ATLANTA — Patients with a drug-eluting coronary stent had a 4.9% rate of cardiac death or myocardial infarction during the first year after they stopped daily treatment with clopidogrel, more than three times the rate in patients with bare-metal stents in a study with 743 patients.
The incidence of clinical events related to late stent thrombosis was 2.4% in patients with drug-eluting stents (DES), compared with 0.8% in those with bare-metal stents (BMS), Dr. Matthias E. Pfisterer said at the annual meeting of the American College of Cardiology.
On the basis of these findings, he calculated that for every 100 patients who receive a DES instead of a BMS in a coronary artery, the consequence is an extra 3.3 late deaths and myocardial infarctions. This was balanced against a DES benefit of 5 fewer patients who needed target-vessel revascularization for every 100 treated, compared with BMS, said Dr. Pfisterer, head of the division of cardiology at University Hospital Basel (Switzerland).
“You're trading late restenosis for deaths and MI. It's a very important and worrisome trade-off,” commented Dr. William W. O'Neill, corporate chief of cardiology at William Beaumont Hospital in Royal Oak, Mich.
The finding “may be practice changing,” commented Dr. Robert Harrington, of the department of medicine at Duke University in Durham, N.C.
The risk of late thrombosis lasted throughout a yearlong follow-up period after discontinuation of clopidogrel treatment. Events did not cluster early after the drug was stopped, but were scattered; some thromboses did not occur until the end of the follow-up period.
For the time being, the best way to treat late stent thrombosis in patients who have received a DES is to continue dual antiplatelet therapy with clopidogrel and aspirin indefinitely, Dr. Pfisterer said in an interview. Another way is to use the DES judiciously. In patients with focal lesions, the advantage of a DES over a BMS is small. In the future, new types of stents that are bioabsorbable or that contain different drug coatings and formulations may provide the solution.
The rate of thrombosis-related events linked with DESs was higher in this study than in past reports, most likely because the study enrolled all patients who needed a coronary stent at University Hospital Basel during May 2003-May 2004, regardless of their indication for stenting. The only exclusion was in patients with a target vessel diameter of 4.0 mm or greater, because the largest DES available at the time had a diameter of 3.5 mm.
As a result, the cohort included a large fraction of patients with unstable coronary disease, Dr. Pfisterer said. More than two-thirds of patients had multivessel disease, and 60% had acute coronary syndrome at the time of enrollment. The patients received an average of 1.9 stents each.
The Basel Stent Kosten Effektivitäts Trial (BASKET) was initially designed to compare the cost effectiveness of sirolimus-eluting stents (Cypher), paclitaxel-eluting stents (Taxus), and BMS. The study did not have any industry funding. The primary results of the study, which followed patients for 6 months after stent implantation, showed that the incremental cost-effectiveness ratio of DES to BMS was about 20,000 euros to avoid one major adverse cardiac event, and more than 50,000 euros per quality life year gained (Lancet 2005;366:921–9).
Dr. Pfisterer and his associates suggested limiting the use of DES to certain high-risk subgroups, such as elderly patients with triple-vessel disease, or patients with long lesions or lesions in small vessels.
All of the patients in the trial were taken off clopidogrel treatment after 6 months and continued to take aspirin daily, which provided an opportunity to assess the risk of late stent thrombosis after dual antiplatelet therapy was stopped. For this analysis, the 502 patients who had been initially randomized to receive either a sirolimus- or paclitaxel-eluting stent were combined into a single DES group and were compared with the 244 patients initially randomized to receive a BMS. Three patients from the DES group did not have follow-up, which reduced the group to 499 patients.
During the year off dual therapy, 4.9% of the DES patients had cardiac death or a nonfatal MI, compared with 1.3% of patients in the BMS group. The incidence of restenosis that required target-vessel revascularization was 6.7% in the BMS group and 4.5% in the DES group. When late thrombosis occurred in either group, it was usually clinically significant, with 88% of thrombotic events leading to death or MI.
In a multivariate analysis of risk factors for late stent thrombosis, three items emerged as significant predictors of risk. The use of a DES was the most potent risk, raising the incidence of thrombosis 3.9-fold, compared with patients who got a BMS. The other factors were the use of a glycoprotein IIb/IIIa inhibitor during stent placement (a marker for acute coronary syndrome at the time of stenting), which raised the risk 3.4-fold, and a history of MI, which raised the risk 3.0-fold, Dr. Pfisterer said.