate long-term survival rates after percutaneous coronary intervention (PCI) or surgical coronary revascularization (CABG) in patients with multivessel coronary artery disease (CAD) and high-risk clinical features.
Patients Enrolled: 6,033
Mean Follow Up: Median 5.2 years
Mean Patient Age: Mean age 64 years
Mean Ejection Fraction: Mean baseline ejection fraction 55% in PCI cohort and 46% in CABG cohort (p<0.0001)
Consecutive patients with multivessel CAD revascularized from 1995-1999 at the Cleveland Clinic
Prior CABG, required valve surgery in addition to revascularization, if a cardiac surgeon had refused them because of severe comorbidity or lack of appropriate target vessels, if they died during the procedure (two patients), if they were undergoing primary PCI for an acute MI, or if they did not have a valid US Social Security number
The study evaluated consecutive patients with multivessel CAD revascularized from 1995-1999 at the Cleveland Clinic (n=6,033). PCI was performed in 872 patients (14%) and CABG was performed in 5,161 patients (86%). In order to adjust for the probability of undergoing PCI based on baseline characteristics, a propensity analysis was conducted.
Patients undergoing CABG were generally higher risk than patients undergoing PCI, with higher rates of hypertension (76% vs. 67%), prior myocardial infarction (MI) (58% vs. 48%), and peripheral vascular disease (34% vs. 14%, all p<0.0001) and lower ejection fractions (46% vs. 55%, p<0.0001). CABG patients also more frequently had left main disease (23% vs. 2%), proximal left anterior descending artery disease (52% vs. 28%), and triple vessel disease (74% vs. 33%, all p<0.0001). Among the patients undergoing PCI, stents were used in 70% and glycoprotein IIb/IIIa inhibitors were used in 51%.
After a median follow-up of 5.2 years, mortality was higher in the PCI cohort than the CABG cohort (16% vs. 14% unadjusted Kaplan-Meier estimates at five years, unadjusted hazard ratio [HR] 1.13; 95% confidence interval [CI] 1.0-1.4; p=0.07). After risk adjustment, PCI was associated with an increased risk of death (adjusted HR 2.1; 95% CI 1.7-2.6; p<0.0001).
Similar results were reported in a model adjusting for the propensity score for performance of PCI (HR 2.3; 95% CI 1.9-2.9; p<0.0001). In an analysis of patients undergoing stenting (n=609) versus CABG, stenting was associated with a higher risk of death (adjusted HR 2.2; 95% CI 1.7-2.9; p<0.0001). Mortality was also consistently directionally higher in the PCI cohort when stratified by deciles of propensity score for PCI or high-risk subgroup analysis such as diabetes and left ventricular (LV) dysfunction.
Among patients with multivessel CAD and high-risk clinical features, revascularization with PCI was associated with an increased risk of death at five years compared with revascularization with CABG. Prior randomized trials showed PCI was associated with similar mortality rates compared with CABG for the treatment of multivessel disease. However, the trials generally did not enroll patients as high-risk as are reported in the present registry, such as those with severe LV dysfunction and a high proportion of diabetics. Additionally, subgroup analysis of several of the randomized trials such as BARI suggested CABG was associated with a lower mortality compared with PCI in high-risk subgroups such as diabetics and patients with triple vessel disease.
There are several limitations in the present study, which should be noted. The study was a registry analysis, not a randomized trial. Although propensity analysis was used to adjust for the choice of revascularization strategy, not all factors influencing the choice of strategy can be known or can be included in the adjusted analysis. Additionally, the registry was of a single center that had very low perioperative mortality.
Brener SJ, et al. Propensity analysis of long-term survival after surgical or percutaneous revascularization in patients with multivessel coronary artery disease and high-risk features. Circulation 2004;109:
Clinical Topics: Invasive Cardiovascular Angiography and Intervention, Prevention, Vascular Medicine, Atherosclerotic Disease (CAD/PAD), Interventions and Coronary Artery Disease, Interventions and Vascular Medicine, Hypertension
Keywords: Kaplan-Meier Estimate, Myocardial Infarction, Coronary Artery Disease, Propensity Score, Follow-Up Studies, Peripheral Vascular Diseases, Percutaneous Coronary Intervention, Risk Adjustment, Stents, Survival Rate, Confidence Intervals, Hypertension, Diabetes Mellitus
< Back to Listings