CME/CEU Activity:Clearing the Fog of Clinical Trial Results: What is the Best Measurement to Compare Stent Effectiveness?

James P. Zidar, MD, FACC, FSCAI, Associate Professor of Medicine, Division of Cardiology, Duke University Medical Center, Durham, North Carolina; Richard E. Kuntz, MD, MSc, Associate Professor of Medicine, Harvard Medical School, Brigham & Women's Hospital, Boston, Massachusetts
James P. Zidar, MD, FACC, FSCAI, Associate Professor of Medicine, Division of Cardiology, Duke University Medical Center, Durham, North Carolina; Richard E. Kuntz, MD, MSc, Associate Professor of Medicine, Harvard Medical School, Brigham & Women's Hospital, Boston, Massachusetts
The following activity is supported by an unrestricted educational grant from Cordis Corporation. This presentation contains discussion of published and/or investigational uses of agents that are not indicated by the FDA. Neither HMP Communications nor Cordis Corporation recommends the use of any agent outside of the labeled indications. Please refer to the official prescribing information for each product for discussion of approved indications, contraindications and warnings. Topics: Clearing the Fog of Clinical Trial Results: What is the Best Measurement to Compare Stent Effectiveness? Are Drug-Eluting Stents Important? Faculty/Credentials: James P. Zidar, MD, FACC, FSCAI, Associate Professor of Medicine, Division of Cardiology, Duke University Medical Center, Durham, NC; Richard E. Kuntz, MD, MSc, Associate Professor of Medicine, Harvard Medical School, Brigham & Women's Hospital, Boston, MA Learning Objectives. At the conclusion of this activity, the participant should be able to: 1. Describe the time course issues in comparing drug-eluting and bare metal stent trials; 2. Name the three most important determinantsof restenosis for bare metal stents; 3. Discuss the relevance of stent strut thickness for restenosis; 4. Describe the concept of late loss. Activity instructions. Successful completion of this activity entails reading the article, answering the test questions and obtaining a score of over 70%, and submitting the test and completed evaluation form to the address listed on the form. Tests will be accepted until the expiration date listed below. A certificate of completion will be mailed to you within 60 days. Estimated time to complete this activity: 1 hour Initial release date: August 31, 2003 Expiration date: August 31, 2004. Target audience. This educational activity is designed for physicians, nurses and cardiology technologists who treat patients with coronary artery disease. Accreditation statements. This activity is sponsored by HMP Communications. Physicians: HMP Communications is accredited by the Accreditation Council for Continuing Medical Education to provide continuing medical education for physicians. HMP Communications designates this continuing medical education activity for a maximum of 1 category 1 credit toward the AMA Physician’s Recognition Award. Each physician should claim only those credits that he/she actually spent in the educational activity. This activity has been planned and produced in accordance with the ACCME Essential Areas and Policies. Nurses: Provider approved by the California Board of Registered Nursing, Provider Number 13255 for 1 contact hour. Radiologic Technologists: Activities approved by the American Medical Association (AMA Category 1) are eligible for ARRT Category B credit as long as they are relevant to the radiologic sciences. Radiologic Technologists, registered by the ARRT, may claim up to 12 Category B credits per biennium. SICP: Society of Invasive Cardiovascular Professionals (SICP) approved for 1 CEU. Commercial support disclosure. This educational activity has been supported by an educational grant from Cordis Corporation. Faculty disclosure information. All faculty participating in Continuing Education programs presented by HMP Communications are expected to disclose to the meeting audience any real or apparent conflict(s) of interest related to the content of their presentation. Dr. Zidar disclosed that he has received a research grant, is a member of the speakers' bureau and/or is a consultant for Medtronic, Cordis, Guidant and Abbott Laboratories. Dr. Kuntz disclosed that he is a consultant for Cordis. Clearing the Fog of Clinical Trial Results: What is the Best Measurement to Compare Stent Effectiveness? Are Drug-Eluting Stents Important? James P. Zidar, MD, FACC, FSCAI, Associate Professor of Medicine, Division of Cardiology, Duke University Medical Center, Durham, North Carolina; Richard E. Kuntz, MD, MSc, Associate Professor of Medicine, Harvard Medical School, Brigham & Women's Hospital, Boston, Massachusetts More than thirty different coronary stents have been developed for use in the United States and Europe over the past decade. Over 900,000 percutaneous coronary intervention procedures alone were performed in the U.S. last year, and stents were used in well over 80% of those procedures. Stents have converged in their cell design primarily as stainless-steel, balloon-expandable stents. They have differed significantly in secondary characteristics that are important to clinicians: optimal scaffolding, flexibility upon delivery, and conformability of the stent once it is deployed. Newer stents have resulted in lower profiles, better ability to track distal vessels, and are available in multiple sizes and lengths with less balloon overhang on the stent edge. Despite these advances, over 50,000 US patients each year develop recurrent symptoms due to the process of clinical restenosis. The questions we are addressing today are whether new designs or new metals can affect restenosis rates and late loss. In other words, is drug elution a requirement? The process of restenosis is well documented and has been reviewed on many occasions. From the era of balloon angioplasty, it is important to obtain a large residual lumen. Yet durable results from balloon angioplasty have been limited by elastic recoil and negative arterial remodeling. To solve this problem, bare metal stents were invented. Stents have done an excellent job of overcoming these issues. Unfortunately, metal stents come with a caveat increased injury to the vessel. In many cases, an exuberant healing response within the stent occurs, referred to as neointimal proliferation or hyperplasia. None of the bare metal stents, despite new design changes, have been able to overcome neointimal hyperplasia. The hypothesis is that drug-eluting stents (DES) will make a significant impact on restenosis by altering the pathway of neointimal hyperplasia and result in long term clinical benefits for patients. Angiographic & Clinical Restenosis Restenosis is broadly divided into two broad categories: angiographic and clinical restenosis. The more stringent criteria of angiographic restenosis has been classically defined as the binary definition of >50% residual stenosis at the site of injury. The new world of drug-eluting stents has indicated a necessary change in how we define angiographic restenosis. Historically, the definition was confined to the area within the margins of the stent. However, the profound effects of the new drug-device combinations have pressed us to include 5 mm at either stent edge, creating the term in-segment restenosis, which includes the area in-stent and 5mm proximal and distal to the stent. Ultimately, angiographic restenosis may or may not be clinically relevant, since physicians do not perform angiographic follow up on every patient. Clinical restenosis is the most important issue, and there are many ways to define clinical restenosis. The most popular are target lesion revascularization (TLR) or target site revascularization (TSR). TLR is defined as revascularization at the lesion where the initial intervention occurred. In some trials, one occasionally also refers to the more broadly based target vessel revascularization (TVR), which often includes the entire length of the vessel. Three Predictors of Restenosis A German group led by Kastrati followed over three thousand patients who had received Palmaz Schatz (PS) bare metal stents and reviewed their TLR in clinical practice. They found three essential predictors of restenosis: 1. Presence of diabetes 2. Lesion length 3. Vessel size The presence of diabetes conferred a 50% higher risk for clinical restenosis. The presence of multiple stents, a result of lesion length [at that time there was only one length of the PS stent (15mm)] and small vessels (minimal lumen diameter 3 mm) vessel will have a low 7% risk of clinical restenosis (see chart 1, top left-hand side). There is a 24% clinical restenosis risk for a non-diabetic with a long lesion in a small (Restenosis Rates at 6 Months vs. 9 Months Another important parameter is the time course of clinical restenosis. Figure 1 reviews data from the STARS trial, a large, multi-center trial looking at three arms of anti-coagulation (aspirin, aspirin/warfarin, and aspirin/ticlopidine) in the setting of ideal stent outcomes with the Palmaz-Schatz stent. STARS sampled clinical restenosis every three months for the first year. Angiographic follow up was not designed into this trial, eliminating its potential for bias. The 6.6% rate of clinical restenosis at 6 months nearly doubled to 11% at one year. As one compares bare metal stent data to drug-eluting stent (DES) data, we move the follow-up assessment from 6 to 9 months. This complicates trial comparisons of new bare metal stent registries at 6 months versus new DES trials at 9 months. Figure II reviews data published in JACC by Dr. Cutlip and colleagues at the Harvard Clinical Research Institute (HCRI). They reviewed data from 9 randomized stent trials and noted a similar phenomenon. From 6 months to one year, one sees an increase in TLR, TVR, and target vessel failure (TVF). With TLR as the benchmark, the 8% risk of clinical restenosis at 6 months increased to 12% at one year, confirming the concept that restenosis rates increase with time. Strut Thickness & Restenosis Recently, the impact of strut thickness on restenosis has become apparent. In an effort to address this issue, Dr. Kalon Ho et al (Harvard) reviewed his HCRI stent database. He performed an analysis assessing strut thickness. His trial data included very thin stents like the ACS Multi-Link (.0022) to the thicker Bard XT stent (.0059) and the AVE MS2 stent (.0068). After a series of analyses, he concluded that the strongest predictors for clinical restenosis were: 1. Final minimal luminal diameter (MLD) 2. Vessel size 3. Stent length 4. Prior myocardial infarction (MI) 5. Diabetes Strut thickness was put into the equation, but it was not a significant predictor of TLR (refer to Figure III). This analysis offers further evidence that clinical restenosis was more dependent on the clinical case mix analyzed than the thickness of the stent strut. TLR rates vary with changes in the patient population. Dr. Kalon Ho et al analyzed the HCRI data and compared diabetes, non-diabetes, small vessels and long lesions. His analysis of angiographic restenosis mimics the clinical restenosis data we just reviewed. Angiographic restenosis remains a significant problem in our diabetics with small vessels and long lesions (Figure IV, next page). Dr. Ho also reviewed angiographic binary restenosis in his 9000 stent patient data set (see Figure V). Two models, A and B, were postulated. He concluded that strut thickness was a weak predictor for angiographic restenosis when compared to the more powerful variables of MLD, diabetes and lesion lesion. Only in model B did strut thickness appear as powerful as stent length (p = .02 “ .06), an order of magnitude below the bigger predictors seen in either model. Late Loss: Becoming the New Gold Standard? In an effort to find a pure assessment of restenosis, we have come upon the concept of late loss. Late loss is a luminal dimension change over time, after coronary intervention. The initial minimal luminal diameter (MLD) is markedly enlarged by the effect of the PCI (acute gain). Over time, the loss of a portion of the gain due to intimal hyperplasia is termed late loss, yielding the final long term result or net gain. The concept of late loss has been advanced in the clinical arena. Late loss is the direct angiographic measure of lumen loss. It is the measure of pure neointimal hyperplasia. Late loss avoids errors that have occurred when assessing loss index, which is essentially late loss over acute gain. Since it is a continuous variable, it is much more powerful when assessed in a statistical model than binary angiographic restenosis. The impact of late loss can be well appreciated by the following example. Let us assume that the late loss is 0.2 mm. In a 2.0 mm vessel, you have lost 10% of your gain, but in a 3.0 mm vessel, you lose 7%. However, the late loss of most bare stents is 0.8 mm. In a 3.0 mm vessel, you have lost 46% of your acute gain, while in a much smaller 2.0 mm vessel, you have lost 64% of your gain. Late loss has a powerful impact on the long-term outcome of patients. In Figure VI, one can compare the late loss in various bare metal stents. The thinnest strut stents are on the left, starting with the Multi-Link stent, to the thicker stents like the BX Velocity, on the right. Outlined in various colors are the different trials that involved these stent designs. This chart clearly demonstrates that late loss is independent of strut thickness. A late loss between 0.8 and 1.0 mm is also independent of the metal material. Cobalt-chromium stents like the Multi-Link Vision (.0032 strut) had a late loss of 0.83 mm and the Medtronic Driver stent (.0036 thickness), had a late loss of 0.91mm. The thin strut BiodivYsio stent had a late loss of 0.8 mm, while the Multi-Link Penta stent had a late loss of 0.9 mm. None of these numbers are different than the thicker strut BX Velocity stent in the VENUS trial, the VELVET trial, or the control arm of the RAVEL trial. Conclusions 1. When assessing restenosis, we must review the time course assessment. Are we reviewing 6-month, 9-month, or 12-month data? This has confused the interventional community, as new stent designs have 6-month outcomes while the drug-eluting stent data are sampled at 9 months. 2. Case mix is important. The most crucial factors are vessel size, lesion length and the presence of diabetes. The most important determinants of restenosis for bare metal stents are these clinical variables. If we don’t have these variables, it is difficult to compare trials. 3. Strut thickness. Strut thickness is important for flexibility and conformability, and to optimize scaffolding. It may be a more important factor in smaller vessels, but it does not have an major effect on clinical outcomes or late loss. In addition, its effect on angiographic binary restenosis is relatively small. 4. Late loss is the final arbitrator. It is not influenced by vessel size and remains a robust measure of clinical effect. I believe it will become the gold standard for assessing the potency of various drug-device combinations. To obtain your CME/CEU credit, please download the PDF adjacent to this article. Instructions are included.