Barcelona—Late in-stent thrombosis is plaguing the first generation of drug-eluting stents, and the results of 2 meta-analyses suggest that this may be causing myocardial infarction (MI) and death.
Late in-stent thrombosis may doom drug-eluting stents
These findings are prompting calls for an independent evaluation of not only drug-eluting stents but also percutaneous coronary intervention (PCI) for the treatment of patients with stable coronary disease.
Case reports of in-stent thrombosis followed by autopsy studies showing worse endothelialization and healing with drug-eluting stents compared with bare metal stents initially raised eyebrows about potential problems with drug-eluting stents. Then, at the annual meeting of the American College of Cardiology this past March, the results of BASKET-LATE (Basel Stent Kosten Effektivitäts Trial—Late Thrombotic Events) were presented. This study showed that during 7 to 18 months of follow-up, the rate of cardiac death and nonfatal myocardial infarction (MI) was significantly greater ( = .01) in subjects receiving a drug-eluting stent than those receiving a bare metal stent.
In exchanging bare metal stents for drug-eluting stents, “you’re trading restenosis, which is a relatively benign disorder, for late thrombosis, which is a very malignant or fatal disorder,” said Steve Nissen, MD, chief of cardiology at the Cleveland Clinic, and president of the American College of Cardiology.
Edoardo Camenzind, MD, presented the results of a meta-analysis of 9 randomized trials of first-generation drug-eluting stents, looking specifically at rates of death and Q-wave MI (when these data were made available). Using the latest available follow-up for each trial (up to 4 years), he found that the relative risk of death or MI was 38% greater with the sirolimus-eluting stent ( = .03) and 16% greater with the paclitaxel-eluting stent ( = .68) compared with bare metal stents.
The findings apply only to the first-generation drug-eluting stents, said Dr Camenzind, associate professor of cardiology at University Hospital, Geneva, Switzerland. Based on these data, indiscriminate use of drug-eluting stents should cease, he said, and a risk-benefit analysis of their use is warranted.
In a second meta-analysis, Alain J. Nordmann, MD, looked at 17 randomized controlled trials with follow-up of 1 to 4 years in which sirolimus- or paclitaxel-eluting stents were compared with bare metal stents.
In these trials, as follow-up progressed, both overall and cardiac mortality increased with the drug-eluting stents compared with their bare metal counterparts. Furthermore, noncardiac mortality increased starting at 2 years in the groups randomized to drug-eluting stents.
Among the noncardiac deaths in patients receiving the sirolimus-eluting stent were 15 cases of cancer, said Dr Nordmann, from the Basel Institute for Clinical Epidemiology. He concluded that “drug-eluting stents for the treatment of coronary artery disease do not reduce total mortality when compared to bare metal stents. Long-term follow-up and assessment of cause-specific deaths in patients receiving drug-eluting stents are mandatory to determine long-term safety of these devices.”
The clinical findings are already having an impact at hospitals, said Dr Nissen, with fewer now opting for drug-eluting stents. “There’s a subtle trend in the United States back toward bare metal stents as people get uneasy about this problem,” he said.
However, 6 million patients have already received drug-eluting stents, and now decisions must be made on how to limit their risk. Until now, up to 6 months of dual antiplatelet therapy with aspirin and clopidogrel (up to 12 months in patients without a risk of bleeding) were recommended. The thinking is that this duration may need to be extended.
“I would just be guessing but I would leave patients on clopidogrel and aspirin for longer than a year,” said Salim Yusuf, MD, professor of medicine, and chair in cardiology, McMaster University, Hamilton, Ontario.
Dual antiplatelet therapy does not solve the problem of in-stent thrombosis entirely, however, as some patients who experience a clinical event with a drug-eluting stent have done so while still on dual antiplatelet therapy.
Dr Yusuf questioned the use of not only drug-eluting stents but PCI in general for stable coronary disease, saying that PCI has never been shown to decrease the rates of death or MI compared with maximal medical therapy.
“The only difference (with PCI) is a short-term effect on angina relief that disappears when compared with medical therapy,” he said. “People have rationalized PCI and then treated the complications of PCI. We have zero data that PCI is useful in stable disease.”
In older patients with chronic heart failure (CHF) and preserved left ventricular function, perindopril failed to reduce the incidence of mortality or unplanned heart failure—related hospitalizations but did improve symptoms and functional capacity over 2 years. So reported John Cleland, MD, lead investigator of a study known as PEP-CHF (Perindopril in Elderly People with Chronic Heart Failure).
ACE inhibitor improves function but not mortality in diastolic heart failure
In PEP-CHF, 852 subjects with clinical evidence of CHF secondary to left ventricular diastolic dysfunction were randomized to placebo or perindopril, 2 mg/day titrated to 4 mg/day. The average age of the study population was 75 years. Consistent with the demographics of diastolic CHF, 56% of those enrolled were women, 79% had hypertension, and the mean left ventricular ejection fraction was 65%. The median follow-up was 26 months and the primary end point was all-cause death and hospitalization due to heart failure.
Planned recruitment was 1000 subjects, but recruitment was stopped early because of the slow pace of entering participants into the trial, which is not unusual when conducting clinical trials of older individuals, said Dr Cleland, University of Hull, Castle Hill Hospital, Kingston upon Hull, United Kingdom.
In addition, blinded therapy was stopped in 40% of subjects assigned to perindopril and 36% assigned to placebo at 18 months. Crossover to open-label angiotensin-converting enzyme (ACE) inhibitor therapy was the main reason for stopping blinded therapy, he said.
The incidence of the primary end point was reduced by 8% in perindopril recipients relative to the placebo group, but this difference failed to achieve statistical significance. According to Dr Cleland, the high crossover from placebo to perindopril at 12 to 18 months and a lower-than-anticipated event rate may have been responsible for the nonsignificant effect of perindopril on the primary end point.
Subjects assigned to perindopril spent a median of 5 fewer days in the hospital for any reason ( = .16) and 3 fewer days for cardiovascular reasons ( = .0557) compared with those assigned to placebo.
Despite the high crossover rate, subjects randomized to perindopril were significantly more likely to improve in New York Heart Association functional class (P < .03) and had a greater increase in their 6-minute walk distance (21 vs 8 m; P = .02).
Therefore, PEP-CHF represents the “first study to show a clear effect in improving symptoms (in CHF with preserved ventricular function) with a therapy other than diuretics,” said Dr Cleland.
During the first year of treatment, there was a strong trend toward clinical benefit with perindopril therapy, with a significant reduction in unplanned CHF hospitalizations, he noted. During that first year, 10.8% assigned to perindopril and 15.3% assigned to placebo experienced a primary outcome event, which approached significance ( = .055). There was a 37% reduction in unplanned CHF hospitalizations in the perindopril group during the first year, a significant result (P = .033).
Many physicians have already been treating patients with diastolic heart failure with an ACE inhibitor or angiotensin receptor blocker. “The results encourage people to continue to do what they’re doing,” he said. “If anything, they have new evidence that they can use an ACE inhibitor.”
Kenneth Dickstein, MD, at the University of Bergen, Norway, agreed that the high crossover rate led to a substantial loss of the study’s power, which contributed to the lack of a significant difference between perindopril and placebo on the primary end point. The 1-year data “may be the most reliable,” he said, including the impressive reduction in heart failure hospitalizations.