New editorial fuels DES safety debate
October 12, 2006
Shelley Wood
Washington, DC - Stirring a pot already overflowing with conflicting opinions, a new editorial suggests that choosing drug-eluting stents (DES) over bare-metal stents is akin to substituting an ever-present peril for the "cosmetic problem" of angiographic restenosis [1].
"The DES may well have done away with the Achilles' heel of stenting (restenosis), perhaps only to be replaced by the Damocles' sword of stenting (stent thrombosis)," Drs Sanjay Kaul and George A Diamond (Cedars-Sinai Medical Center, Los Angeles, CA) write in the editorial, published on the CardioSource website.
But one of the early investigators of DES takes issue with Kaul and Diamond's views and warns against alarmist and "irresponsible" conclusions based on incomplete or inaccurate data.
"Is there a signal of late thrombosis? I'm sure there is," Dr Jeffrey Moses (Columbia University, New York), coinvestigator on the original SIRIUS trial, told heartwire. "The real question is the duration and the magnitude and whether this is offset by the benefits. And people who think that restenosis was nothing were not practicing interventionalists: we were radiating these patients by the hundreds, we were sending them to bypass surgery, and 10% of these people presented with infarction. The enormous cost and lifestyle disruption that it caused can not be overstated."
Editorial cites recent WCC 2006 presentations
Echoing concerns that have been voiced increasingly throughout 2006, Kaul and Diamond posit that the clinical benefits of DES in terms of reducing restenosis and revascularization have been overestimated by clinical trials that enrolled highly selected patients and mandated angiographic follow-up. In routine clinical practice, thin-strut bare-metal stents are used and associated with much lower restenosis rates than those seen for the bare-metal stents used in the pivotal DES trials. As well, in "real-world" patients, much more complicated lesions are stented with DES, leading to higher rates of restenosis and revascularization than those seen in the trials, they write.
We nevertheless estimate that using DES in 80% out of one million percutaneous coronary intervention cases would translate into 2160 excess deaths per year attributable to late stent thrombosis in the US alone—a risk far worse than that of tainted spinach.
On top of less impressive reductions in target vessel revascularization, the true incidence of death and MI—presumed due to stent thrombosis—outside of clinical trials is also cause for concern, the editorialists write, citing data presented by Camenzind and Nordmann at the recent World Congress of Cardiology (WCC) 2006 meeting, also reported by heartwire. According to Kaul and Diamond, a fundamental problem is that the pivotal randomized clinical trials of DES were not powered to evaluate death and MI; the signal from these recent studies is therefore all the more troubling. "Although these preliminary estimates are by no means conclusive (they require a more formal evaluation and thorough peer review), they sound a disquieting alarm, because they are consistent with the trial-by-trial pattern of findings with respect to death or MI over 18 months to five years of follow-up," they write.
It's ironic, Kaul and Diamond suggest, that a recent FDA advisory—recommending no change in use of DES pending an upcoming review—was released on the same day as an FDA warning about E coli-contaminated spinach. The spinach advisory—advising cessation of spinach consumption—was based on 50 cases of illness and one death among millions of potential spinach eaters.
"In the absence of a definitive trial, based on the reported estimate of 0.6% excess late stent thrombosis per year, and the attendant case fatality rate of 45%, we nevertheless estimate that using DES in 80% out of one million percutaneous coronary intervention cases would translate into 2160 excess deaths per year attributable to late stent thrombosis in the US alone—a risk far worse than that of tainted spinach, no matter how profound the reduction in restenosis!" Kaul and Diamond write.
Moses faults "unsupported" data
To start calling this a Vioxx alarms millions of people unnecessarily.
It's a calculation that doesn't sit well with Moses, who dismisses the data presented by Camenzind and Nordmann at WCC 2006 as "simply not accurate." Moses flatly rejected the findings from Camendzind's meta-analysis that showed an increased death/MI rate among sirolimus-eluting stent-treated patients. "There is no such data set that is present within the SIRIUS trial, at any time point you look at; there is nothing to support it," Moses said.
As for Nordmann's study showing an increase in noncardiac deaths with the sirolimus-eluting stent, Moses called the results "unsupported in the data set" and "hardly credible."
"We use rapamycin [oral sirolimus] in our heart-transplant patients here like crazy. These people are immunosuppressed, and I'm not aware of any major cancer or any other ancillary problems, and that's with major doses of rapamycin as opposed to the slight picogram exposure that stent patients are getting for a few weeks."
Moses continued: "To do those back-of-the-envelope calculations is frankly irresponsible. We don't know the frequency of late thrombosis, and we don't know if it translates into mortality."
But Kaul, for one, thinks otherwise, telling the New York Times that the signal of increased mortality with DES is "eerily reminiscent of Vioxx." The prudent line of action, Kaul and Diamond suggest in their editorial, is a thorough FDA review and labeling modifications if warranted, as well as a "more accurate, timely, and comprehensive postmarket surveillance program."
Unnecessary alarm, or right on the money?
In fact, the date for the FDA review is now set for December 7-8, 2006, and Moses advises a wait-and-see approach. The upcoming TCT meeting will also feature new data, including a special hotline session and press conference dedicated to the topic of stent thrombosis.
The Achilles' heel may have developed a new spur called DES.
"Let's approach this prudently and scientifically," Moses advises. "To start calling this a Vioxx alarms millions of people unnecessarily. The fact is, since day one of this endeavor with DES, we've always been conscious of the thrombosis issue because of our radiation experience. . . . A lot more data are coming out in the next few weeks, and my personal opinion is that it will show that [stent thrombosis] is a problem, but not of the magnitude that's being proselytized about or even close to that, but it's clearly something that's going to need to be addressed and refined."
Despite calls for calm from Moses and others, an increasing number of cardiologists are growing more and more concerned. "I actually think Kaul and Diamond are right on the money with this one," Dr Paul Armstrong (University of Alberta, Edmonton) commented to heartwire. Armstrong listed his concerns as being a "slippage in indications," overuse of DES, patient noncompliance with antiplatelet therapy, and the signal that the stent thrombosis risk doesn't appear to abate over time, as suggested by what has been dubbed the "Rotterdam-Bern" analysis presented at WCC 2006 and quoted in the editorial.
"The extra costs [of DES] are often unjustified in my view and could be put to better use," Armstrong commented, adding that while an FDA review would be well received, the clinical community, hospital administrators, and device makers may also need to rethink DES usage. "The Achilles' heel may have developed a new spur called DES," Armstrong quipped. |
