Carotid Stenting Postmarket Studies May Not Reflect Real World

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

A comparison of patients enrolled in postmarketing surveillance studies following carotid artery stenting (CAS) vs. those not enrolled reveals significant differences in patient characteristics and outcomes. The findings, published online March 21, 2011, ahead of print in Circulation, challenge the assumption that the results of such studies are representative of the larger population of patients undergoing CAS.

Using data from the National Cardiovascular Database Registry-Carotid Artery Revascularization and Endarterectomy (NCDR CARE) registry, researchers led by Robert W. Yeh, MD, MSc, of Massachusetts General Hospital (Boston, MA), compared characteristics and outcomes of 7,115 patients who underwent CAS between January 1, 2005, and October 31, 2009 and either were (n = 3,689) or were not (n = 3,426) enrolled in postmarketing surveillance studies.

Overall, postmarket study participants had lower rates of combined in-hospital death, stroke, or MI, driven by lower stroke and death rates (table 1).

Table 1. In-Hospital Outcomes by Postmarket Surveillance Participation


(n = 3,689)

Not Enrolled
(n = 3,426)

P Value

Combined In-Hospital Death, Stroke, or MI



< 0.001

New Stroke







< 0.001

After propensity score adjustment based on patient, procedural, and hospital variables, there was no significant difference in the combined endpoint between study participants and nonparticipants (OR 0.76; 95% CI 0.55-1.05; P = 0.10). However, in-hospital and all-cause mortality continued to be lower in study participants compared with nonparticipants (OR 0.44; 95% CI 0.21-0.95; P = 0.04 for in-hospital mortality; and HR 0.80; 95% CI 0.66-0.97; P = 0.02 for mortality at 2 years).

Baseline Characteristics Help Explain Outcome Disparity

Baseline differences in neurological history explained the largest proportion of the difference in outcomes between the groups. Postmarketing study participants had significantly lower rates of prior stroke (13.7% vs. 19.2%; P < 0.001), symptomatic target lesions (31.5% vs. 58.2%; P < 0.001), and acute evolving stroke (1.0% vs. 3.9%; P < 0.001). A comparison of procedural characteristics showed that study participants also had significantly higher rates of attempted and successful embolic protection device deployment, among other differences.

Hospital characteristics (teaching status, geographic region, profit type, urban vs. rural, and public vs. nonpublic status) did not account for any of the observed difference in outcomes between participants and nonparticipants.

According to Dr. Yeh and colleagues, the reason for the differences in neurological history between the 2 groups is not clear. However, they hypothesize that acute evolving stroke may have precluded some patients from participating in postmarketing surveillance or that symptomatic events and worse National Institutes of Health Stroke Scale scores may have rendered some patients less able to give the informed consent necessary for participation.

Findings a Surprise

In a telephone interview with TCTMD, Dr. Yeh said the findings were somewhat unexpected.

“When we started this project, our initial goal was actually to compare event rates after CAS in premarket randomized studies to those in the broader community, expecting that data from such randomized controlled trials would not truly be generalizable to routine practice,” he said. “When we saw the signal that, in fact, postmarket studies perhaps suffered this same limitation in generalizability, we were quite surprised, as the presumption of all of us had been that such studies would be representative of community practice, as is their intent.”

Dr. Yeh said he suspects that many people “may be surprised because they look to postmarketing studies as examples of how devices or other therapeutics are doing in real- world populations.”

Overall, postmarketing studies are “one piece of the puzzle” for studying device safety but should be interpreted carefully, he said, adding, “Our results suggest that we need to be cautious in assuming that these [post marketing surveillance] results will be reflective of the results we can expect to see as CAS disseminates more broadly.”


Yeh RW, Kennedy K, Spertus JA, et al. Do postmarketing surveillance studies represent real-world populations? A comparison of patient characteristics and outcomes after carotid artery stenting. Circulation. 2011;123:1384-1390.



Related Stories:


  • The study was supported by the American College of Cardiology Foundation’s NCDR.
  • Dr. Yeh reports being an investigator for the Harvard Clinical Research Institute and a consultant for the Kaiser Permanente Division of Research.