Data presented at the 2006 European Society of Cardiology (ESC) meeting suggested that drug-eluting stents (DES) might not be as safe as bare-metal stents (BMS). Now, it appears the tide has changed. Why?
Concerns about the safety of DES were initially prompted by data from studies such as the Swedish Coronary Angiography and Angioplasty Registry (SCAAR), the Bern-Rotterdam Registry, and two meta-analyses assessing somewhat unconventional endpoints which suggested DES may cause harm. Since that time, an abundance of data — from both randomized clinical trials and registries — has been published and presented that has been largely reassuring, including a re-analysis of SCAAR using more updated data which observed no increase in mortality with DES. In total, these studies do not suggest an excess of events such as mortality or myocardial infarction (MI) occurring with DES compared to BMS.
The SCAAR and Bern-Rotterdam studies additionally brought to light the phenomenon of late-stent thrombosis. While there has been a growing appreciation of the risk of late-stent thrombosis since this time, most studies have shown that the incidence of late-stent thrombosis is relatively low overall. In addition, there are benefits associated with DES when used within their approved indications — particularly in terms of reducing the rate of repeat revascularization procedures — that seem to outweigh the risks, particularly given that repeat revascularization due to restenosis occurs not infrequently with BMS.Describe the design and objective of your analysis of DES and BMS presented at this year’s American College of Cardiology (ACC) meeting.
We wanted to synthesize the vast amount of data comparing DES and BMS in order to systematically determine whether there was any safety or efficacy signals with either type of stent (DES or BMS). We searched the published and presented literature to identify all DES and BMS outcomes studies. Studies had to include 100 patients or more, and had to report outcomes continuously for one year or more. We identified more than 20 randomized clinical trials (enrolling 9,470 patients) and 30 registries (enrolling 174,302 patients) that met our inclusion and exclusion criteria. These studies were combined using a technique known as meta-analysis, which weights each study to calculate an average estimate of safety and efficacy rates across studies. It is important to note that we performed separate analyses of randomized clinical trials and registries. To the best of our knowledge, this is the largest meta-analysis comparing DES and BMS to date.
What were the findings of this meta-analysis?
The findings were two-fold. First, there did not appear to be an adverse safety signal associated with DES in either the randomized trials or the registries. In fact, we observed reductions in mortality and MI with DES among the registry studies. The reductions that we observed were 20 percent for mortality, and 11 percent for MI. It is important to note that the actual magnitude of the benefit that we observed should not be the take-home message from this analysis, because the comparison between DES and BMS in these studies was not randomized.
Second, we found a significant reduction (approximately 50%) in terms of reducing repeat revascularization for DES both in randomized trials and in registry studies. This finding is perhaps equally important to the safety finding, because some have criticized the design of randomized clinical trials of DES as biased in favor of these stents through trial design characteristics such as the implementation of routine angiographic follow up. In our analysis, we found that the effects seen in the randomized trials were parallel to the findings in registries.
What was the duration of Plavix® used in these registries?
It is likely varied among the different registries, because not all registries actually report the duration of dual anti-platelet therapy. Most registries that were conducted in the early DES era were only using anti-platelet therapy for three to six months, and the later registries may have used it for longer periods. Further, the BMS patients in registries were almost certainly using dual anti-platelet therapy for shorter periods of time. Therefore, I can’t answer that question using our summary-level analysis that we conducted, but I would estimate that in these trials dual anti-platelet therapy generally was used for one year or less.
How do you expect the current guidelines for one year of dual anti-platelet therapy to support the safety of DES?
Because of the late-stent thrombosis issue, we tend to keep our patients on anti-platelet therapy longer than with BMS, and that was the spirit of the joint committee guidelines for one year of dual antiplatelet therapy. Unfortunately, there is limited — and conflicting — data addressing the issue of prolonged dual anti-platelet therapy. For example, whereas one publication (Eisenstein et al, JAMA 2007) suggested that continuation of dual anti-platelet therapy might be better, others have suggested that stopping dual anti-platelet therapy at six months does not appear to be associated with higher risk of late-stent thrombosis over long-term follow up. In the absence of a definitive trial, we do not really know the optimal duration of dual anti-platelet therapy, and therefore clinicians have been advised to follow the current guidelines (ACC/AHA/SCAI PCI Practice Guidelines) that state dual anti-platelet therapy be administered for at least 12 months for all DES.
Is there any new data coming out on this topic of DES safety?
A large randomized trial called the HORIZONS AMI trial, sponsored by the Cardiovascular Research Foundation, will address the issue of DES use in the setting of acute MI, which many clinicians feel puts patients at the highest thrombotic risk. In this study, we prospectively compare DES to BMS in AMI patients. The initial primary endpoint data regarding the efficacy of DES in reducing clinical restenosis will be presented at this year’s Transcatheter Cardiovascular Therapeutics (TCT) meeting, but as far as safety issues are concerned, we need longer-term follow up from this trial.
Has the recent DES vs. BMS data affected your practice and/or DES adoption at your institution?
Not really. Our group has historically had a lot of experience with DES. We use DES in approximately 80 percent of our cases at Columbia and we systematically collect outcomes on all patients treated with DES in our cath lab. We have seen cases of late-stent thrombosis, but in terms of our overall rate, it is quite low. In light of the initial data from SCAAR and the Bern-Rotterdam analyses, we thought that the best way to assess whether there was overall harm or benefit was to conduct rigorous analyses. That was one of the motivations behind the current analysis. Additionally, we have presented data from our MATRIX registry which is registered with the US FDA, and we conducted one of the original meta-analyses of randomized, controlled trials that was published in the New England Journal of Medicine at the same time as the SCAAR registry. Data from this meta-analysis showed that overall rates of death and MI were not higher with DES. Our conclusion was that there did not appear to be adverse safety signals with DES, but DES were efficacious.
Plavix is a trademark of Sanofi-Aventis.
The HORIZONS AMI Trial is co-funded by Boston Scientific.
Sponsored and prepared by Boston Scientific Corporation