Author + information
- James T. Dove, MD, FACC, ACC President⁎
- ↵⁎Address correspondence to:
James T. Dove, MD, FACC, American College of Cardiology, c/o Padmini G. Rajagopal-Moorehead, 2400 N Street NW, Washington, DC 20037
While we were preparing the American College of Cardiology (ACC) statement regarding drug-eluting stents (DES) for the December 2006 U.S. Food and Drug Administration (FDA) Circulatory Systems Devices Panel meeting, it became evident that we had many gaps in our current knowledge of this new technology. There were no questions about the superiority of DES over bare-metal stents (BMS) in terms of reducing angiographic restenosis. However, the long-term safety of these devices is far less certain (1). A clear emerging theme from the deliberations is that there is a critical need for a better mechanism to systematically collect, collate, and share postmarket real-world data.
Stent thrombosis is an uncommon yet gravely serious risk after BMS and DES implantation and has commanded the attention of every interventional cardiologist. The risk of stent thrombosis attenuates relatively quickly following placement of BMS, but this is not the case with DES (2). The potential magnitude of this problem was brought to light by the publication of the BASKET-LATE study from Switzerland (3). The recent publication of several additional studies presented at the FDA hearing reinforces the concern about late stent thrombosis (LST) in DES-treated patients and the need for longitudinal data (4).
Within 6 months of approval, DES were used in approximately 80% of percutaneous revascularization procedures in the U.S. and BMS were relegated to large vessels, which have a low rate of restenosis with any type of percutaneous revascularization (5). Bare-metal stents also remain a first choice in those cases needing a short course of dual antiplatelet therapy, such as those with a planned near-term surgery or a history of bleeding (6).
Current real-world registries are inherently limited by a lack of valid control groups and often use historical controls. Efforts to compare LST among DES patients to data from older BMS registries have limited value, since the use of these stents is distinctly different from when they were first introduced. Most would agree that DES are implanted in patients with more comorbidities and complex lesions than would have been attempted in the BMS era. It is also recognized that many of the DES randomized trials performed for device approval had restricted enrollment criteria, making extrapolation of their findings to the greater population questionable. In addition, randomized trials and short-duration registries sponsored by industry may raise a concern of bias, whether real or perceived.
The current published literature for stents is also challenging to collate because of inconsistent criteria for enrollment, inconsistent definitions for acute stent thrombosis, LST and very late stent thrombosis (VLST), and clinical end points, and varied intervals of follow-up after stent implantation. A research consortium has proposed standard definitions for definite, probable, and possible stent thrombosis (7). Bare-metal stents remain a major advance in decreasing early angioplasty failures, and DES have an added target vessel revascularization (TVR) advantage, but the issue is the incidence of long-term complications (8).
The recent publication of the Bern-Rotterdam cohort experience suggests this higher rate may steadily continue over time (9). Beyond limited data and a low event rate, it is likely that the etiology of LST is multifactorial. One certain factor is the discontinuation of clopidogrel by many patients. Dual antiplatelet therapy has infrequently been continued beyond 1 year (10). The optimal duration of dual therapy is unknown, and whether dual therapy is protective from LST is uncertain (11). It is known that the risk of stent thrombosis increases with premature discontinuation of dual antiplatelet therapy. A recent multisociety statement stressed the importance of 12 months of dual antiplatelet therapy after placement of a DES in patients at low risk for bleeding (12).
The risk of LST when stopping clopidogrel before surgery is unclear. Conversely, the risk of bleeding associated with continuing dual therapy for noncardiac surgery is also unknown. Many studies have identified potential patient characteristics such as disease comorbidities, lesion types, platelet resistance, and compliance with therapy as independent contributors to stent thrombosis. Stent design, delayed endothelialization, and deployment issues may be added contributors (13).
While it is useful to follow patients in postapproval, industry-sponsored registries, it would be of greater value to follow real-world application of pharmaceuticals and devices in a much broader and larger population of patients. The Society for Thoracic Surgeons (STS) and the American College of Cardiology National Cardiovascular Data Registry (NCDR™) databases provide excellent platforms on which to build this capability and further our understanding of the role of intensive medical therapy, percutaneous coronary intervention, and coronary artery bypass grafting. The NCDR™ data registries have more than 5 million patient records and add nearly 250,000 patients each quarter. There are 3 million patients in the STS data registries. Currently, all Medicare patients who receive an implanted cardioverter-defibrillator (ICD) require entry in the NCDR’s ICD Registry™, and it is an option for non-Medicare patients. Since there is mandatory reporting for Medicare patients, about 85% of all ICD implants are reported to this registry.
A strategy for longitudinal follow-up is under consideration to better define devices used, complications, and factors influencing clinical risk and benefit. This may help improve our current ability to treat those who have the most to gain and the least to lose. A longitudinal database for other devices, procedures, and pharmaceuticals would be invaluable.
What can the ACC do to help? A large-scale longitudinal NCDR™ database for patients receiving coronary artery stents needs to be created and funded. This database should have open-entry to fully capture all U.S. procedures and have unique patient identifiers to protect confidentiality and facilitate the tracking of patients admitted to multiple institutions. Standardized definitions and systematic intervals of follow-up must be created and agreed upon. Postrelease information on devices and pharmaceuticals used in contemporary practice is an essential component of quality patient care. Appropriately, funding such a database is crucial and should not be an obstacle, provided manufacturers are allowed to shift funding from otherwise required approval costs, postrelease surveillance, and registries. Such a database could:
• Capture key elements of patient demographics and management at the point of entry
• Identify specific devices and pharmaceuticals used
• Capture in-hospital complications and outcomes
• Collate deaths from social security database or state-maintained records
• Identify patient, hospital, and physician characteristics that contribute to outcomes
• Confirm discharge status, management, and instructions
• Follow ambulatory compliance with practice guidelines as well as patient compliance
• Identify cohorts that benefit from specific therapies and those who do not
• Follow both complications and freedom from events for devices and pharmaceuticals in the general population
• Avoid some limitations associated with small sample size registries that confound the ability to identify infrequent complications
• Quantify faster and more accurately low frequency adverse events
• Reduce selection and reporting bias
Perhaps the next-generation stents with changes in stent design, polymers, nonpolymers (5), drug delivery, and new antiproliferatives will improve stent delivery and decrease stent thrombosis. Only when we are able to follow patients longitudinally will we be able to better understand the long-term effects of our new treatments. With these added data, physicians will be able to provide their patients with better and more complete information on short- and long-term risks and benefits. When a patient presents with an acute coronary syndrome, the interventionalist will have better data on which to base management decisions. There needs to be a discussion with the patient and his or her family about the treatment choices including the lack of data and the concerns about LST. Informed consent is important for all patients and essential in the off-label use of DES.
The device manufacturers, pharmaceutical companies, and FDA need to support longitudinal data collection built on the foundation of the ACC/NCDR and STS databases to help identify those who will likely benefit the most or who are at risk for harm from a particular procedure or pharmaceutical. A comprehensive database will help determine future recommendations and also uncover new problems. The DES story highlights the current weaknesses in our device surveillance programs and the pharmaceuticals that are used in these patients. We need to better understand the risks and effects of our new treatments.
The authors wish to acknowledge the work of Linda Bell in the preparation of this document.
- American College of Cardiology Foundation
- Pfisterer M.,
- Brunner-La Rocca H.P.,
- Buser P.T.,
- et al.
- ↵Summary from the Circulatory System Devices Panel Meeting, December 7-8, 2006. U.S. Food and Drug Administration. Available at: http://www.fda.gov/cdrh/panel/summary/circ-120706.html. Accessed December 12, 2006.
- Spertus J.A.,
- Kettelkamp R.,
- Vance C.,
- et al.
- Bliden K.P.,
- DiChiara J.,
- Tantry U.S.,
- et al.
- Grines C.L.,
- Bonow R.O.,
- Casey D.E.,
- et al.
- Joner M.,
- Finn A.V.,
- Farb A.,
- et al.