Others caution, however, that comparing recent outcomes of DES with bare-metal-stent results spanning more than a decade does not reflect current standard of care.

Dr Kent G Meredith (LDS Hospital, Salt Lake City, UT) presented the results of the study here at the American Heart Association (AHA) 2006 Scientific Sessions; coauthor Dr Joseph B Muhlestein (LDS Hospital) presented the study to the media during a press briefing.

Dr Joseph B Muhlestein

The study tracked the incidence of all-cause death, nonfatal MI, target vessel revascularization (TVR), CABG, or the composite of these events (MACE) among all patients who underwent PCI with stenting between November 1993 and November 2005. In all, 7022 patients received a bare-metal stent over this period and 1949 received a DES after the devices became available in the US. Patient outcomes were then followed over an average of 2.1 years (to a maximum of 3.0 years).

At follow-up, target lesion revascularizations (TLR), as expected, were significantly reduced in the DES-treated patients, measured as days to TLR. But strikingly, days to non-TLR, non-TVR, and any coronary revascularization were all significantly less in the bare-metal-stent patients.

"Potential explanations for these findings include the possibility that the presence of a DES affects endothelial function and contributes to acceleration of atherosclerosis in nonstented coronary segments," Muhlestein hypothesized. DES-treated patients were also, in general, a more complex disease group, such that progression of disease may have been the reason for increased revascularizations, and the study may not have been able to control for these "unmeasured confounders," he said.

But alarmingly, while rates of nonfatal MI were no different between the two groups, death and MACE rates were significantly higher in the DES patients over follow-up. "Why was our death rate different between the two groups when the MI rate was no different? We concluded that it might be that we don't have the ability to ascertain the difference between a fatal MI and another cause of death," Muhlestein said.

Rates of subacute stent thrombosis were no different between the DES- and bare-metal-stent-treated groups, Muhlestein added: "Stent thrombosis in both groups was less than 2%, but our MI rates were higher than that, leading us to have some concern over whether or not these stents are having an adverse effect on the rest of the vasculature, triggering MI in other parts of the vessel that had nothing to do with where the stent is, but possibly having a proinflammatory effect." Indeed, a separate study presented today by Dr Jin Won Kim (Korea University Guro Hospital, Seoul) suggested that both sirolimus- and paclitaxel-eluting stents might adversely affect coronary endothelial function.

Dr David O Williams

During the press conference, Dr David O Williams (Rhode Island Hospital, Providence) downplayed the findings, pointing out that the randomized clinical trials showed no increased risk of death or TVR over extended follow-up. "I don't think there's a difference between death and MI that's been demonstrated between DES and bare-metal stents at this point, and I do think that the bulk of the data shows effectiveness in terms of revascularization. That's with the understanding that the DES patient receives dual antiplatelet therapy," he added.

Muhlestein, however, defended his data, saying, "Until recently, my opinion was exactly the same as Dr Williams's, and to some degree I still agree with his thoughts. But our individual study unexpectedly found this increase in death when we looked at three-year outcomes in our own patient population. That led me to wonder whether we're being too aggressive at our institution in doing too many sick and complex patients, and that we ought to consider some alternatives."

As with every press conference tackling the DES controversy in recent months, all of the spokespeople during the AHA's media briefing emphasized the need for large-scale randomized trial data to clarify long-term safety of DES and the role of extended dual antiplatelet therapy.

Williams is a coinvestigator on a study being presented later in the meeting by Dr Matthew Voss (University of Pittsburgh, PA). The study, another registry analysis, also compares DES with bare-metal-stent outcomes, this time in the National Heart, Lung, and Blood Institute's Dynamic Registry. Williams presented results from the study to the press in advance, showing no differences in death or MI and a reduction in target vessel revascularization and all repeat revascularizations in DES-treated patients at one year. But as press conference moderator Dr Sidney Smith (University of North Carolina, Chapel Hill, NC) pointed out following the presentation, the signal of harm in other studies—believed due to late stent thrombosis—has typically emerged after the 12-month mark.

Dr Sidney Smith

To heartwire, Voss acknowledged the need for longer-term data and said that he and his colleagues will be continuing to track outcomes in the Dynamic Registry for several years. But even now, he said, the findings are meaningful.

"I think our study can add to our comfort in using DES," Voss said. "It certainly doesn't answer the question past one year, but it does add useful information in that we did not see increased death or MI at one year, and no separation at all in the Kaplan-Meier curves at any point in time."

Voss also urged caution in the interpretation of the Salt Lake City data. Unlike the Dynamic Registry, which looked at relatively contemporary bare-metal stent and DES use, the study design for the Intermountain Heart Collaboration Study covered a much longer period of stent use.

"This study was looking at three-year outcomes, but it started in 1993 and went to the present time," Voss said. "Medical therapy and standard care were not the same back then. And when you compare 7000 patients treated since 1993 with the 1900 treated [more recently], that difference, under scrutiny, may reveal some bias in those numbers."

Responding to heartwire, Muhlestein said that he and his coinvestigators controlled for the differences in patient characteristics and standards in care as much as possible in their analysis. "Given the current dominance of DES in clinical practice, these findings raise concern that long-term outcomes need more scrutiny," he insisted.