EXAMINATION Surprise: Few Late Event Gaps Between DES and BMS in STEMI

Despite their early superiority over bare-metal stents, drug-eluting stents were associated with similar event rates and survival in the latter half of follow-up in the EXAMINATION trial, which compared stent performance for STEMI over a decade.

“The 10-year follow-up confirms the safety and efficacy of the everolimus-eluting stent (EES) as compared to bare-metal stenting in the setting of primary PCI for ST-elevation MI,” said Salvatore Brugaletta, MD (Hospital Clínic de Barcelona, Spain), in a presentation last week at TCT Connect 2020.

The event curves for both devices remained surprisingly parallel for the primary endpoint (a composite patient-oriented endpoint of all-cause death, any recurrent MI, or any revascularization) from approximately 5 years to 10 years, despite the DES maintaining an edge over BMS for cumulative incidence of events over the study period (log-rank P = 0.013). The same was true for the secondary endpoint (a composite of cardiac death, target-vessel MI, or TLR; log-rank P = 0.012).

Both the primary and secondary endpoint curves were also relatively flat beyond 5 years, with few events and no differences in rates of late definite/probable stent thrombosis. Brugaletta showed in landmark analyses that the majority of the benefit of the EES over the 10 years was restricted to the first 2 to 3 years after stent implantation.

Prior findings from EXAMINATION showed no superiority of EES over BMS on the primary endpoint at 1 year, but did demonstrate lower risk for the EES group versus the BMS group at 5 years, driven by reductions in mortality and TLR. At 10 years, rates of all-cause death trended lower at 19.4% in the EES group versus 23.2% in the BMS group (P = 0.06), with no difference in rates of cardiac death or MI. Revascularization also trended lower with EES compared with BMS (16.5% vs 19.9%; P = 0.07), but was significant only for TLR (P = 0.01).

Unexpected Parallel Event Rates

Anthony Gershlick, MBBS (University of Leicester, England), who served as discussant following the presentation, called the results “fascinating,” adding that he never would have predicted the parallel event curves at 10 years.

“I was expecting, on previous data, that the curves were likely to separate over time, not least because in-stent restenosis, of course, leads to acute coronary syndrome,” he remarked. “Even the stent thrombosis rates were really low. I think what it tells you somewhat is that the two devices used were probably the best that were around at the time.”

In the EXAMINATION trial, STEMI patients undergoing emergent PCI within 48 hours of symptom onset were randomized to the Xience V EES (n = 751) or Multi-link Vision BMS (n = 747; both Abbott Vascular). At 10 years, 94.6% of the EES group and 95.8% of BMS group completed follow-up.

Whereas 1-year rates of definite/probable stent thrombosis were lower with the EES at 0.9% versus 2.5% for the BMS (log-rank P = 0.01), the difference was gone at 10 years, with rates of 0.6% and 0.4%, respectively (log-rank P = 0.70). For TLR, the significant advantage of the EES over BMS at 1 year in terms of lower rates (2.1% vs 5.0%; log-rank P = 0.003) again was no longer apparent at 10 years (1.4% vs 1.3%; log-rank P = 0.96).

Gershlick asked Brugaletta if the results change his view of BMS. “In most people's practice they're getting ditched because of all the negative hype around stent thrombosis due to restenosis and high revascularization rates,” Gershlick said.

Brugaletta acknowledged that the EXAMINATION investigators themselves were surprised. The results confirm, he added, that while there is significant benefit early on, the advantage does not seem to last.

“You're not advocating the reintroduction of bare-metal stents based on these data? I mean, they're gone [but] they're still relevant in certain parts of the world where they still have value," Gershlick pressed.

Brugaletta noted that in his view, the findings really attest to the durability of the polymer over many years, and said published 5-year cost analyses by the EXAMINATION investigators showed the EES was cost-effective in comparison with the BMS.

In an email, Gershlick reiterated his comments to TCTMD, saying “essentially, unless it is for real financial issues (eg, a country where the patient has to pay and cannot afford the tariff of a DES), in my book there is no place for BMS in contemporary interventional cardiology.”

Even session moderator George Dangas, MD, PhD (Icahn School of Medicine at Mount Sinai, New York, NY), expressed his surprise, calling the EXAMINATION results “controversial” and “groundbreaking.” To Gershlick, Dangas added, “Salvatore doesn't have to advocate for bare-metal stents. Clearly, the results do that.”