CREST Substudy: Even Biomarker-Only MIs Raise Long-term Mortality

Download this article's Factoid (PDF & PPT for Gold Subscribers)

Patients who undergo carotid endarterectomy (CEA) are more likely to experience periprocedural myocardial infarction (MI) or elevated cardiac biomarkers alone than those who receive carotid artery stenting (CAS), according to a new analysis of the CREST trial published online May 23, 2011, ahead of print in Circulation. Moreover, both cardiac events are linked to increased long-term mortality, suggesting that cardiovascular risk should be factored into any decision regarding revascularization strategy.

In the main CREST (Carotid Revascularization Endarterectomy Versus Stenting) trial, 2,502 symptomatic and asymptomatic patients with significant carotid stenosis were randomized to CAS or CEA. Rates of the primary composite endpoint—stroke, MI, or death during the periprocedural period or any ipsilateral stroke within 4 years—were similar for the 2 groups (7.2% vs. 6.8%; P = 0.51). On the other hand, stroke was more common among stenting patients (4.1% vs. 2.3% for CEA; P = 0.01), while MI was more frequent among surgical patients (2.3% vs. 1.1% for CAS; P = 0.03).

Both Clinical and Biomarker-Only MIs Included

To explore the prognostic importance of MIs, investigators led by Thomas G. Brott, MD, of the Mayo Clinic (Jacksonville, FL), conducted a substudy analyzing outcomes for patients who had either clinical MI or MI defined by biomarker elevation only. The cohort included 42 patients with adjudicated clinical MIs, of which 14 occurred in the CAS group vs. 28 in the CEA group (HR 0.50; 95% CI 0.26-0.94; P = 0.032). In addition, 20 patients had biomarker-positive only events: 8 in the CAS group vs. 12 in the CEA group (HR 0.66; 95% CI 0.27-1.61; P = 0.36).

Overall, the 62 patients who experienced either type of MI were older and had a greater frequency of prior cardiovascular disease and reduced creatinine clearance than those without MI. Multivariable analysis showed that the only independent predictor of clinical MI was a history of cardiovascular disease or CABG (HR 2.22; 95% CI 1.12-4.35; P = 0.02). Using the broader definition of either clinical or biomarker-only MI, predictors included both prior cardiovascular disease or CABG (HR 1.73; 95% CI 1.02-2.95; P = 0.04) and creatinine clearance below 30 mL/min (HR 2.97; 95% CI 0.97-9.05; P = 0.06), although the latter relationship did not reach statistical significance.

In patients with clinical MIs, the median biomarker ratio—an indication of MI size—was 40, while in patients with elevated biomarkers only, the median ratio was 14. There were no differences in peak biomarker ratio for patients who received CAS compared with CEA.

During 2.5-year follow-up (range 1-4 years), there were 177 deaths, with an estimated 4-year mortality of 7.1%.

In both unadjusted and adjusted analyses, patients who experienced either clinical or biomarker elevation only MI were over 3 times more likely to die during follow-up than patients who had neither event (table 1).

Table 1. Long-term Mortality Risk After MI


HR (95% CI)

P Value

Clinical MI

3.40 (1.67-6.92)
3.67 (1.71-7.90)

< 0.001

Biomarker-Positive Only MI

3.57 (1.46-8.68)
2.87 (1.16-7.14)


Additional adjustment for sex and symptomatic status had little effect on estimated risk differences. Among clinical MI patients, there were 5 deaths in the CEA group and 3 deaths in the CAS group. For isolated biomarker patients, there were 2 deaths in the CEA group and 3 deaths in the CAS group. None of the differences were significant.

The authors observe that CREST was unique among carotid revascularization trials because it not only included both symptomatic and asymptomatic patients but also systematically collected biomarker and other clinical data for adjudication of periprocedural MI. And even though the degree of biomarker rise was relatively small compared with spontaneous MIs, the presence of either MI or biomarker elevation only was linked to increased mortality risk.

Should CAD, Renal Dysfunction Influence Revascularization Strategy?

Although baseline patient factors such as prior CAD or renal dysfunction were found to contribute to the risk for periprocedural MI, “[i]t is premature to speculate whether these data should be used to guide patient selection for CAS versus CEA in clinical practice,” Dr. Brott and colleagues write. “Nonetheless, it seems advisable to implement or test strategies for the prevention of MI in these higher-risk patients, especially if selected to undergo CEA.” These approaches might include high-dose statins or more robust dual antiplatelet therapy, they suggest.

The authors acknowledge several limitations to the study, including the small number of deaths among those who had either type of MI and the absence of a central core laboratory for biomarker analysis. In addition, a longer hospital stay among surgical patients may have increased the number of biomarker elevations detected in that group.

In an accompanying editorial, Scott Kinlay, MBBS, PhD, of Harvard Medical School (Boston, MA), writes that among patients who clearly need revascularization, “carotid stenting will be favored in patients with higher periprocedural surgical risk and endarterectomy will be favored in patients at high periprocedural risk from stenting.”

Importantly, “both modes of revascularization require operators who are adequately trained, credentialed, and audited for low periprocedural adverse events,” he adds.

Stroke vs. MI: A Phony Debate 

In a telephone interview with TCTMD, Christopher J. White, MD, of the Ochsner Heart and Vascular Institute (New Orleans, LA), said the finding that “heart attacks are bad for you” is pretty obvious in light of many prior studies linking MI after vascular surgery to mortality. The real added value of this report is the inclusion of biomarker elevation in that conclusion, he said.

When the main CREST results were first published, Dr. White explained, many surgeons were quoted as saying that having a stroke—the predominant complication with CAS—is worse than having an MI—the predominant complication with CEA. In particular, “the surgeons jumped on the term ‘enzyme leak,’ which tends to minimize the injury to the myocardium,” he said. “You can make it sound like not such a big deal, but [this paper shows that] it kills you at the same rate as a full-blown MI.”

William A. Gray, MD, of Columbia University Medical Center (New York, NY), agreed. He told TCTMD in a telephone interview that, in the context of the controversy over the relative merits of CAS vs. CEA, the link between MI and mortality is not as important as the fact that the finding shores up the equipoise between the 2 treatment groups.

Dr. White emphasized that arguing over which complication is worse—stroke or MI—misses the point, which is to strive to minimize all complications. In that regard, he said, some patients are better candidates for surgery (eg, if they have tortuous arteries or heavy calcification) and some for stenting (eg, if they have congestive heart failure or prior neck surgery or radiation).

“We really need to individualize treatment, and we need the freedom as physicians to make those decisions,” Dr. White contended. “But today we don’t have that freedom because Medicare won’t pay for a lot of stenting.”

This paper is largely aimed at encouraging “payers to allow patients to have access to both surgery and stenting,” he said, adding, “My job is to integrate as much information as I can about what the patient wants, what their condition is, and what their risk is. All I would ask is that I have as many tools at my disposal as possible so that I can take the best care of my patients.”

In a similar vein, Dr. Gray underlined that the 2 revascularization strategies are complementary. “Fortunately, it’s unusual for one patient to have risks for both procedures,” he said. “The major message is: This is a choice between 2 very good therapies, both of which prevent stroke in the long term and both of which are done at very low risk in the short term in skilled hands.” 

Study Details 

Cardiac biomarkers were collected before revascularization in 90% of patients and at 6 to 8 hours after the procedure in 88%. Troponin only was obtained in 37%, troponin with CK plus CK-MB in 23%, and CK plus CK-MB without troponin in 30%. Cardiac biomarkers were analyzed at local-site laboratories, yielding a mixture of biomarkers and definitions.


1. Blackshear JL, Cutlip DE, Roubin GS, et al. Myocardial infarction after carotid stenting and endarterectomy: Results from the Carotid Revascularization Endarterectomy Versus Stenting trial. Circulation. 2011;123:2571-2578.

2. Kinlay S. Fire in the hole: Carotid stenting versus endarterectomy. Circulation. 2011;123:2522-2525.



Related Stories:

CREST Substudy: Even Biomarker-Only MIs Raise Long-term Mortality

Patients who undergo carotid endarterectomy (CEA) are more likely to experience periprocedural myocardial infarction (MI) or elevated cardiac biomarkers alone than those who receive carotid artery stenting (CAS), according to a new analysis of the CREST trial published online May
  • The trial was supported by the National Institute of Neurological Disorders and Stroke and the National Institutes of Health with supplemental funding from Abbott Vascular Solutions (formerly Guidant).
  • Dr. Kinlay reports no relevant conflicts of interest.
  • Dr. White reports having served as principal investigator of the CABANA study, which was sponsored by Boston Scientific.
  • Dr. Gray reports having served as a principal investigator for CREST.