Late-Breaking Trials 

Conference Coverage 

Editor’s Note: The following reports from the annual scientific sessions of the American College of Cardiology held March 30 to April 2, 2003, were obtained by handwritten notes, press releases, and news reports. 


Elevated plasma homocysteine has been correlated with increased rates of adverse cardiac events in patients with established atherosclerosis, and a growing body of data have suggested that reduction of plasma homocysteine with vitamin therapy reduces the risk of atherosclerotic coronary events even in those patients without overtly elevated homocysteine levels. Thus, the potential benefit of homocysteine-lowering vitamin therapy for prevention restenosis and adverse cardiac events after percutaneous coronary intervention (PCI) has been an area of interest in the interventional cardiology community. Enthusiasm for this concept was further heightened by a relatively recent study reported by Schnyder and colleagues from the University of Bern,1 which compared angiographic and clinical outcomes in 553 patients undergoing PCI who were randomized to receive homocysteine-lowering therapy (folic acid [1 mg/d], vitamin B12 [0.4 mg/d], and vitamin B6 [10 mg/d]) for 6 months after the index procedure. The Swiss Heart Study showed significant increases in minimal lumen diameter (MLD) and reduced restenosis at 6-month angiographic follow-up, a significant reduction in the composite clinical end point (15.4% vs 22.8%, P = 0.03) and the need for ischemia-driven repeat revascularization, and as well as trends toward decreased death and myocardial infarction (MI) at 1-year follow-up for patients receiving folate +B12+ B6 therapy. Based on the results of this study and enthusiasm for vitamin therapy among US patients in general, many interventional cardiologists began recommending the addition of folate +B12+ B6 supplementation to the pharmacologic regimen of their post-PCI patients.

The results from FACIT were presented by Dr. Helmut Lange (Bremen Heart Center, Bremen, Germany). The FACIT investigators studied 626 patients who underwent successful stent implantation and were randomized to receive a vitamin preparation containing folic acid (1.2 mg/d), vitamin B12 (0.06 mg/d), vitamin B6 (48 mg/d), or placebo. The primary end point was angiographic MLD at 6 months. Secondary end points included angiographic restenosis, major adverse cardiac events (MACE), and target vessel

revascularization (TVR). As in previous studies, homocysteine levels decreased in patients receiving folate +B12+ B6 supplementation. However, in contrast to the Swiss Heart Study, FACIT showed a significant reduction in 6-month MLD in patients receiving folate +B12+

B6 supplementation (1.59 mm vs 1.74 mm), as well as higher percent diameter stenosis and angiographic restenosis (35% vs 27%). In addition, higher rates of TVR (15.8% vs 10.6%) and MACE (16.8% vs 10.9%) were also reported in the folate +B12+ B6 supplementation group with clinical event curves separating within 2-3 months of follow-up. Subgroup analysis showed beneficial effects of folate +B12+ B6 supplementation on only women and diabetics. The FACIT investigators were somewhat surprised and "disappointed" to conclude that, based on the results of their study, folate+B12+ B6 supplementation after successful coronary stent implantation was associated with increased in-stent restenosis and higher rates of adverse clinical events.

Comment by Sarah M. Vernon, MD

The results of FACIT were unexpected, on the basis of what is known about the pathogenesis of homocysteine-induced vascular injury, the predictive value of plasma homocysteine levels on outcomes after PCI, and the beneficial effect of homocysteine-lowering therapy on cardiovascular event rates in other patient populations with vascular disease. In addition, and even more perplexing, the results of FACIT are in direct contradiction with those of the previously published Swiss Heart Study. The most obvious major difference between these 2 studies is that only 50% of patients in the Swiss Heart study underwent stent implantation compared with 100% of patients in FACIT. There are also differences in mode and timing of initiation of therapy, as well as differences in dosing of all 3 elements included in the homocysteine-lowering vitamin "cocktails" administered in the 2 studies. The Swiss study included more diabetics and patients on statins, and additional distinctions between the patients included in the 2 studies may become evident when the results of FACIT are published. As Lange pointed out in his discussion, further study will be needed to reach definitive conclusion about the use of homocysteine-lowering therapy in patients undergoing PCI. However, in the meantime, the potential untoward effects of folate + B12 and +B6 therapy as administered in FACIT would suggest that homocysteine-lowering therapy should not be initiated routinely after coronary stent implantation. In turns out that, contrary to my commentary on this issue in October,2 despite the results of the Swiss Heart Study and the fact that folate +B12+ B6 therapy is "inexpensive, readily available, well tolerated and readily embraced by patients," there may, in fact, be a real "down-side" to recommending it for our post-PCI patients.

Dr. Vernon is Assistant Professor of Medicine Director, VAMC Cardiac Catheterization Laboratory University of New Mexico Health Sciences Center Albuquerque, NM.


1. Schnyder G, et al. JAMA. 2002;288:973-979.

2. Vernon S. Clinical Cardiology Alert. 2002;21:73-74.


Dr. Adnan Kastrati (Deutsches Herzzentrum, Munich, Germany) presented the results of Intercoronary Stenting and Antithrombotic Regimen-Rapid Early Action for Coronary Treatment (ISAR-REACT).

This study evaluated the effectiveness of the glycoprotein IIb/IIIa inhibitor abciximab compared to placebo in low-risk patients undergoing elective percutaneous coronary interventions (PCI)—all of whom had been pretreated with high-dose clopidogrel. The clopidogrel regimen

used in this study was a 600-mg oral loading dose given at least 2 hours before the index procedure, followed by 75 mg twice daily for 3 days, then 75 mg daily for at least 4 weeks. All patients received 100 mg of aspirin. Patients with acute coronary syndromes, recent myocardial infarction (within 14 days), insulin requiring diabetes mellitus, saphenous vein procedures, or reduced left ventricular systolic function (EF < 0.30) were excluded from the study. The primary end point was a composite of death, myocardial infarction (new Q-waves or CK elevation = 3 times normal) or target vessel revascularization (TVR) at 30 days. Safety end points included major and minor bleeding, thrombocytopenia, and need for transfusion. A total of 2159 patients were randomized in the study—1079 to abciximab and 1080 to placebo. Ninety-one percent of the patients underwent stent implantation.

There were no significant differences between abciximab- and placebo-treated patients with regard to the composite primary end point (4.2% vs 4.06%) or to any individual component of the composite. There were no differences between the groups with respect to bleeding events, although thrombocytopenia was more common and transfusion rates were significantly increased in the abciximab group. Kastrati concluded that in the setting of elective PCI in the low-risk nondiabetic patient who has been pretreated more than 2 hours in advance with high-dose clopidogrel, there is no additional benefit of abciximab administration. Kastrati pointed out that these data cannot be extrapolated to higher-risk patients such as those with ACS and stated that a similar trial in a high-risk PCI population was planned.

Comment by Sarah M. Vernon, MD

The most recent ACC/AHA Practice Guidelines for management of ACS (2002) recommend clopidogrel as the thienopyridine of choice given its rapid onset of action and superior safely profile. Also, they recommend a loading dose of 4-8 tablets (300-600 mg) when "rapid onset of action is required." This recommendation is based, at least in part, on data from the CURE and PCI-CURE trials, which used a 300-mg loading dose of clopidogrel, and smaller studies evaluating the effect of a clopidogrel loading dose on ADP-induced platelet aggregation or ex vivo thrombus formation. Even careful reading of the guideline and its bibliography does not elucidate the origins of the recommendation for a 600-mg loading dose. More recently, the CREDO trial1 evaluated the benefits of clopidogrel administration in patients undergoing elective stent implantation both in terms of timing and duration of clopidogrel administration, demonstrating a 38.6% reduction of adverse events in patients pretreated more than 6 hours before their index procedure and a 26.9% reduction in death, MI, and stroke when continued for 1 year. Data from CREDO has convinced operators in our catheterization laboratory to initiate clopidogrel loading as early as possible before diagnostic cardiac catheterization with a moderate-to-high likelihood of follow-on PCI.

The results from ISAR-REACT would suggest that in the low-risk elective PCI patient, pretreatment with clopidogrel (albeit at a higher dose than in CREDO) provides antiplatelet activity adequate to obviate the addition of a GP IIb-IIIa inhibitor in the management of these patients. This result is, in fact, not particularly surprising, as there have never been data demonstrating significant benefit of GP IIb-IIIa inhibitor administration in low-risk, nondiabetic patients undergoing PCI. While many interventional operators will tell you that the low-risk elective patient included in ISAR-REACT constitutes the minority of his/her practice, these patients do exist. With GP IIb-IIIa inhibitor administration rates estimated on the order of 70-90% of PCI procedures performed in the United States, it’s clear that at least some of the patients receiving GP IIb-IIIa inhibitors, drugs with extremely high cost and some bleeding risk, are unlikely to obtain any real benefit from receiving them. While it’s true that the results from ISAR-REACT can only be applied to a relatively narrow sector of patients undergoing PCI, perhaps they will give us the fortitude to choose clopidogrel preloading over abciximab administration for our low-risk PCI patients. 

Dr. Vernon is Assistant Professor of Medicine Director, VAMC Cardiac Catheterization Laboratory University of New Mexico Health Sciences Center Albuquerque, NM.


1. Steinhubl SR, et al. JAMA. 2002;288:2411-2420.

Indoor Smoking Ban Ordinance

Smoking activates platelet aggregation, decreases endothelial-dependant vasodilatation, increases heart rate, and decreases heart rate variability. It has been shown that passive smoke increases the incidence of myocardial infarction in spouse studies. Other studies show that a smoke-free workplace decreases the number of cigarettes smoked per day in smokers. These data suggest that indoor smoking bans may reduce the risk of myocardial infarction in smokers and others.

Thus, investigators took advantage of a new smoke-free ordinance in Helena, Mont, to test this hypothesis. The results were presented by Dr. Richard Sargent. On June 4, 2002, Helena passed an ordinance that banned smoking in the workplace, public places, and even the Native American casinos. This ordinance was rescinded on December 3, 2002, by a court order, resulting in 6 months of intervention data, which they compared to data from the previous 4 years (historical control). Compliance with the law was excellent while it was in force. Since almost all myocardial infarctions are admitted to 1 hospital in Helena, the investigators believed that they did not miss any. However, infarctions secondary to other conditions (ie, post surgery) were excluded. In addition, seasonal variation was taken into consideration. Finally, to be included, infarction patients had to have been in Helena for at least 1 night or 1 meal. The primary end point was acute myocardial infarction per month, which was reduced by 60% during the time the ordinance was in force vs the seasonally adjusted historical control period, P < .002. Patients who resided in nearby counties not affected by the ordinance, who used the same hospital, showed no change in infarction rates.

The investigators concluded that a ban on indoor smoking in public and workplaces had an immediate and substantial beneficial effect on acute myocardial infarction rates. They suggest that communities who wish to enact such ordinances first educate their state legislators so that the court problems they had can be avoided.

Comment by Michael H. Crawford, MD

This is an impressive result in this simple study and points to the power of the law to affect public health. Similar results have been seen with seat belt laws, helmet laws, etc, but this is the first study showing the power of law to affect cardiovascular health. Of course, it would be preferable if people just didn’t smoke or at least didn’t in public places, but achievement of such an aim seems impossible despite nearly 40 years of antismoking publicity since the first Surgeon General’s report in 1964. Unfortunately, the citizens of Helena, Mont, no longer benefit from this ordinance, since a court order rescinded it after 6 months. The investigators did not go into the details of the court challenge, but did state that education of the state legislature and ensuring their support probably would have avoided the court challenge. They are hopeful they can get the law reinstated soon. This would be good since it would be advantageous to see more than 6 months of data. A more robust study with the same results would be a powerful tool to help enact similar ordinances in other cities.

Dr. Crawford is Professor of Medicine, Mayo Medical School; Consultant in Cardiovascular Diseases, and Director of Research, Mayo Clinic, Scottsdale, AZ.


Drs. Arthur Feldman and Michael Bristow presented the Comparison of Medical, Resynchronization, and Defibrillation Therapies in Heart Failure (COMPANION) trial at the Late-Breaking Clinical Trials session at the ACC. COMPANION was a randomized trial of New York Heart Association Class 3 or 4 heart failure patients with an ECG QRS duration > 120 msec, a left ventricular ejection fraction < 35%, and an end-diastolic dimension > 60mm. Patients were randomized to either optimal pharmacologic therapy, cardiac resynchronization therapy with a biventricular pacemaker, or cardiac resynchronization therapy with a biventricular defibrillator. The randomization was in a 1:2:2 format, so that only 20% had no device. Optimal medical therapy was carefully monitored and included beta-blockers, diuretics, ACE inhibitors or angiotensin receptor blockers, spironolactone, and digoxin, whenever tolerated, for at least 1 month. A total of 1520 patients were randomized in the trial. Implant success rate for the resynchronization pacemaker or defibrillator was about 90%.

The mean left ventricular ejection fraction in each group was close to 22%. About 55% of the patients had ischemic heart disease and 45% nonischemic cardiomyopathies. The primary end point was a combination of all-cause death and all-cause hospitalizations over 12 months. Deaths and hospitalizations were reduced by both cardiac resynchronization alone and cardiac resynchronization plus defibrillation by 19%, which reached the efficacy boundaries of the trial so it was stopped early.

Mortality in the medical therapy group was 19%. There was a nonsignificant 23.9% reduction in all-cause mortality with cardiac resynchronization and a highly significant reduction of 43.4% with the addition of defibrillation to resynchronization therapy. Benefit was seen in patients with both ischemic and nonischemic disease.

Feldman and Bristow conclude that cardiac resynchronization plus defibrillation therapy produces both symptomatic and mortality benefits in patients who meet these entry criteria. Resynchronization alone results in reduced hospitalization and a trend toward decreased mortality with a delayed onset of effect on mortality.

Comment by John DiMarco, MD, PhD

Cardiac resynchronization therapy is now being offered to many patients with left ventricular dysfunction and a widened QRS. The COMPANION trial provides further data supporting the benefit of this approach. The important observations here are that the effects of adding a defibrillator are additive to effects seen with cardiac resynchronization alone. This results in a very substantial reduction in hospitalization mortality even though patients were treated with pharmacologic therapy for heart failure.

There were several limitations in the trial. Since the devices that provide resynchronization therapy were introduced during the trial, there was an excess dropout rate among patients in the optimal medical therapy group. Complete data on those patients were not presented. In addition, the hospitalization for the device was not counted as an end point in the trial. If there were changes in medical therapy during that initial hospitalization, that might have accounted for some of the improvement in hospitalization rates. Hopefully, once the full data from the trial become available, many other clinical questions will be answered.

Dr. DiMarco is Professor of Medicine Division of Cardiology University of Virginia, Charlottesville.


At the late-breaking clinical trials session of the American College of Cardiology meeting Dr. Jonathan Halperin presented the results of SPORTIF III, a trial investigating the use of a direct thrombin inhibitor, ximelagatran in patients with atrial fibrillation. Ximelagatran is an oral direct thrombin inhibitor that produces the onset of anticoagulation within 2 hours after oral ingestion. In previous trials, it has been shown to have a wide therapeutic margin. It is not extensively metabolized so it has a low potential for food and drug interactions. Because of these factors, no anticoagulation monitoring is required and a single dose (36 mg b.i.d.) can be used except in the presence of severe renal dysfunction.

In SPORTIF III, 3407 patients with atrial fibrillation were randomized to either adjusted dose warfarin to maintain an INR between 2 and 3 and fixed dose ximelagatran (36 mg b.i.d.). Patients had to have atrial fibrillation and at least 1 additional risk factor for stroke. Treatment was open label to simplify monitoring of prothrombin times in the warfarin group. The primary end points were strokes and systemic embolic events based on an intention-to-treat analysis. An on-therapy analysis was also reported. The hypothesis was that ximelagatran would not be inferior to warfarin in these patients.

The patients enrolled were representative of an elderly atrial fibrillation population. More than 80% had been in atrial fibrillation for more than a year. More than 90% had persistent or permanent atrial fibrillation. One-third of patients were older than 75; 40% were between 65 and 75. Sixty-five percent of the patients were male. Hypertension was seen in 66% of the patients; prior stroke, TIA, or embolism in 24% of the patients. Warfarin therapy was carefully monitored. Sixty-six percent of the INR values obtained during the study were within the prescribed therapeutic range, and 81% were within an expanded range of 1.8 to 3.2. In the intention-to-treat analysis, there were 56 strokes or embolic events in the warfarin group for an annual rate of 2.3% per year. Among the ximelagatran patients, there were 40 events for a rate of 1.6% per year. This met the statistical criteria for noninferiority. In the secondary on-treatment analysis, the relative risk reduction for stroke and systemic embolism was 41% (P = 0.018). Major and minor bleeding events were also evaluated. The rates for intercranial hemorrhage (0.2% vs 0.5%) and major bleeding (1.3% vs 1.8%) both favored ximelagatran, but the differences were not statistically significant. When major and minor bleeding events were combined, the rates were 25% per year on ximelagatran and 29.5% per year on warfarin (P = 0.007). There were no significant differences in major cardiovascular events or in all-cause mortality between the 2 groups. Liver enzyme elevation was more common with ximelagatran. A total of 6.5% of the patients on ximelagatran developed an ALT greater than 3 times the upper limit of normal vs only 0.7% on warfarin. Most of the liver enzyme elevations occurred in the first 6 months of therapy and resolved even if the drug was continued. There were no cases of hepatic failure. Halperin concluded that ximelagatran is as effective as well-controlled warfarin in preventing stroke and systemic embolism. Ximelagatran was associated with less total bleeding than was seen during warfarin therapy, but elevated liver enzymes were more common.

Comment by John DiMarco, MD, PhD

Warfarin has been the standard for anticoagulation for more than 50 years. In the late 1980s, a number of trials conclusively showed the benefits of warfarin in patients with atrial fibrillation who were either elderly or had other risk factors. Warfarin therapy, however, has many limitations. Because the drug has many potential interactions with food and other medications, frequent monitoring of the prothrombin time is necessary. Bleeding is also common during therapy. Warfarin’s onset of action is delayed, and reversing warfarin’s effects can be difficult. Ximelagatran is a new antithrombotic agent. After oral administration, it is converted to melagatran, which is a direct inhibitor of thrombin. The drug is absorbed in the small intestine. Peak concentrations of melagatran are reached 30 minutes after oral ingestion. Peak ximelagatran concentrations are seen about 2 hours after absorption. In melagatran, the active metabolite is predominantly renally eliminated. In prior trials, the drug had been shown to have a wide margin of safety, and dosage adjustment is not thought to be required unless the creatinine clearance is below 30 mL/min. In this important study, ximelagatran was shown not to be inferior to warfarin. All of the measurements appear to favor ximelagatran over warfarin. Given the many disadvantages of warfarin, it is likely that ximelagatran will be an attractive option for many patients who require anticoagulant therapy.

Ximelagatran has also been studied in a number of trials with venous thromboembolism for both primary and secondary prevention. Further data, however, will be needed concerning its use in other situations. In particular, for cardiologists, it will be important to have data about its use in patients with prosthetic valves and also in patients with decreased renal function.

Dr. DiMarco is Professor of Medicine Division of Cardiology University of Virginia, Charlottesville.