CAS Postmarketing Studies May Not Represent Real-World Populations

March 21, 2011—Robert W. Yeh, MD, et al conducted an investigation to determine if the postmarketing surveillance (PMS) studies mandated by the US Food and Drug Administration for approved carotid artery stenting (CAS) systems are representative of routine clinical practice. The findings were published online ahead of print in Circulation.

In the study, the investigators compared patient and procedural characteristics, in-hospital outcomes, and subsequent all-cause mortality after CAS in PMS study participants and nonparticipants within the National Cardiovascular Database Registry–Carotid Artery Revascularization and Endarterectomy Registry. Both crude and propensity score–adjusted comparisons for all outcomes between groups were conducted.

The investigators found that participants in PMS studies for CAS have different clinical and procedural characteristics and lower mortality compared with nonparticipants. Therefore, extrapolating results from PMS studies of CAS to larger real-world settings should be done only with great caution, concluded the investigators.

As detailed in Circulation, the investigators reported that compared with nonparticipants, participants in PMS studies had lower rates of symptomatic carotid artery disease within the preceding 6 months, previous stroke, and acute evolving stroke at baseline. The PMS study participants had lower unadjusted rates of combined in-hospital death, stroke, or myocardial infarction (2.3% vs 4.1%; P < .001), driven by lower rates of stroke (1.7% vs 2.7%; P = .005) and death (0.3% vs 1.4%; P < .001). Differences in survival persisted after propensity score adjustment (odds ratio, 0.44; 95% confidence interval, 0.21 to 0.95; P = .04 for in-hospital mortality; and hazard ratio, 0.80; 95% confidence interval, 0.66 to 0.97; P = .02 for 2-year mortality). Baseline differences in neurological history explained the largest proportion of the difference in outcomes between groups.