A July discussion of the blood clot (thrombus) issue pointed out by JNJ. The article was offered by Aslan on the Yahoo thread. Now, 3 months later comes a Dear Doctor letter from the FDA. Maybe that is not so long and indicates that followup of JNJ's July actions indicates that corrections have not been as complete as possible. I'll post the article by Leon (who I believe was one of the original researchers and perhaps has some affiliation with JNJ) and try to post the URL in edit. L.
tctmd.com
Are CYPHER Stents Associated with Increased Thrombosis? Perspectives from the Available Data
M. B. Leon, J. W. Moses
July 14, 2003
The FDA approval of the Cypher™ sirolimus-eluting coronary stent on April 24th, 2003 has been met with both excitement and controversy. The anti-restenosis benefits of Cypher™ are profound, consistent across all patient and lesion subgroups studied, and may transform the texture of catheter-based interventional therapies in the future. However, many unresolved issues remain, including guidelines for more widespread clinical applications, economic concerns, and the dissemination of appropriate operator techniques to ensure optimal safety and efficacy. These issues have been compounded by limitations in Cypher™ stent availability, resulting in chronic shortages of necessary quantities and sizes of stents to service the growing clinical demands. Most recently, safety concerns have been widely publicized and have added to the confusion and controversy surrounding this new and potentially breakthrough technology. Sound bites from a "Dear Colleague" letter circulated by Cordis (a Johnson & Johnson company) to all U.S. interventional physicians on July 7th have exploded onto the pages of The New York Times, The Wall Street Journal, and many other medical news forums, suggesting that the Cypher™ stent may be associated with increased thrombosis. The purpose of this editorial comment is to discuss the available data, share some perspectives, and to suggest guidelines for optimal use.
Historically, stent thrombosis (acute and subacute) in bare metal stents has been associated with patient, lesion, and procedural factors. Multiple previous studies have indicated that the frequency of stent thrombosis is increased in poorly deployed stents (underexpansion with incomplete stent strut to vessel wall apposition), multiple stent implantations, smaller vessels, multivessel stenting, after vascular brachytherapy, stents implanted in akinetic infarct zones, in patients who have surgical procedures within 1-2 months after stenting, and in patients who discontinue the assigned chronic antiplatelet regimens (ASA plus thienopyridines for one month with bare metal stents). For instance, the per patient subacute stent thrombosis rate in ARTS (multivessel bare metal stenting) was 3.2%, compared with the more familiar < 1%, often cited as the current standard in less complex patient cohorts. In previous animal studies with the Cypher™ stent, the frequency of stent thrombosis and the timecourse of re-endothelialization was similar to control Bx Velocity™ (bare metal) stents. Human data relating to stent thrombosis is derived from multiple sources: blinded, randomized clinical trials, large prospective registries, and the rapidly accumulating observational reports from U.S. clinical sites, post-FDA approval. These data sources differ in rigor (reporting accuracy and event adjudication), patient populations (simple vs. complex), and operator training and technique.
The Clinical Data...
There are four important blinded, randomized clinical trials (RAVEL, SIRIUS, E-SIRIUS, and C-SIRIUS) with Cypher™ stents, conducted both inside and outside the U.S., involving 1748 patients. Patients received one or two Cypher™ (or bare metal) stents in one coronary artery (2.5 mm – 3.5 mm diameter vessels) and they were treated with ASA plus clopidrogrel post-procedure for two or three months. The overall acute (within 24 hours), subacute (from 1 to 30 days), and late (after 30 days) stent thrombosis frequency was 0.6% for both Cypher™ and control bare metal stents.
The 317 site internet-based e-Cypher registry comprising over 6,500 patients is the most representative "real world" sample of Cypher™ stent treated patients. Thusfar, there have been seven cases of stent thrombosis in e-Cypher (0.1%) and > 70% of patients were treated with ASA and clopidogrel for two, three, or six months duration. The Rotterdam Cypher™ registry recently reported three-month follow-up of 510 consecutively treated Cypher™ stent patients. This registry includes many complex patients (25% multivessel stents, 42% overlapping stents, and 26% small vessels) and the combined acute and subacute thrombosis frequency was 0.4%. Finally, data from the Milan Cypher™ registry in 335 patients with "ultra-complex" disease (880 lesions, of which 766 were treated with Cypher™ stents), indicated an overall thrombosis frequency of 1.5% per patient and 0.7% per Cypher™ treated lesion. Of the 5 thrombotic events, all were in complex lesions, including bifurcations (1 patient), chronic total occlusions (1 patient), diffuse disease (> 6cm in length, 2 patients), and left main disease (1 patient).
Since approval of the Cypher™ stent in the U.S., slightly more than two months ago, there have been > 50,000 stents implanted in widely varying clinical circumstances at hundreds of clinical sites. At present, the reported frequency of acute stent thrombosis is 0.01% and subacute stent thrombosis is 0.07%. The mortality of patients with stent thrombosis was 12.9%. which is consistent with the 25% mortality usually reported after stent thrombosis in bare metal stents. Admittedly, in a voluntary post-market surveillance milieu, there may be "under-reporting" of these events, but even if the accurate event rate is 10 or 20 times greater, the per stent thrombosis frequency would be 0.8-1.6%.
Perspectives...
As best as can be determined from the available data, there does not appear to be a higher frequency of stent thrombosis associated with the Cypher™ stent, in either the rarified environment of randomized clinical trials, or in large prospective "real world" registries (often including complex patients), or in the open market U.S. release. Nevertheless, there persists a crescendo of rumors of event clustering in certain institutions which has evolved into an almost deafening outcry in the lay press. Is it possible that the actual Cypher™ stent thrombosis frequency during the current U.S. market release at particular clinical sites is increased beyond the < 1% standard? The answer is clearly yes! Are the causes of increased Cypher™ stent thrombosis understood and preventable? The answer is probably yes! Undoubtedly, the many factors responsible for increased stent thrombosis in bare metal stents have become exaggerated in the current climate of Cypher™ stenting. Cypher™ stents are being used preferentially in the highest restenosis risk patients, including diffuse disease, small vessels, bifurcations and multivessel stenting...all of which are predictors for increased stent thrombosis. Many operators are unfamiliar with important subtleties of Cypher™ stent implantation, such as the need for more careful lesion preparation, more frequent post-dilatation, longer stents to cover mild or moderate reference vessel disease, and confining the balloon injury zone to within the stent treatment zone, thereby avoiding vessel wall trauma to unstented regions. These technique issues are further distorted by the unnatural use of either undersized or oversized stents to accommodate chronic stent size shortages. Thus, we suspect that many undersized stents are inappropriately overexpanded, or paradoxically, left chronically underdeployed. Similarly, oversized stents implanted in smaller vessels, result in increased deep vessel injury and undiagnosed edge dissections. Such suboptimal stent deployment has historically been the single greatest contributor to increased stent thrombosis. Finally, the increased duration of combination anti-platelet therapy to three or more months has been confusing for some referring physicians and patients. Given the same set of circumstances – complex lesions, improper technique, suboptimal stent sizing, and relative non-compliance with anti-platelet therapy – the frequency of stent thrombosis in bare metal stents would be similarly elevated. It would seem that the supposed increased Cypher™ stent thrombogenicity resides more in the operator (technique and device selection) than in the stent!
Recommendations...
In the face of incomplete data, creative conjectures and presumed hypothetical scenarios, how should we approach the current frustrating dilemma? The following recommendations are an attempt restore the balance and provide sensible constructive suggestions to interventional physicians and institutions who remain concerned that some patients may be at increased risk after Cypher™ stent implantation.
Ongoing and more rigorous data collection is essential to better understand the true nature of the current situation. This requires a collaborative effort from Cordis, the FDA, and physicians who now must scrupulously report all events.
Safety is always more important than efficacy – restenosis is tolerable, but stent thrombosis is not – all efforts must be directed to minimize the frequency of stent thrombosis.
During the time of product shortages, interventional physicians should not compensate by using incorrect stent sizes for the intended anatomy! Expectations are that inventories will be restored to full capacity with all stent sizes in the next few months...this is a temporary situation!
Interventional physicians should completely understand the use characteristics of the Cypher™ stent, especially the recommended stent expansion ranges and optimal implantation pressures (at least 14 atm).
All other technique considerations including, meticulous attention to pre- and post-dilation strategies, and increased stent length to cover all inflow and outflow disease are essential to optimize stent deployment.
Concerned physicians should (for the time being) avoid complex lesions and fully adhere to the approved clinical indications...lesions <30 mm in length in vessels 2.5 – 3.5mm in diameter, excluding bifurcations, total occlusions, ostial lesions, instent restenosis, saphenous vein grafts, and acute MI or thrombus-containing lesions.
The increased use of intravascular ultrasound should be considered to clarify the target lesion and to verify optimal stent deployment (stent cross-sectional area and apposition).
Glycoprotein IIb/IIIa platelet inhibitors are recommended in complex lesions or in cases of intra-procedural complications and imperfect stent implantation results.
Strict compliance with recommended post-procedure anti-platelet regimens is mandatory – at least three months of combination ASA and clopidogrel therapy, with strong consideration of longer courses (up to one year), especially in higher thrombosis risk situations.
The introduction of potent new cardiovascular devices with great potential and high expectations to provide incremental patient benefits is never simple or without "dark side" considerations. Over-reaction to the recent claims of increased Cypher™ stent thrombosis are not supported by the facts and should not dampen our long-term enthusiasm for drug-eluting stents. However, these claims cannot be idly dismissed and require a thoughtful response by all concerned parties. We believe that the increased awareness engendered by the current publicity will result in improved interventional operator technique, more careful patient and device selection, and meticulous attention to intra- and post-procedural adjunctive pharmacology.
-------------------------------------------------------------------------------- |
