Abstract 6129: COURAGE Revisited: Can Noninvasive Identification of Ischemia Identify a Long-term Therapeutic Benefit of CABG or PCI versus Medical Therapy
Background: In light of the COURAGE trial results, the relationship between ischemia and benefit from revascularization in patients (pts) with known CAD (HxCAD) has been questioned. We examined whether stress SPECT myocardial perfusion imaging (MPI) can identify which HxCAD pts may accrue a long-term relative survival benefit with either early PCI (within 90 days of testing; EPCI) or early CABG (ECABG) compared to medical therapy (MT).
Methods: We identified 13,969 patients (pts) who underwent SPECT MPI between 1991 and 1997 using adenosine or exercise (EX) stress MPI. Patients lost to follow-up (FU) (2.8%) and those without HxCAD were excluded (no prior PCI/CABG/MI; 8,797), leaving 4,758 HxCAD pts with FU (59% EX, 72% male, age 68±11). Short-term FU was performed at 12–18 months for EPCI and ECABG and pts were matched against the Social Security Death Index for all cause death (ACD). Mean FU was 8.7±3.3 years. Cox proportional hazards modeling (CPH) with adjustment for logistic-based propensity scores was utilized to identify a significant interactions between MPI ischemia (ISC) and either EPCI or ECABG.
Results: On FU, 1852 ACD (39%, 4.5%/year), 449 EPCI (9.4%), and 251 ECABG (5.3%) occurred. After risk-adjustment for clinical/historical data using CPH, no significant EPCI-ISC interaction was present (interaction p=0.75, overall model chi-square 1330, p<0.0001), indicating no survival benefit with EPCI. However, a significant ECABG-ISC interaction was present (interaction p<0.02, overall model chi-square 1318, p<0.0001), indicating improved survival with ECABG in HxCAD pts with ISC but not in pts with small or no ISC. The threshold for benefit was ≥10% myocardium ISC. Importantly, this interaction was present in pts without prior MI, but not in pts with MI.
Conclusion: In this observational series of HxCAD pts with long-term follow up, a threshold of ~10% ISC defined pts with a survival benefit from ECABG over medical therapy, but no such threshold could be identified for EPCI.