BARCELONA — The cloud of uncertainty that has recently formed around the safety of drug-eluting coronary stents grew thicker when results from two metaanalyses suggested an excess of deaths and myocardial infarctions in patients who received these stents.
The new findings “raise concerns. I don't believe that they're convincing [of harm], but they're disconcerting,” Dr. Salim Yusuf said in a talk that commented on the new studies at a joint meeting of the European Society of Cardiology and the World Heart Federation.
One of the metaanalyses used all of the published data from the four studies that led to approval of the sirolimus-eluting stent (Cypher), and from the five pivotal studies that tested the paclitaxel-eluting stent (Taxus).
The second study used data from 17 trials that were supplied by the two companies that make these stents. Sirolimus-eluting stents appeared to fare the worst in the new analyses.
In the published studies, sirolimus-eluting stents were tested in a total of 878 patients, including 828 who were followed for 3 years. The studies also involved 870 comparator patients who received bare-metal stents, with 820 followed for 3 years. The fully pooled analysis showed an incidence of death or nonfatal MI of 6.3% in the sirolimus-eluting stent arms and 3.9% in the bare-metal stent arms, Dr. Edoardo Camenzind reported at the meeting. The 2.4% absolute difference (a 38% relative difference) in event rates in favor of bare-metal stents was statistically significant.
The paclitaxel-eluting stents were tested in more than 1,700 patients in the published studies, with a similar number of patients getting bare-metal stents. About 900 patients in each arm were followed for 3 years. The fully pooled analysis showed a death or MI rate of 2.6% in the paclitaxel-eluting stent group and 2.3% in the bare-metal stent group, a 0.3% absolute difference (a 16% relative difference) that was not statistically significant, said Dr. Camenzind, a cardiologist at University Hospital in Geneva.
The apparent excess of adverse clinical events with drug-eluting stents warrants a systematic risk-benefit analysis, he said.
The second report at the meeting included mortality data that had not previously been released by the manufacturers, and included studies with follow-up periods that ranged from 1 to 4 years. An analysis of overall deaths showed some trends toward increased mortality with sirolimus-eluting stents, compared with bare-metal stents, reported Dr. Alain J. Nordmann, an epidemiologist at the Basel (Switzerland) Institute for Clinical Epidemiology. No trend toward excess total deaths was seen for paclitaxel-eluting stents, but the results also showed that these stents provided no mortality benefit.
The analysis did not show any notable excess of cardiac deaths among patients who received either type of drug-eluting stent, and for some follow-up periods there were trends toward reduced cardiac deaths with both types of drug-eluting stents.
The results also showed an unexpected trend toward an excess of noncardiac deaths in patients getting sirolimus-eluting stents, including 15 deaths from cancers and several other deaths caused by sepsis or pneumonia. In five studies with 2-year follow-up, the excess of noncardiac deaths in patients getting sirolimus-eluting stents was 2.7-fold higher than in the comparator patients getting bare-metal stents, a significant difference, Dr. Nordmann reported, but he cautioned that the noncardiac death finding must be interpreted very cautiously. He called for longer-term mortality follow-up for these drug-eluting stents.
“I'm completely puzzled by the excess of noncardiac deaths,” commented Dr. Yusuf, director of cardiology at McMaster University in Hamilton, Ont., but it's a finding that deserves greater scrutiny, he said.
The new analyses follow several reports over the past few years of cases of late stent thrombosis in patients with drug-eluting coronary stents when they stopped dual antithrombotic therapy with aspirin and clopidogrel. Because of these studies and other reports of adverse events linked to drug-eluting stents, the Food and Drug Administration will convene a meeting of the Circulatory System Devices Advisory Panel by the end of the year, the agency said in a statement.
“It's important not to use drug-eluting stents indiscriminately in all patients,” commented Dr. William Wijns, a cardiologist at OLV Hospital in Aalst, Belgium, and a collaborator on the analysis reported by Dr. Camenzind.
Also, he said, physicians should place drug-eluting stents in patients only when it's certain that they will be compliant with dual-antiplatelet therapy and will continue it for some period of time.
He also cautioned against thinking that all drug-eluting stents act the same. “It's not a class effect. You need to look at each type individually,” Dr. Wijns said.