Winston-Salem, RI - A drug-eluting stent (DES) is a better choice than bare metal for reducing the risk of not only restenosis but also death or MI in patients treated "off-label," a new study suggests [1]. Two-year follow-up data from the Wake Forest registry study—first announced as nine-month outcomes at the American College of Cardiology 2006 Scientific Sessions, as reported by heartwire—should help assuage concerns that DES somehow unduly increase the risk of adverse outcomes in people who do not meet their approved indications for use.

The study, appearing in the February 12, 2008 issue of the Journal of the American College of Cardiology, comes within weeks of another registry study, by Marroquin et al, with just one year of follow-up, that found no differences in outcomes by stent class [2].

In an interview with heartwire, lead author Dr Robert J Applegate (Wake Forest University, Winston-Salem, NC) explained that concerns about the risks of off-label DES use, first singled out for special attention during a December 2006 FDA hearing, came from studies that compared outcomes in all patients treated with DES, on-label and off. But missing, he argued, were data examining outcomes in patients treated off-label with DES as compared with bare-metal stents. Even during the FDA hearing, committee members concluded that risks of off-label use were higher for both DES and bare-metal stents, but whether or not the degree of risk was exponentially greater for DES was unclear.

For their study, Applegate and colleagues compared two-year outcomes among 1164 consecutive patients treated with a bare-metal stent (before the introduction of DES) with 1285 consecutive patients treated with DES, the "routine choice" after the devices came on the market. They report that overall risk of MI or death was reduced 23% in DES-treated patients as compared with bare-metal-stent-treated patients, but for both groups, risks of MI, death, or stent thrombosis were lower for on-label vs off-label procedures, regardless of stent type. However, when off-label results were examined separately, risks of MI, death, or all-cause mortality were significantly reduced in DES-treated patients as compared with bare-metal-stent-treated patients. Even in specific high-risk subgroups—including patients with AMI, diabetes, serious heart failure, and renal failure; or older patients and female patients—outcomes consistently favored DES over bare-metal stents.

Of note, clopidogrel usage was roughly the same for the two stent groups at two years, at approximately 40% for both groups. Applegate explained that the seemingly high rate of clopidogrel use in bare-metal-stent-treated patients related to the high proportion of patients with ACS undergoing stenting.

In an accompanying editorial, Dr Cindy Grines (Beaumont Hospital, Royal Oak, MI) says Applegate et al's findings should on the whole be "reassuring" to the interventional cardiology community [3].

"Although additional trials are warranted, it appears that DES are safe and effective both for FDA-approved indications as well as for many off-label indications," she writes. This leaves the field open for implanting physicians to use their best judgment when choosing a stent for "off-label" lesions at increased risk of restenosis—namely, small vessels, long lesions, in-stent restenosis, chronic occlusions, or vein grafts, she adds. In her own practice, Grines says she now uses DES liberally, with the exception of primary PCI, where she says the choice of a DES is more controversial, due to a lower antirestenosis benefit, difficulties prejudging a patient's compliance with clopidogrel, and higher risk of inappropriate stent sizing and malaposition.

In response, Applegate counters that in his practice, there is no patient group or lesion characteristic for whom he believes a bare-metal stent is a better choice than a DES. Among 847 patients in their study treated for a recent MI (seven days old or less), the hazard ratio for MI or death still favors DES (0.77; 95% CI 0.57-1.04).

He calls the concerns about DES use in AMI "largely theoretical," based largely on pathological studies and one study pointing to a link between visible thrombus during PCI and worse outcomes. TYPHOON, one of the first studies to look at this issue, showed a benefit of a Cypher stent compared with a bare-metal stent in AMI, he adds, as have other meta-analyses.

"Based on our own data, we can't identify a single aspect of the usual things about our patients in their presentation, comorbid illnesses, or lesion characteristics that identify a high-risk group that shouldn't receive a DES. What we do is we ask the patient, the family, and the referring physician whether or not the patient can take clopidogrel for at least a year. That's the single piece of information that we use at this point in time to make our decision on whether or not to use a DES."

But Applegate's approach is not universal, as underscored by Dr P Gabriel Steg (Université de Paris XII, France) during the 2007 ESC meeting, where he presented the result of the GRACE registry. The as-yet unpublished analysis suggested that STEMI patients treated with DES face more than a fourfold increased risk of dying than do those who receive bare-metal stents, prompting Steg to state at the time: "I personally never implant a DES into a STEMI patient these days."

Asked about Steg's GRACE registry presentation, Applegate explained, "I am a little conservative in extrapolating clinical practice from abstract presentations—-I would like to see the complete published results first. Having said that, it does provide concern for using DES in STEMI patients."

In the current study, he and his coinvestigators did not specifically single out STEMI patients within the MI subgroup who underwent PCI within seven days or less. But Applegate told heartwire that his group has subsequently done a statistical analysis on this smaller subgroup that suggests there is a significantly better outcome at two years with DES than bare-metal stents in STEMI patients. This analysis will be presented at the ACC 2008 Scientific Sessions, he added.

"It is always challenging to provide optimal patient care when relevant data about treatment strategies diverge. We need more information and data. We also need to resist the urge to base clinical decisions on observations that have not been thoroughly scrutinized and confirmed."

Steg, in turn, commented on Applegate et al's study for heartwire, noting that he does not think it directly contradicts his findings from GRACE, since these dealt specifically with STEMI patients, and he said he has "no major problem" using a DES for non-STEMI or elective PCI post-STEMI.

That said, Applegate et al's paper, more generally, "does not address a fundamental concern I have with the use of DES for primary PCI in STEMI, which is that it is very difficult to reliably assess the risk of bleeding, comorbidities, and potential compliance of the patient to long-term dual antiplatelet therapy in this setting," Steg commented. "Conversely, restenosis, which is what DES really reduce, is not so much a concern in primary PCI for STEMI, so why take the risk of implanting a DES routinely in this setting? If we find out on the next day that the patient has cancer or is not compliant with therapy, there is no way out, whereas we can usually easily handle his or her restenosis eventually. It is a setting where I clearly prefer to be safe than sorry."

Some answers on the safety and efficacy use of DES in the setting of AMI will come from the DES arm of the HORIZONS-AMI trial, expected to be released at the TCT 2008 meeting later this year.