Dr. Stone reported that at 5 years, the primary composite of death, stroke, or MI occurred in 22.0% of patients in the PCI group and 19.2% of patients in the CABG group, a nonsignificant difference at P = 0.13).
However, when the researchers broke the results into three distinct risk periods within the 5-year time frame, they found that, with longer follow-up, came more of an advantage for CABG. The relative risk of PCI vs. CABG for the primary outcome favored PCI over CABG in the first 30 days (4.9% vs. 8%; hazard ratio, 0.61; P = .008), was neutral at 30 days to 1 year (4.1 vs. 3.8%; HR, 1.07; P = .76), and reversed at 1-5 years (15.1% vs. 9.7%; HR, 1.61; P less than .001). Using restricted mean survival time analysis, Dr. Stone and his colleagues found that, at the end of the 5-year follow-up period, event-free survival time was 5.2 days longer after PCI, compared with CABG. This translates into “a very similar event-free survival of a burden of disease from these two therapies at the end of 5 years,” he said.
In their analysis of secondary endpoints, some differences were noted, including an elevated risk of all-cause mortality in the PCI group, compared with the CABG group (13% vs. 9.9%, respectively; odds ratio, 1.38), yet no differences in definite cardiovascular mortality (5% vs. 4.5%; OR, 1.13) or in MI (10.6% vs. 9.1%; OR 1.14). In addition, there were fewer cerebrovascular events in the PCI vs. CABG groups (3.3% vs. 5.2%; OR, 0.61). “Overall, all of these differences were relatively small given the 5-year perspective,” Dr. Stone said at the meeting sponsored by the Cardiovascular Research Foundation. He concluded that the early benefits of PCI attributable to reduced periprocedural risk “were attenuated by the greater number of events occurring during follow-up with CABG, such that at 5 years the cumulative mean time free from adverse events was similar with both treatments.” He noted that a 10-year or longer follow-up is required to characterize the very late safety profile of PCI and CABG as both stents and bypass grafts progressively fail over time.
Discussant Dharam Kumbhani, MD, an interventional cardiologist at UT Southwestern Medical Center, Dallas, said that the findings from EXCEL “help us move the field forward and help us understand this concept of risk with PCI versus CABG. It really does help inform shared decision-making with patients.”
Results of the study were published online at the time of presentation (N Engl J Med. 2019 Sep 28. doi: 10.1056/NEJMoa1909406). The EXCEL trial was funded by Abbott Vascular. Dr. Stone disclosed having relationships with numerous device and pharmaceutical companies but had no relevant disclosures for this study.
SOURCE: Stone G et al. TCT 2019. N Engl J Med. 2019 Sep 28. doi: 10.1056/NEJMoa1909406.