10-Year Follow-up of Early DES Shows Steady Risk of Late Events
The SORT OUT II study raises questions about the ongoing surveillance of patients with first-generation stents.
First-generation DES have now been replaced by newer, improved iterations, but researchers say the millions of patients with older stents in their coronary arteries remain at risk for late events and should be followed appropriately.
Reporting the 10-year follow up from the SORT OUT II trial in the Journal of the American College of Cardiology, Anders M. Galløe, MD, PhD (Zealand University Hospital, Roskilde, Denmark), and colleagues say MACE and stent thrombosis remain serious concerns in patients treated with the original Cypher, (Cordis, Johnson and Johnson, Miami, FL) and Taxus (Boston Scientific, Natick, MA) stents.
With more than 70% of the nearly 2,100 patients in the original trial still alive, the study has implications for millions of real-world patients and physicians, and highlights the fact that “patients are not ‘home free’ just because nothing has happened for 9 years,” Galløe told TCTMD in an email. He and his colleagues found that the event curves were straight lines, indicating that the risk of a first event happening after 9 years was almost the same as the risk in the second year after DES implantation.
Increased risk of late stent thrombosis with first-generation DES took center stage during the European Society of Cardiology 2006 Congress based on results from two large meta-analyses. Responding to the controversy, the US Food and Drug Administration (FDA) at the time convened a special panel to review the risk of stent thrombosis with drug-eluting stents. A rapid and significant decline in DES use following the “DES firestorm” lasted for more than a year.
Now, with 10-year follow-up, SORT OUT II data add to evidence from registry reports showing ongoing risk of stent thrombosis out to 5 years with first-generation devices, with no clear evidence of an attenuation of this effect over time. The annual rate of Academic Research Consortium (ARC)-defined definite, probable, and possible stent thrombosis after the first year was 1.3% through 10 years in SORT OUT II.
“There is no unique surveillance or preventive strategy that can reduce the risk of very late events with first-generation DES,” write David J. Moliterno, MD, and Khaled M. Ziada, MD (both University of Kentucky, Lexington, KY), in an editorial accompanying the study. “Certainly, it is imperative to continue secondary prevention measures to limit disease progression and reduce the risk of non-stent-related events.”
Questions Regarding Surveillance Remain
SORT OUT II enrolled 2,098 unselected all-comer patients between 2004 and 2006 who were randomly assigned to the first-generation Cyper sirolimus-eluting stent or the Taxus paclitaxel-eluting stent.
MACE, the primary endpoint, occurred at a rate of 32.5% at 10 years in the sirolimus group and 33.1% in the paclitaxel group (HR 0.96; 95% CI 0.83-1.12). Rates of all-cause death were similar at 27.4% vs 26.3% (P = 0.60). Target lesion revascularization (TLR) rates also were similar between the sirolimus group and the paclitaxel group with no suggestion of a late increase. Cardiac death occurred at a rate of approximately 9% in each group, while MI was 18% in each group.
According to Galløe and colleagues, the findings “invite speculation about the rationality of prolonging the surveillance of patients treated with first-generation DES” and raise questions about prescribing or reinitiating long-term dual antiplatelet therapy. While the researchers conclude that there is “no apparent need for extraordinary medical attention” for patients who have gone years after a first-generation stent was implanted and not had an event, Galløe told TCTMD that his group believes continued surveillance is appropriate to ensure detection of warning signals that might necessitate medical intervention.
The editorialists note that the newest SORT OUT II data, as well as those from SIRTAX VERY LATE, “suggest that patient-related and noncardiac variables play substantial roles in determining 10-year outcomes, likely larger roles than that of procedure-related factors, especially later in the follow-up course.”
Galløe AM, Kelbæk H, Thuesen L, et al. 10-year clinical outcome after randomization to treatment by sirolimus- or paclitaxel-eluting coronary stents. J Am Coll Cardiol. 2017;Epub ahead of print.
Moliterno DJ, Ziada KM. More time to SORT OUT clinical outcomes after first-generation drug-eluting stents: related versus unrelated events. J Am Coll Cardiol. 2017;Epub ahead of print.
- The study was supported by unrestricted grants from Boston Scientific and Cordis. Galløe and Ziada report no relevant conflicts of interest.
- Moliterno reports having served on the Data Safety Monitoring Board for Janssen Pharmaceuticals, and research grant support from AstraZeneca.