DES and Stent Thrombosis Revisited

SAN FRANCISCO, CALIF.—In the opening presentation of the DES Summit at TCT 2011, Edoardo Camenzind, MD, of the University Hospital of Geneva in Switzerland, a central figure in the series of reports that 5 years ago set off a firestorm of concern over an apparent increased mortality risk with DES due to late stent thrombosis, reviewed some of the data that sparked the controversy.

“For me, the essential issue is the appropriate understanding of current generation devices while we are looking for the best possible utilization,” Camenzind said. Current DES refer to those that are on the market as well as the latest iterations. Appropriate utilization, he asserted, depends on factors related to clinical presentation, the lesion, and the patient. Every DES is different, and it is important to be aware of each device’s specifications, Camenzind said. For example, first-generation DES are characterized by thick struts; a thick, stable polymer; and cytotoxic drugs. Based on the results of the RAVEL trial, among others, “we can say without any doubt that first-generation devices are potent,” Camenzind said. 

Relevant patient characteristics include the presence of diabetes and the number of diseased vessels. With regard to clinical presentation, Camenzind noted, unstable patients now represent more than half of stented cases. Based on the SYNTAX trial, lesions can now be assessed for key angiographic parameters that are related to prognostic value.

The data that sparked the firestorm

In the pooled analysis of the Cypher trials, rates of target lesion revascularization were only 7.8% in the sirolimus-eluting stent (Cordis) arm vs. 23.6% in the BMS arm at 4 years (P<.001).  The SES devices showed a trend toward a higher incidence of  stent thrombosis (SES: 1.2% vs. BMS: 0.6%). In addition, in the Bern-Rotterdam registry, the cumulative incidence of stent thrombosis was quantified at a mean of 0.6% per year up to 5 years after the first 30 days. In the literature, the associated mortality range was about 23% to 45%.

The key issue raised by these data was whether the persistence of stent thrombosis would lead to more MI and death in the real world, Camenzind said. RAVEL results presented at the World Congress of Cardiology in 2006 showed an excess incidence of death and MI with DES vs. BMS of 1.64% per year, while a pooled analysis of randomized trials presented showed an excess of 0.7% per year. Three-year data from BASKET-LATE revealed a difference of 0.87% per year.

For perspective, it is important to remember the progression of angioplasty from earlier days, Dr. Camenzind said. With the arrival of BMS, mortality was equivalent between stenting and surgery for up to 10 years.  After 4 years of SYNTAX, however, paclitaxel-eluting stents (Taxus, Boston Scientific) lost ground to surgery, with mortality rates slightly higher and MI rates significantly higher (0.3% vs. 1.5%; P=.01).

Metrics for assessing efficacy, inflammation

One major question is whether efficacy and safety are interrelated, Dr. Camenzind said, suggesting that the same morphological indicators that are used to assess efficacy can also be used to evaluate the presence of inflammation. From an angiographic point of view, the main index would be late loss, while from the standpoint of IVUS and OCT, it is late-acquired malapposition. In some patient populations, the use of DES results in negative late loss. However, this may reflect ongoing inflammation and delayed healing, with the potential for higher risk of death and MI. “A promisingly low or even negative angiographic measurement of late luminal loss may not be associated with a favorable long-term outcome,” Camenzind said, quoting his 2006 study.

Simple criteria can help predict poor clinical performance of DES, Camenzind suggested: at 6 to 9 months, strong efficacy or luminal widening or late ongoing late loss may be indicative of local inflammation, he asserted.

“From a pragmatic point of view, when we are using a device, we have to try to categorize it as being potent, intermediate, or ‘light,’” Camenzind said. The ongoing randomized PROTECT trial, which is comparing a ‘light’ zotarolimus-eluting stent (Endeavor, Medtronic) with the potent SES with a primary endpoint of ARC-defined definite or probable stent thrombosis at 3 years, may help.

Understanding DES technology and its biological effect is of utmost importance because of the substantial effect on long-term outcomes, Camenzind concluded. Equipotent DES are likely to have similar long-term effects. In the ultimate analysis, clinicians should strive for appropriate DES utilization, in the sense of selection of a stent for a particular clinical indication.