Among patients who undergo primary percutaneous coronary intervention (PCI) for ST-segment elevation myocardial infarction (STEMI), the risk of very late stent thrombosis persists for many years, according to a single-center registry study published in the January 2011 issue of JACC: Cardiovascular Interventions. Moreover, after the first year, the event rate is higher in patients who receive drug-eluting stents (DES) vs. bare-metal stents (BMS), but that DES disadvantage does not translate into increased mortality.

Investigators led by Bruce Brodie, MD, of the LeBauer Cardiovascular Research Foundation (Greensboro, NC), looked at 1,463 consecutive STEMI patients who underwent primary PCI with stenting at their institution from 1995 through 2009, dividing them into those who received BMS (n = 1,095) and DES (n = 368).

Academic Research Consortium-defined definite or probable stent thrombosis was observed in 107 patients; 65 cases occurred early (≤ 30 days after implantation) or late (31 days-1 year), while 42 occurred very late (> 1 year). All cases of very late stent thrombosis resulted in reinfarction, with STEMI tending to be more common in DES patients than BMS patients (84.6% vs. 55.2%; P = 0.08), and in 6 cases this reinfarction caused acute death (14.3%).

The incidence of stent thrombosis at 1 year was high and similar for DES (4.0%) and BMS (5.1%). Moreover, events continued to accumulate at a slow but steady pace for at least 4.5 years of follow-up with DES and at least 11 years with BMS. After 1 year, however, the average annual increase was greater among DES patients than BMS patients (1.9% vs. 0.6%). And a landmark Kaplan-Meier analysis found that beyond 1 year, rates of very late stent thrombosis, reinfarction, and reinfarction of the target vessel were more frequent in the DES group than the BMS group (table 1).

Table 1. Outcomes After 1 Year (Landmark Analysis)

^a According to Academic Research Consortium criteria.

Also in a landmark analysis, DES use was the only predictor of very late stent thrombosis (HR 3.79; 95% CI 1.64-8.79; P = 0.002) as well as a predictor of reinfarction (HR 2.01; 95% CI 1.15-3.50; P = 0.014) and reinfarction of the target vessel (HR 3.16; 95% CI 1.50-6.64; P = 0.002).

The researchers also analyzed separately data from the so-called DES era (2003-2009), in which patients received DES or BMS at operator discretion. There were no significant differences between this era and the overall study period for stent thrombosis or the other outcomes, even after 1 year of follow-up.

Selection Bias or Stent Type?

Dr. Brodie and colleagues say the finding that BMS thrombosis continues up to 11 years after implantation is somewhat unexpected. They note that thrombotic occlusion in BMS patients typically was associated with severe restenosis, indicating “new strategies might be needed to deal with this problem.”

In addition, the difference in very late stent thrombosis rates for DES vs. BMS in this registry contrasts with results of several randomized trials that found little or no difference between the stent types out to 3 to 5 years, the investigators note. It is unclear whether the current finding represents “a true difference or might be related to selection bias,” they write. “DES might be chosen for diabetic patients and patients with small vessels and long lesions, because of the higher risk of restenosis in these patients. This could increase the risk of [very late stent thrombosis] with DES.”

On the other hand, some data support the hypothesis that the frequency of very late stent thrombosis could be related to stent type, the researchers say. Late malapposition and poor healing are more common with DES in STEMI patients, they point out, and the relatively short follow-up and smaller sizes of randomized trials comparing DES and BMS in primary PCI may have obscured any differences.

The current data may have implications for patient management, Dr. Brodie and colleagues say. Optimizing stent deployment and using longer or more intensive antiplatelet therapies after primary PCI could curb the relatively high risk of very late stent thrombosis. Development of newer-generation stents that employ novel polymers or bioabsorbable scaffolds may also minimize this risk.

Jury Still Out for All-Comers

In an accompanying editorial, Khaled M. Ziada, MD, Richard Charnigo, PhD, and David J. Moliterno, MD, of the University of Kentucky (Lexington, KY), offer a meta-analysis of 7 randomized trials comparing DES and BMS in primary PCI with long-term follow-up involving more than 3,000 patients. They calculate an equivalent likelihood of stent thrombosis with both device types: OR 0.99 (95% CI 0.68-1.45).

The finding suggests that DES do not disproportionately influence stent thrombosis in this patient population, the editorialists say. Moreover, the multiple randomized trials showing reduction in TVR and no increase in stent thrombosis or mortality out to 5 years, “should reassure the interventional community and authors of future practice guideline documents,” they add.

Nonetheless, “the signal toward increased [stent thrombosis] in long-term registries raises concern about the use of DES in patients who are not typical of randomized trial cohorts,” the authors note, adding, “The jury may still be out on the use of DES for all-comers undergoing primary PCI.”

Finding of Very Late BMS Thrombosis New

“I think you can explain the differences in DES vs. BMS very late stent thrombosis by selection bias,” said David A. Cox, MD, of Lehigh Valley Hospital (Bethlehem, PA), in a telephone interview with TCTMD. “We know that when you have more diabetics [in the DES group] and you put in more stents in smaller vessels, your stent thrombosis rate is going to be higher. But the fact that you’re seeing very late stent thrombosis in the BMS patients at a slow but steady pace, that was intriguing to me. I don’t think anybody has reported that before.

“I can’t imagine that [improving] technical measures—like using more IVUS or post dilatation—is going to affect very late stent thrombosis,” he added.

These registry data are unlikely to change practice in regard to primary PCI, Dr. Cox said, explaining that they could support arguments for either stent type. “If you’re a person who likes BMS for STEMIs, then this paper would support your continuing [with that practice] as long as you’re willing to accept a higher restenosis rate,” he said. “On the other hand, if you like DES, this paper might also support you, because it’s saying that although you get a higher stent thrombosis rate, there’s no trade-off in mortality and a much lower revascularization rate.”

Letting Lesion Risk Be the Guide

In choosing a stent for primary PCI, “I think it probably comes down to what the HORIZONS-AMI study group found,” Dr. Cox commented. “Characteristics such as diabetes, smaller vessels, and longer lesions should prompt [clinicians] to think about using a DES—with the caveat that you are able to assess beforehand whether a patient is likely to be compliant with dual antiplatelet therapy.

“If we saw a higher stent thrombosis rate with DES [plus] a higher mortality rate, that would get everybody pretty concerned,” he continued. “But the fact that we’re not seeing that means that, at the very least, we want another report confirming these data, because the [stent thrombosis] numbers are very high.”

Jeffery W. Moses, MD, of Columbia University Medical Center (New York, NY), agreed, noting that the stent thrombosis rates in this registry “don’t conform to any other datasets.” Part of the reason may be a liberal definition of in-stent thrombosis or how angiograms are analyzed, he suggested in a telephone interview with TCTMD. Suboptimal DES technique—especially early on—may also have contributed to this finding, along with DES-delayed healing or inflammation, he said.

Magnitude of Difference Questioned

“There is an undeniable difference in very late stent thrombosis between DES and BMS,” Dr. Moses said. “The question is the magnitude. The concern was that as stent thromboses accumulated, they would offset the mortality and MI reduction seen with reduced TLR. But at least in this length of follow-up, you don’t see that.

“Registries are fraught with confounders, and the randomized data are out there 3 or 4 years now without any real alarms,” Dr. Moses observed, adding, “Frankly, [this registry finding] doesn’t seem to conform to our experience [at Columbia].”

Like Dr. Cox, Dr. Moses looks to HORIZONS-AMI for guidance about when to deploy DES. “You get the most net benefit in medium- to high-risk lesions, whereas in lower-risk lesions there is no major benefit, so why accept the disadvantage of a small increment of late [thrombotic] events?” he asked. Moreover, “if there are clinical unknowns in the rapid-fire situation of an acute MI, most clinicians will defer to a BMS,” Dr. Moses added.

Study Details

Between 1995 and 2002, only BMS were implanted, whereas between 2003 and 2009 patients received BMS or DES at operator discretion.

Patients receiving BMS were older, had more cardiogenic shock and triple-vessel disease, and were less apt to exhibit TIMI flow grade 2 or 3 on initial angiography, while those receiving DES were more likely to have diabetes, smaller vessels, and anterior wall infarction.

 

---

Sources:
1. Brodie B, Pokharel Y, Fleishman N, et al. Very late stent thrombosis after primary percutaneous coronary intervention with bare-metal and drug-eluting stents for ST-segment elevation myocardial infarction: A 15-year single-center experience. J Am Coll Cardiol Intv. 2011;4:30-38.

2. Ziada KM, Charnigo R, Moliterno DJ. Long-term follow-up of drug-eluting stents placed in the setting of ST-segment elevation myocardial infarction. J Am Coll Cardiol Intv. 2011;4:39-41.

 

 

Related Stories:

• Long-Term Data on DES Use in STEMI Patients Raise Concerns
• First-Generation DES Safe, Effective vs. BMS in STEMI Patients Over Long Term
• SES, PES Appear Safe, Effective vs. BMS in STEMI Patients